Transcript Diapositiva 1 - Serbian Chemical Society

ESSEE 4
4th European Summer School on Electrochemical Engineering
Palić, Serbia and Montenegro
17 – 22 September, 2006
METHODS OF MEASUREMENTS IN ELECTROCHEMICAL
ENGINEERING
Dr. Manuel A. Rodrigo
Department of Chemical Engineering. Facultad de Ciencias Químicas.
Universidad de Castilla La Mancha. Campus Universitario s/n. 13071
Ciudad Real. Spain.
Department of Chemical Engineering.
Universidad de Castilla La Mancha.
Spain
CONTENTS
1.
CURRENT DISTRIBUTION
1.1 Importance of current distribution visualization
1.2 Measurement of current distribution
1.2.1 TYPES OF MEASURING METHODS
1.2.2 PARTIAL-CELL APPROACH
1.2.3 SUBCELLS APPROACH
1.2.4 SEGMENTED ELECTRODES
1.2.5 RESISTORS NETWORK
1.2.6 PRINTED CIRCUIT BOARD APPROACH
1.2.7 TYPES OF MEASUREMENTS OF THE LOCAL CURRENT IN PASSIVE RESISTOR NETWORK
1.2.8 MATHEMATICAL MODELLING
1.2.9 MAGNETOTOMOGRAPHY
1.3. Some new applications: calculation of mass diffusion overpotential distribution in a PEMFC
2.
MEASUREMENT OF MASS TRANSFER COEFFICIENTS BY ELECTROCHEMICAL TECHNIQUES
2.1 Why?
2.2 How?
2.3 Typical setup for measuring average cell mass transfer coefficients
2.4 Experimental procedure
2.5 Calculation of the mass transfer coefficient
3.
LOCAL MASS TRANSFER DISTRIBUTION
3.1 Importance of local mass-transfer distribution visualization
3.2 Limit current mapping
3.3 Measurement of mass transfer by electrochemiluminiscence
3.4 Mathematical modelling
4.
WALL SHEAR STRESS
4.1Importance of wall-shear stress distribution visualization
4.2 Measurements of wall-shear stress
4.3 Measurement of local shear in three-phase fluidized beds
4.4 Wall shear stress in multiphase flow
1. CURRENT DISTRIBUTION
1.1 Importance of current distribution visualization
It is one of the more important parameters in the performance of an
electrochemical cell, but unfortunately in the electrochemical industry and in
the electrochemical literature, current distribution has not received the
attention that it deserves
I
I
Through the wire flows the
same current, but the current
distribution on the electrode
surface is different
Uniform current
distribution
Non-uniform current
distribution
Some examples of the importance of uniform current distribution
Electroplating: non-uniform current distribution can cause a local variation of the
thickness of the deposited metal
Electrolyses cell
non uniform corrosion of electrodes
Small contactsurface current
feeder
Aluminium surface after an
electro-dissolution process
Poor efficiency, changes in the products conversion ratio
i
Current efficiency 100%
(if reagents arrives to the
electrode at the same rate
that they are consumed)
1/3i
2/3 i
Current efficiency 83.3%
Part of these electrons are consumed by an
electrochemical side reaction because the desired
reactant does not arrive to the anode surface at the
required rate
PEM fuel cell
Uniform current distribution
Produce maximum power densities
Ensure maximum lifetime for the cell
components
Causes of non-uniform current distribution in FC during fuel cell operation:
inhomogeneities in the reactant concentration,
contact pressure,
temperature distribution,
water management along the flow field
etc.
Examples of local current distribution in a circular-shape electrode
uniform
Current scale
high
low
Factors affecting current distribution:
Geometry of the cell system. Current feeders or
collectors
Conductivity of the electrolytes and the electrodes.
Activation overpotentials at the electrodes which
depend on the electrode kinetic.
Concentration overpotentials which are mainly
controlled by the mass transport processes.
Other factors
1.2 Measurement of current distribution
Load or
power supply
electrolyte
electrodes
Purpose of the measurement
a) Current distribution in the cell
anode
cathode
b) Current distribution in one electrode
1.2.1 TYPES OF MEASURING METHODS
Invasive
methods
yes
Partial approaches
Subcells
Segmented electrodes
Passive resistor network
Is the cell modified
for the
measurement?
(is current distribution
measurement associated
with constructional
modifications of the cell? )
no
Mathematical modelling
Non-invasive
methods
Magnetic measurements
1.2.2. PARTIAL-CELL APPROACH.
Portions or segments of the cell are
tested independently by inactivating
other portions.
electrode
electrolyte
electrode
The inactivation can be carried out either by masking or by other procedure (e.g. in
FC some parts of the MEA can be prepared without catalyst
subcell 3 inactive
subcell 1 inactive
To increase the accuracy more
partial cells should be studied
CELL VOLTAGE
Subcell 1 Subcell 2 Subcell 3
the specific performance is
determined by difference.
whole cell
Subcell 1 inactive
Subcell 3 inactive
INTENSITY
Advantages: very simple, easy to
manufacture
Disadvantages: it can only be
used as a first approach
1.2.3. SUBCELLS APROACH
Several electrically isolated subcells
are placed are conveniently placed
at different locations in the cell
a section of the anode is punched out
a section of the cathode is punched out
The step is repeated in several
determined locations inside the cell
The former anodes and
cathodes are replaced with
smaller ones.
Main cell
subcells
The resulting
empty space is
filled with a
isolating gap
The subcells are separately controlled.
To measure current distribution every
subcell voltage has to be adjusted to
fit approximately the mail cell voltage
SUBCELL 5
MAIN CELL
SUBCELL 3
INTENSITY
Advantages
Gives more information on a
much smaller scale about the
localised current density than
the partial approach
Disadvantages
Complex manufacture. Great care
has to be taken to ensure proper
alignment during assembly of the
cell
L1
Ln
Lm
Main cell
Subcell m
Lmain cell
Subcell n
CELL VOLTAGE
Subcell 1
1.2.4. SEGMENTED ELECTRODES
Measurement circuits
Segmented electrode or
segmented BPP (in a FC)
isolation
electrolyte
Counter electrode
This approach allows a
very accurate current
distribution mapping
Coverage of the whole electrode area
Good spatial resolution
Example of measuring device for each piece of electrode
ohmic resistor
Volt-meter
Piece of electrode
To assume a high ratio between
through-plane and in-plane
conductivity segmented electrodes
must be manufactured in a thin
shape. This generates problems
related to mechanical strength
Very invasive method. It can affect significantly to the current
distribution. Big differences can exist between the measure and the
actual current distribution
1.2.5 RESISTORS NETWORK
Buss plate
Passive resistor network
electrode
Main problem - appearing of lateral currents
Main advantage: It does not
require any modification of the
electrodes (or of the BPP or
MEA in FC)
It is less invasive
Coverage of the whole electrode area
Good spatial resolution
current
Volt-meter
Drawbacks
Electrical properties of the resistors depends on temperature
Completely isolated resistors
Buss plate
Isolated wires
Resistor matrix
electrode
Advantages
Improved mechanical
strength
To assume a high ratio between through-plane and in-plane
conductivity resistor matrix must be manufacture in a thin
shape. This generates problems related to mechanical strength
interconnected resistors
Buss plate
Isolated wires
Resistor matrix
electrode
Advantages
Less affected by in-plane
current distribution
1.2.6 PRINTED CIRCUIT BOARDS APPROACH
Current collector
current
backside
Through-holes
frontside
Easy to manufacture
Possibility of multilayer manufacture
Easy to add electrical components
Can be used as BPP in FC
1.2.7 TYPES OF MEASUREMENTS OF THE LOCAL CURRENT IN PASSIVE
RESISTOR NETWORK
Ohmic resistors
passively
Hall-effect sensors
(only measure)
Current transformers
actively
Multichannel potentiostats
(Measure and
manipulation)
Ohmic resistors
current
Volt-meter
 Very simple
 Frequently used
 Very invasive. It can affect the cell current distribution
Hall-effect sensors
When a current-carrying conductor is placed into a magnetic
field, a voltage will be generated perpendicular to both the
current and the field. This principle is known as the Hall effect.
The figure shows a thin sheet of semiconducting
material (Hall element) through which a current is
passed. The output connections are
perpendicular to the direction of current. When no
magnetic field is present, current distribution is
uniform and no potential difference is seen
across the output.
When a perpendicular magnetic field is present,
a Lorentz force is exerted on the current. This
force disturbs the current distribution, resulting in
a potential difference (voltage) across the output.
This voltage is the Hall voltage (VH). Its value is
directly related to the magnetic field (B) and the
current (I).
Hall effect sensors can be applied in many types of
sensing devices. If the quantity (parameter) to be
sensed incorporates or can incorporate a magnetic
field, a Hall sensor will perform the task
Current follower circuit
Standard operational
amplifier circuit for currentto-voltage conversion
To working electrode
+
+
For very low currents
To data acquisition card
1.2.8 MATHEMATICAL MODELLING
Current
distribution
model
e.g.
simulation
New proposal
Modelled results
no
Experimental
conditions
Agreement?
experiments
experimental results
e.g. product conversion
yes
1.2.9 MAGNETOTOMOGRAPHY
Non invasive method
xcellvision Instrumentation
for Fuel Cells
and Fuel Cell
System
Simulation
Patented technology
z
x
y
Sensor 1
Sensor 2
Sensors are used for magnetic field data
acquisition as a function of the position. The
experimental setup allows the sensor to
measure the magnetic field strength (H) at
different positions around the cell
Ij
Hi
high
low
 H1   a 11

 
 H2  
 ...   

 
 H n 1  
 H  a
 n   n1
...
a 1m   I1 
 

  I2 
   ... 
 

  I n 1 
a nm   I n 
Map of the current intensity
high
low
2. MEASUREMENT OF MASS TRANSFER
COEFFICIENTS BY ELECTROCHEMICAL TECHNIQUES
2.1 Why?
Bulk
solution
Electrode
surface
high
current
Concentration of the electroactive species
low
influence the
current
distribution
Affect to the product distribution
Affect to the efficiency
e-
2.2 How?
relectrochem 
Electrode
Ssurface
R
j·A
n·F
Sbulk
rmasstransfer  k mA(Sbulk  Ssurface )
The method is based on a diffusion-controlled reaction at the
electrode surface:
3
6

Fe(CN )  e  Fe(CN )
If the
cathode is
used as a
probe
4
6
Typical concentration 5 mM of ferrocyanide and
20mM of ferricyanide to make sure a cathodic
controlled electrochemical process
The area of the anode should be larger than that
of cathode for a cathodic controlled-process
A large quantity of inert electrolyte (NaOH, Na2SO4, KSO4, …) has to
be added as supporting electrolyte to minimize the migration effects (to
make them negligible compared to diffusion and convection)
2.3 Typical setup for measuring average cell mass transfer coefficients
The reservoir
contains the
electrolyte
The electric measurement
devices are used to obtain high
accuracy of voltage and
current values, than those
provided by the power supply.
The electrical energy is applied
with the power supply
connected to the electrodes
Oxygen and hydrogen
generated in the
electrochemical cell can
be stripped with
nitrogen.
V
A
The flow rate is
measured by the
rotameter.
The pump propels the electrolyte through
the electrochemical cell.
The heat exchanger keeps the
electrolyte temperature at the
desired set point.
The heterogeneous
processes take place in
the electrochemical cell,
where mass transfer
processes are studied.
I
0
Concentration
Cb
0
Current measured
2.4 Experimental procedure
Distance from
the electrode
Applied potential
V
0
a) No potential is applied to cell. No current
0
0
Current measured
I
Concentration
Cb
Distance from
the electrode
Applied potential
V
0
a) Small potential is applied to cell. No current
Concentration
0
0
Current measured
I
Cb
Distance from
the electrode
Applied potential
V
0
b) Potential scan begins
0
I
0
Current measured
Concentration
Cb
Distance from
the electrode
Applied potential
V
0
I
0
Concentration
Cb
0
Current measured
I limit
Distance from
the electrode
Applied potential
V
0
c) Current limit is reached
0
I
Concentration
Cb
0
Current measured
I limit
Distance from
the electrode
Applied potential
V
0
d) Plateau zone
I
0
Concentration
Cb
0
Current measured
I limit
Distance from
the electrode
Applied potential
V
0
I
0
Concentration
Cb
0
Current measured
I limit
Distance from
the electrode
Applied potential
V
0
e) Other electrochemical processes
(e.g. Electrolyte decomposition)
0
I
Concentration
Cb
0
Current measured
I limit
Distance from
the electrode
Applied potential
V
0
2.5 Calculation of the mass transfer coefficient
e-
relectrochem 
jlim ·A
n·F
Electrode
Ssurface=0
R
relectrochem  rmasstransfer
Sbulk
rmasstransfer  k mA(Sbulk  Ssurface )  k m ·A·Sbulk
jlim ·A
 k m ·A·Sbulk
n·F
jlimit
km 
n  F  Sbulk
3. LOCAL MASS-TRANSFER DISTRIBUTION
3.1 Importance of mass-transfer distribution visualization
Why?
Mass transfer
greatly influence
current distribution
Mass transfer can be easily
improved in a cell by using
turbulence promoters
Local mass transfer distribution can depend on a lot of factors:
Design of the inlet
Design of the outlet
Flow characteristics
Turbulence promoters
Smooth or uneven surfaces
…
How?
By measuring the limit current at different positions on the electrode
By using other techniques
3.2 Limit current mapping
cathode
Push-button
switch
anode
Power
supply
V
voltmeter
A
A
A
ammeter
Arrays of microelectrodes
Drawback  many measuring sites
Corner plate
centre
Total current
Measuring
device
Current of the main electrode
3.3 Measurement of mass transfer by electrochemiluminescence
+ N2 + light
Direct
electrolyses
H2O2
Direct electrolyses
Very slow rate
Iridium tin dioxide
electrode
HO
2
2
3.4 Mathematical modelling
Mass transfer
distribution
model
e.g.
simulation
New proposal
Modelled results
no
Experimental
conditions
Agreement?
experiments
experimental results
e.g. product conversion
yes
4. WALL SHEAR STRESS
4.1 Importance of wall-shear stress distribution visualization
Theories of wall turbulence
considers the existence and
interaction of turbulent bursts,
ejections, sweeps and wall
streaks. A turbulent bursts is an
ejection of fluid from the wall,
which also causes fluid to impige
on the wall by simultaneous
formation of sweeps, or
movement of fluids towards the
wall. Turbulent bursts and
sweeps occur through the
formation of vortices and the liftup of wall streaks.
Analyses of mass flux fluctuations
I limit  n· f ·J
I limit(t )  I limit  I limit (t )
'
In the diffusion regime Faraday’s law
allows to link the mass flux to the wall of
electroactive ions (J) to the limit current
Statistical analyses of this parameter
allows to obtain important information
concerning the turbulent transfer
characteristics within the viscous
sublayer
Turbulent flow visualization
Traditional methods
Laser doppler anemommetry
Particle imaging velocimetry
Thermoanemometry
Electrochemical method
Main
advantage
Information about the
wall turbulence in the
viscous sublayer
 u 
  · 
 y 
Oxide layer
metal
Schematic description of initiation of flow induced localized
corrosion phenomena
4.2 Measurements of wall shear stress
flow
u(t)
cathode
A small dimension probe
allows the measurement
of only a local velocity
gradient which can be
related to local wall shear
stress.
anode
Diffusion
boundary
layer
Viscous
boundary
layer
The electrochemical method is
based on measurement of mass
transfer coefficients. This
coefficients are related to
velocities in the proximity of the
probes
dH
u(t)
dN
microelectrode
I limit  a  1/ 3
This method can be applied with high
resolution using microelectrodes or
microelectrodes arrays incorporated flush
and isolated into flat surfaces exposed to
tangential flows
The time-dependent diffusion
limited current density
correlates with the timedependent gradient of the
streamwise flow velocity
perpendicular to the wall
which is proportional to the
wall shear stress
dH
dN
c  c
c, concentration of the
electroactive species
microelectrode
 u 
S   
 y  y  0
   S
Local wall shear gradient
For a newtonian fluid with dynamic
viscosity  the wall shear stress can
be expressed
I limit  ( D 2 S )
1
3
D S

I  0.8075·n·F ·A·c ·
 l 
2
2
3

For a steady-state flow, the small electrode
mounted flush with the insulating wall delivers a
current I. This measured intensity increases with
the applied potential between the two electrodes
until the process becomes controlled by the
diffusion of the reacting species to the surface of
the working electrode. Then the value of the
intensity is the limiting current. The probes
behaves as a perfect mass sink
1
3
I  0.8075·n·F ·L·l ·c · D 2 S
Levêque formula (valid for a circular
electrode of area A)

1
3
Extension of the Levêque formula for a non
circular electrode: L length of the electrode in
the flow direction (m), l length of the
electrode transverse to the flow direction (m)
D, diffusion coefficient (m2s-1), n number of electrons exchanged in the
electrode reaction, F Faraday constant (96500 C/mol)
The single wall probe is applicable only for
nonreversing conditions
If flow reversal occurs in the
proximate wall flow region and
additional information about the
flow direction is needed a
“sandwich probe” should be used
The sandwich probe
consists of two active
segments separated in the
mean flow direction by a thin
insulating gap
The size of this probe should be equal or
smaller to the typical size of the large
flow structures to ensure homogeneity
x
z
Photolithography probes
Counter electrode
100 m
X velocity component i1 + i2
To current followers
Z velocity component
i1
i2
Insulating gap
i1 - i2
4.3 Measurement of local shear in three-phase fluidized beds
Plastic sphere
Support rigid tube
Gold wire
4.4 Wall shear stress in multiphase flow
Bubble flow
Slug flow
Gas slug
-1
i(A)
Annular flow
Gas slug
Gas current limit
-2
-5
Liquid current limit
Current collectors
Shunt resistors
Conductive layer (backside of the
PCB)
Cooling channels
Printed circuit board
Anodic BPP
load
MEA +GDL
Cathodic BPP
Shunt resistors are integrated into the PCB using a multilayer design
PCB can be easily manufactured in a way that guaranties the compatibility with the elements
of the cell
High flexibility to modular configuration (the same PCB can be used to study different
configurations of the cell)
The sense wires associated with the individual resistors can be integrated into the PCB and
connected to the data acquisition system from the edge of the PCB
The invasive method does not affect to the fluid dynamic properties of the reactant gases and
the electrical and thermal conductivity of the cell are not importantly modified.
PCB can be introduced inside a BPP. This enable to measure current distribution in a stack
1.3. Some new applications: calculation of mass diffusion overpotential distribution
in a PEMFC
UNIFORM OXYGEN
CONCENTRATION
OF OXYGEN ON THE
CATHODE BY FLOW
PULSE APROACH
AND SEGMENTEDELECTRODE
APROACH
V
hW
Cell potential
ea + h
hdiff
CURRENT
INTERRUPTION
METHOD
hW
ea + h + hreaction
hW
Electrolyte
ANODE
CATHODE
In PEMFC uneven
current distribution are
caused by non uniform
oxygen distribution
inside the fuel cell
Direction of
charge flux
CURRENT
DISTRIBUTION
MEASUREMENT
WITH UNIFORM
OXYGEN
CONCENTRATION
CELL
RESISTANCE
MATHEMATICAL
MODEL
MASS-DIFFUSION
OVERPOTENTIAL
DISTRIBUTION
CONDITIONS
Cell operated galvanostatically
For each current the cell was allowed to stabilize and then the
current distribution was measured
A oxygen flow pulse of 10 s is introduced and the current distribution
is measured again
To ensure that the oxygen concentration along the reaction surface is
uniform, the flow pulse has to be strongly over stoichiometric and long
enough to remove all excess water from the electrodes. At the same
time the duration of the flow pulse must be short enough in order not to
change the resistance of the proton conductive phases of the MEA
E  E0  b ln(i j )  rji j  hconc, j
E0  Erev  b ln(i0,c )
MATHEMATICAL MODEL
Ehom  E0  b ln(i hom, j )  rji hom, j
i
hconc, j  E hom  E  b ln( hom, j )  rj (i hom, j  i j )
ij