SETHI - Hassanizadeh Eccellenza 1 · PDF fileRajandrea SETHI, Tiziana TOSCO, Alberto TIRAFERRI...

1

1Sethi R., Tosco T. www.polito.it/groundwater

Part of an Excellence Ph.D. Course

Politecnico di Torino – June 27th, 2012

A talk on:

Transport of colloids and nanoparticles in saturatedporous media for environmental remediation

Rajandrea SETHI, Tiziana TOSCO, Alberto TIRAFERRI and GROUNDWATER ENGINEERING GROUP

DIATI – Politecnico di Torino

DITAG


Program

� Transport in porous media of:

1. Ideal colloids: latex particles

2. Non-ideal colloids: iron particles for the remediation of contaminated aquifer systems

2


What is a colloid?

� A colloid is a chemical mixture (not a solution) in which one

phase (dispersed phase) is evenly dispersed into another

phase (dispersion medium).

� The colloids has and homogeneous aspect.

� A colloidal system may be solid, liquid, or gaseous.


What is a colloid?

Solid SolExample: glass

GelExamples: cheese

silicagel, opal

Solid FoamExamples: aerogel,

styrofoam, pumice

Solid

SolExamples:

pigmented ink, blood

EmulsionExamples: milk,

mayonnaise, hand cream

FoamExample: whipped

creamLiquid

Solid AerosolExamples: smoke,

cloud, air particulates

Liquid AerosolExamples: fog,

mist, hair sprays

None(All gases are

mutually miscible)Gas

Dispersion Medium

SolidLiquidGas

Dispersed PhaseMedium / Phases

3


Colloids in aquifer systems(porous media)

Anthropogenic or

natural nano- and

micro-particles:

� Chemical

� Farmaceutical

� Fly ashes

� Bacteria

Natural colloids

+ sorbed contaminats

Nanoscale

zerovalent iron

Contaminants

Colloid facilitatedtransport

Remediationtechnologies

(adapted from Freyria, 2007)


Bacteria Clay particles

Natural colloids in aquifersystems

4


Nanoscale iron

15 – 100 nm

Engineered colloids


Transport of colloids in porousmedia

� Mechanisms:

� Advection

� Hydrodynamicdispersion

� Interaction withsolid phase

� Particle-particleinteraction

5


Colloids in porous media: interaction forces

Particle - particle

interactions

Sand-particle interactions

SAND

PORE

WATER

Fluid-particleinteractions


me

ch

an

ism

s

Interaction with porous media and particle-particle:

first

then

Transport of colloids in porousmedia

Rip

en

ing

Str

ain

ing

Blo

ck

ing

CF

T

6


Particle-particle interaction

� Colloidal interactions have different origins:

� Van der Vaals

� Electrostatic

� Magnetic

DLVO & extended DLVO theory

)/141(12 λss

AaVVdW

+−=

s

rESeaV

⋅−= κγζεπε 22

032

( )3

32

0

29

8

+

−=

a

s

aV s

M

ρσπµ


Particle-particle interaction:

Particle – Particleinteractions

7


Sand – particles interaction

� Favorable deposition

(different charge)

� Unfavorable deposition

(same charge)


The presence of ions screens the surface charge of the colloids thusreducing electrostatic repulsion

Unfavorable depositionRole of ionic strength

8


Role of particle size


Colloids in porous media: different approaches

� Microscale

Laboratory tests

Mathematical models

Bradford et al., 2005

Messina, Boccardo, 2012 Icardi, 2012

9


Forces acting on a particle

Drag force

Gravity force

Brownian force

Electric doublelayer

Van der Waalsforce

( )vuFaF ppD

rrr−= ρπ 3

3

4

( )gaF fppG

rrρρπ −= 3

3

4

( )( ) ( )

−

+−

+= −

−

−

h

cp

cp

h

h

cp

prEDL ee

eaF

κ

κ

κ

ξξ

ξξκξξεε

222

22

0 212

r

( )2214

)28(

6 h

h

h

HaF

p

VdW+

+−=

λ

λλr

t

kTRFB

∆=

ξ2

Messina, 2012


Macroscale

10


Macroscopic approach

Mobile particles

Immobile particles

Momentum balance

Flow equation( ) ( ),i

EB

i

qhS s Q Q c s

t xρ

∂∂+ = +

∂ ∂

( ) ( )0 0

0 0

ij om m m

i i j

j

gK s chv q e

x

ρ ρ ρµε

µ µ ρ

−∂= = − + ∂

( )( ) ( ) ( ) ( )f

m m m mol disp

b i ij ij

i i j

cs c s v c D D

t t x x xε ρ ε ε

∂ ∂ ∂ ∂ ∂+ = − + +

∂ ∂ ∂ ∂ ∂

( ) ( ),bs f c s

tρ

∂=

∂


Colloidal transport

� 2 coupled equations:

� Mobile phase

(water)

� Immobile phase

(sand)

( ) ( ) ( )

( )

=∂

∂

∂

∂−

∂

∂

∂

∂=

∂

∂+

∂

∂

scft

s

x

cq

x

cD

xt

s

t

c

b

mm

bm

,ρ

ερε

11


me

ch

an

ism

s

Exchange term

� Different formulations/mechanisms:

Str

ain

ing

Rip

en

ing

Blo

ckin

gL

inear

C

FT

skckd

xd

t

sdbamb

str

ρερ

β

−

+=

∂

∂−

50

50

( ) skcksAt

sdbaripmb

rip ρερβ

−+=∂

∂1

skcks

s

t

sdbamb ρερ −

−=

∂

∂

max

1

skckt

sdbamb ρερ −=

∂

∂

Langmuirian

Tosco, Sethi

Bradford

ckt

samb ερ =

∂

∂

Linear

CFT (irreversible)


General use of ripening equation

( ) skcksAt

sdbaripmb

rip ρερβ

−+=∂

∂1

=

=

0

0

d

rip

k

A

>

>

0

0

rip

ripA

β0=ripA

� TSE can be used to simulate different mechanisms:

clean bed blocking

lineare rev. ripening

**c

k

ks

db

am

ρ

ε=

*

*

*

cAkk

cks

ripamdb

am

ερ

ε

−=

=

−=

1

1

max

rip

rips

A

β

( ) aripm

db

ksA

skc

ripβε

ρ*

*

*

1+=

12


Experimental setupLaboratory scale

Sonicating bath

Peristaltic pump

� 1D system, saturated porous medium

� Colloidal particles: latex, d=2 µm

� Porous medium: sand, d50=250 µm

� Column: � dint = 1.6 10-2 m� L = 0.1 m� n = 0.42� aL = 5.8 10-4 m

� Water:� q = 7.9 10-5 m/s� pH = 7.0� Ionic strength = 0 - 300 mM

Spectrophotometer

column


� Silica Sand

� SEM image Particle Size distribution

Column tests:Materials

0

10

20

30

40

50

60

70

80

90

100

0 100 200 300 400 500

size (um)

po

pu

lati

on

(%

)

13


� Particles: latex microspheres



0 2000 4000 6000 8000 10000 120000

10

20

30

t (s)

Ct (m

M)

0 2000 4000 6000 8000 10000 120000

0.5

1

t (s)

C/C

0 (-)

0 2000 4000 6000 8000 10000 12000-0.5

0

0.5

1

C/C

0 (-)

Constant I. S. Transient I. S.

Column

Input

Column

Output

Available models ?? … New model

Column tests:Design

Sa

lt (

mM

)C

ollo

idC

ollo

id

14


Column tests:Results

Constant IS Variable IS


One site model

� One-site models are usually presented in these forms:

Linear attachment function Langmuirian attachment function

These models can be useful in many cases, but did not proved to be

satisfactory when modeling our column experiments.

−=∂

∂

∂

∂−

∂

∂=

∂

∂+

∂

∂

skcnkt

s

x

cv

c

cnD

t

s

t

cn

dbab

b

ρρ

ρ2

2

−

−=

∂

∂

∂

∂−

∂

∂=

∂

∂+

∂

∂

skcks

sn

t

s

x

cv

c

cnD

t

s

t

cn

dbab

b

ρρ

ρ

max

2

2

1

15


0 1 2 3 4 5 6 7 8 9 100

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Lab data

Stanmod

Matlab

0 1 2 3 4 5 6 7 8 9 100

0.2

0.4

0.6

0.8

1

1.2

1.4

lab data

Stanmod

Matlab

One-site model…

� Can’t fit experimental curves at high IS

F.I. = 1mM F.I. = 100 mM

C/C

0(-

)

C/C

0(-

)

PV (-) PV (-)


� Heteroneous silica sand + feldspar

SEM image Size distribution


0

10

20

30

40

50

60

70

80

90

100

0 100 200 300 400 500

size (um)

po

pu

lati

on

(%

)

16


Two sites model

� Differential equations for the two-interaction-site model:

blocking attachment

function

linear attachment

function

� Model implementation (MNM1D www.polito.it/groundwater):

� Matlab code, finite-differences approach

� Non linear-problem, solved with an iterative scheme

� Code validated with well-established analitycal (Stanmod) and numerical (Hydrus1D) solutions

−=∂

∂

−

−=

∂

∂

∂

∂−

∂

∂=

∂

∂+

∂

∂+

∂

∂

22,2,2

11,1,

1max,

11

2

2

21

1

skcnkt

s

skcks

sn

t

s

x

cv

c

cnD

t

s

t

s

t

cn

dbab

dbab

bb

ρρ

ρρ

ρρ


Colloids in porous media: modeling the influence of I.S.

� MNM1D (Micro-and Nanoparticle transport Modelin porous media in 1D geometry):� Development of the model at CONSTANT ionic strength

� Fitting of the first part of experimental curve (I.S. 1 – 300 mM);

� Definition of semi-empirical relationships for

attachment/detachment coefficients VS ionic strength

� Development of the model in TRANSIENT ionic strength

� Application of MNM1D for the fitting of the whole experimental

curves

17


Colloids in porous media: modeling the influence of I.S

� Fitting of the first part of the breakthrough curves, at constant ionic strength (1 – 300 mM)

C/C

0(-

)C

/C0

(-)




� Fitting of the first part of experimental curve (I.S. 1 – 300 mM);

� Definition of semi-empirical relationships for

attachment/detachment coefficients vs ionic strength


conditions

� Application of MNM1D to the whole test

18


Two sites model:influence of the IS

� Dependence of coefficients on IS:

1,

1

,1

1,

1

a

t

a

a

c

CDC

kk

β

+

= ∞

1,

1

0,1

1,

1

d

CRC

c

kk

t

d

d β

+

=1,

1,1max,

s

ts csβ

α=




� Application for the fitting of the first part of experimental (I.S.

1 – 300 mM);

� Definition of semi-empirical relatioships for

attachment/detachment coefficients VS ionic strength


conditions

� Application of MNM1D to the whole test

19


� A further PDE is added to account for the evolution of IS:


MNM1D

( )( ) ( )

( ) ( )

−=∂

∂

−

−=

∂

∂

∂

∂−

∂

∂=

∂

∂+

∂

∂+

∂

∂

∂

∂−

∂

∂=

∂

∂

22,2,

2

11,1,

1max,

11

2

2

21

2

2

1

sckccnkt

s

sckcckcs

sn

t

s

x

cv

x

cnD

t

s

t

s

t

cn

x

cv

x

cnD

t

cn

tdbtab

tdbta

t

b

bb

ttt

ρρ

ρρ

ρρ

1

1

,1

1

1

a

t

a

a

c

CDC

kk

β

+

= ∞

1

1

0,1

1

1

d

CRC

c

kk

t

d

d β

+

=

1

11maxS

tS csβα=

Conservative tracer


Two sites model + IS dependence

( )( ) ( )

( ) ( )

−=∂

∂

−

−=

∂

∂

∂

∂−

∂

∂=

∂

∂+

∂

∂+

∂

∂

∂

∂−

∂

∂=

∂

∂

22,2,

2

11,1,

1max,

11

2

2

21

2

2

1

sckcckt

s

sckcckcs

s

t

s

x

cq

x

cD

t

s

t

s

t

c

x

cq

x

cD

t

c

tdbtamb

tdbta

t

mb

mbbm

ttm

tm

ρερ

ρερ

ερρε

εε

di

i

t

di

id

CRC

c

kk

β

+

=

1

0,

,

ai

t

i

ai

ia

c

CDC

kk

β

+

= ∞

1

,

,

1

11max,

S

tS csβ

α=

20



� MNM1D (Micro-and Nanoparticle

transport Model in porous media in

1D geometry):

� Development of the model at CONSTANT ionic strength

� Application for the fitting of the first part of experimental (I.S. 1 – 300

mM);

� Definition of semi-empiricalrelatioships forattachment/detachment coefficientsvs ionic strength


� MNM1D can be downloaded from: http://areeweb.polito.it/ricerca/groundwater/software/MNM1D.html



� Fitting of the complete column testI.S. = 1mM

21







22





Solid phase concentration

23




References

� Tosco, T.; Tiraferri, A.; Sethi, R. Ionic Strength Dependent Transport of Microparticles in Saturated Porous Media: Modeling Mobilization and Immobilization Phenomena under Transient Chemical Conditions.Environmental Science & Technology 2009, 43(12), 4425-4431.

� Tosco, T.; Sethi, R. MNM1D: a numerical code for colloid transport in porous media: implementation and validation. American Journal of Environmental Sciences 2009, 5(4), 517-525.

� Tiraferri, A., T. Tosco, e R. Sethi (2010), Transport and Retention of Microparticles in Packed Sand Columns at Low and Intermediate Salinities: Experiments and Mathematical Modeling. Environmental Earth Sciences(submitted) 2010.

� Tiraferri, A.; Chen, K.L.; Sethi, R.; Elimelech, M. Reduced aggregation and sedimentation of zero-valent iron nanoparticles in the presence of guar gum.Journal of Colloid and Interface Science 2008, 324(1-2), 71-79.

� Tiraferri, A.; Sethi, R. Enhanced transport of zerovalent iron nanoparticles in saturated porous media by guar gum. J Nanopart Res 2009, 11(3), 635-645.

SETHI - Hassanizadeh Eccellenza 1 · PDF fileRajandrea SETHI, Tiziana TOSCO, Alberto TIRAFERRI...

Documents

Transcript of SETHI - Hassanizadeh Eccellenza 1 · PDF fileRajandrea SETHI, Tiziana TOSCO, Alberto TIRAFERRI...