Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02686831 2009-12-02
Application number/numero de demande: &3 l
Figures: (v Wgote- 30n a
ern nom- C E I I on j 2 i s , - )2 _ SVAA n
~, I Q_ Ih n_~ ) S
Pages:
Unscannable Item(s) received with this application.
To inquire is you can order a copy of the unscannable item(s),
Please visit the CIPO Website at HTTP://CIPO.GC.CA
Item(s) ne pouvant titre balayes.
Documents regus avec cette demande ne pouvant titre balayes.
Pour vous renseigner si vous pouvez commander
une copie des items ne pouvant titre balayes,
veuillez visiter le site web de I' OPIC au HTTP://CIPO.GC.CA
CA 02686831 2009-12-02
1
PROCESS FOR FLOCCULATING, DEWATERING, DEPOSITING AND DRYING OIL
SAND MATURE FINE TAILINGS
FIELD OF THE INVENTION
The present invention generally relates to the field of treating oil sand fine
tailings.
BACKGROUND
Extraction of bitumen from the Athabasca Oil Sands deposits of north-eastern
Alberta,
Canada has primarily been through a process known as the Clark Hot Water
Extraction
process (CHWE). In its purest form, the CHWE involves the addition of hot
water to the
oil sand ore together with pipeline mixing and floatation in separation cells.
The tailings
from this process include slurries of sand, silts, clays and residual bitumen.
When
deposited into holding ponds, this slurry forms segregating deposits with the
sand
settling rapidly to form beaches and a percentage of the fines and clays
remaining
suspended within the water column. Over time, the fines settle to form a thick
material
known as Mature Fine Tailings (MFT). This material is characterised as being
above
30% solids by weight and typically less than 50%; having 90% of the solid
particles less
than 44 microns in size; approximately 50% of the particles less than 2
microns; and
having very slow settlement rates. Several decades of research into this
material have
resulted in an excellent fundamental understanding of behaviours, but had not
produced
a fully implemented, proven technology for dealing with the ever increasing
fluid MFT
inventories.
One potential method of dealing with these inventories was to utilise chemical
addition to
modify the characteristics of the tailings stream sufficiently to allow for
thin lift deposition
of the material onto shallow slopes, resulting in dewatering through
evaporation and
freeze/thaw processes. While there are significant advantages to the use of
this method,
there are also limitations resulting from the climate in northern Alberta.
SUMMARY OF THE INVENTION
The present invention provides a process for flocculating, dewatering,
depositing and
drying undiluted oil sand mature fine tailings comprising:
a dispersion and floc build-up stage comprising adding a polymer focculant to
tailings to cause a rapid increase in yield stress as the polymer flocculant
contacts the minerals of the tailings;
a gel state of high shear yield stress stage, which can be a plateau over time
depending on the applied shear rate and %solids of the tailings;
a decreasing shear strength and floc breakdown stage, wherein a significant
amount of substantially polymer-free water is released; and
avoiding an oversheared zone characterized by rapidly decreasing shear
strength
where the material reverts to a tailings-state and releases very little water.
BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 is a graph of Shear strength progression of flocculated MFT
highlighting four
distinct stages.
CA 02686831 2009-12-02
2
Figure A is a graph of Shear strength progression of flocculated MFT
highlighting four
distinct stages.
Figure B is a graph of Maximum water release from polymer-treated MFT during
mixing.
Figure 2 is a graph of Variation of polymer dosage with yield stress and water
release.
Figure 3 is Scanning electron micrographs of 40 wt% MFT showing the fabric at
different
shear regimes (a). Untreated MFT, (b) high yield strength and (c) dewatering
stage.
Figure 4 is a graph of Shear strength progression for optimally dosed MFT
samples
diluted to varying solids concentration.
Figure 5 is a graph of Yield stress progression of MFT optimally dosed with
preferred
polymer (Poly A) and a high molecular weight anionic polyacrylamide aPAM (Poly
B).
Figures 6a-b are Computational fluid dynamic model images of polymer injection
systems as a mist (top, a) and as a stream (bottom, b).
Figure 7a is a graph of Shear progression curves of the pilot scale
flocculated MFT (35
wt% solid).
Figure 7b is a photograph of Jar samples taken at each sample point in Figure
7a at
ideal dosage and low shear.
Figure 8 is a photograph of Flocculated MFT discharged on a slope showing high
yield
strength material and water channeling around the discharge.
Figure 9 is a graph of Water release rate of flocculated MFT at various
distances from
the injection point.
Figure 10 is a photograph of Crack formation in an optimally flocculated MFT
after a few
days.
DETAILED DESCRPTION
Preliminary advances
Regarding the preferred performance requirements for an additive chemical, the
focus
was put on on strength gain and resistance to shear. Another objective was
enhanced
dewatering, as several previous attempts to flocculate MFT (Golder 2007,
Matthews, J.
2008 and Silawat J. 2008) required dilution of the material prior to mixing
with the
flocculant, and then only achieved clay to water ratios similar to or slightly
less than that
found in the source MFT. Commercial application of polymeric flocculation in
oil sands is
restricted to rapid dewatering of low solids content thin fine tails. In
short, flocculants
had been unable to collapse the clay matrix any further than that found in the
ponds.
During the course of bench scale tests, a new polymer type showed promise in
both
material strength gain as well as shear resistance. In addition, the polymer
appeared to
promote initial dewatering of the MFT shortly after mixing by generating a
highly
permeable floc structure. This means that the process no longer relies on
evaporative
drying alone, but rather a combination of initial accelerated dewatering and
drainage in
CA 02686831 2009-12-02
3
the deposit slope as well as evaporation. No dilution of the MFT was required
beyond
the polymer make up water and the polymer could be injected in line without
the use of a
thickener. The polymer was quite effective for MFT up to 40 percent by weight
(roughly
0.4 clay-to-water ratio).
Initial field tests produced surprising results, allowing for 20-30cm lifts to
reach
80%solids in less than 10 days. Given the weather conditions at the time, the
minimum
amount of water released as free water was 85% as the potential evaporation
rates were
too low to account for the dewatering rate. This initial success appeared to
be robust
and relatively insensitive to changes in fluid density and injection
locations.
Subsequent testing began to illustrate, however, that there was a basic
understanding of
the behaviour of the flocculated material that was not obtained during the
initial
laboratory or field tests. Deposits were attempted with lower levels of
control on the
density and flowrates of the source MFT, resulting in a wider variety of
deposit
dewatering rates. Many of these deposits did not behave as previously
observed, and
several attempts at enhancing the dewatering performance through additional
mixing,
changes in the deposition mechanisms, or mechanical manipulation of the
deposits met
with limited success. It became apparent that more testing was required.
Investigation of Undiluted MFT Flocculation
It was attempted to manipulate the MFT floc structure such that initial
dewatering is
maximized and the MFT gained just enough strength to stack in a thin lift when
deposited on a shallow slope. Dewatering occurs as a function of mixing and
applied
shear during pipeline transport as well as on the deposition slopes.
Bench and pilot scale experiments were conducted to replicate the field
observations
and to investigate the dewatering potential as a function of polymer dosage,
injection
type, mixing, total applied shear and clay-to-water ratio of the MFT. The
experiments
highlight several key factors.
1. Polymer dosage is best determined by clay content, measured as clay
activity
using methylene blue adsorption method.
2. Mixing of the polymer-treated MFT using laboratory or in-line static mixers
can cause less than optimum dewatering potential and stacking in the
deposition slopes
3. Shear energy applied to the flocculated materials can greatly affect the
dewatering and strength performance. Insufficient shear often create a high
strength material with minimal dewatering and excess shear reduces the
strength to MFT-like strengths with reduced permeability and dewatering.
CA 02686831 2009-12-02
4
Polymer Dosage
Although it is recognized that the rheology of flocculated systems is governed
by the
finest particles in a slurry, polymer is often added on a gram per tonne of
solids basis.
This is often adequate for a homogeneous slurry. However, fine tailings are
deposited in
segregating ponds and the mineral size distribution of MFT depends on the
sampling
depth. Therefore dosing on a solid basis would often result in an underdosed
or an
overdosed situation affecting maximum water release. This is highlighted in
Table 1 for
three MFT samples that show large swings in the optimum polymer dosage on
solids or
fines basis. The MFT samples were sourced from two Suncor ponds at different
depths
and with similar water chemistries.
Table 1 - Optimum polymer dosage for maximum initial water release.
Optimum polymer dosage
Sample ID t% solids t% clay* t% fines(g/tonne o (g/tonne o (g/tonne o
on solids on solids solids) fines < 44 pm) clay)
MFT A 44.0 8.9 59.8 300 1424 1742
MFT B 32.6 78.9 89.3 1200 1428 1616
MFT C 22.3 99.6 98.8 1700 1707 1693
*Wt% clay is based on the surface area determined from methylene blue
adsorption and
could be greater than 100 % for high surface area clays (Omotoso and Mikula
2004).
Rheology of Flocculated MFT
A static yield stress progression over time was used to evaluate optimal yield
stress for
deposition and water release in the laboratory (Spicer P.T., 1996), pilot and
field
experiments. The shear yield stress was measured by a Brookfield DV-III
rheometer.
The water release was measured by decanting the initial water release and by
capillary
suction time (CST). The capillary suction time measures the filterability of a
slurry and is
essentially the time it takes water to percolate through the material and a
filter paper
medium, and travel between two electrodes placed 1 cm apart. The method is
often
used as a relative measure of permeability.
Figure 1 shows an optimally dosed MFT mixed in a laboratory jar mixer with the
rpm
calibrated to the mean velocity gradient. The figure shows the shear yield
stress
progression curve for a 40 wt% solids MFT. The polymer was injected within a
few
seconds while stirring the MFT at 220 s-1. Mixing continued at the same mean
velocity
gradient until the material completely broke down. At each point on the curve,
mixing
was stopped and the yield stress measured. Water release during mixing is
often
dramatic and was clearly observed. The extent of water release is given by the
capillary
suction time. A low suction time correlates to high permeability and a high
suction time
correlates to low permeability. MFT dosed at ideal rates released the most
water and
about 20-25% of the initial MFT water was released at the lowest CST.
CA 02686831 2009-12-02
3 Shear yield stress 3
5 Capillary suction time
2
2
1000
100 8
9- -7
`O 8
6
7 5 0
1b S- X
Z 9 /
100
s
y
0 20 40 60 80 100 120 140 160
Time(s)
Figure 1: Shear strength progression of flocculated MFT highlighting four
distinct
stages
In further studies, MFT was mixed with a shear-resistant polymer flocculant in
a
laboratory jar mixer with the rpm calibrated to total mixing energy input.
Figure A shows
the shear yield stress progression curve for a 40 wt% solids MFT dosed at
different
polymer concentrations. The experiment was conducted in two mixing stages. In
the
first stage, MFT was mixed at 220 s' during polymer injection. This stage
lasts for a few
seconds and defines the rate of floc buildup. In the second stage, the
material was
mixed at 63 s' until the material completely broke down. At each point on the
curve,
mixing was stopped and the yield stress measured. Water release during mixing
is often
dramatic and was clearly observed. MFT dosed at 1000 g/tonne of solid released
the
most water (Figure B). The material released about 20% of the initial MFT
water
immediately whereas the under-dosed and over-dosed MFT released very little
water
through complete floc breakdown.
CA 02686831 2009-12-02
6
Figure A: Shear strength progression of flocculated MFT highlighting four
distinct stages.
Figure B: Maximum water release from polymer-treated MFT during mixing.
In summary, four distinct stages were identified in the shear progression
curve:
CA 02686831 2009-12-02
7
= Polymer dispersion or floc build-up stage defined by a rapid increase in
yield
stress as the polymer contacts the minerals and poor water release.
= A gel state of high shear yield stress which can be a plateau depending on
the
applied shear rate and %solids of the MFT. The rates of floc build-up and
breakdown in this stage appear to be roughly the same.
= A region of decreasing shear strength and floc breakdown where significant
amount of polymer-free water is released.
= An oversheared region characterized by rapidly decreasing shear strength
where the material quickly reverts to an MFT state and releases very little
water.
These stages are used to quantify the behaviour of polymer-dosed MFT and to
compare
behaviours under different shear regimes and the third stage is the target
design basis.
An optimal dose of polymer with a good initial dispersion into MFT will
achieves
preferred permeability to release water. Without an optimal dose and good
dispersion,
the MFT remains in the gel state and only dries by evaporation. This is
highlighted in
Figure 2 where the same MFT in the underdosed or the overdosed state fail to
release
significant amount of water despite developing significant yield stresses. A
key
advantage of preferred polymers is having prolonged resistance to shear which
allows
operational flexibility when pipelining flocculated MFT to deposition cells.
1000 -#-Underdosed Mix
-*--Ideal Dosage
a a -.H--Overdosed Mix
y 2
y
100
y
4
2
0 20 40 60 80 100 120 140 160
Time (seconds)
E
1000,
6
4-
2- -+-Underdosed Mix
S 100 --*-Ideal Dosage
U 6 -H--Overdosed Mix
I I
0 20 40 60 80 100 120 140 160
Time (seconds)
Figure 2: Variation of polymer dosage with yield stress and water release
Shown in Figure 3 are the microstructures corresponding to different shear
regimes in
the preferred flocculated MFT in Figure 1. The MFT and flocculated slurries
were flash
dried to preserve the microstructure to some extent. Samples were platinum
coated and
examined in a scanning electron microscope. The starting MFT showed a more
massive
CA 02686831 2009-12-02
8
microstructure on drying and a greater tendency for the clays to stack along
their basal
planes in large booklets. This results in a low concentration of
interconnected pores and
poor dewatering. The middle micrographs in Figure 3 show microstructures
exhibited by
flocculated MFT in the second stage (383 Pa) at the onset of floc breakdown
and water
release. The microstructure is dominated by dense aggregates and randomly
oriented
clay platelets with more interconnected pores. The third set of micrographs
(86 Pa) show
less massive aggregates and a more open structure most likely responsible for
the large
water release observed in the third stage. The starting MFT is highly
impermeable,
whereas the flocculated MFT contains large macropores and significant amounts
of
micropores not visible in the starting MFT. At higher mixing time, the
porosities start to
collapse with an attendant reduction in the dewatering rates.
Figure 3: Scanning electron micrographs of 40 wt% MFT showing the fabric at
different shear regimes (a). Untreated MFT, (b) high yield strength and (c)
dewatering stage
Optimally dosed MFT with varying solids content were also investigated (Figure
4). As
the solids content decreases polymer dispersion becomes easier. The maximum
yield
strength of the material also decreases with increasing water content. A
substantial
amount of water is released at lower solids content (for example, 10 wt%
settles to 20
wt% immediately - the water release at a lower solids content was much greater
at 10
wt% solids (51% of the water in the original MFT) than at 40 wt% solids where
20% of
the water in the original MFT was released); however the floc structure is
weak and
unable to stack in a deposition slope without being washed off.
CA 02686831 2009-12-02
9
1000
8-
6:
a
2-
1 0010 4-
2-
j: 10$
61
6 -9-- 10 wt% solids
-U-- 20 wt% solids
-A- 30 wt% solids
2 -+- 40 wt% solids
1
-1 1 t -7
0 50 100 150 200 250
Time (seconds)
Figure 4: Shear strength progression for optimally dosed MFT samples diluted
to
varying solids concentration.
Further laboratory testing has shown that the strength gain and dewatering
effects are
possible with many anionic polymers, and are not limited to the particular
formulation
used in the first successful tests. Figure 5 compares a 40 wt% MFT optimally
dosed with
the preferred shear-resistant polymer and a high molecular weight anionic
polyacrylamide (aPAM) typically used for flocculating oil sands tailings. The
optimum
dosages for both polymers, in terms of maximum water release, were the same
(1000
g/tonne of solids) and were compared at two different shear rates. Polymer
dispersion
and shear stress response of the polymers differ significantly. Increasing the
dispersion
rate by increasing the mixer speed increases the yield stress instantaneously,
but the
traditional aPAM required additional mixing before the onset of flocculation.
This
decrease in the dispersion rate means that MFT treated with traditional
polymer is more
likely to stay in a gel state and not release as much water. In preferred
aspects of the
process, the polymer flocculant is highly shear-resistant especially during
the second
and third stages, and is also highly shear-responsive especially in the first
stage of
dispersing and mixing.
It is generally expected for an aPAM that a higher mixing energy rapidly
builds up the
yield stress but the floc breakdown also occurs at a faster rate. The lower
viscosity of the
preferred polymer coupled with a high resistance to shear allow the
flocculated MFT to
be transported over long distances to deposition cells without significant
floc breakdown.
CA 02686831 2009-12-02
700
-~- Poly A High Shear
600 -- Poly B High Shear
-4- Poly A Low Shear
500 f Poly B Low Shear
400
300
b
200
100
0 100 200 300 400
Time (seconds)
Figure 5: Yield stress progression of MFT optimally dosed with preferred
polymer
(Poly A) and a high molecular weight anionic polyacrylamide aPAM (Poly B).
5 Various polymers that have been developed with high shear resistance may be
used in
the process to improve the dewatering. Preferably, such shear-resistant
polymers would
also be in the general class of high molecular weight 30% anionic
polyacrylamide-
polyacrylate co-polymer flocculants.
10 In order to optimise the behaviour of the flocculated material, it is
preferable to limit the
variance in the shear energy applied to the various flocs which are created
during
mixing. This is achieved with an in-line orifice injector system. The concept
here is to
inject the polymer as a "mist" through the orifice (Figure 6a) instead of as a
stream
(Figure 6b). However, it should be understood that the quill-shaped injector
device of
Figure 6b may be modified by adapting the size of the perforations to approach
a mist-
like injection into the flow of MFT. When injected into a turbulent back-flow
regime as
shown in Figure 6a, the polymer is evenly distributed and flocculation is
occurring
throughout the pipeline cross section within 4 pipe diameters of the injection
point. This
rapid dispersion allows for precise control of the shear energies from the
injection point
to the point of deposition, and increases the percentage of the material that
falls within
the dewatering zone at a design point in the system. This fundamental
behavioural
understanding is key to the improved application of this technique, and allows
results
obtained from bench scale testing to be used in CFD modelling and scaled up to
field
operations.
CA 02686831 2009-12-02
11
Figure 6: Computational fluid dynamic model of polymer injection systems as a
mist (top) and as a stream (bottom).
Pilot Test for the Determination of Mixing Parameters
A 20-m long and 0.05-m diameter pipe loop fitted with the in-line orifice
injector was used
to investigate the shear response and dewatering behaviour of flocculated MFT
(Swift
J.D. 2004, Heath A.R. 2003, and D.N. Thomas 1999). Sample ports are fitted to
two
locations along the length of the pipe. Figure 7a shows that the yield
strength
CA 02686831 2009-12-02
12
progression in the pipe loop is similar to that observed in the laboratory jar
mixer
although the mixing energies are not directly comparable. MFT flow at 30 Umin
corresponds to a mean velocity gradient of 22 s' compared to 63 s' in the
bench scale
test. Another test conducted at 100 Umin (176 s) showed a more rapid floc
buildup
and breakdown similar to the the 220 s' test in the jar mixer. Figure 7b shows
flocculated MFT sampled at different locations during the test run for the
optimally dosed
MFT at 30 Umin (1000 g/tonne of solids in this case). Such data from the pilot
and field
tests may be used to develop a mixing model for process design and monitoring
of the
commercial scale MFT drying plant.
1000-
7-
6-
5-
4-
3- -I-- 50% Ideal Dosage Low Shear
2 -U- 80% Ideal Dosage Low Shear
--A- Ideal Dosage Low Shear
-- Ideal Dosage High Shear
100-
=A
7-
6-
5-
4-
3-
2-
0
40 60 80 100
Time (seconds)
Figure 7a: Shear progression curves of the pilot scale flocculated MFT (35 wt%
solid)
CA 02686831 2009-12-02
13
Figure 7b: Jar samples taken at each sample point in Figure 7a at ideal dosage
and low shear
Field observations
The rapid polymer dispersion by the orifice mixer caused the yield strength of
flocculated
material to increase very rapidly and results in the deposition of a two-phase
fluid. This is
shown in Figure 8 where the flocculated MFT and a separate water stream are
observed
at the discharge in one of the pilot tests.
A scaled up version of the orifice mixer was investigated in the field with
optimally dosed
35 - 40 wt% MFT flowing at - 7500 Umin (32 s) in 0.3 m pipe diameter, and
deposited
in cells at various distances from the injection point. Figure 9 shows the
extent of water
release for each cell, both from actual sampling after 24 h and a capillary
suction test
conducted on the as-deposited flocculated MFT. The dewatering trend is
analogous to
the shear progression profile for the laboratory and pilot tests. Over 25% of
the MFT
water was released immediately after injection up to 175 m. Beyond this
length, the
water release rate decreased rapidly and the flocculated material properties
resemble
MFT.
Further dewatering occurs in the deposition slopes through drainage enhanced
by the
slope and by evaporation. The under-mixed material deposited at roughly 7 m
from the
discharge was further dewatered by mechanically working the material to reach
the floc
breakdown stage where more water is released from the flocculated material.
Aggressive mechanical working however could break the deposit structure
resulting in
lower permeability and a restricted water release. Once the permeable
structure is
broken, dewatering is only by evaporation.
Evaporation results in crack formation as shown in Figure 10. Deepening cracks
through
dewatering allow for side drainage of release water into cracks and down the
slope.
Typical deposits up to 20 cm thick was found to dry beyond 80 wt% solids in 6-
10 days
after which a subsequent lift could be placed.
CA 02686831 2009-12-02
14
Figure 8: Flocculated MFT discharged on a slope showing high yield strength
material and water channeling around the discharge
CA 02686831 2009-12-02
Time in pipe (s)
0 20 40 60 80 100 120
5 50
J Mechanical working --W- Capillary suction
during discharge -i -- % Water loss after 24 h
-40
1:0
1000-
.2 9-
7- 20
5 "
3
25 0
n en Inn Icn inn
V )V Iv/ IJV tA!
Distance from polymer injection (m)
30 Figure 9: Water release rate of flocculated MFT at various distances from
the
injection point.
35 Figure 10: Crack formation in an optimally flocculated MFT after a few days
CA 02686831 2009-12-02
16
Conclusions
Industry based research efforts into oil sands tailings and reclamation
represent the final,
crucial step in developing successful technologies to deal with the legacy and
future fluid
MFT inventories. Fundamental research, while a first critical step, can only
identify
potential solutions and directions for development. Scale issues are major
determining
factors in successful oil sands technology implementations. Use of the present
techniques can reduce fluid inventories, minimise the need for tailings pond
storage, and
allow for accelerated reclamation of mining areas. Further focus on
determining the
technical controls of the process will result in ever more effective systems,
reducing
costs, land disturbance, and lower water requirements.
References
1. Wells, P. S and Riley, D. MFT Drying - case Study for the use of
Rheological
Modification and Dewatering of Fine Tailings Through Thin Lift Deposition in
Oil Sands
Tails. In Paste 2007 - A.B Fourie and R. J. Jewell (eds). 2007. Austalian
Centre for
Geomechanics, Perth.
2. Matthew J. Past Present and Future Tailings Experience at Albian Sands.
International Oil Sands Tailings Conference. December 7 - 10, 2008.
3. Silawat Jeeravipoolvarn. Deposition of inline thickened fine tailings.
International Oil
Sands Tailings Conference. December 7 - 10, 2008.
4. Golder Paste Technology Ltd. Conceptual study on applying paste technology
to oil
sands tailings management at Suncor, Fort McMurray. 2007.
5. Spicer P. T. Pratsinis S. E. (1996): Shear Induced flocculation: The
evolution of floc
structure and the shape and the size distribution at steady state. Water
Research 30 (5):
1049-1056.
6. Omotoso, O. and Mikula, R. J. (2004): High surface area caused by smectitic
interstratification of kaolinite and illite in Athabasca oil sands. Applied
Clay Science 25
(1-2): 37-47.
7. Swift J.D. , Simic K., Johnston R.R.M. , Fawell P.D. , Farrow J.B. (2004):
A study of
the polymer flocculation reaction in a linear pipe with a focused beam
reflectance
measurement probe. International journal of mineral processing 73: 103-118.
8. Heath A.R., Koh P.T.L (2003): Combined population balance and CFD modeling
of
particle aggregation by polymeric flocculant. Third International Conference
on CFD in
the Minerals and Process Industries. December 10-12 2003.
9. Thomas D.N., Judd S.J., Fawcett N. (1999): Flocculation Modeling: A review.
Water
Research 33 (7): 1579-1592.
45
