Nonoparticles modeling .pdf
Nom original: Nonoparticles_modeling.pdf
Titre: Diapositive 1
Ce document au format PDF 1.5 a été généré par Microsoft® Office PowerPoint® 2007, et a été envoyé sur fichier-pdf.fr le 26/05/2014 à 15:38, depuis l'adresse IP 89.87.x.x.
La présente page de téléchargement du fichier a été vue 483 fois.
Taille du document: 729 Ko (1 page).
Confidentialité: fichier public
Télécharger le fichier (PDF)
Aperçu du document
Reactive-transport mechanisms of manufactured TiO2 nanoparticles in anthropic hydrosystems. Interfacial
physical chemistry, aggregation kinetics, and geochemical reactivity
3nd year PhD student : Izzeddine SAMEUT BOUHAIK(1,2), Supervisor : Lionel Mercury(1), Mohamed Azaroual(2), P. Leroy(2), P. Ollivier(2) and M. A. Sbai(2)
ISTO Orleans Institute of Earth Sciences , 14 Férollerie Street, 45071 ORLEANS Cedex 2
BRGM Water service, 3 Claude Guillemin avenue, Box 36009, 45060 ORLEANS Cedex 2
The reactive-transport processes in water saturated porous media are characterized by their complexity due to coupled and complex physical, chemical, physical-chemical
and kinetic mechanisms expressed at different scales of space and time. To better understand and quantify these processes, we need a thorough understanding of the key phenomena
that occur at each scale of the space. In this study, we consider two space scales. In first, the interface or microscopic scale in which we study the process of aggregation and deposition of
NPs in porous media. And second, the pore or mesoscopic scale in which we study the process of transport coupled with the reactivity and the physical chemical processes at interfaces.
2- Processes at mesoscopic scale
1- Processes at microscopic scale
A- Modeling of the surface potential:
An already existing pore network model (PNM)  will
be adapted to simulate the reactive transfer of NPs in
porous media. Pore volumes are modeled using a
lognormal probability distribution. It is correlated to the
pore-throat hydraulic conductance distribution. Our
calculations are performed on a regular PNM of size
101010 representing a real cubic volume of 111
The zeta potential () is directly predicted
by Extended Stern Model (ESM model, [1,
2]). Zeta potentials calculated by our ESM
model are found to be significantly higher
in amplitude than apparent zeta potentials
estimated by Snoswell and al., ,
because of the strong influence of surface
mobilities of TiO2 NPs (Fig. 1).
Fig. 5. Pore Network model illustration.
Isosurfaces at 2.510-4 ,0 and -2.510-4 Pa
are visualized. Irregular flow patterns are
related to the heterogeneous distribution of
pore-throats hydraulic conductance.
The isosurface at 1.110-12 m2 showing
spatial locations of biggest pores inside
the PNM is visualized
Fig.1: Modeling of zeta potential with the ESM model over a broad salinity
range between 10-4 and 10-1 M KCl .
B- Modeling the interaction energy
The DLVO theory is used to describe the
deposition and aggregation process. The
interaction energy profiles of aggregation
are calculated using the two approaches,
the Derjaguin approximation (DA) and the
surface element integration (SEI) . The
double layer interaction energy is
estimated with linear superposition
Fig. 6. Spatial pore volume distribution in SI units (m3).
Fig. 2. Modeling the interaction energy as a function of the NP-NP
separation distance for a salinity of 10-2 M in KCl .
Fig. 7. Pore pressure distribution inside the 3D PNM.
The isosurface at 0.09 kg/m3 shows
preferential flow paths where nanoparticles
transport phenomena is most active,
favoring higher deposition kinetic rates.
The isosurface at 0.045 kg/m3 shows that a
good correlation exists between preferential
stick and ball sites for surface deposition
C- Modeling the aggregation kinetics
Stability ratios (W) predicted by LSADA, and LSA-SEI are compared to the
measured stability ratios of Snoswell et
al (2005) (see Fig. 3). The DA and SEI
methods predict similar stability ratios ,
except at the lowest ionic strengths
(lower to 10-3 M KCl).
Fig. 8. 3D representation of pore-throat deposits inside the PNM.
Fig. 9. 3D representation of pore deposits inside the PNM.
We also find that, in these physicochemical
conditions, TiO2 NPs aggregation kinetics are
controlled by the local structure of the aggregate
(Fig.4), whereas, at high ionic strengths, when the
interaction range is shorter than the size of the
nanoparticles, TiO2 NPs aggregation kinetics are
controlled by primary particles [1, 5].
Both porosities in the pores and in the
throats connecting them decrease with
the same global rate porosity of the
PNM shows down with an increased
Fig. 3. Modeling the aggregation kinetic as a function of the KCl
In the future, we will introduce the effect of
the surface charge of NPs (electrokinetics)
in the modeling of their reactive transfer.
Fig. 4. Range of interaction compared to primary NP size
Fig. 10. Porosity change during the simulation
timeframe (1 min) due to 10 nm NP transport
and deposition in the PNM.
for the interaction between NPs in aggregate form.
 I. Sameut Bouhaik, P. Leroy, P. Ollivier, M. Azaroual, L. Mercury, J. Colloid Interface Sci. 406(2013).
 P. Leroy, C. Tournassat, M. Bizi, J. Colloid Interface Sci. 356 (2011) 442.
 D.R.E. Snoswell, J.M. Duan, D. Fornasiero, J. Ralston, Int. J. Miner. Process. 78 (2005) 1.
 S. Bhattacharjee, M. Elimelech, M. Borkovec, Croat. Chem. Acta 71 (1998) 883.
 H.C. Schwarzer, W. Peukert, Chem. Eng. Sci. 60 (2005) 11.
 M.A. Sbai, M. Azaroual, Adv Water Resour. 34 (2011) 62–82.