Microreaction Engineering:Is small really better?
Jan J. Lerou
Velocys, Inc., Plain City OH www.velocys.com
2
Overview
Background
Fundamentals
Reaction Applications
Scaleup Methodology
Summary
3
BackgroundHistory
•In public domain since 1995
Definition of microstructured reactors•3dimensional structures from submm to mm range•Main characteristic:
specific surface 10,000 –50,000 m2/m3
Potential benefits•Process intensification•Inherent process safety•Broader reaction conditions incl. explosion regimes•Distributed production•Faster development
4
Background (cont.)
Companies currently moving microchanneltechnology from R&D to commercialization:
•Degussa: running a demonstration project for theevaluation of microreaction technology or DEMiSTM forpropylene epoxidation with hydrogen peroxide
•Clariant: opened its Competence Centre for MicroreactorTechnology (C3MRT) to increase efficiency, improvesafety and reduce the costs of pharmaceutical synthesis Continuous pilot plan for the synthesis of azopigments
•Axiva developed a process for continuous polymerizationof acrylates (8 kg/h) using micro mixers
•Velocys…
5
~ 0.11 mm
Microchannels enablecompact unit operations
with high capacity perunit volume by reducing
transport distances
~ 10100 mm
Conventional
Characteristicdimension
Enabling Technology
Microchannel
6
Conventional Technology
Conventional Reactors§ Steam Methane
Reformer
§ 20 millionstandard cubicfeet/day
§ ~30m x ~30m x~30m
7
Velocys® Technology Reactors
q MicrochannelSteam MethaneReformer
q Same capacity
q 90% volumereduction
q ~25% reductionin overall plantcosts
8
Microchannel Hardware Performance
< 1 W/cm21250 W/cm2Boiling
< 1 W/cm2120 W/cm2ConvectiveConventionalMicrochannel
Intense Heat Transfer Increases Productivity
Fast Reactions Shrink Hardware Volume
110 seconds1100 msGasphaseConventionalMicrochannel
9
dkNuh ×
=Nu: Nusselt numberh: Heat transfer coefficientd: Hydraulic diameterk: Thermal conductivity
Velocys Heat Exchanger Conventional Heat Exchanger
High surface area/volume ratioHigh heat transfer per volume
Low surface area/volume ratio
Low heat transfer per volume
Heat Transfer in Microchannels
10
inlet outlet
Diffusion across laminar streamlines (100<Re<2000)
Mass Transfer in Microchannels
avgconvection vel
L=τ : Characteristic convection time
L: Flow lengthvel: average laminar velocity
Sh: Sherwood numberkA: Mass transfer coefficientd: Hydraulic diameterD: Diffusivityd
DShkA×
=: Characteristic diffusion time
d: Hydraulic diameterD: DiffusivityD
ddiff
2)2(=τ
11
Mass Transfer in Microchannels
Velocys technology enhances bothintraparticle and interparticle
mass transfer
Velocys technology enhances bothintraparticle and interparticle
mass transfer
~ 0.002 cm ~ 0.22 cm
vs.Velocys Conventional
~ 0.02 cm ~ 0.2 cm
flow
12
Low pressure drop
Laminar flow in microchannels•Orderly flow –less fluctuations
•Laminar Flow
Turbulent Flow
flowh VDfLP µ)(=
∆
Turbulent flow• Random flow –more fluctuations
•75.175.025.0)( VDf
LP
h ρµ=∆
flow
(Blasius friction factor)
13
0
0.5
1
1.5
2
0 10 20 30 40 50Flow (SLPM)
Pres
sure
Dro
p (p
si)
Experimental DPPredicted DP
Pressure Drop Case Study
48 channel deviceN2, ambientFlowtot = 40 SLPM40”x 0.035”x 0.160”Velocity = 4.2 m/sRe = 391DP = 1.47 psig (model)
DP = 1.51 psig(experimental)
14
Design for Reaction Applications
Design for Constraints•Pressure Drop•Temperature / Materials
Tailor Catalyst Form•Match kinetics, heat removal, and pressure drop
Select Thermal Management Method•Convective heat transfer•Phase change•Chemical reaction
15
Conventional Catalyst TechnologyPowders for stirred tank or fluidized bed applications
Extrudates for fixedbed applications
Monoliths for gas purification applications
16
Catalysts for Microchannel Reactors
Porous Metallic Felts
~0.050.1 cm
Microchannel Walls
Metallic Foams
Metallic Foils
Metallic Fins
17
Catalyst Selection Strategy forMicrochannel Reactors
kineticsslow fast
Pressure drop and dT allowable
high
low Wall coatWall coatFin coatFin coat
Flowbyengineeredstructures
Flowbyengineeredstructures
Flowthroughengineeredstructures
Flowthroughengineeredstructures
PackedchannelsPackedchannels
18
Endothermic Reaction Examples:Tailor flux and thermal profile
airfuel
reactantexhaustproduct
• Add heat by adjacent exothermic reaction• Add heat by convective heat transfer• Add heat by phase change
19
Thermal Performance:Endothermic Steam Reforming
SMR Response
0.0 0.2 0.4 0.6 0.8 1.0
Normalized Distance
Com
bust
ion
Wal
l Axi
al H
eat
Flux
(W/c
m^2
)
SMR
Axi
al G
as T
empe
ratu
re (C
)
Legend: Temperature Heat Flux
20
High Heat Transfer to AchieveTemperature Uniformity
Temperature profile downtypical FischerTropsch fixed
bed reactor∆ T 25 C
Temperature profile downfixed bed FischerTropsch
microchannel∆ T 2 C
21
Velocys Technology Applications
Reactions
Methane Reforming FischerTropsch
Oxidation Methanol
Dehydrogenation ReactiveDistillation
Separations
Thermal swing adsorption Distillation
Mixing
Emulsions
Heat Exchange
Compact Heat Exchangers LNG
22
Scaleup
q Manufacturable and Cost Effective Design
q Sufficient Flow Distribution
q Robust Operation
q Integration with Commercial Plants
23
Velocys Scaleup Methodology
FullscaleReactor
CELL
Cell•Internal channel dimensions same as
commercial chemical processor•Number of channels increase;
size of channels does not
MultiCell•Many channels•10100 lb/hr
FullScale•>1000 channels•10005000 lb/hr
FullScale Reactor is the basicbuilding block of a commercialplant
Gas flow
MULTI
CELL
24
Commercial Microchannel Reactor:Summary for Hydrogen Generation
Productivity•1MM SCFD H2 per reactor
Size•~0.6 m x 0.8 m x 0.6 m•2 tons•>5000 channels
Streams•Air ~70 ºC•Fuel ~70 ºC•Reactant ~200 ºC•Product ~300 ºC•Exhaust ~300 ºC
Emissions•NOx < 10 ppm•Meets California regulations Eighthscale DeviceEighthscale Device
25
Stacking
DiffusionBonding Machining
Shim
Manufacturable Design:Metal thinness creates microchannel dimension
Bonding creates hermetically sealed microchannelsBonding creates hermetically sealed microchannels
Featurecreation
Wall shim without features
26
Device Manufacturing
27
Diffusion Bonding Large Stacks
Protocol
•Time
•Temperature
•Pressure
28
Scaleup Considerations
q Manufacturable and Cost Effective Design
q Sufficient Flow Distribution
q Robust Operation
q Integration with Commercial Plants
29
Tailor Pressure Drop in Flow CircuitsTo achieve sufficient flow distribution
Inlet
1 2 3
max min1
max
100%m mQm
−= ×
2 3
30
Flow Distribution Validation
31
Channel Flow Distribution
Sufficient flow distribution measured in test deviceSufficient flow distribution measured in test device
Run 16: 214.0 SLPM of air
0.0E+00
1.0E05
2.0E05
3.0E05
4.0E05
5.0E05
6.0E05
7.0E05
8.0E05
9.0E05
1.0E04
0 12 24 36 48 60 72
Channel number
Mas
s flo
w ra
te (k
g/s)
Model
Experiment
32
Manifold Model vs. Experiment dP
0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
4.00
0 50 100 150 200 250 300Total air flow rate (SLPM)
Inle
t pre
ssur
e (p
sig)
Model
Experimental average
Predicted manifold pressure drop matches experimentsPredicted manifold pressure drop matches experiments
33
Scaleup Considerations
q Manufacturable and Cost Effective Design
q Sufficient Flow Distribution
q Robust Operation
q Integration with Commercial Plants
34
Robust Performance to UpsetsRecovery after combustion lost and catalyst in situ refurbishment
Conditions:3:1 S:C23 atm6 ms CT9000 scfd H2
Reactor performance
20%
0%
20%
40%
60%
80%
100%
0 10 20 30 40 50
Time on stream since refurbishment (hr)0
2
4
6
8
10
12
CH4 conversion (%)
Select ivit y t o CO (%)
SMR C balance (% change)
840 °C equilibrium conversion
830 °C equilibrium conversion
upset
SMR dP (psid, r ight axis)
upse t l ost c ombust i on
35
Scaleup Considerations
q Manufacturable and Cost Effective Design
q Sufficient Flow Distribution
q Robust Operation
q Integration with Commercial Plants
36
Microchannel Reactor Plant Interface
Lowcost standardized reactor assemblies:•5 to 30 fullscale reactors in each assembly•Shopbuilt fabrication reduces
construction costs and time•Minimizes utility runs to cut
installation costs•Modular for additions
of incremental capacity.
37
Summary
Microchannel technology’s advantages•Tailored thermal profiles•Optimized catalyst performance and form•Enhanced selectivity•Higher productivity•Operation near stoichiometric feed ratio
Model driven design and optimization process
Scaleup principles modeled and validated
38
Contact Information
Dr. Jan J. LerouManager, Experimental OperationsVelocys Inc.7950 Corporate Blvd.Plain City, OH 43064Phone: (614) 7333300Email: [email protected]
www.velocys.com