Understanding the Path to High- Efficiency Chemical Engines

Size: px

Start display at page:

Download "Understanding the Path to High- Efficiency Chemical Engines"

Sherman Clark
5 years ago
Views:

1 Understanding the Path to High- Efficiency Chemical Engines Chris F. Edwards Kwee Yan Teh, Shannon Miller, Matthew Svrcek, Sankaran Ramakrishnan, and Adam Simpson Advanced Energy Systems Laboratory Department of Mechanical Engineering Stanford University

2 40% 34% >74% of U.S. CO 2 is emitted by engines.

3 Engines All engines have three essential features: they produce work (by definition) they require a resource (1 st Law) they reject energy to surroundings (2 nd Law) Energy Resource Engine Rejected Energy (surroundings) Work

4 Efficiency Limits Only four ways to transfer energy: work (entropy free) heat (energy transfer due to ΔT ) matter (internal and external) External: Internal: K.E., gravitational P.E., electrostatic P.E. thermal, chemical, nuclear radiation (not considered here) It is the combination of energy resource and surroundings that determines the ultimate efficiency limitation of an engine (exergy).

5 Classifying Engines by Energy Resource Space Atmosphere Sun High K.E. Moon Surface Anthrosphere Biosphere Water Lithospher Accumulated Hydrosphere e Geothermal Resources Radiation Engines (e.g., PV) Kinetic Engines (e.g., Wind) Heat Engines (e.g., Geo) Gravity Engines (e.g., Hydro) Nuclear Engines (e.g., BWR) Chemical Engines (e.g., ICE)

6 Chemical Exergy of Some Fuels Fuel Chemical Chem. Exergy ΔH Reaction* ΔG Reaction* ΔS Reaction* Exergy Species+ Formula MJ per fuel MJ per fuel MJ per fuel kj/k per fuel to LHV kmol kg kmol kg kmol kg kmol kg Ratio Methane CH Methanol CH3OH Carbon Monoxide CO Acetylene C2H Ethylene C2H Ethane C2H Ethanol C2H5OH Propylene C3H Propane C3H Butadiene C4H i-butene C4H i-butane C4H n-butane C4H n-pentane C5H i-pentane C5H Benzene C6H n-heptane C7H i-octane C8H n-octane C8H Jet-A C12H Hydrogen H All species taken as ideal gases. Environment taken as: 25 C, 1 bar, 363 ppm CO 2, 2% H 2 O, 20.48% O 2, balance N 2. *Reaction with stoichiometric air at 25 C, 1 bar. All products present as ideal gases, including water. Fuel Conversion Efficiency potential (maximum first-law efficiency based on LHV) of most fuels is ~100%.

7 Classification & Architecture Classification: Chemical Engines (1) Restrained Reaction Unrestrained Reaction (2) Electrical Work (e.g., SOFC) Mechanical Work (e.g., None) Electrical Work (e.g., MHD) Mechanical Work (e.g., GT) (3) Architecture: the set of components & connections, and the corresponding set of thermodynamic idealizations & device limitations that constitute a particular engine.

8 Two Approaches to Reaction Unrestrained Reactants are initially internally restrained, i.e., frozen in chemical non-equilibrium (e.g. combustion, fuel reforming). Internal restraint is released, allowing reaction to proceed. Reaction stops when equilibrium is achieved or kinetics are so slow as to be negligible (frozen again). Inherently irreversible. Restrained Reactants are initially externally restrained, i.e., in chemical equilibrium (e.g. electrochemistry, solution chemistry). External restraints are changed, allowing reaction to proceed. Never stops; always dynamically balanced. Reversible only in the limit of infinitesimal rate and constrained chemical pathway (chemical reversibility).

Irreversibility reduced via facile kinetics (reaction and transport) Efficiency improves with load

9 Restrained vs. Unrestrained Architectures Restrained (SOFC) Unrestrained (DI Diesel*) Efficiency declines with load Irreversibility reduced via facile kinetics (reaction and transport) Efficiency improves with load Irreversibility reduced by reaction at extreme states * After Primus, et al. Proceedings of International Symposium on Diagnostics and Modeling of Combustion in Reciprocating Engines, (1985) p

10 Entropy Generation with Unrestrained Reaction Stoichiometric propane/air mixture modeled as ideal gases. Includes the effects of variable specific heats, reaction, & dissociation. Four ways to transfer energy

11 Efficiency Achievable with Simple- Cycle Extreme Compression First-Law Efficiency (%) SI 20 First Law (per LHV) 70-80% First Law Fuel Exergy/LHV CI Compression Ratio Stoichiometric propane/air

Stoichiometric propane/air mixture modeled as ideal gases.

12 Extreme-Compression Post-Combustion Conditions 3300K! 1000 bar! Must be fast! Must be balanced! Stoichiometric propane/air mixture modeled as ideal gases. Includes the effects of variable specific heats, reaction, & dissociation.

13 Free-Piston Engines Example: Junkers Compressor M. Nakahara and H. Kohama, Junkers High Pressure Air Compressor-A Case of Technology Transfer through the Imperial Japanese Navy, in The 1st international conference on business and technology transfer, 2004.

14 Van Blarigan/Aichlmayr Linear Alternator Concept

15 Experimental Apparatus

16 16 Operating Space

17 Combustion Visualization CR = 30:1 CR = 100:1 1 ms injection duration, finishing at TDC

18 Combustion Data at CR = 70 Pressure (bar) Air-only Combustion Isentrope 10 0 φ = Volume (V/V 0 )

19 First-Law Efficiency: Initial Results First law (per LHV), φ = % first law Combustion data Efficiency (%) Compression Ratio

20 First-Law Efficiency: Initial Results Efficiency (%) First law (per LHV), φ = % first law Combustion data Theoretical efficiency with air losses Losses in air experiments Additional losses due to combustion Compression Ratio

21 First-Law Efficiency: Initial Results Efficiency (%) First law (per LHV), φ = % first law Combustion data Theoretical efficiency with air losses Low blowby 53%, 20 C walls Confident we can demonstrate 60% indicated Speculate 70% is achievable regeneratively Losses in air experiments Additional losses due to combustion Compression Ratio

22 Simple-Cycle Steady Flow What is the optimal action to be taken (transfer or transformation) at each step in order to minimize S gen?

23 Challenges w/steady Flow Irreversibility Chemical reaction Reactant mixing Rejection of non-equilibrium exhaust Polytropic compression and expansion (Friction, viscous dissipation) Material Limitations Temperature limit Pressure limit

24 Polytropic work a a' b b' c' Equilibrium Attractor Trajectory h-h i (MJ/kg mix ) Reversible Work Cycle i P i f ' -0.8 f Premixed Reactants, GRI 3.0 Polytropic efficiency Net Work Out c s-s i (kj/kg mix K) Irreversible Work Cycle P i

25 Optimal Pressure Ratio Entropy Generation (kj/kg mix K) Nonpremixed reactants Polytropic efficiency Combustion Fluid Friction Total Pressure Ratio P*

26 Effect of Polytropic Efficiency Pressure Limit P * (bar) Nonpremixed reactants Polytropic Efficiency η Maximum Temperature T max (K) In the absence of material limitations, the pressure ratio of today s engines is well below optimum.

27 Temperature Limit h-h i (MJ/kg mix ) i Temperature Limit : 1650K Brayton (18.5:1) CT (40:1) CT(160:1) Attractor (160:1) Attractor (40:1) s-s i (kj/kg mix K) c Decreasing S gen f Increasing work-output Nonpremixed reactants Polytropic efficiency A temperature-limited, extreme-state cycle gives the optimal simple-cycle GT architecture.

28 T-Limited Simple-Cycle GT

29 Take-Home Messages (1 of 2) Despite three centuries of effort, engine efficiency remains well below theoretical limits (resource exergy) often by more than a factor of two. Misconceptions about what ultimately limits engine efficiency (e.g., Carnot) are sometimes to blame. Working in the space between the exergy limit and real engines, we have found the ideas of classification and architecture to be useful. Our approach is to use the principles of optimal control to identify the most efficient architecture for any given set of allowable devices, resources, and environment. For chemical engines, a key to understanding is whether the architecture uses restrained or unrestrained reaction.

30 Take-Home Messages (2 of 2) Irreversibility in restrained reaction engines can be reduced by improving kinetics. To date, the only examples of restrained reaction engines are electrochemical (i.e., fuel cells). Irreversibility in unrestrained reaction engines can be reduced by reaction at states of high energy density (extreme-states principle). For simple-cycle engines, we believe that architectures capable of delivering 60% first-law efficiency are possible. For regenerative engines, we believe a systematic approach to identifying optimal architectures can be developed. We speculate that such engines are capable of 70% first-law efficiency. For combined-cycle engines, we speculate that a systematic approach is again possible and can lead to the development of engines with first-law efficiencies in excess of 80%.

32 Conversion Efficiency of Engines First-Law Efficiency (%) % Time (Years A.D.) Savery, Newcomen (<0.5%) Watt/Boulton Steam Engines Post-Watt Steam Engines Lenoir, Hugon Coal-Gas Engines Otto/Langen Coal-Gas Engines Atkinson, Tangye Coal-Gas Engines Banki Spirits Engine Priestman's Oil Engine Diesel's Oil Engines Automotive SI Engines Truck Diesel Engines Large Bore DI Diesels Steam Turbines Gas Turbine/Steam Turbine Polymer Electrolyte Membrane FC Phosphoric Acid Fuel Cells SOFC/Gas Turbine After three centuries of development, combined-cycle efficiency just exceeds 50%, simple-cycle remains below.

33 Work Extraction During Combustion Otto Cycle Processes Detailed Chemical Kinetics Slider-Crank Piston Profile 2nd Law All complete rxn solutions resulted in increased irreversibility! Conclusions invariant with changes in fuel (methane, methanol, propane), rate, piston profile, etc.

34 Optimal Control Problems

min The key is to manage the location of the u-v attractor

35 Optimal Piston Motion Linear system wrt control input q bang-bang control S gen is minimized when reactions occur at V min The key is to manage the location of the u-v attractor Strategy has no explicit dependence on kinetics (Pontryagin Max. Principle)

states 2nd Law The key to reducing irreversibility in

36 Work Extraction During Combustion Otto Cycle Processes Detailed Chemical Kinetics Slider-Crank Piston Profile u-v attractor states 2nd Law The key to reducing irreversibility in unrestrained reaction (combustion) is to drive the reactants to the highest u state.

cylinder ) O ( aq, reactor ) 2 2 H O( cylinder ) H O( reactor ) 2 2 Reaction

37 Restrained Reaction w/out Electrochemistry? Transfer Restraint Requires: μ μ μ H ( g, cylinder) H ( aq, reactor) 2 2 O ( g, cylinder ) O ( aq, reactor ) 2 2 H O( cylinder ) H O( reactor ) 2 2 Reaction Restraint Requires: μ HOreactor ( ) 2 μ = μ = μ = μ = + μ 1 H ( aq, reactor) 2 O ( aq, reactor) 2 2 Gibbs-Duhem Relation: dμ = sdt + vdp

38 Chemical Equilibrium, 300K H 2 + ½ O 2 H 2 (kj/mol H2 ) H 2 O O 2 (kj/mol O2 ) μ (kj/mol) H 2 + ½ O 2 (kj/mol Reaction ) Incompressible H 2 O(l) (kj/mol H2O ) P/P 0 P o = 1 bar

39 Pressure Retarded Osmosis Sat. NaCl: π ο Dead Sea: π o = 380 atm > 500 atm

40 A Restrained Chemical Engine Nafion Protons Hydrogen Heat Electrons M Work Anode: H 2H + 2e 2 μ = 2 % μ + 2 % μ + gas, a nafion, a anode H + 2 H e Nafion connected, open circuit to motor: % μ = % μ nafion, anode nafion, cathode H + + H μ + 2% μ + 0.5μ = 2 % μ + μ gas, a cathode gas, c anode gas, c H2 e O2 e H2O gas, a gas, c gas, c anode cathode μh O μ 2 H2O = 2 % μ % μ e e cathode anode 2F( φ φ ) Δ rgoverall= A ( Chemical Affinity) ( ) Humid Air Water and Depleted Air + Cathode: 2H + 2e + 0.5O H O % μ + 2 % μ + 0.5μ = μ nafion, c cathode gas, c gas, c + H e O2 H2O

41 Electrochemical Cell w/losses Reactant Preparation Triple Junc. Triple Junc. Triple Junc. Triple Junc. Cathode Reactants Reactant Crossover Electrolyte Membrane Anode Contact Cathode Contact Reactants Anode Contact Cathode Contact Bipolar Plates - + Electrolyte Membrane Reactant Crossover Anode Reactants Cathode Reactant Channel Anode Reactant Channel Transport Losses Activation Losses

42 Restrained/Unrestrained Expansion Wout = Wmax Wout < Wmax 42

43 Restrained/Unrestrained Reaction Restrained Reaction Unrestrained Reaction Wout = Wmax Wout < Wmax 43

Implementing Restrained Reaction W = X = T S lost

must be slow compared to the internal relaxation time

44 Implementing Restrained Reaction W = X = T S lost destroyed 0 gen ξ2 ξ2 i i A i Sgen = dξ = d T T ξ ξ 1 1 νμ ξ dxpiston dξ A if is small for a given dξ T S gen will be small The rate of change of the restraint must be slow compared to the internal relaxation time of the resource in order to be fully restrained (reversible). 44

45 Requirements for Restrained Chemical Engines The reaction pathway must be open no constraints or additional restraints on the reaction reaction affinity equals zero before work is produced Work must couple to the chemical reaction temperature pressure composition electrical potential determining parameters for electrochemical potential of reacting species 45

47 Minimum S gen Solution 1. Bang-bang solution, switching at P = P eq 2. Optimal over set of all possible piston motions

48 Constant-UV Equilibrium Attractor U V 1 +dv U 1 +du S eq,1 +ds eq V 1 U 1 S eq,1 V S 1 ds eq S P < P eq compression Control: dv < 0 1st. law: du = P dv T eq ds eq = du eq + P eq dv eq = du + P eq dv T eq ds eq < 0 = ( P + P eq ) dv Extracting energy during combustion can only decrease efficiency. We are going the wrong way!

50 Exergy Destruction via Reaction Stoichiometric propane/air mixture modeled as ideal gases. Includes the effects of variable specific heats, reaction, & dissociation.

51 Effect of Compression Stoichiometric propane/air mixture modeled as ideal gases. Includes the effects of variable specific heats, reaction, & dissociation.

52 Effect of Heating & Cooling Stoichiometric propane/air mixture modeled as ideal gases. Includes the effects of variable specific heats, reaction, & dissociation.

54 Extreme Compression Concept High compression ratio, ~100:1 Multiple pistons (balanced forces, ~unity aspect ratio) High speeds, M~0.3 (reduced time for heat transfer) - air at 300 K, speed of sound ~ 350 m/s 100 m/s - for reference: 3000 RPM and 90 mm stroke 9 m/s

55 Combustor Design Material stress strategy: Use pressure profile to our advantage Exhaust valve Combustor bore below 100 bar after 200 mm Injectors (5) Combustor tested to 2000 bar 55

56 Combustor Design and Injection 5 Bosch, diesel injectors (1500 bar) with customized nozzles 56

57 Volume Measurement Piston Design Steel ring for VR sensors Optical bar code. Outer diameter reflects light, inner diameter does not. Graphite bearings Copper ring for thermally protecting the rings Graphite sealing ring 57

58 Air Compression: Initial Findings g piston 350 g piston MPS = 90 m/s Volume (V/V 0 ) MPS = 60 m/s CR = Time (ms) CR = g piston g piston Pressure (bar) Time (ms) Pressure (bar) Time (ms)

59 Simulating Acoustic Waves Method of Characteristics simulations show that high piston accelerations cause acoustic waves Method of Characteristics Experimental Shock forms due to lack of dissipation in model. Pressure (bar) Time (ms)

60 Air Data 500 CR = 99 Isentrope Pressure (bar) Compression 100 Expansion Volume (V/V 0 )

61 Total Losses Over an Air Cycle Losses consist of heat and mass transfer (~50:50). Percentage work lost per LHV Percentage isentropic pressure achieved W net /LHV (%) P peak /P isentropic (%) bar Compression Ratio Compression Ratio 61

62 Critical Questions Critical Question Material stress Wall temperatures and heat transfer NOx Seal survivability Sealing ability Ignition phasing Combustion control Extreme Compression Apparatus Extreme Compression Engine?? solutions available and understood more research required, but no obvious barriers high priority for research

64 Aquifer Sequestration as Commonly Envisioned 64

65 Potential Problems with Aquifer Sequestration 65

66 Pre-equilibrated Aquifer Sequestration 66

67 Storage Security Adapted from IPCC Special Report on Carbon Dioxide Capture and Storage 2005, p. 208

68 Conceptual Plant Schematic Pre-equilibrated Storage Zero Emissions to Atm. 42.1% Efficiency (LHV, 1600 K, 38 C) Indirectly Fired Combined- Cycle Engine

69 Modeled System

70 Thermodynamic Analysis Combustor Outlet T=1600 K Condenser T=38 C Component Power (MW) Brayton Cycle Compressor Turbine Net Rankine Cycle Condensate Pump Feed Pump Turbine Net ASU Water Pumps Overall Plant Heat Rate (LHV basis) Overall Efficiency 42.1%

71 Experimental Schematic

72 Experimental Combustor

74 Mixing Entropy Generation Unmixed NG/air at the same temperature and pressure. (~2% of comb. S gen )

75 Next: Regeneration Work, Heat, and Matter with Closure Constraints and Environmental Interactions What is the optimal action to be taken (transfer or transformation) at each step in order to minimize S gen?

76 Thermodynamic State-Space Natural Gas Air, Equivalence Ratio 0.5

78 Work-Specific Exergy Consumption (MJ-Exergy/MJ-Work) n tio p ) m rk u o s n o -W J C y /M y rg x e rg e x E t-e c ific u p e p -In J -S rk (M o W 4?? 90% Seq.? 50% Seq.?? No Sequestration 25% n e g SCWC, ro SI 3.5 y doxy OxyPC-SFA GE GCEP SI H SI Super-SFA GE3Super Shell SIH SIH Shell-SFA Super OxyPC SIMeOH SI-H SI 3 IGCC MCGC-RA MCGC-RB Sub Sub. IGMC HFGT HCCI 6FB SIH GE2 Sub. GE Super Super SIMeOH Super-SFA GE1 GE3 Super Mat ADGT CLSC IGCL ADGT LM6000D Shell LM6000 Elsam Shell-SFA Egas Ultra Shell 2.5 CES MAT CI CI SIH2 PEM MC-Ultra-Coal-NG DICI OxyNG MCUltra PCCI P WC-G ICGT Ultra EU IGCC E ATR-GTCC Z SIH FGC1 CLCC ATRCC LMS100 EU DOE AZEP WC-K A WC-G ATR MCNG-RA DOE FGC2 GTCC-SFA AZEP CLCC MCCC Graz RGT SOGT-Z LBCI 2 GTCC STIG LBCI PEM MCNG-RB SOGT-BA EUPC Graz GTCC-SFA DOEPC PEM GTCC GTCC-H PEM SOGT-M SOGT-CA SOGT-BB SOGT-RA SOGT-CB SOGT-C SOGT-RB SOGT-K SOGT 1.5 SOGT-BC SOGT SOGT-RC EC, GCEP 29% 33% 40% 50% 67% 1 100% Work-Specific Carbon Emission to Atmosphere (kg-c/mj-work) Work-Specific Carbon Emission to Atmosphere (kg-c/mj-work) c y n ie fic E tic e rg e x E ) y rg x e t-e u p -In J /M rk o -W J (M

Sankaran Ramakrishnan, and Adam Simpson. Department of Mechanical Engineering

Understanding: di The Path to High- Efficiency Chemical Engines Chris F. Edwards Kwee Yan Teh, Shannon Miller, Matthew Svrcek, Sankaran Ramakrishnan, and Adam Simpson Advanced d Energy Systems Laboratory