Optimization of the Caustic Treatment of Jet-Fuel to Reduce the
Consumption of Caustic Soda in an Oil Refinery
ROXANA CORTÉS MARTÍNEZ1,*, ERENIO GONZÁLEZ SUÁREZ2, NANCY LÓPEZ BELLO2,
MERLYN LEITER BORMEY3
1Department of Chemical Sciences,
Autonomous University of San Luis Potosí,
Álvaro Obregón n.º 64 Col. Centro; C.P. 78000, San Luis Potosí, San Luis Potosí,
MEXICO
2Chemical Engineering Department,
“Marta Abreu” Central University of Las Villa,
Highway Camajuaní km 5 1/2, Villa Clara,
CUBA
3Department of Process and Hydraulic Engineering,
Metropolitan Autonomous University,
Av. San Rafael Atlixco, No. 186, Col. Leyes de Reforma, C.P. 09310, Mexico City,
MEXICO
*Corresponding Author
Abstract: - Jet-Fuel desulfurization is one of the most important processes in the petroleum industries due to the
high quality required for aviation fuels. In the pre-sent work, the desulfurization of Jet-Fuel with a content of
sulfurous compounds was studied using a process assisted with sodium hydroxide. It was considered that the
composition of the turbo fuel varies because different blends of crude oil are processed in the refinery under
study. The effects of operating conditions, such as reaction time (15 to 30 min) and the amount of caustic soda
(NaOH) in its solution (0.024 to 0.2 mol/L) were investigated. The objective functions (responses) were the
NaOH consumption and the total cost of production while the acidity and the total sulfur content of Jet-A1 were
considered as restrictions. The optimal conditions of the proposed model were obtained using the optimization
with the fmincon function of the MATLAB software. These optimal conditions were found to vary according to
the content of sulfur compounds present in the process feed. From the optimal points obtained, it was possible
to determine the frequency of change of the NaOH solution that allows its maximum use.
Key-Words: - Desulfurization, jet fuel, neutralization reactions, objective function, optimization, sulfurous
compounds.
Received: January 23, 2023. Revised: February 21, 2024. Accepted: March 12, 2024. Published: April 24, 2024.
1 Introduction
Significant industrial developments, especially those
using fossil fuels as the general source, intensely
increase environmental pollution and corrosion
problems. Exhaust gases of industrial towers
consuming fossil fuels contain ample quantities of
sulfur compounds that bring about many problems,
[1]. Petroleum impurities mainly contain
compounds such as , mercaptans, thiophene, or
other sulfur compounds, [2]. Desulfurization, one of
the fundamental fields in the prevention of
environmental pollution, has been proposed in
previous research, [3].
At the Refinery of Cienfuegos, Jet-Fuel
desulfurization is carried out using sodium
hydroxide (NaOH). The Jet-Fuel produced at this
refinery contains mainly hydrogen sulfide (),
light mercaptans (R-SH) and naphthenic acids
(RCOOH). The reactions that occur in this process
are, [4]:
(1)
(2)
International Journal of Chemical Engineering and Materials
DOI: 10.37394/232031.2024.3.4
Roxana Cortés Martínez,
Erenio González Suárez,
Nancy López Bello, Merlyn Leiter Bormey
E-ISSN: 2945-0519
25
Volume 3, 2024
(3)
(4)
One of the major operational problems in these
systems is the possible formation of stable sodium
naphthenate emulsions in the treated raw material.
Acid extraction is enhanced by a flow of a high
NaOH solution, but the possibility of NaOH eddies
in the treated stream also increases. The formation
of stable emulsions is more evident when using high
concentrations of NaOH, [5]. Generally, when the
feed fraction contains high concentrations of
naphthenic acids, low concentrations of NaOH
should be used to decrease emulsion formation. The
fresh NaOH injection flow must be regulated
according to the value of the acidity number in the
feed, [6].
After the desulfurization treatment, the NaOH is
termed spent caustic soda because it is the one that
captures sulfur, phenolic, and naphthenic
compounds, [7]. Spent soda is a waste classified as
hazardous and it has a high pH, [8]. High values of
reacted NaOH give the benefits of a lower
consumption of fresh NaOH and lower volumes of
spent soda to drain. As compensation, there may be
problems with the properties of the treated product
(acidity value higher than requested) and the
tendency to emulsion formation increases, [9].
Today, in the Refinery of Cienfuegos,
operational changes have been made to improve the
jet fuel desulfurization process. However, large
amounts of spent caustic soda are still generated
having a high content of free NaOH and it is not
treated in the waste plant. The objective of this
paper is to optimize the jet fuel desulfurization
process to minimize NaOH consumption at the
Refinery of Cienfuegos.
2 Problem Formulation
A great variety of crude oils from different parts of
the world such as Venezuela, Russia and Algeria
enter the Refinery of Cienfuegos, including Mesa
30, Merey 16, Sahara, and Lagomar, blended in
various proportions. This causes the amount of
sulfur compounds to vary in the raw material of the
process being analyzed. That is why seven different
blends are used for the jet-fuel desulfurization
optimization process. Table 1 characterizes these
blends.
At the Refinery of Cienfuegos, the caustic
treatment plant for the jet-fuel fraction is designed
to process 73 m3/h. The jet-fuel fraction (feed)
reaches the caustic treatment where sulfur
compounds are neutralized. This stream is mixed
with a sodium hydroxide solution through a mixing
valve before entering the vessel where the treatment
occurs. The treated stream exits the top of the vessel
towards the water wash stage. Chemically, the
vessel where the caustic treatment occurs is not
particularly sensitive to changes in either
temperature or operating pressure. The operating
temperature of this unit depends on the temperature
of the feed fraction, which must be kept close to 40
°C. The temperature of the feed fraction, the
concentration of fresh NaOH, and the degree of
consumption of the soda are the most important
parameters. These must be regularly monitored to
avoid operational problems.
It is necessary to determine the reaction rates for
each of the reactions involved. The initial and final
NaOH concentrations and the NaOH depletion time
are known. These values are obtained from the
operational control of the studied processes. With
the results obtained from the operational control, the
simulation, and the general expressions of reaction
speed, the mathematical models that define the
reaction speed of NaOH are obtained. The
differential method was used, which is based on the
real rates of the reactions and measures the slopes of
the concentration-time curves. In these systems,
bimolecular irreversible reactions are observed, with
different initial concentrations of the reactants. [10],
state that the equilibrium constants of reactions (Eq.
1) and (Eq. 2) are K1 = 9.0×106 L/gmol and K2 =
0.12 L/gmol respectively, at 25°C and infinite
dilution. Since the value of K2 is small, it may be
assumed that reaction (Eq. 2) does not occur at all
and the only reaction taking place in the liquid is
reaction (Eq. 1). Considering a batch reactor, the
kinetic equations of the reactions that occur in
desulfurization are as follows:
(5)
(6)
(7)
Polynomial regression was applied and the
equations show a deviation of less than 5%, the
adjustments being satisfactory. The global order of
the reactions of each process (n1=4.15, n2=3.678 and
n3=1.384) confirms the reaction rate and the affinity
of NaOH with each sulfur compound present. It is
confirmed that high concentrations of NaOH favor
the reaction rate with naphthenic acids.
International Journal of Chemical Engineering and Materials
DOI: 10.37394/232031.2024.3.4
Roxana Cortés Martínez,
Erenio González Suárez,
Nancy López Bello, Merlyn Leiter Bormey
E-ISSN: 2945-0519
26
Volume 3, 2024
Table 1. Characterizes these blends
Blends
Mesa 30 (%v)
Merey 16 (%v)
Sahara (%v)
Lagomar (%v)
(mol/L)
(mol/L)
(mol/L)
M1
73.14
-
20.86
-
0.0149
6.51 x 10-05
1.18 x 10-14
M2
19
5
76
-
0.0055
2.58 x 10-05
1.04 x 10-14
M3
-
30.24
69.76
-
0.0104
6.47 x 10-05
1.04 x 10-14
M4
60.86
-
-
39.14
0.0247
0
1.18 x 10-14
M5
100
-
-
-
0.0211
1.05 x 10-05
1.26 x 10-14
M6
-
-
-
100
0.0293
0
1.08 x 10-14
M7
23.06
-
23.06
53.88
0.0207
0
1.09 x 10-14
2.1 Optimization Method
[11], established that multiobjective optimization
implies a series of objective functions that are going
to be optimized. As in the nonobjective optimization
problem, the multiobjective optimization problem
usually has many constraints that must satisfy any
feasible solution (including the optimal solution).
The term "optimize" means to find a solution that
provides values, for all objective functions and that
is acceptable to the designer. However, if the same
values minimize or maximize all the objectives
simultaneously, the multiobjective optimization
problem can be considered as monobjective. In this
investigation, there are 2 objective functions with
various constraints. It is desired to optimize as a
function of two variables (NaOH concentration and
time) and both objective functions must be
minimized. The two variables affect the objective
functions in the same direction. This means that
there is a multi-objective optimization that can be
considered a single objective. MATLAB® R2015a
software was used to optimize the process; it
contains a set of tools to optimize all kinds of
equations. Fmincon function was used as the target
certain functions are nonlinear functions and the
system has equality constraints and inequality.
3 Problem Solution
To perform the optimization, the objective functions
and the process constraints were defined.
3.1 Objective Functions
The objective functions are determined by the
consumption of NaOH and the total cost of
production. In both cases, to achieve the minimum
values it is necessary to obtain the optimal NaOH
concentration and reaction time.
NaOH consumption
It is necessary to know the optimal concentration of
NaOH that allows an adequate quality of the final
product, [12]. This function is defined by the
reaction rate of NaOH against sulfurous compounds.
The kinetic expressions of each reaction are used for
that purpose. By design, the process should work
with a concentration of 3.3% wt of fresh NaOH.
Due to practical experiences, the concentration of
soda was initially reduced between 1 % wt and 2 %
wt, finally to 0.8 % wt, considerably reducing stable
emulsions and satisfactorily complying with the
acidity of the finished product. The NaOH solution
is considered depleted when it has a concentration
of 0.1% wt. According to the literature, vessel
volumes are frequently sized to provide a 15 to 30
minutes hold time, [12]. Taking into account that the
volume of solution is 25 m3 and that the molar mass
of NaOH is 40 kg/kmol:
(kg) (8)
Constraints:
Total cost of production ( )
To define the total cost of production function, the
procedure proposed by [13], was used (Table 2).
According to data available at the refinery, it
reached a direct estimate of the fixed capital
investment (FCI) of FCI = 24,493,024.8 CUP (24
CUP = 1 USD) and the total capital invested TCI =
26,942,325.6 CUP. Being the price of water of 37.2
CUP/m3 and of NaOH of 19.1 CUP/kg, [14], the
total cost of production is defined as follows:
(9)
(10)
(11)
For the calculation of general expenses (),
only administrative expenses were considered (0.04
), in this case the other aspects have no impact.
Being is the volume of water (m3) used to
prepare the NaOH solution used in the
International Journal of Chemical Engineering and Materials
DOI: 10.37394/232031.2024.3.4
Roxana Cortés Martínez,
Erenio González Suárez,
Nancy López Bello, Merlyn Leiter Bormey
E-ISSN: 2945-0519
27
Volume 3, 2024
desulfurization of jet-fuel and the mass of
NaOH (kg) used in said solution.
(12)
The refinery operates 330 days in a year, the
volume of the solution in the drum is 25 m3.
Therefore, the volume of water and the amount of
NaOH consumed in a year is defined as follows:
(m3) (13)
(kg) (14)
From the economic point of view, the unit
production cost () is the annual production
cost divided by the annual production volumes. This
will depend on variables that decide the process and
its productivity, such as reaction time and NaOH
concentration, which allows obtaining an expression
as follows, [12]:
(15)
And
(16)
Substituting Eq. (12) – Eq.(14) and Eq. (16) in Eq
(15):
(CUP) (17)
Table 2. Estimation of production costs
Components
Composition
Cost (CUP)
Direct Costs (
)
Raw Material
20 % TCP
Labor
10% TCP
Supervision
15 % TCP
Requirements
1 % TCP
Maintenance and repair
6 % FCI
1,469,581.49
Supply
Water
37.2
NaOH
19.1
Electricity
10 % TCP
Laboratory expenses
1 % TCP
Fixed Charges (
Depreciation
10 % FCI
2,449,302.48
Taxes
1 % FCI
244,930.25
Insurance
0.4 % FCI
97,972.10
Indirect costs ( )
Other costs
5 % TCP
3.2 Constraint Function
For future marketing of Jet-A1 fuel as a product of
high added value is essential that this meets the
quality standards of the market. In this case, only
the acidity and total sulfur content will be
considered.
Jet-A1 Acidity
The final acidity of Jet-A1 only depends on the
reaction of NaOH with naphthenic acid (RCOOH).
According to quality standards the final product
should have an acid content of 0.011 mgKOH/gJet-
A1, then,
(18)
Jet-A1 total Sulfur (ST)
For total sulfur, only the reactions of NaOH with
H2S and mercaptans (R-SH) are considered.
According to quality standards, the final product
should have a total sulfur of 0.3 ppm, then,
(19)
3.3 Optimization Results
The models obtained are fractional polynomials
with two variables to optimize that represent an
industrial process.
Table 3. Results obtained in the optimization of the desulfurization of the turbo-fuel
International Journal of Chemical Engineering and Materials
DOI: 10.37394/232031.2024.3.4
Roxana Cortés Martínez,
Erenio González Suárez,
Nancy López Bello, Merlyn Leiter Bormey
E-ISSN: 2945-0519
28
Volume 3, 2024
Blends
Initial Sulfur (kg)
(mol/L)
(mol/L)
(mol/L)
(mol/L)
(min)
M1
609.48
0.0149
6.51 x 10-5
1.18 x 10-14
0.157
15
M2
142.44
0.0055
2.58 x 10-5
1.04 x 10-14
0.0563
15
M3
193.35
0.0104
6.47 x 10-5
1.04 x 10-14
0.126
15
M4
1006.75
0.0247
1.18 x 10-14
0.2
15
M5
787.80
0.0211
1.05 x 10-5
1.26 x 10-14
0.18
15
M6
1188.37
0.0293
1.08 x 10-14
0.2
15
M7
463.29
0.0207
1.09 x 10-14
0.2
15
Table 4. Results of the evaluation of optimal points
Blends
NaOH consumed (kg)
Rate of change (day)
TCPU (CUP)
A(mgKOH/gJet-A1)
ST (ppm)
t (min)
M1
609.48
0.0149
6.51 x 10-5
1.18 x 10-14
0.157
15
M2
142.44
0.0055
2.58 x 10-5
1.04 x 10-14
0.0563
15
M3
193.35
0.0104
6.47 x 10-5
1.04 x 10-14
0.126
15
M4
1006.75
0.0247
1.18 x 10-14
0.2
15
M5
787.80
0.0211
1.05 x 10-5
1.26 x 10-14
0.18
15
M6
1188.37
0.0293
1.08 x 10-14
0.2
15
M7
463.29
0.0207
1.09 x 10-14
0.2
15
The objective functions must be minimized for
what they were added, turning the multiobjective
problem into a monobjective problem. The software
calculation time was approximately 0.5 s.
In addition, it has the advantage that it can be
used for other crude oil mixtures where the
concentrations of naphthenic acids and total sulfurs
vary and it will always have the standard quality
standards. However, it is only contemplated for a
caustic treatment process, that is, for concentrations
greater than 1.0 x 10-4 mol/L and 1.5 x 10-14 mol/L
of mercaptans and naphthenic acids respectively,
this process does not apply and therefore neither
does the model.
Table 3 shows the result of the optimization.
The optimal concentration of NaOH and the optimal
reaction time are obtained for each blend studied.It
is observed that the blends that need the highest
content are those with the highest content of
naphthenic acids and mercaptans. The high
concentrations of NaOH favor the extraction of H2S
and mercaptans but cause the formation of
emulsions when reacting with naphthenic acids. It is
established that the operating temperature of the
studied process is 40 °C, reducing the formation of
emulsions that lead to the entrainment of NaOH. On
the other hand, increasing the temperature increases
the thermal energy of the reaction. This alters the
interaction between sodium hydroxide and
naphthenic acid, thus decreasing the degree of
extraction. Due to this, the compromise that exists
between the NaOH concentration and the operating
temperature must be taken into account.
From the points obtained, compliance with the
quality parameters can be estimated using the
functions defined above. Table 4 shows the result of
the evaluation of the optimal points. By evaluating
the optimal points in Eq. (8), the amount of NaOH
consumed in desulfurization can be determined.
From this value, it is calculated how often must
change the NaOH solution. Using Eq. (17), the
minimum production cost in the desulfurization of
the jet-fuel of each mixture is calculated. To
determine the acidity Eq. (18) was used and the total
sulfur content of Eq. (19) was used.
All the blends comply with the established
quality parameters. When the solution reaches, the
minimum allowed concentration must be changed. It
is observed that the NaOH solution can be changed
for a longer period than is currently happening in
the refinery studied. The NaOH is not completely
reacted and a part of it is being sent to waste. If any
crude oil blend does not meet the quality parameters
for Jet-A1, it is recommended to recirculate the
product for a second cleaning step.
4 Conclusion
This study investigated the jet-fuel desulfurization
process and the factors affecting NaOH
consumption based on operational control data. It
was confirmed that high concentrations of NaOH
favor the rate of reaction with naphthenic acids. The
compromise between temperature and NaOH
concentration must always be taken into account to
avoid the formation of emulsions. Defined
mathematical models for optimization can be
considered predictive tools for technologists of the
studied processes. Therefore, it is possible to extend
its implementation in the analysis of blends used by
the refinery in the future.
International Journal of Chemical Engineering and Materials
DOI: 10.37394/232031.2024.3.4
Roxana Cortés Martínez,
Erenio González Suárez,
Nancy López Bello, Merlyn Leiter Bormey
E-ISSN: 2945-0519
29
Volume 3, 2024
[1] Esmaeili Faraj, Seyyed Hamid, Bijani,
Abdolmotaleb, & Saei Moghaddam,
Mojtaba. (2019). Simulation and
Optimization of Demercaptanization of
Propane and Butane in South Pars Gas
Refineries. Applied Research in Chemical
Polymer Engineering, 3(2), 3-14. SID,
[Online]. https://sid.ir/paper/267787/en
(Accessed Date: April 5, 2024).
[2] Motahari K, Abdollahi-Moghaddam M,
Rashidi. Mechanism study and determination
kinetic of catalytic oxidation of mercaptans
in Merox process. South African of Chemical
Engineering, Vol. 33, 2020, pp. 116-124,
https://doi.org/10.1016/j.sajce.2020.06.003.
[3] Hossain MN, Park HC, Choi HSJC. A
comprehensive review on catalytic oxidative
desulfurization of liquid fuel oil. Catalysts,
Vol.9, No.3, 2019; pp.
229, https://doi.org/10.3390/catal9030229.
[4] Pino-Cortés E, Montalvo S, Huiliñir C,
Cubillos F, Gacitúa JJP. Characteristics and
treatment of wastewater from the mercaptan
oxidation process: a comprehensive review.
Processes, Vol. 8, No.4, pp. 425.
https://doi.org/10.3390/pr8040425.
[5] Saha R, Uppaluri RV, Tiwari P. Influence of
emulsification, interfacial tension, wettability
alteration, and saponification on residual oil
recovery by alkali flooding. Journal of
industrial and Engineering Chemistry,
Vol.59, 2018, pp. 286-296.
https://doi.org/10.1016/j.jiec.2017.10.034.
[6] Wu C, De Visscher A, Gates ID. On
naphthenic acids removal from crude oil and
oil sands process-affected water. Fuel,
Vol.253, 2019; pp.1229-1246.
https://doi.org/10.1016/j.fuel.2019.05.091.
[7] Seyedin S, Hassanzadeganroudsari M.
Evaluation of the different methods of spent
caustic treatment. Int. J. Adv. Res. Sci. Eng.
Technol., Vol. 5, 2018, pp. 5275-5283.
[8] Hawari A, Ramadan H, Abu-Reesh I,
Ouederni M. A comparative study of the
treatment of ethylene plant spent caustic by
neutralization and classical and advanced
oxidation. Journal of Environmental
Management, Vol.151, 2015, pp.105-112,
https://doi.org/10.1016/j.jenvman.2014.12.03
8.
[9] Yusra A. Abd Al-Khodor, Talib M. Albayati.
Protection E. Employing sodium hydroxide
in desulfurization of the actual heavy crude
oil: Theoretical optimization and
experimental evaluation. Process Safety and
Environmental Protection, Vol. 136, 2020;
pp.334-342,
https://doi.org/10.1016/j.psep.2020.01.036.
[10] Hikita, Haruo; Ishikawa, Haruo; Murakami,
Yasumasa. Absorption of Hydrogen Sulphide
into Sodium Hydroxide Solutions in an
Agitated Vessel with a Flat Gas-Liquid
Interface. Bulletin of University of Osaka
Prefecture. Series A, Engineering and
natural sciences, 1971, Vol. 19, No. 2, p.
349-358.
[11] Coello CAC. Evolutionary algorithms for
solving multi-objective problems. Springer;
2007.
[12] Cortés Martínez R, Ramos Miranda FE,
González Suárez E. Solution Methods for
Multi-Objective Optimization Problems in
Oil. Chemical Technology Journal (in
Spanish), Vol. 41, No. 1, 2021, pp.75-91,
[Online].
https://tecnologiaquimica.uo.edu.cu/index.ph
p/tq/article/view/5181 (Accessed Date:
March 25, 2024).
[13] Peters MS, Timmerhaus KD, West RE. Plant
design and economics for chemical
engineers: McGraw-Hill New York; 2003.
[14] Lobelles Sardiñas G, López Bastida E,
Pedraza Gárciga, J, Peralta Suárez, L. A
methodology based on an ecological
economy approach for the integrating
management of the sulphurous water in an
oil refinery. Sugar Cane Journal (in
Spanish), Vol.43, No.4, 2016, pp. 50-62,
[Online].
http://scielo.sld.cu/scielo.php?script=sci_artt
ext&pid=S2223-
48612016000400006&lng=es&tlng=es
(Accessed Date: April 3, 2024).
International Journal of Chemical Engineering and Materials
DOI: 10.37394/232031.2024.3.4
Roxana Cortés Martínez,
Erenio González Suárez,
Nancy López Bello, Merlyn Leiter Bormey
E-ISSN: 2945-0519
30
Volume 3, 2024
References:
- Merlyn Leiter Bormey. Research, methodology,
software, writing - first writing and data
conservation.
- C. Roxana Cortés Martínez. Project management,
obtaining funding, resources, writing - first
writing and data retention.
- Nancy López Bello. Supervision, methodology,
conceptualization, writing - revision and editing.
- Erenio González Suarez. Project management,
formal analysis, software, writing - revision and
editing.
Sources of Funding for Research Presented in a
Scientific Article or Scientific Article Itself
No funding was received for conducting this study.
Conflict of Interest
The authors have no conflicts of interest to declare.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the
Creative Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en
_US
International Journal of Chemical Engineering and Materials
DOI: 10.37394/232031.2024.3.4
Roxana Cortés Martínez,
Erenio González Suárez,
Nancy López Bello, Merlyn Leiter Bormey
E-ISSN: 2945-0519
31
Volume 3, 2024
Contribution of Individual Authors to the
Creation of a Scientific Article (Ghostwriting
Policy)
