Upload
others
View
5
Download
0
Embed Size (px)
Citation preview
A research partnership
between Boise State
University, Idaho
National Laboratory,
Idaho State University
and University of Idaho.
Center for Advanced Energy Studies
Experimental Studies of OxygenSparging in Molten Salt Through aTransparent Furnace
Experimental Studies of OxygenSparging in Molten Salt Through aTransparent Furnace
Ryan W. Bezzant and Supathorn PhongikaroonChemical Engineering Department and Nuclear
Engineering Program
University of Idaho, Idaho Falls
Center for Advanced Energy Studies
Michael F. SimpsonPyroprocessing Technology Department,
Idaho National Laboratory
Ryan W. Bezzant and Supathorn PhongikaroonChemical Engineering Department and Nuclear
Engineering Program
University of Idaho, Idaho Falls
Center for Advanced Energy Studies
Michael F. SimpsonPyroprocessing Technology Department,
Idaho National Laboratory
Alternative Proposed ProcessesChopped Fuel
Uranium MetalMetal Waste
Ceramic Waste
Electrorefiner andProduct
Refinement
Ion Exchange Zone Freezing
High PurityElectrolyte
High PurityElectrolyte
Contaminated Electrolyte
Rare Earth Oxidative Precipitations
• MotivationThere is a lack of fundamentalstudy to provide insight intosystematic parameters in moltensalt
• ApproachUse transparent cell, high speedcamera and O2 sensor to obtainconcentration data, bubbledistribution and reaction rates.
Motivation and Approach
23223
2223
223
Cl2
3ORE
2
1ORECl3
Cl2
3REOORECl2
ClREOClO2
1RECl1
Korea Atomic Energy
Research Institute (KAERI)
observed gravity separation
of oxides in molten salt.1
1Yung-Zun Cho, Gil-Ho Park, Dae-Seok Han, Han-Soo Lee, and In-Tae Kim. Journal of Nuclear Science and Technology, 46,2009, 1004-1011;Yung-Zun Cho, Hee-Chul Yang, Gil-Ho Park, Han-Soo Lee, In-Tae Kim. Journal of Nuclear Materials, 384, 2009, 256-261;Yong-Jun Cho, Hee-Chul Yang, Hee-Chul Eun, Eung-Ho Kim, and In-Tae Kim. Journal of Nuclear Science and Technology,43, 2006, 1280-1286.
Experimental Setup and Plans
Oxygen FlowController
High SpeedDigital Camera
TransparentFurnace
Gas CoolingTube
OxygenSensor
HalogenLamp andStrobe Light
Sparging Tube
• Melt 150 g of LiCl-KCl eutectic salt (58.5 mol% LiCl, 41.5 mol% KCl) at500C under an argon environment.
• O2 Sparging Rates: (8.33 10-7, 1.67 10-6, 2.50 10-6, 3.33 10-6 ) m3/s.Note: (0.05, 0.10, 0.15, 0.20) L/min.
• Record bubble images (250 fps) and O2 concentration data ( 3%).• Sparge until salt is O2 saturated.
Corrosion Issues
304 Stainless Steel
Inconel 600
Non-porous Alumina(Aluminum oxide)3/16” OD 0.031”holes.
Solution
Obstacles
Bubble Dispersion Video
Temperature: 500⁰ C Camera Speed: 250 fps
Sparging Rate: 8.33 10-7 m3/s Sample Size: 150 g
Transparent Study Analysis
Bubble velocity is also observed.
Bubble measurement Midas OS Imaging
3/1
EllipsoidEquivalent
2SmalleargLEllipsoid
V6
b
bb6
V
Sparging rate,Q
(m3/s)
Number meandiameter, b10
(m)
Numberstandard
deviation, σ10
(m)
Volume meandiameter, b30
(m)
Sauter meandiameter, b32
(m)
8.33 × 10-7 0.00263 0.00073 0.00301 0.00304
1.67 × 10-6 0.00368 0.00112 0.00421 0.00438
2.50 × 10-6 0.00383 0.00118 0.00466 0.00455
3.33 × 10-6 0.00407 0.00103 0.00451 0.00459
Bubble Shape Validation
/VdRe
/gM
/dgEo
bequiv
324
2equiv
Conditions of Curve
Log M= -10
Eotvos Number 0.2-40
Reynold’s Number 300-4000
Experimental Results
Log M= -10.1304
Eotvos Number 0.54-8.94
Reynold’s Number 344-2406
[7]
100
1000
10000
0.5 1 2 4 8 16
Rey
nol
ds
Nu
mb
er,
Re
Eotvos Number, Eo
8.33 E-7
1.67 E-6
2.50 E-6
3.33 E-6
Log M=-10
Sparging Ratem3/s
Salt viscosity,μc (N s/m2)
Salt density,ρc (kg/m3)
Surface tension,σc (N/m)
Oxygen density,ρb (kg/m3)
0.00223 1,621 0.126 0.457
Bubble Size Distributions
0.0
0.3
0.6
0.9
1.2
1.5
1.8
0 1 2
Pro
ba
bil
ity
Den
sity
Fu
rnct
ion
Normalized Bubble Diameter, b
0.0
0.2
0.4
0.6
0.8
1.0
0 1 2
Cu
mu
lati
ve
Nu
mb
erF
req
uen
cy
Normalized Bubble Diameter, b
8.33 E-7
1.67 E-6
2.50 E-6
3.33 E-6
Gaussian Curve
Sparging Rate(m3/s)
The normally distributed probability density function:
2
2
1exp
2
1)(f
018.0255.0027.0004.1b
b
10
and
Fanning Friction RatioSpargingRate, Q(m3/s)
Mean BubbleRising
Velocity, Vb
(m/s)
StandardDeviation
(m/s)
8.33 × 10-7 0.262 0.05101.67 × 10-6 0.301 0.09982.50 × 10-6 0.299 0.05543.33 × 10-6 0.338 0.0770
0.00001
0.0001
0.001
0.01
300 3000
Fa
nn
ing
Fri
ctio
nR
ati
o,
f/R
e b
Bubble Reynolds Number, Reb
8.33 E-7
1.67 E-6
2.50 E-6
3.33 E-6
Sphere
Fit (Eq. 9)
Sparging Ratem3/s
c
bc
bc
c
b V
g
3
4
Re
f
35.1b
b
Re29.6Re
f
Using this correlation, the rising
velocity of a bubble can be
estimated based on the equivalent
diameter of a bubble.
Oxygen Gas Holdup, f
Sparging rate, Q(m3/s)
Experimental(Dimensionless)
Akita and Yoshida(Dimensionless)
Difference(%)
8.33 × 10-7 0.00169 0.00174 3.09
1.67 × 10-6 0.00294 0.00345 17.4
2.50 × 10-6 0.00427 0.00514 19.2
3.33 × 10-6 0.00523 0.00681 30.3
Experimental Approach:
Akita and Yoshida’s Correlation (1973):
2b
S
D
Q4
V
V
SVand
Fr
068.0
c
b086.0Ga
121.0Bo4
NNN32.0)1(
c
c2
Bo
gDN
Bond Number
2c
3
Ga
gDN
2/1S
FrgD
VN
Galileo Number Froude Number
• Calculate the interfacial area of an averageequivalent bubble:
• Oxygenation model:
• After integration with the I.C. t = 0, C = C0:
32b
6a
k = Mass Transfer Coefficient
C* = Concentration at saturation
C0 = Initial concentration
C = Concentration at specific time
φ = Gas hold up
a = Interfacial area
Vs = Superficial gas velocity
Vb = Bubble rise velocity
Q = Gas flow rate
D = Crucible diameter
b32 = Sauter mean diameter
)CC()1(
ka
dt
dC *
t
1
ka)CCln()CCln( 0
**
O2 Mass Transfer Coefficient
Plot and Resulting k values
0.70
0.75
0.80
0.85
0.90
0.95
1.00
0 20 40 60 80
ln((
C*-
C0)
/(C
*-C
))
Time (s)
8.33 E-7
1.67 E-6
2.50 E-6
3.33 E-6
Sparging Ratem3/s
Sparging rate,Q (m3/s)
Slope,ka/(1-F)
(s-1)
Interfacialarea, a
(m-1)
Mass transfercoefficient, k
(m/s)
95%Confidence
Interval (%)8.33 × 10-7 0.0032 3.33 0.00096 18.91.67 × 10-6 0.0029 4.03 0.00072 18.92.50 × 10-6 0.0035 5.63 0.00068 26.23.33 × 10-6 0.0027 6.82 0.00039 14.7
Diffusion Coefficient• Use the experimental k value to predict DL .
• Use 3 correlations based on two different frames of preferences: (1)physics of the bubble and (2) systematic physical properties.
Penetration theory:
b
322
LV4
bkD
Calderbank and Moo-Young (1960):3/23/1
c3/1
c2
L )g/(kD
Akita and Yoshida (1973): )g/(DkD 08.0c
16.0c
66.006.024.0c
26.02L
Sparging Rate( m3/s )
Diffusion coefficients, DL (m2/s)Penetration
TheoryCalderbank and
Moo-YoungAkita andYoshida
8.33 × 10-7 8.40 × 10-9 1.28 × 10-8 2.43 × 10-8
1.67 × 10-6 5.89 × 10-9 7.21 × 10-9 1.42 × 10-8
2.50 × 10-6 5.82 × 10-9 6.43 × 10-9 1.30 × 10-8
3.33 × 10-6 1.65 × 10-9 2.12 × 10-9 4.44 × 10-9
Conclusion
• Both bubble size and rise velocity with in oxygen sparging rate.
• b10 ranged from 0.00263 m to 0.00407 m.
• These bubbles form an ellipsoidal shape and the equivalent bubblediameters of the populations observed were normally distributed.
• The bubble rise velocities can be predicted using a Fanning frictionfactor-Reynolds number correlation.
• k can be calculated experimentally using the oxygenation model andranged from 3.94 × 10-4 m/s to 9.61 × 10-4 m/s.
• The results show trend with sparging rate or bubble size whichis supported by observations reported by previous researchers(Calderbank et al., 1960; Kulkarni, 2007).
• Diffusion coefficients were calculated using three correlations andcompared.
• Penetration theory is the most robust of these models yieldingdiffusivity values between 1.65 × 10-9 m2/s and 8.40 × 10-9 m2/s,which is supported by literature (Cussler, 1984; Sada et al., 1984).
Acknowledgement
• US Department of Energy/Idaho Nationallaboratory/Battelle Energy Alliance-Drawdown Project .
• Center for Advanced Energy Studies for the use of theRadiochemistry Laboratory.
• Michael Shaltry, Robert Hoover and Ammon Williamsfor experimental and technical supports.