Design of a pyrolyser model for the conversion of thermoplastics into fuels

Abstract

This study focuses on alternatives to the sustainable management of plastic wastes through the development of a pyrolyser model, adapted to the recycling of plastics into fuel. It is a batch reactor, fixed bed, designed and built for the extraction of pyrolysis oil that can be recycled into petrol or diesel. The pyrolyser consists of a reactor with a volume of 0.0424 m³ and a copper spiral condenser with 2.31 m length. The plastics used for this study were Low Density PolyEthylene (LDPE), polypropylene (PP), polystyrene (PS) and polyethylene terephthalate (PET). They were collected from surrounding companies, washed, sampled, cut and sieved. Two different sizes of pyrolysis material: 1–3 cm (G₁) and 3–7 cm (G₂) were obtained and tested. The pyrolysis reactor and the plastics entering at an ambient temperature of 25 °C were heated. Plastics were then melted at 110 °C and vaporised at 450 °C. The hot vapour produced circulated through a copper coil and condensed. The resulting liquid was called pyrolysis oil. The results of this study show that the pyrolysis of LDPE, PP and PS yields two liquids: the heavy and majority fraction which arelike the conventional diesel and the light fraction which is like gasoline. Yields of 6–12.4% for the light fraction and 43.2–63.8% corresponding to the heavy fraction are observed. PE has the highest yield, 63.8% for the heavy fraction and 12.4% for the light fraction. The study further underscores that the size of the pyrolysis material influences the yields, i.e. an increase of 12.5, 9.1 and 7 % for LDPE, PS and PP respectively when the size of the pyrolysis material is increased from G₂ to G₁. In contrast, the results of PET have shown a liquid that solidifies 46 s later. It was also noticed that 2061.34 kJ of energy was required to pyrolyse 1 kg of plastic and produce 0.762 l of fuel. The simple physico-chemical characterisation of the majority fraction shows a great similarity with diesel fuel, as the distillation went beyond 200 °C. Therefore, we can say that the diesel fraction is similar to diesel fuel. We equally observed a high cetane number (52.1–55.1) and a high calorific value (42.9–55.5 MJ/kg). Consequently, there are some points of non-conformity with the European 590 standard and Cameroonian specifications for diesel fuel. These include a low density (767.8–815.1 kg/m³) and a low viscosity at 40 °C (1.108–1.346 mm²/S). A thorough physico-chemical analysis will complete this study before any recommendation for appropriate use.

1. Introduction

Synthesised in 1862 by Alexander Parkes, plastics play an indispensable role in improving our living standard. Its low density, durability and low cost of production makes it an ideal material. The population growth over the years has led to an increase in the use of plastics. Plastics are used in a variety of domains, including packaging, textiles, toys, sports equipment, household appliances, and electronic devices and in agriculture ([1,2]). For example, between 1950 and 2018, global plastic production increased from 1.5 million tonnes to 360 million tonnes, making plastics the third most manufactured material in the world [3]. This increase in the demand of plastics, coupled with the lack of sustainable recycling mechanisms, led to the accumulation of huge amounts of plastic wastes deposits in the environment [4].

At the global scale, it is estimated that only 14% of plastic packaging produced is recovered and only 9% recycled [5]. More than 10% of the world's plastic production is in the oceans, and this flow will exceed the combined mass of all marine fish by 2050 [6]. Moreover, plastic pollution threatens not only the quality of life, but life itself. Indeed, although plastics have greatly contributed to the development of modern societies, they are also responsible for many environmental problems.

Existing literature categorises plastics into two main families: thermoplastics and thermosets [2]. Thermoplastics can be repeatedly softened and melted by heat. On cooling, they are hardened in such a way that they can be used to get new plastic products. They account for over 70% of total plastic production [7]. Thermosets, on the other hand, are hardly recycled. A third family, elastomers can be used, but they are not plastics per se. They are elastic and have a high capacity of deformation. Traditional plastics are resistant to degradation; they resist in the environment for long periods of time with direct negative effects on human and animal health, contamination of the food chain and the inability of ecosystems to absorb these wastes. In addition, they are also sources of visual pollution. However, there are very few end-of-life management options for these wastes.

Inspite of bans on the production of plastics with a density of less than 60 μm (as in Cameroon), local production and consumption is fast growing. These plastic bans even if properly thought out, planned and implemented, are not enough to reduce the amount of plastic waste produced. Therefore, effective waste management to discover new energy sources should be emphasised to sustain the planet's massive population and ecosystem [8].

To address plastic pollution, researchers over the world have proposed many approaches. These include primary recycling, mechanical recycling, chemical recycling and energy recovery [9]. Among plastic recycling techniques, incineration, combustion and gasification face strong opposition from society, particularly because of the toxic gases produced, especially dioxins. Landfilling plastics is not an appropriate option for disposing this waste. Nevertheless, the most attractive method, in line with the principles of sustainable development is chemical recycling. Environmentally, pyrolysis offers an alternative to landfill and does not cause water contamination. Financially, pyrolysis produces fuel with a high calorific value [2]). Among the chemical recycling processes, pyrolysis seems to be the prominent one. According to Weerachai et al., 2022 [4], an alternative fuel can be produced by the pyrolysis of plastic waste. Tapanee et al., 2019 [10] argue that it is an alternative method of recycling plastic into sustainable waste management.

Non-catalytic pyrolysis or thermal pyrolysis of polyolefin is a process of thermal degradation of polymers with long carbon chains into smaller and less complex molecules at high temperatures (300–600 °C) and in the absence of oxygen [11,12]. At the end of this process, three main products are obtained: a mixture of light non-condensable gases, a mixture of light and heavy hydrocarbons and coal or coke [13]. Condensable hydrocarbon oils are composed of paraffins, iso paraffins, olefins, naphthenes and aromatics and non-condensable gases with a high calorific value. The proportion of different fractions and their composition depend mainly on the nature of the plastic waste and the operating conditions such as temperature, pressure, residence time and the type of reactor among others. Consequently, the appropriate choice of a reactor is of paramount importance [14].

This work stretches on the thermal pyrolysis of some thermoplastics. It is true that the use of catalysts during pyrolysis has demonstrated several advantages. There is a high potential for the conversion of plastic waste with improved quality of the liquid product at lower temperatures and reaction times compared with thermal pyrolysis [2]. In addition, a good number of studies dealt with the non-catalytic pyrolysis of thermoplastics ([2,4,9,11,15]). However, a number of key problems remain with this type of degradation, including limited plastic conversion and a high energy balance.

This work aims at developing a pyrolyser model suitable for recycling thermoplastics into fuel; while optimising the conversion rate and reducing the energy balance of the process. In order to improve the energy balance of the thermal pyrolysis of plastics, we have introduced a gas recirculator to make the process autonomous after certain times. The study shows that it is possible to reduce the amount of energy consumed during the pyrolysis of plastics by burning pyrolytic gas. The paper presents the results of the study on the development of a pyrolyser model for the conversion of plastics into petrochemical products. Performance tests were carried out using Low Density Polyethylene, Polypropylene, Polystyrene and Polyethylene terephthalate.

This study has two objectives.

• ⮚Designing and building a pyrolyser model for recycling plastic wastes;
• ⮚Investigating the effect of plastic type and particle size on the yields and quality of pyrolysis oils.

The ambition is to come up with an up to date protocol that is simple, realistic and predictive over a wide range of parameter variations, in transforming plastic wastes into fuel. This will be a major step towards the industrialisation of the process.

2. Materials and methods

This research was conducted at the Renewable Energy Laboratory of the Agricultural Engineering Research Unit of the Faculty of Agronomy and Agricultural Sciences of the University of Dschang, West Region (Cameroon).

2.1. Plastic materials used

Weathered plastics were used for this research. They were collected from surrounding agricultural companies and workshops. Four types were used to perform the pyrolysis tests. These were: Low Density Polyethylene (LDPE), Polyethylene terephthalate (PET), Polypropylene (PP) and Polystyrene (PS). They were chosen because of their abundance and accessibility in the study area. Also, the review of existing literature outlined the possibility of transforming plastics into pre-petrochemical products with good yields [9,[15], [16]].

Practically, plastics were collected, sorted out and thoroughly washed handily with distilled water and then dried under sun; then, they got cut into chips using scissorsand sieved. Two different sizes of pyrolysis material (G₁ and G₂) were considered corresponding to 1–3 cm and 3–7 cm respectively.

2.2. Design of the pyrolysis unit

In the framework of this study, a pyrolysis unit has been designed. It consists of four main parts: a pyrolysis reactor, a heating chamber, a condensation and product recovery system and a support.

The design of the various parts was based on the thermal decomposition parameters of plastics and the laws governing the operation of pyrolysis reactors.

2.3. Tests undertaken

Performance tests were carried out using PE, PET, PP and PS. The aim was to test the designed and manufactured pyrolyser. A total of 120 tests were conducted, i.e., 30 tests for each type of plastic. The pyrolysed plastics were of two different sizes and all the tests were carried out with 1 kg of each type of plastic. In the pyrolysis reactor, plastics entering at an ambient temperature of 25 °C are heated, melted at 110 °C and vaporised at 450 °C. The steam produced circulates through the copper pipe. After a complete turn, the condensable fraction of the hot vapours from the pyrolysis reactor condenses. The resulting liquid is called pyrolysis oils.

Once produced, a simple combustion test was carried out. Thus, 50 ml of each pyrolysis oils was collected and burned.

2.4. Assessment of the energy consumption of the pyrolyser developed

Fir wood was used in the experiments as a source of heat. With a moisture content of less than 15%, it has an PCI of 3.8 kWh/kg. The quantities of wood ($m_b$) used for pyrolysis of 1 kg of plastic were 3.25, 3.1, 2.7 and 2.3 respectively.

Consider the total energy available in the fuel (Q) and the useful energy ($Q_U$). Calling ($Q_P$) the various heat losses, we can write:

$$Q=Q_U+Q_P$$ (1)

Other losses include.

• -Smoke losses$({\mathit{Q}}_{\mathit{f}}$)

$$Q_f=V_f*C_{pf}*T_{fu}$$ (2)

$$V_{fu}=V_{f0}+\left(V_{a0}*e/100\right)$$ (3)

with:

$V_{a0}$: Volume of air used in the complete combustion of one unit of mass;

$e$: Excess air expressed in a percentage;

$V_{fu}$: Volume of flue gas produced by combustion (in stoichiometry this volume is called $V_{f0}$);

$T_{fu}$: Furnace flue gas outlet temperature (in °C);

$C_{pf}$: Average volumetric heat capacity at constant pressure in the temperature range 0 °C to $T_{fu}$.

• -Heat loss through walls$({\mathit{Q}}_{\mathit{v}}$)

$$Q_v=K\left(T_{ip}-T_e\right)S$$ (4)

$$K=\frac{1}{\mathrm{Ʃ}\mathrm{e}/{\lambda }_i)+(\frac{1}{k_e})}$$ (5)

with:

$K$: Convection coefficient

$\mathrm{e}$: Thickness of the i successive layers of refractory or insulation;

${\lambda }_i$: Thermal conductivity of the i successive layers of refractory or insulating materials;

$k_e$: Exchange coefficient of the external wall with the environment;

$T_{ip}$: Temperature of the inner wall of the furnace;

$T_e$: External ambient temperature;

$S$: External surface of the furnace.

• -Energy required to heat the walls$({\mathit{Q}}_{\mathit{m}\mathit{v}}$)

$$Q_{mv}=m_i{C}_i\left(T_{fin}-T_{in}\right)$$ (6)

with:

$m_i$: Mass of the material under consideration;

$C_i$: Mass heat capacity of the material under consideration;

$T_{in}$: Initial temperature of the material;

$T_{fin}$: Final temperature of the material.

• -Coal losses ${\mathit{Q}(}_{\mathit{c}}$)

$$Q_c=m_c*P{CI}_c$$ (7)

with:

$m_c$: Mass of coal remaining;

$P{CI}_c$: Higher heating value of coal.

2.5. Physicochemical characterisation of pyrolysis oils

For this study, only the majority fraction of pyrolysis oils is considered. The physicochemical characterisation tests of the pyrolysis oils were carried out in the Quality Control Laboratory of the National Refining Company Limited (SONARA). Based on the American Society for Testing and Materials and French standards, parameters such as: distillation point (D 86); Pensky flash point (D 93); density (D 4052); cetane number (D 4737/976); kinematic viscosity (D 445) and gross calorific value (NF M 03–005) were determined for pyrolysis oils from PE (Sample A), PP (Sample C) and PS (Sample B) (Fig. 1). The quality of this fuel was then checked by comparing the results obtained with the European standard for diesel fuel (EN 590). The conformity evaluation was done in relation to the requirements/specifications of the Cameroonian market.

Fig. 1.

Samples of pyrolysis oils collected for analysis. A: Majority fraction of oils from LDPE pyrolysis; B: Majority fraction of oils from PS pyrolysis; C: Majority fraction of oils from PP pyrolysis.

2.5.1. Distillation point (D 86)

The pyrolysis oils obtained underwent a distillation phase from the initial to the final point. The methodology described in ASTM D 86 was applied. Oil from the LDPE or PS was introduced separately into a 100 ml test flask and heated. The heating was set at 60–70 °C (recommended for diesel). After boiling, the vapours formed rose towards a condenser bathed in a water mixture. The initial point is the temperature at which the first drop of condensate appears on the tube outlet. This takes about15 min for diesel engines. The diesel's characteristic temperatures are regularly recorded at 10%, 20%, 50% and 90%. At the end of distillation, the temperature reaches its maximum value, the end point of distillation, and decreases.

2.5.2. Pensky flash point (D 93)

The flash point was determined using a Pensky Martens’ apparatus. The apparatus vessel was filled in with the fuels produced up to the gauge line. The heating block was placed at the centre, and then the lid was fitted and locked up. The stirring hose was then fitted and the thermometer inserted. The test flame was then lightened with the diameter adjusted at 4 mm and the heating set at 30 °C. It was then applied on top of the crucible at every degree until a slight explosion occurred and the temperature of the product, which corresponded to the flash point was observed.

2.5.3. Density (D 4052)

Density was calculated using the methodology described in ASTM D 4052. A volume of 500 ml of diesel was poured into a 500 ml test tube. A few minutes were allowed to the densimeter to find its balanced position. It was then rotated slightly until the densimeter was completely stable and all the air bubbles had reached the surface. The temperature and density of the product analysed were then recorded and converted into density at 15 °C using the density at 15 °C conversion table.

2.5.4. Cetane number (D 4737/976)

This describes the calculated formula for the cetane number, particularly applicable to fuels such as diesel. It has been calculated using the following data: density at 15 °C; 50% distilled point, 90% distilled point and 10% distilled point.

2.5.5. Kinematic viscosity (D 445)

The viscometer is filled in with pyrolysis oil up to the gauge line, then immersed in the bath and tilted vertically at approximately 40°. The first opening is then plugged and a pump is set over the second opening to draw into the liquid until it reaches the centre of the second viscometer ball. The suction is then disconnected and the capillary tube unblocked. At the same time, the timer was started to measure the time used for the oil to flow from the first to the second line. The kinematic viscosity in mm2/s is given by Eq (1).

Kinematic viscosity = C*t (8)

C: Constant of the Ubbelohde viscometer_capillary which is equal to 0.04652 mm2/s2;

t: Flow time in seconds.

2.5.6. Gross calorific value (NF M 03–005)

The method used is described in the French standard NF M 03–005. Measurements are carried out in a 250 ml calorimetric flask immersed in a calorimetric vessel itself enclosed in an insulating jacket. The sample is placed in a dish of 25 mm in diameter and 14–19 mm high. Combustion takes place in the presence of oxygen and is triggered electrically by a tungsten wire. The gross calorific value is then determined at constant volume from the temperature rise observed.

3. Results and discussion

3.1. Pyrolysis unit

A semi-slow, batch pyrolyser with a continuous feed option has been designed. To facilitate its mobility and transportation, the pyrolysis unit has been designed to be easily dismantled when necessary. It consists of a pyrolysis reactor with a hermetic lid. A feed hopper and a distillation column are attached to the lid. The distillation column is connected directly to the condenser. The reactor rests in a combustion chamber where the required energy for pyrolysis is transmitted by the combustion of the biomass. The combustion chamber consists of a double wall in which an insulator (clay type 1/1) is placed. Table 1 shows the dimensions of the designed pyrolysis unit.

Table 1.

Characteristics of the developed pyrolyser.

Pyrolyser	Settings	Symbols	Results	Unit
Reactor	Internal diameter Outside diameter	D d	300 302	mm mm
	Length	H	600	mm
	Volume	V	0.042	m3
Heating unit	Cylinder diameter 2 Cylinder diameter 1	Du2 Du1	480 380	mm mm
	Height	Hu	900	mm
Capacitor	Total heat required	Q	2061.3	KJ
	LMTD corrected	ΔTm	120.3	°C
	Length Internal diameter	L Di	2.31 10.92	m mm

The manufactured condenser is used to cool the steam resulting from the breakdown of 1 kg of plastic. To do that, the amount of heat to be transferred (q) to the water for condensation to occur was determined to be 1 kJ/s. The heat exchanger in which the heat given off by the hot fluid is entirely transferred to the cold fluid consists of a copper coil, formed of 5 coils, and spaced 5 cm from each other. This condenser model works in counter-current to optimise the heat exchange between the hot vapours from the pyrolyser and the cooling water in the tank where the condenser is set. Fig. 2, Fig. 3, Fig. 4, Fig. 5, Fig. 6 below show the technical diagrams of the various parts of the pyrolysis unit.

Fig. 2.

Technical specification of the reactor.

Fig. 3.

Technical specification of the heating unit.

Fig. 4.

Technical specification of the condenser.

Fig. 5.

Designed pyrolysis unit.

Fig. 6.

Manufactured pyrolysis unit.

This pyrolysis unit is made up of 11 compartments (Fig. 5): a reactor (1), insulation (2), a heating chamber (3), a distillation column (4), a spiral condenser (5), a container (6), a chimney and filter membrane (7), gas recirculator (8), water inlet area (9), a water outlet area (10) and a hopper (11).

The cylindrical stainless steel reactor is 0.6 m high and 0.3 m in diameter and is fed from a hopper. A heating unit heats the feedstock inside to temperatures of about 550 °C using biomass. The reactor is thermally insulated with a clay jacket. A thermocouple (Th) placed inside the reactor is used to monitor temperature changes during the process. The pressure inside the reactor was not measured. The distillation column is connected to the outlet of the reactor, enabling the hot vapour to be delivered to the spiral condenser. The latter opens into a container where the condensable products are stored. The non-condensable products are discharged through a pipe to the burners to support the combustion.

In the study by Otoikhian et al. (2021) [17] on a co-pyrolysis system generating waste-plastic extract fuels of 8 mm thick, 300 mm high, 180 mm shell internal diameter and condensing pipes of 1000 mm long; the pyrolysis of 2 kg of HDPE, PP and PS mixed gave yields of 1.7 l–1.85 l for an average duration of 3 h. Elsewhere, Jasan et al. (2018) [18], in their experiment designed and built a pyrolyser for the conversion of plastic waste. In a stainless steel reactor tube of 145 mm long with 37 mm internal diameter, the pyrolysis at about 500 °C gave fuels. Furthermore, Ajay et al. (2017) [19] also designed a 474 mm high, 494 mm internal diameter reactor with Shell and Tube Condenser. Tests on the unit showed that 10 kg of LDPE yielded 6.1 l of fuel oil.

The special feature of this model is its energy savings. A non-condensable gas recirculation system reduces the fuel input 30 min after the start of the first vapours, which improves the overall energy balance compared with other pyrolysis reactors. In fact, Kunscer showed that gas from the pyrolysis of PE could be recovered because its calorific value is 65 MJ/Nm³. As a result, it was found that the theoretical energy required for the pyrolysis of 1 kg of plastic was not supplied. During the experiment, the actual demand was lower than the calculated theoretical demand. With this device, the process becomes autonomous from the 169th, 163rd, 153rd and 65th minutes for PP, PE, PS and PET respectively.In addition to the overall energy balance of the pyrolysis process, which is improved with the pyrolyser model developed, there is an initial separation between the heavy and light fractions on leaving the condensation stage. Indeed, the device implemented in this study, in addition to generating a condensable fraction with a liquid appearance, enables an initial separation of this liquid into two sub-fractions. This differs from the results of many other authors (Kassagry, Kuncser). In the experiment by Kuncser, 2011, a condensable fraction with a gel-like appearance for the pyrolysis of PE was obtained. This author continued with the experiment by varying the heating rate. He ended up concluding that the pyrolysis of PE in a conventional electrically heated cylindrical reactor, even at different heating rates, did not improve the viscosity of the condensable fraction. He therefore proposed the design of a new pilot plant to carry out catalytic pyrolysis in order to improve this important energy characteristic for internal engine applications. This result illustrates the specificity of the present pyrolyser model designed and its advantage in optimising the conversion rate.

3.2. Pyrolyser efficiency tests

The start-up of the pyrolyser was done in 4 steps. A first temperature stage of 10 min, not only to eliminate disturbances that may occur at the beginning of the heating, but also to ensure the stability of the pyrolyser. Another rapid heating stage lasted between 50 and 75 min and ranged from 150 to 450 °C. This was followed by a temperature stabilisation stage which extended from 450 °C to 600 °C and lasted between 90 and 120 min. Finally, a 25-min stage corresponded to the cooling of the pyrolyser from 600 °C to 250 °C.

In each series of trials, 1000 g of LDPE, PET, PP and PS were individually loaded into the reactor. The pyrolysis times were approximately 228, 220, 206 and 120 min for PP, LDPE, PS and PET respectively. The average yields of liquid and solid products from non-catalytic degradation were determined and presented in Fig. 7.

Fig. 7.

Average yields (%) of products from pyrolysis trials.

For all the tests carried out in general, the yields of the heavy fraction ranged from 57.3 to 65.1%, 55.9–60.4% and 44.8–53.1% for LDPE, PS and PP respectively. For the light fraction, it ranged from 3.4 to 16.3% for all the plastics tested.

Here, G represents the two different sizes of pyrolysis material, i.e. G₁ of 1–3 cm and G₂ of 3–7 cm.

In general, three products are obtained during pyrolysis. They are a mixture of light non-condensable gases, a condensable fraction and coal (coke). This result corroborates with some previous works ([20,21]). In this study, the non-condensable fraction of the gas yield was not measured as the gases were oriented directly to the combustion chamber; thus, sustaining the combustion. For PE, PP and PS in particular, pyrolysis yielded two liquids, a mixture of heavy (light brown) and light (yellowish) hydrocarbons. In some other cases, tar was obtained in small proportions. Comparison of the results for each type of plastic in different sizes shows that the smaller the particle, the shorter the reaction time and the higher the yields of the liquid fraction. Thus, for class 1 (1–3) cm plastics, the liquid yield is higher than the yield of the same class 2 (3–7) cm plastics. This increase in the heavy fraction is 7, 9.2 and 12.5 % for PP, PS and PE respectively.

The non-catalytic pyrolysis of thermoplastics (LDPE, PET, PP and PS) leads to a relatively low yield compared to the data quoted by several other authors ([9,12,19]). This difference can be explained by the rather long observation period, which favours the conversion of the condensable fraction into a non-condensable fraction [21]. The pyrolysis of the PE yielded three products, namely: a major fraction (Fig. 8a), a minor fraction (Fig. 8b) and coke (Fig. 8c).

Fig. 8.

Minor fraction (a), major fraction (b) and coke (c) products of LDPE pyrolysis.

The pyrolysis treatment of PET resulted into a liquid that solidified 46 s after leaving the condenser (Fig. 9c). This result is equally maintained by the works of authors who argue that pyrolysis of PET results into - solid and - gas [12] and differs from the work of [21] when he sustains that PET like many other plastics reacts during thermal conversion.

Fig. 9.

Pyrolysis products of PS (a) PP (b) ant PET (c).

3.3. Energy balances

Table 2 shows the use of the energy supplied by the fuel (in MJ). Let Q be the energy available in the fuel. The useful energy (Qu) was calculated by deducting from Q all the heat losses (Q_P) during the plastics pyrolysis process (G1). The following losses were considered.

• -Losses in the coal (Qc);
• -Flue gas losses (Qf) ;
• -Wall losses (Qv);
• -The energy required to heat the walls (Qmv);
• -Losses through the furnace openings (Qe).

Table 2.

Use of the energy provided by the fuel.

Plastic	$\mathbf{Q}$ (MJ)	${\mathit{Q}}_{\mathit{u}}$ (MJ)	Losses (MJ)
			${\mathit{Q}}_{\mathit{c}}$	${\mathit{Q}}_{\mathit{m}\mathit{v}}$	${\mathit{Q}}_{\mathit{f}}$	${\mathit{Q}}_{\mathit{v}}$	${\mathit{Q}}_{\mathit{e}}$
LDPE	45.14	2.05	20.68	3.23	10.45	1.97	6.77
PP	42.27	1.73	19.22	3.23	9.78	1.97	6.34
PS	36.94	1.63	16.02	3.23	8.55	1.97	5.54
PET	31.46	1.45	12.82	3.23	7.28	1.97	4.72

These different losses can be calculated from the simple equations governing heat transfer in such reactors.

The useful energy is quite low. In fact, one of the major problems observed during the experiments is the loss of energy. However, the low rate of conversion of available energy into useful energy does not depend on the pyrolyser model or its technology. Rather, findings point to the need of improving combustion parameters (fuel, air inlet conditions and inlet and outlet temperatures).

In order to quantify the contribution of non-condensable gases, the energy balance of the pyrolyser developed was calculated. According to the mass conservation equation, the pyrolysis of LDPE, PP, PS and PET produced 0.22, 0.36, 0.33 and 0.28 kg of non-condensable gas respectively. From up to date literature, the mass heat capacity of gases from plastic pyrolysis, under thermal conditions similar to those of our operation, would be about 1.08 kJ/kg.k (Kassagry, 2018). This approximate value was used to calculate the amount of energy provided by the non-condensable fraction during pyrolysis. Table 3 compares the amount of energy involved during pyrolysis in this experiment with that echoed in the literature.

Table 3.

Energy involved in the pyrolysis of plastics.

Items	Energy supplied during pyrolysis (MJ) (Literature)	Calculated energy demand (MJ)	Energy supplied during pyrolysis (MJ)
			Energy provided by non-condensable gases	Energy from biomass
LDPE	2.1 [21]	2.06	0.12	2,05
PP	1.76 [9]	1.9	0.21	1.73
PS		1.73	0.19	1.63
PET		1.7	0.55	1.45

For the present experiment, it was observed that the amount of energy required for the pyrolysis of 1 kg of plastic is provided jointly by the combustion of the biomass and that of the non-condensable fraction resulting from the said reaction. Hence, the useful energy of the biomass taken individually is less than the theoretical energy demand (calculated) necessary for the pyrolysis of 1 kg of plastic and cannot therefore ensure the complete pyrolysis reaction. It can therefore be stated that recirculation of the non-condensable fraction has reduced the energy demand. Accordingly, the pyrolyser model developed optimises the conversion rate and reduces energy consumption in the pyrolysis process.

3.4. Combustion tests

When burning the majority fraction of the liquids from the pyrolysis of PE in the open air, it was observed the difficulty to ignite and that it had a very high smoke smell. This result shows that the majority fraction of the plastic pyrolysis can be assimilated to diesel (Fig. 10).

Fig. 10.

Combustion tests of conventional diesel (a) and majority fraction from LDPE (b).

Table 4 shows a comparison of the combustion test of the majority fraction from PE pyrolysis and diesel fuels.

Table 4.

Comparison of the combustion test between conventional diesel and the majority LDPE fraction.

Settings	Majority fraction	Diesel
Combustion time	1 h 29 min and 16 s	15 min and 5 s
Colour of the flame	Light yellow	Light yellow
Smell of smoke	High, black	Very high, black
Difficulty lighting up	Very difficult	Difficult
Continuity of the flame	Slight discontinuity	Continuous
Noise	Small spark	No noise

In the combustion tests (Fig. 11), no differences in activation, colour, and odour and combustion continuity were observed between the minority fraction of the LDPE pyrolysis and conventional petrol. However, a difference was recorded in the combustion time. Indeed, for the same volume, the combustion time was longer for the LDPE liquid than for the gasoline.

Fig. 11.

Combustion tests of gasoline (a) and the minority fraction of LDPE (b).

Table 5 reports the combustion test of the minority fraction from LDPE pyrolysis.

Table 5.

Comparison of combustion tests between minority LDPE and petrol.

Settings	Minority products	Petrol
Combustion time	1 h 18 min and 50 s	24 min and 4 s
Colour of the flame	Yellow	Yellow
Smell of smoke	Low	Low
Difficulty lighting up	No	No
Continuity of the flame	Continuous	Continuous

From the findings, it can be concluded that the minority fraction resulting from the pyrolysis of LDPE can be assimilated to conventional gasoline. But, complete physicochemical analyses will readjust this first test.

3.5. Physicochemical analysis of pyrolysis oils

The pyrolysis oils were analysed using standard methods. In general, the LDPE, PP and PS samples are the good ones according to Cameroonian specifications in terms of diesel fuel except for some characteristics such as density, distillation, flash point (Pensky Martens) and viscosity. Other characteristics such as sulphur content (very low), cetane index (high) and Gross calorific value make this new type of fuel excellent. Table 6, shows the values of the physicochemical characteristics of oils from LDPE, PP and PS.

Table 6.

Physicochemical and combustion characteristics of the majority fraction of pyrolysis oils.

Items		End point distillation	Flash point	Density	Viscosity at 40 °C	Cetane index	Gross calorific value
Measurement units		°C	°C	Kg/m3	mm/s2		MJ/kg
LDPE		328.7	37.0	761.8	1.108	52.1	54.8
PP		339.9	37.0	796.5	1.189	53,4	49,3
PS		373.5	37.0	815.1	1.346	55.1	45.1
Specified values (EU)	Min		>55	820	2	>46
	Max			845	4.5
Specified values (Cameroon)	Min		60.0	820	1.6	45.0	45.9
	Max	390.0		880	6.0

In previous explorations, plastic waste-derived oil had similar properties to commercial diesel [22]. But in the present study, some points of divergence were revealed. In fact, the formulated fuels have a lower flash point than diesel. This is 37.0 °C for both LDPE and PS. The minimum value (55) is therefore not reached, and fire risks may arise when the fuel is used in a diesel engine. This result differs from those of Wasif et al., 2020 [11] where they obtained a flash point of 33.6 °C and 26.1 °C for LDPE and PS respectively. The density of the pyrolysis oils obtained was lower than that of diesel and is close to that obtained by Weerachai et al. (2022) [4]. The cetane numbers of the formulated fuels are high compared with those of conventional diesel. In terms of viscosity however, it was found that the viscosity of the fuels obtained from the pyrolysis of these three thermoplastics did not comply with the standards, and the values obtained were very low. This result equally differs from the work of Weerachai et al. (2022) [4], who obtained a viscosity of 3.065 mm2/s for oils derived from PE.

When the minority fraction was distilled, significant differences were observed compared with the majority fraction. During distillation (ASTM D 86), the boiling temperature at 95% of the light fraction was 176, 184.2 and 189 °C for PE, PS and PP respectively. This value is a characteristic of gasoline. At 90%, the boiling point of the heavy fraction was 277.5, 345.5 and 348.2 °C for PE, PS and PP respectively. These are the characteristic values for diesels. It can therefore be concluded that “the heavy and majority fraction is like the conventional diesel and the light fraction is like gasoline.”

Based on the characteristics of pyrolysis oils, several authors opine that modifications of this oil are necessary before it can be used in an engine ([2,9]) or that pyrolysis can be carried out using catalysts [23]. This improves upon its viscosity, flash point and density, and consequently the engine's energy performance. It also appears from this characterisation that the majority fraction of the oils from pyrolysis still contains a considerable proportion of light fraction. It would be interesting to carry out a cross-sectional separation to isolate the different fractions.

4. Conclusion

This work centered on the development of a thermochemical recycling device by pyrolysis of plastic wastes with the aim of producing alternative fuels. By so doing, a semi-slow, batch pyrolyser was designed and built. It consists of a reactor with a volume of 0.04 m³. Four types of plastics (LDPE, PP, PS and PET) in two classes (1–3) cm and (3–7) cm, were used for the tests. It was demonstrated that it is possible to transform some thermoplastics into petrochemical products. The smaller the particle size, the better the oil yield. The highest yield (63.8%) was obtained in the pyrolysis of LDPE with a particle size between (1–3) cm. The oil combustion tests assert that the pyrolysis of PE, PS and PP yields liquid products that can be used as fuel. In contrast, the pyrolysis of PET gives a product that cannot be used as a fuel. From the physicochemical analysis of the oils from LDPE, PP and PS, it has been realised that the majority fraction can be assimilated to diesel fuel. However, further analyses must be carried out to make this fuel match up with the European 590 standard. We can therefore conclude that pyrolysis is a promising alternative for the recycling of some plastic wastes.

Data availability statement

Data will be made available on request.

CRediT authorship contribution statement

Nadia H. Dassi Djoukouo: Writing – review & editing, Writing – original draft, Methodology, Investigation, Funding acquisition. Boris Merlain K. Djousse: Writing – review & editing, Supervision. Henri G. Doukeng: Writing – review & editing, Methodology. Daniel A.M. Egbe: Writing – review & editing, Funding acquisition. Julius K. Tangka: Writing – review & editing, Methodology. Martin Tchoffo: Writing – review & editing, Supervision, Methodology.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:DJOUKOUO DASSI Nadia Hillary reports administrative support was provided by None.

Appendix A. Supplementary data

The following is the Supplementary data to this article.