Optimized design and operation strategy of a Ca-Cu chemical looping process for H 2 production

Transcription

1 Optimized design and operation strategy of a Ca-Cu chemical looping J.R. Fernández*, J. M. Alarcón, J.C. Abanades jramon@incar.csic.es CO 2 Capture Group INCAR-CSIC (Spanish Research Council) LOGO World Hydrogen Energy Conference, June, 13 th -16 th, 2016, Zaragoza

2 Outline 1. Introduction 2. Model description and experimental validation 3. Dynamic operation of the Ca-Cu looping process 4. Process design 5. Conclusions

3 Outline 1. Introduction 2. Model description and experimental validation 3. Dynamic operation of the Ca-Cu looping process 4. Process design 5. Conclusions

4 Sorption Enhanced Reforming (SER) H 2 rich gas Carbonation (Reforming+Shift) CO 2 sorbent Fuel+ Steam Thermodynamics Equilibrium is shifted to H 2 production (>90 vol.% H 2 on a dry basis) The reforming process in one single stage Overall reaction is slightly exothermic

5 Sorption Enhanced Reforming (SER) H 2 rich gas CO 2 Carbonation (Reforming+Shift) CaCO 3 CaO Calcination Heat IN Fuel+ Steam High T for P CO2 calcination = 0.1 bar required T > 750 o (T>850ºC) C Efficiency P CO2 penalties = 1 bar T > 850 o C P CO2 = 15 bar T > 1100 o C Unsurmountable challenge?

6 Proposed solutions for sorbent regeneration Oxy-combustion of additional fuel in a regenerator External heating through high temperature heat transfer surfaces Direct heating by contact with hot solids or hot gases obtained from additional fuel combustion Use of the waste heat from a fuel cell coupled to the SER process Low thermal efficiencies and/or High equipment cost

7 The Ca/Cu chemical looping process (Abanades & Murillo, CSIC, EP , 16th Sep 2009) Based on the Unmixed reforming concept (Kumar et al., 2000): Calcium looping + Chemical Looping Combustion CaO for CO 2 capture in the reformer (higher H 2 yield) The reduction of oxigen carrier (CuO) with fuel gas supplies the heat for CaCO 3 calcination Endothermic/exothermic reactions coupled in the same bed matrix Higher efficiency Lower equipment cost

8 The Ca/Cu chemical looping process Steam Reforming + CaO Carbonation

9 The Ca/Cu chemical looping process Cu Oxidation

10 The Ca/Cu chemical looping process CuO Reduction + CaCO 3 Carbonation

11 The Ca/Cu chemical looping process CO 2 -rich stream suitable for storage CuO Reduction + CaCO 3 Carbonation

12 H 2 PROCESS SCHEME PSA PSA off-gas (CH 4, H 2, CO 2,CO) GT N 2 (CO 2 ) ~ N 2 (CO 2 ) CH 4 +H 2 O H 2 -rich gas A B N 2 (CO 2 ) B C C H 2 O N 2 (CO 2 ) Syngas (CO,H 2,CO 2,H 2 O) CH 4 +H 2 O Air CO 2

13 Outline 1. Introduction 2. Model description and experimental validation 3. Dynamic operation of the Ca-Cu looping process 4. Process design 5. Conclusions

14 Pseudo-homogeneous model Mass and thermal dispersion in axial direction Ideal gas behaviour Negligible intraparticle mass and temperature gradients Constant bed void fraction (ε=0,5) Uniform particle size (d p =5 mm) Uniform mixture of the solids in the packed bed CuO-based particles: 60% active Ca-based particles: 35% active Cycle duration: 15 min MODEL DESCRIPTION Assumptions

15 Mass balance: MODEL DESCRIPTION Energy balance: Axial pressure distribution (Ergun equation): / ;

16 MODEL DESCRIPTION Axial mass dispersion coefficient (Edwards and Richardson, 1968) Effective axial heat dispersion (Vortmeyer and Berninger, 1982) (Krupiczka, 1967) (Gunn and Misbah, 1993) Gas-to-particle heat transfer coefficient (Gunn, 1987)

17 Kinetics of SMR and WGS MODEL DESCRIPTION (Xu and Froment, 1989) Kinetics of CuO reduction/cu oxidation / (Abad et al., 2007) Kinetics of CaCO 3 calcination (Martínez et al., 2012)

18 TEST RIG AT INCAR-CSIC EXPERIMENTAL VALIDATION Material: Inconel Internal diameter: m External diameter: m Height: 1 m Solids mass: around 1 kg Multipoint type K termocouple (15 points) Insulating material: quartz wool IR and Paramagnetic Gas Analizers

19 EXPERIMENTAL VALIDATION Reduction/calcination tests Temperature, ºC min 5min 10min 2min_exp 5min_exp 10min_exp 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 L, m Temperature out, ºC Time, min Mole percent out (dry basis) 100 H2exp 90 CO2exp 80 CO exp 70 H2mod CO2mod 60 COmod Time, min The dynamic model gives a good description of the experimental results Evolution of the temperature along the fixed bed Dynamic profiles at the reactor exit (gas temperature and gas composition) Prediction of the breakthrough period

20 EXPERIMENTAL VALIDATION Cu oxidation tests The dynamic model gives a good description of the experimental results Evolution of the temperature along the fixed bed Dynamic profiles at the reactor exit (gas temperature and gas composition) Prediction of the breakthrough period

21 Outline 1. Introduction 2. Model description and experimental validation 3. Dynamic operation of the Ca-Cu looping process 4. Process design 5. Conclusions

22 STAGE A: SORPTION ENHANCED REFORMING PSA H 2 -rich gas H 2 PSA off-gas (CH 4, H 2, CO 2,CO) ~ GT N 2 (CO 2 ) N 2 (CO 2 ) CH 4 +H 2 O Reference case: Nm 3 H 2 /h (after PSA) Inlet flow: 130 mol CH 4 /s (around 100 MW th ) A B N 2 (CO 2 ) B CH 4 +H 2 O Air Operating Conditions T gin = 700ºC P= 10 bar S/C ratio =3 H 2 O C CO 2 C N 2 (CO 2 ) Syngas (CO,H 2,CO 2,H 2 O) To avoid CaO hydration Steam partial pressure, bar 20 Shaube Halstead Samms 15 Nikulshina Barin Temperature, ºC

23 STAGE A: SORPTION ENHANCED REFORMING Axial profiles t=0 min t=1min t=5 min t=8 min t=10 min t=15min Heat plateau at T max =730ºC 0,50 0,45 0,40 t=0 min t=3 min t=6min t=10 min t= 15 min Temperature, ºC X_CaO 0,35 0,30 0,25 0,20 0, , , , L, m L, m T min downstream around 600ºC (no CaO hydration) Initial T profile resulting from a previous stage C Initially carbonated from a previous stage C Sharp carbonation front (enhanced Reforming and WSG) Partial carbonation downtream the heat plateau

24 STAGE A: SORPTION ENHANCED REFORMING Profiles at the reactor exit Cycle duration Cycle duration Mole percent out (dry basis),% Time, min H2 CH4 CO CO2 Temperature, ºC Time, min Product gas composition (dry basis): 90% H 2 8% CH 4 up to 4% CO (heat plateau) <1% CO 2 T min at 590ºC Heat plateau at T max =730ºC

25 STAGE B: Cu OXIDATION H 2 Operating Conditions PSA H 2 -rich gas PSA off-gas (CH 4, H 2, CO 2,CO) ~ GT N 2 (CO 2 ) N 2 (CO 2 ) CH 4 +H 2 O T gin = 300ºC P= 20 bar A B N 2 (CO 2 ) B C C H 2 O N 2 (CO 2 ) Syngas (CO,H 2,CO 2,H 2 O) CH 4 +H 2 O Air Recycle ratio=0,9 Inlet flow=4300 mol/s O 2,in =2,6 %.vol CO 2,in =1 %.vol CO 2 T max =860ºC To minimize CO 2 loss by sorbent calcination

26 STAGE B: Cu OXIDATION Temperature, ºC t=0min t=0min t=1min t=5 min t=10 min t=15min RF HF L, m Initial Reaction T profile Front resulting (RF) advances from previous slowerstage than A Heat (most Exchange of solids Front bed(hf): initially O 2 dilution at 520ºC) T sufficiently high to allow a rapid Cu oxidation

27 STAGE B: Cu OXIDATION Temperature, ºC t=0min t=1min t=5 min t=10 min t=15min RF HF Solids conversion 1,0 0,9 0,8 0,7 0,6 0,5 0,4 0, , L, m Reaction Front (RF) advances slower than Heat Exchange Front (HF): O 2 dilution 0,1 0, L, m Narrow oxidation fronts T sufficiently high to allow a rapid Cu oxidation Part of the bed at reactor exit is left calcined

28 Profiles at the reactor exit STAGE B: Cu OXIDATION Cycle duration Cycle duration Mole percent out, % O2 CO2 Temperature, ºC To be recirculated To GT Time, min Time, min Product gas composition: u RF <u HF t<11 min: no O 2 (prebreakthrough) 8<t<13 min: increase of CO 2 in the flue gas (heat plateau at Tmax) up to 2% vol. CO 2 T gout low during the first part of stage B Heat plateau at T max =860ºC (t>10min)

29 STAGE B : Solid/gas Heat Exchange PSA Temperature, ºC H 2 -rich gas H 2 PSA off-gas (CH 4, H 2, CO 2,CO) CH 4 +H 2 O ~ t=15min GT N 2 (CO 2 ) N 2 (CO 2 ) A B N 2 (CO 2 ) B 100 Air CH 4 +H 0 2 O CO 2 L, m C C H O N 2 (CO 2 ) Syngas (CO,H 2,CO 2,H 2 O) Operating Conditions P= 20 bar Not all gas recycle to B, only the fraction at the highest temperature ( 860ºC) Inlet flow=1300 mol/s At the end of stage B most of the bed is left at around 300 ºC

30 STAGE B : S-G HEAT EXCHANGE ,40 Temperature, ºC t=0min t=2min t=5min t=10min t=15 min L, m X_Ca 0,35 0,30 0,25 0,20 0,15 0,10 0,05 0, L, m t=0min t=1min t=2min t=15min A large part of the bed is left at favourable T for the reduction/calcination stage (around 50% at T>800ºC) Most of previously calcined Ca-sorbent is carbonated (1.5% CO 2 in gas feed))

31 STAGE C: REDUCTION/CALCINATION PSA H 2 -rich gas H 2 PSA off-gas (CH 4, H 2, CO 2,CO) CH 4 +H 2 O ~ GT N 2 (CO 2 ) N 2 (CO 2 ) A B N 2 (CO 2 ) B C C H 2 O N 2 (CO 2 ) Syngas (CO,H 2,CO 2,H 2 O) Air CH 4 +H 2 O CO 2 Operating Conditions P= 1 bar Fuel gas: PSA off-gas and syngas from SMR stage C T gin =700ºC Composition inlet fuel gas: 40%H 2, 30%CH 4, 15% CO, 5%CO 2 Cu/Ca molar ratio=2 CO 2 stream suitable for storage

32 STAGE C: REDUCTION/CALCINATION Temperature, ºC min 5min 10min 15min L, m Rapid increase of temperature due to the reduction of CuO Heat plateau at T max around 900ºC u RF >>u HF (most of the bed is left at 900ºC) Solids conversion 1,0 0,8 0,6 0,4 0,2 0,0 X_Ca X_Cu t=0 min L, m t=5 min t=15 min t=10 min For t=0 min, part of the solids are already converted. But this does not cause hot spots during the operation because both Ca- and Cuconverted solids are at reactor inlet. Both reduction and calcination fronts advance together leaving behind the solids totally converted

33 STAGE C: REDUCTION/CALCINATION Cycle duration Cycle duration Mole percent out, % H2 CH4 CO H2O CO Time, min Temperature, ºC Time, min t<4 min: <2% CO 2 (low T of product gas) t>5 min: maximum CO 2 content (55%) Complete fuel conversion during pre-breakthrough T gout low during the first part of stage C Heat plateau at 900ºC (t>8min)

34 STAGE C : STEAM METHANE REFORMING H 2 PSA H 2 -rich gas PSA off-gas (CH 4, H 2, CO 2,CO) ~ GT N 2 (CO 2 ) N 2 (CO 2 ) A B N 2 (CO 2 ) B C CH 4 +H 2 O C P= 1 bar Tgin=700ºC S/C=1.5 Operating Conditions H 2 O N 2 (CO 2 ) Syngas (CO,H 2,CO 2,H 2 O) CH 4 +H 2 O Air CO 2 Cooling solid bed for Stage A (SER) High content of CO+H 2 Low CO 2 formation (minimum Ca-solids carbonation) Readily available source of CO and H 2 for Stage C (reduce the Cu/Ca ratio required)

35 STAGE C : STEAM METHANE REFORMING Temperature, ºC L, m t=0min t=1min t=5 min t=8 min t=10 min t=15min cch 4, kmol/m 3 0,006 0,005 0,004 0,003 0,002 0,001 0, L, m t=1min t=5min t=8 min t=10 min t= 15 min X_CaO 0,40 0,35 0,30 0,25 0,20 0,15 0,10 0,05 0, L, m t=5min t=10 min t= 15 min Drop in temperature at reactor inlet because of the steam methane reforming Almost complete conversion of CH 4 when it reaches the solids at 900ºC Only carbonation at the reactor inlet where the low temperature favours carbonation reaction Most of the bed is left at a temperature around 600ºC (suitable for the subsequent SER stage)

36 Outline 1. Introduction 2. Model description and experimental validation 3. Dynamic operation of the Ca-Cu looping process 4. Process design 5. Conclusions

37 PROCESS DESIGN Pressure drop (ΔP/P), % 80 5 t=10min 70 t=15min 60 t=20min ,0 1,5 2,0 2,5 3,0 L/D ratio Gas velocity, m/s Reactor cross-sectional area, m 2 20 t=10min t=15min 15 t=20min ,0 1,5 2,0 2,5 3,0 L/D ratio Total volume, m 3 10min 15min 20min Cycle duration Shorter cycle times lead to smaller areas, but also to higher pressure drops Larger L/D ratios reduce the total area required, but at the expense of higher ΔP Limiting stage: Stage B (with the highest inlet gas flow) H 2 efficiency (LHV basis)=76% Assumptions: CO 2 capture efficiency=96% Minimum L/D=2 Maximum pressure drop (ΔP/P)=10% Minimum duration of single stage=15 min 5 reactors, L=6 m, D=2,9 m

38 Outline 1. Introduction 2. Model description and experimental validation 3. Dynamic operation of the Ca-Cu looping process 4. Process design 5. Conclusions

39 CONCLUSIONS Each stage of the Ca-Cu looping process has been simulated using a pseudo-homogeneous model and dealing with solids temperature and solids conversion profiles left in previous stages SER stage has been designed at a relatively low pressure (10 bar) and low S/C ratio (3) to avoid CaO hydration The use of a PSA unit downstream SER reactor allows H 2 efficiency and CO 2 capture efficiency to be improved A process design for a reference H 2 plant (30000 m 3 NH 2 /h) shows that 5 reactors (6 m long with an inner diameter of 2,9 m) operating in cycles of 15 min are required. Future work: this reactor model will be incorporated into a comprehensive process model for the complete integration of the Ca-Cu looping process.

40 THANKS FOR YOUR ATTENTION! Any questions..? LOGO World Hydrogen Energy Conference, June, 13 th -16 th, 2016, Zaragoza