Assoc. Prof. Carolina P. Naveira Co2a Laboratory of Nano and - - PowerPoint PPT Presentation

▶
assoc prof carolina p naveira co2a
SMART_READER_LITE
LIVE PREVIEW

Assoc. Prof. Carolina P. Naveira Co2a Laboratory of Nano and - - PowerPoint PPT Presentation

Apr 24, 2023 482 likes •1.08k views

Experimental-TheoreHcal Analysis of Biodiesel Synthesis in Micro-reactors with Inverse Problem SoluHon for Parameter EsHmaHon Assoc. Prof. Carolina P. Naveira Co2a Laboratory of Nano and Microfluidcs and Micro-Systems LabMEMS Mechanical

slide-1
SLIDE 1
  • Assoc. Prof. Carolina P. Naveira Co2a

Laboratory of Nano and Microfluidcs and Micro-Systems – LabMEMS Mechanical Engineering Program – PEM/COPPE Engineering of Nanotecnology Program – PENT/COPPE Universidade Federal do Rio de Janeiro - UFRJ

Experimental-TheoreHcal Analysis of Biodiesel Synthesis in Micro-reactors with Inverse Problem SoluHon for Parameter EsHmaHon

slide-2
SLIDE 2

Fields/Areas:

  • Heat & Mass Transfer and Fluid Flow in Micro and Nano Scale
  • ConHnuum Mechanics and Complex Fluids
slide-3
SLIDE 3

Fields/Areas:

  • Heat & Mass Transfer and Fluid Flow in Micro and Nano Scale
  • ConHnuum Mechanics and Complex Fluids

Experim. Analysis Foward Analysis Inverse Analysis

Micro- Systems FabricaHon

slide-4
SLIDE 4

Fields/Areas:

  • Heat & Mass Transfer and Fluid Flow in Micro and Nano Scale
  • ConHnuum Mechanics and Complex Fluids

Experim. Analysis Foward Analysis Inverse Analysis

Micro- Systems FabricaHon

μ-PIV : ParHcle Image Velocimetry μ-LIF : Laser Induced Fluorencense Infrared Thermography Non-Intrusive Measurements

slide-5
SLIDE 5

Fields/Areas:

  • Heat & Mass Transfer and Fluid Flow in Micro and Nano Scale
  • ConHnuum Mechanics and Complex Fluids

Experim. Analysis Forward Analysis Inverse Analysis

Micro- Systems FabricaHon Hybrid Methods (analiHcal -numerical ) Integral Transforms (GITT) Bayesian Inference

Fast, accurate and robust! computaHonally intensive!

slide-6
SLIDE 6

Nanocomposites and Nanofluids Micro reactors Micro-Heat Exchangers Micro-Heat Spreaders Micro-Sensors Micro-Models of Porous Media EOR – Enhanced Oil Recovery Human on a chip (cell toxicology) Cell culture Cell separaHon Cell encapsulaHon Bio-prinHng Micro-Models of Porous Media

slide-7
SLIDE 7

Nanocomposites and Nanofluids Micro reactors Micro-Heat Exchangers Micro-Heat Spreaders Micro-Sensors Micro-Models of Porous Media

Combining

EOR – Enhanced Oil Recovery Human on a chip (cell toxicology) Cell culture Cell separaHon Cell encapsulaHon Bio-prinHng Micro-Models of Porous Media

slide-8
SLIDE 8

R&D Challenges : Biodiesel Production Intensification with Heat Recovery from High Concentration Photovoltaic Cells (HCPV)

8

  • Rejected heat can be used for other

purposes (desalination, heating, cooling, biodiesel production, etc) ;

slide-9
SLIDE 9

R&D Challenges : Biodiesel Production Intensification with Heat Recovery from High Concentration Photovoltaic Cells (HCPV)

9

  • Rejected heat can be used for other

purposes (desalination, heating, cooling, biodiesel production, etc) ;

slide-10
SLIDE 10

R&D Challenges : Biodiesel Production Intensification with Heat Recovery from High Concentration Photovoltaic Cells (HCPV)

10

HCPV system High Concentration: 800 suns Power Generation: 1.5kVA

  • Rejected heat can be used for other

purposes (desalination, heating, cooling, biodiesel production, etc) ;

slide-11
SLIDE 11

R&D Challenges : Biodiesel Production Intensification with Heat Recovery from High Concentration Photovoltaic Cells (HCPV)

11

HCPV system High Concentration: 800 suns Power Generation: 1.5kVA

  • Rejected heat can be used for other

purposes (desalination, heating, cooling, biodiesel production, etc) ;

OpHmized Micro Heat Exchanger For the HCPV cooling system

slide-12
SLIDE 12

R&D Challenges : Biodiesel Production Intensification with Heat Recovery from High Concentration Photovoltaic Cells (HCPV)

12

  • Rejected heat can be used for other

purposes (desalination, heating, cooling, biodiesel production, etc) ;

OpHmized Micro Heat Exchanger For the HCPV cooling system Integrated micro-heat exchanger and micro reactor;

slide-13
SLIDE 13

R&D Challenges : Biodiesel Production Intensification with Heat Recovery from High Concentration Photovoltaic Cells (HCPV)

13

OpHmized Micro Heat Exchanger For the HCPV cooling system Integrated micro-heat exchanger and micro reactor;

SIGNIFICANT POTENTIALS…

  • Portable biodiesel production;
  • Short residence time;
  • Continuous mode;
slide-14
SLIDE 14

14

Take advantage of the high surface area-volume ratio To achieve improved mass and heat transfer rates Time scales reduction Which results in lower energy consumptiom Micro reactor

  • Fig. : Advantages of microreactors in reactional systems.

R&D Challenges : Biodiesel Production in micro reactors

slide-15
SLIDE 15

Take advantage of the high surface area-volume ratio To achieve improved mass and heat transfer rates Time scales reduction Which results in lower energy consumptiom Micro reactor

  • Fig. : Advantages of microreactors in reactional systems.

R&D Challenges : Biodiesel Production in micro reactors

15

15

  • Fig. Complete manifold of micro

reactors

  • Fig. Single micro reactor;
  • Fig. Module of micro reactors
slide-16
SLIDE 16

Ø Design of optimized micro-reactor Experimental Analysis Theoretical Analysis

R&D Challenges : Biodiesel Production in micro reactors

Alcohol + catalyst Vegetable oil biodiesel

1 2 3 4 5 6

k k k k k k

Triglyceride(TG) + Alcohol(A) Diglyceride(DG) + Biodiesel(B) Diglyceride(DG) + Alcohol(A) Monoglyceride(MG) + Biodiesel(B) Monoglyceride(MG) + Alcohol(A) Gly ⎯⎯ → ←⎯ ⎯ ⎯⎯ → ←⎯ ⎯ ⎯⎯ → ←⎯ ⎯ cerol(GL) + Biodiesel(B)

slide-17
SLIDE 17

Ø Design of optimized micro-reactor Experimental Analysis Theoretical Analysis

R&D Challenges : Biodiesel Production in micro reactors

Inverse Analysis Kinetic constants estimation

k1, k2, k3, k4, k5, k6

1 2 3 4 5 6

k k k k k k

Triglyceride(TG) + Alcohol(A) Diglyceride(DG) + Biodiesel(B) Diglyceride(DG) + Alcohol(A) Monoglyceride(MG) + Biodiesel(B) Monoglyceride(MG) + Alcohol(A) Gly ⎯⎯ → ←⎯ ⎯ ⎯⎯ → ←⎯ ⎯ ⎯⎯ → ←⎯ ⎯ cerol(GL) + Biodiesel(B)

slide-18
SLIDE 18

18

Micro-fabrication Experimental Analysis Metal – Glass micro reactor

R&D Challenges : Biodiesel Production in micro reactors

Ø Design of optimized micro-reactor

glass

Oil Alcohol Interface

Fig.: Exp. observation of stratified flow pattern formed by the alcohol and the vegetable oil. Theoretical Analysis

slide-19
SLIDE 19

19

Mathematical modelling Steady state diffusive-advective mass transfer equations with nonlinear chemical reaction terms Parametric analysis Residence time Reaction temperature Cross section Theoretical Analysis fully developed stratified laminar flow

R&D Challenges : Biodiesel Production in micro reactors

Ø MODELLING Experimental Analysis

slide-20
SLIDE 20

20

Fase álccol Fase Triglicerídeo H W L

x z y

HTG HA Alcohol phase Triglyceride phase

Velocity profiles for stratified flow: Ø FLOW PROBLEM: FORMULATION AND SOLUTION

uA(y,z) = ! Ψvel,i(z)S5i

i=1 ∞

∑

1+ µTG µA ⎛ ⎝ ⎜ ⎞ ⎠ ⎟ -sinh Hiπ W ⎛ ⎝ ⎜ ⎞ ⎠ ⎟ + sinh iπ y W ⎛ ⎝ ⎜ ⎞ ⎠ ⎟ ⎡ ⎣ ⎢ ⎤ ⎦ ⎥ + ⎧ ⎨ ⎪ ⎩ ⎪ + 1- µTG µA ⎛ ⎝ ⎜ ⎞ ⎠ ⎟ sinh H - 2HTG

( )iπ

W ⎛ ⎝ ⎜ ⎞ ⎠ ⎟ + sinh iπ 2HTG - y

( )

W ⎛ ⎝ ⎜ ⎞ ⎠ ⎟ + ⎡ ⎣ ⎢ ⎢

  • sinh iπ H - HTG - y

( )

W ⎛ ⎝ ⎜ ⎞ ⎠ ⎟ - sinh iπ H + HTG - y

( )

W ⎛ ⎝ ⎜ ⎞ ⎠ ⎟ ⎤ ⎦ ⎥ ⎥ + 2sinh iπ(H - y) W ⎛ ⎝ ⎜ ⎞ ⎠ ⎟ ⎫ ⎬ ⎪ ⎭ ⎪

R&D Challenges : Biodiesel Production in micro reactors

slide-21
SLIDE 21

21

Ø MASS TRANSFER MODELLING Dimensionless reaction-convection-diffusion :

( ) ( ) ( ) ( )

2 2 2 2 2 2

, , , , , , , , ( , )

s s s s TG s s

F X Y Z F X Y Z F X Y Z F X Y Z U Y Z G X X Y Z ξ γ δ ς ⎛ ⎞ ∂ ∂ ∂ ∂ = + + + ⎜ ⎟ ∂ ∂ ∂ ∂ ⎝ ⎠

Table: Dimensionless kinetic relations for the species in the transesterification reaction.

( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )

1 2 1 3 5 2 4 6 1 3 2 4 3 5 4 6 5 6 1 3 5 2 4 6

Re

s TG A DG B TG DG MG A DG MG GL B TG DG A DG MG B DG MG A MG GL B MG A GL B TG DG MG A DG MG GL

Species action termsG Fs TG k F F k F F A k F k F k F F k F k F k F F DG k F k F F k F k F F MG k F k F F k F k F F GL k F F k F F B k F k F k F F k F k F k F F − + − − − + + + − + − + − + − + − + + + − − −

B

3D Model:

R&D Challenges : Biodiesel Production in micro reactors

slide-22
SLIDE 22

22

Ø MASS TRANSFER MODELLING

  • Solved thougth GITT (Generalized Integral Transform Tecnique)

Hybrid numerical-analytical solution with automatic error control

R&D Challenges : Biodiesel Production in micro reactors

1 - Choose the associated eigenvalue problem. 2 - Develop the integral transform pair. 3 - Integral transform the original PDE. 4 - Numerically (or analytically) solve the resulting coupled ODE system for the transformed potentials. 5 - Recall the analytical inversion formula to reconstruct the desired potential.

STEPS in the Generalized Integral Transform Technique (G.I.T.T.)

slide-23
SLIDE 23

23

Micro-fabrication Metal – Glass micro reactor

R&D Challenges : Biodiesel Production in micro reactors

Experimental Analysis Theoretical Analysis multiple Metal – Metal micro reactor + micro heat exchanger

slide-24
SLIDE 24

24

Micro-fabrication Metal – Glass micro reactor

R&D Challenges : Biodiesel Production in micro reactors

Features of the Device:

  • Composed of 10 micro reactors
  • Composed of 11 micro-heat exchanger
  • Total Dimensions 2,5cm x 4cm x 1.27cm.
  • Microchannels with square section 400µmX400µm
  • Total Length of the microchannel of the reactor of 43.26 cm.

3D printed (FSL) Theoretical Analysis Experimental Analysis multiple Metal – Metal micro reactor + micro heat exchanger

slide-25
SLIDE 25

25

Micro-fabrication Experimental Analysis Metal – Glass micro reactor multiple Metal – Metal micro reactor + micro heat exchanger

R&D Challenges : Biodiesel Production in micro reactors

3D printed (FSL) Theoretical Analysis E t h a n

  • l
slide-26
SLIDE 26

26

Experimental Analysis Inverse Analysis Kinetic constants estimation k1, k2, k3, k4, k5, k6 Theoretical Analysis

1 2 3 4 5 6

k k k k k k

Triglyceride(TG) + Alcohol(A) Diglyceride(DG) + Biodiesel(B) Diglyceride(DG) + Alcohol(A) Monoglyceride(MG) + Biodiesel(B) Monoglyceride(MG) + Alcohol(A) Gly ⎯⎯ → ←⎯ ⎯ ⎯⎯ → ←⎯ ⎯ ⎯⎯ → ←⎯ ⎯ cerol(GL) + Biodiesel(B)

Implemented Method MCMC

R&D Challenges : Biodiesel Production in micro reactors

✔ ✔

Ethanol

slide-27
SLIDE 27

27

R&D Challenges : Biodiesel Production in micro reactors

Ø SENSITIVITY ANALYSIS Figure: Reduced sensitivity coefficients.

k

k = 10−k

It increases the sensitivity

  • f parameters and reduces

the search region

Range of search:

k : 10−9 to 10−1

Range of search :

k : 1 to 9

slide-28
SLIDE 28

28

R&D Challenges : Biodiesel Production in micro reactors

Ø SENSITIVITY ANALYSIS

Table – Cases for each set of measures of the species used in the analysis of the │JTJ│.

1 2 3 4 5 6

k k k k k k

Triglyceride(TG) + Alcohol(A) Diglyceride(DG) + Biodiesel(B) Diglyceride(DG) + Alcohol(A) Monoglyceride(MG) + Biodiesel(B) Monoglyceride(MG) + Alcohol(A) Gly ⎯⎯ → ←⎯ ⎯ ⎯⎯ → ←⎯ ⎯ ⎯⎯ → ←⎯ ⎯ cerol(GL) + Biodiesel(B)

Good Exp.

slide-29
SLIDE 29

29

R&D Challenges : Biodiesel Production in micro reactors

Ø SENSITIVITY ANALYSIS

1 2 3 4 5 6

k k k k k k

Triglyceride(TG) + Alcohol(A) Diglyceride(DG) + Biodiesel(B) Diglyceride(DG) + Alcohol(A) Monoglyceride(MG) + Biodiesel(B) Monoglyceride(MG) + Alcohol(A) Gly ⎯⎯ → ←⎯ ⎯ ⎯⎯ → ←⎯ ⎯ ⎯⎯ → ←⎯ ⎯ cerol(GL) + Biodiesel(B)

Table – Cases for each set of measures of the species used in the analysis of the │JTJ│.

“Bad” Exp.

slide-30
SLIDE 30

30

Ø MASS TRANSFER MODELLING Dimensionless reaction-convection-diffusion : 3D Model:

R&D Challenges : Biodiesel Production in micro reactors

UTG(Y,Z) ∂Fs X,Y,Z

( )

∂X = ξs γ ∂2 Fs X,Y,Z

( )

∂X 2 + ∂2 Fs X,Y,Z

( )

∂Y 2 +δ ∂2 Fs X,Y,Z

( )

∂Z 2 ⎛ ⎝ ⎜ ⎞ ⎠ ⎟ +ςGs s = TG, MG,DG,B,GL, A

( ) ( ) ( )

2 2 2 2

, , , , , , ( , )

s s s TG s s

F X Y Z F X Y Z F X Y Z U Y Z G X X Y ξ γ ς ⎛ ⎞ ∂ ∂ ∂ = + + ⎜ ⎟ ∂ ∂ ∂ ⎝ ⎠

2D - Parallel plates :

slide-31
SLIDE 31

31

3D Model:

R&D Challenges : Biodiesel Production in micro reactors

UTG(Y,Z) ∂Fs X,Y,Z

( )

∂X = ξs γ ∂2 Fs X,Y,Z

( )

∂X 2 + ∂2 Fs X,Y,Z

( )

∂Y 2 +δ ∂2 Fs X,Y,Z

( )

∂Z 2 ⎛ ⎝ ⎜ ⎞ ⎠ ⎟ +ςGs s = TG, MG,DG,B,GL, A

( ) ( ) ( )

2 2 2 2

, , , , , , ( , )

s s s TG s s

F X Y Z F X Y Z F X Y Z U Y Z G X X Y ξ γ ς ⎛ ⎞ ∂ ∂ ∂ = + + ⎜ ⎟ ∂ ∂ ∂ ⎝ ⎠

2D - Parallel plates :

....

i=1 1

∑

....

i=1 40

∑

Ø MASS TRANSFER MODELLING Dimensionless reaction-convection-diffusion :

slide-32
SLIDE 32

32

3D Model:

R&D Challenges : Biodiesel Production in micro reactors

UTG(Y,Z) ∂Fs X,Y,Z

( )

∂X = ξs γ ∂2 Fs X,Y,Z

( )

∂X 2 + ∂2 Fs X,Y,Z

( )

∂Y 2 +δ ∂2 Fs X,Y,Z

( )

∂Z 2 ⎛ ⎝ ⎜ ⎞ ⎠ ⎟ +ςGs s = TG, MG,DG,B,GL, A

( ) ( ) ( )

2 2 2 2

, , , , , , ( , )

s s s TG s s

F X Y Z F X Y Z F X Y Z U Y Z G X X Y ξ γ ς ⎛ ⎞ ∂ ∂ ∂ = + + ⎜ ⎟ ∂ ∂ ∂ ⎝ ⎠

2D - Parallel plates : 1D - Lumped-Differential Model :

UTG dFs X

( )

dX = ςGs , s = TG, MG,DG,B,GL UTG dFA X

( )

dX = ξ A 3P*FA X

( )+ Q*

( )+ςGA

Ø MASS TRANSFER MODELLING Dimensionless reaction-convection-diffusion : CIEA - Coupled Integral EquaHons Approach (Improved Lumped Analysis)

....

i=1 1

∑

....

i=1 40

∑

slide-33
SLIDE 33

R&D Challenges : Biodiesel Production in micro reactors

Ø BAYE’S FORMULA

π posterior(P) = π(P Yexp) = π prior(P)π(Yexp P) π(Yexp)

x posterior prior likelihood ∝

Ø MCMC (MARKOV CHAIN MONTE CARLO METHODS)

METROPOLIS-HASTINGS ALGORITHM 1. Sample a Candidate Point P* from a jumping distribution q(P*,P(t-1)). 2. Calculate: 3. Generate a random value U which is uniformly distributed on (0,1). 4. If U a, define P(t) = P* ; otherwise, define P(t) = P(t-1). 5. Return to step 1 in order to generate the sequence {P(1) , P(2) , …, P(n)}.

* ( 1) * ( 1) * ( 1)

( | ) ( , ) min 1, ( | ) ( , )

t t t

q q π α π

− − −

⎡ ⎤ = ⎢ ⎥ ⎣ ⎦ P Y P P P Y P P

slide-34
SLIDE 34

R&D Challenges : Biodiesel Production in micro reactors

Ø LIKELIHOOD

  • The exp. errors are additive, with zero mean and normally distributed.
  • The statistical parameters describing the errors are known.
  • There are no errors in the independent variables.
  • P is a random vector with known mean µ and known covariance matrix W.
  • P is independent of Yexp.

π(Yexp P) = (2π)− I/2 W−1 −1/2 exp − 1 2 (Yexp - Ymodel)T W(Yexp - Ymodel) ⎡ ⎣ ⎢ ⎤ ⎦ ⎥

slide-35
SLIDE 35

R&D Challenges : Biodiesel Production in micro reactors

Ø LIKELIHOOD

π(Yexp P) = (2π)− I/2 W−1 −1/2 exp − 1 2 (Yexp - Ymodel)T W(Yexp - Ymodel) ⎡ ⎣ ⎢ ⎤ ⎦ ⎥

Ymodel = Yreduced

model

+ ε(P)

Ø APROXIMATION ERROR MODEL ε P

( ) = ε reduced

model

P

( )+ εexp

slide-36
SLIDE 36

R&D Challenges : Biodiesel Production in micro reactors

Ø LIKELIHOOD

π(Yexp P) = (2π)− I/2 W−1 −1/2 exp − 1 2 (Yexp - Ymodel)T W(Yexp - Ymodel) ⎡ ⎣ ⎢ ⎤ ⎦ ⎥

ε P

( ) = ε reduced

model

P

( )+ εexp

Ymodel = Yreduced

model

+ ε(P)

π(Y P) = (2π)− I/2 ! W−1 −1/2 exp − 1 2 Y − Yreduced

model

P

( )− ε

⎛ ⎝ ⎜ ⎞ ⎠ ⎟

T

! W(Y − Yreduced

model

P

( )− ε)

⎡ ⎣ ⎢ ⎢ ⎤ ⎦ ⎥ ⎥

ε ≈ ε

reduced model

! W ≈ W

reduced model

+ W

exp

Ø APROXIMATION ERROR MODEL

ε = ε

reduced model

+ ε

exp + Γε Pr Γ Pr

−1 Pr − µ

( )

! W = W

reduced model

+ W

exp - Γε Pr Γ Pr

−1Γ Prε

ε

exp → 0, measurement uncertainties have zero mean

Γε

Pr → 0, neglecting the linear dependence between ε and Pr
slide-37
SLIDE 37

37

Ø SIMULATED MEASURES

( )

model

0.05Max σ = Y

3D Model x 2D Parallel Plates Model Triglyceride Diglyceride Monoglyceride Biodiesel

...

i=1 40

∑

...

i=1 1

∑

slide-38
SLIDE 38

38

Ø Parameter estimation Triglyceride Diglyceride Monoglyceride Biodiesel 3D Model x 2D Parallel Plates Model

...

i=1 40

∑

...

i=1 1

∑

slide-39
SLIDE 39

39 Iterations Step size Acceptance Burning

100000 0.003 41.94% 20000 Ø Minimization of objective function Minimization of objective function with 2D parallel plate model and approximation error model 3D Model x 2D Parallel Plates Model

...

i=1 40

∑

...

i=1 1

∑

slide-40
SLIDE 40

40

Ø Markov Chains EsHmated parameter Reference parameter (Al-Dhubabian, 2005) 3D Model x 2D Parallel Plates Model

...

i=1 40

∑

...

i=1 1

∑

slide-41
SLIDE 41

41

Ø Markov Chains EsHmated parameter Reference parameter (Al-Dhubabian, 2005) 3D Model x 2D Parallel Plates Model

...

i=1 40

∑

...

i=1 1

∑

slide-42
SLIDE 42

42

Ø ERROR MODEL Parameters estimated with 2D parallel plate model WITH approximation error model Parameters estimated with 2D parallel plate model WITHOUT approximation error model 3D Model x 2D Parallel Plates Model

...

i=1 40

∑

...

i=1 1

∑

slide-43
SLIDE 43

43

Ø ERROR MODEL Parameters estimated with 2D parallel plate model WITH approximation error model Parameters estimated with 2D parallel plate model WITHOUT approximation error model 3D Model x 2D Parallel Plates Model

...

i=1 40

∑

...

i=1 1

∑

...

i=1 1

∑

...

i=1 1

∑

reduced model

≈ ε ε ε ε

reduced model

≈ ε ε ε ε

slide-44
SLIDE 44

44

Ø ERROR MODEL Parameters estimated with 2D parallel plate model WITH approximation error model Parameters estimated with 2D parallel plate model WITHOUT approximation error model 3D Model x 2D Parallel Plates Model

...

i=1 40

∑

...

i=1 1

∑

...

i=1 1

∑

...

i=1 1

∑

reduced model

≈ ε ε ε ε

reduced model

≈ ε ε ε ε model

  • Comp. Hme for

1000 states

  • Comp. Hme for

200 000 states 3D with 40 terms 21h 175 days 3D with 1 terms 13s 44 min 2D with 40 terms 10h 83 days 2D with 1 terms 20s 1h

slide-45
SLIDE 45

45

Ø ERROR MODEL Triglyceride Diglyceride Monoglyceride Biodiesel 3D Model x 1D Lumped Model

...

i=1 40

∑

slide-46
SLIDE 46

46

Ø ERROR MODEL Trigliceride Digliceride Mongliceride Biodiesel

...

i=1 40

∑

3D Model x 1D Lumped Model

slide-47
SLIDE 47

47

Ø ERROR MODEL Trigliceride Digliceride Mongliceride Biodiesel

...

i=1 40

∑

3D Model x 1D Lumped Model model

  • Comp. Hme for

1000 states

  • Comp. Hme for

200 000 states 3D with 40 terms 21h 175 days 3D with 1 terms 13s 44 min 2D with 40 terms 10h 83 days 2D with 1 terms 20s 1h 1D - lumped 20s 50min

slide-48
SLIDE 48

48

Ø ERROR MODEL Triglyceride Diglyceride Monoglyceride Biodiesel Real Experimental data + 1D Lumped Model

slide-49
SLIDE 49

49

Experimental Analysis Theoretical Analysis

R&D Challenges : Biodiesel Production in micro reactors

EsHmated constant

Validation Case Residence Time Biodiesel Conc. Experim. Result [mol/m3] Biodiesel Conc. Predicted

  • Math. Model

[mol/m3] Percentage error 1 0.78 min 2.676,96 2.626,26 2,0% 2 1.55 min 2.646,54 2.623,22 1,1% EsHmated biodiesel concentraHon, based on the esHmated kineHc constants, in comparison with two addiHonal cases of residence Hme.

ValidaJon Experiments

slide-50
SLIDE 50

50

Experimental Analysis Theoretical Analysis

R&D Challenges : Biodiesel Production in micro reactors

EsHmated constant

Validation Case Residence Time Biodiesel Conc. Experim. Result [mol/m3] Biodiesel Conc. Predicted

  • Math. Model

[mol/m3] Percentage error 1 0.78 min 2.676,96 2.626,26 2,0% 2 1.55 min 2.646,54 2.623,22 1,1% EsHmated biodiesel concentraHon, based on the esHmated kineHc constants, in comparison with two addiHonal cases of residence Hme.

ValidaJon Experiments

slide-51
SLIDE 51

R&D Challenges : Biodiesel Production in microreactors

Biodiesel Production Device Total weigth Device Total volume 1 module 10 micro-reactors 1,33 L/day 123 g 2,5cm X 4cm X 1,27cm

Ø SCALING UP...

slide-52
SLIDE 52

R&D Challenges : Biodiesel Production in microreactors

Biodiesel Production Device Total weigth Device Total volume 1 module 10 micro-reactors 1,33 L/day 123 g 2,5cm X 4cm X 1,27cm

Ø SCALING UP...

slide-53
SLIDE 53

R&D Challenges : Biodiesel Production in microreactors

biodiesel WCO Ø WASTE COOKING OIL (WCO) ...

slide-54
SLIDE 54

R&D Challenges : Biodiesel Production in microreactors

Ø WASTE COOKING OIL (WCO) ... biodiesel WCO

slide-55
SLIDE 55

55

Ø Conclusions

  • Complex physical problem can take advantage of both Computational and Experimental Analysis ;
  • Successful fabrication of micro reactors – metal/glass and metal/metal 3D printed
  • Estimation of kinetic constants from inverse analysis using real experimental data and taking account the

error model;

  • Demonstration of micro reactors for the synthesis of ethanol based biodiesel with promising results

(99,61% of biodiesel production in 35 seconds residence time);

R&D Challenges : Biodiesel Production in microreactors

slide-56
SLIDE 56

56

Ø Conclusions

  • Complex physical problem can take advantage of both Computational and Experimental Analysis ;
  • Successful fabrication of micro reactors – metal/glass and metal/metal 3D printed
  • Estimation of kinetic constants from inverse analysis using real experimental data and taking account the

error model;

  • Demonstration of micro reactors for the synthesis of ethanol based biodiesel with promising results

(99,61% of biodiesel production in 35 seconds residence time);

R&D Challenges : Biodiesel Production in microreactors

Ø Future work

  • Fabricate a pilot plant with up to 200-500 microrreactors;
  • Enhance the conversion of waste cooking oil in the esterification process in microreactor;
  • Estimate the kinectic constant for the waste cooking oil esterification;
  • Optimize the micro channel structure based on the estimated constant;
slide-57
SLIDE 57

57

ACKNOWLEDGEMENT Sponsoring Agencies: Our students :

  • DSc. José Martim Costa Junior
  • MSc. Pericles Crisiron Pontes
  • MSc. Kelvin Chen

CTI Collaborators:

  • DSc. Jorge Vicente Lopes da Silva
  • MSc. Paulo Inforçatti
  • DSc. Izaque Alves Maia

UCL Collaborators:

  • Prof. Stavroula Balabani
  • Prof. Manish Tiwari

UFRJ Collaborators:

  • Prof. Luiz Antônio d’Avila
  • DSc. Cristiane Gimenes de Souza
  • Prof. Donato Alexandre Aranda
  • Prof. Yordanka Reyes Cruz
  • Prof. Helcio Orlande
  • Prof. Marcelo Colaço
  • Prof. Antonio Leitao
  • Prof. Bernd Hofmann
  • Eng. Diego Busson
  • Eng. Saxon Paiz
slide-58
SLIDE 58

58

ACKNOWLEDGEMENT Our students :

  • DSc. José Martim Costa Junior
  • MSc. Pericles Crisiron Pontes
  • MSc. Kelvin Chen

CTI Collaborators:

  • DSc. Jorge Vicente Lopes da Silva
  • MSc. Paulo Inforçatti
  • DSc. Izaque Alves Maia

UCL Collaborators:

  • Prof. Stavroula Balabani
  • Prof. Manish Tiwari

UFRJ Collaborators:

  • Prof. Luiz Antônio d’Avila
  • DSc. Cristiane Gimenes de Souza
  • Prof. Donato Alexandre Aranda
  • Prof. Yordanka Reyes Cruz
  • Prof. Helcio Orlande
  • Prof. Marcelo Colaço
  • Prof. Antonio Leitao
  • Prof. Bernd Hofmann
  • Eng. Diego Busson
  • Eng. Saxon Paiz

carolina@mecanica.coppe.ufrj.br

LabMEMS – Nano and Microfluidics and Microsystems Laboratory Mechanical Engineering Dept. (PEM) & Nanoengineering Dept. (PENT) Universidade Federal do Rio de Janeiro – UFRJ/COPPE Rio de Janeiro - Brazil