Mixing Sludges & Slurries with

Pulsed Jets: Some mixing theory &

Test Results
Slurry Retrieval, Pipeline Transport & Plugging & Mixing Workshop
January 14 - 18, 2008, Orlando, Florida.

Perry A. Meyer
Pacific Northwest National Laboratory
 Unsteady Jet Mixers at Hanford
• Retrieving from storage • Treating & vitrifying waste
– Underground, 1 - 2ft risers – Closed “black” cells
– Limited access for equipment – No maintenance for 40 years
– 2 - 300hp mixer pumps (baseline)

Rotating horizontal opposed jets Pneumatic pulsed jets

2
 Turbulent Jets
dj uj
Near Field δ(z) Far Field

z u(z)
High Reynolds number far field
z Constant spread angle δ(z) = θz
z Peak & ave. velocity decrease u(z) = c ju jd j / z u A
d
z Thrust/force is constant F(z) = Fj
z Flow rate increases (entrainment) q(z) / q j ~ z / d
z Energy decreases e(z) / e j = d / z
z Constant Reynolds number Reδ (z) = Red

3
 Turbulent Jets, cont.
Same results for impinging & attaching jets
z Different constants
z Wall shear stress τ w (r) ~ ρu 2j (d j / r) 2
True independent of nozzle cross-sectional area
Approximately true in near-far-field transition z / d j , r / d j = 15 − 30
Allows one to approximately obtain flow fields, fluxes, forces, etc
Similar relations for dense jets

δ(r)
u(r)

4
 Jets as mixers
Axial flow impeller: ND ~ uj
z dj/T <<1 (careful about blindly applying agitator results)
z Power, thrust, and flow numbers = ~ 1
„ Much higher power than agitators for same thrust
„ Lower flow, higher head
Highly directional
z point them where you want them
z Must design for thrust reaction
Return placement
z Can be important

5
 Downward vertical jet mixers
Centered Jet(s) Jet rings(s)

uuw

~T/2

uT
uT ~ u j (d j / T) uuw ~ u j Nj (d j / T) × f(H / T)

Ω ~ u jd j / T 2 t uw ~ T 2 / u jd j Nj × f 2 (H / T)

6
 Geometry
Nozzle geometry
z Cross-section: No effect in far field- only area counts
z Convergence: extra thrust from pressure
Stand-off
z No effect for h/dj < 6, little effect for h/T<<1
Number of jets
z N1/2dj momentum/thrust effect
z T/N1/2 ZOI geometric effect
Return location
z Can be important- Avoid short-circuiting
Dish shape
z Impingement angle- flow distribution
Other internals
z Wakes/blockages

7
 Intermittent Jets

Dimensionless pulse time Steady

u
determines regime uS
z
N p = t p ud / d
Short pulse
uV
NP < 4 vortex ring
uC
uP
4 < NP vortex ring with tail
Long pulse
4 << NP developing steady uF

uP
PJMs: NP = 80 - 500
Steady

8
 Unsteady effects on mixing/mobilization
Would like to utilize steady mixing knowledge base
z Can we find simple corrections for unsteady effects or are we
dealing with fundamentally new phenomena?
Must consider relative time scales
z Flow establishment/mixing times compared with pulse time
z Duty cycle
„ What happens when the jet is off?
z Other time scales
„ Erosion rates
„ Settling rates
„ Etc.
Two new parameters are introduced
z Relative pulse volume
z Duty cycle

9
 Pulse jet mixers
H
u j(t)
H
V
tp
tc
Vpjm

PJMs in the WTP

dj z V (range)
uj
z N (range)
T z Pvf (range)
Mixing modes z DC (range)
z drive z Dpjm (range)
z refill
10
 Important parameters

Geometry Waste physical

N number jets configuration
Uj jet velocity (peak) z Normal/off-normal
dj/T nozzle diam. operations
Φp = Vp/V pulse size z Uniform
DC = tp/tc duty cycle z Settled layers
geometry

Operational Physical &

rheological
properties

11
Pulse Jet Mixing Studies at Battelle/PNNL
Physical regimes
z Transitional flow
z unsteady
z Non-settling/non-Newtonian
z Settling- wide particle size &
density range, agglomerates
z Heels- cohesive/non-cohesive
z In situ gas generation
Mixing requirements
z Stagnation/caverns
z Off-bottom suspension- VJS
z Vertical distribution
z Gas hold-up & release behavior
Scaled testing program
z Simulant development
„ Physical/chemical
„ Transparent/opaque
„ 1/2/3 phase
z Testing
„ Bench scale - 40m3
„ Single & multi jets
„ simplified & prototypic
geometries
z Scale up
„ Rating, not designing
„ Similarity, physical,
empirical
z Instrumentation
12
 Non -Newtonian PJM Test Program
Non-Newtonian

Technical basis
z Develop scaled testing approach
z Validate approach- limited testing at scales
Rate existing designs
z (3 unique designs in WTP)
Improved PJM designs
PJM/sparge hybrid designs

13
 Theory of PJM Operation with Non
Non--
Newtonian Materials
Typical Pulse Jet Mixer System
Model problem: Cavern
formation
z Initially gelled material Un-yielded
material
z Representative of restart after
mixing shutdown
Distinct
z Good mixing system will interface
eliminate cavern

Rheological model Turbulent flow

z Static gel formation with shear
strength τs Illustrating Rheological Characteristics of
Waste Slurry
z Bingham plastic laminar flow τs stat ic
rheology with yield stress τ0 and K
consistency K Shea r
Stress
z Turbulent flow characteristics (Pa)
τ0 lam inar
~K
determined by high shear
consistency ~K turbu lent
Stra in R ate

2/26/2008
 Cavern Formation from a Steady Jet
Turbulent wall jet
u(z) = cJ u d d / z τf = Cf ρu 2 / 2
Force balance at static Stagnant Vp Cavern
interface material boundary
at zC ≈ HC + T / 2 τf = τs
zc
HC / T = a(d / T)Re1τ / 2 − 1/ 2 Turbulent
u(z) mixing
Yield Reynolds Number HC Path of
z wall jet cavern
d
Reτ = ρu 2d / τs ud
T
Reynolds number dependence
−β
Red = ρu d d / k C f , cJ = f (Red ) a ~ Red
15
Theory of PJM Operation in Non-Newtonian
Non-Newtonian
Materials
Pl ug m otion

Cavern Formation from a U nd is tu rbed

Steady Jet m ater ia l

zc
z Turbulent jet theory with force
u(z)
balance at interface predicts
=ρu/τ
2
HC z
cavern height
Yield Reynolds number
T urbu lent
C a ve rn Re
τ u0 s 0

z Ratio fluid force to material DT

strength 1.6

Reτ = ρu 20 / τ s 1.4

1.2

Effects of pulsation 1

0.8
z Ratio PJM drive time to flow
establishment time 0.6 0.875 inch Steady Jet

tD / t ss ~ Vp / d30 Re τ
0.4 1.0 inch Pulse Jet

2.0 inch Pulse Jet

0.2

Predicted cavern height 0

0 1000 2000 3000 4000 5000

Yield Reynolds Number Re τ

⎛
1/ 2
Hc d Vp ⎞ 1
= a 0 Re τ ⎜1− exp( −c 3 )⎟ −
1/ 2
Non-dimensional cavern height as a function of yield
DT DT ⎝ d 0 Re τ ⎠ 2 Reynolds number for a single PJM in Laponite
16
Single-PJM Cavern Tests ((laponite)
Single-PJM laponite)

1.6

Normalized Cavern Height H C/T

1.4

1.2

0.8

0.6

0.4 2.2cm Steady Jet (Eq. 10)

2.5cm Pulse Jet (Eq. 10)
0.2 5.1cm Pulse Jet (Eq. 10)

0
0 1000 2000 3000 4000 500
Yield Reynolds Numberτ Re

17
 Test to Verify Scaled Testing Approach
Breakthrough
1PJM Tests Location

z Simulant selection
z Verify cavern formation theory
Upwell
4PJM Tests Velocity

z Downward firing PJMs v

z Performed at 3 scales
HC
Simulants
H
z Laponite Turbulent
„ Transparent Cavern u0
„ Adjustable shear strength
z Kaolin/Bentonite Clay DT
„ Opaque Types of measurements
„ Adjustable yield
stress/consistency z Cavern height (Laponite)
z Breakthrough velocity (clay &
Test conditions Laponite)
„ Rheology (20 -120 Pa) z Upwell velocity (clay)
„ Velocity (3-30 m/s)
18
Small Scale Test Stands
Battelle 1/4-scale 4
PJM Test Vessel
z 34 in. diameter
z 250 gallons
z Acrylic vessel
z Compressed
air/vacuum PJM
drive system

SRNL 1/9-scale 4
PJM Test Vessel
z 17 in. diameter
z ~30 gallons
z Acrylic vessel
z Compressed
air/vacuum PJM
drive system
19
 Large -Scale Test Stand at Battelle
Large-Scale
Battelle 336 4 PJM Test
Vessel
z ~13 ft. diameter, ~12,000
gallons
z Steel construction
z Prototypic AEA
Compressed air PJM drive
system

Pulse tube prior to

installation
z 24 in. diameter
z 2 in. conical nozzle
20
 Scaling Data Comparisons
1.6

1.4
Non-dimensional Cavern Height Hc/D

1.2

0.8
336
APEL
0.6 SRS
Break Through 336
APEL Break Through
0.4
SRS Break Through
)Linear (336
0.2 )Linear (APEL
)Linear (SRS

0
0 200 400 600 800 1000 1200 1400 1600 1800

Yield Reynolds Number Reτ

Comparison of cavern position for tanks of 3 different scales with

with
Laponite simulant.
21
 Scaling Data Comparisons

Laponite Breakthrough

Clay Breakthrough Laponite Breakthrough

Clay Breakthrough

Comparison of surface breakthrough velocity for tanks

of 3 different scales with Laponite & clay simulants.
22
 Gas hold -up & release
hold-up release-- tests at 3 scales
12
SRNL 12/13/03 (1:9 - 16 Pa)
10
APEL 12/02/03 (1:4.5 - 20 Pa)

Retained Gas (vol%)

8 336 7/23/04 (1:1 - 20 Pa)

0
APEL 2-25-04 (18 Pa, 14 cP) 0 100 200 300 400
336 7-22-04 (20 Pa, 18 cP)
2 SRNL 12/13/03 (16 Pa, 19 cP)
Number of PJM Cycles
Retained Gas (vol%)

0 50 100 150
Time (min)
23
Baseline Designs

UC

I Cavern only II Breakthrough, ŅfrozenÓzones

III Breakthrough with slow IV Full turbulent mixing

peripheral movement

24
Improved PJM designs

25
 Air sparging in Bingham Plastic Slurry
150
Liqu id Su r face
140

118" 83" 66" ROB Avg. 130

118" 83" 66" ZOI 120

160 110
Characteristic Diameter (DROB, D ZOI ; in)

100
140

Tank Elevation (in.)

90

120 0.34 80
DZOI = 34Q
70
100
60

80 50

0.34 40
60 (DZOI + D ROB)/2=22Q
30 40 ac fm
15 ac fm
40 20 5 ac fm
0.34 Sparge Tube
DROB = 11Q 10 Liquid Surfac e
20 0
Co n e T an k Bo tto m Tank W all
0 10 20 30 40 50 60 70 80
0 Dis tan ce fr o m Ce n te r o f T ank (in .)

0 10 20 30 40 50 60 3.0
Flow (Q, ACFM) 206 acfm
2.5 68 acfm

Gas Volume (vol%)

2.0
ROB 2/3 ZOI

1.5

1.0

ZOI 0.5
SPARGE
TUBE 0.0
DS 0 5 10 15 20
Time (min)

26
 PJM/air -sparge hybrid designs
PJM/air-sparge
34 in

Recirculation Pump
Discharge Line
70 in. Diameter
16 1/2 in. Diameter
48
Perimeter PJM
Sparger
40
59 1/8 in. Diameter
In Center PJM
37

Pump Discharge
Pump Suction Line Line
40 30 in. Diameter
61 3/4 in. Diameter

17 1/8in
PJM Tube
14 3/4in

Recirculation Pump
Discharge Line ~ 4 in 1 1/4in
1 1/4 in
3 7/8in
45
Recirculation Pump
Suction Line

27
 Final Design Mixing Performance

PJM Only PJMs + Sparging

PJMs + Pump (1 Disch. Noz) Linear (PJM Only)

LS 35-36 Pa QSLS 34 Pa Clay

HSLS 33-43 Pa QSLS 13 Pa Clay
QSLS 32 Pa AZ+AFA
QSLS 13 Pa AZ+AFA
Gas Volume Fraction (vol%)

QSLS 3 Pa AZ+AFA
0
10

-1
10
-2 -1
10 10
Superficial Velocity (mm/s) 28
 M3
M3-- Rating WTP Mixing Systems

Rate mixing system designs for balance of WTP vessels

z Normal operations & mixing restart
Broad range of potential waste conditions
z Non-cohesive (settling) solids - cohesive solids
z Wide range of solids size, density, slurry rheology
18 different vessel/mixing system geometries
Primary metrics
z Off-bottom suspension
z Vertical solids distribution
z Blend times
Work in 3 phases: non-cohesive, cohesive, gas handling

29
Preliminary tests with non -cohesive solids
non-cohesive
Simulants
z Glass spheres (low grade), S ~ 2.47
z 3 sizes: ds = 63-100, 150-210, 600-800μm
z 2 solids loadings: φs = 0.005 & 0.015
Vessel geometries
z 34-in., 1/13.4-scale of HLP-22
z 12 tubes, 0.3 & 0.45-in nozzles (4 & 6-in. full scale)
Operational
z Pulse volume fraction φp = 0.025 - 0.10
z Duty cycle: DC = 0.18, 0.36, 0.5, 1 (steady)
Measurements
z Ujs & peak cloud height

30
 Some off -bottom suspension results
off-bottom
pulsed, 0.5% solids pulsed, 1.5% solids DC=0.18 DC=0.33 DC=1.0, Steady
steady, 0.5% solids steady, 1.5% solids
10
9
0.23
y = 1.24x
-
y = 19.20x
0.27
y = 0.96x
7
-
y = 33.36x 0.49
y = 0.22x
-
5 y = 25.69x

-
y = 34.73x
3
1
3 4 5 6 7
100 1000
Nozzle Diameter, dj (equiv. plant scale, in.) Volume Mean Particle Diameter, dS (mm)

4-in. pulsed, DC=0.18 4-in. pulsed, DC=0.33 4-in. 6-in.

6-in. pulsed, DC=0.18 6-in. pulsed, DC=0.33
4 in. Steady 6 in. Steady 7

8
6
-
7 y = 3.75x y = 5.72x -

6
5
5
-
y = 1.99x
4 4
-
3 y = 3.46x
3
2
0.00 0.05 0.10 0.15 0.00 0.25 0.50 0.75 1.00
Pulse Volume Fraction, fp Duty Cycle, DC

Preliminary data for information only 31

 Correlating just -suspended velocity
just-suspended
Assume Zwietering values for un-tested parameters

Ucs = k(H/ D′)0.14 g0.5(s −1)0.43(D′)1.3

× (ds )a5 (d j )a6 (100sφs )a 7 (DC)a8 (φ p /(1+ φ p ))a9
D′ = D/ N
Steady Pulsed
k 0.78 0.23
ds 0.47 0.26
dj -1.3 -1.06
Sφs 0.23 0.34
DC - -0.06

φp - -0.18

Preliminary data for information only 32

 Data correlation: off -bottom suspension
off-bottom
Steady Ujs Unsteady Ujs based on peak. ve
10 10

y = 1.00x + y = 1.00x +
8 0.00 8 0.01
2
R = 0.97 R2 = 0.97
6
6

4
4

2
2

0
0
0 2 4 6 8 10
0 2 4 6 8 10
measred Ucs
measred Ucs

Suggests pulsation effects small at low concentration

Scale-up to plant conditions: design likely inadequate
z More testing at additional scales & higher solids required

Preliminary data for information only 33

 Cloud height data

100 micron pulsed 200 micron pulsed 800 micron pulse DC=1 DC=0.5 DC=0.33 DC=0.1
100 micron steady 200 micron steady 800 micron stead
1.25
1.00

1.00
0.75

0.75

0.50
0.50

0.25
0.25
1 2 3 4 5 6 7 8
2 3 4 5 6 7 8
Jet Velocity (m/ Jet Velocity (m

Preliminary data for information only 34

 Correlating cloud-height
cloud-height
Simple energy argument
z Energy per pulse ~ change in potential energy of solids
φs u2
φ pFH ~ FH =
φd 2(s −1)gHc

HC φ p φd u2
~ FD FD =
D φs 2(s −1)gD

Attempt correlation of the form

HC a1 a 2 a 3 a 4
~ FD φ p φd φs include d s / D or u s / u
D

Preliminary data for information only 35

 Correlation of cloud -height data
cloud-height
Correlation of steady-jet cloud-heigh Correlation of pulsed-jet cloud-height
1.2 1.4

y = 1.00x + 0.12 1.2

1.0 y = 1.00x - 0.01
2
R = 0.94
R2 = 0.95
1.0
0.8
0.8
0.6
0.6

0.4
0.4

0.2 0.2

0.0
0.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Hc/T - 0.25
Hc/T -0.25

k U φs φd ds/T φp DC
Steady 2.8 2 -0.56 1.7 -1.1 - -
Pulsed 7.1 2 -1.1 1.0 -0.5 0.3 0.25

Preliminary data for information only 36

 Summary of findings

Just suspended velocity

z Unsteady effects minor
„ DC effects negligible
„ There is evidence this breaks down at higher concentration
where time to suspend > drive time
z Similar solids size effect
z Concentration exponent 2x
z Effect of nozzle size as expected
z To be sure, need more data

37
 Summary, cont.

Vertical distribution
z Strong bulk density stratification effect
z Unsteady effects appear to dominate
„ Exponents on DC & PVF
„ Fundamental behavior
z Weak solids size dependence:
„ Define UJH (“just to H…”). Then UCH ~ ds0.25
z Strong concentration effect: UCH~ φs 0.5
z Strong pulsation effect: UCH~ φp -0.5

38