Use of Asphaltene Stabilizers for the Production of Very Low Sulphur Fuel Oil

Abstract

Marine fuel oil stability has always been an issue for bunkering companies and ship owners all around the world and the problem has become even more apparent with the introduction of the Global Sulphur Gap by the International Maritime Organization (IMO) in 2020. In this article, the historical background and the technical reasons why marine fuel oils lose their stability, as well as methods for preventing such instability from occurring, are presented. While it is possible to make fuel compositions stable by adjusting their composition in such a way that the components of the fuel are compatible, considering that marine fuel oils are often comprised of the least value-added products, the method of adding special fuel oil stabilizers (also known as “asphaltene dispersants”) is usually preferred. An overview of such stabilizers is presented; their chemical composition, based on the information provided by the manufacturers and/or inventors is studied. In addition, the experimental research of the produced marine fuel oil and its components is carried out. The results of the model composition studies show that adding even as little as 10% of residual asphaltene-rich components can make a composition with a high stability reserve unstable. It was also shown that the content of the asphaltene-rich component in a stable fuel can be increased from 3% to 10% by introducing stabilizers in low amounts (up to 2000 ppm), thus lowering the amount of higher value-added, mostly naphthene-paraffinic-based components. Different methods of fuel stability evaluation were studied and tested, most of them being in correlation with one another. Several types of stability enhancers were tried out on unstable fuel, with stabilizers based on alkylphenol formaldehyde resin showing the best results.

1. Introduction

On 1st January 2020, the IMO regulation came into force which limited the sulphur content of all marine fuels to 0.5% wt. unless ships are equipped with scrubbers (IMO 2020) [1,2]. Such fuel, if it contains residual components of oil refining, is called very low sulphur fuel oil (VLSFO). The requirements for sulphur content in marine fuel were tightened gradually in stages: until 1 January 2012, the limit was 4.50% wt., and from 1 January 2012 to 1 January 2020 it was 3.50% wt.

IMO 2020 did not affect ships within Emission Control Areas (ECAs), as these regions are subject to a stricter limit of 0.1% wt. of sulphur. Since 2015, the maritime area of Northern Europe, which includes the Baltic Sea, the North Sea, the English Channel, as well as territory within 200 nautical miles from the coasts of the United States and Canada, have been considered as ECA, and since 2022 the Mediterranean Sea, namely the area bounded by the Strait of Gibraltar, the Dardanelles and the Suez Canal, has become an ECA as well [3].

One way to meet the IMO 2020 regulations is to use alternative marine fuels, such as liquified natural gas (LNG), liquified petroleum gas, methanol, ammonia, hydrogen, etc., which scarcely contain any sulphur [4,5,6]. However, since most of the marine fuel used in the world today is of oil origin, the problem still persists.

For petroleum marine fuels, sulphur content is determined by its content in the components. For this reason, there are two ways to reduce the mass fraction of sulphur in the fuel composition—the so-called direct and indirect hydrotreatment.

Direct hydrotreatment is a straightforward process of hydrodesulphurization, hydrotreating, hydrocracking and other methods of direct sulphur removal from heavy oil residues. The indirect method implies the compounding of low-sulphur (which have undergone some kind of hydrotreatment process or initially contained low amounts of sulphur) and high-sulphur components. The sulphur content in the final fuel is proportional to its content in the components and to the content of the components in the fuel. However, changing the composition of the fuel system by mixing stable high-viscosity fuel with a low-viscosity paraffin fraction, caused by the need to achieve the required value for sulphur content or viscosity, can also affect the stability of asphaltenes and make the entire fuel mixture unstable. If this happens, the components are called incompatible (in the current blending ratio). Stable fuel does not form residue, it does not delaminate, and has constant composition and properties during storage and use [7].

The cause of fuel instability is known to be the precipitation and agglomeration of asphaltenes. This results in sludge formation and depends on different parameters, including the thermodynamic condition of the system (the temperature in the storage tank), structure and the concentration of structural units of asphaltenes, their origin, and the medium that surrounds the asphaltenes [8,9,10].

The asphaltenes present in the compositions containing residual petroleum products function as coagulation centres of high-molecular-weight compounds. Asphaltenes form associates, the so-called micelle nuclei, the outer layer of which is made of polar aromatic hydrocarbons. This layer prevents the mutual fusion of micelles and the formation of large associates. It is important to note that asphaltenes formed during thermic oil refining processes, as a result of chemical reactions caused by thermobaric conditions, due to their less complex structure and composition, show poorer aggregative stability compared to native asphaltenes [11]. Moreover, in the straight-run residues, the resins that are naturally present in the crude oil are often able to stabilize the electrostatically charged asphaltenes. During refining processes, these resins can be altered, and thus the dispersion of asphaltenes and fuel stability deteriorate [12].

Poor stability of marine residual fuel and sedimentation can cause problems both at the storage stage (accumulation of residue in tanks, leading to deterioration in the quality of discharged oil products and reducing the useful volume of tanks) and at the stage of fuel operation (blocking of filters, clogging of pipes in an engine, compromised combustion leading to turbocharger damage, operational problems with fuel centrifuges, etc.). In January 2019, Lloyd’s Register (LR) issued a FOBAS alert showing that a number of samples of ISO-F-RMG380 grade fuel that did not meet total sediment potential requirements (failing to filter in 25 min) had been discovered in Singapore. LR had urged shipowners whose vessels planned to bunker in Singapore to take particular care when checking whether their fuel met the regulatory standards, particularly in terms of stability [13]. A similar situation was recorded in November 2015 in West Africa, when the total draft values in the ports of Lomé (Togo) and Abidjan (Ivory Coast) exceeded the norm by 2–5 times (0.25–0.46% wt.). The high sediment results were partially due to dirt present in the fuel, but mainly due to precipitated asphaltenic material [14].

According to ISO 8217 [15], the stability of marine fuel is determined with the total sediment via a hot filtration (HFT) test (ISO 10307-1 [16]) for distillate fuels and total sediment aged (TSA) test (ISO 10307-2 [17]) for residual fuels, and the tests are carried out with the help of special equipment. Besides these two, other methods for the determination of fuel stability are known. In Russia, a complex of methods called “Residual fuel oils. Test for straight-run” under the GOST R 50,837 standard assembles a set of methods that through different means show the stability of a fuel. This set includes toluene equivalent (GOST R 50837.3), xylene equivalent (GOST R 50837.4), stability and compatibility by spot (so-called spot test, GOST R 50837.7), and total sediment (GOST R 50837.6), the latter being virtually the same as the ISO 10307-2 method [18,19,20,21]. Additionally, a number of studies over the years have presented different ways of predicting the stability of asphaltenes, some of them actually being focused on the stability of crude oil and mixed feeds for refinery units, but finding application in marine fuel as well, due to the similarity of the problems [22].

Wiehe et al. [23] presented the Oil Compatibility Model which was designed to determine the mutual solubility of various fuel components. It is based on the calculation of two parameters: the insolubility number, which determines the degree of insolubility of asphaltenes, and the solubility blending number, which, on the contrary, determines the degree of solubility of the dispersion medium in relation to asphaltenes. To calculate these indicators, at least two tests must be carried out. Using the first test, the maximum volume of n-heptane is determined, which is mixed with the test component without precipitation. The second test is to determine the toluene equivalent, the essence of which is described later in the present study. Based on these two parameters, the insolubility number and the solubility blending number are calculated and the fuel is considered to have high stability provided that the solubility blending number is higher than the insolubility number.

Moura et al. [24] studied some of the indices and models of crude oil stability, including using the BMCI (Bureau of Mines Correlation Index) and toluene equivalent, which Demidova et al. [25] later proved to be applicable to heavy fuel oil as well. Comparing the values of BMCI and toluene equivalent for a large number of residual fuels, both compatible and incompatible, showed that for all stable mixtures, the difference between the two parameters (BMCI minus toluene equivalent) must be no lower than 7–9, going as high as 11–14 for the fuel to be stable for a long period of time.

The colloidal instability index (CII) has also been widely used for determining the stability of asphaltenes in crude [26] and heavy fuel oils [27]. This method is based on determining the SARA (saturates, aromatics, resins, and asphaltenes) contents of the fuel and calculating the CII. The lower the CII, the better the stability of the asphaltenes in the medium, with the preferable values being less than 0.7.

It is important to note that none of the methods for assessing the stability of marine fuels can utterly guarantee the absence of sediment formation and its harmful effects on ship equipment. That is why an accurate determination of the suitability of fuels, especially those obtained by mixing different batches, is possible only by testing it in practice [28].

There are two options for increasing the stability of marine fuels with improved environmental quality.

The first option is to regulate the composition of marine residual fuel. It means that at the stage of developing the fuel oil composition, the selection of components is carried out and their compatibility is tested according to stability assessment methods.

Colloidal–chemical transformations that occur in petroleum disperse systems when different components are combined determine the final properties of the final mixture. Blending the residual stable fuel containing asphaltenes with low-viscosity distillate paraffin fraction can make the whole fuel mixture unstable, and vice versa [28]. Therefore, when blending fuels, there is a need to take into account their compatibility. The compatibility of fuels is considered as the stability of their mixture [29].

To date, many research items have been published regarding the patterns and regularities of the stability of marine fuel with improved environmental properties obtained by mixing residual and distillate components. Scientists from Saint Petersburg Mining University have developed a method of presenting the results of stability studies of fuel systems obtained by mixing several components in the form of three-phase stability diagrams. Such a result of stability assessment allows one to establish stability areas and determine the maximum permissible concentrations of each fuel component in a multi-component system [30]. In addition, studies have been carried out to optimize marine fuel compositions [31,32,33].

The second way to increase the stability of marine fuel is to introduce special stabilizer additives (also called compatibility or stability enhancers, asphaltene dispersants, and even dispersant additives) into its composition. There is often a need to refuel partially filled tanks in port with other fuels, and the compatibility of the two fuels is not checked in such cases, which may cause fuel instability during further transportation and operation. For these reasons, the use of stabilizers often becomes a necessity [30].

Dispersing additives are used as stabilizers for marine fuels—these are surfactants (both anionic and non-anionic) added to fuels in small quantities, usually no more than 0.1–0.25%, and soluble in them [34,35,36]. These additives change the structure and properties of the formed insoluble associates; similar to natural resins, they have an affinity for polar asphaltene molecules. Thus, when introducing such additives, the formed insoluble products are not deposited on the equipment and vessels. The operating principle of dispersing additives is based on keeping the insoluble products in a finely dispersed state and preventing their sedimentation and coagulation, which is due to their surface-active and dissolving properties [37]. Asphaltene stabilizers include inhibitors, which slow down or prevent the initial precipitation of asphaltene molecules from solution, and dispersants, which slow down or prevent the flocculation of precipitated asphaltene molecules into larger particles [12]. Effective stabilizers are compounds that have a peptizing effect, as well as those that block particle adhesion areas, reducing the strength of the contacts between layers of the dispersion medium through which the dispersed phase particles contact each other.

The operating principle of each stabilizer depends on its chemical structure and the way it interacts with the asphaltene surface. Zhang et al. studied the interactions between asphaltenes in crude oil and various stabilizers, finding that polymeric additives with an acidic group were superior to those of small molecule size and that stabilizers with an acidic group are generally better than those with basic groups. In turn, acid–base interactions were considered to be most effective in dispersing asphaltenes, followed by π-π interactions and hydrogen bonds [38].

Table 1. Patent review of marine fuel stabilizers.

Patent No.	Company	Main Component	Dose Rate	Reference
US 5,925,233 A	Clariant	Secondary alkylsulfonic acid C11–C22	0.0002–0.2%	[39]
US 6,488,724 B1	Clariant	Amine; fat-soluble salt with a metal from the first group of the periodic table	0.0002–0.2%	[40]
US 6,063,146 A	Clariant	Carboxylic acid ethers; alkoxylated alcohols	0.0002–0.2%	[41]
EA 018,052 B1	Clariant	Substituted phosphoric acids	0.00005–1%	[42]
EP 1,362,087 A2	Clariant	Cardanolaldehyde resins or their mixture with alkylphenol–formaldehyde resins and alkoxylated amines	0.0002–0.2%	[43]
US 6,204,420 B1	Clariant	C4+ carboxylic acids or esters thereof, alkyl substituents C18–C22	0.0002–0.2%	[44]
WO 0,218,454 A2	Cognis Deutschland GmbH & Co.	Polyether amides	0.015–0.05%	[45]
US 11,180,588 B2	Stepan Company	Alkylsuccinic anhydride	0.0005–0.5%	[46]
US 5,948,237 A	Clariant	Sarcocynates, alkylphenol formaldehyde resins, oxyethylated amines, alkylbenzenesulfonic acids	0.0002–0.2%	[47]
US 6,180,683 B1	Clariant	Alkylphenol tars, oxalkylated amines	0.0002–0.2%	[48]
US 5,021,498 A	Ecolab USA Inc.	Alkyl-substituted phenol–formaldehyde liquid tar	0.0001–1.0%	[49]

reviews the patent documentation related to marine fuel stabilizers [39,40,41,42,43,44,45,46,47,48,49].

One of the most common types of stabilizers is alkylphenolformaldehyde resin with a molecular weight from 2 to 5 thousand g/mol and an alkyl chain length of 8 to 12 carbon atoms (with 9 being the most common). Its functional efficiency is related to the fact that the polar -OH group of the phenolic part of the resin forms hydrogen bonds with asphaltene molecules, and in particular with their hydroxyl and amine groups, thus being adsorbed on the surface of particles. At the same time, the long alkyl radical of the resin provides, on the one hand, good solubility of the stabilizer and asphaltenes in the hydrocarbon part of the fuel, and, on the other hand, steric hindrances preventing the formation of larger structural units of asphaltenes [50,51]. Dispersants based on alkylphenol formaldehyde resin are produced by Clariant and Wilhelmsen [52,53].

Oxyethylated fatty amines are also a common type of dispersant, used, for example, by the marine chemical manufacturer Drew Marine. While the nature of the interaction between these substances and asphaltenes is not fully understood, it is at least partially due to an acid–base interaction between the amine part of the dispersant and the acid groups of the asphaltenes [54].

Alkylbenzenesulfonic acids, and in particular dodecylbenzenesulfonic acid, are also known as stabilizers for marine fuels. Dodecylbenzenesulfonic acid shows even higher solubilizing ability against asphaltenes compared to alkylphenol formaldehyde resin because its sulfonic group (-SO₃H) has a higher polarity compared to the -OH group. Furthermore, dodecylbenzenesulfonic acid is known to have a synergetic effect on both π-π interaction and acid–base reaction [36].

Polyisobutylene succinimide, an ashless dispersant based on polyisobutylene, in which the nitrogen functional group is linked by maleic anhydride to a hydrocarbon tail, is also known to be used as a marine fuel stabilizer [51]. The interaction between the molecule and asphaltenes is primarily based on the π-π bond, with a possibility of hydrogen bond formation and acid–base reaction depending on the hydrocarbon radical attached to the aromatic head. The polyisobutylene branch helps the stabilizer to break the asphaltene aggregates as well as create steric hindrances so it is hard for them to coagulate again in the future [55].

Biodiesel (FAME—fatty acid methyl esters) has also been studied by numerous researchers and found to have a dispersing ability when mixed with marine fuel oil, though the amount in which it is added to the fuel (up to 60%) makes it more of a component rather than an additive. The study of biodiesel–asphaltene molecular dynamics shows that the interaction energy between these molecules is higher than that within their own kind, making it possible for FAME molecules to adsorb on asphaltene molecules and inhibit the precipitation process [56].

Thus, when solving the problem of meeting the requirements for low sulphur content in marine fuel, there is a related problem of instability of the obtained compositions, and the ways to solve it are the subject of this paper.

The purpose of this work is to study the influence of the group composition of marine residual fuel and dispersants of different origins on its stability; moreover, to study the compatibility of different fuel oil components and the way the stability pattern can be changed by using asphaltene stabilizers of different origins. While the influence of group composition of marine fuel oil on its stability has been studied before, this work explores the correlation between different stability inspection methods and addresses the knowledge gap by comparing the use of stabilizers of different origins in the same unstable fuel.

2. Materials and Methods

In this work, 8 types of petroleum products selected from different Russian refineries were used as objects for the research (

Table 2. Objects of study.

VLSFO Components	Code Name
Straight-run diesel fuel	SRDF
Hydrotreated diesel fuel	HTDF
Light Vacuum Gasoil	LGO
Heavy Vacuum Gasoil	HGO
Vacuum residue	VR
Hydrocracking residue	HCR
Visbreaking residue	VisR
Catalytic Cracker Light Cycle Oil	LCO

). The presented objects were chosen as representative because they can act as the model components for the sake of this study: VisR and VR are residual components with high asphaltene content; LCO is a component with a predominantly aromatic base; HTDF and HCR are used as low-sulphur blending components to improve the environmental quality of marine fuels and have a paraffin-naphthenic base.

Table 3. Physico-chemical parameters of the research objects.

Objects	Parameters
	Kinematic Viscosity at 50 °C, mm2/s	Density at 15 °C, kg/m3	Sulphur, % wt.	Flashpoint (Closed Cup), °C	Water Content, % wt.	Carbon Residue, % wt.
SRDF	3.7	856.02	0.368	78	traces	0.020
HTDF	2.6	833.66	0.001	60	absence	0.015
HGO	140.55	936.06	0.942	226	absence	1.52
HCR	10.66	855.12	0.0155	168	traces	0.022
LGO	3.99	827.15	0.555	58	traces	0.020
VisR	- 1	989 2	1.020	- 3	traces	16.25
VR	- 1	963 2	1.106	- 3	traces	9.65
LCO	2.90	951.01	0.998	88	traces	0.036

¹ The kinematic viscosity was not determined due to the high values for these components. ² The density was determined using a calculation method when the sample was mixed with toluene because of the high value of the sample’s density. ³ The closed cup flashpoint for visbreaking residue and vacuum residue has not been determined as it is known to be significantly higher than the requirements for residual marine fuels.

presents the physicochemical parameters of the objects used in this work; studies were conducted according to the requirements of ISO 8217. Additionally, since the low viscosity of a fuel is ensured by the high content of the low-boiling distillates in it, distillation testing of some of the components was also performed, and the distillate composition is presented in

Table 4. Distillate composition of some of the fuel components.

Distillation	Temperature, °C
	SRDF	HTDF	LGO	HCR
IBP	207.1	185.4	246.0	323.0
5%	230.2	206.5	266.7	339.2
10%	245.6	221.8	276.3	343.2
20%	260.6	244.2	289.5	347.6
30%	271.2	260.3	301.4	353.2
40%	280.3	273.9	312.2	360.2
50%	289.1	284.5	326.4	369.6
60%	298.9	295.1	337.0	380.0
70%	310.1	306.6	349.1	392.5
80%	323.7	319.7	360.0	407.0
90%	339.6	336.8	384.6	425.1
95%	354.9	349.1	396.2	432.1
FBP	364.8	356.9	402.6	432.4
% of residue	1.8	1.8	1.6	2.2

.

In this study, three commercially available fuel oil stabilizers were used, which were given the code names S1, S2, and S3. S1 and S2 additives are based on alkylphenol–formaldehyde resin, the two being produced by different facilities but supposedly having similar structures (it is impossible to say for sure since this information is deemed confidential), and the S3 additive is based on polyisobutylene succinimide. While there are definitely more available stabilizers on the market, these three were the ones that could be easily acquired for research purposes, while being representative of the most effective and widely used commercial ones. Some of the physical properties of the S1, S2, and S3 stabilizers were measured and presented in

Table 5. Physical properties of stability enhancers used in this study.

Parameter	S1	S2	S3
Appearance	Homogenous semi-viscous liquid of dark brown to black colour
Density at 15 °C,	902.04	927.40	891.25
Kinematic viscosity at 50 °C	51.54	41.48	69.23
Pour point, °C	minus 30	minus 44	minus 48

.

The test methods used in this study are presented in

Table 6. Study methods.

Parameter	ISO 8217 Regulation	Method
Density at 15 °C, kg/m3	depends on the category	ISO 3675 [57]
Kinematic viscosity at 50 °C, mm2/s	depends on the category	ISO 3104 [58]
Flashpoint, closed cup, °C	min. 60	ISO 2719 [59]
Mass fraction of sulphur, % wt.	statutory requirements, max. 0.5 for VLSFO	ISO 8754 [60]
Water content, % wt.	depends on the category	ISO 3733 [61]
Carbon residue, % wt.	depends on the category	ISO 10370 [62]
Distillation, °C	not regulated	ISO 3405 [63]
Pour point, °C	not regulated	ISO 3016 [64]
Total sediment (Hot Filtration Test, HFT), % wt.	not regulated	ISO 10307-1 [16]
Total sediment aged (TSA), % wt.	max. 0.10	ISO 10307-2 [17]
Separability number	not regulated	ASTM D7061 [65]

[15].

For some of the samples, density at 15 °C was too high for it to be determined using the ISO 3675 method, as stated in Table 6. For these samples, the density was determined by a calculation method when the sample was mixed with toluene. It was assumed that density can be, with a certain degree of error, considered an additive property of petroleum products. Therefore, this principle was applied to the samples. The samples were mixed with toluene in a 1:1 mass ratio, and then the density of the mixtures was measured. Finally, a calculation was carried out according to the following equation:

1/d_blend = 0.5/d_sample + 0.5/867

(1)

where d_sample is the density of the studied sample and d_blend is the density of the toluene/sample blend, which is determined as stated in Table 6, while 867 is the density of the toluene at 15 °C.

The following indicators were also used as additional methods for studying the stability of fuel oil:

• Toluene equivalent. This method of stability assessment was carried out according to GOST R 50837.3 [18]. The essence of the method is to dissolve the sample in standard mixtures of toluene and heptane at varying concentrations of toluene with an evaluation of the stains formed on the membrane filter. The toluene equivalent is calculated based on two values of toluene concentration, the first being the minimum concentration that gives a ring pattern on the stain and the second is the lower concentration at which no ring is observed. The stability criterion is a value of 30% vol. units or less.
• Xylene equivalent. A stability study using this method was carried out according to GOST R 50837.4 [19]. The essence of the test was dissolving the sample in a mixture of xylene and normal heptane and examining a drop of such a mixture on a paper filter. The value of xylene equivalent is determined by the minimum volume content of xylene in the solution, which does not form a ring on the filter. The stability criterion is a value less than or equal to 25/30% vol.
• Determination of stability and compatibility by spot. This method is also used to assess the stability of samples according to GOST R 50837.7 [21]. The determination is based on a visual evaluation of a spot on a paper filter formed by a heated and homogenized sample. After applying a drop of the sample on the paper filter, it is kept for 1 h at 100 °C in a desiccator. The method can be used to determine the stability of the fuel and the compatibility of its components. The stability criterion is a spot number of no more than no. 2. For the spot reference, please refer to the method documentation.
• Determination of the group composition was performed using liquid adsorption chromatography with gradient displacement on a chromatograph apparatus called “Gradient-M”. A Gradient-M chromatograph can be used for the quantitative determination of group component composition of heavy oil fractions—fuel oils, vacuum residues, visbreaking residues, cracking oils, oxidized and natural bitumen, and other oil products, including distillates. The principle of operation of the liquid chromatograph consists of the separation of the analysed product in the chromatographic column by the flow of mobile phase consisting of a mixture of solvents, transfer of the eluent in the form of a thin layer on the conveying chain, removal of the components of the mobile phase in the evaporator, thermo-oxidative destruction of the separated components of the analysed substance in the oxidation cell, and detection of the formed carbon dioxide by the detector via thermal conductivity. The chromatogram is then processed by the lab staff, acquiring the group composition, consisting of paraffin-naphthenes, light, medium, and heavy aromatics, resins, and asphaltenes. The absorbent used in the column was silica gel, the temperature inside the reactor was 750 °C, air flow in the katharometer was 11 cm³/min, evaporator air flow was 9 L/min. The operating pressure at the first part of the reactor was 0.15 excessive atm, and at the second part of the reactor it was 0.20 excessive atm.
• Testing of samples according to the ASTM D7061 method for determining the stability and ageing of emulsions and dispersions was carried out on a MultiScan MS 20 apparatus, manufactured by DataPhysics Instruments GmbH, similar to the Turbiscan MA2000 and Turbiscan Heavy Fuel by Formulaction. Scanning was performed every 60 s at a speed of 12.5 mm/s and a resolution of 0.04 mm at a temperature of 25 °C. The separability number was determined using the software included with the device. A separability number from 0 to 5 corresponds to a high fuel stability margin; 5 to 10 means that the stability margin is much lower, but sediment is not likely to form unless the fuel is exposed to adverse conditions. If the separability number is greater than 10, the fuel is considered unstable. As an additional analysis, graphs of light transmission and backscattering of fuel were also examined as a function of tube height and examination time for 15 and 60 min, as well as after stabilizing the sample for one day at room temperature.

3. Results and Discussion

The experimental part consisted of four stages:

• Study of the physicochemical properties, hydrocarbon composition, and stability of the research objects, and prediction of their compatibility.
• Development of the model compositions of marine fuel, and analysis of their physical and chemical properties and stability. Comparing different stability inspection methods.
• Study of the properties and stability of the real composition of a marine residual fuel from a refinery.
• Comparative analysis of the performance of three fuel oil stabilizers.

3.1. Study of the Properties of the Research Objects

The Materials and Methods section of this study presents the physicochemical properties of the objects chosen for the experiments. To study the influence of hydrocarbon composition on the aggregative stability of marine fuel, the hydrocarbon-type composition of each sample was analysed using liquid adsorption chromatography, with the results presented in Figure 1.

Thus, the components presented above can be divided into three conventional groups:

• Components with a predominant content of paraffin-naphthenic hydrocarbons (more than 60%), small or zero amount of resins, and complete absence of asphaltenes: SRDF, HTDF, HGO, HCR, LGO;
• Components with a high asphaltene content: VR, VisR;
• Components with high aromatic hydrocarbon content: LCO.

Stability was assessed for the components using the assessment methods presented in the Materials and Methods section, and the results are summarized in

Table 7. Stability properties of the research objects.

Objects	Parameters
	HFT, % wt.	Total Sediment after Thermal Ageing, % wt.	Total Sediment after Chemical Ageing, % wt.	Toluene Equivalent	Xylene Equivalent	Spot Test, Spot Number
SRDF	less than 0.01	less than 0.01	less than 0.01	Absence 1	Absence 1	1
HTDF	less than 0.01	less than 0.01	less than 0.01	Absence 1	Absence 1	1
HGO	less than 0.01	0.017	less than 0.01	Absence 1	Absence 1	1
HCR	less than 0.01	0.052	0.065	4	Absence 1	2
LGO	less than 0.01	less than 0.01	less than 0.01	Absence 1	Absence 1	1
VisR	0.03	- 2	- 2	32	20/25	1
VR	0.01	- 2	- 2	7	Absence 1	1
LCO	less than 0.01	0.024	0.018	Absence 1	Absence 1	1

¹ The absence of xylene and toluene equivalents indicates that there is no ring pattern on the stain obtained during the experiment. ² Total sediment after ageing was not determined due to the limited scope of the method in terms of maximum viscosity (up to 55 mm²/s at 100 °C).

.

For the most accurate result, the stability assessment was carried out using a set of methods, the analysis of which allows us to draw conclusions for each of the components.

From the results of total sediment with ageing according to ISO 10307-2, it is observed that the components are stable as the residue value does not exceed 0.1% wt. for each component that was tested with this method. Studies using the toluene equivalent method have shown that, since the stability criterion is a value of not more than 30% vol., the visbreaking residue sample is considered unstable based on this method, while the other samples meet the requirements of the standard. According to the xylene equivalent method, the stability criterion is a value less than or equal to 25/30, so all components are stable. According to the results of the visual method of evaluation by spot, all components are stable, as the stains numbered 1 and 2 meet the criterion of no more than no. 2.

The study of samples shows that if the pure components were to be used as a marine fuel, they would be considered stable, except the VisR sample, which can be categorized as unstable by the toluene equivalent method, and this can be explained by the borderline stability of this component since the xylene equivalent test is also barely passed. However, the spot test shows excellent results for VisR. Thus, it is important to note that the results of different stability assessment methods do not always correlate with each other, which is why in this study aggregative stability is assessed in a comprehensive manner using several methods. By aggregating the full set of experiment results, not only will the stability be measured accurately but the methods can be assessed on how much they correlate with one another.

3.2. Development and Study of Properties of the Model Compositions of Residual Marine Fuels

As mentioned before, one of the options for enhancing stability is regulating the composition of the fuel, i.e., assessment of the compatibility of fuel components before blending and subsequent verification of this assessment by laboratory methods.

Compatibility should be assessed on the basis of the hydrocarbon-type group composition of the components.

To study the influence of group hydrocarbon composition, model compositions were prepared and studied (

Table 8. Structure of model compositions.

Composition	Component content, % wt.
	HTDF	HGO	VisR	LCO
1	50	50	-	-
2	90	-	10	-
3	40	-	10	50
4	60	-	10	30

and

Table 9. Results of the experimental study of the compositions.

Parameter	Regulation	Composition No.
		1	2	3	4
Kinematic viscosity at 50 °C, mm2/s	depends on the category	14.054	4.267	3.96	3.95
Density at 15 °C, kg/m3	depends on the category	886.24	852.03	908.96	886.54
Mass fraction of sulphur, % wt.	Statutory requirements, max. 0.5 for VLSFO	0.48	0.10	0.60 1	0.40
HFT, % wt.	max. 0.10	less than 0.01	0.21 1	0.06	0.14 1
Total sediment after thermal ageing, % wt.	max. 0.10	0.02	0.36 1	0.09	0.27 1
Total sediment after chemical ageing, % wt.	max. 0.10	0.01	0.23 1	0.07	0.18 1
Toluene equivalent, % vol.	max. 30	absence	40 1	30	35 1
Xylene Equivalent, % vol.	max. 25/30	absence	35/40 1	25/30	30/35 1
Spot test, spot number	max. no. 2	1	4 1	2	3 1

¹ These results do not meet the requirements of the ISO 8217 standard/IMO 2020 regulations.

).

To confirm the hypothesis that the cause of the instability of residual marine fuels is the coagulation of asphaltenes and the incompatibility of components in the composition, the composition of model composition no. 1 was selected. According to the obtained results of the study of this composition, it can be concluded that such fuel composition meets the requirements of IMO 2020 on sulphur content and is stable due to the hydrocarbon composition of the mixture, in which neither of the components contains asphaltenes and has a paraffin-naphthenic base. The total absence of asphaltenes makes it impossible to form any kind of sediment when operating in normal conditions.

However, today, due to the deepening of refining processes and striving for more rational use of the distillate fractions of oil for economic purposes, the involvement of heavy residues in marine fuel containing asphaltenes is preferable, which requires the study of model compositions that contain residual components.

Composition no. 2 consists of two components with different bases, paraffin-naphthenic in hydrotreated DF and predominantly aromatic with a high content of asphaltenes in visbreaking residue. Prior studies [66,67,68] have shown that such components show incompatibility, which is confirmed by the results of the experimental study of composition no. 2. None of the stability indices for composition no. 2 meet the requirements for stable fuels. However, the environmental qualities of such a composition are improved (sulphur content 0.10% wt.). Thus, it can be concluded that the introduction of even a small amount (10% wt.) of VisR into diesel fuel leads to the instability of the whole system.

Due to the fact that the residual components contain asphaltene substances, which are highly condensed polycyclic aromatic compounds, their best solubility is observed in an aromatic environment.

For composition no. 3, visbreaking residue was used as a residual asphaltene-containing component and LCO was used as the aromatic-based blending component. Earlier, it was noted that by adding components with a paraffinic base to the aromatic medium with dissolved asphaltenes, the latter would be able to remain in the solution. The paraffinic base component in composition no. 3 is hydrotreated diesel fuel.

According to the results of the experimental study of composition no. 3, it can be noted that the prediction of the stability of the system based on the study of the group hydrocarbon composition of the components included in its composition was correct. At the same time, the environmental properties of the composition do not meet the requirements of IMO 2020, since the sulphur content is 0.6% wt.

It is important to note that the environmental qualities of composition no. 3 do not meet the requirements due to the high sulphur content in LGO (0.998% wt.), which can be solved at the refinery, e.g., hydrotreating the vacuum gas oil before the catalytic cracking process.

Model composition no. 4 has the same components as composition no. 3, but in a different ratio: the content of the paraffin component is increased by decreasing the content of the aromatic base component and, consequently, decreasing the aromaticity of the whole system.

As can be seen from the results of the study of composition no. 4, the environmental qualities of the model of marine residual fuel improved in comparison with composition 3, as the sulphur content decreased from 0.6 to 0.4% wt., since the proportion of the hydrogenated low-sulphur component was increased. However, due to the decrease in the aromaticity of the system, the aggregative properties of the system deteriorated and did not meet the requirements of the selected methods.

Thus, the analysis of model blends has shown that the main cause of instability of marine residual fuel is the incompatibility of group hydrocarbon composition due to the addition of residual components to blends with low aromaticity, i.e., consisting mainly of naphthene-paraffin base components.

In this case, all of the selected stability assessment methods were in agreement, i.e., none of the tests showed dissimilar results regarding whether or not the sample was stable. In addition, some conclusions can be drawn from analysing the model compositions alone. The total sediment test results with and without ageing (ISO 10307-2 and ISO 10307-1 methods, respectively) increase in the following sequence: HFT < total sediment with chemical ageing < total sediment with thermal ageing. The HFT values are lower than those of chemical and thermal ageing tests, since the sample is, in fact, not artificially aged, and the asphaltene state is much more stable due to the micelle outer layer being intact. Thus, since the three methods are basically the same test with different preparation procedures, the HFT test naturally provides results with lower accuracy when applied to residual fuels’ potential stability. Total sediment with chemical ageing is supposed to be a sort of express alternative to total sediment with thermal ageing (testing time is 1 vs. 24 h, respectively) but tends to give results with lower values than the latter, while total sediment with thermal ageing is widely recognised to better represent the actual conditions of fuel storage on board and use in actual marine engines. Apparently, this is due to the fact that the use of hexadecane in chemical ageing tests does not represent the process of fuel ageing in the storage tanks as accurately as thermal ageing does. Still, both methods are now widely used in both research and bunkering applications.

3.3. Study of the Aggregative Properties of Compositions of Residual Marine Fuel from Refineries

For research purposes, a promising composition from one of the Russian refineries was studied, and the composition is shown in

Table 10. Composition no. 5.

Component	Content, % wt.
HGO	20
HTDF	50
LGO	5
VisR	25

.

The physicochemical and stability properties were investigated for this composition and are presented in

Table 11. Results of the experimental study of composition no. 5.

Parameter	ISO 8217 Regulation	Composition No. 5
Kinematic viscosity at 50 °C, mm2/s	depends on the category	48.776
Kinematic viscosity at 100 °C, mm2/s	not regulated	4.524
Density at 15 °C, kg/m3	depends on the category	892.26
Mass fraction of sulphur, % wt.	statutory requirements, max. 0.5 for VLSFO	0.48
Flashpoint, closed cup, °C	min. 60	75
HFT, % wt.	max. 0.10	unfiltered
Total sediment after thermal ageing, % wt.	max. 0.10	0.97
Total sediment after chemical ageing, % wt.	max. 0.10	0.87
Toluene equivalent, % vol.	max. 30	65
Xylene Equivalent, % vol.	max. 25/30	45/50
Spot test, spot number	max. 2	4
Separability number	not regulated	0.867

.

The research results show that the composition meets the requirements of IMO 2020 in terms of sulphur content, but is unstable, which is associated with the group hydrocarbon composition of components. From the analysis of the group hydrocarbon composition of the composition (

Table 12. Hydrocarbon-type structure of the composition no. 5.

Objects	Hydrocarbon Group Composition, % wt
	Paraffin-Naphthenes	Light Aromatics	Medium Aromatics	Heavy Aromatics	Resins	Asphaltenes
Composition no. 5	70.1	5.5	2.7	5.3	7.9	8.5

), we can draw a number of conclusions about the reasons for the instability of the studied system.

Composition no. 5 consists of five components, four of which have a predominantly naphthene-paraffin composition, and the fifth component is residual and makes up a quarter of the composition, which causes a high content of asphaltenes in the resulting fuel. Thus, the cause of instability is the incompatibility of the components.

One way to solve the instability is changing the composition of the mix. A series of tests were carried out, which made it possible to experimentally select the modified composition of composition no. 5, named composition 5–1, that meets the requirements for marine residual fuel (

Table 13. Composition no. 5–1.

Component	Content, % wt.
HGO	18
HCR	74
LGO	5
VisR	3

).

The low amount of VisR accounts for the low asphaltene content, and thus the composition does not have enough asphaltenes to form any major deposits.

The second way to solve the problem of stability of marine residual fuel is the use of stabilizers. Therefore, if the composition of the fuel is already developed and even in use, the solution can be to add a stabilizer.

For composition no. 5, a commercially available stability enhancer sample based on alkylphenol formaldehyde resin (S2) was investigated as a stabilizer.

The experimental results show that the introduction of the stabilizer in the amount of 1000 ppm (0.1% wt.) allowed an increase in the potential of the involvement of the residual fraction, i.e., visbreaking residue, up to 10% wt. (Figure 2).

Also, within the framework of this study, the effective concentration for composition no. 5 was found. After introducing the stabilizer in the amount of 2000 ppm (0.2% wt.) into composition no. 5, stability tests of the obtained marine residual fuel were carried out (

Table 14. Results of the experimental study of composition no. 5 after the introduction of the stabilizer.

Parameter	Composition no. 5	Composition no. 5 + S2 Stabilizer (2000 ppm)
HFT, % wt.	unfiltered	0.06
Total sediment after thermal ageing, % wt.	0.97	0.09
Total sediment after chemical ageing, % wt.	0.87	0.08
Toluene equivalent, % vol.	65	30
Xylene equivalent, % vol.	45/50	25/30
Spot test, spot number	4	2
Separability number	0.867	0.211

).

In the course of studying composition no. 5, apart from the previously introduced test methods, n-heptane-induced phase separation using an optical scanning device according to ASTM D7061 was also carried out. The degree of dilution in accordance with Appendix A1 of ASTM D7061 was chosen based on the kinematic viscosity of the fuel at 100 °C as a determining indicator of its classification as grade no. 5 according to ASTM D396 criteria. Since the viscosity of the fuel at this temperature is lower than 14.9 mm²/s but exceeds 24.0 mm²/s at 50 °C, such fuel can be classified as grade no. 5, for which the dilution degree according to ASTM D7061 is 1:6.

The separability number of composition no. 5 both with and without the addition of S2 stabilizer is quite low (albeit the no-stabilizer composition value is higher), making it seem as if the stabilizer-free composition is, in fact, stable with a margin. At the same time, other stability methods unanimously show the opposite, all of them exceeding the stability limit. To further explore this issue, the light transmission and backscattering graphs of composition no. 5 were studied. In order to investigate the issue more thoroughly, the analysis was performed not for 15 min, as stated in ASTM D7061, but for 1 h. The results are presented in Figure 3.

The 0 mm position on the x-axis in Figure 3 corresponds to the bottom of the test tube, and 55 mm to its top. The y-axis shows the degree of light transmittance of the fuel, where 100% means that the sample is transparent and light fully passes from the light source to the detector, and 0% means the receiving sensor does not detect light because it is completely reflected or absorbed. The colour of the line on the graph corresponds to the time when the measurement was made, gradually changing from dark blue at the first measurement (0 min) to light grey at the last (60 min). For user convenience, there is a timeline presented to the right of the graph.

The first measurement (dark blue line) shows that after intensive mixing the fuel transmits light evenly. A sharp increase in light transmittance at the position of approximately 3 mm corresponds to the transmittance of the bottom of the tube, a sharp decrease at the position of 47–48 mm corresponds to the fuel meniscus, and a constant state of light transmittance in the region of 50–55 mm corresponds to the part of the test tube above the fuel. Thus, only regions from 5 to about 45 mm should be taken into consideration.

Further measurements (lighter lines) show that the light transmittance of the fuel increases over time, and in the upper part of the test tube, this growth occurs slightly more intensely than in the lower. This fact indicates that precipitation occurs at the bottom of the test tube. The change in fuel homogeneity can be further observed in the graph of average light transmission per measurement versus measurement time presented in Figure 4.

As can be seen from the graph, light transmission change did not slow after one hour of study, hence the test was performed again after 24 h of keeping the sample at room temperature. This way it was ensured that all of the asphaltenes were precipitated and the final measurement could be taken, as presented in Figure 5.

This graph shows that the fuel has stabilized, and light transmittance has increased to about 80%. At a test tube height of approximately 4 mm, there is a complete absence of light transmittance, and after that, it sharply increases to a stable value. This fact indicates that the fuel contains an interface between two phases—one that transmits light and one that does not.

The formation of sediment can be more clearly observed in the backscattering graph presented in Figure 6. It can be seen that the degree of backscattering gradually decreases over time (dark blue lines are higher than light grey ones) throughout the entire height of the tube, except in the region up to approximately 4 mm, where it gradually increases (light grey lines are higher than dark blue ones). This growing peak indicates the formation of sediment. In the light backscattering plot after fuel stabilization for 24 h, this peak is more obvious (Figure 7).

The results of a study of light transmission and backscattering for composition no. 5 with the addition of 2000 ppm of S2 stabilizer are presented in Figure 8.

When adding a stabilizer to the composition of marine fuel, no precipitation of particles is observed, the fuel is dark and even throughout the entire volume (light transmission stays the same over the entire height of the test tube), and stable over time. No precipitation (in the form of a peak) is observed in the backscattering graph either. Although the standardized separability number assessing the fuel did not bear fruit in the form of any comprehensible results, from these light transmission and backscattering graphs it can be seen that by adding the stabilizer to the fuel, the asphaltenes do not precipitate and form a sediment.

Thus, the results of the experiment after the introduction of the stabilizer confirm that in the absence of the possibility of adjusting the composition of marine residual fuel, the introduction of a stabilizer is the only way to improve stability.

3.4. Comparative Analysis of the Effectiveness of Different Stabilizers

The effect of different stabilizers on the aggregative stability of marine fuel was compared by experimenting with composition no. 6 (

Table 15. Composition no. 6.

Component	Content, % wt.
VR	20
HCR	50
HTDF	30

).

To compare the effectiveness of different types of stabilizers, the following stability enhancers were considered: S1 and S2 (based on alkylphenol–formaldehyde resin), and S3 (based on polyisobutylene succinimide). All samples were injected in an amount of 1500 ppm. This time, the evaluation was only carried out using the total sediment after thermal ageing method according to ISO 10307-2 (

Table 16. Results of the experimental study of composition no. 6 after the introduction of stabilizers.

Parameter	Composition no. 6	Composition no. 6 + S1 (1500 ppm)	Composition no. 6 + S2 (1500 ppm)	Composition no. 6 + S3 (1500 ppm)
Total sediment after thermal ageing, % wt.	0.481	0.052	0.046	0.063

).

The results show that the alkylphenol–formaldehyde-resin-based stabilizer samples perform better than the polyisobutylene succinimide-based sample. For both types of stabilizers, the interaction between the additive and the asphaltene is supposed to be based on the π-π bond and hydrogen bonds, while in the case of polyisobutylene succinimide additive, it can also be strengthened by acid–base interactions. In general, prior studies show that the latter is known to have a better effect on asphaltene precipitation prevention. However, in this case, the opposite seems to be true, even if the results are not that far apart. We believe it is safe to say that in this situation the amount of the solvent in the S3 additive is the issue, the amount of which must be more than in the S1 and S2 stabilizers, due to the fact that concentrated polyisobutylene succinimide is a liquid with a very high viscosity that makes it inconvenient to use as is.

4. Conclusions

In this paper, the stability properties of residual marine fuels of different compositions were studied. The influence of group hydrocarbon composition of the components used was explored.

Model compositions consisting of components with different hydrocarbon-type bases were developed for the experiments. The results showed that components with a high content of paraffin-naphthenic compounds have extremely poor compatibility with components with a high content of asphaltenes. Highly aromatic components in this case can act as mediators, meaning that their addition to a binary paraffinic-asphaltene mixture in a certain concentration can help to stabilize the composition. In order to obtain a stable composition and to keep asphaltene associates in suspension, it is necessary for the fuel mixture to have a predominantly aromatic base. This way the introduction of paraffin-naphthenic base components (up to a certain extent) will only contribute to reducing the total sulphur content of the system (indirect enrichment) without impairing the stability of the entire marine fuel composition.

The use of stabilizers in different compositions showed that their addition increases the potential of involvement of residual components, sometimes even being able to completely solve the problem of aggregative stability. This study showed that adding 1000 ppm of alkylphenol–formaldehyde-resin-based stabilizer can increase the total admissible content of asphaltene-rich visbreaking residue from 3% to 10% wt. without exceeding the acceptable level of total sediment.

Different sediment potential methods were studied, and tests based on them were performed. Overall, for most of the tests carried out during this research, such methods as total sediment with ageing (both chemical and thermal), toluene equivalent, xylene equivalent, and spot test, showed a decent correlation with one another when performed on both stable and unstable compositions, and questionable correlation when performed on semi-stable fuels. The separability number, however, in itself, did not seem to conform with other methods, with the former showing more optimistic results regarding fuel stability than is considered to be veritable. However, studying the light transmission and backscattering graphs may help better understand and establish actual fuel’s stability.

Different types of fuel oil stabilizers were studied, with alkylphenol–formaldehyde-resin-based stability enhancers showing a better result than ones based on polyisobutylene succinimide; however, this result is affected by the fact that the solution of polyisobutylene succinimide additive was used.

While the research presented in this work is plentiful, a lot more can still be done to expand and make the research more thorough. Firstly, since only three stabilizers were studied (with two of them being of the same type), further exploration of different stabilizer types can expand the range of comparative analysis between them while also helping determine the recommendations of which stabilizer is best used for each kind of fuel. Secondly, the concentration range of the stabilizers can be broadened in order to study the tendency of stability change with an increase in additive dose rate. Lastly, using more fuel stability methods in the course of this experiment could both benefit the research accuracy and help find the correlations between different methods of determining fuel stability.