Petroleum Research 8 (2023) 309e315

Contents lists available at ScienceDirect

Petroleum Research
journal homepage: http://www.keaipublishing.com/en/journals/
petroleum-research/

Design and construction of a continuous pilot flotation facility: A case

study for water-based oil sand extraction
Feng Lin
Natural Resources Canada, CanmetENERGY, Devon, Alberta, T9G 1A8, Canada

a r t i c l e i n f o a b s t r a c t

Article history: In this work, the design and construction of a continuous and pilot-scale flotation facility are demon-
Received 20 April 2022 strated. The performance of the new facility was evaluated from series of pilot flotation tests, carried out
Received in revised form using oil sands slurry to extract bitumendan extra-heavy form of petroleum. Results indicated that
3 November 2022
bitumen recoveries of the new pilot plant for an identical ore and water chemistry were largely
Accepted 3 November 2022
dependent on air injection method, slurry conditioning time, flotation residence time, and slurry tem-
Available online 7 November 2022
perature. Importantly, when compared with those of bitumen extraction tests using a bench-scale Batch
Extraction Unit (BEU) operated at the most optimal conditions, it was suggested that the new pilot plant
Keywords:
Enhanced oil recovery
produced flotation recoveries of bitumen and froth qualities at a level as good as, or even higher than, the
Flotation level accomplished using the BEU at an identical ore, water chemistry and operating temperature. This
Pilot facility continuous, pilot flotation plant could potentially serve as a pre-commercial production system that
Minerals processing verifies a new processing aid, or as an alternative extraction technology for oil sand, coal, and mineral
Bitumen extraction processing.
© 2022 The Authors. Publishing services provided by Elsevier B.V. on behalf of KeAi Communication Co.
Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/
by-nc-nd/4.0/).

1. Introduction petroleum, is an important portfolio of natural resources in several

countries including Canada. The commercial method for mineable
Flotation is largely dependent on differences in surface Canadian oil sand operations is the Clark Hot Water Extraction
wettability of valuable species and gangue in an aqueous slurry (CHWE) process, which is literally based on the principle of flota-
or pulp (Hanumanth and Williams, 1988). The surfaces of valu- tion (Zhou et al., 2013). Caustic (e.g., sodium hydroxide) is often
able particles are either inherently hydrophobic by natural sur- used as a process aid to enhance bitumen recovery and froth
factants or selectively rendered to be hydrophobic by an quality. Caustic addition could help release natural surfactants
additivedsynthetic surfactant or collector (Yu et al., 2017). When beneficial for bitumen liberation and reduce the detrimental slime
an aqueous slurry is aerated, the valuable particles could be coating of fine particles on bitumen, thereby improving aeration of
attached to air bubbles and rise to the top of the vessel as “froth”, bitumen droplets and the overall flotation outcome (Zhou, 2021).
therefore separating unwanted materials, or gangue, that are Nevertheless, the use of caustic is not applicable to provide a suf-
hydrophilic and remain in the slurry (Nesset et al., 2006). ficient improvement for low-grade or problematic processing ores.
Flotation plays a central role in many processes of great tech- In addition, caustic increases the pH of process water, creating more
nological importance, including minerals and coal processing difficulty in the management of waste stream tailings (Lin and Xu,
(Chen et al., 2020; Han et al., 2020b), deinking (Luo et al., 2003), 2019).
waste treatment and recycling (Huang et al., 2007), selective To address the issue of caustic usage and other drawbacks
separation of microorganisms, and oil sand extraction (Nassif associated with the CHWE, new processing aids (Han et al., 2020a;
et al., 2014). Hou et al., 2021) and novel extraction processes (Lin et al., 2018;
As a globally reliable energy supply and petrochemical feed- Zhou, 2021) have been intensively studied at the bench scale. For
stock, oil sand bitumendan extra-heavy and viscous form of transforming a new chemical or process from laboratory-into in-
dustrial-scale performance, a pilot facility is necessary to cross “the
so-called commercialization valley of the death” (Hay and Rule,
2003). However, direct information on the design of pilot
E-mail address: [email protected].

https://doi.org/10.1016/j.ptlrs.2022.11.001
2096-2495/© 2022 The Authors. Publishing services provided by Elsevier B.V. on behalf of KeAi Communication Co. Ltd. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
F. Lin Petroleum Research 8 (2023) 309e315

flotation facility was scarce, and the minerals and oil sand pro-
ducers retained the technology as a confidential.
In this technical note, we first describe the design and con-
struction of a continuous, pilot flotation facility scaled to approxi-
mately 100e600 kg/h slurry flow-through. As a case study using
the custom-built facility, an oil sand ore is used to study the per-
formance of pilot flotation tests. Included in this note are the effects
of physical parameters on the bitumen recovery and froth quality.
Impacts of chemical parameters (e.g., processing aids or additives)
were published on separate articles (Lin et al., 2019; Lin and Pang,
2020).

2. Design and construction of pilot flotation facility

The process flow diagram for our design of a continuous, pilot

flotation facility is illustrated in Fig. 1. The entire facility consists of
two integrated units: slurry production and froth flotation. The pur-
pose of the slurry production unit was to generate the ore and water
mixture, the so-called ‘slurry’, to allow for conditioning of the slurry,
and to produce a mixture that can be pumped from the slurry tank to Fig. 2. An image of continuous, pilot-scale flotation facility.
the flotation cell. The aim of froth flotation unit was to recover as
much valuable species from the conditioned slurry as possible in the
form of froth. A photograph of the constructed pilot flotation facility is
shown in Fig. 2. In this facility, the main pieces of machinery and 2.2. Hydraulic mixer
equipment are the slurry tank, hydraulic mixer, hydro-transport
pipeline, flotation cell, pumps, and variable frequency drives. A variable-speed mixer with hydraulic lifting was suitable for
dispersing and mixing the viscous and high-solids-content sus-
pensions e the mixtures of oil sands ore and water e in the slurry
2.1. Slurry tank tank. The mechanical agitation needed for mixing was also ex-
pected to further break down any remaining lumps of ore into
A slurry tank was necessary to mix the crushed ore with water
smaller-size pieces. The strength of the agitation (or the speed of
to form the ore slurry, and to serve as a pool where the ore slurry mixing) could be controlled by changing the frequency of the VFD
could be recircuited for improved conditioning. A stainless-steel
connected to the mixer. The motor of the mixer was volatile-solvent
tank about 60 cm in diameter and 210 L in volume was modified proof and was stable and produced little noise. The mixing blade
for use. The tank had to be able to store at least 40 min of slurry
was easy to replace and could be changed to a different type of
supply to the flotation cell, assuming a 2.5 L/min flow rate. As blade if needed. The lift frame using a hydraulic cylinder as actu-
suggested by the commissioning test, the wall of the tank was also
ator, allowed the mixer to be easily lifted up and down as required.
equipped with four baffles that could prevent the formation of
vortex flow and provide good dispersion.
2.3. Hydro-transport pipeline

The purpose of the hydro-transport pipeline was to circulate the

slurry back to the pool, providing sufficient time and favourable
conditions for the ore conditioning (e.g., bitumen liberation from
the oil sand grains), and then to transport the well-conditioned
slurry to the flotation cell.
Rubber was chosen as the material for the pipes due to its
availability, low cost, and erosion resistance, especially considering
that in this pilot scale the pipes were rinsed after each run. How-
ever, it is recommended that carbon steel should be selected as the
material of pipelines. From the tests of sediment transport, we
found that the slurry velocity should be kept above 3 cm/s to pre-
vent settling of solids with 0.3 mm in diameter e average size for
sand particles e in the slurry. For a velocity of 3 cm/s, a pipe with a
nominal diameter of 1 inch is calculated to give a flow rate of 1.5 L/
min.

2.4. Flotation cell

The flotation cell (FC) was built similar to a WEMCO depurator

(Kind, 1976; Degner, 1988), with some modifications. The cell was a
single chamber of approximately 25 L in volume, with height of
35 cm and diameter of 30 cm, including a variable-speed agitation
rotor and its motor. Other components included: a cover to mini-
Fig. 1. Schematic diagram of a continuous, pilot flotation facility. mize the evaporation of materials, a loop for circulating and
310
F. Lin Petroleum Research 8 (2023) 309e315

discharging tailings and a hose for discharging froth. Two air in- moveable electrical panel with five VFDs was built. The relation
jection methods were considered and denoted as AIM-1 and AIM-2. between flowrate of peristaltic pump and the frequency of a VFD
As shown in Fig. 3, AIM-1 was a conventional induced gas design was determined, as were the relations between the agitation speed
that air was directly injected into the FC motor-driven rotor (Kind, of the motor for flotation cell or the skimmer and the VFD
1976; Degner, 1988), while AIM-2 was relatively novel in which air frequency.
was injected via the FC circulation loop. The amount of injected air
from either AIM-1 or AIM-2 was controlled by an accurate flow 3. Case study: Materials and method
meter. For a given residence time, air bubbles interact with the
valuable species and form the lower-density aggregates that float 3.1. Materials
upward. The phase separation in the FC involved at least two layers:
the froth phase at the top of the cell that contained most of the As a case study evaluating the performance of above new pilot
valuable species and could be removed by skimming and collected facility, a medium-grade, Athabasca oil sand ore were used. Bitumen,
as product, and the bottom phase containing most of the unwanted solids, and water contents in this ore were about 11.3 wt%, 85.6 wt%,
wastes. The bottoms were circulated back towards the cell to pre- and 3.0 wt%, respectively. The fine content (i.e., mineral solids
vent clogging and to further enhance the phase separation. One smaller than 44 mm) in the total solids of the ore were 16.3 wt%. This
stream of circulated bottom phase was discharged as gangue or type of ore was chosen because it is the minimum grade for current
tailings. water-based extraction process, and it is to ensure that the study
bears direct relevance to industrial applications. For a low-grade or
2.5. Pumps ill-defined ore, the use of water-based extraction process is unfea-
sible without a novel chemical aid. Prior to the tests, the partially
As mentioned previously, there were two areas that require the frozen ore was crushed to less than 1 cm in size using a crusher with
use of pumps. One area was to pump the oil sands slurry into the a screen of 1 cm, and then well homogenized. For aqueous phase,
hydro-transport pipeline for circulation and transportation to the Town of Devon tap water was used and its pH was 8.0 ± 0.1.
FC, while the other was to pump the bottom phase in the FC for
circulation and discharge. The requirement for pumping in these 3.2. Protocol of pilot aqueous bitumen extraction
two cases was similar; the types of pumps selected were deter-
mined with special emphasis on their ability to deal with the high- Approximately 25e30 kg of the size-reduced ore was trans-
density slurry that contains a large amount of solids. Based on this ferred to the tank where tap water equal to the mass of oil sand was
requirement most types of pumps were ruled out. The two cate- added to generate an oil sand-water mixture, the slurry. The slurry
gories of pumps well-suited to cope with high-density suspensions temperature was controlled by optimizing the temperature of the
were the air operated diaphragm (AOD) pump and peristaltic added water to the desired degree. The tank also allowed for
pump. Considering the different flow rates required, AOD pumps thorough conditioning of the slurry. The AOD pump produced the
that have ability to handle large volumetric flow rates were used for pressure required to transport the slurry through the hydro-
circulation of either the slurry or the FC tailings, and peristaltic transport pipeline back to the slurry tank, which served as a con-
pumps whose rates could be easily controlled by the frequency of ditioning pool, promoting bitumen separation from solid grains.
VFDs were chosen for hydro-transporting the slurry to the FC or After a certain period of conditioning at 700 rpm, a second round of
discharging the tailings. tap water was added to the slurry tank, and a stream of slurry was
then distributed and fed into the FC at a desired feed flow rate (i.e.,
controlling residence time). Simultaneously, air was dispersed at a
2.6. Variable frequency drive constant rate of 3.6 L/min into the system via either AIM-1 or AIM-2
to facilitate flotation of the separated bitumen to the top of the cell.
A variable frequency drive (VFD) was a type of motor controller The agitation speed in the FC was controlled at approximately
that drives an electric motor by varying the frequency and the 150 rpm. As illustrated in Fig. 4, bitumen froth was removed by
voltage supplied to the electric motor. VFDs had at least two skimming the surface of the fluid on the top of the FC with an
functions: one was to protect the motor of a connected machine automatic skimmer. Bitumen froth was collected and analyzed by a
from sudden and high voltage charges and the other was to control Dean-Stark apparatus to determine bitumen recovery and froth
the pumping and the agitating speed precisely. In this work, a quality. Bitumen recovery was defined as the ratio of the mass of
bitumen recovered in froth to the total bitumen mass in the oil sand
ore. The mass ratio of bitumen to solids (B/S) and bitumen to water
(B/W) in froth was used as a measure of froth quality. The bottom
layer of the FC contained a mixture of water and sand, with trace
amounts of bitumen. This bottom layer was circulated back to the
FC. After a given residence time, a portion of suspension from the FC
circulation loop was discharged into a bucket as oil sand tailings at
a controlled flow rate.

4. Results and discussion

4.1. Effect of air injection method

Bitumen flotation experiments were first conducted using two

different air injection methods (AIM-1 and AIM-2) to choose a
proper AIM for evaluating the performance of the new pilot facility.
The tests were conducted by keeping flotation residence time of
Fig. 3. Images of two air injection methods denoted as AIM-1 and AIM-2. 10 min at 20  C. Fig. 5 shows the bitumen recovery rates using the
311
F. Lin Petroleum Research 8 (2023) 309e315

generated during the mild mechanical agitation by the rotor in

AIM-1 were generally micro-sized or greater, i.e., microbubbles
(Kind, 1976; Degner, 1988). In contrast, sparging air into the FC
circulation loop in AIM-2 could produce a large quantity of tiny,
submicron-size bubbles, (i.e., nanobubbles) due to the two
contributing factors (Zimmerman et al., 2011; Mankosa and
Wyslouzil, 2012; Xiong and Peng, 2015). First, the valve installed
for AIM-2 had a contractive (narrower) internal diameter than the
connecting pipes. This structure was similar to that of venturi tube,
the most economic and efficient method of creating hydrodynamic
cavitation which leads to the generation of nanobubbles (Mankosa
and Wyslouzil, 2012; Xiong and Peng, 2015). The second contrib-
uting aspect was the high intensity agitation in high-speed circu-
lation loop, aiding in facilitating the formation and stabilization of
nanobubbles (Zimmerman et al., 2011). As a result, nanobubbles
generated in AIM-2 could not only increase the ability of collision
and attachment with bitumen droplets (Fan et al., 2010a), but also
improve the hydrophobicity of bitumen droplets, acting as a
nanobubble bridging role (Fan et al., 2010b). Fine bitumen droplets
would form aggregates with the nanobubbles and grow to be larger
droplets, which were beneficial for increasing bitumen recovery
(Esmaeili et al., 2012; Lin et al., 2014).
Note that the values of error bars in Fig. 5 and the following
figures represent the standard deviations of two repeated pilot
tests for each condition investigated. As seen, the errors were less
Fig. 4. Images of (a) bitumen froth removed by a skimmer; (b) bitumen froth draining than 10%, showing good repeatability of our pilot data and
down a hose into a pail. reasonable agreement with the typical errors occurred in bench
extraction tests.

4.2. Effect of slurry conditioning time

Fig. 5 also shows the effect of slurry conditioning time on

bitumen recovery using two air injection methods (AIMs). As seen,
bitumen recovery rates for two AIMs followed similar trend as a
function of conditioning time: bitumen recovery in pilot aqueous
extraction process at 20  C increased significantly as conditioning
time increased from 5 min to 20 min, after which bitumen recovery
did not increase any further. Taking AIM-2 as an example, the
bitumen recovery increased from 45.9% to 52.1% at 10 min, and then
to about 72.6% at 20 min, respectively. A further increase to 30 min
of conditioning time retained the recovery rate similarly to 72.1%. In
view of the above findings of two AIMs on bitumen recovery, AIM-2
was considered as a better option than AIM-1, and only data for
AIM-2 are discussed in the rest of this document.
Results presented in Fig. 6 show that when the slurry condi-
tioning time was increased from 5 min to 30 min, bitumen froth
quality in term of B/S and B/W ratios noticeably increased in the
first step and then leveled off. B/S ratio reached a plateau at 10 min
of conditioning time, while it took 20 min of conditioning time for
B/W ratio achieved a maximum in the range of this study.
The improved bitumen recovery and froth quality for longer
Fig. 5. Bitumen recovery as a function of slurry conditioning time in pilot aqueous
bitumen extraction for two different air injection methods (AIM-1, and AIM-2) at 20  C. conditioning time were ascribed to higher degree of bitumen
liberation (DBL) from the sand grain, a prerequisite step for
bitumen flotation (Lin et al., 2014). Previous laboratory studies
showed that DBL often depends on the conditioning time and fol-
two AIMs. In AIM-1, bitumen recovery rates were relatively low. lows the first-order process kinetics. Longer conditioning time
However, when using AIM-2 of sparging air into the FC circulation would provide greater DBL until a maximum state is reached (Lin
loop, bitumen recovery was increased significantly as compared to et al., 2018). The results in Figs. 5 and 6 suggest that 20 min were
the case using AIM-1 at an identical slurry conditioning time. For estimated sufficient to allow good separation without being
conditioning time of 20 min, for example, AIM-2 gave bitumen impractical.
recovery of about 72.6%, compared to only 40.0% using AIM-1.
The much higher bitumen recovery by injecting air into the 4.3. Effect of flotation residence time
circulation loop (i.e., using AIM-2) was possibly attributable to the
smaller size of air bubbles generated by AIM-2. The air bubbles The bitumen recovery as a function of flotation residence time
using AIM-2 is given in Fig. 7 (a). For these pilot extractions,
312
F. Lin Petroleum Research 8 (2023) 309e315

Fig. 8 (a). These pilot tests were done at a constant conditioning

time of 20 min using AIM-2. As seen, for the same residence time,
the higher bitumen recovery for a pilot extraction was clearly
observed with increasing slurry temperature from 20  C to 40  C.
No further gain in bitumen recovery was seen beyond 40  C. Taking
15 min of residence time as an example, the bitumen recovery
improved gradually from 72.6% at 20  C to 79.1% at 30  C, and then
to 86.9% and 94.6% at 35  C and 40  C, respectively. With a further
increase to 50  C, the recovery rate was remained at 94.5%. Fig. 8 (a)
also indicates that at a given temperature, the bitumen recoveries
for the two residence times were of a very similar value, confirming
that increasing residence time from 10 to 15 min had a negligible
effect on the bitumen recovery using the new pilot facility.
Fig. 8 (b) and (c) show the effect of operating temperature on the
Fig. 6. Effect of slurry conditioning time on (a) bitumen/solids (B/S), and (b) bitumen/ froth B/S and B/W ratios for the two residence times of pilot ex-
water (B/W) ratios in pilot aqueous bitumen extraction for AIM-2 at 20  C.
tractions. As in Fig. 8 (b), provided that other process conditions
were kept the same, the increase of slurry temperature in our range
conditioning time was fixed at 20 min and the operating temper- of study showed insignificant effects on the B/S ratio, considering
ature was 20  C. As can be seen in Fig. 7 (a), bitumen recovery for the experimental error. The data in Fig. 8 (c) suggests that the
aqueous pilot extraction was increased with increasing flotation impact of slurry temperature on B/W ratio was not straightforward.
residence time from 2.5 min to 10 min. A further increase in resi- For 10 min residence time, the B/W ratio in pilot aqueous extraction
dence time to 20 min showed no further gain in the bitumen re- initially increased when temperature was increased from 20  C to
covery rate. 40  C, then the B/W decreased with a further increase in temper-
The results in Fig. 7 (b) and (c) show a continued increase in ature to 50  C. However, for 15 min residence time, the B/W ratio
bitumen froth quality for both B/S and B/W ratios when residence remained similar with increasing temperature in the beginning and
time was increased from 2.5 min to 20 min. As Fig. 7 (b), the B/S then decreased in the range of 35  Ce40  C.
ratio increased from 2.29 to 2.77, and then to 2.86 and 3.40 with the
increase of residence time in our range of study. As illustrated in 4.5. Comparison with results of bench extraction tests
Fig. 7 (c), similar trend but with even more profound improvement
in the B/W ratio by longer residence time was observed in the range Bench extraction tests using a 1.2-litre batch extraction unit
of this study. (BEU) were also conducted as a reference, since the BEU is often
The increase on bitumen recovery and froth quality for larger considered as a reliable source of evaluating oil sand extractability
residence time could be largely accounted by improved bitumen at laboratory scale. Details of the BEU apparatus and its test pro-
aeration, which is an essential factor for bitumen flotation (Zhou tocol are given elsewhere (Lin et al., 2018). It should be noted that
et al., 2013). Larger residence time would increase the probability the operating conditions of the BEU were tuned the most optimal,
of bitumen-air attachment and give more time allowing aerated including the 20 min of slurry conditioning and 15 min of flotation
bitumen droplet floating up to the top and for solids and water time. Data of the BEU tests is also shown in Fig. 8. As can be seen,
particles to settle down in the flotation cell, translating to more the same trend in term of the effectiveness of the slurry tempera-
bitumen and less contaminants recovered in the collected froth (Lin ture on oil recovery rate, was obtained among pilot and the BEU
et al., 2018). extractions. However, at an identical temperature, pilot extractions
yielded a slightly higher oil recovery than the BEU. For example, the
oil recoveries from the pilot extractions for 10 min, and 15 min of
4.4. Effect of slurry temperature residence time and from the BEU at 40  C were determined to be
93.9%, 94.6%, and 89.8%, respectively. Similarly, as illustrated in
The effect of slurry temperature on bitumen recovery in pilot Fig. 8 (b) and (c), when compared to the BEU extraction that
aqueous extraction for two flotation residence times is presented in operated at 15 min of flotation time, pilot extraction for 15 min of

Fig. 7. Effect of flotation residence time on (a) bitumen recovery, (b) bitumen/solids (B/S), and (c) bitumen/water (B/W) ratios in pilot aqueous bitumen extraction for AIM-2 at
20  C.

313
F. Lin Petroleum Research 8 (2023) 309e315

Fig. 8. Effect of slurry temperature on (a) bitumen recovery, (b) bitumen/solids (B/S), and (c) bitumen/water (B/W) ratios in pilot aqueous bitumen extraction for two residence
times. RT10, RT15 represent residence time of 10 min and 15 min, respectively. Bitumen recovery for bench-scale tests using a batch extraction unit (BEU) was used as the reference.

residence time gave comparable or slightly higher values for both Development (PERD) and Alberta Innovates. © His Majesty the King
B/S and B/W ratios at the same temperature. in Right of Canada, as represented by the Minister of Natural Re-
It is worthwhile to point out that, depending on what kind of sources, 2022.
processing aid or additive, the relative importance of the above-
mentioned physical parameters may be varied if considering the
presence of a chemical parameter. For example, when cyclohexane References
was used to soak an ore prior to the addition of water, the effect of
Chen, L., Zhao, Y., Bai, H., Ai, Z., Chen, P., Hu, Y., Song, S., Komarneni, S., 2020. Role of
temperature would be negligible or little on bitumen recovery; the montmorillonite, kaolinite, or illite in pyrite flotation: differences in clay
other factors (conditioning time, residence time) would signifi- behavior based on their structures. Langmuir 36, 10860e10867.
cantly affect bitumen recovery but showed less noticeable effects Degner, V.R., 1988. Recent Wemco flotation technology advancements. Resources
Processing Tech 35, 27e44.
on froth quality in the range of our study (Lin et al., 2019). Inter- Esmaeili, P., Lin, F., Yeung, A., 2012. Stability of emulsified heavy oil: the combined
estingly, the general trend and magnitude of data for our pilot tests effects of deterministic DLVO forces and random surface charges. Langmuir 28,
in presence of a chemical additive were consistent with those of 4948e4954.
Fan, M., Tao, D., Honaker, R., Luo, Z., 2010a. Nanobubble generation and its appli-
BEU tests. cations in froth flotation (part III): mechanical cells and specially designed
column flotation of phosphate. Int. J. Min. Sci. Technol. 20, 317e338.
5. Conclusions Fan, M., Tao, D., Honaker, R., Luo, Z., 2010b. Nanobubble generation and its appli-
cations in froth flotation (part II): fundamental study and theoretical analysis.
Int. J. Min. Sci. Technol. 20, 159e177.
This work described in detail the design and construction of a Han, C., Li, R., Lu, Y., 2020a. Study on synthesized thermoresponsive block copol-
continuous, pilot flotation facility scaled to 100e600 kg/h slurry ymer for water-based oil sands extraction. Energy Fuel. 34 (8), 9473e9482.
Han, H., Liu, A., Wang, H., 2020b. Effect of hydrodynamic cavitation assistance on
flow-through. As a case study, an oil sand ore was used to study the
different stages of coal flotation. Minerals 10, 221e233.
physical operating parameters on the aqueous bitumen extraction Hanumanth, G.S., Williams, D.J.A., 1988. Design and operation characteristics of an
outcomes using the new pilot facility. Physical parameters included improved laboratory flotation cell. Miner. Eng. 1, 177e188.
Hay, M.P., Rule, C.M., 2003. SUPASIM: a flotation plant design and analysis meth-
air injection method (AIM), slurry conditioning time, flotation
odology. Miner. Eng. 16, 1103e1109.
residence time, and slurry temperature. It was shown that AIM-2 of Hou, J., Du, J., Sui, H., Sun, L., 2021. Surfactants enhanced heavy oilesolid separation
sparging air into the flotation cell circulation loop produced a much from carbonate asphalt rocks: experiment and molecular dynamic simulation.
higher bitumen recovery than AIM-1 of injecting air into the rotor. Nanomaterials 11, 1835.
Huang, Y., Takaoka, M., Takeda, N., Oshita, K., 2007. Partial removal of PCDD/Fs,
The results also clearly demonstrated the importance of the other coplanar PCBs, and PCBs from municipal solid waste incineration fly ash by a
three physical parameters as they significantly affect the measured column flotation process. Environ. Sci. Technol. 41, 257e262.
bitumen recovery and froth quality for pilot aqueous bitumen Kind, P., 1976. Design criteria and recent developments in large-capacity Wemco
flotation cells. J. S. Afr. Inst. Min. Metall 3, 345e362.
extraction. Interestingly, when compared with those of bitumen Lin, F., Pang, C.J., 2020. Impact of a hybrid bitumen extraction process on the
extraction tests using bench-scale Batch Extraction Unit (BEU) destabilization of resulting bitumen froth emulsion diluted with heptane.
operated at the most optimal conditions, the new pilot plant yiel- Miner. Eng. 145, e106069.
Lin, F., Suda, J., Yeung, A., 2014. A small point regarding DLVO coagulation condi-
ded a comparable or slightly higher bitumen recovery rate and tions. J. Colloid Interface Sci. 430, 113e115.
froth quality at an identical operating temperature. The new pilot Lin, F., Xu, Y., 2019. Impact of a novel bitumen extraction process on mined oil sands
flotation plant could be viewed as a useful pre-commercial pro- tailings behavior. AIChE J. 65, 250e258.
Lin, F., Xu, Y., Nelson, R., 2018. Determination and understanding of solvent losses to
duction system for verifying new process aids for water-based oil
tailings from a laboratory hybrid bitumen extraction process. Energy Fuel. 32,
sand, coal, and minerals processing and for developing an alter- 9202e9210.
native extraction technology. Lin, F., Xu, Y., Nelson, R., 2019. Pilot evaluation of aqueous nonaqueous hybrid
bitumen extraction process for mineable oil sands. Miner. Eng. 131, 241e248.
Luo, Q., Deng, Y., Zhu, J., Shin, W.T., 2003. Foam control using a foaming agent spray:
Declaration of competing interest a novel concept for flotation deinking of waste paper. Ind. Eng. Chem. Res. 42,
3578e3583.
The authors declare no conflict of interest. Mankosa, M., Wyslouzil, H., 2012. Cavitation Tube Sparging System. Australian
Patent AU2012100021.
Nassif, M., Finch, J.A., Waters, K.E., 2014. Determining frother-like properties of
Acknowledgement process water in bitumen flotation. Miner. Eng. 56, 121e128.
Nesset, J.E., Hernandez-Aguilar, J.R., Acuna, C., Gomez, C.O., Finch, J.A., 2006. Some
gas dispersion characteristics of mechanical flotation machines. Miner. Eng. 19,
This research was financially supported by the government of 807e815.
Canada's interdepartmental Program of Energy Research and Xiong, Y., Peng, F., 2015. Optimization of cavitation venturi tube design for pico and

314
F. Lin Petroleum Research 8 (2023) 309e315

nano bubbles generation. Int. J. Min. Sci. Technol. 25, 523e529. Zhou, J.Z., 2021. Role of mineral flotation technology in improving bitumen
Yu, Y., Ma, L., Cao, M., Liu, Q., 2017. Slime coatings in froth flotation: a review. Miner. extraction from mined Athabasca oil sands III. Next generation of water-based
Eng. 114, 26e36. oil sands extraction. Can. J. Chem. Eng. 99, 755e777.
Zhou, Z., Li, H., Chow, R., Roberge, K., 2013. Role of carrier flotation in accelerating Zimmerman, W.B., Tesar, V., Bandulasena, H.C.H., 2011. Towards energy efficient
bitumen extraction recovery from mineable Athabasca oil sands. Can. J. Chem. nanobubble generation with fluidic oscillation. Curr. Opin. Colloid Interface Sci.
Eng. 91, 1340e1348. 16, 350e356.

315