Effect of Ultrasonic Irradiation Treatment on Rheological

Behaviour of Extra Heavy Crude Oil: A Solution Method for

Transportation Improvement

Mohammad Amin Rahimi, Ahmad Ramazani S.A.,* Hossein Alijani Alijanvand,

Mohammad Hossein Ghazanfari and Mahdi Ghanavati
Department of Chemical and Petroleum Engineering, Sharif University of Technology, Tehran, Iran

The highly viscous property of heavy oil often causes problems in its transportation in pipelines. Mixing heavy oil with light oil as well as ultrasound
treatment are viable solutions to this problem. In this study, extra heavy crude oil samples were first diluted with 0, 0.05, 0.1, and 0.15 mL/mL (0, 5,
10, and 15 vol%) of a light crude oil; then the mixture was irradiated by ultrasonic waves for 0, 5, 10, 15, and 20 min; finally the viscous shear
functions of all mixtures was measured at different values of shear rate at different temperature levels. The results revealed that the minimum
viscosity of the diluted extra heavy crude oil samples was obtained at 10 min of ultrasonic irradiation. Moreover, the viscosity reduction rate in
relation to temperature decreases as temperature increases. In other words, the maximum viscosity reduction rate occurred at 0.05 mL/mL (5 vol%)
of light crude oil. Using the experimental data, the parameters of common rheological models were obtained and a new modified Power Law model
was presented to calculate the effect of shear rate and temperature simultaneously.

Keywords: heavy crude oil, rheological model, ultrasonic irradiation, viscosity reduction, asphaltene

INTRODUCTION Various factors such as temperature, pressure, and compositional

changes may influence the asphaltene stability and result in

H
eavy crude oil or extra heavy crude oil is a highly viscous asphaltene precipitation.[12] Asphaltene causes several problems in
oil which cannot easily flow from a reservoir to production the oil industry in the stages of production, transportation, and
wells in normal reservoir conditions or through land refining.[13,14] For instance, asphaltene precipitation causes perme-
pipelines under ambient temperature.[1] Accordingly, heavy and ability reduction of reservoir rock,[15] blockage of the pore
extra heavy crude oil has been defined as any type of crude oil with throats,[16] and wettability alteration of rocks from water wet to
an API gravity of less than 208 and 108, respectively.[2] The oil wet,[17] which does not have a desirable effect on production
adjective heavy is used to describe this type of oil because its from the reservoir rock. Additionally, asphaltene may be precipi-
density or specific gravity is much higher than that of conventional tated through oil transportation via pipelines; thus, cleaning
light crude oil. Besides, unlike light crude oil, the molecular processes are required to remove the asphaltene deposition and
composition of heavy or extra heavy crude oil consists of prevent the blockage of pipelines.[18] Asphaltene is also one of the
hydrocarbons with more carbon atoms. main reasons why heavy and extra heavy crude oils have high
According to estimates, heavy and extra heavy crude oil viscosity.[19] Mechanical, chemical, thermal, and biological meth-
constitute more than half of the global recoverable crude oil ods or a combination of them are used to solve difficulties involved
reserves. Despite the various problems involved in their produc- in asphaltene precipitation during oil production.[20] However,
tion, transportation, and refining, they have recently drawn the apart from the time-consuming nature of such solutions and their
attention of energy consumers across the word. This is because of heavy financial costs, they may have health hazards, environmen-
the ever-increasing global demand for crude oil and the tal damage, and limitations in application.[20] Therefore, transpor-
diminution of light and medium crude oil reserves. Such problems tation of heavy and extra heavy crude oil is a highly technical and
can be largely attributed to their physical and chemical properties. complex operation in the petroleum industry.
For one, the high viscosity of heavy and extra heavy crude oil In light of the above-mentioned considerations, several methods
makes them difficult to pump through pipelines. For another, as have been applied to solve the problems regarding the transporta-
different studies suggest, heavy and extra heavy crude oil may tion of heavy and extra heavy crude oil through pipelines. For
have high amounts of salt, sulphur, and metals such as vanadium example, viscosity reduction by means of dilution with light
and nickel, causing problems in their refining.[3] Furthermore, compounds,[21] heating,[22] or formation of oil-in-water
heavy crude oils may contain complex compounds like wax and emulsions[23–26] is a common method. Another solution is
asphaltene with high molecular weights.[4] Asphaltene is defined
as the heavy fraction of the crude oil which is not soluble in normal
alkanes (such as n-heptane, n-hexane, and n-pentane) but is
soluble in aromatic solvents (such as toluene and benzene).[3,5–7] * Author to whom correspondence may be addressed.
Asphaltene is made up of polar aromatic compounds with E-mail address: [email protected]
high molecular weight, which is partly dispersed and partly Can. J. Chem. Eng. 95:83–91, 2017
dissolved in crude oil.[3,6–8] Crude oil also has non-polar resins, © 2016 Canadian Society for Chemical Engineering
DOI 10.1002/cjce.22676
which surround asphaltene molecules and do not let them Published online 11 October 2016 in Wiley Online Library
aggregate.[9–11] (wileyonlinelibrary.com).

VOLUME 95, JANUARY 2017 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING 83

reducing the pipeline friction[27] by drag reduction[28,29] or
Table 1. General specifications of the extra-heavy crude oil
annular flow through pipelines.[30,31] The initial upgrading of
heavy and extra heavy crude oil (i.e. separating its heavy fraction) Specification Value Units
and then transporting the residue is another method to make the
C1 0 mole percent
transportation easier.[32]
C2 0 mole percent
Recently, ultrasonic treatment has been applied in different
C3 0 mole percent
industrial oil issues.[33] The relevant literature show that C4 0 mole percent
ultrasonic treatment has considerable effects on various processes C5 0 mole percent
in producing and refining crude oil like removing asphaltene C6 0 mole percent
precipitation from porous rocks,[21,22] well drilling,[23] increasing C7 0 mole percent
oil recovery,[24] desalting,[25,26] and dehydration.[27] C8 0 mole percent
With this in mind, the authors of this article were encouraged to C9 1.57 mole percent
study the influence of ultrasonic treatment on the viscosity C10 2.85 mole percent
reduction of asphaltenic extra heavy crude oil. In particular, this C11 3.13 mole percent
paper investigates the implications of applying the ultrasonic C12þ 92.45 mole percent
treatment method for conventional transportation of extra heavy Saturates 19.62 weight percent
crude oil. Furthermore, this study examines the effect of dilution Aromatics 62.45 weight percent
Resins 8.04 weight percent
with a light crude oil, as the oldest method, together with
Asphaltenes 9.89 weight percent
temperature on the viscosity reduction of an extra heavy crude oil
sample.

MATERIALS AND METHODS Table 2. Properties of the used extra heavy and light crude oil samples

Materials Asphaltene Density Viscosity

content (kg/m3) at API (Pa  s) at
Two types of crude oil, extra heavy and light, which were used in
Sample (wt%, g/100 g) 15 8C gravity 15 8C
this study were obtained from two major Iranian oil fields.
Densities of the crude oils were measured by a digital Anton Paar Extra heavy 9.89 1025 6.55 10.4
densitometer (Anton Paar, DMA 35N). The measured density and crude oil
API gravity of the used extra heavy crude oil were 1025 kg/m3 and Light crude oil 0.57 825 40.02 0.2
6.558API at 15 8C, respectively. The parameters for the light crude
oil were 825 kg/m3 and 40.028API at 15 8C, respectively. The
ultrasonic irradiation was applied into the extra heavy crude oil by vertical and horizontal directions in the samples and accurately
using an ultrasonic wave generator (Hielscher Ultrasound keeps the temperature stable (Anton Paar Physical MCR technical
Technology, Germany) with an average output power of 100 W specifications).
and a constant ultrasonic frequency of 20 kHz. The purity of n- For conducting the experimental measurements with an Anton
heptane (Merck, Germany) used for measuring the asphaltene Paar rheometer, a thin layer of crude oil sample is placed on the
content of light and extra heavy crude oils is 99.9 %. fixed plate of the rheometer. Afterwards, the upper plate is
adjusted to obtain the desired gap width. For all measurements,
Asphaltene Content Measurements
the gap was set to 0.8 mm. Figure 1 shows a schematic diagram of
The asphaltene contents of the extra heavy and light crude oil the Anton Paar rheometer in parallel plate geometry, which has
samples were measured based on the ASTM D6560-00 method. been used for measurements in this study.
According to the measurements, the asphaltene contents of the To achieve the reproducibility of the experimental results, the
extra heavy and light crude oil were 0.0989 and 0.0057 mL/mL viscosity of every sample was measured three times and
(9.89 and 0.57 wt%), respectively. In addition, the composition of the average of these three values is reported in the results.
heavy crude oil is reported in Table 1; the density, API, and The relative error in these calculations has a maximum value of
viscosity of that can be seen in Table 2. 2.6 %.

Rheometer
The rheological measurements were conducted by a Physica MCR
301 rheometer from Anton Paar, which is operated by RheoPlus
software. The measurements were carried out in a parallel
geometry setup. This geometry contains a fixed lower plate and
a rotatable upper plate with a diameter of 25 mm. Since
temperature greatly influences the viscosity of crude oil, it should
be controlled carefully during measurements. In this regard, the
Anton Paar rheometer uses a new temperature control system
which can keep the temperature fixed on any optional temperature
in the range of
40 to 200 8C with an accuracy of 0.01 8C. It does so
by a Peltier plate at the bottom of the rheometer. In addition, an
active Peltier hood was utilized as a cover, which controls the
temperature within the hood to match that of the bottom Peltier
plate.[28] This technique minimizes temperature gradients in both Figure 1. Schematic diagram of rheometer in the plate-plate geometry.

84 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING VOLUME 95, JANUARY 2017

Experimental Procedure
To examine the influence of ultrasonic irradiation on the
rheological behaviour of extra heavy crude oil, with and without
dilution of light crude oil, the following experimental procedure
was done:

1. Samples of extra heavy crude oil (each measuring 100 mL)

were placed in a beaker and ultrasonically irradiated for
different time intervals of 0, 5, 10, 15, and 20 min.
2. The ultrasonic-treated crude oil samples were cooled to room
temperature for 2 h, and then rheological measurements were
done at temperatures of 25, 35, 45, 55, and 65 8C respectively.
3. To study the effect of dilution on viscosity reduction, the extra
heavy crude oil was diluted with 0.05, 0.1, and 0.15 mL/mL
(5, 10, and 15 vol%) of the light crude oil.
4. The mixture was homogenized by agitating to ensure that all Figure 2. Plot of viscosity versus ultrasonic irradiation time of extra heavy
points of it have the same physical properties. Then, steps 1 crude oil at different temperatures.
and 2 were applied to the mixture.

RESULTS AND DISCUSSION

Effect of Ultrasonic Irradiation without Dilution

Figure 2 shows the measured viscosity of the extra heavy crude oil
samples versus different ultrasonic irradiation time intervals at five
different temperatures of 25, 35, 45, 55, and 65 8C. It shows a
decreasing trend for viscosity from 0 to 10 min of ultrasonic
irradiation time and then an increasing one. This increasing trend
continues into the final time interval, which is 20 min. Moreover, it
illustrates that the maximum viscosity reduction of the extra heavy
crude oil sample occurs at 10 min of ultrasonic irradiation time for
all examined temperatures. The probable explanation is that
applied ultrasonic energy degrades the resin intermolecular bonds,
particularly the hydrogen bonds and charge-transfer p-p bonds
between the resins and the asphaltene particles at the outmost
layers, resulting in the separation of asphaltene particles.[29] Figure 3. Plot of viscosity versus ultrasonic irradiation time of extra heavy
Therefore, the viscosity of the sample decreases. Afterwards, crude oil with 0.05 mL/mL (5 vol%) light crude oil at different
temperatures.
increasing the applied ultrasonic time to more than 10 min can
break down the asphaltene cluster to more finely divided particles;
the solubility of the asphaltene particles in the extra heavy crude oil
increases; and consequently, the viscosity of the samples
increases.[20] Moreover, this reduction in viscosity is more apparent
at low temperatures. As temperature increases from 25 to 65 8C, the
value of maximum viscosity reduction reduces.

Effect of Ultrasonic Irradiation with Dilution

Figures 3 to 5 show the viscosity of the extra heavy crude
oil samples diluted with 0.05, 0.1, and 0.15 mL/mL (5, 10, and
15 vol%) of the light crude oil at five different temperatures of 25,
35, 45, 55, and 65 8C.
According to these figures, the minimum viscosity of all samples
with three different volume fractions of light crude oil was
obtained at ultrasonic irradiation time interval of 10 min, which is
similar to that of the non-irradiated extra heavy crude oil. Figures 3
to 5 also indicate that as the light crude oil volume fraction Figure 4. Plot of viscosity versus ultrasonic irradiation time of extra heavy
increases, so does viscosity reduction. This may be attributed to crude oil with 0.1 mL/mL (10 vol%) light crude oil at different
the solubility of asphaltene particles. The reason is that at higher temperatures.
light crude oil volume fractions, more asphaltene can be dissolved
in the mixture at all applied ultrasonic irradiation time intervals. resins; hence, resins can be separated more easily from asphaltene
The point is, for all diluted extra heavy crude oil samples, at a particles. Consequently, viscosity of the samples reduces.[30] That
10 min time interval, the ultrasonic irradiation energy can break is why more viscosity reduction is observed at higher concen-
hydrogen and charge-transfer p-p bonds between asphaltene and trations of the added light crude oil and 10 min of ultrasonic

VOLUME 95, JANUARY 2017 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING 85

 Figure 6. Plot of viscosity versus temperature of extra heavy crude oil at
Figure 5. Plot of viscosity versus ultrasonic irradiation time of extra heavy different ultrasonic irradiation time intervals.
crude oil with 0.15 mL/mL (15 vol%) light crude oil at different
temperatures.

irradiation for all samples. In short, it is the result of the

simultaneous effect of resins separated from asphaltene and
ultrasonic energy.

Effect of Temperature on the Viscosity at Different Ultrasonic

Time Intervals
Figure 6 presents the viscosity of the extra heavy crude oil samples
versus temperature, for five different time intervals of 0, 5, 10, 15,
and 20 min. According to this figure, viscosity decreases as the
temperature rises from 25 to 65 8C at all applied time intervals.
This is due to the fact that as temperature rises, kinematic energy
of the extra heavy crude oil samples becomes higher, resulting in
an increase in the flow ability of the extra heavy crude oil.
Furthermore, at higher temperatures more resins at the outmost
layers can be desorbed from the asphaltene cluster and the Figure 7. Plot of viscosity versus temperature of extra heavy crude oil with
interactions between asphaltene particles decrease.[31] Accord- 0.15 mL/mL (15 vol%) light crude oil at different ultrasonic irradiation time
ingly, the viscosity of the crude oil samples reduces. The viscosity intervals.
of the diluted mixtures also shows the same trend. For instance,
Figure 7 shows the viscosity of the extra heavy crude oil sample
with 0.15 mL/mL (15 vol%) of light crude oil versus temperature
at different ultrasonic time intervals. Again, the viscosity
decreases as temperature increases.
To analyze the effect of temperature on the crude oil viscosity at
different ultrasonic irradiation time intervals, the Arrhenius
equation has been used. This equation can be derived from the
Eyring’s theory of viscosity as follows:[32]



E 1 1
mðT Þ ¼ mðT 0 Þ  exp 
ð1Þ
R T T0

It is notable that the reference temperature in this study is 65 8C.

To determine the activation energy from the Arrhenius
   
equation, the slope of the plot of ln mmððTT0ÞÞ versus R1 T1
T10 gives Figure 8. Plot of activation energy versus ultrasonic irradiation time of the
extra heavy crude oil with different light crude oil concentrations.
the activation energy of the viscous flow for each sample at the
given ultrasonic irradiation time interval. Figure 8 shows the
activation energy of the diluted samples at different ultrasonic sample, the less the activation energy. This adverse correlation
irradiation time intervals with different light crude oil volume results from the simultaneous effect of the dissolution of
fractions. The minimum activation energy is observed at 10 min of asphaltene particles on light crude oil and the applied ultrasonic
applied ultrasonic irradiation. This is due to desorption of resins energy. The lowest activation energy at 10 min of irradiation
from asphaltene particles with ultrasonic energy. Moreover, the treatment means that at this time interval, the attractive inter-
greater the amount of light crude oil added to the extra heavy particle interaction between the asphaltene particles is weakest

86 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING VOLUME 95, JANUARY 2017

Figure 9. Plot of shear stress versus shear rate for extra heavy crude oil after Figure 10. Plot of shear stress versus shear rate for extra heavy crude oil
10 min of ultrasonic irradiation at five temperatures. diluted with 0.05 mL/mL (5 vol%) of light crude oil after 10 min of
ultrasonic irradiation at five temperatures.
due to applied ultrasonic irradiation and the dissolution effect of
the light crude oil addition. Consequently, lower activation
energy of viscous flow is needed to overcome these attractive
inter-particle interactions and enable the samples to flow.
Table 3 shows the activation energy of the extra heavy crude
oil with four different light crude oil concentrations. According
to this table, the activation energy quickly decreases from 34.51
to 28.80 kJ/mol and from 28.80 to 24.10 kJ/mol when dilution
concentration increases from 0 to 0.05 mL/mL (0 to 5 vol%) and
0.05 to 0.1 mL/mL (5 to 10 vol%), respectively. This is because
of more solubility of heavy components of the extra heavy
crude oil such as asphaltene in the light crude oil. It should
also be noted that the least decrease in activation energy is
observed when the dilution concentration increases from 0.1 to
0.15 mL/mL (10 to 15 vol%). This might be due to the fact that
the greatest portion of asphaltene has been dissolved at two
previous dilution concentration changes (i.e. 0–0.05 mL/mL and
0.05–0.1 mL/mL (0–5 vol% and 5–10 vol%)) and increasing the
light crude oil concentration does not greatly affect the solubility Figure 11. Plot of shear stress versus shear rate for extra heavy crude oil
of the asphaltene compared to other dilution concentration diluted with 0.1 mL/mL (10 vol%) of light crude oil after 10 min of
intervals; therefore, the activation energy does not significantly ultrasonic irradiation at five temperatures.
change.

Rheological and Statistical Evaluation

The purpose of this section is to assess the effects of shear rate and
temperature on the rheological properties of extra heavy crude oil
and diluted ones through graphical observation of experimental
data plots and statistical investigations. The effects of temperature
and volume percentage of the light crude oil will be further
elucidated through the comparison of parameters obtained from
four well-known empirical rheological models. The experimental
data are fitted to the four rheological models of Newtonian,

Table 3. Activation energy of the extra heavy crude oil with different
light crude oil concentrations at 10-min ultrasonic irradiation time
interval

Light crude oil concentration, Activation energy,

LC (vol%, L/100 L) E (KJ/mol)

0 34.51
5 28.80 Figure 12. Plot of shear stress versus shear rate for extra heavy crude oil
10 24.10 diluted with 0.15 mL/mL (15 vol%) of light crude oil after 10 min of
15 22.44 ultrasonic irradiation at five temperatures.

VOLUME 95, JANUARY 2017 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING 87

 Table 4. Rheological models of the extra heavy crude oil sample after 10 min of ultrasonic irradiation and the calculated models' parameters and
statistical errors at different temperatures

Parameters

Rheological models T ¼ 25 8C T ¼ 35 8C T ¼ 45 8C T ¼ 55 8C T ¼ 65 8C

Newtonian mp ¼ 3.1853 mp ¼ 3.1060 mp ¼ 1.8620 mp ¼ 1.0825 mp ¼ 0.8785

R2 ¼ 0.9977 R2 ¼ 0.9988 R2 ¼ 0.9991 R2 ¼ 0.9986 R2 ¼ 0.9998
RMSE ¼ 3.7773 RMSE ¼ 2.6805 RMSE ¼ 1.3710 RMSE ¼ 1.0115 RMSE ¼ 0.3057

Bingham Plastic t0 ¼ 2.8950 t0 ¼ 0.9293 t0 ¼ 0.9293 t0 ¼ 0.6727 t0 ¼ 0.2007

mp ¼ 3.1334 mp ¼ 2.5132 mp ¼ 1.8453 mp ¼ 1.0705 mp ¼ 0.8749
R2 ¼ 0.9987 R2 ¼ 0.9999 R2 ¼ 0.9994 R2 ¼ 0.9990 R2 ¼ 0.9999
RMSE ¼ 2.8708 RMSE ¼ 0.7367 RMSE ¼ 1.1218 RMSE ¼ 0.8353 RMSE ¼ 0.2534

Power Law k ¼ 4.4803 k ¼ 3.2352 k ¼ 2.1293 k ¼ 1.2817 k ¼ 0.9847

n ¼ 0.9187 n ¼ 0.9306 n ¼ 0.9737 n ¼ 0.9665 n ¼ 0.9695
R2 ¼ 1.0000 R2 ¼ 0.9997 R2 ¼ 0.9999 R2 ¼ 0.9999 R2 ¼ 0.9999
RMSE ¼ 0.3619 RMSE ¼ 3.6849 RMSE ¼ 1.4560 RMSE ¼ 1.0285 RMSE ¼ 0.3978

Casson t00.5 ¼ 0.3824 t00.5 ¼ 0.1977 t00.5 ¼ 0.1274 t00.5 ¼ 0.1227 t00.5 ¼ 0.0558
mp0.5 ¼ 1.7497 mp0.5 ¼ 1.5672 mp0.5 ¼ 1.3564 mp0.5 ¼ 1.0322 mp0.5 ¼ 0.9319
R2 ¼ 0.9992 R2 ¼ 0.9999 R2 ¼ 0.9996 R2 ¼ 0.9994 R2 ¼ 0.9999
RMSE ¼ 2.5380 RMSE ¼ 0.4006 RMSE ¼ 1.1580 RMSE ¼ 0.8517 RMSE ¼ 0.2223

Bingham Plastic, Power Law, and Casson (Appendix A) using models along with their corresponding calculated parameters and
Microsoft Excel. To assess the adequacy of each rheological model two errors of R2 and RMSE. Based on the reported values in
in describing the rheological behaviour of extra heavy crude oil Tables 4 to 7, all four models can predict the rheological behaviour
samples and the diluted ones, statistical parameters of root mean of the samples accurately; this accuracy is acceptable based on the
square error (RMSE) and squared correlation coefficient (R2) are adequately close-to-unity values of R2 and low values of RMSE for
determined (Appendix B). all four different concentrations of light crude oil and five different
temperatures. Additionally, as the amount of light crude oil
Effect of temperature on rheological behaviour of extra heavy crude increases, the values of R2 and RMSE increase and decrease,
oil and diluted samples respectively; in other words, the minimum value of error is
Figures 9 to 12 show the experimental data of shear stress versus for the sample of extra heavy crude oil diluted with 0.15 mL/mL
shear rate for the extra heavy crude oil sample with four different (15 vol%) of light crude oil. Moreover, in all four tables, in the
volume fractions of light crude oil of 0, 0.05, 0.1, and 0.15 mL/mL Bingham Plastic model, the value of t0 is very low and the value of
(0, 5, 10, and 15 vol%). These were fitted to the four previously mp is very close to that of the Newtonian model. The values of t00.5
mentioned rheological models. Tables 4 to 7 show the results of and mp0.5 have values close to those of the Bingham Plastic model.
the fitting process of the experimental data to the rheological Additionally, in the Power Law model the values of n are close to

Table 5. Rheological models of extra heavy crude oil diluted with 5 vol% of light crude oil after 10 min of ultrasonic irradiation and the calculated
models' parameters and statistical errors at different temperatures

Parameters

Rheological models T ¼ 25 8C T ¼ 35 8C T ¼ 45 8C T ¼ 55 8C T ¼ 65 8C

Newtonian mp ¼ 1.6117 mp ¼ 1.6113 mp ¼ 1.3381 mp ¼ 0.8226 mp ¼ 0.5346

R2 ¼ 0.9991 R2 ¼ 0.9881 R2 ¼ 0.9994 R2 ¼ 0.9995 R2 ¼ 0.9997
RMSE ¼ 1.2117 RMSE ¼ 4.3076 RMSE ¼ 0.8184 RMSE ¼ 0.47200 RMSE ¼ 0.2160

Bingham Plastic t0 ¼ 0.7175 t0 ¼ 0.5034 t0 ¼ 0.7925 t0 ¼ 0.2813 t0 ¼ 0.1932

mp ¼ 1.7915 mp ¼ 1.4228 mp ¼ 1.0555 mp ¼ 0.8176 mp ¼ 0.5311
R2 ¼ 0.9991 R2 ¼ 0.9999 R2 ¼ 0.9987 R2 ¼ 0.9996 R2 ¼ 0.9999
RMSE ¼ 0.4180 RMSE ¼ 0.3822 RMSE ¼ 0.9349 RMSE ¼ 0.4073 RMSE ¼ 0.1406

Power Law k ¼ 2.3798 k ¼ 1.1840 k ¼ 1.3539 k ¼ 0.9633 k ¼ 0.6480

n ¼ 0.9227 n ¼ 0.9326 n ¼ 0.9466 n ¼ 0.9596 n ¼ 0.9502
R2 ¼ 0.9997 R2 ¼ 0.9997 R2 ¼ 1.0000 R2 ¼ 0.9999 R2 ¼ 0.9999
RMSE ¼ 5.6439 RMSE ¼ 2.1222 RMSE ¼ 0.4700 RMSE ¼ 0.3794 RMSE ¼ 0.3320

Casson t00.5¼.1944 t00.5¼.1413 t00.5 ¼ 0.1547 t00.5 ¼ 0.0791 t00.5 ¼ 0.0794

mp0.5 ¼ 1.1393 mp0.5 ¼ 1.1800 mp0.5 ¼ 1.0227 mp0.5 ¼ 0.8994 mp0.5 ¼ 0.7225
R2 ¼ 0.9999 R2 ¼ 0.9999 R2 ¼ 0.9993 R2 ¼ 0.9999 R2 ¼ 0.9999
RMSE ¼ 0.1555 RMSE ¼ 0.2023 RMSE ¼ 0.9210 RMSE ¼ 0.3693 RMSE ¼ 0.0886

88 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING VOLUME 95, JANUARY 2017

 Table 6. Rheological models of extra heavy crude oil diluted with 10 vol% of light crude oil after 10 min of ultrasonic irradiation and the calculated
models' parameters and statistical errors at different temperatures

Parameters

Rheological models T ¼ 25 8C T ¼ 35 8C T ¼ 45 8C T ¼ 55 8C T ¼ 65 8C

Newtonian mp ¼ 1.2938 mp ¼ 1.2776 mp ¼ 0.8215 mp ¼ 0.7264 mp ¼ 0.4431

R2 ¼ 0.9982 R2 ¼ 0.9993 R2 ¼ 0.9989 R2 ¼ 0.9988 R2 ¼ 0.9995
RMSE ¼ 1.3496 RMSE ¼ 0.8381 RMSE ¼ 0.6707 RMSE ¼ 0.6138 RMSE ¼ 0.2429

Bingham Plastic t0 ¼ 0.9475 t0 ¼ 0.2595 t0 ¼ 0.4644 t0 ¼ 0.4405 t0 ¼ 0.1638

mp ¼ 1.2768 mp ¼ 0.9580 mp ¼ 0.7181 mp ¼ 0.6136 mp ¼ 0.4401
R2 ¼ 0.9988 R2 ¼ 0.9999 R2 ¼ 0.9993 R2 ¼ 0.9987 R2 ¼ 0.9997
RMSE ¼ 1.0844 RMSE ¼ 0.1065 RMSE ¼ 0.4707 RMSE ¼ 0.5572 RMSE ¼ 0.1991

Power Law k ¼ 1.6739 k ¼ 1.2326 k ¼ 0.9043 k ¼ 0.7412 k ¼ 0.5131

n ¼ 0.9402 n ¼ 0.9267 n ¼ 0.9506 n ¼ 0.9662 n ¼ 0.9618
R2 ¼ 0.9999 R2 ¼ 0.9995 R2 ¼ 0.999 97 R2 ¼ 0.9999 R2 ¼ 0.9997
RMSE ¼ 0.3999 RMSE ¼ 1.8626 RMSE ¼ 0.2520 RMSE ¼ 0.7284 RMSE ¼ 0.1869

Casson t00.5 ¼ 0.1839 t00.5 ¼ 0.1157 t00.5 ¼ 0.1201 t00.5 ¼ 0.0988 t00.5 ¼ 0.0571
mp0.5 ¼ 1.1220 mp0.5 ¼ 0.9657 mp0.5 ¼ 0.8423 mp0.5 ¼ 0.7826 mp0.5 ¼ 0.6607
R2 ¼ 0.9994 R2 ¼ 0.9999 R2 ¼ 0.9995 R2 ¼ 0.9992 R2 ¼ 0.9998
RMSE ¼ 1.0186 RMSE ¼ 0.1089 RMSE ¼ 0.4451 RMSE ¼ 0.5825 RMSE ¼ 0.1909

unity. According to these values, it can be inferred that the Results of fitting the Arrhenius model to the experimental data
rheological properties of the extra heavy crude oil and diluted show that the Arrhenius equations are able to estimate the
samples show Newtonian behaviour. relationship of extra heavy crude oil samples and temperature.
These findings indicate that the shear stress depends on the
Development of a New Model for Describing the Dependency of temperature exponentially and may have the following form:
Extra Heavy Crude Oil Viscosity on Temperature and Shear Rate

A
It is useful to analyze the effects of temperature and shear rate on t / exp ð3Þ
the viscosity of extra heavy crude oil in one equation. Based on T
results from the previous section, although all samples show where A is a constant. It is assumed that viscosity is the product of
Newtonian behaviour, the error of the Power Law model is less shear rate and temperature. Thus, combining the effect of shear
than those of all the other three models. Therefore, the proposed rate, Equation (2), and the effect of temperature, Equation (3),
model is better described based on the Power Law model. gives the following form:
Accordingly, the shear stress is a function of shear rate as follows:

A
t ¼ k:g_ n :exp ð4Þ
t ¼ kg_ n ð2Þ T

Table 7. Rheological models of extra heavy crude oil diluted with 15 vol% of light crude oil after 10 min of ultrasonic irradiation and the calculated
models' parameters and statistical errors at different temperatures

Parameters

Rheological models T ¼ 25 8C T ¼ 35 8C T ¼ 45 8C T ¼ 55 8C T ¼ 65 8C

Newtonian mp ¼ 0.9641 mp ¼ 0.8037 mp ¼ 0.6169 mp ¼ 0.4327 mp ¼ 0.3253

R2 ¼ 0.9999 R2 ¼ 0.9998 R2 ¼ 0.9998 R2 ¼ 0.9996 R2 ¼ 0.9994
RMSE ¼ 0.2718 RMSE ¼ 0.2878 RMSE ¼ 0.1978 RMSE ¼ 0.2244 RMSE ¼ 0.1907

Bingham Plastic t0 ¼ 0.2857 t0 ¼ 0.2788 t0 ¼ 0.1910 t0 ¼ 0.2106 t0 ¼ 0.1065

mp ¼ 0.9590 mp ¼ 0.7987 mp ¼ 0.6134 mp ¼ 0.4289 mp ¼ 0.3234
R2 ¼ 0.999 97 R2 ¼ 0.9999 R2 ¼ 0.9999 R2 ¼ 0.9999 R2 ¼ 0.9996
RMSE ¼ 0.1232 RMSE ¼ 0.1642 RMSE ¼ 0.1135 RMSE ¼ 0.1359 RMSE ¼ 0.1679

Power Law k ¼ 1.2496 k ¼ 1.0188 k ¼ 0.7701 k ¼ 0.5678 k ¼ 0.3756

n ¼ 0.9242 n ¼ 0.9345 n ¼ 0.9376 n ¼ 0.9291 n ¼ 0.9589
R2 ¼ 0.9996 R2 ¼ 0.9998 R2 ¼ 0.9998 R2 ¼ 0.9999 R2 ¼ 0.9994
RMSE ¼ 1.8384 RMSE ¼ 1.0315 RMSE ¼ 0.8182 RMSE ¼ 0.4333 RMSE ¼ 0.2835

Casson t00.5 ¼ 0.1246 t00.5 ¼ 0.1119 t00.5 ¼ 0.0890 t00.5 ¼ 0.1030 t00.5 ¼ 0.0379
mp0.5 ¼ 0.9654 mp0.5 ¼ 0.8828 mp0.5 ¼ 0.7744 mp0.5 ¼ 0.6459 mp0.5 ¼ 0.5676
R2 ¼ 0.9999 R2 ¼ 0.9999 R2 ¼ 0.9999 R2 ¼ 0.9998 R2 ¼ 0.9998
RMSE ¼ 0.1150 RMSE ¼ 0.0661 RMSE ¼ 0.0387 RMSE ¼ 0.0837 RMSE ¼ 0.1628

VOLUME 95, JANUARY 2017 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING 89

 mp plastic viscosity in Bingham-plastic model (Pa  s)
Table 8. Parameters and errors of new rheological model.
t shear stress (N/m2)
Model parameters Fitting errors t0 yield stress in Bingham-plastic model (N/m2)

Vol% (L/100 L)
of light crude oil k N A R2 RMSE
APPENDIX A: RHEOLOGICAL MODELS
0 9.599 0.938 0.031 0.9721 9.3490
5 4.333 0.925 0.023 0.9714 5.2940 A.1. Newtonian
10 2.814 0.984 0.025 0.9831 1.5536
15 1.895 0.977 0.023 0.9954 1.1664 t ¼ mg_ ðA:1Þ

This proposed model expression includes the effect of both shear Parameter constraints: m > 0:
rate and the temperature. The experimental data of shear rate and
shear stress were fitted to Equation (4), using the Genetic A.2. Bingham Plastic
Algorithm function in MATLAB and the three constants of the t ¼ t 0 þ mp g_ ðA:2Þ
model were obtained. Table 8 shows the parameters and errors of
fitting experimental data to Equation (4). According to this table, Parameter constraints: t0 > 0; m > 0:
as the concentration of light crude oil increases, parameter A,
generally, decreases; resulting in less dependency of shear rate A.3. Power Law
upon temperature. Furthermore, by increasing the light crude oil
concentration, the value of R2 increases and that of RMSE t ¼ kg_ n ðA:3Þ
decreases. This means that the greater the light crude oil vol%, the
more accurately the model can predict the shear stress as a Parameter constraints: k > 0; 0 < n < 1.
function of shear rate and temperature. A.4. Casson

CONCLUSIONS t0:5 ¼ t0 þ mg_ 0:5 ðA:4Þ

In this work, we examined the role of ultrasonic treatment as well
Parameter constraints: t0 > 0; k > 0.
as dilution with light oil for the improvement of rheological
behaviour of extra heavy oil samples. Based on the results of the
study, the following conclusions can be drawn: APPENDIX B: STATISTICAL FORMULAS
 The ultrasonic irradiation treatment can significantly decrease
the viscosity of the extra heavy crude oil and there is an B.1. Coefficient of Determination
optimum irradiation time, i.e. 10 min, which maximizes the XN  experiment
2
relative (compared to the reference value of viscosity measure i¼1
y model
i
yi
at room temperature) viscosity reduction. This can be R2 ¼ 1
X  2 ðB:1Þ
N experiment
attributed to the solubility of the fine asphaltene particles in i¼1
y model
i
averageðy i Þ
the crude oil. At this time interval, minimum activation energy
of the samples occurs. Beyond this optimum time, the viscosity
undergoes an increasing trend. B.2. Root mean square error (RMSE)
 The rheological model parameters for Newtonian, Bingham
0X  2 11 =2
Plastic, Power Law, and Casson models are obtained for N
y model

y experiment
different dilution ratios at optimum irradiation time. The B i¼1 i i C
RMSE ¼ @ A ðB:2Þ
Power Law model best matched the experimental data. N
 A new rheological model has been proposed which accounts
for the role of temperature in the Power Law model. The
proposed model has a better performance when the dilution
ratio increases.
REFERENCES
[1] A. Mai, J. Bryan, N. Goodarzi, A. Kantas, J. Can. Petrol.
NOMENCLATURE
Technol. 2009, 48, 27.
E activation energy of viscous flow (kJ/mol) [2] M. B. Dusseault. “Comparing Venezuelan and Canadian
k flow consistency index (Pa  s) Heavy Oil and Tar Sands,” Canadian International Petroleum
n flow behaviour index Conference, Petroleum Society, Calgary, 12–14 June 2001.
R universal gas constant (kJ/(mol  K))
[3] H. Nakhli, A. Alizadeh, M. S. Moqadam, S. Afshari, R.
T temperature (K)
Kharrat, M. H. Ghazanfari, J. Petrol. Sci. Eng. 2011, 78, 384.
T0 reference temperature (K)
vol% volume percent [4] S. W. Hasan, M. T. Ghannam, N. Esmail, Fuel 2010, 89, 1095.
wt% weight (mass) percent [5] C. Leyva, J. Ancheyta, C. Berrueco, M. Millan, Fuel Process.
Technol. 2013, 106, 734.
Greek Letters [6] M. D. Deo, J. Hwang, F. V. Hanson, Fuel Process. Technol.
g_ shear rate (1/s) 1993, 34, 217.
m viscosity (Pa  s) [7] K. Dunn, T. F. Yen, Fuel Process. Technol. 2001, 73, 59.

90 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING VOLUME 95, JANUARY 2017

 [8] H. W. Yarranton, J. Disper. Sci. Technol. 2005, 26, 5.
[9] A. Chamkalani, A. H. Mohammadi, A. Eslamimanesh, F.
Gharagheizi, D. Richon, Chem. Eng. Sci. 2012, 81, 202.
[10] H. Nourbakhsh, M. Yazdizadeh, F. Esmaeilzadeh, J. Petrol.
Sci. Technol. 2011, 80, 61.
[11] H. H. Ibrahim, R. O. Idem, Fuel 2005, 84, 311.
[12] R. A. Almehaideb, J. Petrol. Sci. Technol. 2004, 42, 157.
[13] M. Tavakkoli, M. Masihi, M. H. Ghazanfari, R. Kharrat, Fluid
Phase Equilibr. 2011, 308, 153.
[14] K. Akbarzadeh, H. Alboudwarej, W. Y. Svrcek, H. W.
Yarranton, Fluid Phase Equilibr. 2005, 232, 159.
[15] S. R. Panuganti, F. M. Vargas, D. L. Gonzalez, A. S. Kurup,
W. G. Chapman, Fuel 2012, 93, 658.
[16] Z. Yang, C.-F. Ma, X.-S. Lin, J.-T. Yang, T.-M. Guo, Fluid
Phase Equilibr. 1999, 157, 143.
[17] A. Y. Zekri, S. A. Shedid, J. Petrol. Sci. Technol. 2004, 42, 171.
[18] T. Tavassoli, S. M. Mousavi, S. A. Shojaosadati, H. Sale-
hizadeh, Fuel 2012, 93, 142.
[19] M. Tavakkoli, V. Taghikhani, M. R. Pishvaie, M. Masihi, S. R.
Panuganti, W. G. Chapman, J. Disper. Sci. Technol. 2014, 35,
1155.
[20] S. A. Shedid, J. Petrol. Sci. Technol. 2004, 42, 57.
[21] O. G. Gunal, M. R. Islam, J. Petrol. Sci. Eng. 2000, 26, 263.
[22] I. Najafi, S. M. R. Mousavi, M. H. Ghazanfari, A. Ramazani,
R. Kharrat, C. Ghotbi, Petrol. Sci. Technol. 2011, 29, 966.
[23] N. Bjorndalen, M. R. Islam, J. Petrol. Sci. Eng. 2004, 43, 139.
[24] M. H. Rad, M. Tavakolian, I. Najafi, M. H. Ghazanfari, V.
Taghikhani, M. Amani, Sci. Iran. 2013, 20, 611.
[25] G. R. Check, D. Mowla, Ultrason. Sonochem. 2013, 20, 378.
[26] Y. E. Guoxiang, L. Xiaoping, H. Pingfang, P. Fei, W. Yanru, S.
Xuan, Chem. Eng. Process. 2008, 47, 2346.
[27] S. Nii, S. Kikumoto, H. Tokuyama, Ultrason. Sonochem.
2009, 16, 145.
[28] W. P. Schellart, J. Struct. Geol. 2011, 33, 1079.
[29] P. Luo, Y. Gu, Fuel 2007, 86, 1069.
[30] D. A. Storm, R. J. Barresi, E. Y. Sheu, Energ. Fuel. 1995, 9,
168.
[31] R. B. Bird, W. E. Stewart, E. N. Lightfoot, Transport
Phenomena, revised 2nd edition, John Wiley & Sons,
New York 2002.
[32] E. C. Tynan, T. F. Yen, Fuel 1969, 48, 191.
[33] T. Nakamura, H. Okawa, Y. Kawamura, K. Sugawara,
Ultrason. Sonochem. 2011, 18, 85.

Manuscript received May 9, 2015; revised manuscript received

March 1, 2016; accepted for publication March 9, 2016.

VOLUME 95, JANUARY 2017 THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING 91