Microscale reacting flows and power generation

1
Microscale reacting flows and power generation

• Micropower generation what and why (Lecture 4)
• Microcombustion science (Lectures 4 - 5)
• Scaling considerations - flame quenching,
  friction, speed of sound,
• Flameless catalytic combustion
• Effects of heat recirculation
• Devices (Lecture 6)
• Thermoelectrics
• Fuel cells
• Microscale internal combustion engines
• Microscale propulsion
• Gas turbine
• Thermal transpiration
• Rocket
• Reference Fernandez-Pello, A. C., Micropower
  Generation Using Combustion Issues and
  Approaches, Proc. Combust. Inst., Vol. 29, pp.
  883 - 899 (2002)

2
Thermoelectric power generation

• Same principal as thermocouple, material
  optimized for power generation
• In heat-recirculating combustor, can imbed in
  wall between hot (outgoing product) and cold
  (incoming reactant) streams

Overall configuration - integrated heat
recirculating combustor power generation - wall
is electrical conductor
Typical thermoelectric configuration -
alternating n- and p-type elements
3
Thermoelectrics

• Widely used in deep space missions, some
  commercial applications, e.g. small refrigerators
• TE efficiency typically 15 of Carnot with same
  ?T - not bad - but how to get large ?T?
• Quantum well materials potentially much higher
  efficiency (Ghamaty Elsner, 2003)

4
Thermoelectrics

• Thermal efficiency (?) depends on figure of merit
  (ZT), which depends on material T

5
Thermoelectrics

• Figure of merit ZT S2T/?k
• S Seebeck coefficient (Volts/K) T
  temperature (K) ? electrical resistivity
  (ohm-m) k thermal conductivity (W/mK)
• Why ZT?
• Electrical power V2/R S2(?T)2/(?(?x)/A)
• Thermal power kA(?T)/?x
• ? (Electrical power)/(thermal power) S2?T/?k
  (ZT)(?T/T)
• but this is gross power need to consider
  source resistance load resistance (optimal
  match resistances) and i2R heating of TE
• Note as ZTa ? 8, ? ? 1 - Tc/Th ?Carnot

6
Thermoelectric microgenerator problem

• TE wall material thermal conductivity k 1
  W/mC
• Gas k 0.025 - 0.1 W/mC
• ? Thermal resistance between gas TE wall
• Rgas 1/hA 7.5d/kairA gtgt resistance
  across TE
• RTE ?X/kTEA
• ? Most ?T between gas TE wall, not across TE
• No power generation!
• Note with surface catalytic combustion on hot
  side of TE, RTE can be much lower - but what
  about cold side? (Could flow cold liquid fuel
  through channels in contact with cold-side of TE,
  but what about pumping power?)

7
Thermoelectric microgenerator problem

• Macroscale devices - can use turbulence Nu
  Re0.8 Re1
• Heat transfer rate hA(?T) (Nu1)k/dd2 ?T
  Re1kd?T Ud2k?T/?
• Heat transfer per unit mass flow rate
  (Ud2k?T/?)/(?Ud2) k?T/?? constant
  (independent of U)
• Pumping power (?P)Ud2 (f?U2)Ud2 for
  turbulent flow f constant, pumping power per
  unit mass flow U2
• Heat transfer / Pumping power k?T/??U2 1/U2 -
  low U more energy efficient (but less space
  efficient) (also too low U, no turbulence!)
• Microscale devices - laminar flow
• Heat transfer rate hA(?T) (Nu0)k/dd2?T
  kd?T
• Heat transfer per unit mass flow rate
  (kd?T)/(?Ud2) k?T/?Ud
• Pumping power (?P)Ud2 (f?U2)Ud2, f 1/Re
  ?/Ud, pumping power per unit mass flow U?/d
• Heat transfer / Pumping power k?T/??U2 1/U2
  (again!) - low U more energy efficient (and no
  effect on space efficiency)
• For compactness, want high heat transfer per unit
  mass flow rate 1/U (laminar) ? low U (low Re or
  M), but then streamwise wall heat transfer
  becomes important!
• Either case, can use fins to improve space
  efficiency, but heat transfer pressure drop
  fin area, no benefit in Heat transfer / Pumping
  power
• Microfire - need dirty tricks !
• Special fin designs

8
Dirty tricks

• Integrated TE wall T-fin design greatly reduces
  Rgas/RTE - without massive pressure drops due to
  aggressive fins in flow channel
• Metal fins (blue) have high thermal conductivity
  - act as thermal short-circuit air acts as
  thermal open-circuit (insulator)
• Elongating base of T-fin and TE walls (increasing
  Lfin LTE) reduces Rgas/RTE
• Laminar flow Rgas 1/hA 7.5 dgas/kairLfinWfin
• RTE ?X/kTEA LTE/kTEdTEWTE
• Rgas/RTE 7.5 (kTE/kair)(dgasdTE/Lfin2)
• note Rgas/RTE 1/Lfin2 !!!
• US Patent No. 6,613,972 (9/2/2003)

9
Thermoelectric generator design

• To maximize power from a TE module of fixed
  cross-section area A, there is an optimum
  thickness ?x
• Too little ?x, total heat transfer is large but
  portion of temperature drop across TE module is
  small
• Too high ?x, temperature drop across TE module is
  large but total heat transfer is too small
• Hot-side gas temperature TH, cold-side TC
• Hot side of TE element TTE,H, cold side TTE,C
• Hot-side thermal resistance RH 1/hHA, cold-side
  RC 1/hCA
• Thermoelectric thermal resistance RTE ?x/kTEA
• As with resistors in series (Temperature ?
  Voltage)
• Note that the areas cancel out of Rs in each
  term

10
Thermoelectric generator design

• Determine the ?x that maximizes power for given
  hH, hC and kTE
• Pick a value of ?x
• Compute TH,TE and TC,TE
• Compute the efficiency
• Compute the power generation
• Efficiency (heat transfer)
• ? (TH - TC)/(RH RTE RC)
• Adjust ?x until power is maximum

11
Thermoelectric generator design

• Small ?X - very little ?T across TE element - too
  little thermal resistance
• Large ?X - too much thermal resistance, so
  thermal power is low, BUT most of ?T is across TE
  element, so efficiency is good
• Power peaks at intermediate ?X

TC 300K, TH 500K hC hH 1000 W/m2K kTE 2
W/mK, ZTa 1
12
Thermoelectrics - Princeton - Dryer et al.

• Note flat Swiss roll, (losses in 3rd dimension),
  large fins to minimize thermal resistance

13
Thermoelectrics - Princeton - Vican et al. (2002)

• Heat loss ?T
• Transport limited Heat generation H2 mass flow
• Steady-state Heat loss heat generation
• ?T H2 mass flow
• Low mass flow low T, kinetically limited, ?T
  decreases faster as mass flow decreases
• Peak efficiency (30 mW)/(6.8 W) 0.44

14
Thermoelectrics - Michigan - Zhang et. al. (2001)

• Minimal thermal resistance between hot side of
  TEs and gas due to catalytic combustion, but what
  about resistance between cold side of TEs and
  ambient air?

15
Fuel cells (S. Prakash, USC Chemistry Dept.)
?GT ?H - T?S (T constant)
16
Fuel cells (S. Prakash, USC Chemistry Dept.)
17
Micro H2 fuel cells - CWRU group

• Reference Wainright et al. (2003)
• Hydrogen-air fuel cell using Proton Exchange
  Membrane (Nafion)
• Simple, can use hydrides H2O Pt catalyst at
  room temperature as H2 source
• Low conductivity at low relative humidity (RH) -
  need H2O to conduct protons (H) through PEM
• PEM swelling at high RH
• Have to ensure no leakage of H2 or O2 across
  membrane

18
H2 PEM fuel cells

• Up to 5 mW/cm2 demonstrated
• Potential somewhat low - 1.23 V ideal open
  circuit potential, decreases as I increases
  (obviously) due to internal resistance of
  membrane
• See Wainright et al. (2003)

19
Honeywell fuel cell balance of plant

• Uses CWRU fuel cells with LiAlH4 H2O to make H2
• Self-regulating water valve controlled by H2
  pressure
• Biggest drawback no burst of power mode like
  battery

20
Honeywell fuel cell balance of plant

• Demonstrated performance for complete system
• 0.95 watt-hours per gram (4x better than CR123A
  battery)
• 0.90 watt-hours per cm3 (1.7x better than CR123A)

21
Hydride storage revisited
22
Methanol fuel cells

• Methanol good feedstock for fuel cells
• Anode CH3OH H2O ? 6e- CO2 6H
• Cathode 1.5O2 3H 6e- ? 3OH-
• 3 OH- 3 H ? 3 H2O
• but need unobtainum membranes for good
  performance in Direct Methanol Fuel Cells
• Nafion Proton Exchange Membrane allows
  substantial CH3OH cross-over from anode to
  cathode
• CO poisoning a problem - preferentially adsorbed
  on anode catalyst (Pt-Ru or Pt-Mo) compared to H2
  even at 10 ppm level which prevents further H2
  electrochemistry
• USC (Loker Hydrocarbon Institute, Prakash et al.)
  is a major player in DMFCs - improved PEM and
  catalyst materials
• Can reform methanol to H2 and CO2 via CH3OH H2O
  heat ? 3H2 CO2 (or other overall reactions)
  then use H2
• but need thermochemical plant for reforming
• Lower energy efficiency (not utilizing C atoms)
• Must remove CO to avoid poisoning Pt

23
Methanol fuel cell
24
Silicon methanol fuel cell - UCLA/PSU group

• Lu et al. (2003)

25
Devices - methanol reformers - Battelle/PNNL

• Holladay et al, 2002

26
Methanol fuel cells (using reformer)
27
Formic acid (HCOOH) fuel cells - UIUC

• Low energy density (5.53 x 106 J/kg, 8.4x less
  than hydrocarbons), but good electrochemistry
• In water HCOOH ? H HCOO-
• On Pt or other catalysts (anode side) HCOO- ?
  CO2 2e- H
• Cathode side 4 H 4 e- O2 ? 2 H2O
• Less tendency for cross-over than methanol fuel
  cells

28
Formic acid fuel cell (lecture 4)

• Zhu et al. (2004) Ha et al. (2004)

29
Single chamber SCFC chemistry

• Solid Oxide Fuel Cells (SOFCs)
• Use hydrocarbon fuels directly
• Fuel flexible - methane, propane, butane, octane,
  
• Unlike PEM-type fuel cells, loves CO (but still
  poisoned by sulfur)
• but need high T (but no need for T gradient)
• Usually use oxygen-ion conducting membranes
• Anode
• CH4 .5 O2 ? CO 2H2 (may be done in separate
  reformer upstream of fuel cells)
• H2 O ? H2O 2e-
• CO O ? CO2 2e-
• Cathode
• .5 O2 2e- ? O2-
• Also proton-conducting membranes
• - different reactions and materials
• - fewer electrons per fuel molecule
• Anode
• CH4 .5 O2 ? CO 2H2
• H2O CO ? CO2 H2
• H2 ? 2H 2e-

30
Power generation - SOFC in a Swiss roll

• Collaboration with Prof. Sossina Haile, CalTech
• SOFC with Swiss roll for thermal management
• Catalytic after-burner with secondary air to
  oxidize rich products
• Single chamber fuel cell
• No seals required
• Reduced temp. (400-600ºC)
• Minimize thermal stress
• NO REFORMING required, but eed rich mixtures for
  in situ reforming to CO H2
• Patent pending (filed 6/23/04)

31
Single Chamber Solid Oxide Fuel Cells
Single chamber SOFC
conventional SOFC
H2O CO2
CxHy O2
O2


anode
cathode
.5 O2 2e- ? O
CH4 4O ? CO2 2H2O 8e-
seals
O
e-
electrolyte
e-

CH4 .5 O2 ? CO 2H2 H2 O ? H2O 2e- CO
O ? CO2 2e-
.5 O2 2e- ? O

• Introduced by Hibino et al. Science (2000)
• Fuel oxidant mixed - no sealing issues!
• Reforming done directly on anode
• Less coking problems
• Excellent anode cathode catalyst selectivity
  essential in SCFC

32
SCFC construction fuels

• Three material choices required Hibino et al
  (2000b)
• Electrolyte (ion-carrying material)
  La0.9Sr0.1Ga0.8Mg0.2O3 (LSGM), Ce0.8Sm0.2O1.9
  (SDC) yttria-stabilized zirconia (YSZ)
• Cathode Sm0.5Sr0.5CoO3 (SDC)
• Anode Ni 10 wt SDC
• Hibino et al. only very rich mixtures tested
  (provides good thermochemical reforming), not
  practical
• Performance sensitive to fuel type - desired
  electrochemistry same for all but reforming
  characteristics vary widely

33
SCFC performance

• Results sensitive to temperature
• Low T electrolyte conductivity low, reforming
  reactions slow
• High T Complete oxidation instead reforming
  (partial oxidation)
• Hibino et al (2000b)

34
Single-Chamber Fuel Cells (Caltech)

• Both anode-supported (Caltech) cathode
  supported (LBL) fuel cells examined
  anode-supported somewhat better, probably due to
  increased area for reforming

Sinter, 1350oC 5h
Anode supported
Dual dry press
SDC
NiO SDC
NiOSDC
600oC 5h, 15H2
Spray cathode
Calcine, 950oC 5h, inert gas
Porous anode
35
Worlds smallest self-sustaining SOFC (?)
1.3 cm
7 cm
0.71 cm2
 
36
Single Chamber Fuel Cell in Swiss roll

• Maximum power density 375 mW/cm2 at T 540C
  demonstrated with direct utilization of
  hydrocarbon fuel - much higher than PEM fuel
  cells using methanol or formic acid

 
37
Effect of cell temperature and O2fuel ratio

• Much lower T than conventional SOFC significant
  power even at 400C
• Performance not to sensitive to temperature -
  range of T within 20 of max. power  50C
• Performance sensitive to O2fuel ratio - best
  results at lower O2fuel ratio (closer to
  stoichiometric but still fuel-rich)
• Butane (not shown) similar performance

 
38
SCFC Operation on Methane

• Ni SDC SDC (20 mm) SDC Ba0.5Sr0.5Co0.8Fe0.
  2 O3 (BSCF)
• Haile et al., Nature, Sept. 9, 2004
• Monotonic increase in power output with
  temperature
• Higher power outputs than with propane (less fuel
  decomposition at cathode, higher Octane number)

730 mW/cm2
39
Higher (liquid) hydrocarbons

• Iso-octane (2, 2, 4 trimethylpentane) used as a
  surrogate for various hydrocarbon fuels including
  gasoline, diesel JP-8
• 1.5 chamber fuel cell
• Cathode Ni-SDC, reactant air
• Anode LSCF-GDC, reactant fuel-rich (7
  iso-octane in air) mixture
• Electrolyte SDC
• Enabling technology special catalyst layer on
  anode (Zhan and Barnett, 2005)

40
Iso-octane / air SOFC

• Power density 550 mW/cm2 at 600C
• Power density 250 mW/cm2 at 450C (temperature
  limit for polymer Swiss rolls)
• Iso-octane power comparable to hydrogen
• Cell stable over 60 hr test, no coking observed
• Needs to be tested in single-chamber cells
• Results should transfer well to other
  hydrocarbons

41
Iso-octane / air SOFC

• Catalyst layer greatly increases longevity

Time (hours)
42
Automotive gasoline / air SOFC

• Catalyst/Ni-YSZ/YSZ/LSCF-GDC cell
• Power density 900 mW/cm2 at 800C
• No coking except at T lt 650C
• SEM-EDX measurements showed sulfur on the
  catalyst layer is responsible for degradation
  over time

43
Devices - IC engines

• Berkeley system - not just an engine!

44
Overall energy budget
45
Testing
46
Heat recirculation concept

• What problems might this have?

47
Wiring

• Another issue - what to do about power wires
  (high thermal as well as electrical conductivity)

48
Performance

• Engine speeds low for scale - similar to
  macroscale RX-7 Wankel engine
• Power shown is net combustion power, i.e.
  difference between measured power with without
  combustion (i.e. with without igniter)
  (igniter spark or glow plug)
• Intake at 1.5 atm - pumping gain
  (Pintake-Pexhaust)VN (0.5 atm)(348 mm3)(9000
  RPM) 2.64 watts - net indicated power 3.8 -
  2.6 1.2 Watts at 9000 RPM
• Brake power ( net indicated power - friction (
  1.5 watt at 9000 RPM) ) -0.3 Watt

Fu et al. (2001)
49
Free piston engine - Ga. Tech.

• Inductive coupling - can use closed-loop control
  of piston motion (power in power out) to
  optimize compression ratio, scavenging, )
• Not microscale (PDR definition)
• Typical speed 2000 RPM, 0.64 cm thick chamber,
  4.4 cm stroke ? Re 1250
• Thickness gt quenching distance
• Turbulence model used for simulation! (Menon et
  al. (2002))

50
Free piston engine - Ga. Tech.

• High compression - higher ideal thermodynamic
  efficiency but nowhere near isentropic
  compression
• Leaks
• Heat losses
• Lower compression ratio optimal
• Spark ignition - how much power?
• Honeywell / U. Minn. group free piston
  Homogeneous Charge Compression Ignition (HCCI)
  engine - no need for spark plug

51
Gas turbine (MIT)

• Friction heat losses
• Manufacturing tolerances
• Very high rotational speed ( 2 million RPM)
  needed for compression (speed of sound doesnt
  scale!)
• Design thrust 0.1 N
• References Epstein, Waitz

52
MIT gas turbine

• Pressure ratio (r) across compressor stagnation
  pressure ratio
• (1(?-1)/2 M2)?/(? -1) (M compressor blade
  Mach )
• Thermal efficiency (ideal Brayton cycle) 1 -
  1/r(?-1)/?
• Need M  1 for 15 efficiency of ideal device (no
  losses)!
• M 1, 4 mm rotor diameter 1.8 million RPM
• (Macroscale gas turbines use 10 compressor
  stages to get decent efficiency)

Calculations from Excel spreadsheet http//ronne
y.usc.edu /AME436S06 /AirCycles4Propulsion.xls
53
MIT gas turbine

• Not microscale according to PDRs definition
• Re (1000 cm/s)(0.13 cm)/(2.3 cm2/s) 56
  (qualifies as µscale) but quenching distances
  1.5 mm _at_ ? 0.4 to 0.7 mm _at_ ? 0.6 (doesnt
  qualify as µscale)
• Mixing time vs. chemical time - mixing time
  scales with combustor size but reaction time does
  not - need larger relative chamber size as scale
  decreases
• Heat transfer along casing rotor, from turbine
  to compressor

54
Combustion properties

• Dual combustion limits
• Various other design limitations - flashback
  (upstream burning, materials T limit)

55
Combustor designs
Basic
Heat recirculating
Dual-zone combustion
56
Gas turbine (MIT)

• Benefits of heat recirculation
• 3-stack no recirculation, 0.066 cm3 annular
  volume
• Static 6-layer design with recirculation,
  0.195 cm3

57
Gas turbine (MIT)

• Need to use hydrocarbons, not H2! Ethylene fuel
  tests
• Residence time at max T 0.4 kg/m3 190 mm3 /
  0.02 g/s 0.0038 s - longer than chemical time
  ?/SL2 0.001 s for stoichiometric hydrocarbons,
  BUT maximum combustor T 1600K, SL only 10
  cm/s, chemical time 0.016 s gt residence time!
• Moving toward catalytic combustion for
  hydrocarbons

58
Thermal transpiration for pumping or propulsion

• Q How to produce gas pressurization (or thrust)
  without mechanical compression (i.e. moving
  parts)?
• A Thermal transpiration - occurs in narrow
  channels or pores with applied temperature
  gradient when Knudsen number 1
• Kn ? mean free path ( 70 nm for air at STP) /
  channel or pore diameter (d)
• First studied by Reynolds (1879) using porous
  stucco plates

Reynolds (1879)
59
Aerogels for thermal transpiration

• Q How to reduce thermal power requirement for
  transpiration?
• A Vargo et al. (1999) aerogels - very low
  thermal conductivity
• Gold film electrical heater
• Behavior similar to theoretical prediction for
  straight tubes whose length (L) is 1/10 of
  aerogel thickness!
• Can stage pumps for higher compression ratios

60
Aerogels

• http//eande.lbl.gov/ECS/Aerogels/
• Nanoporous (typ. 10 nm) materials with low
  density (typ. 0.1 g/cm3)
• Typically made by supercritical (to avoid surface
  tension, which would destroy the structure)
  drying of silica gel using CO2 solvent
• Outstanding insulator (k lt kgas), outstanding for
  thermal transpiration (Kn 5 for air at STP),
  but generally fragile

61
Thermal transpiration

• Maximum pumping power Mach ?P occurs at Kn
  1
• ?P higher at low Kn (narrow channels) but flow
  speed very low
• (Results in right plot shown are at maximum
  pumping power (?P/?Pno flow 0.5))

62
Theoretical performance of aerogel jet engine

• Can use usual propulsion relations to predict
  performance based on Vargo et al. model of
  thermal transpiration in aerogels
• Non-dimensional TFSC of silica aerogel (k
  0.0171 W/mK) only 2x - 4x worse than theoretical
  performance predictions for commercial gas
  turbine engines

Except as noted Hydrocarbon-air, T1 300K, T2
600K, P1 1 atm, L 100 µm, d 100 nm
63
Fuel-driven jet engine with no moving parts

• Q How to provide thermal power without electric
  heating as in Vargo et al.?
• Answer catalytic combustion!
• Can combine with nanoporous bismuth
  (thermoelectric material, Dunn et al., 2000) for
  combined power generation propulsion

64
Transpiration (porous) membranes
Ahlstrom Glass Microfiber filter Purchased from
VWR Thermal conductivity .038 W/mK Effective
pore diameter 4.4 ?m Diameter 25 mm Thickness
.45 mm
Silica aerogel disks machined using traditional
techniques Thermal conductivity .017 W/mK Mean
Pore diameter 27 nm Diameter 21 mm Thickness 1
mm
65
Transpiration Membranes cont.
SEM image of typical aerogel structure created by
supercitical drying of silica gel using CO2
solvent
SEM image of commercially available glass-fiber
filter membrane
66
Feasibility testing

• Performance 50 of theoretical predictions in
  terms of both flow and pressure (even with thick
  membrane no sealing of sides)

67
Prototype transpiration pump (1-D)

• Incorporates basic principles as described in
  integration theory
• Cold side made of high conductivity material
  (Al) for uniform temperature gradient

68
Performance and limitations of (1-D)

• Best performance was 7 ml/min H2, 30 ml/min Air,
  at max. ?T of 100C (overall lean) would not work
  with hydrocarbons
• 50 of surface heat loss was to radiation and
  50 to buoyant convection
• To much heat loss area
• Solution replace exhaust plenum with opposing
  transpiration pumps

Convective energy loss
Radiative energy loss
Exhaust
Reactants
Intake plenum (warm)
Exhaust plenum (hot)
69
Better prototype (3-D)

• No exhaust plenum, so less heat loss
• All sides are transpiration pumps
• Exothermic reaction zone (catalytic) is confined
  to center of device

Exhaust or access port
Air/fuel
Air/fuel
Exothermic reaction
Thermal guard
Porous membrane
Air/fuel
70
Current prototype (3-D) cont.

• Fuel inlet connect via a T fitting to pump,
  with base of T open to ambient
• This ensures that device must draw in BOTH FUEL
  AND AIR (were not cheating by forcing fuel in
  and entraining air)
• If device was not drawing its own air, fuel would
  escape through base of T
• Of course, other 3 sides of tetrahedron also draw
  air

Fuel Inlet
Exposed to ambient
Intake plenum
71
Preliminary results

• It works!
• Maximum T 280C with 3.9 ml/min of propane (6.1
  watts if complete combustion)
• Performance not sensitive to orientation (not
  buoyancy driven)
• (Probably) worlds first self-pressurizing
  combustor / reactor with no moving parts
• Uses fuel (not electricity) as the energy
  feedstock
• New design cubic reactor

72
Proposed integration with SOFC
73
Really really preliminary ideal design

• Airbreathing, single stage, TL 300K, TH 600K,
  ?P 0.042 atm, 5.1 W thermal power
• Hydrocarbon fuel, thrust 3.1 mN, specific thrust
  0.36, ISP 2750 sec
• Theoretical performance
• Total weight 0.22 mN, Thrust/weight 14
• Hover time of vehicle (engine fuel Ti alloy
  fuel tank, no payload) 2 hours flight time
  (lifting body, L/D 5) 10 hours

74
MEMS rockets

• Berkeley maximum thrust gt 20 mN, gt 2 seconds
• Maximum Height 5 m (specific impulse
  (2h/g)1/2/ln(mi/mf) 1.5 s for mi/mf 2 (SSME
  370 s)
• MIT turbopump-fed rocket!

75
MEMS rockets

• http//www.aero.org/publications/crosslink/spring2
  005/headlines.html
• Addressable MEMS array of (?) sodium azide (NaN3)
  propellant (used in auto air bags)
• Test flown on Space Shuttle with tethered
  picosatellite

76
References

• Alan H. Epstein, Stuart A. Jacobson, Jon M.
  Protz, Luc G. Frechette, Shirtbutton-sized gas
  Turbines The Engineering Challenges of Micro
  High Speed Rotating Machinery, 8th International
  Symposium on Transport Phenomena and Dynamics of
  Rotating Machinery (ISROMAC-8), Honolulu, HI,
  March 2000
• K. Fu, A. Knobloch, F. Martinez, D.C. Walther, C.
  Fernandez-Pello, A.P. Pisano, D. Liepmann, K.
  Miyaska and K. Maruta, Design and Experimental
  Results of Small-Scale Rotary Engines, Proc.
  2001 International Mechanical Engineering
  Congress and Exposition (IMECE),
  IMECE2001/MEMS-23924, New York, November 11-16,
  2001.
• T. Hibino, A. Hashimoto, T. Inoue, J.-I. Tokuna,
  S.-I. Yoshida and M. Sano, A Low-Operating-Temper
  ature Solid Oxide Fuel Cell in Hydrocarbon-Air
  Mixtures, Science 288 (2000a) 2031-2033.
• T. Hibino, A. Hashimoto, T. Inoue, J.-I. Tokuna,
  S.-I. Yoshida and M. Sano, Single-Chamber Solid
  Oxide Fuel Cells at Intermediate Temperatures
  with Various Hydrocarbon-Air Mixtures, Journal
  of The Electrochemical Society, 147 (8) 2888-2892
  (2000b)
• J. D. Holladay, E. O. Jones, Ma Phelps, J. Hu
  (2002). Microfuel processor for use in a
  miniature power supply, Journal of Power Sources
  108, 2127.
• M. Kirtas, M. Disseau, D. Scarborough, J. Jagoda
  and S. Menon (2002). Combustion Dynamics in a
  High-Aspect Ratio Engine Proceedings of the
  Combustion Institute, Vol. 29, 917-923.
• M. Knudsen. Eine revision der gleichgewichtsbedin
  gung der gase. Thermische molekularströmung. Ann.
  Physik 31205229 (1910) Ibid., Thermischer
  molekulardruck der gase in röhren. Ann. Physik,
  3314351448 (1910).
• G. Q. Lu, C. Y. Wang, T. J. Yen and X. Zhang
  (2004). Development and characterization of a
  silicon-based micro direct methanol fuel cell,
  Electrochimica Acta, Volume 49, Pages 821-828.

77
References

• Mehra, A., Zhang, X., Ayon, A., Waitz, I., and
  Schmidt, M., Spadaccini, C., "A 6-Wafer
  Combustion System for a Silicon Micro Gas Turbine
  Engine," Journal of Microelectromechanical
  Systems, Volume 9, Number 4, December 2000,
  pp.517-527
• Shao Z. P., Haile S. M. (2004) A
  high-performance cathode for the next generation
  of solid-oxide fuel cells Nature 431170-173.
• C. M. Spadaccini, J. Lee, S. Lukachko, I. Waitz,
  A. Mehra, X. Zhang, "High Power Density Silicon
  Combustion Systems for Micro Gas Turbine
  Engines," GT-2002-30082, Proceedings of ASME
  Turbo Expo, Amsterdam, The Netherlands, June 2002
• S. E. Vargo, E. P. Muntz, G. R. Shiflett, W. C.
  Tang, Knudsen compressor as a micro- and
  macroscale vacuum pump without moving parts or
  fluids, Journal of Vacuum Science and Technology
  A, Vol. 17, p. 2308 (1999).
• J. S. Wainright, R.F. Savinell, C.C. Liu, M. Litt
  (2003). Microfabricated fuel cells,
  Electrochimica Acta 48, 2869-2877.
• Vican, J., Gajdeczko, B. F., Dryer, F. L.,
  Milius, D. L., Aksay, I. A., Yetter, R. A.,
  Development of a Microreactor as a Thermal
  Source for MEMS Power generation, Proceedings of
  the Combustion Institute, Vol. 29, pp. 909-916
  (2002).
• Zhang, C. Najafi, K., Bernal, L.P., Washabaugh,
  P.D. An Integrated Combustor-Thermoelectric
  Micro Power Generator 11th Int. Conference on
  Solid State Sensors Actuators, Munich, Germany,
  June 10-14, 2001.
• Zhan, Z., Barnett, S. A., An Octane-Fueled Solid
  Oxide Fuel Cell, Science 308, 844-847 (2005).