Non-Thermal Plasma Pyrolysis of Fuel Oil in the Liquid Phase

: A pulsed plasma pyrolysis reactor with an efﬁcient control system was designed for fuel oil processing. Non-thermal plasma pyrolysis was carried out in the liquid phase at low temperatures (not higher than 100 ◦ C) in a 300 cm 3 reactor without additional reagents or catalysts. The main process parameters and characteristics of non-thermal plasma fuel oil products were investigated within the DC source voltage range of 300–700 V. An increase in the energy of pulsed discharges led to an increase in the productivity of the plasma pyrolysis process and the yield of hydrogen but reduced the yield of acetylene and ethylene. The resulting gas consisted predominantly of hydrogen (46.5–50.0 mol%), acetylene (28.8–34.3 mol%), ethylene (7.6–8.6 mol%), methane (4.2–6.2 mol%), and C3–C5 hydrocarbons. The solid-phase products were in the form of disordered graphite and multilayer nanotubes.


Introduction
Heavy oil reserves are estimated to contain 5.5 trillion barrels, which accounts for over 70% of the world's oil reserves [1,2].Heavy oil has a density of 920-1000 kg/m 3 and extra-heavy oil has a density above 1000 kg/m 3 .The sulfur content in these oils reaches up to 7% [3,4].The low content of light fractions and hydrogen to carbon ratio, as well as high viscosity, a large number of heterocompounds, asphaltenes, resins, and highmolecular-weight paraffins also complicate the processing of heavy oils via conventional methods [5][6][7][8].Current processing methods for heavy hydrocarbon feedstocks (visbreaking, delayed coking, hydro pyrolysis, and catalytic cracking) have several disadvantages: high temperature (400-600 • C) and pressure (up to 20 MPa) requirements, the need for hydrogen, and high costs for equipment and coke removal [9][10][11][12][13].Heating the raw materials and maintaining high temperatures during the process requires burning a large volume of hydrocarbon fuel with high CO 2 emissions [14][15][16].Replacing high-temperature reactors and furnaces with plasma reactors using carbon-free electricity will significantly reduce CO 2 emissions [17][18][19].Plasma technologies for refining heavy oils reduce the use of hydrogen and also exclude the use of expensive catalysts [20][21][22][23].
The action of plasma on hydrocarbons leads to primary reactions such as excitation, dissociation, and ionization.Chemically active particles resulting from these reactions (radicals, ions, and excited particles) undergo rapid recombination to form products at the relaxation stage.A distinction is made between thermal plasma (TP) and non-thermal plasma (NTP) according to particle energy and plasma temperature [24].TP requires tens of kilowatts of power at a plasma temperature of ~10,000 • C, which helps increase the yield of gas phase products such as hydrogen, acetylene, and other gaseous hydrocarbons [25,26].However, a significant portion of the input power for TP generation is lost through heating the system.In contrast, NTP requires relatively little power and does not cause an obvious increase in system temperature.
The generation of highly reactive species, such as radicals and other active particles, is a significant advantage of NTP for chemical processes.These species can selectively activate specific chemical bonds or functional groups in a molecule, leading to specific reaction pathways that may not be accessible through conventional chemical methods [27,28].The utilization of NTP for chemical processes enables strongly endothermic reactions at moderate temperatures, limiting joule heating effects and thereby minimizing energy loss through heating the reactant mass [29][30][31].This attribute renders plasma processes energetically efficient and promising for thermocatalytic reactions, which are customarily executed at elevated temperatures and pressures.
While examples of large-scale plasma applications have been demonstrated, problems hinder widespread introduction into the chemical industry [32][33][34].These relate mainly to scaling up laboratory plasma processes, which requires solving technical challenges in designing an entire high-power plasma system including a reactor and power supply.Solving these problems is required to launch large-scale plasma processes.
The transition to a low-carbon economy has led to the intensification of the development of plasma-chemical technologies for hydrocarbon processing [35].Plasma-chemical pyrolysis makes it possible to produce turquoise hydrogen together with olefins and carbon materials [36][37][38].Plasma-chemical technologies are also being developed for waste recycling [39][40][41] and for the decomposition of CO 2 [42,43].
Previous studies [44,45] show the possibility of using NTP pyrolysis for processing heavy oil fractions and thermostable toxic organochlorine substances.A 40 cm 3 reactor with pulsed discharges in the liquid phase generated by one automatically moving electrode was used.Pulsed discharges in the liquid phase were generated by the automatic movement of an electrode according to a predetermined algorithm using the control system [46,47].The developed principle for generating electric discharges, control system, and plasma reactor are effective for inducing chemical processes.
This paper describes a 300 cm 3 plasma pyrolysis unit with an efficient control system for generating electric discharges in the liquid phase.The modified unit is used to study the NTP pyrolysis of fuel oil using 300-700 V DC electric discharges.The results will allow the creation of small modular plasma-chemical reactors with high productivity and process selectivity.

Experimental Setup
The plasma pyrolysis unit (Figure 1) includes a reactor, a system for controlling and recording electric discharge parameters, and an off-gas collection system.The reactor is made of carbon steel.The reactor volume is 300 cm 3 .Graphite electrodes are inside the reactor.An EA-PSI 9750-06 2U DC power supply up to 750 V is used to generate electric discharges.Fuel oil is poured between graphite electrodes.Before starting, helium was pumped through the entire installation to create an inert atmosphere and prevent possible ignition of the resulting gases (hydrogen, acetylene, and others) inside the reactor.Plasmachemical pyrolysis is carried out at a slight excess pressure of 1-2 atm.Gaseous products from the reactor pass through a seamless gas sampler and are then burned on a Bunsen burner.The pyrolysis process is carried out until the gas formation stops.Low-voltage discharges are generated by a semiconductor switch.Discharge frequency, duration, number, and operating sections are set by the control system.The adjustable power supply (DC VS) sets the voltage at which discharges occur in a multisection system.The microprocessor control system (MCS) measures the voltage value at the DC VS output using a voltage sensor (VS) and compares it with the set value.When the voltage reaches the set value with the measured one, it generates control pulses for the transistor VT1.The number of discharges is counted by the MCS using the current sensor (CS).
During the process of NTP pyrolysis, the oscilloscope was employed to record the instantaneous values of current and voltage at a 2 min interval (600,000 measured values).MATLAB was employed to process the recorded values, obtain time-dependent representations of transients (discharges), and determine current pulse parameters.This analysis enabled the determination of current pulse parameters, including the number of pulses within each 2 min interval, the duration of each pulse, the total discharge exposure time, and the amplitude (maximum current) of each pulse.The average values (Table 1) were then calculated using the method described in [47].Typical oscillograms of current and voltage pulses of plasma-chemical pyrolysis of fuel oil at 300-700 V are shown in Figure 2.

Sample Analysis and Characterization
The raw material and liquid NTP pyrolysis products of fuel oil were analyzed for several characteristics.The fractional composition was determined via ASTM D7169 using a Chromatec-Crystall 5000.2 gas chromatograph equipped with a flame ionization detector (JSC SDO Chromatec, Yoshkar-Ola, Russian Federation).The kinematic viscosity at 100 • C was determined according to ASTM D7042.The density at 20 • C was measured using an Anton Paar SVM 3000 Stabinger viscometer-densitometer (Anton Paar GmbH, Graz, Austria), following ASTM D4052.The sulfur content was analyzed according to ASTM D4294, using a Lab-X 3500 X-ray fluorescence analyzer.IR spectra were measured using an FSM-1202 Fourier-transform infrared spectrometer.Gas product composition was determined using a Chromatec-Crystall 5000.2 chromatograph (JSC SDO Chromatec, Yoshkar-Ola, Russian Federation).
NMR spectra were recorded on a Bruker Avance III 400 spectrometer at 400 MHz for 1 H NMR and 100 MHz for 13 C NMR using residual CHCl 3 as an internal reference.
1 H and 13 C NMR spectra were acquired on a Bruker Avance III spectrometer.Fuel oil samples were dissolved in CDCl 3 solutions following the ASTM D5292 method, with a 5% volume concentration.The signals of chloroform-d (δ C = 77.0)and residual CHCl 3 in the solvent (δ H = 7.26 ppm) were used as an internal standard.To enhance the sensitivity of the analysis and reduce noise, the solid particles were removed from the samples via filtering through a 0.22 µm pore size fluoroplate filter.The aromatic hydrogen and carbon coefficients were calculated using the following equations: F HA = H ar /(H ar + H al ) and F CA = C ar /(C ar + C al ), where H ar and C ar are the total integrals of aromatic hydrogen and aromatic carbon atoms, respectively, and H al and C al are the total integrals of aliphatic hydrogen and carbon atoms, respectively [48].
The still bottoms obtained after NTP pyrolysis of fuel oil were extracted with heptane, and the solid phase was filtered using a paper filter.The solid products were then dried in a muffle furnace at 110 • C for one hour.
Solid product analysis was conducted using various analytical techniques, including scanning electron microscopy (SEM) together with energy-dispersive X-ray spectroscopy (EDXS), transmission electron microscopy (TEM), scanning transmission electron microscopy (STEM), electron diffraction (ED), energy-dispersive X-ray (EDX) microanalysis, X-ray phase analysis (XRD), X-ray fluorescence analysis, and Raman spectroscopy.SEM/EDX studies were performed in a Supra 50 VP electron microscope (Carl Zeiss AG, Jena, Germany) with the INCA microanalysis system (Oxford Instruments, Abingdon-on-Thames, UK).The SEM images were obtained in the secondary electron (SE) registration mode.
Samples for TEM/STEM/EDXS studies were prepared through depositing sample particles on an electron microscopic copper grid with a Lacey carbon film.Samples were examined using an Osiris TEM/STEM (Thermo Fisher Scientific, Waltham, MA, USA) equipped with a high-angle annular dark-field detector (HAADF) (Fischione, Export, PA, USA) and a Super-X EDXS (Bruker, Billerica, MA, USA) at 200 kV accelerating voltage.
X-ray fluorescence analysis was performed using the Orbis (EDAX, Mahwah, NJ, USA).The accelerating voltage in the X-ray tube was 40 kV, the current was 100 µA, and the diameter of the X-ray beam was 2 mm.The studies were conducted without air pumping.Such conditions make it possible to register elements starting with sodium.
Raman (combination scattering) spectra of the samples were measured using a spectrometer consisting of a monochromator (Solar Laser Systems M266, Minsk, Belarus) with a CCD detector (charge-coupled device, U2C-16H10426, Japan).A 532 nm reflector (532 nm StopLine ® single-notch filter, Semrock, Rochester, NY, USA) was also used in the setup.This filter reflects laser beams at the specified frequency, allowing only scattered light to pass to the spectrometer.Source imaging and laser beam alignment were performed using an optical system consisting of a video camera (The Imaging Source DFK 22AUC03, 644 × 484 pixels, Germany), a light source, a lens (50×, Plan Apo, ULWD, NA = 0.42, WD = 22.5 mm, Edmund Optics, Burlington, NJ, USA), two lenses, and two light beam dividers.Measurements of the Raman spectra were performed at an exposure time of 200 s, and the laser irradiation power on the sample was 0.47 mW at a laser wavelength of 532 nm.
The ABK-1B calorimeter was used, which belongs to variable-temperature calorimeters, in which the amount of heat is determined by the change in the temperature of the calorimeter vessel to determine the energy of combustion of the solid product.

Characteristics of NTP Pyrolysis and Analysis of Gaseous Products
The study of the effect of voltage on the phase and component composition of NTP pyrolysis products of fuel oil was carried out at voltages of 300-700 V. Table 2 presents the experimental results of the effect of stress on NTP pyrolysis of fuel oil.The main substances in the gas phase are hydrogen (46.5-50.0mol%), acetylene (27.8-34.4mol%), ethylene (7.6-8.3 mol%), and methane (4.2-6.2 mol%).
When conducting the NTP pyrolysis process at higher voltages, the energy consumption for the formation of gaseous products is significantly reduced from 39.7 to 8.3-11.7 kWh/kg (Table 2) as the process speed and gas flow rate increase from 7 to 555-709 mL/min.This effect can be attributed to an increase in the specific energy density and the number of radical and active particles, which result from an increase in the average pulse energy from 1.5 to 4.6 J (Table 1).The highest productivity of NTP pyrolysis is observed at 700 V; however, at a voltage of 500 V, the lowest energy consumption (3.9 kWh/kg of fuel oil), the highest yield of valuable gaseous hydrocarbons (46.5 wt%), and a high content of acetylene (30.4 mol%) are observed.

Still Bottoms Characteristics from the NTP Pyrolysis of Fuel Oil
The fractional composition (Figure 3) of the initial fuel oil and still bottoms after NTP pyrolysis at 300-700 V was determined using simulated distillation (ASTM D7169).The results showed that the NTP pyrolysis process led to the production of heavier fuel oil (Table 3 and Figure 3) with increased stress and degree of conversion compared to the initial fuel oil.Additionally, the NTP pyrolysis process increased the boiling point of the products while reducing the proportion of volatile components from 69% to 43-44%.Table 3 presents the results of the analysis of fuel oil and still bottoms obtained from NTP pyrolysis of fuel oil.An increase in the density and viscosity of the pyrolysis products was observed.However, it was not possible to measure the viscosity and density of the pyrolysis still bottoms accurately at 500 and 700 V due to their high viscosity, solids content, and stickiness.The high content of solid structures (33.6-33.9wt%) leads to an increase in the electrical conductivity of the fuel oil suspension and limits the conversion to 46.2-49.0wt% (Table 2).The fuel oil suspension exhibited low resistance and started to conduct electric current, which resulted in the heating of the reaction medium.The increase in the carbon-hydrogen ratio in the still bottoms agreed with the composition of the obtained gaseous products (Table 2).As the stress increased in the NTP pyrolysis of fuel oil, the hydrogen content increased from 46.5 to 50.0 mol%, while the acetylene content decreased from 34.3 to 27.8 mol%.
Figure 4a-c show photos of still bottoms and solid products (d-f) isolated from still bottoms after plasma-chemical pyrolysis of fuel oil.The heat of combustion of the solid product (Figure 4f) is 36,439.6± 187.2 kJ/kg.

IR and NMR Spectroscopy of Still Bottoms of the NTP Pyrolysis of Fuel Oil
Figure 5 shows the IR spectra of fuel oil and still bottoms obtained from NTP pyrolysis of fuel oil.The FTIR analysis results (Figure 5) are in agreement with the 1 H NMR analysis.The IR spectra revealed an increase in the content of aromatic (1380 cm −1 ) and polyaromatic hydrocarbons (737 cm −1 ), which may be attributed to the dealkylation of aromatic compounds [49,50].The ratio of -CH 3 and -CH 2 -groups in aliphatic hydrocarbons in the range of 2850-3970 cm −1 showed little to no change.
In the 1 H NMR spectra, the signals observed in the ranges of 7.24-6.5,8.3-7.3, and 9.0-8.3ppm (Table 4) were assigned to monoaromatic CH bonds, diaromatic CH, and tri(and more) aromatic CH, respectively [51,52].The results of NMR analysis showed that the plasma pyrolysis process affects both aliphatic and aromatic hydrocarbons (Table 4).As the fuel oil conversion increased from 28.3 to 49.0 wt%, the proportion of protons in aromatic hydrocarbons (H ar ) increased significantly compared to aliphatic hydrocarbons (H al ), leading to an increase in the F HA value from 0.0499 to 0.0697.This observation suggests that the alkyl hydrocarbon decomposition process intensified along with the ring condensation process at deeper stages of the NTP pyrolysis of fuel oil.The destruction of single-ring aromatic hydrocarbons (7.24-6.5 ppm) occurred initially, resulting in a decrease in their content from 2.55 to 2.18%.Conversely, triaromatic hydrocarbons (9.0-8.3 ppm) increased significantly from 0.19 to 0.57%, while diaromatic hydrocarbons (8.3-7.3 ppm) increased from 2.26 to 4.23%.In the 13 C NMR spectrum (Figure 6), the signals in the 115-155 ppm region correspond to aromatic carbon atoms, which belong mainly to condensed molecules.Signals of carbon atoms belonging to aliphatic fragments appear in the 5-60 ppm range.The fraction of condensed aromatic molecules in the NTP pyrolysis cube residues increases with increasing stress (Table 5).Signals 13 C NMR in the 115-155 ppm range refer to aromatic carbon (Figure 7a).In the 13 C NMR spectrum DEPT-135 (Figure 7b), the signals of aromatic hydrocarbons coupled with protons are presented, so there are no signals in the 135-155 ppm region related to condensed structures.The NTP pyrolysis of fuel oil results in the formation of condensed aromatic rings, which significantly reduces the proportion of volatile components and increases the boiling point of the products (Figure 3).The excitation of condensed aromatic hydrocarbons requires more energy and results in a higher hydrogen yield [53], leading to increased energy costs for gas production (Table 2).However, the involvement of condensed aromatic hydrocarbons in the pyrolysis process can also result in coking.This occurs through a consecutive scheme involving a series of consolidation monomers and intermediates formed during condensation, polymerization, dehydrocyclization, binding of aromatic rings, and hydrogen depletion processes, ultimately leading to the formation of carbon structures [54].These findings have implications for the optimization of the NTP pyrolysis process for the production of value-added products with reduced coking and energy costs.

Transmission Electron Microscopy and Microanalysis
The solid products of the NTP pyrolysis of fuel oil are morphologically similar for the whole range of the power supply, namely 300-700 V.A typical low-magnification brightfield TEM image of a solid product of plasma-chemical pyrolysis of fuel oil at 700 V of demonstrates a particle conglomerate (Figure 8).The conglomerate consists of round-shape particles stacked together.The size of the particles ranges between one and tens of µm.The presence of an amorphous/glassy phase was confirmed via HR TEM study of the particles in the conglomerates, and one example is presented in Figure 10.Close inspection of HR TEM images did not reveal any traces of atomic ordering.However, particles of other types, namely, graphite flakes and multilayer carbon nanotubes (MCN), were also found in the solid product of fuel oil pyrolysis, and TEM images of an MCN are shown in Figure 11a-c.Interestingly, the thickness of the wall is not uniform: the left wall (Figure 11b) is at least twice as thick as the right one (Figure 11c).Inside the tube, there are carbon flakes (see Figure 11a).The EDX microanalysis was performed on a sample of the solid product of fuel oil NTP pyrolysis.The EDXS spectra obtained from the group of particles (Figure 12) reveal the presence of C, S, O, Cl, Na, K, and traces of Ca and Fe.The results of the quantitative EDX microanalysis of the elemental composition of the solid product obtained from the area marked in the compositional map (Figure 13) by a black square are presented in Table 6.The high oxygen content (1.1-1.4 wt%) in the solid product may indicate the presence of oxidized asphaltenes, which can be used as economically advantageous catalysts [55,56].
In some conglomerates, the nanocrystals were observed to exhibit cubic morphology (Figure 14).Surprisingly, the SAED pattern (see Figure 14 inset), together with EDX mapping (not shown here), revealed the mixture of NaCl and KCl nanocrystals.NaCl and KCl are used together with calcium compounds in drilling fluids.During oil rectification, these compounds are concentrated in fuel oil and then remain in solid products after pyrolysis.

X-ray Fluorescence Analysis
The X-ray fluorescence analysis was performed from an area with a diameter of about 10 mm (see one example as the inset in Figure 15.This method allowed us to obtain composition information from a significantly larger sample volume compared to EDX microanalysis.The X-ray fluorescence spectra are shown in Figure 15, and the quantitative results are presented in Table 7.These results were quite similar (within 5% accuracy) to other areas of the sample.The extremely high content of impurities in the solid product was due to the absence of C, O, and N in the spectra, which are not registered during X-ray fluorescence analysis.The high content of nickel and vanadium is associated with the concentration of the initial heavy oil [57].

X-ray Phase Analysis
A diffractogram of the solid product of the NTP pyrolysis of fuel oil at 700 V is shown in Figure 16.The peaks from the graphite (2H graphite Space Group #194 P6 3 /mmc, a = 0.2464 nm, c = 0.6711 nm) [58] are marked by asterisks.There is a good match of the most intense peak, which corresponds to the 0002 reflection of the graphite (at 2θ = 26.382• ), but there is a slight discrepancy between the experimental peak and the 0004 reflection of cited graphite.That discrepancy could be explained by the high density of defects in the material.The reflection marked by the cross at 2θ = 21.342• (see Figure 15) matches with sodium stearate's (C 18 H 35 NaO 2 PDF-2 00-001-0418) most intense peak.The estimation of the average size of graphite crystallites according to the Scherer formula for the 0002 main peak with a half-width of 0.2 • corresponds to ~50 nm.At the same time, this character of the spectrum can be associated with the superposition of reflections from C 5 H 10 O 2 , C 29 H 60 and C 14 H 28 compounds as well as asphaltene [59,60] compounds.

Raman Spectroscopy
The Raman spectrum (Figure 17) of the solid product of the NTP pyrolysis of fuel oil at 700 V has broad peaks with the maximum positions at 1350-1360 cm −1 and 1580-1583 cm −1 .Peaks with maxima around 1350 cm −1 are characteristic of disordered graphite and are denoted by the letter D (disordered).Peaks with a maximum of 1580 cm −1 are denoted by the letter G (graphitic), and their location refers to the bond length in sp 2 -coordinated carbon [61].The ratio of peak intensities, I D /I G , is directly related to the size of small (on the order of several nanometers) sp 2 -linked clusters of ordered aromatic rings in carbon films.The proportion of sp 3 links is inversely proportional to the I D /I G ratio.The area of the G peak decreases as the sp 3 bonds increase for samples 20-30 I D /I G ~1.2.Thus, the Raman spectra indicate that the solid product samples are disordered graphite.

Conclusions
NTP pyrolysis of fuel oil was carried out via the application of electric discharge in the liquid phase at a voltage of 300-700 V at the DC source.Increasing the power of energy exposure leads to higher productivity, energy efficiency of the process, and gaseous product yields, and also affects the composition of the gaseous products of NTP pyrolysis.The gaseous products of the NTP pyrolysis of fuel oil include substances that are widely used in the chemical industry: hydrogen (46.5-50.0mol%), acetylene (28.8-34.3mol%), ethylene (7.6-8.6 mol%), methane (4.2-6.2 mol%), and hydrocarbons C3-C5.
NTP pyrolysis at a voltage of 500 V at the DC source is optimal because the energy consumption is the lowest (3.9 kWh/kg of fuel oil), yield of valuable gaseous products is the highest (46.5 wt%), and there is low methane content (4.2 mol%) and high acetylene content (30.4 mol%) in the gas flow (555.5 mL/min).
The pyrolysis process accumulates polyaromatic hydrocarbons, which are further transformed into carbon structures.The yield of solid-phase products is 53.5-70.1 wt%.The solid-phase products are disordered graphite and multilayer nanotubes.In the elemental composition of solid products, S, O, V, and Ni were determined, which indicates the possible use of these carbon structures as catalysts.
Further work will involve improving the energy efficiency of the process and studying carbon structures for commercial applications.

Figure 4 .
Figure 4. Photos of still bottoms (a-c) and solid products (d-f) of plasma-chemical pyrolysis of fuel oil at 300, 500, and 700 V, respectively.

Figure 5 .
Figure 5. IR spectra of fuel oil and still bottoms after NTP pyrolysis at 300-700 V.

Figure 6 .
Figure 6.NMR 13 C spectrum of the NTP pyrolysis residue of fuel oil at 300 V.

Figure 7 .
Figure 7. NMR 13 C spectrum of the still bottoms of NTP pyrolysis of fuel oil at 300 V. Region of aromatic groups (a) and only CH-carbons of aromatic groups (b).

Figure 8 .
Figure 8. Bright-field TEM image of a solid product.A selected-area electron diffraction (SAED) pattern of the solid product of NTP pyrolysis (obtained at 700 V DC) and the histogram of radially averaged SAED are shown in Figure 9a,b, respectively.The SAED and the histogram indicate the glassy and/or nanocrystalline nature of the solid product of fuel oil pyrolysis.Several diffuse rings and corresponding maxima on the histogram are close to the 002, 100/101, and 102/004 hexagonal graphite interplanar distances.

Figure 9 .
Figure 9. SAED of the solid product of NTP pyrolysis (a) and the histogram of radially averaged SAED (b).

Figure 11 .
Figure 11.TEM images of an MCN: low-magnification image-(a), carbon flakes inside the wall are arrowed.The HR TEM images of the "left" wall-(b), and the "right" wall-(c).

Figure 12 .
Figure 12.The EDX spectra from the group of particles.After the inspection of the EDX spectra, elemental mapping was carried out, and the maps are shown in Figure13.The maps demonstrate the uniform distribution of C, S, and O, and the compositional map confirms that Na and K form precipitates of NaCl and KCl.

Figure 13 .
Figure 13.Dark-field HAADF STEM image; elemental maps of C, S, O, Cl, Na; and compositional map.The scale bar is 3 µm on all maps.The Cu grid is visible in the right upper corner.The EDX microanalysis spectrum quantification was performed from the area marked by a black square in the compositional map.

Figure 15 .
Figure 15.X-ray fluorescence analysis areas-optical images.X-ray fluorescence spectra of the corresponding areas.

Figure 16 .
Figure 16.X-ray diffractogram of the solid product of fuel oil pyrolysis.Asterisks indicate the hexagonal graphite reflections and the cross indicates Sodium stearate.

Figure 17 .
Figure 17.Raman spectrum of the solid product of fuel oil pyrolysis at 700 V.

Table 1 .
Characteristics of electric discharges during plasma-chemical pyrolysis of fuel oil.

Table 2 .
Characteristics of NTP pyrolysis of fuel oil and composition of gaseous products at 300-700 V.

Table 3 .
Characteristics of fuel oil and still bottoms of NTP pyrolysis of fuel oil.

Table 4 .
1H NMR analysis of fuel oil before pyrolysis and NTP pyrolysis still bottoms *.

Table 5 .
NMR13C analysis of fuel oil before pyrolysis and NTP pyrolysis cube residues *.
* Product solutions in CDCl 3 filtered through a membrane filter.

Table 6 .
Elemental composition of the solid product of the NTP pyrolysis of fuel oil.

Table 7 .
Results of the X-ray fluorescence analysis of areas.