An Experimental Analysis of the Influence of Pyrolytic Oil on the Spray Breakup Process

Abstract

Solid waste presents a very large problem in the developed world. Waste plastics, which make up a large part of solid waste, have high energy value, which is discarded if they are not treated properly. Most of the plastic found in solid waste is produced from petrochemical material, so it can be used in resource recovery processes to produce various materials. One promising resource recovery process is the pyrolysis process, from which pyrolytic oil, gas, and solid residue are obtained. Pyrolytic oils have properties that are similar to conventional fossil fuels, and are promising fuels for use in heat engines or heating applications. In the present work, HDPE plastic in the form of plastic bottles caps was collected from solid waste and used in a thermal pyrolysis process for the production of pyrolytic oil. The obtained oil was characterised, and the obtained results were compared to conventional fuels. The obtained oil was used further in an oil burner fuel injection application, in which the spray breakup characteristics were monitored and analysed using VisiSize particle characterisation systems. The obtained results were compared to those of conventional fuel. The results indicate that the difference in fuel properties influences the spray breakup process slightly, but the differences are rather small. This indicates that from a spray development perspective, pyrolytic oil can be used as a substitute for conventional fuels in oil burners.

1. Introduction

Plastic has become one of the key materials used in modern society. To achieve sustainable development, it is necessary to minimise the negative impacts of plastics while maintaining the positive effects by considering their entire life cycle [1,2]. Recently, one of the biggest challenges has been pollution by waste plastic or plastic residual, as it has a significantly negative impact on ecosystems, the health of living beings and climate change [3]. According to data from the trade association Plastic Europe, plastic production already amounted to 413.8 million tons in 2023 [4,5]. Today, plastic waste represents 10–12% of all waste produced worldwide [6].

Plastic is derived from fossil fuels and petrochemicals, such as naphtha, coal, and natural gas, through processes like polymerisation and polycondensation [7]. According to Walker [8], plastic production accounts for 6% of global oil consumption and other non-renewable resources. Engelbeen [9,10] estimated that, for the production of 1 kg of PVC plastic, it takes approximately 53 MJ of energy, PE plastic requires 70 MJ, PP requires 73 MJ, and PS plastic requires 80 MJ. However, Marczak [11] pointed out in his research that the value of the energy consumed depends primarily on the type of raw material, the final form of the polymers, the source of process energy, and the efficiency of the processing procedures. Zappitelli et al. [12] estimated that, in 2018, plastic production, which includes plastic resin production, raw material processing, and energy processes, generated 350 million tons of greenhouse gas emissions, 8.9 EJ of cumulative fossil fuel consumption, and 95 million tons of embedded carbon that remains stored in the plastic itself until it is destroyed or decomposed. Due to the high embodied energy content of plastic waste, disposing of plastics in landfills represents a significant loss of recoverable energy and resources.

One of the innovative alternatives for waste plastic treatment is pyrolysis, which is a process of thermal decomposition of material in an inert atmosphere without oxygen, and which produces products with high energy value. The main products that are produced are solid residue, oil, and combustible gases [13]. Pyrolysis is a process that represents a promising solution for reducing the growing problem of plastic in solid waste and in the environment. It is also a highly flexible process, as it ensures the optimisation of product yield by adjusting the process parameters [14,15]. The most promising product of thermal pyrolysis of waste plastics is pyrolytic oil, which can present an alternative fuel for many applications.

With increasingly stringent emission regulations, the atomisation process has recently become a central topic of research [16,17,18,19]. The efficiency of fuel atomisation is influenced most by fuel physical properties, such as density, viscosity, and surface tension. A high degree of fuel atomisation affects the combustion efficiency and the overall performance of the thermal system significantly, while also having a major impact on emission formation. Khan et al. [20] characterised pyrolytic oil from waste HDPE plastic and compared the physical–chemical properties using conventional fuels such as diesel, kerosene, gasoline, etc. The viscosity of pyrolytic oil obtained at 40 °C was 1.98 mm²/s. It was found that the viscosity of HDPE pyrolytic oil was lower than that of diesel fuel, but higher than that of kerosene. In terms of density, it was found that the density of pyrolytic oil was comparable to the density of kerosene, diesel fuel, and gas oil. In terms of caloric value, they found that the caloric value was slightly lower, but still comparable to diesel fuel and kerosene. Jahirul et al. [21] also investigated the physical–chemical properties of pyrolytic oil obtained from HDPE, PS, and PP plastic. First, they compared the properties of crude oils with diesel fuel. They found that, in terms of higher calorific value (HCV), HDPE and PP oils reached values comparable to diesel fuel (42–46 MJ/kg), while the calorific value of PS oil was slightly lower, at38.1 MJ/kg. In terms of kinematic viscosity and density, it was found that HDPE pyrolytic oil had a much higher kinematic viscosity than diesel fuel, while PS and PP oil had slightly lower viscosity. In terms of density, the values were comparable. By distilling pyrolytic oil, they achieved satisfactory properties that meet the minimum requirements of the automotive standards. Through distillation, they managed to increase the HCV slightly and reduce the kinematic viscosity slightly. Palomar-Torres et al. [22] compared the elemental composition, density, and higher heating value of HPDE pyrolytic oils obtained by thermal and catalytic pyrolysis with the properties of heavy fuel oils (HFO), which are used traditionally to power gas turbine power plants. They found that HDPE pyrolytic oil obtained from both types of pyrolysis has advantages over HFO, such as higher HCV, much lower density, and lower sulphur content.

In terms of the use of waste plastic pyrolytic oil, there is already quite a bit of research available, which mainly examines combustion efficiency, emission generation, and the performance efficiency of heat engines and devices. Most often, the research has evaluated diesel engine performance [23,24,25,26,27], with less attention given to spark ignition engine performance [28,29], gas turbine performance [30,31,32], and the performance of other thermal devices, such as heater combustion systems [33]. Additionally, very little research can be found on the efficiency of the spray and atomisation of pyrolytic oil. Given that waste plastic pyrolytic oils represent an alternative type of fuel, it is extremely important to determine what physical properties these oils have and how these properties affect the efficiency of oil atomisation. Szwaja et al. [34] investigated the spray characteristics of distilled pyrolytic oil derived from HDPE plastic waste experimentally using a high-pressure direct injection system. The main spray parameter, the Sauter Mean Diameter (SMD), was compared to 91-octane gasoline. The results showed that the distilled HDPE pyrolytic oil had similar spray dynamics under the same conditions, with comparable SMD values ranging from 7 to 9 μm. The study concluded that this oil can be used both as an additive and as a standalone fuel in internal combustion engines. Malode et al. [35] conducted pyrolysis of waste polypropylene (PP) face masks to produce pyrolytic oil. They performed thermogravimetric analysis, differential scanning calorimetry, FT-IR, and proximate and ultimate analyses. Spray experiments were conducted using a custom 3D-printed air-assisted atomiser with a mixture of PP pyrolytic oil, diesel, and two surfactants. The mixture had lower density, but higher kinematic and dynamic viscosity and a low calorific value. High-speed imaging at 10,000 fps was used to analyse spray characteristics such as jet breakup length, spray cone angle, critical wavelength, and breakup frequency. The results showed that higher air injection velocity reduced jet the breakup length and increased the spray cone angle, while the breakup frequency decreased.

The presented work studies the influence of differences in pyrolytic oil (PO) properties on the primary and secondary spray breakup processes in an oil burner nozzle. In the first step, pyrolytic oil was obtained from waste high-density polyethylene (HDPE) plastic using the thermal pyrolysis process in a batch pyrolytic reactor. Plastic bottle caps were used as a source of HDPE plastic, since they present a big problem for the environment. Due to their colour and size, they are very attractive for animals, which can swallow them easily, and they are harder to collect compared to bigger plastic parts. In the second step, the properties of the obtained pyrolytic oil were determined and compared to heating oil (ELKO), corresponding to the SIST 1011 Standard [36]. In the third step, the obtained pyrolytic oil, without additional treatment, was used further in an oil burner fuel injection application in which the spray breakup characteristics were monitored and analysed. The used test method allowed for monitoring the whole spray area, which can give more detailed information regarding the spray properties compared to monitoring only specific points of the spray jet. The results indicate that the differences in the oil properties influenced the spray breakup process slightly, but the differences were rather small.

2. Materials and Methods

2.1. Plastic Material

In the presented work, we used HDPE plastic in the form of plastic bottle caps, which were collected from municipal waste. The caps were checked further, and the caps with an additional layer of plastic on the inner side were removed.

2.2. Pyrolysis Process

The pyrolytic oil was obtained in a single-batch pyrolytic reactor with a fixed bed. At the start of the process, the whole system was flushed with nitrogen (N₂) gas to remove all the remaining air from the system. The system was heated using a Leister Mistral 6 System (Leister Technologies AG Kaegiswil, Sarnen, Switzerland) electrical heater with 4500 W of power. K-type thermocouples were used, along with a National Instruments DAQ USB-6255 with an SCC-68 terminal block (NI, Emerson, Austin, TX, USA). The program was written in LabVIEW 2020, and enabled temperature measurements, heater (temperature and air flow) control via AO signals, and logging of the process parameters. The pyrolysis process is presented schematically in Figure 1.

First, 100 g of plastic was inserted into the reactor and sealed in each experiment. The temperature inside the reactor was raised from ambient to a maximal temperature of 420 °C. The temperature inside the reactor was regulated by setting the heater air flow and air temperature using our DAQ system. The air flow and temperature were regulated based on the temperature measured inside the reactor and the desired temperature for the pyrolytic process. The vapours of melted plastic were driven through the condenser, which was cooled with water. In the condenser, some of the plastic vapours were condensed into PO, which were collected in a recovery tank. The non-condensable gases were not collected for further analyses. In total, three experiments were performed, and the oils obtained from all the experiments were mixed for further testing. The results of the pyrolytic process product yields are presented in Figure 2, in which error bars of the standard deviation for the product rates of the three experiments are shown. As can be seen, the error bar interval for oil was very small, while those for the solid residue and gas were slightly larger. This trend was expected since the raw material used for pyrolysis process was HDPE plastic waste, which does not always have consistent composition, since different bottle caps may contain varying amounts of fillers and additives.

The average yields of the pyrolysis process products from all three experiments are presented in

Table 1. Average product yield ratios from the pyrolytic process.

Product
Oil [%]	76.0
Solid residue [%]	12.6
Gas [%]	11.4

.

The pyrolytic oil and solid residue yields were determined by weighing the mass of each product obtained in each experiment. The total mass of each product from all the experiments was divided further by the total mass of plastic used. The gas phase yield was determined from the difference between the total mass of plastic used in all the experiments and the mass of all the obtained pyrolytic oil and solid residue.

The presented properties of the pyrolytic oil present the average property for the oil from all three experiments. The obtained pyrolytic oil was filtered before use using MN 615 filter paper with a retention range of 4–12 µm. This prevents clogging in the built-in filter in the used nozzle.

2.3. Oil Characterisation

The properties of the obtained pyrolytic oil were tested further (characterised) and compared to the properties of heating oil. Anton Paar’s DMA 35 density meter (Anton Paar, Graz, Austria) with an accuracy of 0.001 g/cm³ and repeatability of 0.0005 g/cm³ according to the ISO 5725 Standard [37] was used for density determination. The oil’s viscosity was determined using a FUNGILAB VISCOLEAD ONE rotational viscometer (Obsnap Group, Selangor, Malaysia) with ±1% full-scale precision, and oil surface tensing using a Krüss K12 tensiometer (KRUSS Scientific, Hamburg, Germany) with an accuracy of ±0.15 mN/m and resolution of 0.05 µm. The calorific value was determined using an IKA C200 bombe calorimeter (IKA-WERKE GmbH&Co. KG, Staufen, Germany), and the FT-IR spectra were recorded using a Perkin Elmer Spectrum GX (PerkinElmer, Hopkinton, MA, USA). The chemical composition of the oils was determined using a TruSpec Micro analyser from LECO (LECO Corporation, St. Joseph, MI, USA).

2.4. Density

The effect of fuel density on atomisation in pressure-swirl atomisers is often considered less significant compared to viscosity and surface tension. While density influences the mass and momentum of the liquid jet, its role in the breakup dynamics is limited, because atomisation depends primarily on centrifugal forces and aerodynamic interactions. Some studies have suggested that higher-density fuels may increase the droplet size slightly due to increased inertia. This means that denser fuel carries more mass at a given velocity and increases the inertial forces, resisting breakup. This can result in a longer breakup length, larger droplet size, and slower atomisation, but the overall impact on spray quality and breakup length is minor in swirl atomisers. Therefore, density is typically considered a secondary parameter influencing atomisation performance [38,39,40,41,42,43,44].

In the presented paper, the oil density was determined at 15 °C and 20 °C. The obtained results of the used oils densities are presented in Figure 3.

As can be seen from the presented results of oil density in Figure 3, the pyrolytic oil had slightly lower density compared to the heating oil. The difference was very small, measuring 4.3% at 15 °C and 4.1% at 20 °C. The increase in temperature influenced the slight increase in both oils’ densities.

2.5. Viscosity

Viscosity is a measure of a fluid’s internal resistance to flow and shear deformation. It is one of the most influential properties in spray formation, especially in the pressure-swirl atomisers that are used in oil burner applications. High viscosity reduces the internal motion of the swirling liquid film within the nozzle, lowering the angular velocity and weakening the centrifugal atomisation mechanism. This leads to thicker liquid sheets, delayed breakup, larger ligaments, and, ultimately, coarser droplets. It also dampens the growth of the surface instabilities that promote atomisation. In contrast, low-viscosity fuels allow the swirl motion to develop more fully, facilitate faster formation of instabilities on the liquid film, and lead to more efficient breakup into fine droplets. Therefore, the atomisation quality improves with decreasing viscosity, particularly in systems where a fine spray and rapid evaporation are essential for efficient combustion [38,39,40,41,42,43,44].

The presented results of viscosity in Figure 4 indicate that the pyrolytic oil had lower dynamic and kinematic viscosities compared to the heating oil. The dynamic viscosity of PO was 24.4% lower compared to that of ELKO, while the kinematic viscosity of PO was 19.6% lower in comparison to ELKO.

2.6. Surface Tension

Surface tension, defined as the force acting along the surface of a liquid that minimises its area, plays a key role in determining the droplet formation mechanism. Fuels with high surface tension resist the formation of surface instabilities and ligament disintegration, leading to slower breakup, larger droplet sizes, and a reduced spray cone angle. In pressure-swirl atomisers, where centrifugal force and aerodynamic drag initiate sheet breakup, high surface tension impedes these mechanisms and delays atomisation. Oils with lower surface tension form waves and ligaments on the spray surface faster, which promotes spray disintegration, generating finer oil droplets. Thus, surface tension acts as a stabilising force at the liquid–gas interface, which must be overcome for effective atomisation, while reducing it typically enhances atomisation performance in oil burners [38,39,40,41,42,43,44].

Table 2. Results of oil surface tension.

ELKO	27.66 mN/m
PO	24.78 mN/m

presents the surface tension values of ELKO and PO. The obtained value of surface tension for the pyrolytic oil was 11.6% lower than the value of surface tension obtained for the heating oil. Due to the lower surface tension, a wider spray angle may be obtained when using PO.

2.7. Calorific Value

The heating, or calorific value, of oil determines the amount of chemical energy contained in each oil. This energy is released during combustion, so it also influences the total heat release during the combustion process. Higher calorific values of oil were determined in the presented paper. The obtained higher calorific values of the used oils are presented in

Table 3. Results of oil higher calorific values.

ELKO	48.088 MJ/kg
PO	45.359 MJ/kg

.

The HCV results obtained indicate that PO has a very high calorific value. The value of HCV for pyrolytic oil was 6% lower compared to heating oil, which indicates that PO has comparable chemical energy to ELKO.

2.8. ATR FT-IR Spectroscopy

The results of the FT-IR spectra for pyrolytic oil and heating oil are presented in Figure 5. The spectra were recorded at intervals from 4000 to 900 cm⁻¹, with a resolution of 4 cm⁻¹.

The values of the peaks for each oil are presented in

Table 4. Peaks of the FT-IR spectra.

ELKO	PO
Wavenumber [1/cm]
2922.8	2922.6
2854.1	2853.5
1459.4	1641.9
1377.3	1464.9
	1377.8
	991.77
	909.24

.

The FT-IR spectrum for ELKO shows the typical peak signals of aliphatic hydrocarbons at 2922.8 cm⁻¹ and 2854.1 cm⁻¹. These values indicate C–H stretching vibrations from the –CH₂– and –CH₃ groups, typical of aliphatic hydrocarbons. The 1459.4 cm⁻¹ peak corresponds to scissoring of the –CH₂– groups. The 1377.3 cm⁻¹ peak is typical for symmetric bending of the –CH₃ groups. These signals suggest the presence of long aliphatic chains.

The spectra for PO also indicate C–H stretching in the –CH₂– and –CH₃ groups (peaks at 2922.6 cm⁻¹ and 2853.5 cm⁻¹), again suggesting aliphatic hydrocarbons. The 1641.9 cm⁻¹ peak typically represents C=C stretching, possibly from an alkene. For the peaks at 1464.9 cm⁻¹ and 1377.8 cm⁻¹ we again assumed that they correspond to the bending vibrations of the –CH₂– and –CH₃ groups. The bands at 991.77 cm⁻¹ and 909.24 cm⁻¹ are characteristic of out-of-plane C–H bending, supporting the presence of alkene structures.

Both spectra presented similar results in the ranges of 2960 cm⁻¹ to 2850 cm⁻¹, 1460 cm⁻¹, and at 1375 cm⁻¹, but the peaks around 1650 cm⁻¹ and from 991 cm⁻¹ to 909 cm⁻¹ were only detected in the spectra of pyrolytic oil. The spectra of PO indicate a more complex composition compared to that of ELKO, which could also include olefinic or aromatic components.

2.9. Elemental Composition

The results of the PO elemental composition are presented in

Table 5. Elemental composition of PO.

	ELKO	PO
C [%w/w]	86.1	79.7
H [%w/w]	13.8	13.5
N [%w/w]	0	0.2
O2 [%w/w]	0	6.6

. The information about the elemental composition of PO is very important, since the composition of the oil influences its calorific value, emission formation, etc., directly.

The results of the elemental composition of both oils indicate that PO has a slightly lower content of carbon (C), which has a further influence on the 5% lower C/H ratio. The lower C/H ratio has further influence on the lower calorific value of PO oil compared to ELKO, which has a higher C/H ratio.

3. Spray Visualisation and Characterisation

In the present paper, the VisiSize N60 High-Speed Particle and Spray Imaging System was used for the visualisation and characterisation of the spray images. This system allows for non-intrusive, image-based diagnostics of sprays, and measures the size, velocity, and trajectory of spray droplets.

In the present work, 100 pictures of each spray were recorded using the VisiSize system with a picture frame rate of 15 fps. The field of view was 10,965 × 6242 microns for all the measurements and was set to capture the spray at the exit from the injection nozzle. This allowed us to study the primary and secondary breakup process of the spray. A schematic presentation of the designed spray visualisation system is presented in Figure 6. The VisiSize system operates on the principle of Particle/Droplet Image Analysis (PDIA), a direct imaging technique used to analyse droplet size, shape, and distribution across a wide dynamic range. In this method, droplets are illuminated from behind using a laser beam, creating shadow images based on the Mie scattering theory. As light interacts with the droplets, it is scattered and partially absorbed, forming shadow contours that are captured by a high-resolution camera.

What distinguishes PDIA from other image-based techniques is its automated image processing. The captured greyscale images (0 = black, 255 = white) undergo segmentation, where a threshold value is applied to separate the droplets from the background. A threshold level of 85 was used in this study. The segmentation algorithm generates a binary image, enabling precise measurement of the droplet geometric features. The system supports analysis of both spherical and non-spherical droplets by calculating the projected shadow area. To minimise the influence of overlapping or deformed droplets, a sphericity criterion called the shape parameter was applied. A threshold of 0.7 was set in this study to filter out droplets with irregular shapes. The factory calibration data for the VisiSize N60 system were used for all the measurements and remained unchanged throughout the experiments [45,46].

The PDIA method was also used successfully by Kashdan et al. [46,47,48], who tested the technique on a hollow cone pressure-swirl atomiser for fuel injection and compared it with the PDA (Phase Doppler Anemometry) technique. Garai et al. [49] tested the PDIA technique on a hybrid atomiser for diesel fuel injection, Feng et al. [50] tested the technique on diesel fuel injection using a high-pressure common rail injection system, and Jiang et al. [51] tested it using a CCD camera and LED illumination.

From the obtained images, spray characteristics were determined using VisiSize software and an application made in the LabVIEW program, with which the spray angle and atomisation intensity were determined.

3.1. Spray Angle and Atomisation Intensity

The spray angle and atomisation intensity were determined from the obtained images of spray injection using an application made in the LabVIEW program.

The image analysis was made by applying the following procedure written in LabVIEW. First the 100 images were loaded and converted into a 2D array of numbers. The arrays were overlapped, and the values were analysed for each point. The average and standard deviation were calculated. To compensate for the background, the average value was used as a background. The whole procedure was repeated, and the average image was subtracted from each image. Then, the values obtained from each image were overlapped, and the obtained results were used again for average and standard deviation calculations across all the images.

From the obtained results, it was possible to extract the part that presented the spray boundary. This was used further for calculation of the spray angle.

3.2. Experiment Parameters

Pure heating oil and pure pyrolytic oil were injected at 8 bar and 10 bar of injection pressure in the presented work. The pressures used present typical injection pressures in domestic oil burners. The nozzle used for injection was a Steinen Twin Filter oil burner nozzle with a factory-specified mass flow rate of 1.91 kg/h and a 70° solid cone spray angle according to the European CEN Standards EN 293 [52] and EN 299 [53]. The temperature of the ambient air and both oils was 24 °C. Before the start of image recording, the valve was open for 3 s, to ensure that a constant oil flow was formed.

4. Results and Discussion

The following section presents the results of the influence of pyrolytic oil on the spray breakup process in oil burner applications.

4.1. Mass Flow Rate

The obtained results of the mass flow rate for both oils are presented in Figure 7. The obtained results indicate that the oils’ mass flow rates were influenced by the oil properties and injection pressure. Higher injection pressure increased the oils’ mass flow rate. For ELKO, the mass flow rate was only increased slightly, but, for PO, the mass flow rate was increased by almost 22% when increasing the pressure from 8 bar to 10 bar.

Regarding the influence of oil properties, the results indicate that ELKO had a higher oil mass flow than PO in both tested pressures. The higher mass flow rate for ELKO can be attributed to its higher oil density compared to PO.

4.2. Spray Angle

The spray angle is one of the most important spray parameters. The spray angles were determined from the obtained images of oil spray.

A typical image of oil spray with a defined spray (red colour) angle is presented in Figure 8.

The obtained results of the oil spray angles are presented in Figure 9.

The results of the spray angles indicate the expected trend. The spray angle for PO was higher compared to that of ELKO at both injection pressures. This can be attributed to the lower surface tension and lower viscosity of pyrolytic oil, as a decrease in viscosity reduces the internal energy losses within the nozzle, resulting in a wider spray angle. Higher viscosity of ELKO contributes to greater internal friction of the liquid in the swirl section of the nozzle, which reduces tangential velocity inside the nozzle, as viscous dissipation consumes a large portion of the pressure and kinetic energy. Consequently, the dispersion of the spray sheet is reduced. These effects have been confirmed experimentally by Dafsari et al. [40] and numerically via CFD by Laurila et al. [54]. A similar influence of viscosity was also observed by the authors in [55,56]. As can also be seen, the increase in injection pressure leads to an increase in the spray angle, which is consistent with the findings of Chen et al. [55]. The oil spray angle of pyrolytic oil is 8% wider at 8 bar of injection pressure and 9% wider at 10 bar of injection pressure.

4.3. Droplet Characterisation

The following chapter presents the results of the arithmetic mean droplet diameter (AMD), also known as $D_{10}$, and Sauter mean droplet diameter (SMD), also known as $D_{32}$, for both oil and injection pressures obtained from images of the oils sprays using the VisiSize system software.

The arithmetic mean droplet diameter in the VisiSize system is defined with Equation (1) and Sauter mean droplet diameter with Equation (2):

$$AMD=D_{10}=\sum _k{D}_k/N$$

(1)

$$SMD=D_{32}=\sum _k{D}_k^3/D_k^2$$

(2)

where D presents the droplet diameter, and N represents the total number of droplets.

The results of AMD are presented in Figure 10, and the results for SMD are presented in Figure 11.

The results of the arithmetic mean droplet diameter are influenced by oil type and injection pressure. The results for 8 bar of injection pressure indicate that the AMD for ELKO was around 1% higher compared to that of pyrolytic oil. The trend for 10 bar was the same as at lower pressure, with the AMD for ELKO being 6% higher compared to PO. The lower arithmetic mean droplet diameter can be attributed to the lower surface tension and lower viscosity of pyrolytic oil, since both properties affect the development of surface instabilities on the liquid jet according to Bremond et al. [41]. As previously mentioned, lower viscosity and surface tension affect the conditions inside the nozzle. We assume that both the tangential and axial velocities of the liquid are higher due to the lower viscosity, which can influence the thickness of the spray sheet and promote a faster breakup of the sheet into smaller ligaments and droplets [57]. Meanwhile, surface tension represents resistance to the creation of a new surface; consequently, in the case of ELKO, which has a higher surface tension, it is likely that the breakup of ligaments into smaller droplets is slightly delayed. The effect of surface tension on atomisation was also mentioned in [44].

The trends for the Sauter mean droplet diameter were not the same for 8 bar and 10 bar of injection pressure. At a lower injection pressure, the SMD was higher for PO, while, at the higher injection pressure, the SMD was higher for ELKO. The differences were again very small, measuring 3% at the lower pressure and 4% at the higher pressure. The difference in trends of SMD can be attributed to the definition of SMD, as shown in Equation (2). As presented in the results of the droplet diameter distribution for 8 bar and 10 bar, shown in Figures 13 and 4, the spray of PO at 8 bar contained slightly more droplets with a higher diameter (higher than 250 µm) compared to ELKO. This can influence the calculation of SMD further, which can be higher in cases when a very small number of droplets with a larger diameter is obtained in a specific oil spray. The occurrence of larger droplets, which influence the SMD values, can also be attributed to the large difference in mass flow rate at an injection pressure of 8 bar. The mass flow rate differed by 30%, which suggests that, despite the same injection pressure, the oils may exhibit different flow regimes due to their physical properties. At 8 bar, the surface of the ELKO oil spray showed more disturbances caused by aerodynamic interaction between the surrounding air and the oil, due to the higher mass flow rate. These disturbances may contribute to a more stable breakup in terms of a more uniform droplet size distribution. At 10 bar, the difference in mass flow rate was around 7%, and a similar disturbance rate was observed on the surface of the pyrolytic oil spray, indicating that the two oils were in a comparable breakup regime. Consequently, the trends of AMD and SMD were also comparable. The dynamics and formation of disturbances on the oils’ spray/jet surfaces at different injection pressures are shown in Figure 12. At the higher injection pressure, it can be observed that the SMD of pyrolytic oil was slightly lower than that of ELKO oil, which can again be linked to the lower viscosity and lower surface tension of the pyrolytic oil. This is consistent with the findings of Wang and Lefebvre [58], who observed that lower surface tension and lower viscosity also resulted in a lower SMD.

Comparing the results for 8 bar and 10 bar of injection pressure shows that the Sauter mean diameter and arithmetic mean diameter of droplets at 8 bar were lower than at 10 bar. The results indicate an 8% higher droplet AMD and 11.7% higher SMD for ELKO at 10 bar of injection pressure compared to lower injection pressure. For pyrolytic oil, a 3% increase in AMD and 9.3% increase in SMD was obtained at the higher injection pressure. This anomaly can also be explained by the difference in mass flow rates at 8 bar and 10 bar, since increasing the injection pressure increases the mass flow rate through the nozzle. According to the findings of [55,59,60], an increase in mass flow rate through the nozzle leads to a longer breakup length, and, consequently, an increase in SMD. An increase in breakup length also has a direct impact on the field of view of the VisiSize system, which was the same for all the tests. As will be presented in later sections of the presented work, the breakup length of oils at higher pressure was longer, which reduced the view area in which the droplets were monitored. This resulted in a higher number of droplets with larger diameter at 10 bar of injection pressure, which influenced the calculation of AMD and SMD and their values further. This phenomenon can be seen in Figure 14, where a higher number of droplets with diameters between 100 µm and 120 µm were obtained for both oils at 10 bar of injection pressure. At an injection pressure of 10 bar, we also obtained some droplets with a higher diameter (higher than 420 µm) compared to 8 bar of injection pressure, which also influenced a higher droplet AMD and SMD calculation.

Another explanation for a higher AMD and SMD calculation at higher pressure can be found in the work of [34]. They concluded that, at higher injection pressure, higher oil velocities can be expected, which can lead to the collision and coalescence of droplets, which can also result in a slight increase in AMD and SMD.

The results of droplet diameter distribution for both test pressures are presented in Figure 12 and Figure 13.

As can be seen from the results of the droplet diameter distribution for 8 bar of injection pressure in Figure 13, the distribution of droplet diameters for both oils was the same. Both oils produced the most droplets in the range of 25 µm to 80 µm. For HDPE, slightly more droplets were obtained in the range of 30 µm to 60 µm. Pyrolytic oil also produced a small share of droplets with higher diameters, up to 420 µm.

The results for droplet distribution at 10 bar of injection pressure in Figure 14 show similar trends to 8 bar of injection pressure. Again, the distribution of droplet diameters for both oils was the same. There were slightly more particles with diameters between 45 µm and 220 µm for the ELKO oil. We also obtained very small amounts of droplets with a high diameter, up to 690 µm, in the spray of PO. The obtained difference in droplet diameter between the oils was small and can be considered marginal.

4.4. Spray Breakup

The application to monitor the spray breakup was developed in the LabVIEW program. The spray breakup was determined as the standard deviation of the image pixel values, as explained in the chapter “Spray angle and atomisation intensity”. The standard deviation results are presented in Figure 15. There are two areas in which the difference can be observed in spray breakup. They are marked with yellow (the first area is close to the injection nozzle—the left side of the image) and red lines (the second area). A high standard deviation value (the darker colour of the pixels on each image) means that there is a big difference between the pixel values. This indicates a high fluctuation of the oil spray in this specific area (big differences between the presence of oil spray between the images). If the standard deviation values are small, the difference between the presence of oil spray between the images is not as high.

As can be seen in the first marked area of spray breakup, the length of the spray-to-air boundary was slightly longer at 10 bar of injection pressure compared to 8 bar. The boundary was also thicker at lower pressure, which indicates that the spray exhibited more dynamic behaviour. Consequently, this means that the spray layer was also thicker. This was also observed in the theoretical study by Rizk and Lefebvre [38], where they derived an equation for the thickness of the oil liquid sheet, and found that, with an increasing pressure difference, the sheet thickness decreased. Looking at the second marked area, the intensity of the standard deviation values was higher at the lower pressure, which can indicate that the spray break process was more intense in this area. This all indicates that the breakup process at higher pressure occurred at a greater distance from the injection nozzle compared to lower pressure. Since the field of view was the same for all the sprays, the observed area below the spray breakup was decreased, which can influence the droplet diameter distribution at higher injection pressure and the results of the AMD and SMD.

5. Conclusions

In the present paper, the influence of pyrolytic oils produced from waste HDPE plastic on the spray breakup process in an oil burner fuel injection application was tested experimentally. The obtained results were analysed further and compared to the results of heating oil. The main conclusions from the presented study are as follows:

• Using the pyrolysis process for the production of synthetic fuels can recover the embedded energy of waste plastic.
• The calorific value of PO was found to be merely 6% lower compared to conventional ELKO oil, which indicates the possible usage of pyrolytic oils in heating applications, but a slight increase in fuel consumption can be expected.
• The obtained pyrolytic oil (PO) has physical properties that are similar to conventional heating oil properties.
• PO has slightly lower density than ELKO, which consequently affects the fuel mass flow rate through the nozzle. In addition to density, the mass flow rate is also influenced by the injection pressure. An increase in mass flow rate through the nozzle also leads to a longer breakup length.
• The viscosity and surface tension of PO are also lower than those of ELKO. Both properties are very important in terms of atomisation as they affect the internal flow conditions within the nozzle, as well as flow conditions at the nozzle inlet. It was found that PO exhibits a larger spray angle due to its lower viscosity, which can be explained by the findings of other researchers showing that lower viscosity leads to reduced viscous dissipation within the swirl chamber, which preserve more of the available pressure and kinetic energy for generating tangential and axial momentum. Moreover, both properties also effect the spray sheet thickness and the rate of sheet breakup.
• Viscosity, surface tension, and injection pressure also influence the arithmetic mean droplet diameter (AMD) and Sauter mean droplet diameter (SMD). At lower pressure, the AMD for ELKO was 1% higher, whereas at higher pressure it was 6% higher. This is attributed to the lower viscosity and surface tension of PO. For SMD, smaller differences were also observed. At lower pressure, a large difference between the mass flow rates of the two oils was detected, indicating that oils operate in different flow regimes. For ELKO, more instabilities were observed on the jet surface, which may suggest that these instabilities contribute to more uniform breakup at lower injection pressure. At higher injection pressure, disturbances on the jet surface were of similar magnitude, so the SMD is slightly lower for PO due to its lower viscosity and surface tension.
• Overall, the droplet diameters of PO were very similar to the droplet diameters of conventional heating oil, which indicates that the combustion process of both oils may be similar regarding fuel evaporation and mixing with air.

Overall, the presented results indicate the possible usage of pyrolytic oils produced from waste in domestic oil burners from the perspective of spray development. Since the differences in droplet size and spray angle are not significant with the use of pyrolytic oil, the authors suspect that no changes are needed in the injection pressure and combustion chamber design. This can help in the reduction of fossil fuel consumption, utilising the embedded energy of waste plastic, which represents a large part of solid waste. In conclusion, this study demonstrates that plastic waste has the potential for resource recovery and can contribute to minimising the environmental effects associated with plastic usage. The study also indicates that smarter solutions in solid waste treatment present an innovative approach to sustainable management of solid waste, which can also have a positive environmental impact. Before massive usage of pyrolytic oils in oil burners, further studies are needed to determine pyrolytic oil’s impact on atomisation, combustion, and the emission formation process, since PO contains unidentified trace components, which can lead to nozzle blockage or carbon production and deposition during long-term usage in domestic oil heaters, and may also affect the energy rate of return for pyrolytic oil production.