Examination of the Emission of Gas-Phase Components, Including Some Not-Conventional Ones from a Parking Heater, While Increasing the Bioethanol Content of the Fuel

Abstract

The air pollutant emissions of a motor vehicle do not only mean the emissions from the engine used for propelling the vehicle. Another possible source is the passenger compartment heater. In the focus of our research, we investigated the effect of mixing gasoline-powered cabin air and ethanol (E10, E30, E100 on a volumetric basis) on the emissions of the equipment. Among the pollutant components examined, there were conventional components and so called not-conventional components. The chosen length of the test was 1800 s, while the intake air parameters temperature (t_in) and relative humidity (h) have been kept constant (t_in = 15 °C, h = 30%). Bioethanol mixing has a significant reducing effect on NO_x (oxides of nitrogen) and CO₂ (carbon dioxide). As for the components CO (carbon monoxide), THC (total hydrocarbons), CH₄ (methane) and N₂O (nitrous oxide), the values of the components reach usable values only in the start-up and burnout phases, while in the stable-operation phase, their values are outside the limit of detectability. A small part of THC is only CH₄; a more significant part is NMHC (non-methane hydrocarbons). The results of the developed vehicle fleet model for calculating the GHG (greenhouse gas) emissions of a vehicle fleet equipped with such a device showed that the fleet’s GHG emissions are less than 1% of the annual emissions from the combustion of transport fuel.

1. Introduction

As far as the emission of air pollutants from vehicles is concerned, we usually only think of the air pollutant emissions produced by the engine. However, there are other emissions from vehicles. One such emission, commonly referred to in the literature as non-exhaustive emissions, is the pollution released into the environment from, for example, tyre and road surface wear, or clutch wear [1,2,3]. Ageing of rubber hoses and seals can lead to leakage, both in fuel supply systems (as the ethanol content of fuels increases, this process accelerates, and new sealants are needed) and in the cooling circuit or air conditioning system of engines, which can release many greenhouse gases. That also includes the evaporation of fuel during refuelling, which in the case of petrol, is converted into hydrocarbon vapour, which also contains gases that are harmful to the environment [4,5,6]. Exhaustive emissions, but not from the vehicle’s engine, include air pollution from the parking heater device. It is important to stress that the combustion products from the operation of these heaters are released directly into the environment as raw gas without any after-treatment. But if it could be possible, e.g., in non-road applications, to introduce the flue gas into the catching system of the propulsion engine, then the after-treatment system could convert the harmful components with very good efficiency [7]. Today, a wide range of devices (fuel operated heaters) is available on the market, from different manufacturers, with different technical solutions and performances. The number of vehicles equipped with combustion heaters is increasing. These devices can be divided into two groups, depending on the fuel they are fuelled with and the fuel used by the engine. These devices receive their fuel from the vehicle’s original fuel system. There may also be two other groups, one heating the engine cooling water and one supplying hot air directly to the passenger compartment [8,9,10,11]. These devices use liquid fuel to produce heat energy, thus polluting the air environment. They are connected to the vehicle’s original fuel system. So, if the fuel contains a renewable component, the device uses it. If the renewable component to be blended into the motor gasoline used in Europe changes, the emissions of these appliances will be affected [12,13]. Heaters are available in different power ranges, in the order of magnitude of 1–10 kW [8,9,10,11].

For the scope of this article, the conventional air pollutants include NO_x (nitrogen oxides), N₂O (nitrous oxide), CO₂ (carbon dioxide), CO (carbon monoxide), and THC (total hydrocarbons). CH₄ (methane) and NMHC (non-methane hydrocarbons) are part of THC, so they are discussed here. NO_x (oxides of nitrogen) are generally used for a gas containing two gaseous components. The first is nitrogen monoxide, a colourless, odourless gas, and nitrogen dioxide, a reddish-brown, pungent-smelling gas. Nitrogen oxide reacts with oxygen in the air or with ozone in the air to form nitrogen dioxide. Inhaling pure gases can quickly become fatal. Nitrous oxide is a vital greenhouse gas and also damages the ozone layer [14]. Carbon monoxide exerts its health effects by reacting with the blood’s haemoglobin. The affinity of haemoglobin for CO is more than 200 times greater than for oxygen, so even at low concentrations, it can replace oxygen in the blood [15]. CO₂ is a gas produced naturally by living organisms (humans, animals) during respiration and the decomposition of biomass and is used by plants during photosynthesis. Although it makes up only 0.04% of the atmosphere, it is one of the most important greenhouse gases [16]. Total hydrocarbons (THC) are the sum of all volatile compounds measured by a flame ionisation detector (FID). Methane is the second most important greenhouse gas contributing to climate change after carbon dioxide. In addition, methane is a potent local air pollutant and contributes to ozone formation, which causes serious health problems [17,18,19]. Under normal environmental conditions, nitrous oxide is a colourless gas with a slightly sweet taste and odour. It is non-flammable but supports combustion and is only slightly soluble in water. It has anaesthetic and analgesic properties, and is analgesic if inhaled sufficiently [20].

In the present study, SO₂ (sulfur dioxide), NH₃ (ammonia), C₂H₂ (acetylene), C₂H₄ (ethylene), C₂H₆ (ethane), C₃H₆ (propylene), and C₄H₆ (butene) were treated as unconventional components. Sulphur dioxide is typically released during the combustion or oxidation of fuels or other sulphur-containing materials. It is a pollutant that contributes to acid deposition, which can lead to changes in soil and water quality. The downstream effects of acid deposition can be significant, including adverse effects on aquatic ecosystems in rivers and lakes and damage to forests, plants and other vegetation. Sulphur dioxide emissions also aggravate asthmatic conditions and can reduce lung function and cause respiratory inflammation. A secondary effect is that they also contribute to forming atmospheric particulate matter in sulphate bonds, which are essential air pollutants because of their adverse effects on human health [21,22]. Ammonia forms acidic deposits and contributes to so-called eutrophication, which can lead to potential soil and water quality changes. The downstream effects of acid deposition can be significant, including adverse effects on aquatic ecosystems in rivers and lakes, and damage to forests, plants and other vegetation [23,24]. In general, in industrial practice, acetylene is not considered a severe toxic hazard. However, acetylene is a serious fire and explosion hazard. The literature has a wealth of information on deaths due to acetylene explosions. Acetylene is a simple asphyxiant. Symptoms of exposure include dizziness, headache, fatigue, nausea and vomiting. Exposure to high concentrations can cause unconsciousness and death [25,26]. Ethylene is a colourless gas at room temperature. It is used as a coolant and in the welding and machining of metals. It is also used to produce ethylene oxide, mustard gas and other organic substances, and to accelerate fruit ripening. Ethylene gas, when inhaled, has harmful effects on the human body. Skin contact with liquid ethylene can cause frostbite. Exposure to ethylene can cause headache, dizziness, fatigue, confusion and loss of consciousness. Ethylene is a highly flammable and reactive chemical and a dangerous fire and explosion hazard [27,28]. Ethane is an organic chemical compound with the chemical formula C₂H₆. At standard temperature and pressure, ethane is a colourless, odourless gas. Like many hydrocarbons, ethane is isolated on an industrial scale from natural gas and as a petrochemical by-product of petroleum refining. Its primary use is as a feedstock for ethylene production [29,30]. Propylene is a colourless, slightly odorous gas or liquid under pressure. It produces many organic chemicals, including resins, plastics, synthetic rubber and gasoline. Propylene is harmful to the body when inhaled. Exposure to high concentrations can cause dizziness and fainting. Propylene can damage the liver. Exposure may affect the heart and nervous system. Propylene is a highly flammable gas and a fire hazard [31]. 1,3-butadiene is an organic compound with the formula CH₂=CH-CH=CH₂. It is a colourless gas that readily condenses into a liquid. It is industrially essential as a precursor for synthetic rubber. The molecule can be considered as the union of two vinyl groups. Although butadiene is rapidly degraded in the atmosphere, it is still found in the ambient air in urban and suburban areas due to constant emissions from motor vehicles [4,32].

There needs to be more literature available on the emissions of such devices. Research on heaters is diversified, e.g., from a control engineering point of view, comfort management research, retrofitting of the heater to an electric vehicle, and detailed particulate emission studies.

In [33], using a stationary heater with a control system, the study optimises the appliance’s operation regarding the cabin’s comfort. It mainly deals with heat quantity and temperature parameters. The study investigates particle number, particle mass, and black carbon emissions from heaters installed in vehicles and operated under actual driving conditions. The tests were conducted in Finland during winter at low ambient temperatures. Six different vehicles, three petrol-engine and three diesel-engine passenger cars with different mileages were selected for the tests. The test time was set to 30 min. The total NO_x emissions of diesel vehicles was 20% lower than the total emissions of petrol vehicles. THC emissions, on the other hand, were lower for petrol vehicles compared to diesel vehicles. The devices were fuelled with conventional commercially available fuels [12].

In [34], the authors present and evaluate retrofit heaters for electric cars. One of the options is fuel-operated heaters, and one of these is to heat the air. It gives results comparing the heating capabilities of each solution in terms of cabin temperature and electrical consumption and also examines the impact of each heater on the range. It concludes that electric heaters significantly impact the range of electric cars and that fuel-operated heaters (FOH) are an attractive solution in cold regions. Furthermore, retrofitting is not a particular problem. In a well-to-cabin assessment, a vehicle with a purely electric drive will show CO₂ emissions if equipped with a FOH. In contrast, a vehicle with a purely electric drive is expected to have zero emissions.

Ref. [35] presents studies on a battery-powered electric vehicle (BEV), a passenger car with a retrofitted bioethanol fuel-operated heater. The battery charge was tested under real driving conditions, when the heater was running and when it was not. Ambient temperatures were −11 °C and −13 °C. FOH operation corresponds to a range increase of approximately 21.6%. The same comparative test was performed when the ambient temperature was +1 °C. This represents a calculated increase of about 5.5%. The heater was operated with bioethanol only, and no fuel comparison existed.

The research investigated the effect of a water heater with a 5 kW thermal output on the warm-up behaviour of a 2.2 L, four-cylinder, 105 kW common rail direct injection diesel test engine. The test engine was mounted on an engine test bench with an asynchronous machine. The exhaust system and the passenger compartment heating circuit were arranged in series. The cabin heat exchanger was insulated to prevent heat transfer. Unacceptably high emissions of HC (hydrocarbons) and CO from in situ combustion at low start-up temperatures were measured. A fuel-fuelled auxiliary heater, which ensures the rapid heating of the combustion chamber elements, particularly the cylinder head, reduces the crude emissions of the propulsion engine and contributes to faster catalyst warm-up. Fuel-fired air heaters are classically used in the cabs of commercial vehicles as preheaters during waiting and rest periods and night stops. This is confirmed by measurements of air temperatures at the air intake of a small van in preheating mode at an ambient temperature of −14 °C. Whereas a 5 kW water heater reaches a vent outlet temperature of +20 °C in 720 s, a 4 kW air heater achieves this in 240 s. The benefits of air heating would ideally meet the comfort requirements of high-end luxury cars and vans [36].

Bioethanol heating for electric vehicles has been developed as a CO₂-friendly solution. The thermal output as a lambda function has been investigated from a bioethanol share of 0% to 85%. They presented appliance types for bioethanol operation in different power ranges (2–4 kW). No other fuel was tested, only bioethanol. To compare the effect of an ethanol fuel heater and an electric heater, a comparative test was carried out at −7 °C with a small-car-class electric vehicle with four suitable seats, according to the TÜV Süd, TSECC (TÜV Süd Electric Car Cycle) driving cycle. The TSECC cycle lasts 60 min for each run. A distance of 60 km is covered. Therefore, the range of the fuel heater compared to the electric heater was arithmetically determined through energy consumption. The total electricity consumption during the TSECC cycle was 15.72 kWh. During this time, the ethanol heater introduced an average of 3.17 kWh of thermal energy into the water cycle. If this energy had been drawn from the vehicle’s battery, based on an electric water heater with a thermal efficiency of about 85%, 3.7 kWh should have been used. If an electric heater were used, the battery capacity available for displacement would be reduced by just this amount to 12.02 kWh. This would result in a range of only 45.88 km, corresponding to a range reduction of about 30% [37].

In one study [38], there was an attempt to reduce the emissions of FOHs. Emissions and their variation over time were analysed over about 16 min. CO emissions were the highest during the start-up and shutdown phases, showing how much they can be reduced using a catalyst. They also saw emission reduction potentials in the use of different renewable fuels. Gasoil and its alternatives, such as PME (palm oil methyl ester), BtL (biomass to liquid) and OME (oxymethylene ether) are being investigated, and petrol and its alternative E85 are also being investigated. CO, NO_x and soot emissions were investigated as a function of air pollution factors. All components were reduced when using renewable fuel over the entire range of air pollutants tested compared to conventional fuel, and emissions were further reduced with a catalyst.

One article [13] deals with the EU’s (European Union) fuel quality regulation, alternative fuel infrastructure and renewable energy. It presents the alternative fuels expected in the future and the modular devices developed for them. It also presents the results for a diesel conventional (B7) and a diesel alternative (HVO, hydrogenated vegetable oil), a petrol conventional (E10) and a petrol alternative (1-butanol) fuel, in terms of the CO air pollutant component, as a function of the air to fuel excess ratio (λ). Emissions were measured at two different power outputs. One is at 5 kW, which is considered the full load of the combustor, and at 2.5 kW, which is considered a part load. The result can be summarized that, at full load, the different fuels (diesel and gasoline) have the same CO emissions, and at part load, the alternative fuels have CO-reducing effects, especially at higher lambda values.

Test series have been conducted regarding the emissions of buses and coaches at idling speed before and after fitting a fuel-operated device. The tests cover several buses and longer operating times. Summary results show that with the installation of the device, the idling time of buses is reduced, and all measured pollutant components (CO, NO_x, HC, CO₂ and PM (particulate mass)) are also reduced by an average of 28.5%, but to different extents for each component. The emissions of the installed device have not been investigated [39].

It was not the aim of this research to investigate this device. Nevertheless, it could be an exciting area of research to study the conversion of such a device to run on gas fuel, as has been studied for internal combustion engines [40].

There are few papers in the literature on the gaseous pollutants emitted by these appliances; so, this work aims to investigate this. By default, we investigate the commercially available motor gasoline, the base fuel. However, our studies aim to investigate the change in emissions when the renewable component in motor gasoline is increased and with a fully renewable component (E100). In addition to the conventional gaseous components, our research also investigates other non-traditional, toxic components, mainly carbon–hydrogen components. We extended our investigations with calculations of total vehicle fleet emissions.

2. Materials and Methods

The base fuel for our tests was commercially available motor gasoline that complies with EU and Hungarian standards [41,42]. It was purchased in Hungary from the well of an Austrian manufacturer. As a renewable fuel used as a blending component and on its own in the test series, bioethanol for automotive use, produced according to the standard given in [43], was chosen and obtained from a plant in Hungary. The two materials were used to form 3 different test fuels: (i) the commercially available motor gasoline E10, containing a 10 v/v% renewable component; and (ii) a blend of the two fuels with a bio-component content of 30 v/v%. We chose the experiments with the mixture called E30 because of the following: (i) In addition to the brave renewable goals of the EU (European Union) [44], a more extensive series of E30 tests were also carried out on motor vehicles [45]. The last (ii) fuel tested was pure bioethanol (E100).

Table 1. Some combustion relevant properties of gasoline and ethanol [41,47,48,49].

Property	Gasoline	Ethanol
Formula (liquid)	C8H18	C2H6O
Density [kg/dm3]	765	785
Heat of Vaporization [kJ/kg]	305	840
Lower Heating Value [kJ/kg]	44,000	26,900
Stoichiometric Air-to-Fuel Ratio [kgair/kgfuel]	15.13	9.00
Molecular Weight [kg/kmol]	114.15	46.07
(Final) Boiling Point [°C]	210	78
Kinematic Viscosity [cSt] (at 20 °C)	0.71–0.88	1.2–1.52

shows some properties of petrol and ethanol that are important for oxidation. In the case of gasoline, there is a surrogate chemical formula. The fuel is composed of several chemical compounds [46]. Density is a fundamental property; it shows the mass of fuel delivered to the combustion chamber for a given fuel volume. The densities of the two fuels are similar. The heat of vaporisation of ethanol is more than twice that of gasoline; so, it takes more than twice as much heat to vaporise a unit mass. The lower heating value of petrol is almost double that of ethanol; hence, twice as much work can be obtained from a unit mass of fuel in the case of gasoline. The stoichiometric air requirement for ethanol is significantly lower due to its lower molecular carbon and hydrogen content. The same applies to the molecular weight. A comparison of boiling points shows that ethanol converts from a liquid to a gas at much lower temperatures, which improves combustion. Kinematic viscosity affects the combustion efficiency by the quality of atomisation. Gasoline has a lower kinematic viscosity and, therefore, better atomisation properties.

We have designed a 30 min test procedure for the following reasons. On the one hand, the so-called WLTC (Worldwide Harmonised Light Vehicles Test Cycle), which is used in emission-related type approval tests of road passenger cars has the same duration [50]. In this way, comparing the emissions with those of a passenger car engine is possible. On the other hand, in real-life winter conditions, statistics show that the heater is used to warm up the cab once in the morning [8]. The test duration of 1800 s, the period of the switched-on state, and the harmful components were also tested in part of the cool-down phase; thus, the complete emission recording cycle was 1900 s. No international standard, specification or regulation for testing these kinds of devices regarding emissions could be found. Manufacturers use their own developed cycles for testing appliances. An international standard or specification can be expected when so many FOHs are on the market that their total emissions reach some significant value compared to the total emissions of vehicles.

Our tests were conducted in a state-of-the-art engine test bench under controlled conditions. That was necessary to ensure repeatability. Details of the laboratory equipment are given in the next section. During the series of measurements, the parameters of the intake air of the device were kept constant at t_in = 15 °C and a relative humidity of 30%. That was the minimum value achievable by the equipment for both parameters.

The measurements were performed three times for a temperature–humidity–mixing ratio parameter triple. The mean of the three results is shown, and no standard deviation is calculated.

Before any measurement with different fuels, the stationary heater was dismantled and cleaned. The fuel evaporated through the burning mesh into the combustion chamber, where deposits and soot were visible. A new part was permanently installed to ensure the same starting and measuring conditions. Before start-up, the engine test bench environment and the heater were tempered. The measuring equipment was switched on prior to the start-up of the instrument, and a self-cleaning (purge) process was carried out before each measurement. For each measurement, the heater was shut down after 1800 s after start-up, where it was left to cool down in the measuring chamber. The measurements were repeated three times for each fuel mixture, and the averaged values were used for the evaluations. During the series of tests, the instrument was operated at full load, i.e., the fundamental frequency of the fuel pump was not interfered with. The device was operated in the Normal mode of operation, out of three possible modes (Economy, Normal, Boost).

The conversion of gas components from ppm to grams was performed as described in [51]. The components are plotted as a function of time. On the other hand, calculations have been performed to calculate the following parameters: (i) the total emissions of the cycle and (ii) the emissions of the start-up and shutdown phases. The following parameter (iii) is the ratio of the first two. Parameter (iv) is the emission level when the unit is in a steady state, and parameter (v) is also a calculated parameter of the cycle’s total emissions if it were to operate in a steady state all the time.

Other vehicle fleets and total emissions calculations were carried out, and a model was created. The model aims to obtain a more significant scale emission of a vehicle fleet because we are curious to see how calculated total emissions might compare to the total transport emissions in Hungary. The model is simple and constructed as follows: the measured emission values are multiplied by the number of vehicles in the given category. The vehicles selected are goods vehicles with a gross vehicle weight of 12,000 kg and above, goods vehicles with a gross vehicle weight between 3500 and 12,000 kg, goods vehicles with a gross vehicle weight of 3500 kg and 10% of passenger cars with a gross vehicle weight of 3500 kg.

The emissions obtained from the above are further multiplied by 60, as if a vehicle were to operate every working day during a winter period, assuming three months of winter and taking only working days in the three months, with a single daily use in the morning. The model results are for one year. The model results are very difficult or impossible to validate. The authors’ aim with this model is only to obtain an order of magnitude compared with the total transport emissions from fuel combustion in Hungary.

3. Experimental Setup

The schematic diagram in Figure 1 is not an integral part of the experimental setup. Nevertheless, we would like to show the heater from a more distant perspective. It includes the fuel being supplied from the vehicle’s tank, and the flue gas vented to the environment. The air to be heated is drawn in from the passenger compartment, which is also discharged into the passenger compartment. In the picture, the passenger compartment is the passenger compartment of a truck. Still, it can also be the passenger car’s passenger compartment, caravan or boat cabin.

The experimental setup is shown in Figure 2, and the main characteristics of the parking heater are shown in

Table 2. Technical parameters of the stationary heating device (own editing).

Nominal Heat Power	5.5 [kW]
Type	cabin (air) heater
Fuel supply	by electric pump, pulsing
Injection	burner mesh
Evaporation	glow plug
Fuel	gasoline

. The parking heater is a German-made, gasoline-operated cabin air heater with a rated power of 5.5 kW. The appliance is constructed in such a way that it is necessary to combust the liquid fuel, i.e., pump, burning mesh (atomization), and combustion chamber are followed by an exhaust chamber.

The measuring units fitted to the instrument are shown in both Figure 2 and

Table 3. Parts of measurement system (own editing).

Path	Parameter	Instrument, Device	Make, Type
Air	Intake air humidity and temperature	Humidity and temperature sensor	Vaisala HMT310
Combustion	Flame temperature	Thermo couple	N type sensor with QuantumX MX1609B
	Air excess ratio	Lambda sensor	Bosch LSU 4.9 wide band sensor with ETAS ES636.1 module
Exhaust	Exhaust temperature	Thermo couple	K type sensor with QuantumX MX1609KB
	Gaseous exhaust components	Exhaust gas analyser	AVL SESAM i60 FT SII

. The gas analyser is relevant to the present study, that is, an AVL-manufactured instrument of the type AVL SESAM i60 FT SII. The principle of operation is Fourier Transform Infrared Spectroscopy [53]. The AVL PUMA 2^TM [54] and Webasto Thermo Test [55] programs were used for data acquisition. The acquired data were evaluated using AVL CONCERTO 5™ [56] and Microsoft Excel programs. Calibrations of the instruments were performed according to the annual calibration plan for the laboratory and are valid.

The device and the measurement system built on it were placed in a state-of-the-art engine test bench, where the ambient air and the air supplying the combustion were tempered using a device for supplying air to the internal combustion engines used on the engine brake bench. That is what the AVL ConsysAir 2400 [57] was designed for.

4. Results and Discussion

4.1. Exploring and Defining the Device’s Operating Phases

The operation of stationary heating systems can be divided into four stages. The first is the start-up (i), the second is the steady-state (ii), the third is the burnout (iii), and the fourth is the cooling (iv) stage.

(i) Start-up phase: Preheating occurs at about 80% glow plug power at low combustion air fan speeds for 12 s. Heating continues for 18 s at a lower glowing rate of approximately 60%. Then, with the glow plug operating, fuel delivery is started, and with it, the excess air supply. Fuel delivery starts at a higher pump frequency called preheating for 5 s, during which combustion is initiated. For 54 s, a continuous fuel delivery of 2.05 Hz (with a pumping frequency of 1 Hz, the fuel delivery is approximately 0.114 L/h) and a slight decrease in the glow plug power stabilise the flame. After about 92 s from start-up, the flame stabilises, the glow plug stops working, and the pump delivery and the fan speed increase until the steady state is reached, which takes about 280 s after the heater is started; the whole start-up process is shown in Figure 3. The variation of emission values is shown in later sections.

(ii) Steady-state phase: The fan speed and pump frequency are nearly constant in a steady state. In this mode, at a constant fan speed of 610 per second, the control unit of the stationary heater reduces the pump fuel delivery for 12 s. As a result, as shown in Figure 4, the lambda value increases and then stabilises again.

(iii) Phase burnout: When the device is switched off, the fuel pump stops working. In order to burn out the fuel remaining in the burning mesh, the control unit briefly operates the glow plug at a lower fan speed (Figure 5).

(iv) Cool-down phase: After the burnout phase, the device cools down by gradually reducing the fan speed (Figure 4) and then stops altogether. Since the design of the device means that the combustion air and the heating air fan blades are driven by the same electric motor, the combustion chamber and outside are cooled evenly. When the outside of the device is cooled, the circulated air is delivered to the room to be heated, which cools from about 100 °C to 20 °C, so even this can be considered helpful in heating air at average usage. During the cooling of the combustion chamber, some gaseous components are still emitted by the device, which will be described in more detail later.

4.2. The Tested Air Pollutant Components as a Function of Time and Bioethanol Mixing

4.2.1. Conventional Components

Figure 6 shows the emission of NO_x, CO and CO₂ components as a function of test time and bioethanol mixing. In the case of all three components, the start, stable operation, and shutdown effect on emissions and the mixing of bioethanol have a significant effect. However, the reduction in fuel injection every 600 s is visible. With E10 fuel, the NO_x emission in stable operation is 89 ppm/s (0.0006 g/s), with E30 at 65 ppm/s (0.0004 g/s), with E100 at 25 ppm/s (0.0002 g/s) and the % value of the change compared to the E10 base fuel is −27.8% and −72.2%, respectively. It can also be observed that the reduction in emissions becomes smaller as the bio ratio increases if the fuel is reduced every 600 s. Regarding CO emissions, significant emissions appear in the start, fuel-reduction and shutdown phases. The device was operated with the same intake air volume and fuel delivery parameters regardless of the fuel composition tested. Since the stoichiometric air demand for ethanol was significantly lower, it operated with ethanol at a much higher air-to-fuel excess ratio than for gasoline. Furthermore, the calorific value of ethanol is almost half that of gasoline. Therefore, the same amount of ethanol can cause less heat generation. NO_x generation depends mainly on the O₂ content of the environment, i.e., the ambient air conditions and temperature [58]. Due to these factors, the NO_x generated decreased significantly with increasing ethanol content.

In regard to CO emissions, unlike NO_x, in the stable operating range, the emissions are negligible. The values of the specific emissions are 49 ppm/s for E10, 31 ppm/s for E30 and 19 ppm/s for E100. So, the value of the emissions in the steady state is practically independent of the fuel used if we look at the emissions compared to the start-up and burnout stages. We obtain large values if we express the change from the values as a percentage. Changing the amount of fuel delivery results in growth peaks in the case of CO, which may be due to a change in lambda or combustion stability. The peak values of CO during this process reach very similar values for all tested fuels. The function of CO₂ emissions follows the function of NO_x emissions. It increases from zero in the start-up phase, reaches a stable value in the stable-operation phase, and continuously decreases from the stable value in the shutdown phase. The individual values of CO₂ at the stable operating points are E10—96,237 ppm/s, E30—87,137 ppm/s and E100—65,629 ppm/s. The change rate is −10% and −33.7%, respectively. As ethanol content increases, CO₂ content decreases for several reasons. The air excess ratio mentioned in the literature changes the calorific value of mixture because ethanol has fewer carbon molecules [34,37].

For NO_x, the start-up phase emissions are 7% and the burn-up phase are 0.2% of full cycle emissions. For this reason, it is best to run the two-seater as little as possible during a start-up. These parameters are 25% and 8% for CO; so, stable operating emissions are key. If we look at CO₂ emissions from the same point of view, we obtain values similar to those for NO_x, which are exactly 9% and 0.3%, respectively, so that, again, it is the stable operating emissions during a single operating cycle that are the decisive factor.

It is very important to note the phenomenon that can be seen in Figure 6. This is why the CO diagram has been magnified for the steady-state mode. But, the same can be seen in the NO_x and CO₂ charts without magnification. You can see that emission functions oscillate stochastically for gasoline, while they do not for pure ethanol. With E30, oscillation is also observed, but to a lesser extent than with E10. We hypothesize that this is partly due to the fact that gasoline (in this case E10) is a heterogeneous mixture, while ethanol is homogeneous because it is built up of single and same molecules. On the other hand, as we have seen in the analysis of Table 1, the boiling point of ethanol is much lower than that of petrol; so, it converts to a gaseous vapour much faster than petrol and can therefore improve the uniformity of the combustion–time process.

In Figure 7, the THC, CH₄ and N₂O emission results can be seen. As for the THC results, there is only a significant amount of emissions during the start-up and shutdown phases and very little during the stable-operation phase. It is at the detection limit, and we take it as zero. CH₄ emission is part of the THC; so, its nature is the same as a function of time. The amount of CH₄ emissions during the start-up and shutdown phases is orders of magnitude lower than the values of THC. We also take it as zero in a stable state. No results were found in the literature for THC release from FOHs to gasoline–ethanol. THCs are basically the vaporized, unburned hydrocarbons present in the flue gas. Combustion in the two-seater is not similar to either the Otto or Diesel combustion processes. The combustion chamber has a relatively large volume into which the fuel is introduced by poor-quality atomization and ignition. The high excess air factor reduces it significantly and keeps it low in stable operation.

The N₂O emissions are also at the detectability limit in the entire test range, regardless of the fuel used. With these emission levels, CH₄ and N₂O as greenhouse gases do not significantly contribute to GHG emissions in addition to CO₂. After the burnout phase, in the case of CH₄, there is an additional emission during the device’s cooling, which also affects THC emission. As the amount of air flowed to cool the combustion chamber decreases, the emission value also decreases.

In the case of THC, the start-up and burnout phases account for almost all of the emissions, with values of 0.81% to 99.19% for start-up and burnout, respectively. Also, for CH₄, the total steady-state emissions are negligible compared to the emissions during the two transient phases, which are 12.15 and 87.85 in percentage terms, respectively. That means that for hydrocarbon emissions, the start-up and burnout phases account for almost all of the emissions. From an operational point of view, once the appliance has been started, it is best to keep it on for as long as possible. The ratios analysed above also apply to the other hydrocarbon components shown in Figure 8.

4.2.2. Not-Conventional Components

In Figure 8, the time course of the components is shown, where the components, in the context of this article, are called non-conventional components. A large part of these components are hydrocarbon molecules, which are toxic to the environment, and we also added sulphur dioxide and ammonia. The latter two are outside the detectability range (less than five ppm) and show emission values independent of the fuel used. The other hydrocarbons are part of THC. NMHC, which stands for non-methane hydrocarbons, is the difference between the total and methane hydrocarbons.

For this reason, all HC molecules follow the characteristics of THC, i.e., they only appear in the start-up and shutdown phases. Looking at the NMHC curves for the values, we see that almost all THC is NMHC and very little is CH₄. So, most THC contains directly harmful components to the living environment, rather than a gas component that causes a greenhouse effect.

As mentioned above, in the case of different CH molecules, emissions can only be evaluated in the start and stop phases, and their peak value is independent of the fuel used. If the maximum values of each CH in the start phase are added, approximately 50 ppm is obtained. The peak value of NMHC in the start phase is approximately 1000 ppm; so, we only covered the hydrocarbon emission in detail with these CH molecules. The same ratio can be observed for the firing phase. The peak of NMHC is around 6000 ppm, while the sum of the peak values of C₂H₂, C₂H₄, C₂H₆ and C₃H₆ and C₄H₆ is approximately 100 ppm. During the cooling phase of the device, in addition to CH₄, a gradual decrease in the emission value can also be seen in the case of NMHC, and both together form THC. An exponential decrease is also observed in the case of C₂H₆, C₃H₆ and C₄H₆. While in the case of the other components, the emission value ceases almost immediately at the end of burnout. The low value of sulphur dioxide emissions was as expected. The fuel used contained little or no sulphur [41]. Ammonia emissions were significantly off, outside the detection limit, similar to the values for SO₂ and NH₃ components. No research results were found for these components, and therefore results cannot be compared with them.

To reduce the cost of these kinds of measurements, ANN or some machine learning model can be a good tool [58,59]. There are more and more applications of ANN to simulate the operation of thermal machines with different cycles. Still, examples of such FOH simulations still need to be found.

4.3. Emission Level of Air Pollutant Components Depending on the Rate of Bioethanol

4.3.1. Conventional Components—Full Cycle Emissions

Figure 9 shows the amount of all harmful substances emitted during the measurement cycle (start-up, stable operation, shutdown) represented in grams. A significant decrease can be observed in the case of NO_x (69.5%), CO₂ (28.8%), THC (97.4%), and a more moderate decrease in CO (20.8%), moving from base fuel to pure ethanol. These are indeed due to ethanol’s lower calorific value and oxygen content. The cumulative value of NO_x is 0.89 g/cycle, 66 g/cycle and 27 g/cycle for E10, E30 and E100, respectively. The value of CO₂ emission is 945 g/cycle, 864 g/cycle and 674 g/cycle, showing a decreasing trend with increasing bioethanol proportion in the same order for THC emissions of 53 g/cycle, 0.33 g/cycle and 0.02 g/cycle, respectively. In the case of methane, no clear trend can be observed.

An increasing trend can be seen for CH₄ (6.2%) and N₂O (33.4%), which are greenhouse gases. The CH₄ emission expressed in grams is quite small, and the change shown by adding bio alcohol is also small. A 0.0022 g/cycle belongs to E10, 0.0018 g/cycle to E30 and a 0.0013 g/cycle belongs to the tested fuel E100. In the case of N₂O, the absolute values of the emissions are an order of magnitude higher than CH₄, and the variation is also an order of magnitude higher: 0.0075 g/cycle, 0.0084 g/cycle and 0.0111 g/cycle for E10, E30 and E100, respectively. During the measurements, it was observed that the combustion temperature decreased with an increase in the ethanol content, which may have had a significant effect on the reduction in NO_x formation. This effect (calorific value) could also be effective in the reduction in CO₂ emissions. As a result of the higher ethanol content, the combustion was also more uniform, which can be explained by the more homogeneous composition of the ethanol fuel, and the CO and THC emissions may have decreased due to the oxygen content.

4.3.2. Conventional Components—Emissions during the Start-Up and Burnout Phases

Figure 10 shows the values in grams of the standard components emitted during start-up and burnout. In most cases, with the addition of bioethanol, a tendentious decrease or increase can be observed for the individual components. In the combustion phases of NO_x and CO and the start-up phase of THC, the expected trend is not achieved. That is certainly due to the high-slope transient process. In the case of NO_x, CO₂ and N₂O, the trends for the biological effect seen in the entire cycle are the same as those seen in the start-up and burnout phases. The trends of CO, THC and CH differ from those seen in the whole cycle. It should be emphasized that THC and CH₄ are mirror images of each other in bio-tendency in the initiation and firing phases. The emission value of the entire cycle is determined by the emission generated in stationary operation in the case of NO_x and CO₂. At the same time, the start-up and burnout phases are decisive for the other components. The results of Section 4.3.1 and Section 4.3.2 are not presented in these sections; the specific emission values of non-traditional components can also be found in

Table A1. Emission results of fuel-operated parking heating device with investigated fuels.

Pollutant		NOx	N2O	CO	CO2	THC	CH4	NMHC	C2H2	C2H4	C2H6	C3H6	C4H6
Investigated fuel blend		1. Full cycle’s emission [g/cycle]
E10		0.89	0.0075	0.41	945	0.532	0.0022	0.530	0.0003	0.0005	0.0032	0.0008	0.0005
E30		0.66	0.0084	0.33	864	0.329	0.0018	0.327	0.0004	0.0010	0.0021	0.0005	0.0002
E100		0.27	0.0111	0.34	674	0.019	0.0013	0.017	0.0005	0.0012	0.0000	0.0000	0.0000
		2. Emission during the start and burnout phases [g]
E10		0.065	0.0013	0.136	85.5	0.532	0.0022	0.530	0.0003	0.0005	0.0032	0.0008	0.0005
E30		0.051	0.0014	0.157	77.9	0.329	0.0018	0.327	0.0004	0.0008	0.0021	0.0005	0.0002
E100		0.026	0.0015	0.215	63.9	0.019	0.0013	0.017	0.0005	0.0012	0.0000	0.0000	0.0000
		3. Emission level in a steady-state condition of the cycle [g/s]
E10		0.0006	0.0000	0.0002	0.58	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000
E30		0.0004	0.0000	0.0001	0.53	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000
E100		0.0002	0.0000	0.0000	0.41	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000
		4. The emission of the cycle if the emission were the steady-state emission level during the entire cycle [g]
E10		1.0	0.0073	0.33	1038	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000
E30		0.74	0.0082	0.21	947	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000
E100		0.30	0.0115	0.14	735	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000

.

4.4. Comparison of Cycle Emission Results with Vehicle Emission Type Test Limit Values

During the emission type test of passenger cars and light trucks, we compared the limit values of the regulated gaseous components with the values of the cycle emissions we received. The values are given in

Table 4. Comparison of cycle’s emissions.

Emission Component	E10 [g/cycle]	E30 [g/cycle]	E100 [g/cycle]	Euro 5a Positive Ignition Emissions Limits (N1 CL3) [g/km] [60]	Euro 6b, 6c, 6d Temp, 6d Positive Ignition Emissions Limits (N1 CL3) [g/km] [60]
NOx	0.89	0.66	0.27	0.082	0.082
CO	0.41	0.33	0.34	2.270	2.270
THC	0.53	0.33	0.02	0.160	0.160
NMHC	0.530	0.327	0.017	0.108	0.108
The emission of the cycle if the emissions were at the steady-state emission level
Emission Component	E10 [g/cycle]	E30 [g/cycle]	E100 [g/cycle]	Euro 5a Positive Ignition Emissions Limits (N1 CL3) [g/km] [41]	Euro 6b, 6c, 6d Temp, 6d Positive Ignition Emissions Limits (N1 CL3) [g/km] [41]
NOx	1.0	0.74	0.30	0.082	0.082
CO	0.33	0.21	0.14	2.270	2.270
THC	0.00	0.00	0.00	0.160	0.160
NMHC	0.00	0.00	0.00	0.108	0.108

. In the case of our measurements, the emission values are given for the entire cycle, while the type test limits apply to 1 km. The value of the type test cycle is approximately ten times this. The vehicle category chosen from the point of view of the limit values is N1 and Class 3. The Euro 5a limits are the same as those of Euro 6b, 6c, 6d temp, and 6d regarding the gaseous components presented. There is a difference between the values of particulate mass, which, as a parameter, is not involved in this research. The NO_x cycle emissions decrease significantly depending on the ethanol content but with the base fuel, they reach approximately ten times the type approval limit value, and even in the case of E100, approximately 3.3 times. From the point of view of the CO component, the device’s emission is favourable; in the case of all three tested fuels, it remains well below the vehicle type test limit. This is expected because the device operates with a more extensive excess air. Regarding THC emissions, when testing base fuel, the value measured in the device and the limit value were almost the same. The growing organic component significantly reduced the device’s THC output. The last relevant component is non-methane hydrocarbons. With increasing bioethanol content, the amount of emissions decreased significantly. In the case of E10 fuel, in one cycle, approximately 500% more is produced than the limit value for 1 km in the type test, but in the case of E30, the emission is already approximately 300% of the type test limit value, and in the case of E100, it approximately agrees with the type test limit value.

Table A2. Emission results of calculation model for emission from vehicle fleet with FOH.

Pollutant		NOx	N2O	CO	CO2	THC	CH4	NMHC	C2H2	C2H4	C2H6	C3H6	C4H6
Investigated fuel blend		1. FOHs’ total emission—heavy-duty vehicles (permitted total weight > 12,000 kg) [kg]
E10		1606	13.5	748	1,706,739	961	3.9	958	0.6	0.9	5.8	1.5	1.0
E30		1189	15.1	602	1,559,916	594	3.3	591	0.7	1.8	3.9	0.9	0.4
E100		492	20.0	615	1,216,390	34.5	2.4	31.4	0.9	2.2	0.1	0.1	0.0
		2. FOHs’ total emission—heavy-duty vehicles (permitted total weight 3500 kg–12,000 kg) [kg]
E10		7937	66.7	3695	8,434,470	4750	19.3	4733	2.9	4.4	28.9	7.2	4.9
E30		5878	74.8	2973	7,708,890	2934	16.3	2920	3.3	8.8	19.1	4.6	2.2
E100		2432	99.0	3042	6,011,231	170	11.7	155	4.5	11.0	0.2	0.3	0.0
		3. FOHs’ total emission—light-duty vehicles (permitted total weight < 3500 kg) [kg]
E10		18,971	159	8831	20,158,699	11,353	46	11,312	6.9	10	69.1	17.3	11.6
E30		14,048	179	7105	18,424,535	7013	39	6978	8.0	21	45.6	11.0	5.3
E100		5814	237	7270	14,367,066	407	28	371	11	26	0.6	0.7	0.0
		4. 10% of passenger car fleet [kg]
E10		20,921	176	9739	22,231,099	12,521	51	12,475	7.6	11	76	19	12.8
E30		15,492	197	7835	20,318,656	7734	43	7696	8.8	23	50.2	12.2	5.8
E100		6411	261	8017	15,844,062	449	31	409	12	29	0.7	0.8	0.0
		5. Every vehicle category [t]
E10		49.44	0.42	23.01	52,531	29.59	0.120	29.48	0.018	0.027	0.180	0.045	0.030
E30		36.61	0.47	18.51	48,012	18.28	0.102	18.18	0.021	0.055	0.119	0.029	0.014
E100		15.15	0.62	18.94	37,439	1.06	0.073	0.967	0.028	0.069	0.002	0.002	0.000

contains the average values of steady-state emissions in g/s. From there, we created a cycle index that shows how large the emissions would be if the entire cycle operated with this average cycle. We created this to see “how clean” the operation of the device is when it operates much longer than the 1800 s currently tested; so, in terms of total emissions, stable operation is dominant. For comparison, we also entered this in Table 4. Regarding NO_x, it can be seen that the steady-state cycle is dominant compared to the transient cycle. In the case of CO, transient emissions are more decisive. Since THC and NMHC emissions are practically non-existent in stable operation, long-term operation is also favourable from this point of view. Compared to the type test values, these results do not change in proportion to the results of the initial cycle.

4.5. Results of the Vehicle Emissions Model

In this subsection, we perform emission analyses related to the vehicle stock. The description of the model created for the calculations is included in Section 2. The results of the calculation model can be found in Table A2. According to the data of [61], in 2020, there were 30,101 freight vehicles with a permissible total weight of 12,000 kg or more, while there were 148,755 freight vehicles with a permissible total weight between 3500 and 12,000 kg in Hungary. As for freight vehicles with a permissible total weight of 3500 kg, their number is 355,530. Part of the model stipulated that 10% (392,080 [62]) of passenger vehicles with a permissible gross weight of 3500 kg could be fitted with such a device since they are high-end vehicles. In total, the above four sub-stocks represent 926,466 vehicles. We assumed that due to the number of vehicles, it is worthwhile to calculate the total emissions, even if the emissions of a parking heater are lower than the emissions generated by a vehicle’s engine.

However, it is essential to note that our tests were carried out with a gasoline-powered parking heater. As previously explained, the heating devices mostly obtain their fuel from the tank of the given vehicle, which is diesel in the case of trucks and heavy-duty vehicles. The difference in emissions between the parking heater running on gasoline and diesel can make a difference in the emission values of the individual pollutant components. However, according to the technical judgment of the authors, they are not significant from the point of view of the magnitude of the total emissions.

The value of the total emissions obtained from the model may be lower than it is in reality. The reasons for this can be as follow:

• The measurements were made at 15 °C; so the emissions are certainly lower than if measured in winter temperature conditions.
• The device with a nominal power of 5.5 kW is the largest available in use for heating trucks and ships, campers, and larger cabins. So, they are relevant for more than just road vehicles.
• Proper use for 60 min in the winter can be an underestimation in the case of trucks, since they have to be heated all night.

However, the value of the total emissions from the model can be increased in the direction of reality because, with this calculation, we assume the following:

• All vehicles are in use, and it is winter; the system is used once every weekday. While not all vehicles may be used, not all are used for commuting.
• The heating devices used in passenger vehicles have a lower power, 1–2 kW, and therefore can realize a lower emission than measured.

According to [63], in 2020, GHG (greenhouse effective gas) emissions from the combustion of transport fuel were 12.58 million tons with a CO_2e (carbon dioxide equivalent) measurement unit. From the CO₂, CH₄ and N₂O emissions we measured, we also calculated GHG emissions by considering the values given by [64] when converting CH₄ and N₂O to CO_2e. Then, the total traffic and stock values we measured and calculated were proportioned according to the following table (

Table 5. Comparison of GHG results.

Investigated Fuel	(i) Vehicle Stock Emission CALCULATED with the Model Used [Tons CO2e]	(ii) Hungary’s Total GHG Emissions from the Combustion of Transport Fuel [Million Tons CO2e] [63]	Ratio between (i) and (ii) [%]
E10	52,534	12.58	0.42
E30	48,153		0.38
E100	37,642		0.30

). Therefore, the total GHG emissions calculated according to our calculation model with the tested fuels (E10, E30, E100) are 0.42, 0.38, and 0.30 per cent, respectively, of the GHG emissions from the total transportation fuel combustion. The model used is simple, but in the authors’ opinion, a more detailed model would maintain the same ratio compared to total transport emissions.

The following can be said regarding the model results from the point of view of the other relevant components. The NO_x and CO components reach values of several tens of tons. N₂O can also be measured in hundreds of kilograms and THC in several tons. During the cycle, CH molecules with a significantly low emission value can mean more than 10 kg for the environment for a year and for this number of vehicles.

The model calculation should be further developed and refined. If the number of FOHs increases, emissions will also increase. Emissions may move out of the range of being no longer negligible. And it may even be necessary to develop regulations on the emissions of appliances.

5. Conclusions

In the research presented above, the results of a detailed air pollutant emission test for vehicles designed with stationary air heaters have been presented. This device was operating with a base fuel (E10), another gasoline–ethanol mixture (E30) and pure ethanol (E100), while both conventional and non-conventional (detailed hydrocarbon) components have been measured. Total emission values for the vehicle fleet from the cycle emissions have also been calculated. Based on our results, the most important conclusions are the following:

• With the help of the device’s operating parameters (combustion air fan, fuel pump, glow plug), three phases of operation (the start-up, stable-operation and burnout phases) could be defined. The changes in the individual emission components confirmed the phase definitions.
• For some of the gaseous pollutant components (NO_x, CO₂), the start-up, stable-operation and burnout phases can be separated, and the addition of bioethanol has a significant reducing effect (NO_x: 27.8–72.2%, CO₂: 10–33.7%). As for CO, THC, CH₄ and N₂O emissions, the cycle’s different phases can also be clearly distinguished here. The emission values of these components reach a usable value during the start-up and burnout phases, while in the stable-operation phase, their value is outside the detectability limit. Their emission value is independent of the fuel used.
• Regarding the individual hydrocarbon molecules, sulphur-dioxide and ammonia, there is an appreciable amount of emissions in the start-up and burnout phases, while the emissions in the stable-operation phase are minimal. A tiny part of THC is CH₄, and a more significant part is NMHC. The examined CH molecules have covered only approximately 10% of NMHC quantitatively, so 90% of NMHC has not been investigated. Some deviation from the fuel used can be observed in the start-up and burnout stages but not in the steady-state operating stages.
• The effect of increasing bioethanol mixing on the cycle emissions concerning the individual components developed according to the following trends. NO_x, CO, CO₂ and THC emissions decreased, while CH₄ and N₂O emissions increased.
• For the NO_x, CO and CO₂ emission components and their steady-state values as a function of time, the process oscillates stochastically for E10 and is constant for E100. The authors believe this is due to the homogeneous composition of bioethanol and its much lower boiling point than petrol. That was not found for other components (e.g., THC and its parts, NH₃, SO₂, N₂O) because their values are outside the detection limit in steady-state operation.
• Among the tested components, we selected those included in the emission type test requirements of an N1, Class 3 light truck, and we took the EURO 5 and 6 limit values from the legislation. With the base fuel, the cycle emission values of NO_x, THC and NMHC were higher than the type test limit values, but the values of CO were not. In the case of the other two tested fuels (E30, E100), the cycle values of all components were lower than the limit values. Regarding NO_x, the emission level of the stable-operation phase was decisive, while in the case of CO, THC and NMHC, only small emissions were present.
• The results of the calculation model showed that GHG emissions are less than 1% of the annual emission from the combustion of transport fuel. On the other hand, the low cycle emission values can reach a significant mass in the case of many vehicles and multiple uses (as shown in the model). The NO_x and CO components reach values of several tens of tons. N₂O can also be measured in hundreds of kilograms and THC in several tons. During the cycle, CH molecules with a significantly low emission value can mean more than 10 kg for the environment for a year and for this number of vehicles.

6. Outlook

In addition to the comprehensive examination of the air pollutant emissions of the heater investigated some types of tested fuels, the authors see great importance in further detailed carbon–hydrogen analysis, in the examination of particle emissions and in the exploration of the amount of heat that can be introduced into the vehicle cabin with different fuels. The model calculation should be further developed and refined. If the number of FOHs increases, emissions will also increase. Emissions may move out of the range of being no longer negligible. And it may even be necessary to develop regulations on the emissions of appliances.