multiscale analysis anddesign of industrialand ... · vapor mass transfer flow = ... 15 eindhoven...

Laboratory for Chemical Technology, Ghent University

http://www.lct.UGent.be

Multiscale analysis and design of industrial andemerging thermal processes involving solids

Guy B. Marin

1

Eindhoven Multiscale Institute (EMI) Annual Symposium “Multiscale Challenge of Multiphase Flows” December 14 2016

2

Outline

• Introduction

• Steam cracking

• Vortex reactor

• Conclusions


Multiscale: from molecule to process

3


Molecular level

Process level

4

Outline

• Introduction

• Steam cracking

• Vortex reactor

• Conclusions


Steam cracking: from fossil to renewables

5


atozforex.com; pnnl.org; districtenergy.org; scade.fr; schmidt-clemens.de; Linde Group

Crude oil

Natural gas

Bio-based feeds

Steam cracking Consumer goods fromchemical industry

Naphtha

Light gasoil

crude oil

Natural gas liquids

Gas-condensates

Hydrode-oxygenated

FAME

Ethene

Propene

Butadiene

Aromatics

Thermochemical conversion of biomass

6


LIGNOCELLULOSIC BIOMASS

BIO-OIL CHAR

CO

CO2

H2OCH4

H2

GAS

NH3

HCNC2H4

5wt% Torrefaction35wt% Slow pyrolysis13wt% Fast pyrolysis85wt% Gasfication

20wt%30wt%75wt%5wt%

75wt%35wt%12wt%10wt%

Toraman, H.E.et al.,Bioresource Technology, 207, 229-236, 2016

COILSIM1D

Pilot plant

Simulation andOptimization

Industry


http://www.avgi.be/

Steam cracker: hot section

8

D

A

C

A: Radiation sectionB: TLEC: Steam drumD: Convection section

Endothermic process 1050–1150 K

Preheat feed and other utility streams

Rapidly quenching of reactor effluent

B

45%

50% 800K

700K

1050–1150 K

5%

1450 K

FPH

Saturated steam

BFWECO

Feed

Dilution steam

SSH

HTC

Steam


Coke formation


Optimization by- Feed additives- Metallurgy & surface technology- 3D reactor technology

Deposition of a carbon layer on the reactor surface

Thermal efficiency

Product selectivity

Decoking procedures

Estimated annual cost to industry: $ 2 billion

9

10

Steam cracker convection section

EvaporatorEvaporator

Steam

Feed

Nozzle

Steam

Feed

Nozzle Feed-steam mixture overheater-1

Feed-steam mixture overheater-1

Feed Evaporator

Steam super heater

Mixture over-heater-1

Mixture over-heater-2

Flue gas out ~ 700K

Flue gas in~ 1400 K

Mixing nozzle

To radiation section

Schematic of convection section

Heavy feed


Mahulkar, A.V. et al., Chemical Engineering Science, 110, 31-43, 2014

Droplet trajectories

11

Inlet bend

Inlet

Inlet bend

Inlet

Inlet bend

Inlet

100 micron inlet diameter

Dro

plet

Dia

met

er (m

)

Inlet bend

Inlet




• Large droplets (50 & 100 µm) impinge the inlet bend due to inertial separation from flow.• Large (50 & 100 µm) droplets reduce in size due to splashing while the smaller droplets (1 & 10

µm) reduce in size by evaporation


12

Gas condensate: regime map

• Wider stick regime as compared to that in single componentdroplet regime map

• For splash and limited splash, the no. of daughter droplets formedis greater than predicted by correlations available in literature

Rebound

StickSplash

Limited Splash

0

200

400

600

800

1000

1200

480 530 580 630 680 730 780

No

rma

l W

eb

er

nu

mb

er

Wall temperature (K)

Stick

Rebound

Splash

Limited Splash

820TLF

critNormWe


Mahulkar, A.V. et al., Chemical Engineering Science, 130, 275-289, 2015

Wall film model

• Droplet collection as mass source for film• Film velocity based on interfacial shear and gravity• Energy balance over the film• Film has one pseudo species (physical property of which changes as

evaporation proceeds)

13

Wall

Liquid film

Impinging droplet

Heat transfer from wall

Evaporative heat and mass transferVapor

flow

�� =��ℎ

2�� =

��ℎ�

3��Shear

componentGravitational component


14

Outline

• Introduction

• Steam cracking

• Vortex reactor

• Conclusions


Gas/Solid Fluidization Reactors

15


gravitational technologies centrifugal technologies

ConventionalFluidized Bed 1

Riser/Circulating Fluidized Bed 2

Conventional Rotating Fluidized Bed 3

Gas/Solid Vortex Reactor

1. van Hoef et al., Ann. Rev. Fluid Mech. 40 (2008) 47-702. http://www.fluidcodes.co.uk/fbed.html3. adapted from Watano et al., Powder Tech.131 (2003) 250-255

Real-time Video

16


0.9 mm polyvinylidene fluoride particles (ρ = 1800 kg/m3) ~1 kg/s air flow~5 kg bed mass

Gas/Solid Vortex Reactor (GSVR)

17


GSVR Characteristics:• Gas injection forces bed rotation

& induces fluidization• Centrifugal forces resist drag

� Dense bed � High radial slip velocity

gas flow path solid velocity

Tangential gasinjection

Rotating solidparticles

18

Outline

• Introduction

• Steam cracking

• Vortex reactor: cold flow

• Conclusions


Experimental GSVR Set-ups: cold flow

19


High-Speed Video

20


70 micron particles (5000 FPS)

~1.6 mm particles (10000 FPS)

CFD Example Movies

21


2D (with gravity), 0.74 kg/s air, 3250 g bed

Optimization of design

22


�8 inlet slots�10° with respect to the tangent�1 mm width�Gas inlet velocity 60-140 m/s

Side view ↓Top view ↑

Gas-only simulation

23


� N2 mass flow: 6.67 g s-1

� Inlet temperature: 842K� Turbulence model: RSM� 2.8 million cells mesh� 3rd order discretization

outlet

� Backflow is minimized and displaced upwards towards the outlet

[m/s]

Gas-only simulation

Geometry simplification for speedup of calculation

24


Actual inlet Circular inlet Pie-shape

� Same inner geometry and shape of inlet slots.� In all cases the same amount of gas is fed.

In theory, the flow pattern in the inner part of the reactor should not be affected.

25

Outline

• Introduction

• Steam cracking

• Vortex reactor: hot flow

• Conclusions


Gas-solid simulation procedure

• STEP 1: Gas-only simulation until steady-state– Hot gas (N2) enters via the outer ring

• STEP 2: Feeding of solid particles– Ring-shaped feeding zone right after the slits– Via UDF: add mass source term in the solid

phase continuity equationOR: “patch” a volume fraction of solids

– Particles at room temperature (no sourceterm in energy equation)

– Final goal: 15 g of solids in reactor

• STEP 3: Stabilization– Stop particle feed and monitor bed stabilization– Heating of particles by the hot gas

26


N2

Particles are fed in this ring

Some results

Unstable bed (a lot of bubbles) – Total mass of solids: 16 g

Stable, thin bed – Total mass of solids: 5.5 g

27


Sol

ids

volu

me

frac

tion

Sol

ids

volu

me

frac

tion

Iso-surface of solids volume fraction = 0.2

Iso-surface of solids volume fraction = 0.2

Some results

Gas enters at 1023 K, solids are introduced at 300 K (8 g/s)

Solid temperature evolution :

28


Time = 1 s(during solid feed step)

Time = 2 s(during solid feed step)

Time = 3 s(after solid feed step)

29

Outline

• Introduction

• Steam cracking

• Vortex: flow model

• Conclusions


Flow model: Euler/Euler

Conservation equations� Mass (� = �, �)

� Momentum

� Energy (� = �, �)

30


�

�� + � ⋅ �� = ��

�

�� + � ⋅ �� =−��! − �!� + � ⋅ ��" + �� + #$% ��& − �� + '�(),�

�

��&�&��& + � ⋅ �&�&��&��& =−�&�! + � ⋅ �& + �&�&�� + #$% �� − ��& + '�(),&

�

��ℎ� + � ⋅ ��ℎ� = ��

�!

��+ �̿�: �,� − � ⋅ -�� + .�/

�̿� = ��(�+0,�) �� + ��0 + �� 1� −

2

3(�+0,�) � ⋅ �� 2�Wherein, stress tensor:

Momentum transfer(Gidaspow?)

Turbulent dispersion(Simonin-et-al)

Heat transfer(Gunn)

Flow Model based on experimental program

31


Pressure Profile

Solids VelocityBed Thickness

(Bulk Solids Fraction)

Gas flow rate

GSVU geometry

Wall

Bed mass

Particle

CFD model parameters

Gas

Observable(dependent)

variables

Set (independent)

variables

Experimental Data


32

32

� Static Pressure profilePressure probes at wall

-0.5

0.5

1.5

2.5

3.5

0 0.09 0.18 0.27

Pre

ssur

e [k

Pa]

Radius [m]

∆4

∆4

012345

0.18 0.21 0.24 0.27Par

ticle

vel

ocity

[m

/s]

Radius [m]

� Bed Height: Visual observation, Pressure profile

∆4

PIV

� Solids VelocityParticle Image Velocimetry of solid phase near wall

56,7

56,8

� Particle azimuthal velocity: � Pressure drop over bed:

Pressure

probes

� Void fraction: 9:;<=

:;<=

:;<=

M.N.Pantzali et al., AIChEJ, 4114-4125 (2015)

32

Stereo PIV: single phase (seed particles)

z

r

Instantaneous 3D velocity fields

Average vector field

F=0.394 Nm3/s, vinj=55m/s


33

Single phase flow: model validation

34


Fields of velocity magnitudeExperimental

Radial profiles of azimuthal velocity

Numerical Simulations

Axial profiles of azimuthal velocity

0

20

40

60

80

100

120

140

0 0.1 0.2

Uθ

(m/s

)

r (m)

Niyogi K. et al, AIChE J., in press

35

Outline

• Introduction

• Steam cracking

• Vortex: drag model

• Conclusions


Radial Momentum Balance

Three main equations (simplified) :

1. Gas momentum balance

2. Solids momentum balance

3. Gas-solid combination

36

'> negligible ?

FS : Force exerted on the bed by the wall (N)

∆!

ℎ?@)='A�?@)

�� 1 − CD�,E�

F

F

ℎGH

F

F − ℎ='A�?@)

+'>�?@)

∆!

ℎ?@)= �� 1 − C

D�,E�

F

F

ℎGH

F

F − ℎ−'>�?@)


Radial Momentum Balance

37

Bed pressure drop

• Particles

DiameterDensity

‘1 mm’ ‘1.5 mm’ ‘2 mm’

HDPE 950 kg/m 3 0.9 1.4 1.8

PC 1240 kg/m 3 - - 1.9

Dependent variables studied:

• Air flow rate: Gf 0.4-0.8 Nm3/s

• Solids loading: 2 kg – Maximum Capacity

∆!

ℎ?@)= �� 1 − C

D�,E�

F

F

ℎGH

F

F − ℎ

Experimental validation of balance simplification

Gas-solid momentum balance:

Weight of the bed in centrifugal field

=

'> negligible !

∆�

IJKL[kPa/m]

�� 1 − CMN,OP

Q

Q

IGH

Q

QRI[kPa/m]

0

50

100

150

200

0 50 100 150 200


Drag Force – Modelling

Drag Coefficient definition (single particle)

- Drag force in GSVU:

- Model for SA,T>UM:

FV�,W ≔D&,W�&Y�&

� =2ZF

[,\]V^_`�G_�� ∗ �G_��ℎ�bcHV��b_��


'A = H�SAZY�

�

4

�&D&,WC

�

2

Ff,%g: = Cf,%gi)j

P

k

lmnmP

�

SA = oFV�,W? Cp Sr

Swirl ratio at inlet� ≔D&,7

D&,W

rθ

5t,u

5t,E

vw,E

5t

α

Where:

12

D&,W =y

2ZFz

D&,7 =y ∗ b_��

H,\]V^_`�G_�� ∗ �G_��ℎ�bcHV�� ∗ z

G : Gas mass flow rate (Nm3/s) L : Unit Length (m)

{

(azimuthalvelocityatslots)

(radialvelocityfrommassconservation)

Geometrical Factor

�

�

38

Regression of “measured” drag force FDCalculated: Proposed

Drag Model

Estimation of model parameters a,b,c,d by minimization of sum of squares of residuals

between experimental and calculated drag

forces

14

Experimental: Solids Radial Momentum Balance

FV�,W ≔D&,W�&Y�&

� =2ZF

[,\]V^_`�G_�� ∗ �G_��ℎ�bcHV��b_��

Dependentvariables:

CHp�?@)

Independent Variables:

ℎ?@)

��, yY� ��

D�,ED&,W =

y�

2ZFz

��:solids load (kg), G:gas mass flow rate (Nm3/s)

(�2�)(��,oG)

D&,W

F

ℎGH

F

F − ℎ�?@)�� 1 − C

D�,E�

F= 'A '�� = H��,{6�5

Z

4Y��1

2�&

D&,WC

�

��,{6�5 = �FV�,W; C�S�


Model performance

40


SA,T>UM = (15.00 ± 4.65)C�.� ±¡.¡¢FV�,W

(R¡.�¢±¡.¡£)S(¡.¤¥±¡.¡ )

0

10

20

30

40

50

60

70

0 200 400 600 800 1000 1200 1400

c D,G

SV

Uε

-1.9

3

Rep,r

Friedle M. et al., submitted

Flow Model

41


Pressure Profile

Solids VelocityBed Thickness

(Bulk Solids Fraction)

Gas flow rate

GSVU geometry

Wall

Bed mass

Particle

CFD model parameters

Gas

Observable(dependent)

variables

Set (independent)

variables

Model validation: radial pressure drop

42

Gas momentum balance:∆!

ℎ?@)='A�?@)

Including the model

Leads for c=2 and hbed~0 to:∆!

12�&D&,W

�=15

8H�FV�,W

R¡.�¢S¡.¤¥Y��

Fz


'A = H�SA,T>UMZ

4Y��1

2�&

D&,WC

�

SA,T>UM = 15C�.� FV�,W

R¡.�¢S¡.¤¥

Model validation: parity plot

43

0

200

400

600

800

1000

1200

1400

1600

0 200 400 600 800 1000 1200 1400 1600


∆!

12 �&D&,W

�

44

Outline

• Introduction

• Steam cracking

• Vortex reactor: pyrolysis

• Conclusions


Cold Flow Test with Pinewood

45

Ghent, 12/12/2016

� Pinewood particles� Irregular shapes maximum dimension 2

mm� Gas inlet velocity 90 m s-1

� Continuous feeding for approx. 10 s using a drill to drive a provisional injection screw

� Average bed height 15 mm

� Solids holdup 9-11 g (voidage 0.5-0.6)

Cold Flow Test with Pinewood

46

Ghent, 12/12/2016

8 inlet slots, 10° with respect to the tangent, 0.94 mm width

� Pinewood particles� Irregular shapes max. dimension 2 mm� Gas inlet velocity 90 m s-1

� Continuous feeding for approx. 10 s using a drill to drive a provisional injection screw

� Average bed height 15 mm� Average solids azimuthal velocity 4.5 m s-1

� Solids holdup 9-11 g (voidage 0.5-0.6)

Particle image velocimetry

Biomass Pyrolysis Process: flow diagram

47


http://www.pyne.co.uk

BIOMASS

BIO-OIL (TAR)

Flue gas

Sand + Char

Hot Sand

Air

Gas recycle

Pyrolysis Gas

Combustor

Pyrolyser

Pyrolysis Modeling in a GSVR

481. Xue, Heindel, and Fox, Chem. Eng. Sci. 66 (2011) 2440

• 2D periodic GSVR simulations

• Heterogeneous reactions (solid � gas + char):

• 10-reaction network with pseudo-components 1

• Continuous feeding of biomass

• Cellulose, hemicellulose, and lignin

• Different rates for each biomass component

• 4-phase Eulerian multiphase simulation (3 granular)

• Gas, biomass, char, and sand

• Sand and biomass retained in reactor

• Char leaves with gas flow due to lower density


VirginBiomass

ActiveBiomass

Tar (g)PyrolysisGases

Char (s) + Pyrolysis Gases

biomass

char

sand

Volume fraction

Volume Fraction and Temperature

49


biomass char sand0.030 0.065 0.55

50


BiomassChar

Total solids(biomass, char, sand)

Volume Fraction Animation0.08 0.15

0.63

Ashcraft, R.W. et al.,Chemical Engineering Journal, 207-208, 195-208, 2012

GSVR Process Intensification

51


1.Z.Y. Zhou, A.B. Yu, P. Zulli, Particle scale study of heat transfer in … fluidized beds, AIChE J. 55 (2009) 868–8842.Y. Ma, J.X. Zhu, Experimental study of heat transfer in a co-current downflow fluidized bed, Chem. Eng. Sci. 54 (1999) 41–50

Typical range1,2 static fluidized beds and risers/CFBs: ~100 – 200 W/(m2 K)

Gas/Solid Fluidization Reactors

52


Gas

/Sol

id S

lip V

eloc

ity

Solid Volume Fraction

Terminal velocity

Fluidized

Beds

Risers &

Circulating

Fluidized Beds

GSVR &

Rotating

Fluidized

Bed

Reactors

Improved gas/solid mass transferPotential for intensification

Images from: Watano et al., Powder Tech.131 (2003) 250-255

Gravitational technologiesLonger gas/solid contact time

Centrifugal technologiesShorter gas/solid contact time

Conclusions“ standard “ CFD models and codes • allow to describe and assess “new” reactor technologies involving multiple

scales

PIV data combined with visual observation and pressure me asurement• provide correlation for drag coefficient in GSVR flow model

GSVRs have the potential to intensify processes• High intrinsic mass/heat transfer can yield improved overall rates• High solid volume fractions can reduce equipment size

Biomass Pyrolysis • Stratification of solid phases to retain unreacted biomass: multifunctional reactor• Comparison to a static gravitational fluidized bed

– Order of magnitude larger heat and mass transfer coefficients

53


CRE in the 21st century

54

design

experimenttheory


Acknowledgments


The Long Term Structural Methusalem Funding

55

56

Acknowledgments

• Profs. Geraldine Heyndericks, Kevin Van Geem

• Drs. Marita Torregrosa, Vladimir Shtern, Amit Mahulkar, Ruben Debruycker, Maria Pantzali, Jelena Kovacevic, Robert Ashcraft

• PhD students Pieter Reyniers, Laurien Vandewalle, Kaustav Nyogi, Maximilian Friedl, Arturo Gonzalez-Quiroga


People

•Assistant professors: 4

•Visiting/senior scientists: 4

•Post-docs: 8

•PhD students: 60

•Technical staff: 10

•Administrative staff: 3

57

multiscale analysis anddesign of industrialand ... · vapor mass transfer flow = ... 15 eindhoven...

Documents