• Aucun résultat trouvé

Liquid-liquid extraction of two radiochemical systems at micro-scale predict and achieve segmented flow to optimize mass transfer

N/A
N/A
Protected

Academic year: 2021

Partager "Liquid-liquid extraction of two radiochemical systems at micro-scale predict and achieve segmented flow to optimize mass transfer"

Copied!
20
0
0

Texte intégral

(1)

HAL Id: cea-02438369

https://hal-cea.archives-ouvertes.fr/cea-02438369

Submitted on 14 Jan 2020

HAL is a multi-disciplinary open access

archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Liquid-liquid extraction of two radiochemical systems at

micro-scale predict and achieve segmented flow to

optimize mass transfer

A. Vansteene, J. Jasmin, R. Brennetot, C. Mariet, S. Cavadias, G. Cote

To cite this version:

A. Vansteene, J. Jasmin, R. Brennetot, C. Mariet, S. Cavadias, et al.. Liquid-liquid extraction of two radiochemical systems at micro-scale predict and achieve segmented flow to optimize mass transfer. BIT’s 5th Annual Conference of AnalytiX 2017 (AnalytiX-2017), Mar 2017, Fukuoka, Japan. �cea-02438369�

(2)

«LIQUID-LIQUID EXTRACTION OF TWO RADIOCHEMICAL SYSTEMS AT MICRO-SCALE: PREDICT AND ACHIEVE SEGMENTED FLOW TO

OPTIMIZE MASS TRANSFER»

| PAGE 1

Axel Vansteene, J.P. Jasmin, René Brennetot, Clarisse Mariet 1

1Den – Service d’Etudes Analytiques et de

Réactivité des Surfaces (SEARS), CEA, Université Paris-Saclay, F-91191, Gif sur Yvette,

France

Siméon Cavadias, Gérard Cote2

2PSL Research University, Chimie ParisTech

-CNRS, Institut de Recherche de Chimie Paris, 75005, Paris, France

(3)

OVERVIEW : RADIOCHEMICAL ANALYSIS

Current nuclear procedures :

• Separation and purification is needed before detection

• Hardly implementable in glove boxes • Huge volumes of solvents

Radiochemical

issues

Waste (solvents,

extractants)

| PAGE 2

Microfluidics:

Manipulate fluids at micro-scale i.e. one dimension of the analytical device is below 100 µm [1]

(REACH)

Volumes Analysis time

Operator exposure

Costs

Classical fluid

dynamics

A solution: process intensification

(4)

 Easy retrieval of the two phases  Diffusion-limited

 Set specific interfacial area (depending on the chip)

LIQUID-LIQUID EXTRACTION MINIATURISATION (µ-LLE)

Kagawa, Talanta, 2009, 79, 1001 Ralston, ISEC Conference, 2011

Assets

• Analysis automation and parallelization • Possible coupling with detection devices

Phase 1

Phase 2

Phase 2 Phase 1

Two types of biphasic flows

 Convection

 Adjustable specific interfacial area  Phase separation to be performed

Parallel flows (stratified flows)

Segmented flow

Suitable for all chemical systems Non-suitable for slow kinetics systems

(5)

Comparizon of 2 chemical systems in the same microchip

[2] Coleman et al., AIME Annual Meeting, 1979, New Orleans, LA, USA [3] Weigl et al., Solv. Ext. Ion Exch., 2001, 19, 215-229

U(VI) / Aliquat® 336

Eu(III) / DMDBTDMA

Quick kinetics

[2]

Slow kinetics

[3]

[U(VI)]= 10-5 M

[HCl]= 5 M

Aqueous phase:

[Aliquat® 336]= 10-2 M in

n-dodécane/ 1-décanol 1% (v/v)

Organic phase : Aqueous phase : Organic phase :

[Eu(III)]= 10-2 M

[HNO3]= 4 M

[DMDBTDMA]= 1 M

n-dodécane

R

U,batch, optimal

= (85.2 ± 1.2) %

for Vaq= Vorg

Viscosity ratio

μ

org

/ μ

aq

≈ 1.2

R

Eu,batch,optimal

= (90.1 ± 0.3) %

for Vaq= Vorg Viscosity ratio

μ

org

/ μ

aq

≈ 14

| PAGE 4

(6)

PHD AIMS AND OBJECTIVES

Optimize the specific interfacial area (A/V) =

𝐼𝑛𝑡𝑒𝑟𝑓𝑎𝑐𝑖𝑎𝑙 𝑎𝑟𝑒𝑎

𝑀𝑖𝑐𝑟𝑜𝑐ℎ𝑎𝑛𝑛𝑒𝑙 𝑣𝑜𝑙𝑢𝑚𝑒

 Droplets volume : 𝑉𝑝𝑙𝑜𝑡 = 𝑓 𝑝ℎ𝑦𝑠𝑖𝑐𝑜𝑐ℎ𝑒𝑚𝑖𝑠𝑡𝑟𝑦, ℎ𝑦𝑑𝑟𝑜𝑑𝑦𝑛𝑎𝑚𝑖𝑐𝑠, 𝑐ℎ𝑖𝑝 𝑔𝑒𝑜𝑚𝑒𝑡𝑟𝑦  Droplets frequency 𝑓 = 𝑄𝑑

𝑉𝑝𝑙𝑜𝑡

 Spacing between consecutive droplets 𝑒 = 𝑄𝑐+𝑄𝑑 ℎ𝑤𝑜𝑓

Determine the segmented flow (i.e. droplets population)

characteristics, in order to figure out the specific interfacial area

Physicochemistry

η

𝑖

, σ

Hydrodynamics

𝑄

𝑖

Chip geometry

𝐽𝑢𝑛𝑐𝑡𝑖𝑜𝑛 𝑡𝑦𝑝𝑒 (𝑇, 𝐹𝐹), 𝑑𝑖𝑚𝑒𝑛𝑠𝑖𝑜𝑛𝑠

(7)

JUNCTION TYPE

Which junction best suits our needs?

| PAGE 6

 Available equations for every flow regime

 Squeezing, transition regime, and dripping regimes to be studied

 Available equations for every flow regime

 Squeezing, transition regime, and dripping regimes to be studied

 Very few models in the litterature T-Junction

Focalized Flux (FF)

Co-current Flux

Will only be presented in the following our results concerningthe FF junction

► Flow regimes to be chosen

(8)

FLOW CARTOGRAPHY

– FF JUNCTION

w

c

w

c

w

d 𝒘𝒅 = 𝒘𝒐𝒓 = 𝒘𝒄 = 𝑯  Squeezing  Dripping Available equations :

 Liu and Zhang model [4]  Cubaud and Mason model [5]

[4] Liu et al., Physics of Fluids, 2011, 23, 8 [5] Cubaud et al., Physics of Fluids, 2008, 20, 5

(9)

EXPERIMENTAL SET-UP

Corrosive chemicals (Acids, solvents)

Hydrophilic surface, suited for oil in water segmented flow

• Glass chip (Dolomite, UK)

Dolomite® Pumps

Syrris®

Membrane phase separator Continuous aqueous phase [Eu(III)]= 10-2 M [HNO3]= 4 M To-be-dispersed organic phase [DMDBTDMA]= 1 M n-dodecane | PAGE 8 Microchannel dimensions: Width : 300 μm Depth : 100 μm Sketch of the 100 μm ID hydrophilic FF-junction chip

(10)

ACQUISITION METHOD FOR DROPLETS POPULATION

CHARACTERISTICS

Droplets morphometry and velocimetry analysis

[6]

10.000 fps acquisition – 94 ms Played back at 30 fps

Slowed down by a factor >300 Number of droplets analysed: 31

Experiments performed on 2016/11/22 with phase separation– PHNO3= 1280 mPa– PDMDBTDMA= 1180 mPa

Droplets diameter

Droplets velocity

Droplets spacing

SOFTWARE TREATMENT RAW VIDEO

(11)

VALIDATION OF THE DRIPPING MODEL

| PAGE 10

Results comparison with Cubaud et al. theoretical model

[5]

From [5] Cubaud et al., Physics of Fluids, 2008, 20, 5

Theoretical and experimental comparison of the droplets populations characteristics generated in a FF-junction in the dripping regime, for the following chemical system : [Eu(III) ]= 10-2M – [HNO

3]=4M /[DMDBTDMA] 1M –

n-dodecane

Predicted volumes and frequencies

= Hydrodynamics control

(12)

MASS TRANSFER STUDY

Mass transfer is only ruled by reaction kinetics

[7] Launière, Gelis, ACS, 2016, 55, 2272-2276

𝑡 → +∞ 𝑡ℎ𝑒𝑛 𝐸% → 𝐸𝑏𝑎𝑡𝑐ℎ

𝐸𝑢

3+

+ 3𝑁𝑂

3

+ 2𝐷𝑀𝐷𝐵𝑇𝐷𝑀𝐴 →

𝐸𝑢 𝑁𝑂

3 3

. (𝐷𝑀𝐷𝐵𝑇𝐷𝑀𝐴)

2

𝑘

𝑎𝑜

𝑘

𝑜𝑎

𝐸% 𝑡 = 𝐸

𝑏𝑎𝑡𝑐ℎ

(1 − 𝑒

−𝐴𝑉 1+𝐾1𝑑 𝑘𝑎𝑜𝑡

)

𝐾

𝐷

=

𝐶

𝑜𝑟𝑔,𝑒𝑞

𝐶

𝑎𝑞,𝑒𝑞

=

𝑘

𝑎𝑜

𝑘

𝑜𝑎

With segmented flows, diffusion is not a limiting factor in mass transfer:

The regime is called « kinetic » [7]

(13)

E% 𝑡 = 𝐸

𝑏𝑎𝑡𝑐ℎ

(1 − 𝑒

−𝐴𝑉 1+𝐾1𝑑 𝑘𝑎𝑜𝑡

)

MASS TRANSFER STUDY

Composition of the extraction yield

| PAGE 12

The

extraction

yield

is

dependent on the volume

ratio of the two phases.

A/V= 1000 m-1

A/V=10 m-1

V

aq

/V

org

=1

And

on

the

specific

interfacial area

0 10 20 30 40 50 60 70 80 90 100 0 20 40 60 Ebat c h (% ) (e rr o r b a rs a re d is p la y e d ) Vaq/Vorgratio Experimental results y = -19,81ln(x) + 95,959 R² = 0,9886

(14)

Mass transfer results currently being validated

Extraction yields are slightly superior (~5%) to those expected theoretically, due to a small uncertainty on contact times.

MASS TRANSFER CASE STUDY: EU(III) EXTRACTION BY

MALONAMIDE DMDBTDMA

𝐸

𝑏𝑎𝑡𝑐ℎ

= 𝐸

= 𝑓(

𝑉

𝑎𝑞

𝑉

𝑜𝑟𝑔

)

Dripping regime, Dolomite® FF junction, [Eu(III) ]= 10-2M – [HNO

3]=4M /[DMDBTDMA] 1M - dodecane

K

d

= 9.1 ± 0.3

k

ao

~ (5.9 ± 0.7).10

-5

m/s

[8] E%

(15)

CONCLUSION

| PAGE 14

1. Factual background: the choice of junctions and flow regimes

Focalized flux junction

T-junction

Squeezing

Dripping

2. Development of an observation method for segmented flow characterization

 Droplets size

 Droplets frequency  Droplets velocity

 Spacing between droplets

 Quick and exhaustive analysis of any segmented flow

3. Validation of theoretical equations : produce droplets with desired

characteristics

Used flow rates

Droplets characteristics

a. Validation of equations

(16)

0 2 4 6 8 10 0 20 40 60 Co nc e nt ra tion f a c to r Vaq/Vorg 0 10 20 30 40 50 60 70 80 90 100 0 20 40 60 E% Vaq/Vorg

PERSPECTIVES

The smaller the Vaq/Vorg

ratio, the higher the extraction yield

Analysis

Process

FF-junction chip to be optimized

Same methodology to be developed with T-junctions chips

The whole approach was based on one particular chemical system :

HNO3 4M – Eu 10-2M / Dodecane – DMDBDTDMA 1M

Slow kinetics, η𝑜𝑟𝑔

η𝑎𝑞 ~15

The higher the Vaq/Vorgratio, the higher the concentration factor 𝐶𝐶𝑜𝑟𝑔,𝑓

(17)

PERSPECTIVES

| PAGE 15

Have a generic approach towards mass transfer, independent on the used junction or the chemical system

COMSOL (CFD) model being developed with Chimie Paris-Tech (Pr. Cavadias and Pr. Cote):

- mass transfer model between a droplet and an external phase being tested

«

(18)

Liquid-Liquid Extraction of two Radiochemical Systems at Micro-Scale: Predict and Achieve Segmented Flow to Optimize Mass Transfer

AnalytiX-2017, March 22-24, 2017, Fukuoka (Japan)

ORALS

POSTERS

PAPERS

A Simple and Adaptive Methodology to use Commercial Microsystem as Screening Tool: Validation with the U-TBP Chemical System

Solvent Extraction Ion Exchange

Liquid-Liquid microflow patterns of two radiochemical systems used in the nuclear field: predict the formation of segmented flow

RANC 2016, April 10-15, 2016, Budapest (Hungary)

Predict and compare the formation of segmented flow in microsystems : Interest for radiochemical liquid-liquid extraction

(19)

FORMULAE

– CROSS JUNCTIONS

Model

Regime

Formula

Liu Transition Cubaud Fu Dripping 𝑙𝑝𝑙𝑜𝑡 ℎ ≈ 2.2. 10−4 1 1 + 𝜙𝐶𝑎𝑐 −1 𝑝𝑜𝑢𝑟 𝑙𝑝𝑙𝑜𝑡 ℎ > 2.5 0.5 1 1 + 𝜙𝐶𝑎𝑐 −0,17 𝑝𝑜𝑢𝑟 𝑙𝑝𝑙𝑜𝑡 ℎ < 2.5 𝑙𝑝𝑙𝑜𝑡 ℎ ≈ 0.3𝜙0.23𝐶𝑎𝑐−0.42𝑝𝑜𝑢𝑟 𝑙𝑝𝑙𝑜𝑡 ℎ > 2.35 0.72𝜙0.14𝐶𝑎 𝑐−0.19𝑝𝑜𝑢𝑟 𝑙𝑝𝑙𝑜𝑡 ℎ < 2.35 Cubaud Jetting 𝑑 ℎ ≈ 2.19 𝜙 | PAGE 18 𝑙𝑝𝑙𝑜𝑡 𝑤𝑐 = ( 𝜀 + 𝛼 𝑄𝑑 𝑄𝑐)𝐶𝑎𝑐𝑚 𝜀 = 0.32, 𝛼 = 0.219 𝑒𝑡 𝑚 = −0.243

(20)

MASS TRANSFER STUDY

𝐸𝑢3++ 3𝑁𝑂

3− + 2𝐷𝑀𝐷𝐵𝑇𝐷𝑀𝐴 → 𝐸𝑢 𝑁𝑂3 3. (𝐷𝑀𝐷𝐵𝑇𝐷𝑀𝐴)2

Mass transfer is only ruled by reaction kinetics

hence

[ [5] Launière, Gelis, ACS, 2016, 55, 2272-2276

With Kd= 9.1 ± 0.3 and kao~ (5.9 ± 0.7).10-5 m/s [7]

𝑡 → +∞ 𝑡ℎ𝑒𝑛 𝑅𝑒𝑥𝑡𝑟𝑎𝑐𝑡𝑖𝑜𝑛 → 𝑅𝑏𝑎𝑡𝑐ℎ 𝑘𝑎𝑜

𝑘𝑜𝑎

𝑅

𝑒𝑥𝑡𝑟𝑎𝑐𝑡𝑖𝑜𝑛

𝑡 = 𝑅

𝑏𝑎𝑡𝑐ℎ

(1 − 𝑒

−𝐴𝑉 1+𝐾1𝑑 𝑘𝑎𝑜𝑡

)

With segmented flows, diffusion is not a limiting factor in mass transfer: The regime is called « kinetic »[5]

𝐾𝐷 = 𝐶𝐶𝑜𝑟𝑔,𝑒𝑞 𝑎𝑞,𝑒𝑞 = 𝑘𝑎𝑜 𝑘𝑜𝑎

𝑅

𝑒𝑥𝑡𝑟𝑎𝑐𝑡𝑖𝑜𝑛

𝑡 =

𝐶𝑎𝑞,0𝐶−𝐶𝑎𝑞(𝑡) 𝑎𝑞,0 Yet

Références

Documents relatifs

ARC Association pour la Recherche sur le Cancer ; 2001 : Subvention équipement lourd co-financement du microscope confocal: Rôle des canaux potassiques et du calcium dans

The limited values of Tafel slopes measured at low overpotential (LO) for the small particle based electrodes suggest the electrodesorption (Heyrowsky step (4)) as the limiting

[r]

case const1: instruction break; case const2: instruction break; case const3: instruction break; ... default:

Unite´ de recherche INRIA Lorraine, Technopoˆle de Nancy-Brabois, Campus scientifique, 615 rue du Jardin Botanique, BP 101, 54600 VILLERS LE` S NANCY Unite´ de recherche INRIA

Les cinq entités d’hémopathies malignes les plus fréquemment diagnostiquées dans le département de l’Isère depuis l’ouverture du registre du cancer en 1979, sont

Pour ce qui est des entreprises qui bénéficient de ce « capital humain » à bon marché, l’espoir d’un changement dans les pratiques des entreprises demeure

Dans un premier temps, il ne s’agit pas de trancher entre ludologie et narratologie, mais de se centrer sur la dimension narrative des jeux vidéo, en essayant de déterminer le