Modelling the aerosol synthesis of silica nanoparticles from TEOS. - PowerPoint PPT Presentation

Modelling the aerosol synthesis of silica nanoparticles from TEOS. Shraddha Shekar 1 , Ali Abdali 2 , Mustapha Fikri 2,3 , Christof Schulz 2,3 and Markus Kraft 1 July, 2011 1 University of Cambridge, UK 2 IVG, University of Duisburg-Essen, Germany 3 CeNIDE, Center for Nanointegration Duisburg-Essen

Some basic questions • What is TEOS? • What are silica nanoparticles (SiNP)? • Why are they important? • Why model this system?

Tetraethoxysilane TEOS • Central silicon attached to 4-ethoxy branches • Vibrations and rotations within the molecule • Many possible ways of bond breaking, many possible reactions • Preferred precursor because relatively inexpensive and halide- free.

Silica nanoparticles What are SiNP? Network of Si-O bonds such that Si:O = 1:2 Why are they important? •Support material for functional/composite nanoparticles, catalysis •Bio-medical applications, drug delivery •Optics, optoelectronics, photoelectronics •Fabrics, clothes

Why model this? Silica nanoparticles Precursor(TEOS) Aerosol reactor Industrial P ≥ 1 atm Scale T ≈ 1200 - 2000 K •What are the optimal process conditions? •What are the final product properties? Molecular •What is the final particle size distribution? Scale •What happens in the gas-phase? •How do gas-phase precursors form the particles? •How do these particles grow? •How to describe the overall system from first-principles?

Methods: Ab initio modelling Species generation Quantum Chemistry calculations Statistical Mechanics Thermochemistry Overall Model calculation H(T) S(T) C p (T) Equilibrium Chemical Population Balance Model calculation Kinetics Ref: W. Phadungsukanan, S. Shekar, R. Shirley, M. Sander, R. H. West, and M. Kraft. First-principles thermochemistry for silicon species in the decomposition of tetraethoxysilane. J. Phys. Chem. A , 113 , 9041–9049, 2009

Reaction kinetics • Kinetics • Equilibrium – Reaction set generated – Hints towards the to include all existence of stable intermediates and intermediates & products from products. equilbrium. – Intermediates – Reactions obey Si(OH) x (OCH 3 ) 4-x Arrhenius rate law Si(OH) y (OC 2 H 5 ) 4-y k = AT β e -Ea/RT – Main Product Si(OH) 4 – Rate parameters (A, β , Ea) fitted to experimental vaues (a) (a) J. Herzler, J. A. Manion, and W. Tsang. Single-Pulse Shock Tube Study of the decomposition of tetraethoxysilane and Related Compounds. J. Phys. Chem. A , 101 , 5500-5508, 1997

Reaction Mechanism • Heuristic reaction mechanism C6H16O4SI2 = C2H4 + C4H12O4SI2 C8H20O4SI = C6H16O4SI + C2H4 C4H12O4SI2 = C2H4 + C2H8O4SI2 C6H16O4SI = C4H12O4SI + C2H4 C2H8O4SI2 = C2H4 + H4O4SI C4H12O4SI = C2H8O4SI + C2H4 C8H20O4SI = C2H5 + C6H15O4SI2 C2H8O4SI = H4O4SI + C2H4 C6H15O4SI2 + H = C6H16O4SI2 C8H20O4SI = C7H17O4SI + CH3 C6H16O4SI2 = C2H5 + C4H11O4SI2 C7H17O4SI = C2H4 + C5H13O4SI C4H11O4SI2 + H = C4H12O4SI2 C5H13O4SI = C2H4 + C3H9O4SI C4H12O4SI2 = C2H5 + C2H7O4SI2 C3H9O4SI = C2H4 + CH5O4SI C2H7O4SI2 + H = C2H8O4SI2 CH5O4SI = H4O4SI + CH C2H7O4SI2 = H3O4SI + C2H4 C7H17O4SI = C7H16O4SI + H H3O4SI = H2O3SI + OH C7H16O4SI = C6H13O4SI + CH3 H2O3SI = SIO2 + H2O C6H13O4SI = C6H12O4SI + H C6H15O4SI2 = C2H3 + C4H12O4SI2 C6H12O4SI = C4H10O4SI + C2H2 C4H11O4SI2 = C2H3 + C2H8O4SI2 C4H10O4SI = C2H8O4SI + C2H2 C2H8O4SI2 = H3O4SI + C2H3 C8H20O4SI = C2H4 + C6H16O4SI2

Flux and Sensitivity Analyses Reaction number Main Reaction Pathway

Rate parameter estimation The rate parameters have been fitted to shock-tube experimental data provided by Herzler et al (a) Step 1: Low discrepancy series To perform a pre-scan of parameters for 18 Si reactions. Step 2: Sensitivity Analysis To identify the 4 most sensitive parameters Step 3: Response Surface Methodology Uncertainties in model parameters for reactions R1 and R15 To estimate model uncertainties (a) J. Herzler, J. A. Manion, and W. Tsang. Single-Pulse Shock Tube Study of the decomposition of tetraethoxysilane and Related Compounds. J. Phys. Chem. A , 101 , 5500-5508, 1997

Optimisation Theory • Experimental data η = η ± σ exp exp exp 0 • Model response with paramaters x ( ) ( ) η = η = x ,..., x with x x K 1 = + ξ x x c • and with uncertainty factor c and 0 standard normally distributed ξ For a simple linear model, K =1 ( ) ( ) ( ) η = + η ξ = + + ξ x A B x x , c , A B x c 0 [ ] ( ) ( ) ( ) ( ( ) ) µ = η ξ = + σ = η ξ = 2 2 c Var x , c , B c x E x , c , A B x 0 0 0 0 A. Braumann, P. L. W. Man, and M. Kraft. Statistical approximation of the inverse problem in multivariate population balance modelling. Ind. Eng. Chem. Res., 49: 428–438, 2010. doi:10.1021/ie901230u

Experimental validation (a) J. Herzler, J. A. Manion, and W. Tsang. Single-Pulse Shock Tube Study of the decomposition of tetraethoxysilane and Related Compounds. J. Phys. Chem. A , 101 , 5500-5508, 1997

Gas-phase mechanism

Reactor Plot Conclusion from kinetic model: Si(OH) 4 is the predominant gas-phase precursor

Particle Model Si(OH) 4 molecules undergo dehydration to form particles. These particles change in size and shape through their lifetime. Particle Particles processes Move in type spce

Type Space ( )

1. Inception Inception increases the number of particles in the system.

2. Surface Reaction Si(OH)4 from gas-phase reacts on a particle surface.

2. Rounding due to surface reaction Surface reaction alters the common surface between different primaries of a particle:

3. Coagulation Particles collide and stick to each other.

4. Intra-particle reaction Adjacent OH sites react with each other to form Si- O-Si bonds.

5. Sintering Sintering calculated on a primary particle-level & alters composition.

Particle model parameter estimation Unknown parameters Parameter space: Objective function: Parameter estimation method: Sobol sequences followed by simultaneous perturbation stochastic approximation Ref (a): T. Seto, A. Hirota, T. Fujimoto, M. Shimada, and K. Okuyama. Sintering of Polydisperse Nanometer-Sized Agglomerates, Aerosol Sci. Tech ., 27 , 422-438, 1997

Particle model parameter estimation Ref (a): T. Seto, A. Hirota, T. Fujimoto, M. Shimada, and K. Okuyama. Sintering of Polydisperse Nanometer-Sized Agglomerates, Aerosol Sci. Tech ., 27 , 422-438, 1997

Particle model validation T = 900 o C T = 1500 o C T = 1750 o C

Simulation for industrial conditions Isothermal Batch Reactor (T = 1500 K, P = 1 atm) Precursor + Fuel +Air Sensitive applications of SiNP require highly specific properties TEOS Formation of Particle formation decomposition Si(OH) 4

Simulation results Mean Diameter 20-40nm Low Standard deviation dc

Simulation results TEOS decomposition rate High surface activity

Desirable process zones • Catalysis / functional materials / fillers

Desirable process zones • Drug delivery / bio-medical applications

Thank you!

Particle size distribution evolution T = 1500 K