Clean Technologies and Environmental Policy
https://doi.org/10.1007/s10098-020-01893-0
ORIGINAL PAPER
Recycling of used lubricating oil by solvent extraction: experimental
results, Aspen Plus simulation and feasibility study
Nancy Zgheib¹D. Hosni Takache¹
Received: 30 September 2019 / Accepted: 27 June 2020
© Springer-Verlag GmbH Germany, part of Springer Nature 2020
Check for
updates
Abstract
This study investigates the recycling of used lubricating oil (ULO) by solvent extraction using a ternary solvent consist-
ing of 2-propanol, 1-butanol and methyl ethyl ketone (MEK) followed by a vacuum distillation unit. In order to select the
best combination of solvents for the liquid–liquid extraction process, many parameters were first evaluated on a laboratory
scale such as the effect of the solvent composition, the solvent-to-oil ratio and the temperature. The oil recovery efficiency
and the physicochemical properties of the oil obtained from the extraction process were measured, and the results showed
that a ternary system consisting of 50% 1-butanol/25% isopropanol/25% MEK is the best mixture for effectively reducing
the contaminants in ULO with a good sludge separation prior to the fractionation. Also, a solvent-to-oil ratio of 3:1 and a
temperature of 25 °C are the optimal parameters for the extraction when the aforementioned solvent mixture is used. The
whole re-refining process including the vacuum distillation unit was then simulated for a plant capacity of 2125 kg/h using
Aspen Plus™. Four pseudo-components (saturate, monoaromatic, diaromatic and polyaromatic) together with the nonrandom
two-liquid (NRTL) and universal quasichemical activity coefficient (UNIQUAC) were used to describe the liquid-liquid
equilibrium (LLE) in the system solvent/lubricating oils during the extraction phase. The calculated yields on the extraction
unit performed for different solvent composition, solvent-to-oil ratio and temperature showed a good agreement with the
experimental data performed at the laboratory scale which verifies the right choice of the solvent mixture. The investiga-
tion of the amount and composition of oil loss inside the sludge obtained from the simulation showed that the contaminant
reduction and the oil recovery depend on the percentage of 1-butanol in the solvent mixture. Also, the results confirmed that
the amount of MEK and 2-propanol in the solvent mixture should not exceed 25% and the best results were obtained when
1-butanol was used as a base solvent with an amount of 50%. Finally, to check the feasibility of the process, the investment
payback time and the return on investment of the re-refining process were evaluated. Based on the economic study results,
the production process was found to be highly profitable with a return on investment equal to 55% and a payout period of
1.9
years in the worst-case scenario where the price of lubricating oil is 2.5 $/L.
☑ Nancy Zgheib
1
nancyzgheib@usek.edu.lb
Department of Chemical and Petroleum Engineering, School
of Engineering, Holy Spirit University of Kaslik (USEK),
Jounieh, Lebanon
Published online: 14 July 2020
Springer Graphic abstract
ULO
Re-Refining of ULO
Transport
✓ Experimental Study
Dehydration
✓ Process Simulation
✓ Sensitivity Analysis
Feasibility Study
Solvent Extraction
Solvent Recycling
Fractionation
N. Zgheib, H. Takache
Recycled
Lubricating oil
.
.
.
Keywords Recycling · Used lubricating oil · Solvent extraction Aspen plus simulation · Feasibility study
Introduction
Used motor oil is a hazardous material and harmful for the
environment and for the human health because of the pres-
ence of pollutants, for instance heavy metals, polychlo-
rinated biphenyls and polycyclic aromatic hydrocarbons
(Kanokkantapong et al. 2009). In the absence of an appro-
priate treatment, used motor oil is normally discarded in
the ground and into waterways including sewers making it
an environmental hazard (Hamad et al. 2005). At the same
time, ULOS are considered as valuable resources since the
recycling of such materials will reduce the imports of oil
products and will have a positive impact on the environ-
ment as well (Totten and Bruce 2006). Numerous studies
have reported the recycling of lube oil; however, there has
been very little research reported on the re-refining of used
lubricating oil (Boadu et al. 2019). ULO can be recycled
and used as an alternative fuel in a variety of engine con-
figurations and furnaces or as an important component in
road paving asphalt. The most preferred option by experts
in terms of energy requirements, cost and environment
is the re-refining process (Jafari and Hassanpour 2015).
Springer
.
Re-refining is a process that removes all contaminants
for the purpose of reusing the base oil as lube oil. The
conventional methods of contaminants removal that have
been used on an industrial scale either require the use of
toxic materials such as sulfuric acid or implies high cost
technologies such as vacuum distillation. The acid pro-
cess followed by clay treatment is the oldest one. This
process is capable of producing high-quality base oil with
the lowest cost but have the highest environmental risk.
Membrane technology is another method for recycling the
used motor oils where hollow fiber membranes such as
polyethersulfone, polyvinylidene fluoride and polyacry-
lonitrile are used (Cao et al. 2009). Solvent extraction has
been employed as an alternate method to treat the used oil
and the solvent chosen should have maximum solubility
for the base oil and minimum for additives and impuri-
ties. Vacuum distillation is also used to separate asphalts,
additives and breakdown products from the base oil. This
process operates at high temperature above 300 °C and
proved to be impractical because of coking and fouling of
processing equipment. The extraction process using liq-
uid-liquid solvent performed prior to distillation of the Recycling of used lubricating oil by solvent extraction: experimental results, Aspen Plus...
used lubricating oil is an alternative solution to the sepa-
ration of asphalts and additives by vacuum distillation.
Many studies have been conducted on the re-refining
of used oil using solvent extraction (Whisman et al. 1978;
Dos Reis and Jeronimo 1988; Alves dos Reis and Jeronimo
1990; Martins 1997; Elbashir et al. 2002; Hamad et al. 2005;
Daham et al. 2017). Supercritical fluids propane and ethane
are used as extracting solvents for recycling used motor oils
but give low yield 72–80% (Rincón et al. 2003). According
to Reis et al. (Dos Reis and Jeronimo 1988; Alves dos Reis
and Jeronimo 1990), ketones and alcohols are the most effi-
cient solvents for extraction since they are miscible with the
base oil on the one hand and flocculate some of the additives
and carbonaceous compounds by applying an antisolvent
effect on the other hand. The capability to segregate sludge
is closely related to the difference between the solubility of
the solvent and the polyisobutylene, a viscosity improver
additive. Whisman et al. (1978) showed that a ternary sys-
tem consisting of 25% 2-propanol, 25% butanone and 50%
1-butanol is reasonably effective to reduce contaminants in
the oil with a good sludge separation. They indicated that
2-propanol rich systems produce high ash reduction but
poorly separated sludge, while butanone-rich systems lead to
a good sludge separation, but re-dissolve the contaminants.
To adjust the ash reduction and oil recovery properties, an
equal percentage of 2-propanol and butanone was employed
and 1-butanol was chosen as a base solvent.
Few studies have reported the simulation of an extraction
process using vacuum distillates as main feed (Coto et al.
2006a, b; Espada et al. 2008), and only one study (Botas
et al. 2017) has reported the simulation and design of the
re-refining process of ULO using propane as a solvent for
the extraction process. In fact, the simulation of the extrac-
tion process of such systems is complex due to the huge
number of components present in the lubricating oil. To be
able to properly reproduce the equilibrium properties dur-
ing the extraction operation, an appropriate thermodynamic
model must be used. Also, the number, the composition and
the properties of each component that are taken into con-
sideration to describe the system should be well defined.
The approach usually used to describe the components pre-
sent in petroleum product is based on distillation curves.
This method is not suitable for extraction processes since
the chemical structure of the oil and the solvents used has a
much bigger effect than the boiling temperatures. In order
to estimate the LLE of the systems furfural/lubricating,
Coto et al. (2006a, b) used a model based on a correlation
of pseudo-component properties (specific gravity, density,
refractive index and sulfur content) and NRTL parameters
with the oil average boiling temperature. This method has
proved to be highly effective for any lubricating oil cut and
needs only the average boiling point and three physical prop-
erties as experimental data.
In this study, three mixtures of composite solvents con-
taining 2-propanol, 1-butanol and MEK were first investi-
gated on a laboratory scale to treat the used oil. The choice
of the solvent and the ratio of the ternary system were based
on a literature investigation (Whisman et al. 1978; Dos Reis
and Jeronimo 1988; Alves dos Reis and Jeronimo 1990). The
objective of this investigation is to select the best combina-
tion of solvents as well as the optimal operating conditions
for the extraction unit. In fact, solvent selection is critical for
the development of an economical extraction system and the
optimal solvent selection is based on laboratory effort before
starting the design or the selection of the extraction unit.
For this purpose, the effect of the solvent type and composi-
tion, the solvent-to-oil ratio and the temperature on the oil
recovery efficiency and the physicochemical properties of
the oil obtained from the extraction process will be studied.
Then, the whole re-refining process will be simulated with
Aspen plus™, and the yields of the extraction unit obtained
by varying the solvent type and composition as well as the
solvent-to-oil ratio and the temperature will be compared to
the experimental results. The detailed composition of the
oil in the sludge obtained from the simulation will be inves-
tigated in order to study the effect of the solvent type and
concentration on the contaminant reduction and the oil loss.
Finally, a profit and loss analysis of the recycling process
will be performed to explore if the application of the process
at industrial scale could be recommended.
Experimental section
Materials
Virgin lubricating oil was purchased from the local mar-
ket. The used lubricating oil samples were collected from
Mercedes-Benz lube change center in Lebanon. The sam-
ple used in this study is a mixture of used oil coming from
automobiles operating with different engine conditions. The
sample was stored in a suitable tank away from light and at
room temperature. Solvents used for extraction were 2-pro-
panol, 1-butanol and MEK. The solvents purchased from
Sigma-Aldrich were of analytical grade with 99.5% purity.
Solvent extraction process
The oil was first filtered and then dehydrated using a rotary
evaporator. This operation was carried out to remove water
and some light hydrocarbons. The dehydrated used oil was
collected and then used for the next step of solvent extrac-
tion. Three different compositions of a ternary system con-
taining 2-propanol, 1-butanol and MEK were used in this
study, and the composition of the three mixtures is reported
in Table 1.
Springer N. Zgheib, H. Takache
Table 1 Composition (vol%) of the solvent mixtures used in the
extraction process
Mixture composition
2-propanol 1-butanol (%)
MEK (%)
(%)
Mixture A
25
50
Mixture B
25
40
Mixture C
40
35
25
535
25
5355
The used oil was weighed (mo) and mixed for 20 min with
the composite solvent mixture using an electrical stirrer at
500
rpm. The solvent-to-oil ratios employed in this study
were 2:1, 3:1 and 4:1, and the extraction temperatures used
were 25, 40 and 60 °C. The mixture was then left unshaken
for a period of 24 h to allow extraction-flocculation. Then,
solvent-oil solution was separated from the sludge by decan-
tation and the solvent mixture was recovered. This proce-
dure was repeated in all experiments for three times. The oil
recovered was weighed (mp), and the percentage of lubricat-
ing oil extracted was calculated using Eq. (1).
%R
= 100 ×
mp
то
(1)
where % R is the percentage of lubricating oil extracted,
то is the mass of the ULO, and mp is the mass of the oil
obtained after the separation of the solvent in the extract.
Analysis of samples
The ULO was tested for ASTM D-1160 method. The recy-
cled oil obtained from the extraction was analyzed for differ-
ent physical properties such as: Flash points by ASTM D56,
pour point by ASTM D97-02, viscosity index by ASTM
D445-18, density (specific gravity) by ASTM 1298, ash con-
tent by ASTM 482-13 and sulfur content by ASTM D4294.
Kinematic viscosity
The kinematic viscosity is the measure of the resistive flow
of a fluid under the influence of gravity. A fixed volume of
oil was charged into the capillary of a calibrated viscometer
which has a flow time above 200 s. The diameter of the
capillary was chosen by considering the expected viscosity.
The viscometer was placed into a constant temperature bath
maintained at the temperature on which the viscosity is to
be determined. A chronometer was used to determine the
time the oil takes to progress between two points highlighted
inside the capillary. The viscosity is determined by multiply-
ing the efflux time by the viscometer constant.
Pour point
The
pour point of petroleum products is the lowest temper-
ature at which the oil will flow under cooling prescribed
conditions of test. A specific amount of oil was placed in
a test jar which is closed with a cork engulfing a thermom-
eter. The oil was heated in a bath at 48 °C, then cooled at
intervals of 3 °C and tilted to determine the flow charac-
teristics. The pour point is reached when the fuel no longer
moves upon holding the specimen horizontally for 5 s.
Flash point
The flash point is the lowest temperature at which vapors
of the material will ignite under the application of an igni-
tion source. The test cup was filled with oil to the required
volume, and both the test cup and the specimen were main-
tained at a temperature that is at least 23 ±5 °C below the
flash point expected. The temperature of the specimen was
increased, and the ignition source is applied to the vapor
area in the cup. The sample is considered to have reached
the flash point when a large flame occurs and instantane-
ously propagates over the whole surface of the specimen.
Ash content
This method consists of measuring the amount of ash-
forming material in the oil which provides an idea about
the quality of the product and the adequacy for any
application since the ash represents the impurities and
unwanted components. The sample was placed in a suit-
able vessel and was burned until only ash was left behind.
The evaporating crucible was heated at 775 °C, cooled to
room temperature and weighed. The mass of the ash was
then calculated as a percentage of the original sample.
Sulfur content
This test measures the sulfur content in petroleum prod-
ucts. The sample was placed in the beam emitted from an
X-ray source, and the resulting fluorescent X-rays were
measured and the sulfur concentration in mass% was deter-
mined by comparing the accumulated count with counts
from previously prepared calibration samples.
Vacuum distillation of used lubricating oil
ASTMD 1160-13 standard method was used to determine
at reduced pressures the range of boiling points for used
motor oil. The sample of ULO was vacuum distilled at
Springer Recycling of used lubricating oil by solvent extraction: experimental results, Aspen Plus...
about 65 mm mercury and over a temperature range of
272-400 °C.
Experimental results
Oil recovery is an important parameter to be considered, and
it represents the economy of the process since it accounts for
the oil yield obtained from the extraction process. The ash
reduction is also a vital parameter in the extraction process
since it accounts for the reduction of the concentration of the
contaminants in the oil such as dirt, iron oxide, wear metals
and corrosion products as well as the concentration of ash-
producing additives. The sulfur reduction also accounts for
the removal of sulfur compounds and additives. The recov-
ered oil obtained from the extraction experiments performed
with mixtures A, B and C using a solvent-to-oil ratio of 3:1
was tested for oil recovery, ash reduction % by means of
ASTM 482-13 and sulfur reduction % by means of ASTM
D4294. The results obtained for different compositions of
the ternary solvents used are shown in Table 2.
As it is shown in Table 2, the optimum solvent mixture
that produces maximum oil recovery (96%) with maximum
ash reduction (28%) and an acceptable sulfur reduction
(16.7%) was achieved with solvent mixture B. These results
indicate that mixture B is capable of reducing contaminants
in the oil with a good contaminant separation. The high per-
cent oil recovery obtained for mixture B is in good agree-
ment with Whisman et al. (1978) who reported that using
higher percentage of MEK than 25% produces good sludge
separation and thus high oil recovery since MEK has a lower
solubility parameters value than the two alcohols. However,
increasing the amount of 2-propanol does not decrease the
ash forming material as reported in the literature. This differ-
ence might be associated with the nature of the base oil and
to the fact that these days ash-less materials are used in the
additive package instead of ash forming materials.
The solvent-to-oil ratio investigation was conducted on
mixture B. The investigated solvent-to-oil ratios were 2:1;
3:1 and 4:1, and the extraction temperature used was 25 °C.
The oil recovery results in relation to solvent-to-oil ratio are
shown in Fig. 1. It can be seen that the best condition after
which a further increase in the solvent oil ratio bring no
Table 2 Oil recovery %, ash reduction % and sulfur reduction % val-
ues obtained from the solvent composition experiments performed at
the laboratory scale
Oil recovery (%)
100
90
80
70
60
50
40
30
20
10
0
2
3
Solvent to oil ratio
4
Fig. 1 Percentage of oil recovery for optimum solvent-to-oil ratio
experiments performed on mixture B using an extraction temperature
of 25 °C
change in the percent oil recovery values is above a solvent-
to-oil ratio equal to 3:1. This can be attributed to the fact that
increasing the solvent-to-oil ratio improves the solubility of
the oil in the solvent resulting in decreasing the oil loss in
the sludge and increasing the oil recovery values. Solvent-
to-oil ratios higher than 3:1 are not considered economically
feasible by industry, and the experimental results showed no
improvement in terms of oil recovery. Thus, a solvent-to-oil
ratio of 3:1 will be considered to be the optimum ratio in
this work.
To study the effect of the temperature on the extraction-
flocculation process, the extraction was performed on mix-
ture B at three different temperatures (25, 40 and 60 °C)
using a solvent-to-oil ratio of 3:1. The oil recovery results
in relation to the extraction temperature are shown in Fig. 2.
No significant change in percent oil loss was shown with
increasing the extraction temperature. Thus, the extraction
process at a temperature higher than 25 °C does not lead to
a higher yield of oil recovery.
Oil recovery (%)
100
90
80
70
60
50
40
30
20
Mixtures
Oil recovery (%) Ash reduction
Sulfur
10
(%)
reduction
0
(%)
25
40
60
Temperature (°C)
Mixture A
90
27
20
Mixture B
96
28
16.7
Mixture C
84
20
14.3
Fig. 2 Percentage of oil recovery for optimum extraction temperature
experiments performed on mixture B using a solvent/feed ratio of 3:1
Springer
