Note: Descriptions are shown in the official language in which they were submitted.
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
OLEOPHILIC SEPARATION TO REPLACE BITUMEN
FROTH FLOTATION OF OIL SAND SLURRY AND FFT.
RELATED APPLICATIONS
Patents granted to or applied for in Canada and in the USA by Jan Kruyer of
Thorsby, and previously from Edmonton are related to the present patent and
describe
more than 40 years of oleophilic separation technology development. Very few
patents
applied for in Canada or in the USA by others are related to oleophilic
separation as de-
fined in this patent. Most patents applied for or granted for oil sand
separation are based
on the concept of bitumen froth flotation, which is different from oleophilic
separation.
For more details, see : "Recent patents granted to or pending by the current
inventor"
near the end of the present specifications.
FIELD OF THE INVENTION
The present invention relates to process devices and methods for processing
mined oil sand slurries and oil sand tailings pond fluid fine tailings (FFT).
Accordingly,
it involves the fields of process engineering, chemistry and chemical
engineering.
BACKGROUND OF THE INVENTION
A detailed description of oil sand, tar sand and bituminous sand deposits, and
of
commercial processing these ore deposits to produce bitumen product is
provided in prior
patents of the present inventor. Up to now, only the oil sand deposits in the
Fort McMur-
ray, Alberta area have been and are surface mined in any major way to
commercially
produce bitumen that can be refined to useful hydrocarbon fuels or made into
other useful
products. Unlike other bituminous deposits, the Fort McMurray oil sand deposit
consists
of sand and silt grains covered by a thin envelope of water with bitumen
between the wa-
ter wetted grains. Dispersed clay fines often reside in the water envelope
that surround
1
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
each sand grain. In addition to the above components, oil sands also contain
fatty acids,
such as naphthenic acid, for example.
Current Alberta commercial mined oil sands plants surface mine the oil sand
ore
which may contain, for a medium grade ore,10% bitumen, 5% water and 85% solids
by
mass. Sufficient water, and some caustic soda are currently added to the mined
ore to
commercially enhance the subsequent preparation of an oil sand slurry at 50
degrees cen-
tigrade or higher. The slurry then may contain, for example, 7% bitumen, 70%
water and
23% solids. Commercial separation of that slurry results in a valuable bitumen
froth
product plus an aqueous effluent of separation that will contain sand, silt,
finer minerals,
water, and some unrecovered bitumen. The commercial bitumen froth normally is
de-
aerated and then contains about 60% bitumen for average grade oil sand, with
the rest of
the product being water and fine minerals. Solvent extraction or diluent
centrifuging is
used next to remove water and fine solids and yield a bitumen product that is
suitable for
upgrading to a useful hydrocarbon. Similar to the traditional method of making
soap,
which involves reacting caustic soda with grease (a fatty acid), the current
commercial
mined oil sands plants add caustic soda to the slurry to produce detergents
from fatty ac-
ids in the ore. These detergents help in oil sand slurry preparation and in
the subsequent
slurry separation by bitumen froth flotation.
During froth flotation the oil sand slurry, in the presence of detergents , is
frothed
with air to cause bitumen flecks - adhering to small air bubbles - to rise to
the top of flo-
tation vessels as a bitumen froth. This froth is skimmed off the top and
becomes the
product of separation after clean up. The aqueous effluent or tailings of
separation, con-
taining water, sand, silt, clay and bitumen, are impounded in tailings ponds
for decades
and currently are not processed commercially to recover any discarded bitumen.
This
impounding behind carefully engineered barriers is very expensive and mostly
prevents
the tailings from entering the environment. However, as happens in hydrocarbon
contain-
ing landfills, the unrecovered hydrocarbon (bitumen) in time is altered by
microbial ac-
tivity and releases methane into the surrounding air unless captured. In
landfills this me-
thane may be collected since methane is a valuable fuel. Collecting methane
from tailings
ponds is not done because of the large tailings pond surfaces. When entering
the air, me-
thane is known to be twenty times as environmentally potent as carbon dioxide.
2
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
Bitumen froth flotation is a long duration process which may take up to 6
hours
(360 minutes) of processing time to yield, from mined oil sand slurry, a
useful bitumen
froth product for further cleanup and/or upgrading. The carbon footprint of a
commercial
mined oil sands plant is very large and the use of caustic, used during froth
flotation, hin-
ders natural compaction of its aqueous effluent. Hence, the extraction
tailings cannot be
discarded to the environment but are stored in very large tailings ponds that
may remain
of environmental concern for many decades to come.
Oleophilic separation development was started in 1975 to overcome the environ-
mental impact of bitumen froth flotation, to speed up the recovery of bitumen
from oil
sand slurry, to retrieve discarded bitumen and to do away with long duration
tailings
ponds. It is now ready for field testing and commercial development.
SUMMARY OF THE INVENTION
Oleophilic separation does not alter the mining of oil sand ore, except that
caustic
soda normally is not needed in the production of oil sand slurry for
oleophilic separation.
Preparation of oil sand slurry for oleophilic separation can be faster than
for froth flota-
tion since the agglomerator, drum or cage, which is at the heart of oleophilic
separation,
is less demanding of slurry produced than is froth flotation. Eliminating
caustic soda in
oleophilic separation speeds up tailings settling and opens the way for using
short dura-
tion small tailings ponds instead of large long duration ponds now needed for
commercial
bitumen froth flotation plant effluents. Small ponds will allow, in a matter
of months,
after sand and silt have settled to the bottom and residual bitumen has
concentrated in the
upper levels, effective recovery of bitumen not captured before when
separating oil sand
slurry by oleophilic separation. Normally this does not required heating of
the settled ef-
fluent. Total bitumen recovery from mined oil sand then becomes very high. The
current
commercial mined oil sand plant operators do not recover bitumen from tailings
ponds
because there is no process, other than oleophilic separation, capable of
doing so at a po-
tential profit. However, oleophilic separation is a technology not owned by
governments
nor by oil sand operators, and this has significantly slowed down its
commercial devel-
opment.
3
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
DISCARDED BITUMEN
A few decades ago, bitumen froth flotation of oil sand slurry achieved about
90%
bitumen recovery from average mined oil sand ore. For a plant producing
100,000 bar-
rels of bitumen per day, the mined ore contained 111,111 barrels of bitumen
and 11,111
barrels of bitumen were discarded per day, representing 4 million barrels of
bitumen lost
per year at each commercial plant. That is a lot of unrecovered valuable
bitumen. Most
is still stored in the existing ponds and is subject to microbial activity on
residual bitumen
that results in methane release to the air.
Today, bitumen froth flotation averages 95% bitumen recovery from average
mined oil sand ore. For a modern plant now producing 200,000 barrels of
bitumen per
day, the ore contains 210,526 barrels of bitumen and 10,526 barrels are
discarded per day
(again 4 million barrels of unrecovered bitumen per year at each commercial
mined oil
sand extraction plant).
Oleophilic separation of bitumen from a tailings pond does not require slurry
preparation. It is less expensive than mining and processing new oil sand ore
and can be
very profitable, provided that oleophilic separation technology is used. It is
the only pro-
cess with proven fast and high recovery of the lost bitumen to date.
Oleophilic separation
has proven in parallel field pilot plant studies that it is the only process
that can efficient-
ly retrieve discarded bitumen from the tailings ponds at low cost. Doing so
will result in
the production of a very large amount of bitumen due to the many years of
fluid
fine tailings (FFT) accumulation in current tailings ponds. Fresh tailings
that are a few
years old will yield high quality bitumen and mature tailings a few decades
old may yield
a lower quality bitumen as a result of pond microbial activity. Bitumen
matured for
many years in a pond is still a valuable hydrocarbon that can be converted
into useful
products.
OLEOPHILIC SEPARATION AS A PRIMARY SEPARATION PROCESS
An even greater commercial advantage can be obtained if oleophilic separation
replaces bitumen froth flotation of oil sand slurry altogether. The reason is,
while the res-
4
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
idence time for commercial bitumen froth flotation is about 6 hours (360
minutes), pilot
plant studies have shown that oleophilic separation requires between 2 and 15
minutes to
achieve the same degree of slurry separation but with better purity of bitumen
product.
Reducing total processing time from 360 minutes to, for example, 10 minutes
will result
in a time reduction factor of 36 with the added advantage of reduced apparatus
and ener-
gy costs and superior bitumen product.
This potential reduction in commercial processing time makes it commercially
at-
tractive to consider producing and processing oil sand slurry right at the
mine face. Do-
ing so will eliminate the need for abrasive slurry pipeline transport from
mine to central
processing plant and for abrasive tailings pipeline transport back to the
mined out area for
settling and dewatering, and it will eliminate long duration tailings ponds
that have been
known to leak toxic liquid into the surrounding landscape. These, and
subsequently here-
in described additional benefits will result in a much lower carbon footprint
for bitumen
extraction, fewer environmental concerns, and more cost effective commercial
mined oil
sand processing.
BASIC PROCESS DIFFERENCES
Bitumen froth flotation needs specific chemical feed conditioning and careful
process control to convert the oil sand ore into a suitable slurry and to
cause bitumen con-
tained in that slurry to be dispersed into tiny droplets or flecks at an
elevated temperature.
Twenty years ago, that temperature was close to 100 degrees C. Today, that
temperature
has been reduced to 50 degrees. At lower process temperatures, froth flotation
becomes
unprofitable due to degradation of percent bitumen recovered. Large amounts of
com-
pressed air at high pressure are needed to form a multitude of small air
bubbles during
pipeline transport of oil sand slurry, and more compressed air at lower
pressure may be
needed in the separation vessels of froth flotation. Careful pH and chemical
control is
needed for bitumen to air adhesion and the froth flotation process steps
require several
process recycle loops to achieve the desired product purity and recovery, as
shown in
Figure 8. A multitude of small air bubbles, each with small flecks of bitumen
on their
surfaces, slowly rise through a downward flowing slurry to reach the top of
froth flotation
5
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
vessels. Bitumen droplets that are weighted down too much by mineral fines
leave the
process with the tailings and are not captured as product. The more processing
time and
air is spent on flotation, the higher the bitumen yield but at the expense of
time, energy
and larger and greater numbers of separation vessels. Considerations of
economy normal-
ly limit froth flotation time to 6 hours.
The bitumen product of froth flotation is high in air and water content and
must
be deaerated before it can be purified by solvent extraction or by dilution
centrifuging.
This adds a bit more to the 6 hours of commercial processing needed to obtain
the desired
95 percent recovery of good quality bitumen from an average mined oil sand
ore.
OLEOPHILIC SEPARATION
Similar to bitumen froth flotation, oleophilic separation also requires a
water
based feed for separation. But to recover bitumen from the feed, caustic soda
is not
needed for almost all grades of oil sand and neither is bitumen to air
adhesion needed.
Instead, the oil sand slurry is tumbled inside a rotating cage (agglomerator)
in the pres-
ence of oleophilic balls or oleophilic rods. For short cages oleophilic balls
are needed but
when the cage length is more than 50% greater than its internal diameter,
oleophilic rods
may be used, which are much easier to obtain and more economical to produce
than balls
per volume (See Figure 18).
To avoid thinking that an oleophilic agglomerator for separating bitumen from
oil
sand slurry or from FFT is similar to a ball mill or a rod mill for grinding
mineral ore, the
following differences should be understood. Ball and rod grinding mills are
very heavy
and only use dense solid balls or dense solid rods of abrasion resistant metal
to crush and
break up ore and gangue by attrition into small particles. This is followed in
subsequent
froth flotation equipment to partition crushed gangue from crushed ore by
mineral partic-
ulate froth flotation. Neither the balls nor the rods have to be oleophilic to
function in a
ball or rod mill. Ball and rod mills require large amounts of power to crush
mined rocks.
None of this is similar to oleophilic separation of oil sand slurry or fluid
fine tailings
(FFT). Unlike the power required to turn a ball mill or a rod mill, an
agglomerator cage
6
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
demands much less power (see Figure 18) and ball or rod mills do not do any
oleophilic
separation.
During agglomerator cage rotation oil sand slurry feed or tailing pond fluid
fine
tailings feed intimately mix with oleophilic balls or oleophilic rods tumbling
inside the
rotating cage. This causes the balls or rods, in a few minutes, to collect on
their surfaces
nearly all the bitumen from the feed in the form of adhering viscous bitumen
phase paste
at the operating agglomerator cage temperature. This paste has a viscosity
similar to con-
ventional ketchup, peanut butter or tooth paste. The bitumen paste accumulates
on the
tumbling balls or rods in increasing thickness and transfers from tumbling
ball to turn-
bling ball or from tumbling rod to tumbling rod until arriving at the
apertured agglomera-
tor cage wall along the cage bottom. There, the paste extrudes through the
apertured cage
wall and between oleophilic wraps for adhesion to the oleophilic cable wraps
covering
the cage bottom quadrants. Aqueous phase effluent outflow from which bitumen
has
been removed flows out of the cage bottom quadrants past the rope wraps until
bitumen
accumulation on the wraps becomes so great that it progressively prevents such
aqeous
outflow along the revolving cage circumferential wall in the direction of wall
movement.
For a counter clockwise rotating agglomerator cage, most of the bitumen
reduced aque-
ous phase leaves the cage through its left bottom quadrant and most of the
bitumen paste
adheres to rope wraps along cage right bottom quadrant for conveyance to
internally
heated roller(s) above the cage to produce a free flowing warm bitumen product
of sepa-
ration. For a clockwise rotating cage, most of the bitumen reduced aqueous
phase passes
out of the cage bottom right quadrant.
Thus the bottom half at least of the agglomerator cage apertured
circumferential
wall is covered by an oleophilic sieve in the form of multiple closely spaced
oleophilic
rope wraps, that allow aqueous phase leaving the cage to pass though narrow
spaces be-
tween the wraps along the cage bottom until the spaces are closed off by
bitumen paste
adhering to the wraps. This closing off by bitumen paste progresses in the
direction of
apertured wall movement and may result in a thick cold bitumen paste coating
on the
wraps conveying collected bitumen to the heated roller(s) above the
agglomerator cage.
When that bitumen coating on the wraps becomes too thick, cage rotation speed
must be
increased to allow sufficient bitumen reduced aqueous effluent to leave
through cage bot-
7
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
torn instead of spilling out of one or both top quadrants of the cage. Hence,
cage RPM is
one area of control and process optimization. When the aqueous outflow is
found to be
too fast, process residence time must be increased. This is a design issue
that can be
overcome by adding an extra rope wrap between adjacent hoops , by selecting
larger di-
ameter wraps or in some cases by increasing paste viscosity through lowering
feed tem-
perature. A temporary solution to solve excessively short feed residence time
is to recy-
cle the aqueous effluent back to the agglomerator until the problem is
overcome. When
the aqueous outflow is slow, bitumen recovery is high but at the expense of
processing
time. Increasing cage RPM deminishes the amount of bitumen paste accumulating
on the
cable wraps for a given feed, which also influences the rate of aqueous phase
outflow.
Internally heated rollers mounted above the agglomerator heat the bitumen
paste
on the oleophilic sieve (rope wraps) and cause it to flow off the oleophilic
sieve and roll-
ers as a warm, free flowing, liquid product that does not contain air and is
low in water.
Condensing low pressure steam inside the rollers often is used to heat the
rollers to pre-
vent overheating of bitumen or rope wraps. Hot water or warm oil may be used
as well to
internally heat the rollers. Overheating the wraps may evaporate water in
bitumen on the
wraps and may deposit minerals on the roller and wrap surfaces, which is not
desirable.
OTHER ADVANTAGES
The bitumen product of oleophilic separation of oil sand slurry or of tailings
pond
FFT may be cleaned by washing it with water and reprocessing it by another
oleophilic
separator cage to remove trapped hydrophilic minerals from the bitumen product
The
resulting bitumen product normally is much superior in quality to current
deaerated
commercial bitumen froth. Bitumen loss resulting from such water washing is
very low.
Feed processing by oleophilic separation is much faster than by bitumen froth
flo-
tation, as detailed in the tabulated result detailed in the present patent
specifications and
the bitumen product quality is superior. Another benefit of oleophilic
separation is that
normally the feed does not have to be heated above ambient temperature (e.g.
room tern-
perature). FFT obtained from large tailings ponds (normally at 12 degrees
centigrade
year round) does not need to be heated in winter or summer, provided the FFT
is pumpa-
8
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
ble. For that reason a patent was granted to the present inventor for an
efficient low tur-
bulent hydraulically driven immersible pump for cold FFT (containing viscous
bitumen)
pumping at high pressure from a tailings pond for processing.
Oleophilic separation is very energy efficient because feed volume always is
much greater than bitumen product volume and specific heat of feed always is
higher
than specific heat of product. The product only is normally heated to suit
subsequent
processing but the feed is not heated or only sparingly. None of these thermal
energy ef-
ficiencies are possible with bitumen froth flotation, which requires 50
degrees C in all its
flow and recycle loops. This difference will compute into major energy savings
and ma-
jor carbon credit savings when oleophilic separation is used commercially
instead of bi-
tumen froth flotation.
BRIEF DESCRIPTIONS OF THE DRAWINGS
Figure 1 is a drawing of a small pilot plant that was used to recover
bitumen from mined oil sand ore by oleophilic separation after previous
extensive bench
scale testing.
Figure 2 is a spread sheet of two performance run results with the original
pilot
plant of Figure 1 and of a performance run after major apparatus and process
improve-
ments.
Figure 3 is a drawing of a larger oleophilic separation pilot plant that was
used to
separate bitumen from tailings pond sludge (fluid fine tailings or FFT)
shipped by high-
way tankers from Fort McMurray to Edmonton. After successfully processing
about
40,000 kg of FFT, this pilot plant was skid mounted and shipped to the field
adjacent to
the tailings pond of a commercial oil sands extraction plant and there
obtained the same
separation results as in Figure 3.
Figure 4 is a flow diagram of a large bitumen froth flotation field pilot
plant that
was constructed and modified after preliminary testing to recover bitumen from
tailings
pond FFT for comparing the results of bitumen froth flotation with the results
of oleo-
philic separation in the field.
9
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
Figure 5 is a comparison table of test results obtained when processing FFT by
bi-
tumen froth flotation compared with oleophilic separation.
Figure 6 is a table of oleophilic separation results compared with results of
bitu-
men froth flotation of Syncrude type oil sand beach ore. This ore is
considered to be one
of the most difficult mined oil sands to separate. Single stage oleophilic
separation in this
table is compared with bitumen froth flotation using standard three stage pot
tests devel-
oped by the oil sands industry to correctly simulate commercial bitumen froth
flotation.
Figure 7 is a design footprint diagram of a typical commercial bitumen froth
flota-
tion mined oil sands plant for processing 8000 tonnes of mined oil sand ore
per hour at
50 degrees C.
Figure 8 is a flow diagram of the design bitumen froth flotation commercial
plant
of Figure 7 showing vessels, pumps, cyclones, recycle flows residence times
and power
requirements.
Figure 9 illustrates a proposed concept described in the present patent of
replacing
a remote central large commercial bitumen froth flotation plant with multiple
oleophilic
separators located adjacent to mine faces using working tailings ponds of
short duration
to maximize bitumen recovery, reduce fresh water requirements for extraction
and make
mined oil sands extraction more profitable and environmentally acceptable.
Figure 10 illustrates a typical oleophilic separator using roller chain and
sprockets
to drive rotation of an oleophilic separator agglomerator cage with closely
spaced multi-
ple wraps of endless rope on the cage to collect cold viscous bitumen phase
onto the
wraps for conveyance to heated rollers to produce a free flowing warm
unaerated bitu-
men product of superior quality with high bitumen recovery.
Figure 11 illustrates the use of multiple wraps of endless rope with rope
redirect-
ing mechanisms between rollers and also between cage and roller to prevent
rope wrap
from rolling off the end of roller(s) or cage. Two endless ropes and two
redirecting mech-
anisms are used with the cage of Figure 11B. Using two endless ropes is a
convenient
method to prevent cage tilting as it rotates.
Figure 12 illustrates a typical oleophilic separator using gears on hydraulic
motors
to drive rotation of an oleophilic separator agglomerator cage.
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
Figure 13 illustrates an inside view of an oleophilic separation agglomerator
cage
partly filled with oleophilic balls or oleophilic rods of three densities,
which various den-
sities are indicated by line thickness in the dawing Wrap of endless rope
(dashed line) is
shown to contact the full circumference of bottom half of apertured cage
circumferential
wall between adjacent hoops (only one hoop shown). The hoops are tied to the
agglom-
erator end walls (not shown) by structural rods to form an apertured
circumferential cage
wall to accept closely spaced rope wraps on cage surfaces between adjacent
hoops.
Figure 14 A illustrates the use of bobbins on a mechanical rod to
equidistantly
spaced hoops and rope wraps between adjacent hoops of an agglomerator cage to
control
aqueous phase flow out of the cage. Figure 14 B shows a mechanical rod welded
to
hoops and with tabs welded to the mechanical rod to equally space rope wraps
between
adjacent hoops to control the outflow of aqueous phase from the cage and hence
to con-
trol the residence time of feed in the agglomerator cage. Mechanical rods are
attached to
the end walls (not shown) or pass through holes in the end walls for
attachment. Figures
14C and D illustrate typical tabs suitable for welding to mechanical rods or
pipes.
Figure 15 A shows ribs to assemble into hoops that may be welded to mechanical
rods or pipes as shown in Figure 15B. To save on the cost of steel and on the
cost of rib
cutting, the size of the inside radius of each rib is the same as the size of
its outside radi-
us. This is accomplished by shifting the center of rotation as shown in Figure
15A, by
numbered tags 163 and 164, which means that the inside radius cut of one rib
is also the
outside radius cut of the next rib, as shown in the cutting pattern of Figure
15C. Since
long commercial agglomerators may require many large diameter ribs this will
reduce the
cost of producing agglomerators. Notches (162 of Figure 15 A) are cut into the
rib inside
radius for welding to pipes or rods but that cutting is easily done as part of
continuous
abrasive water cutting of ribs.
Figure 16 provides approximate horsepower calculation for driving agglomerator
cages shorter than 1.5 times cage internal diameter. Oleophilic balls normally
are used in
short cages but oleophilic rods are normally used in longer cagess for reasons
of econo-
my.
11
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
Figure 17 identifies cage quadrants for counter clockwise and clockwise
turning
agglomerators. Hoppers for introducing feed into an agglomerator preferably
are located
in or close to quadrant 4.
Figure 18 provides information for use of sealed conventional steel pipe as
oleo-
philic rods inside an agglomerator longer than 1.5 times its inside diameter.
It also iden-
tifies plastic pipe for use as oleophilic rods and describes the use of balls
and of golf balls
loaded with steel inserts for use as oleophilic balls in agglomerators of
shorter length.
Figure 19 illustrates potential replacement of conventional mined oil sand
condi-
tioning drums with rotating grizzlies to digest mined oil sand with condensing
steam and
remove rocks, gravel and clay lumps from grizzly end whilst digested wet sand,
silt and
bitumen pass out of the grizzly between grizzly bars to subsequently mix with
water and
thus produce oil sand slurry feed to an agglomerator for oleophilic
separation.
Figure 20 illustrates why axial feed distributors have now been replaced by
feed
hoppers along top quadrants of agglomerator cage surfaces not covered by rope
wraps.
DETAILED DESCRIPTION OF THE FIGURES
Figures lA and 1B are historical drawings of a small scale pilot plant that
was
used to recover bitumen from mined oil sand by oleophilic separation. The
drawings il-
lustrate the equipment that was used to develop oleophilic separation after
its potential
for commercial oil sand processing had been confirmed by several years of
extensive
bench scale testing. Mined oil sand ore entered a rotating drum called a
conditioning
drum, or slurry tumbler, where the ore was mixed with warm water to produce a
slurry.
In early testing, caustic soda was added to the mix since that was, and still
is, the conven-
tion in commercial oil sand extraction. Later, caustic soda was eliminated
since it did
not appear to help in oleophilic separation of high and medium grade oil sand
slurries.
Only for the very low grades of oil sand ore, normally commercially bypassed
and not
processed, did sodium hydroxide appear to have some use. Separation
temperature was
gradually lowered as test work with the pilot plant of Figure 1 proceeded. A
screen, at the
exit of the conditioning drum, removed gravel and then the slurry was pumped
to a rotat-
ing agglomerator. This agglomerator was a 30 cm. diameter drum, 6 cm. long,
filled 80%
12
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
with a bed of half inch (13 mm) steel balls with an exit screen that prevented
balls from
leaving the agglomerator drum rotating at about 6 RPM. Slurry feed entered the
rotating
drum and balls, mixing with the feed, stripped finely dispersed bitumen
particles from the
slurry and returned enlarged bitumen particles to the mixture before it left
the drum as the
exit stream. This exit stream spilled onto an endless mesh belt that was made
from oleo-
philic longitudinal and perpendicular strands with mesh opening size about 4
mm square.
The revolving belt collected the enlarged bitumen phase particles onto the
oleophilic
strands whilst the aqueous phase of the slurry passed through the mesh belt
openings.
Augers under the belt removed sand, silt and water from the separator as
illustrated in
Figure 1B. Rubber rollers squeezed the mesh belt and removed the resulting
bitumen
phase product. The oil sand feed, the oversize, the tailings and the bitumen
product were
measured by load cells and were analyzed immediately in an upstairs analytical
laborato-
ry as each test run proceeded. For the analyses, conventional soxlet
extraction apparatus
was used to weigh the extracted minerals. Rotovacs were used to remove toluene
from
the warm extraction flasks, followed by weighing pure bitumen left in the
tared flasks.
Water collected in the apparatus graduate was recorded. This obtained accurate
and
complete analytical data of minerals, bitumen and water content of each sample
within 8
hours after completing each pilot plant run. The results of our laboratory
analyses were
routinely checked for accuracy against commercial laboratory results of
identical samples
taken at the same time.
Figure 2 is a spread sheet of performance run results with the pilot plant of
Figure
1, giving the results of three performance runs to give an indication of
progress made dur-
ing an 2 year program of testing and apparatus improvement. As shown in the
previous
Figure 1 A , the oleophilic separation pilot plant program used only one stage
of separa-
tion and not three stages as is common for bitumen froth flotation. Bitumen
recovery
from slurry in this one stage pilot plant was 70.0%, 99.9% and 96.0% for the
three per-
formance runs here tabulated. The ore that produced these slurries contained
respective-
ly: 6.8%, 12.3% and 9.1% bitumen. The product from these three runs
respectively
contained 44%, 49% and 46% bitumen. An encouraging result from this test
program
was that slurry processing took only 4.5, 9.1 and 9.7 minutes respectively for
the three
performance runs.
13
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
The positive results and the short processing time obtained well before the
process
had developed to the point of combining agglomeration with oleophilic
screening in one
apparatus, indicated the need for continued research. As shown in Figure 8,
commercial
bitumen froth flotation of oil sand slurry normally uses three stages of
separation and
takes at least 6 hours (360 minutes) of processing. The average oleophilic
separation
time was 7.8 minutes for the performance test runs tabulated in Figure 2. As
development
proceeded, separation time progressively became shorter without performance
degrada-
tion.
Bitumen recovery in performance run 25 was very high for low grade 9.1 % bi-
tumen content oil sand ore as a result of several minor changes that were made
in the
flow diagram of Figure 1. In previous test runs, fresh water was used to
produce and di-
lute the oil sand slurry. In process development that led up to performance
run 25, pro-
cess water containing residual bitumen and fine particulate minerals was
recycled back
continuously to make slurry. Only enough fresh water was added to replace
water con-
tamed in moist tailings that were augured out of the bottom of the separation
effluent
tank, using a vertical Moyno pump with long exposed auger to remove moist
tailings.
Fresh water added to produce slurry only replaced water that left the process
in moist tail-
ings augured out of the bottom of the tailings tank under the oleophilic
sieve. This was
achieved by replacing the tank and augers of Figure 1 with a deep cone
tailings tank with
a vertical Moyno pump in the bottom and a moist sand conveyor below the Moyno
pump
to return any run off water back to the process. The recycle liquid at the end
of run 25 had
the same bitumen and fines concentration as at the beginning of the run.
Stirred aqueous
recycle liquid stored from previous test runs was used as process water
recycle at the start
of run 25. The only bitumen that left the process was bitumen contained in the
moist tail-
ings.
The results of performance run 25 suggested that tailings ponds could, in
time, be
eliminated if oleophilic separation were commercialized. Permission was not
granted by
AOSTRA (Alberta Oil Sands Technology Research Authority) to publish these
results in
the scientific literature until recently. The inventor had consistently
refused to assign his
technology to the Alberta government in return for potentially large
government research
contracts. To have assigned his patents would have left him with no future as
a researcher
14
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
or manager and could very well have stopped any further oleophilic separation
develop-
ment.
Figure 3 is a drawing of an oleophilic separation pilot plant that was used to
sepa-
rate bitumen from tailings pond sludge. Mention should be made here that pond
sludge
(fluid fine tailings or FFT) mainly differs from oil sand slurry in being very
low in sand
content and high in mineral fines content. The original name of "sludge" was
changed
by the oil sands industry to "fluid fine tailings" or FFT, possibly to make it
more appeal-
ing in the scientific literature. Figure 3 includes a tabulation of the
results of the perfor-
mance run of oleophilic separation that processed 8000 kg of FFT (fluid fine
tailings) us-
ing an agglomerator, that was 1.108 meter in diameter and 0.095 meter (9.5
cm.) long.
Optimum processing time was 1.54 minutes. This test program more than exceeded
the
required performance criteria stated at the beginning of the contract. Bitumen
recovery
from FFT was 85% and the product contained 58% bitumen, 15% mineral and 27% wa-
ter. For processing times shorter than 1.54 minutes bitumen recovery dropped
below
85%. When the bitumen product of Figure 3 was washed with fresh water and then
pro-
cessed by single stage oleophilic separation, bitumen loss was less than 1%
and the bitu-
men product then contained 61% bitumen, 10% mineral and 29% water. Assays
showed
that the mineral in the reprocessed product was mainly oleophilic, which
suggested that
water washing is a convenient method of beneficiating any valuable oleophilic
particulate
minerals contained in oil sand bitumen product, such as rutile or zircon.
The results in Figure 3 are similar to those obtained in the previous pilot
plan pro-
gram at an earlier date for processing oil sand slurry but which then averaged
7.8 minutes
of processing time detailed in Figure 2. Comparing these historical test
results gives an
indication of development improvements made in oleophilic separation during
many
years of continued research. Noteworthy is that the 85% bitumen recovery in
1.54
minutes was achieved using a single oleophilic separator.
Figure 4 is a flow diagram of the bitumen froth flotation field pilot plant
that was
used to recover bitumen from tailings pond FFT in the comparative field test
program
described with Figure 3. The flow diagram was taken from the Suncor (Collins
and
Webster) handout describing the field study that was not published in a
scientific journal
but was only presented at a Fort McMurray local CIM meeting. This bitumen
froth flota-
1 5
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
tion field pilot plant required five stages of processing. It included air
blowing to drive
off carbon dioxide to change the pH of the FFT feed from 6.8 to 8.0 before
froth flotation
would work and after doubling the FFT water content. Then, an oleophilic
separator was
added at the tail end of the bitumen froth flotation pilot plant to remove air
and sufficient
water from the froth product before it was suitable for dilution centrifuging.
In contrast,
the bitumen product of oleophilic separation (detailed in Figure 3) was
dilution centri-
fuged as produced without the need for deaerating or dewatering.
Figure 5 is a comparison table of the field test results of processing FFT by
oleo-
philic separation and of processing FFT by bitumen froth flotation in the
comparative
field test program of Figures 3 and 4. Noteworthy is that oleophilic
separation took less
than 2 minutes of processing time and bitumen froth flotation took 26 minutes
to achieve
the same 85% bitumen recovery. The amount of effluent leaving oleophilic
separation
was 90% of the FFT feed entering the separator, indicating that the amount of
effluent
produced was less than the amount of FFT feed processed. Bitumen froth
flotation of
FFT, because of the need for dilution water, yielded an effluent equal to 152%
of the
FFT entering the field pilot plant. A commercial pond to receive that effluent
would have
needed to be 52% larger than the commercial pond that contained the original
FFT feed.
Another difference between results of oleophilic separation and results of
bitumen
froth flotation of FFT in these field tests was in the quality of the bitumen
product. From
single stage oleophilic separation the product contained 58% bitumen. The
bitumen
product from five stages of separation in the field pilot plant of Figure 4
contained 24%
bitumen; less than half as good as the product of oleophilic separation of the
pilot plant of
Figure 3. Unfortunately the $3 million oleophilic separation pilot plant was
never re-
turned to the inventor.
Figure 6 is a table comparing oleophilic separation with bitumen froth
flotation of
Syncrude type oil sand beach ore, which was studied with the assistance, under
AOSTRA
contract, of consultants of the Alberta Research Council, Dr. Dean Wallace and
Ms.
Deborah Henry. Syncrude type beach ore was tested in this comparison since it
is con-
sidered to be the most difficult mined oil sand to separate. Pilot plant
oleophilic separa-
tion was carried out by staff of the present inventor and was closely observed
by Wallace
and Henry. Parallel pot test were conducted by Wallace and Henry using the
identical oil
16
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
sand ore. The results of the comparison tests are shown in the table of Figure
6. The ole-
ophilic separation pilot plant used one stage of separation. Pot tests had
been developed
by the mined oil sands industry and were considered to be representative of
commercial
three stage froth flotation at the time when these comparison tests were
conducted. As
shown in Figure 6, the pot tests collected three products of bitumen froth
flotation: pri-
mary tailings, secondary tailings and a toluene wash. Total bitumen recovery
by the pot
tests was18% and total bitumen recovery by oleophilic separation was 67%
representing
an almost 4 fold improvement. The product quality from the pot tests was 5.4%
bitumen
and from oleophilic separation was 43.5% bitumen representing an 8 fold
improvement.
Clearly, single stage oleophilic separation results were superior to three
stage bitumen
froth flotation pot tests of the same low quality ore.
Figure 7 is a design footprint diagram of a typical modern commercial bitumen
froth flotation mined oil sands extraction plant for processing 8000 metric
tons of mined
oil sand ore per hour as designed by staff of the University of Saskatchewan
with advice
and help from industrial oil sand consultants. For the three stages of froth
flotation, pri-
mary separation occupies 1250 square meters, secondary separation 2500 square
meters
and tertiary separation 6400 square meters. Then, to minimize the
environmental impact
of extraction tailings, the tailings thickeners occupy 48,600 square meters,
much greater
than the extraction plant itself. A commercial plant, that large, requires a
large amount of
real estate and necessitates the use of central processing of oil sand slurry
and long dis-
tance expensive slurry pipelines to bring oil sand slurry to the central
extraction plant.
These pipelines are fabricated from abrasion resistant steel and need to be
rotated about 3
or 4 times per year and normally have less than two year lifespan. The
resulting extrac-
tion tailings then have to be removed by slurry pipeline from the central
plant to tailings
ponds some distance away when sterilization of bitumen in ore bodies by
covering the
ore with tailings ponds is not allowed. All this requires much and complex
equipment and
major investment.
Figure 8 is a flow diagram of the bitumen froth flotation plant of Figure 7.
Note-
worthy is that this modern plant design requires 26 (twenty six) very large
vessels, 40
(forty) large capacity pumps and 20 (twenty) hydro cyclones to process 11,000
cubic me-
ters of slurry per hour, all at a constant temperature of 50 degrees
centigrade. The separa-
17
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
tion vessels alone require 450 kw of power, not counting the 40 pumps nor the
high pres-
sure air compressors needed to inject flotation air into slurry pipeline
feeding the plant,
nor steam needed to keep the complete process at a constant 50 degrees C. This
drawing
shows that several recycle loops are needed to obtain enough bitumen product
of ac-
ceptable quality and to minimze loss of bitumen to the effluent tailings.
Figure 9 introduces an option where oleophilic separation can simplify commer-
cial mined oil sand extraction and make it more cost effective. The Figure
illustrates a
concept herein proposed of replacing central large commercial bitumen froth
flotation
plants, such as the plant design of Figures 7 and 8, with a series of
oleophilic oil sand
slurry separators located adjacent to oil sand mine faces. This proposal
involves the use
of short term working tailings ponds to significantly increase total bitumen
recovery and
to produce non or less toxic tailings water for reuse right on site for more
oil sand slurry
production or for disposal.
Thus current commercial plants use 6 hours of 50 degree centigrade aerated
froth
flotation in very large central extraction plants with feed from many km of
long abrasive
slurry pipelines followed by pipeline disposal of abrasive tailings effluent
slurry. That
costly approach to oil sand processing should in time be replaced with
multiple oleophilic
separators that require less than 10 minutes of ambient temperature processing
to achieve
the same or better bitumen recovery and better product quality close to the
mine face. It
is a concept well worth considering to reduce the cost and environmental
impact of mined
oil sand extraction. Illustrative of this concept is Figure 9. Oil sand slurry
(1) is separated
near a mine face using an oleophilic separator (2). Aqueous effluent (3) of
oleophilic sep-
aration flows into a temporary tailings pond (8) and after a few months sand,
silt and
some fines will have settle to the bottom (9) of the pond (8) since tailings
pH will be
close to neutral in the pond (8). Bitumen product (4) of the separator (2) is
pumped or
gravity fed to the inlet of a pump (14) where it is augmented by fresh water
(7) entering
the pump (14). Then the bitumen product (4) and the fresh water (7) are pumped
by the
pump (14) into a liquid pipeline (15) to cause flow of liquid (16) in
turbulent flow
through the pipeline (15) on its way to a central bitumen processing plant.
Turbulent flow
in the pipeline (15) causes dispersion of bitumen (4) into water (7) and
results in the
transfer of hydrophilic minerals from the bitumen phase to the water phase in
the turbu-
1 8
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
lent mixture flowing in the pipeline (15). An emulsifier (not shown) may be
added to the
pump inlet to assist in dispersing the bitumen (4) if required. As the thus
dispersed mix-
ture arrives at the central bitumen processing plant, an oleophilic separator
may be used
to separate liquid leaving the pipeline into a water washed bitumen product
and an efflu-
ent water containing most of the hydrophilic minerals that were present in the
bitumen (4)
that entered the pipeline (15). A demulsifier (not shown) may be added to the
dispersed
pipeline mixture before or during oleophilic separation at the central bitumen
clean up
facility, if needed.
A few months later, or a year or more later, the same or a different
oleophilic sep-
arator (5) may be used to separate the upper layers (10) of the pond. Sand,
silt and some
fines of the effluent (3) of prior separation will have settled to the bottom
(9) of the pond
(8). Bitumen content in the FFT (10) upper pond layers will have increased
dramatically
as a result of sand, silt and fines settling. These upper pond layers may then
be separated
by the same or by an alternate oleophilic separator in providing a feed (6)
for entry into
that separator (5). The resulting bitumen product (12) of FFT separation then
flows into
the same or similar pump (14) as a feed mixture (13) with added water (7), as
was done
months or a few years previous, to produce a turbulent mixture (16) flowing in
the pipe-
line (15) of wash water (7) mixed with bitumen product (12) of the oleophilic
separator
(5).
Since bitumen is recovered from both oil sand slurry and from the resulting
FFT
after settling, the net result of this proposal will be very high total
recovery of bitumen
right near the mine face, yielding a high purity bitumen product at the
central bitumen
clean up plant (not shown). The aqueous effluent (11) of the second separation
will be
very low in bitumen content and low in relatively coarse particulate mineral
content and
will be more environmentally friendly than the contents of the tailings ponds
of the pre-
sent commercial oil sand extraction plants since caustic was not used and more
bitumen
has been recovered. The aqueous effluent (11) leaving the separator (5) may
then be con-
sidered for use directly to produce more oil sand slurry near the mine face
for oleophilic
separation, since oleophilic separation is very tolerant of fines.
Alternately, this effluent
may be centrifuged, cycloned or stored in a short duration pond for a few
extra years be-
fore use or acceptable environmental disposal. The absence of caustic reaction
products
19
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
in that effluent will make it less environmentally sensitive and particulate
fine minerals
mud of that pH neutral effluent will settle relatively fast to encourage water
reuse or rec-
lamation.
The oleophilic separators (2 and 5) of Figure 9 are described in more detail,
for
example, in Figures 10, 12 and13.
Separating mined oil sand slurry to recover bitumen by froth flotation,
followed
by separating the resulting effluent a few months or a few years later by
oleophilic sepa-
ration to increase total bitumen recovery from oil and ore is one of the
claimed objectives
of this present patent.
A number of other devices were developed since 1975 to augment oleophilic sep-
aration for which patents are pending or were granted to the present inventor
to make
oleophilic separation more effective. These included: 1) a confined path hydro-
cyclone
to more effectively water wash and remove rocks, gravel and coarse sand from
oil sand
slurry, 2) a hydraulically driven high pressure positive displacement pump to
transfer
FFT in laminar flow to prevent dispersion of bitumen when pumping fluid fine
tailings
from tailings ponds, 3) much time was spent on developing long lasting
oleophilic
sieves and this led to the use of multi wraps of endless rope to replace mesh
belts that
were fragile and problematic. 4) the redirecting of rope wraps to prevent them
from roll-
ing off cages or rollers was another development. The last two developments
eliminated
the use of fragile mesh belts and replaced these with long lasting steel or
plastc rope
wraps to cover commercial welded separator cages fabricated from metal ribs to
form
hoops for the cages, and use long oleophilic rods instead of oleophilic balls
for tumbling
inside long commercial cages during separation.
The herein described use of closely spaced rope wraps on agglomerator cage bot-
tom quadrants and beyond, serve to very closely control aqueous phase outflow
from the
cage to precisely control the residence time of feed in the cage for
processing. This was
an issue not detailied in previous patents.
One oleophilic separator design is detailed in end view in Figure 10
comprising
an agglomerator cage that is driven to rotate by means of roller chains (31),
connecting
sprockets (32) mounted on each cage end wall (22), and with sprockets (34) on
a drive
shaft mounted in bearings on the frame (29) of the separator. Lever arms (33),
with bear-
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
ings on both ends, provide tension in the roller chains (31) to keep the
agglomerator cage
from swinging as it rotates. An air or hydraulic cylinder (35) at both cage
end walls (22)
between lever arm (33) and separator frame (29) may be used to position the
cage for
stringing cable wraps during construction but also may carry some of the
weight of the
cage and its contents during separation. However most of the weight of the
cage and its
contents is normally carried by a multitude of closely spaced rope wraps
(illustrated in
Figure 11 B). One wrap only (25,26) is shown in the Figure 10 draped in
tension over
rollers (27,28) above the cage. One or more holes with cover plates (35)
normally are
provided in the cage end walls (22) to allow loading of the agglomerator cage
with oleo-
philic rods or oleophilic balls. Balls are used when the cage length is less
than 1.5 times
its internal diameter, but oleophilic rods are cheaper than balls and these
are to be used in
large commercial separators when the cage internal length is greater than 1.5
times its
internal diameter. Oleophilic rods do not tumble well in cages that are
shorter in length
than 1.5 its internal diameter. Feed (20) for separation enters the cage
through a hopper
(21) above one of its top quadrants not covered by cable wraps. This provides
for easy
entry of feed into the agglomerator cage without disturbing movement of rods
or balls in
the cage interior. Normally a feed hopper (21) is used the full length of the
cage for uni-
form distribution of feed into the cage. The cage axis should be level and the
feed hopper
(21) also is level to evenly distribute the feed (20) into the agglomerator
cage. During op-
eration, aqueous effluent (23) of separation leaves through the apertured cage
bottom
quadrants through voids between the rope wraps (25,26) along the bottom cage
quadrants
and most of it through the left bottom quadrant for a counter clockwise
turning cage.
Viscous bitumen paste (not shown for clarity of Figure 10 but is shown in
Figure 13) is
pushed by gravity and by tumbling oleophilic rods or balls out of the cage
mostly through
the apertured bottom right quadrant of the counter clockwise rotating cage. It
adheres to
the oleophilic cable wraps along the bottom cage quadrant. This bitumen paste
(shown in
Figure 13 as tag 85) is conveyed by the wraps to internally heated rotating
rollers (27)
above the cage. Sometimes a bit of bitumen falls off the wraps when the cage
rotates too
slowly but it is collected in a catch basin (24) and is recycled back to the
feed hopper
(21). Bitumen paste adhering to the cable wraps (25) is conveyed to the
surfaces of the
internally heated rollers (27) above the cage. Heating of these cages (27)
normally is by
21
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
internally condensing low pressure steam and the condensate normally is
returned to the
boiler that generated the steam. Heat from those rollers heats up the cold
viscous bitumen
paste on the rope wraps and converts it into a free flowing good quality
bitumen product
that flows off the cable wraps (25) to become the product of separation (30).
In Figure 10
a cold roller (28) is used to cool down the cable wraps. Such a cold roller
may or may not
be required, in which case the cold roller (28) could also be a heated roller,
properly en-
closed. Both the hot rollers (27) and the cold roller (28) are shaft mounted
in bearings
securely mounted to the separator frame (29) to be able to completely or at
least partly
support the weight of the agglomerator cage and its contents plus the weight
of the rollers
(27,28) and their contents. The cold roller (28), which may be kept cool by
flowing cold
water inside the roller (28) serves to cool down the rope wraps (26)
sufficiently before
these return to the agglomerator cage, but only if that is needed to prevent
bitumen loss
with the effluent (23) leaving the cage. Alternately, a bank of air fans
impacting cool air
on the cable wraps (26) may serve to cool the cable wraps if required to
prevent warm
bitumen from leaving the agglomerator cage with the effluent (23)..
Figure 11 shows the previously patented concept of using rope wraps in place
of
mesh belts. In this case two endless ropes are used to prevent the
agglomerator cage tilt-
ing during cage rotation. Figure 11 B shows how the cable wraps are redirected
to pre-
vent wraps from rolling off the cage or roller. It shows a cage completely
hanging in
rope wraps supported from well mounted rollers in framed bearings above the
cage. In
previous patents, the concept of using rope wrap spacing to control outflow of
aqueous
phase from the agglomerator cage or drum was not detailed nor claimed.
Method to prevent rotating cage from swinging is not shown in Figure 11 B but
it
often uses a lever bar shown in Figures 10 and 12 for that purpose.
Alternately two roller
chains at each cage end wall, with the appropriate number of sprockets to keep
the roller
chains tight, may be used to prevent cage swinging.
Figure 12 is very similar to Figure 10, except that the cage is driven by
hydraulic
motors using gears (55) that match with gears (52) securely mounted on the
cage end
walls (50) at both agglomerator cage ends. Lever arms (57) at both cage end
walls with
pivot points (56) at the separator frame (51) and bearings (53) at each cage
end wall pre-
vent cage swinging and provide mounting for the required gears(55) driven, for
example,
22
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
by hydraulic motors. One or four hydraulic motors may be used. Two are shown
on each
end wall in the Figure. The lever arms (57) each are provided with a frame
(54) to mount
the hydraulically driven gears (55) and to provide mounting of an air or
hydraulic cylin-
der (59) to the separator frame (51). This cylinder (59) serves to position
the elevation of
the agglomerator cage when stringing rope wraps between the cage and the
rollers
(47,48) above the cage. It may also serve to reduce some stress in the rope
wraps during
operation, if needed, but normally the rope wraps are strong enough and high
enough in
number to eliminate the need for additional cage support. After passing the
heated rollers
(47) the rope wraps (49) return to the cage cylindrical wall. Aqueous effluent
(43) mostly
leaves through the bottom left cage quadrant and a chute (44) may be provided
to direct
the effluent (43) as it exits. As in Figure 10, the agglomerator of Figure 12
is provided
with one or more holes in each end wall to insert or remove oleophilic rods or
oleophilic
balls and are closed with cover plates (58). Similarly, as is shown in Figure
10, aqueous
effluent (43, 44) leaves the separator- mainly from the bottom left quadrant
of Figure 12.
A recycle catch (45) may be provided for recycling any bitumen and associated
aqueous
effluent back to the feed (40) inlet (41). Cold bitumen paste (not shown here
but shown
in Figure 13) is conveyed by rope wraps (46) of Figure 12 to heated rollers
(47) that serve
to heat the viscous bitumen phase adhering to the cable wraps (46) resulting
in a free
flowing unaerated bitumen product (42). As in Figure 10, in Figure 12 a cold
roller (48)
may be used to cool the rope wraps leaving the heated rollers but such a cold
roller is not
always required. Usually a bitumen product receiver (60) may provide some
temporary
storage of bitumen product (42) but is not always needed.
Figure 13 provides for an inside conceptual view of an oleophilic separator as
it
starts to rotate, filled with oleophilic rods of three densities (70, 71 and
72) The denser
rods (70) tend to concentrate along the right bottom quadrant of the
agglomerator cage
(73). The medium density balls (71) tend to gravitate towards the middle of
the cage and
the light balls (72) tend to occupy the upper regions of the agglomerator
cage. During
rotation, the rods intermingle with each other and with the feed (74) inside
the cage,
which feed during operation enters the agglomerator cage from the feed hopper
(75) at
the top. Feed (76) enters when the agglomerator cage (77) rotates and
establishes a liquid
level in the cage. Often there is air space (78) near the top of the cage
inside that may be
23
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
controlled by liquid level sensing to prevent spillage of feed from
overflowing over the
top of the cage (77) . A seal (79) may be provided and rope wrap (81) location
and direc-
tion between agglomerator cage and heated rollers above the cage may be
designed to
minimize such overflow. This may be done by placing the rollers closer
together in the
horizontal direction and further apart in the vertical direction. The
oleophilic rods tumble
inside the cage (77) as the cage rotates and mix with the feed(74) inside the
cage to strip
bitumen phase paste out of the feed (74) for adhesion to the oleophilic rods
(70,71,72) or
balls. Then the aqueous effluent (82) of the feed, after enough bitumen has
transferred to
the balls or rods, leaves through the bottom quadrants of the cage. With a
counter clock-
wise rotating cage, most of the effluent (82) leaves the cage through its left
bottom quad-
rant. Bitumen phase (83) of the feed in the cage that has transferred to the
oleophilic
balls or rods extrudes out of the cage and can be seen to adhere to the cable
wraps along
the bottom right quadrant (73). At times, a small amount of water and bitumen
fall off
the cage along the right bottom quadrant and this may be collected in a
receiver (84) un-
der the bottom right quadrant for return to the feed hopper (75) instead of
becoming part
of the aqueous effluent (82). Viscous bitumen (85) adhering to the rising rope
wraps (80)
is conveyed to the external surfaces of internally heated rollers (86) where
the viscous
bitumen turns into a low viscosity high quality free flowing bitumen product
(87) that
may be collected in a product hopper (88) to become the bitumen product (89)
of oleo-
philic separation. A Squeeze roller (90) may be mounted, if needed, adjacent
to the last
heated roller to strip remaining warm bitumen from the rope wraps before these
contact a
cold roller (91), which roller is one method for cooling down the rope wraps
before re-
turning to the agglomerator cage wall (92), if that is needed. Ambient
temperature water
circulating through the cold roller normally is sufficient. Air cooled rope
wraps do not
readily release thin layers of viscous bitumen from rope wraps when impacted
by cold
aqueous effluent (82) leaving the cage. Thus a cold roller (91) may not be
needed in
many cases. Often ambient air circulating by the wraps or from a bank of air
fans serves
the same purpose as a cold roller. However a roller (91) , if grooved to
accept and keep
rope wraps aligned, in some cases is useful after the heated rollers (86).
Viscous bitumen paste (85) with a consistency of cold ketchup, peanut butter
or
tooth paste is extruded out of the agglomerator cage along its bottom right
quadrant by
24
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
oleophilic balls or rods. It is very difficult to observe the movement of
bitumen or oleo-
philic rods or balls within an agglomerator cage because, if a glass or
plastic wall were
used to observe such movement, bitumen would quickly coat that clear wall and
thus
prevent such observation. However, observing bitumen phase leaving an
apertured wall
with revolving balls in a rotating experimental cage gives a strong indication
that heavier
balls inside the cage extrude viscous bitumen through the cage wall apertures
to the rope
wraps. Bitumen adheres to the oleophilic sieve (rope) surfaces whilst aqueous
effluent
passes through the sieve apertures( between adjacent wraps) to disposal. Along
the bot-
tom right quadrant, the amount of bitumen adhering to the rope wraps
(oleophilic sieve)
progressively increases in the direction of cage rotation and often is very
significant. For
a counter clockwise rotating cage, most of the aqueous phase passes through
the left bot-
tom quadrant and that flow of aqueous phase progressively diminishes in the
direction of
sieve movement along the bottom cage quadrants as bitumen accumulates on the
sieve
(rope wrap) surfaces and progressively closes the sieve apertures (space
between the
wraps) in the direction of cage rotation. As shown in Figure 15, the cage wall
may be as-
sembled from 90 degree ribs to form hoops. In this particular case, mechanical
rods (93)
pass through the ribs to tie the ribs to cage end walls, as shown in Figure 14
A with bob-
bins on the mechanical rods to space the hoops. However, as shown in Figure 14
B bob-
bins are not needed when mechanical rods or pipes are welded directly to the
hoops or
ribs and tabs or pins are welded to or attached to the longetudinal mechanical
rods or
pipes to equidistantly space the wraps. Examples of such tabs are illustrated
in Figures 14
C and D
Figure 14 A illustrates one way of placing rope wraps (111) on mechanical
struc-
ture of an agglomerator. In this case, bobbins (113) on mechanical rods (112)
between
cage end walls space the hoops (114) of an agglomerator cage. The mechanical
rods
(112) are in tension and pass through holes in agglomerator hoops (114) and
through
holes in the end walls where the mechanical rods are terminated with, for
example, nuts.
Rope wraps fill the space between hoops with enough void space between the
wraps and
between wraps and hoops to allow controlled outflow of aqueous phase effluent
for a de-
sired feed processing rate (residence time) in the agglomerator.
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
Figure 14 B illustrates details for another way of constructing an
agglomerator
with apertured cylindrical wall. Again, two end walls (not shown)are provided
with cov-
er plates for holes in the end walls that are used to insert oleophilic balls
or oleophilic
rods into the agglomerator cage. In this case, mechanical rods, pipes or flat
bar (115) are
welded to spaced hoops(117) between end walls, and to end walls to form - like
a cage -
the apertured cylindrical wall of the agglomerator. Tabs (118), illustrated in
Figures 14C
or 14D are equally spaced between hoops, and welded to the rods, pipes or bar,
to proper-
ly space the multiple wraps. These tabs may be half washers, as illustrated,
or one third
or quarter washers with appropriate holes for welding to the rods, pipe or bar
to tightly or
properly space the waps, as required.
The agglomerator cage is driven to rotate but hangs in multiple wraps of one
or
two endless ropes from rollers above the cage that support all or part of the
weight of the
cage and its contents. The multiple wraps, closely and equally spaced in
groups of one,
two, three, four, or more, contact the cage bottom quadrants between adjacent
hoops.
The wraps serve the multi purpose of (A) supporting part or all of the weight
of the cage
and its contents, (B) controlling the rate of outflow between rope wraps of
aqueous efflu-
ent, (C) collecting on the wrap surfaces bitumen paste of such separation, and
(D) con-
veying the collected bitumen paste to internally heated rollers above the cage
for produc-
ing a warm free flowing bitumen product of separation. Many of these purposes
and their
effect on oleophilic separation have not yet been disclosed or claimed in
previous patents
of the present inventor.
Shown in Figure 14 B are one of the mechanical pipes (115) and two hoops (117)
with nine rope wraps (116) between hoops. To prevent bunching up of wraps
(116) be-
tween hoops (117) and thereby creating uneven spaces between adjacent wraps,
tabs
(118) may be welded to the mechanical pipe (115) to separate the nine wraps
into three
groups of three wraps. Of course, other grouping may be used to achieve the
objective of
controlling the outflow of aqueous effluent and hence control the residence
time of feed
in the agglomerator cage. Two types of tabs are illustrated in Figures 14 C
and D. Tabs
(118) like Figure 14 C would be used on pipes or round rods and in simplest
form would
be washers cut in half or in quarters, for example. For square or rectangular
cross section
mechanical rods, tabs similar to the one illustrated in Figure 14D could be
used. To pre-
26
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
vent cage ovality, hoops would likely not be rolled from bar - but would be
assembled
from ribs - to minimize internal hoop stresses resulting from rolling bars and
from weld-
ing to rolled hoops; and to save on cage construction cost.
When oleophilic rods are used instead of oleophilic balls in the agglomerator
cage, the rods are almost as long as the cage and this reduces the number of
hoops needed
for cage construction since, unlike balls, rods can not pass out of the cage
between widely
spaced hoops. The use of oleophilic rods could also favor thicker metal and
wider space
between ribs provided that tabs (118) were used to control wrap spacing
between hoops.
Bitumen progressively accumulates on the rope wraps in the direction of cage
ro-
tation and progressively limits or stops the outflow of aqueous phase in that
direction.
Hence, as a result of viscous bitumen phase accumulation on the rope wraps,
outflow of
aqueous phase is progressively reduced along the cage bottom in the direction
of cage
rotation. Increasing cage RPM reduce the amount of viscous bitumen
accumulation on
each rope wrap and will tend to shorten feed processing time for a cage of a
given diame-
ter and wrap spacing, but could reduce percent bitumen recovered from a feed..
Figure 15 illustrates ribs that may be used to form hoops for long
commercially
constructed agglomerator cages that use oleophilic rods instead of oleophilic
balls to strip
bitumen from the feed. Figure 15 A shows a typical rib with outside diameter
(160) and
with various cutouts (173,162) along the inside rib diameter (161). Noteworthy
is that
the center of curvature (163) of the outside diameter of each rib is offset by
a vertical dis-
tance from the center of curvature (164) of the inside diameter, but the
radius of the out-
side diameter is the same as the radius of the inside diameter. This is done
to save on the
cost of cutting of many ribs that are needed for long commercial
agglomerators, to make
strong but effective cage like agglomerators. As shown in Figure 15 C the cut
of the in-
side diameter of each rib also is the cut of the outside diameter of the next
rib when cut
from a sheet (169) of metal. A small reversal of cut is needed several times
during each
rib cut to remove the required notches (173, 162 of Figure 15A) along the
inside rib di-
ameters. This method of rib cutting minimizes on scrap metal and halves the
cost of cut-
ting ribs. Also abrasive water cutting of holes (See Figure 13, item 93) is
more expensive
than cutting continuous smooth curves (162) as is done in Figure 15 C.
27
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
Figure 15 B shows longetudinal pipes (166), flat bars (168, 167) or welded to-
gether pipe protrusions (165) projecting inward from longitudinal pipes (167,
168) or flat
bars (168) welded to the ribs at the indentations (Figure 15A, items 162, 173)
of each rib
over the length of the cage. These are used when internal protrusions are
needed in the
agglomerator cage to cause oleophilic rod or ball tumbling. In some cases the
protrusion
of pipe only (166) welded to each rib is sufficient.. Location of rope wrap
(176) between
hoops and between pipes (165, 166) welded to ribs (175) is shown by the dashed
line in
Figure 15 B.. The further the pipes (165,166) are apart, the closer the wraps
(176) will be
to the cage interior (174) and to the oleophilic balls or rods.
Figure 16 provides a preliminary method for calculating the horsepower needed
for turning an agglomerator cage partly filled with oleophilic balls mingling
with the feed
inside the cage. The drawing of Figure 16 shows moment arms resisting initial
motion of
cage and balls for cage segments as a function of distance from the cage
center expressed
in radii for a cage that is 2 meters in diameter, The moment arm of torque of
each seg-
ment containing oleophilic balls is shown by vertical arrows to indicate those
segments of
the cage that resist motion due to buoyed ball weight. The top cage quadrant
portion be-
low 0.4 meter radius contributes very little resistance to cage rotation where
the ball den-
sity in that area is close to the density of the feed and where the moment arm
is relatively
short. The two bottom cage half quadrants contribute a positive and a negative
resistance
to motion and these will tend to cancel each other out. The other half of the
cage only
contains balls after the cage is rotating for a cage half filled with balls.
Further calcula-
tion details are provided in Figure 16. For calculating total required cage
rotation horse-
power it is assumed that actual required power is about 4 times the herewith
calculated
brake horsepower to cause ball tumbling in the cage, which accounts for
friction in bear-
ings, gears and rope wraps, etc., and for motor efficiency. These results
appear to be
close to the actual power required in the tested pilot plant equipment but
will need to be
confirmed for large commercial agglomerators using actual rod buoyed densities
in the
cage radial segments for the calculations. Figure 16 introduces an approximate
method
for calculating agglomerator power requirements. A similar procedure may be
developed
for the use of long oleophilic rods in an agglomerator.
28
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
For clarity of disclosure, Figure 17 illustrates cage quadrant segment
numbering
for clockwise and for counter clockwise rotating agglomerator cages. Most of
the aque-
ous effluent leaves the cage in quadrant 2 and most of the viscous bitumen
phase is trans-
ferred to rope wraps in quadrant 3. Feed enters the cage from a hopper along
the top of
the agglomerator cage , usually along quadrant 4. From the hopper, the feed
passes be-
tween hoops and mechanical structural members not covered with rope wraps for
flow
into the agglomerator cage. The preferred location of the center of that
hopper is at an
angle between 45 and 60 degrees indicated in Figure 17. However, feed hopper
centers
between 30 and 120 degrees are also acceptable for introducing feed into the
agglomera-
tor. In Figure 17 the rope wraps leave and return to the agglomerator
apertured wall verti-
cally to connect to rollers above the cage. However, rope wraps may also
leave the cage at an angle to increase the amount of cage surface covered by
rope wraps
and thus reduce the chance of feed overflow from the cage. Normally the
ascending rope
wraps leaving the bottom right quadrant should be close to vertical to prevent
adhering
bitumen from spilling off the rope wraps. However, the descending rope wraps
returning
to the cage may have an angle with vertical close to 45 degrees, and this may
be accom-
plished with an auxiliary roller above quadrants lor 4 to deflect the
descending wraps.
Figure 18 provides density information for using standard capped empty
schedule
steel pipe for use as oleophilic rods in agglomerator cages. To achieve denser
oleophilic
rods, steel pipes may be filled with water or alternately with foamed
concrete, which may
prevent damage to thin walled pipes due to contact with denser pipes tumbling
inside the
agglomerator cage. Plastic pipes may be used instead and these may be provided
with
steel or reinforced concrete cores to achieve a desired rod density and
rigidity. Infor-
mation on using loaded golf balls in short agglomerator cages is shown as well
in the in-
eluded tables. Oleophilic rods should not be used in cages that are shorter in
internal
length than 1.5 times the internal cage diameter since short oleophilic rods
may get stuck
in the cage structure, which will prevent rod tumbling.
Figure 19 introduces the concept of simplifying the production of oil sand
slurry
for oleophilic separation since an agglomerator of the present invention can
serve to dis-
engage bitumen from sand grains. The figure illustrates an apparatus and
method for the
removal of rocks, gravel and clay lumps, that are low in bitumen content, from
mined oil
29
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
sand while preparing sand, silt, fines, water and bitumen mixture to be
separated in an
oleophilic agglomerator cage used for oleophilic separation of feed. Figure 19
is based
on the concept of using live saturated condensing steam to soften bitumen of
an oil sand
ore to allow wet sand, silt and bitumen to pass through a circular apertured
wall of a ro-
tating grizzly, or through apertures of circular perforated drum wall, while
temporarily
retaining rocks, gravel and clay lumps inside the rotating grizzly or drum
until these leave
through the grizzly or drum end. To achieve that objective, the drum or
grizzly is tilted
slightly downward towards the drum end exit. To achieve the desired
separation, spacing
between grizzly bars or perforations in the circumferential drum wall should
be small
enough to retain rocks, gravel and clay lumps while passing sand, silt and
bitumen out
between the grizzly bars or through the drum perforations.
A hydro-cyclone or a confined path hydro-cyclone may be used in addition to
remove and water wash remaining pea gravel and coarse sand sand before the
resulting
product entrs an agglomertor. Figure 19 A and Figure 19 B show an exposed
rotatable
grizzly with a central entrance (128) in its entrance wall (125) and with
support rings
(126) welded to grizzly bars (127) to form the grizzly drum. The support rings
(126) en-
gage with rollers (129) that are driven to cause grizzly rotation. A receiver
(131) is
mounted under the grizzly to accept bitumen, sand and silt mixture as a result
of steam
condensing in the grizzly interior when the grizzly is enclosed to save on
heating steam.
Rocks, gravel and undigested clay lumps (130) leave the drum exit (141) and
water wet
bitumen, fine sand and silt (132) leave the receiver (131) , followed by
oleophilic separa-
tion by an agglomerator of the present invention after removal of pea gravel
and coarse
sand. Figure 19 C shows an internal view of the grizzly, its entrance (128),
its support
rollers (129) and its receiver (131) enclosed in an enclosure (142) with
saturated steam
(140) entering the enclosure (142). Saturated steam (140) that enters the
grizzly enclo-
sure condenses on the grizzly contents (150) to allow sand, silt and bitumen
to pass be-
tween grizzly bars into the receiver (131). Normally the receiver (131) would
have a
conical bottom shape to provide for ease of removal of is contents and a water
feed (not
shown) may be provided to the receiver (131) contents for effective slurry
flow of sand,
silt and bitumen from the receiver (131) to an agglomerator (not shown) for
separation.
Rocks, gravel and clay lumps (130) exit from partly closed grizzly end (141)to
disposal.
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
Figure 20 A and B are illustrations to explain why axial (central) feed
distributors
have been removed and are not used in current oleophilic agglomerators of this
present
patent. Figure 20A shows an old style agglomerator (92). Feed (89) enters the
agglom-
erator drum axially through a feed distributor using a screen (88) comprising
rods sur-
rounding the entering feed, to prevent balls from entering the feed
distributor. The old
style feed distributor is shaded to show distribution of the feed into the
agglomerator inte-
rior. Three circulation paths are shown. Path one is the path (85) of heavy
balls. Path
two is the path (86) of lighter balls. Path three is the path (87) created by
balls impacted
on by the axial feed distributor in the rotating agglomerator interior. This
path results
from disturbance created by the axial feed distributor (87), and appears to
interfere with
the interaction of paths 86 and 85 in the transfer of bitumen from the lighter
balls to the
heavier balls. for subsequent effective extrusion of bitumen paste out of the
drum along
the bottom (91) to cable wraps (90). While this could not be observed visually
(since bi-
tumen will coat the surface of any transparent agglomerator end wall), the
axial feed dis-
tributor appeared to interfere with the other two flow patterns. It is the
reason why feed is
now introduced through top quadrant or quadrants of the agglomerator cage for
more ef-
fective feed distribution into the cage. Not only does it eliminate concerns
about circula-
tion of oleophilic rods or balls in the cage but it also is simpler since
rotary seals are not
required for a concenric axial feed met. Figure 20 B illustrates feed
introduction into the
agglomerator through the upper quadrants of the cage not covered by rope
wraps. In this
case there are two main flow paths shown. The denser balls have a flow path
(94) close
to the bottom of the cage and the lighter balls have a flow path (93) closer
to the center of
the cage. The rope wraps (103) mainly cover the bottom part of the apertured
cage wall
(97). There is interaction between these two flow paths and there are
indications that bi-
tumen paste transfers from the lighter balls to the heavier balls for
subsequent passage or
extrusion to the rope wraps (103). An uneven internal cage surface of Figure
15 B is
caused, for example, by mechanical rods or pipes of the apertured cage wall
(98,99) of
Figure 20 B along rib ID and results in tumbling of the oleophilic balls (not
shown) inside
the cage. Other possible attachments inside the agglomerator are shown as
items 165,
166,167 and 168 in Figure 15 B to cause ball and rod tumbling.
31
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
In Figure 20 B feed (95) enters a feed hopper (104) that normally covers the
full
length of the agglomerator cage. Baffles (96) may be used to evenly and
uniformly con-
trol feed entry into the cage.
EARLIER PATENTS GRANTED TO THE CURRENT INVENTOR, OR PEND-
ING DEAL WITH:
1. Using rotating drums with perforated cylindrical wall and rubber rollers to
collect
bitumen from the cylindrical walls for oleophilic separation of oil sand
slurry.
2. Using oleophilic mesh belts by themselves to separate oil sand slurry and
many
other mixtures.
3. Using agglomerators with perforated cylindrical walls partly covered by
mesh
belts.
4. Using multiple wraps of endless rope (or cable) on agglomerator drums as
oleo-
philic sieve for agglomerator drums of older designs with axial entrance of
feed
into the drum.
5. Using hydraulically driven positive displacement pumps to pump bitumen
and/or
FFT.
6. Using confined path hydro-cyclones to wash and remove rocks, gravel and
sand
from an oil sand slurry.
7. Using two temperature processing of oil sand slurry in a conventional PSV
(froth
flotation primary separation vessel) to froth flotate warm bitumen in the PSV
and
to recover bitumen from cold middlings of the PSV by oleophilic separation.
8. Using a vena contracta to disperse bitumen into water in order to transfer
particu-
late hydrophilic minerals from bitumen phase to aqueous phase. However the
concept of using a long pipeline in turbulent flow to achieve the same
objective,
as described in the present patent, is new technology.
9. The early work of the present inventor used and patented drums with
perforated
circumferential walls as the agglomerator with mesh belt oleophilic sieves
around
part of the drum circumference.
32
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
10. Agglomerator cages with internal reinforcement members inside the cage to
cre-
ate a rigid cage was patented but, unknown at that time, it interfered with
oleo-
philic ball circulation, which will be more severe when oleophilic rods are
used in
aglomerator cages or drums.
11. The option of driven heated rollers to drive drum rotation.
12. Some pending patents were abandoned as technology development passed them
by.
NEW ITEMS INCLUDED IN THE PRESENT PATENT
1 The use of oleophilic separators near a mine face.
2 The use of short duration tailings ponds in conjunction with
oleophilic separation.
3 The use of oleophilic rods instead of oleophilic balls inside an
agglomertor that
has a greater length than 1.5 times its internal diameter.
4 The location and use of hoppers to introduce feed into the agglomerator cage
along the top portion of the cage not covered by rope wraps, without the use
of a
central (axial) feed distributor.
5 The elimination of structural members inside the agglomerator to
thereby allow
unrestricted tumbling of oleophilic balls or oleophilic rods inside the cage.
6 The use of ribs cut from metal plate to form hoops for constructing
agglomerators
where the ID of each rib is the same as the OD of each rib to save on metal
and
metal cutting.
7 The use of bars, rods or pipes welded to hoop ID for large potential
commercial
separators instead of using mechanical rods with bobbins on the rods to space
the
hoops.
8 The use of longitudinal tabs, pipes or bars on the inside of hoops
of an agglomera-
tor to encourage tumbling of oleophilic balls or oleophilic rods inside the
agglom-
erator.
DETAILS AND SUMMARY
33
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
Froth flotation is an old technology that has been the basis for minerals
extraction from
crushed or ground ore in a water medium for well over a century. In that
technology var-
ious chemicals are used to modify mineral surface properties and then to
separate result-
ing water wetted gangue from oil wetted valuable minerals. Air and chemicals
are used
to froth the oil wetted minerals fractions and cause these to float to the top
of separation
vessels whilst water wetted gangue leaves the vessels through the bottom.
William
Haynes, around 1870, introduced froth flotation to separate minerals from
gangue. Karl
Clark, around 1920 began to perfect it to separate bitumen from mined oil
sand. The first
commercial mined oil sand extraction plant came on stream in 1968. All this
indicates
that technololgy improvement takes time to become accepted by industry.Test
work on
oleophilic separation of mined oil sand was started by the present inventor in
1975 to re-
place bitumen froth flotation as a more environmentally acceptable and cost
effective
process.
The earliest story of oleophilic separation came from Herodotus (484-425 BC)
who saw maidens draw feathers or myrtle branches covered in bitumen through
beach
sand to collect flour gold. Modern oleophilic separation tackles bitumen
extraction from
a completely different approach than frothing an ore or dragging myrtle
branches through
sand. The sieve does not pass through the feed but the feed passes through the
sieve,which is an important difference that clearly defines the apparatus and
process of
the present invention.
Only the rope wrap surfaces need to be oleophilic for the process to function
but
the cage structural members can be oleophobic (bitumen hating) and may
preferably be
oleophobic where possible since oleophilic separation screens out or
separates, with an
oleophilic sieve in the form of multiple cable wraps, bitumen from water, sand
and silt.
Anyone in the oil sands industry who has pushed a shovel into an oil sand ore
deposit
during summer has seen ambient temperature bitumen adhering to that shovel by
oleo-
philic adhesion. Oleophilic separation is simple and much less demanding than
bitumen
froth flotation. Instead of a shovel, it uses a rotating agglomerator with
apertured wall
covered with carefully spaced oleophilic rope wraps to separate bitumen,
generally at
ambient temperature, from oil sand slurry or tailings pond FFT (fluid fine
tailings). The
agglomerator is a metal cage that rotates and is partly filled with a bed of
oleophilic rods
34
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
or oleophilic balls of one or several densities. These balls or rods mix with
the bitumen
containing feed and collect onto their surfaces, by contact, cold bitumen
paste particles
from the feed. With a consistency of ketchup, peanut butter or tooth paste,
this collected
bitumen moves to the cage bottom and is extruded through apertured cylindrical
wall of
the agglomerator to oleophilic rope wraps that cover at least the bottom half
of the rotat-
ing agglomerator. From there the bitumen paste is conveyed by the sieve to a
set of heat-
ed rollers mounted above the agglomerator cage where it is converted into a
free flowing
warm bitumen product that contains no air but that, like bitumen froth,
contains hydro-
philic and oleophilic minerals and water. Aqueous phase effluent of this
separation, con-
taming hydrophilic minerals, such as sand, silt and fines , flows out of the
agglomerator
apertured wall and passes through sieve apertures (restricted space between
adjacent
wraps) to disposal. Based on data collected in bench tests, in pilot plant
tests and in field
comparison tests, instead of six hours, oleophilic separation when perfected
and used in a
commercial setting will take about 15 minutes to achieve the same or better
results than
commercial bitumen froth flotation that now takes 6 hours (360 minutes).
Hence, oleo-
philic separation has the promise of a more than twenty four fold increase in
separation
speed with superior separation results as detailed in the attached pilot plant
figures and
tables. A second stage of oleophilic separation may be used to remove trapped
hydro-
philic minerals from bitumen product to yield a product that is superior to
bitumen froth
and with less loading on dilution centrifuges or solvent extraction.
Oleophilic minerals of
the product are normally not removed by oleophilic separation but, since
hydrophilic
minerals are removed, the water washed bitumen contains beneficiated
oleophilic miner-
als to provide a potentially concentrated source of rutile and zircon, etc.
from the centrif-
ugal tailings or from the product of solvent extraction. The positive results
of short but
effective processing suggest that oleophilic separation should be considered
for mine face
bitumen extraction to reduce processing time, to eliminate costly annual
replacement of
long distance oil sand slurry pipelines, to eliminate long distance tailings
pipelines, to
reduce the consumption of energy for slurry processing, to eliminate the need
for com-
pressed air for flotation, to reduce the carbon footprint of mined oil sands
extraction and
to reduce or eliminate the accumulation of naphthenic acid containing toxic,
high pH,
tailings (FFT) in long duration tailings ponds. The current commercial
tailings ponds
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
contain large amounts of discarded bitumen and oleophilic minerals but the
commercial
mined oil sands industry has not developed an effective method to recover that
discarded
bitumen which, since 1980 has started to release large amounts of
environmentally sensi-
tive methane to the air. Thus far the industry has not shown much interest in
encouraging
others, who do have the technology, to recover that lost bitumen or to
encourage com-
mercialization of technology that does not discard bitumen to permanent
tailings ponds.
The present patent attempts to create awareness, to protect technology and to
encourage
external technology review by or for the oil sands industry of this
technology.
DEFINITIONS
Oleophilic separation:
Oleophilic separation, as defined in the present patent represents a concept,
meth-
od and apparatus for commercially separating - at temperatures well below 50
degrees
centigrade - a feed mixture of bitumen phase and aqueous phase, both
containing particu-
late minerals not exceeding a few millimeters in any direction. It uses a
rotating agglom-
erator cage having two end walls which may have multiple holes along the
periphery to
accept mechanical rods and has a horizontal cylindrical wall in the form of
multiple
hoops, assembled from ribs, that also have holes along the periphery to accept
the same
rods. The hoops may be tied to the end walls by means of the mechanical rods
in tension
that pass through the holes in the end walls and through the holes in the
hoops with bob-
bins on the mechanical rods to equally space the hoops and thus to form a cage
with end
walls and with apertured cylindrical cage wall. Multiple wraps of endless
plastic rope or
metal rope contact the bobbins along the bottom quadrants of the cage and
partly close
off the space created by the bobbins between the hoops along the cage
circumferential
bottom and thereby control the outflow of aqueous phase of processed feed.
This deter-
mines in part the residence time of feed being processed in the cage whilst
bitumen phase
paste of the processed mixture leaving the cage along bottom quadrants adheres
to the
multiple wraps and is conveyed to heated rollers above the cage to cause
bitumen phase
in the form of a warm free flowing liquid to leave the wraps as product of
separation.
Hoops may be rolled from flat bar or may be assembled from curved ribs cut
from metal
36
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
plate with holes to accept the mechanical rods. Alternately the agglomerator
may resem-
ble a welded metal cage.
Oleophilic and oleophobic
In the context of the present patent, oleophilic refers to bitumen loving or
having
an affinity for bitumen. Oleophobic refers to bitumen hating or not having an
affinity for
bitumen.
Welded agglomerator cages:
Welded agglomerator cages may be used instead of having apertured drum cir-
cumferential walls held together by rods in tension passing through holes in
hoops or ribs
between end walls hoops, with bobbins on the rods to space the hoops. A cage
may be
welded together using end walls, hoops and steel structural bars or pipes to
fabricate the
cage, as shown in Figures 14 B, 14 C and 15 B.. Tabs may be welded to or pins
may be
attached to longitudinal cage members to maintain equal space between rope
wraps. Long
oleophilic rods are more economical than oleophilic balls of the same combined
volume
and are preferred inside the cage, instead of balls, for large scale
commercial agglomera-
tor use when the cage length exceeds the cage internal diameter by a factor of
at least 1.5
Agglomerator:
Agglomerator, agglomerator drum, agglomerator cage, and cage, all have the
same
meaning in the present patent with the exception that the word drum and its
plural gener-
ally refers to older technology or other equipment. Cage and its plural
generally refers to
modern oleophilic separation technology.
Process temperature:
Process temperature inside the agglomerator cage normally is well below 50 de-
grees centigrade since the objective of the process is to yield bitumen
products from the
cage that have a viscosity close to or exceeding conventional ketchup, peanut
butter or
tooth paste when viscosities are measured at room temperature. The
agglomerator cage
37
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
and its contents may also be at approximately room temperature to achieve the
desired
bitumen viscosity for effective adherence to rope wraps. The wraps convey
viscous bi-
tumen phase of mixture separation from the cage to internally heated rollers
above the
cage to produce a free flowing bitumen product. As a result, process thermal
energy de-
mands are very low since the quantity of cold effluent leaving the cage is
much larger
than the quantity of warm bitumen product of separation leaving the rollers
and specific
heat of feed and of effluent is much higher than specific heat of bitumen
product.
Optimum separation temperature:
Aqueous phase of the feed mixture that has lost most of its contained bitumen
leaves as effluent of separation through the cable wrap restricted bottom half
of the cage
by passing through available space between the cable wraps. For effective
oleophilic
separation, only a very small amount of viscous bitumen phase should leave
with the
aqueous effluent. When bitumen phase viscosity is too low, too much bitumen
leaves the
agglomerator with the aqueous phase. When bitumen phase particles in the feed
are so
cold that these resemble solid particles, bitumen will not adhere to the rope
wraps but will
also leave the agglomerator with the aqueous phase effluent. Hence, each feed
may have
an optimum temperature range for separation but, in all cases, that
temperature is much
lower than the temperature used in bitumen froth flotation because oleophilic
separation
is based on different concepts than flotation.
Two or more stages of oleophilic separation:
Overall operating conditions may be optimized when two or more stages of oleo-
philic separation are used, each optimized for the feed it is to be
processing. This may
happen when the effluent of one oleophilic separator is to be separated later
by another
oleophilic separator after that effluent has been stored in a temporary
tailings pond for a
few months. In such cases, the first separator may be designed to allow more
bitumen to
pass out with the effluent than normal since the second separator only has to
process the
resulting fluid fine tailings from which sand and silt have been removed by
settling and
thus will capture most of the bitumen contained in the FFT when that second
separator
38
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
has been optimized for FFT processing only. This may be done to reduce total
residence
time of mechanical processing when two separators are used in sequence while
allowing
natural minerals settling for a while between mechanical separations to assist
the separa-
tion process to maximize total bitumen recovery.
Combining oleophilic separation with minerals settling:
An extreme example of combining natural settling with oleophilic sparation
would be to allow oil sand slurry to spray or disperse directly onto a
temporary tailings
pond to allow gravel, sand and silt to settle for a few months followed by
oleophilic sepa-
ration of the resulting liquid layers of such a pond. Such natural settling
could well result
in clean gravel, sand and silt at the pond bottom and clean bitumen lenses and
dispersed
bitumen in water and fines in the pond above the gravel, sand and silt for
oleophilic sepa-
ration. Risky, but well worth a small test and development program to evaluate
the bitu-
men content of the settled gravel, sand and silt, as a function of pond depth.
Rotation drive of the process.
The heated rollers are not driven by a motor. The agglomerator cage is driven
to
rotate but hangs in the rope wraps, which are closely spaced to carry most or
all of the
weight of the cage and its contents. When the cage is driven, there is less
stress on the
rising cable wraps than when the rollers are driven. This reduced stress will
allow rope
wraps to last longer. Normally only a lever bar keeps the cage in rotatable
contact with
the separator frame. Tension in the multiple wraps of endless rope carries all
or most of
the weight of the cage and its contents. The rollers that carry cage weight by
means of
cable wraps are well supported by rotatable rollers with shafts in bearings in
strong
mounts above the cage attached to separator support frame. These rollers are
heated in-
ternally to transfer heat to the wraps and to bitumen adhering to the wraps to
yield a
warm free flowing bitumen product that leaves the heated wraps as product of
oleophilic
separation. For proper operation, rotation of the agglomerator cage is by
driven gears or
by driven roller chains and sprockets. Lever bars are normally used to prevent
cage
39
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
swinging and to facilitate such gear or sprocket drive. Alternate to lever
bars, one set of
two chains at each cage end wall may be used instead of lever bars. In that
case, four
roller chains mating with four sprockets on the cage and four sprockets on the
separator
frame keep the rotating gage positioned and prevent it from swinging.
Multi stage slurry or FFT processing:
Thus far only single stage oleophilicseparation of oil sand slurry or FFT has
been
reported on. However using more than one stage of slurry oleophilic separation
will re-
suit in an increase in bitumen reovery.
Undesirable effluent outflow:
An important consideration in oleophilic separation is that all spaces between
wraps and between wraps and hoops (or ribs) should be small enough to keep
feed in the
agglomerator cage until a desired amount of bitumen has been removed from the
feed.
This is achieved by proper and close placement of multiple wraps on the
apertured cage
wall over the full length of the cage.
Feed entrance into the agglomerator:
Feed for proposed effective commercial mixture separation, using long agglomer-
ator cages, does not enter near the axis of the agglomerator drum, like in
previous patents
of the present inventor, but enters the agglomerator cage from above where
space be-
tween cage hoops is not covered by rope wraps. Feed entry is centered
approximately
between the two top quadrants or may be centered near the 45 degree angle of
the top
right quadrant for a counter clockwise rotating cage, as illustrated in Figure
17. Feed en-
try into the cage requires careful design of feed hopper and feed control to
assure uniform
distribution of feed throughout the full length of the cage..
Internal structural members:
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
Unlike in previous patents of the present inventor, there are no structural
members
inside the cage interior which tend to interfere with interaction of tumbling
oleophilic
rods or oleophilic balls inside the cages of the present patent. Removing all
structural
members from the cage inside allows unrestricted ball or rod circulation with
feed in the
cage. This is a new feature of the present patent not previously disclosed.
Endless rope:
Endless rope can be a multi strand metal wire rope that has been made endless
by joining
its ends by a splice, such as a long splice, in which the rope diameter at the
splice is not
much greater than diameter elsewhere of the rope. It can be a multi strand
plastic, steel
or other material rope made from any suitable material. Normally the endless
rope is a
multi strand twisted rope to simplify splicing. When two twisted strand ropes
are placed
tightly together the twisted strands often provide narrow passages for
controlled passage
of aqueous phase out of the agglomerator cage between rope wraps or for
extrusion flow
of bitumen phase to, between and around the wraps. When a non metal rope is
used, its
properties should not significantly lose strength as a result of heat from
heated rollers
contacting the rope. This means that when using plastic rope, control of
roller internal
heating is an important consideration to prevent deformation or undesirable
stretching of
the wraps. For commercial separators, long lasting, abrasion resistant metal
wire ropes
may be preferred.
Rope for commercial separators:
Plastic ropes can be very strong and abrasion resistant and suitable for pilot
plant
oleophilic separation studies since plastic rope splicing without increasing
rope diameter
is easy. However, large long lasting commercial oleophilic separators will
have cage di-
ameters 2 meters or larger and cage lengths 10 meters or longer, and will use
endless ole-
ophilic wire ropes (multi strand metal cable) instead of multi strand plastic
ropes to ex-
tend the life of the rope wraps in a commercial setting.
41
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
Plating of metal rope:
Cold bitumen paste clings by adhesion to twisted multi strand metal rope and
this
adhesion may be enhanced by coating or plating the individual metal rope
strands with an
oleophilic coating such as copper, tin or oleophilic abrasion resistant metal
prior to fabri-
cating the rope from strands. A twisted metal rope may also be coated with a
plastic cov-
ering but this covering will tend to wear and may eventually break off during
commercial
agglomerator use. An oleophilic plating or coating on individual metal rope
strands will
be more effective since abrasion will not likely remove such plating from
twisted metal
rope surfaces not directly contacting cage or roller surfaces.
Cage weight sensing:
Since the rotating agglomerator cage is hanging in rope wraps from heated
rollers
above the cage, weight sensing of the cage and its contents may provide a
convenient
method of feed level sensing in the cage to control and prevent feed overflow
along the
agglomerator top. When a load cell is attached to a redirecting cable roller
(Figure 11B)
that keeps the wraps from rolling off the cage or rollers, the resulting time
averaged load
cell reading will be proportional to the total weight of the cage and its
contents. In other
words, measuring tension in one wraps will allow calculation of the total
tension in all the
wraps combined and hence will give an indication of the total weight of the
cage and its
contents when the load cell is calibrated accordingly.
Processing capacity:
A single 2 meter diameter, 10 meter long agglomerator cage half filled with a
bed
of oleophilic rods at a process residence time of 2.5 minutes per cage volume
not occu-
pied by bed of balls may have an estimated processing capacity of
approximately 9,000
cubic meters or 13,000 tonnes per day at 95% bitumen recovery. When a second
oleo-
philic separator is used a few months later to process the aqueous effluent
from the first
42
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
separator, after settling of sand, silt and some fines in a working tailings
pond, total bitu-
men recovery may increase to 98%. This is feasible since oleophilic separation
has
shown, in field testing, it does not normally require heating of oil sand
fluid fine tailings
and achieves high bitumen recovery from FFT..
Oleophilic:
Oleophilic in these specifications refers to a substance that has an affinity
for bi-
tumen and does not necessarily refer to a substance that has an affinity for
light oil. For
example, a rope wrap that is coated with a thin layer of light oil will not
allow adhesion
of bitumen to that wrap and hence is not oleophilic as used in these
specifications. Simi-
larly, a hot rope wrap covered with a thin layer of hot bitumen will not
readily allow ad-
hesion of cold bitumen to the hot wrap. It is one of the reasons for cooling
of cable wraps
before returning from hot rollers to agglomerator cage, if so needed. A rope
covered with
a thin layer of very low viscosity oil will not readily allow strong adhesion
of very high
viscosity oil upon contact because a thin layer of low viscosity oil
represents a barrier to
high viscosity oil adhesion to a normally oleophilic plastic or metal surface.
Small bench or pilot scale:
Small bench or pilot scale oleophilic separation refers to units for
separating bi-
tumen or oil from an aqueous feed containing oil or bitumen. I have found that
using
small diameter closely spaced endless rope wraps is an effective method for
separating
viscous or weathered oil from aqueous phase and this could lead to the
development of
viscous oil skimmers which are not detailed herein but are not excluded in the
objectives
of the present invention and its technology protection. The thinner the rope
wraps and
the closer the oleophilic rope wrap surfaces are together, the lower the
viscosity of oil or
bitumen that can be recovered efficiently from a feed by oleophilic
separation. Converse-
ly, the higher the bitumen viscosity of a feed the further the cable wraps may
be apart,
within certain limits consistent with separator construction and objectives.
Residence time in the agglomerator in minutes:
43
CA 02939495 2016-08-17
Jan Kruyer, Thorsby, AB
Residence time in the agglomerator in minutes is conveniently reported herein
on
the basis of total cage volume minus volume of oleophilic balls or oleophilic
rods, divid-
ed by feed rate in minutes.
Valuable minerals production:
Fast oleophilic separation of valuable oleophilic minerals from sedimentary
and/or cushed ore is feasible by passing a mixture of bitumen, water,
hydrophilic miner-
als and oleophilic minerals through the apparatus and process of the present
invention to
produce a product comprising mainly of valuable oleophilic minerals and
bitumen, and
an effluent mainly comprising water and hydrophilic minerals.
Oleophobic cage coating
There will be a potential advantage to oleophilic separation if the cage metal
of
the agglomerator is coates with an oleophobic (oil hating) coating so that
only the rope
wraps will present an oleophilic surface to bitumen contained in the feed
leaving the
cage. For that reason an oleophobic cage or an oleophobic coating on the cage
metal is
recommended where possible.
25
44
