Performance and emissions of a CRDI diesel engine fuelled with swine lard methyl esters–diesel mixture

Fuel 164 (2016) 206–219

Contents lists available at ScienceDirect

Fuel journal homepage: www.elsevier.com/locate/fuel

Performance and emissions of a CRDI diesel engine fuelled with swine lard methyl esters–diesel mixture Maciej Mikulski ⇑, Kamil Duda 1, Sławomir Wierzbicki 1 University of Warmia and Mazury in Olsztyn, Faculty of Technical Sciences, 46 A, Słoneczna St. 10-710 Olsztyn, Poland

h i g h l i g h t s
Different blends of self-made swine lard methyl ester and diesel were examined.
The examination included assessment of physicochemical parameters and engine tests.
A CRDI engine operated on 75% ester content biofuel without any issues.
Major reduction of emissions with slight deterioration of efficiency was observed.
Biofuel operation is possible using both single and divided injection strategy.

a r t i c l e

i n f o

Article history: Received 6 March 2015 Received in revised form 29 September 2015 Accepted 29 September 2015 Available online 13 October 2015 Keywords: Compression-ignition Common-rail Biofuels Biodiesel Swine lard methyl esters Exhaust emissions

a b s t r a c t Biodiesel, as a product of the transesterification reaction of fatty material, is gaining appreciation all over the world and confirms its beneficial effect on the reduction of exhaust gases when (even partly) used to power compression-ignition engines. Although many researchers have focused on biofuel production and its application in combustion engines, continuous development of injection systems, providing new control strategies, encourages the seeking for new ways of optimizing biofuel combustion. In particular, very little attention has been given to the possibility of using biofuels in modern CRDI (Common Rail Direct Injection) engines with divided injection technology. There is also limited information available on engines operating on biodiesel of animal origin and their performance and emissions for such fuels. Because of this gap of knowledge, in the present study, the authors focused on analysing the phenomena of combustion of animal-origin biodiesel mixtures in a CRDI engine. Swine lard methyl esters, obtained in the laboratory by a single-step alkali transesterification process as a biocomponent, and mineral diesel were used to obtain B25, B50, B75 mixtures (25%, 50% and 75% of biocomponent concentration by volume). The physicochemical parameters of B25, B50, B75, pure esters and mineral diesel were examined to determine whether the fuels met quality standards. The mixtures were used to fuel a 2.6 L Andoria CRDI engine placed on a dynamometer test stand. Tests were carried out in steady state operation, at rotational speeds when two different injection strategies occur (single injection and two subsequent injections), also different load conditions were introduced during tests. During the tests, engine performance and exhaust gas emissions were measured and analysed in detail. The study has confirmed the capability of using diesel–biodiesel mixtures containing up to 75% biocomponents in a modern CRDI engine without any operational issues. A minor deterioration of fuel performance parameters with an increasing biodiesel share has been observed. Brake-specific fuel consumption increased on average by 3.2%, 8.5% and 13.8% for B25, B50 and B75, respectively. An average reduction of brake fuel conversion efficiency was observed, amounting to 1.6%, 4.8% and 7.8% for B25, B50 and B75, respectively. Significant reduction of exhaust gas emissions (excluding NOx) and opacity was also observed in all examined operation conditions. Total hydrocarbon concentration was reduced

Abbreviations: B25–B75, biodiesel with 25–75% SLME admixture; BFCE, brake fuel conversion efficiency; BSFC, brake specific fuel consumption; CA, crank angle degrees; CFFP, cold filter plug point; CI, compression ignition; CR, common rail; CRDI, common rail direct injection; EGO, exhaust gasses opacity; EGR, exhaust gasses recirculation; MD, mineral diesel fuel; PM, particulate matter; PPM, parts per million; PPT, parts per thousand; RPM, revolutions per minute; SI, spark ignition; SLME, swine lard methyl ester; SOC, start of combustion; SOI, start of injection; THC, total hydrocarbons; CO, carbon monoxide; CO2, carbon dioxide; NOx, nitrogen oxides; O2, oxygen. ⇑ Corresponding author. Tel.: +48 89 524 52 11, +48 89 524 51 00. E-mail addresses: [email protected], [email protected] (M. Mikulski). 1 Tel.: +48 89 524 51 00. http://dx.doi.org/10.1016/j.fuel.2015.09.083 0016-2361/Ó 2015 Elsevier Ltd. All rights reserved.

207

M. Mikulski et al. / Fuel 164 (2016) 206–219

Nomenclature A G HRR N P Pe Q T Te

Subscripts air air fuel fuel inj injection id ignitron delay avg average r relative

crank shaft rotation angle (CA) mass flow rate (kg/h) heat release rate (J/CA) engine rotational speed (RPM) in cylinder pressure (bar) engine power (kW) heating value (J/kg) temperature (K) engine torque (Nm)

by a maximum of 72% for the B75 mixture for a speed of 1500 RPM and 100 Nm load. The best emission performance was observed for operation conditions when a short pre-injection occurred early in the compression phase, before the main fuel injection. This has proven that advanced injection strategies can be applied to fuel mixtures with high biodiesel share, especially for low engine load conditions. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction Biodiesel is gaining immense popularity as an environmentally friendly source of engine fuel and as an alternative to fossil fuels. The current trend of acquiring increasing portions of energy from renewable sources is attributable to the depleting mineral resources, which, regardless of media speculation on the alleged abundance thereof, will eventually run out. Furthermore, the rising pro-environmental attitudes and environmental awareness of societies synergistically increase the demand for renewable fuels. The injection systems that are currently used in compressionignition engines are optimized for running on diesel fuel. If fuel of different properties is used, a number of operating issues occur [1,2]. Uneven spraying of the fuel jet, imperfect combustion, congestion of injector nozzles, damage to fuel pumps, carbon fouling in cylinders are only some of the issues observed during the study of use of unprocessed fat as fuel in compression-ignition engines. Properties such as: high viscosity, density and poor lowtemperature capabilities disqualify this material as an engine fuel. Kleinova et al. [3] confirm in their study that using unprocessed fat material to fuel a compression-ignition engine is not possible without considerable modification of the engine’s injection system. To optimize the properties of fat material for engine application, a number of thermo- and physicochemical methods have been developed and are being successfully applied [4]. The objective of these methods is to optimize the properties of fat to make them eligible for the injection system without any modifications thereof. These methods include: decomposition at elevated temperature (pyrolysis), emulsification, mixing and transesterification. Of the methods listed above, the most promising results are yielded by transesterification [5] – a series of reversible reactions, wherein ester and glycerol particles are formed from triglycerides and alcohol. Thus far, a number of methods for conducting the

transesterification process have been developed, each yielding a different final product in terms of quality, quantity or economical properties [6–10]. Esters acquired with the use of various methods or raw materials may differ considerably in terms of properties, which are affected by a large variety of factors [11]. One of the key factors determining the physicochemical properties of biocomponents is the type of raw material used to produce the component. Fat material used in the production process is composed of fatty acids of different degrees of saturation, mixed in different proportions. The proportions determine the crucial physicochemical properties of the obtained fuel [4–12]. Table 1 presents the profiles of fatty acids of various raw materials used to produce biofuels. Despite the different properties, all types of biocomponents are capable of being mixed in any proportions, either with each other or with diesel fuel. This makes them eligible as standalone fuels, or as mixtures of different esters and diesel fuel (biodiesel). The physicochemical properties of biocomponents obtained by transesterification are comparable to those of diesel fuel, but their renewable origin, atoxicity, biodegradability and high durability favour widespread application. Biofuel can be obtained both from edible and inedible plants (for example: rapeseed oil, soybean oil, cotton seed oil, jatropha seed oil, palm oil, etc.) and from animal fat (beef tallow, swine lard, yellow grease, fish oil, etc.). Fuels can also be produced from fats generated as by-products of processing (used frying oils, waste animal fat or edible fat noncompliant with food industry standards). Biofuel production should be economically substantiated, as the product should compete with conventional fuels in terms of price. Therefore, all stages of production of biofuel, as well as the choice of methods and raw materials, should be analysed in economic terms. In consideration thereof, the materials for biodiesel on an industrial scale should be sourced from inedible and waste

Table 1 Comparison of fatty acids in rapeseed oil, swine lard and beef tallow [12,13]. Material

Rapeseed oil Swine lard Beef tallow

Fatty acid profile (%) Myristic (Tetradecanoic) C14

Palmitic (Hexadecanoic) C16

Stearic (n- Octadecanoic) C18

Oleic (C18:1)

Linoleic (C18:2)

Linolenic (C18:3)

1–2 2–3

4.9 28–30 25–30

1.6 12–18 21–26

33 40–50 39–42

20.4 7–13 2

7.4 0–1

; 1 cylinder, direct injection, naturally aspirated Beef tallow methyl esters blends B100, B75, B50, B25, B5 John Panneer Selvam (2012) [15]

Cengiz Öner [27]

Barrios [26]

Marine fish-oil methyl esters B100, waste cooking-oil biodiesel Ethyl ester of fish oil blends B100 B80 B60 B40 B20 Animal fat/soy biodiesel bnlens B50, B40, B30, B25, B20, B10 Beef tallow methyl ester blends B50, B20, B5 Cherng-Yuan Lin (2009) [16] Sakthivel [25]

4-cylinder, direct injection, turbocharged, egr (euro 4) 1 cylinder, direct injection

Constatnt torque, variable speed Variable torque, constant speed Variable torque and speed Variable torque and speed Variable torque, constant speed

"

" Low RPM ; high RPM ; "

;

Best 30% animal biodiesel – diesel mixture Best B20 N/A "

;

Decreased NOx emissions N/A

;

" low RPM ; high RPM " "

;

Other Emissions Fuel efficiency BSFC

4 cylinder, direct-injection, naturally aspirated 1 cylinder, air cooled

Test type Engine Type Material Author

materials [14]. Such materials include waste from meat processing and fat obtained in these processes. Using fat material considered to be waste raises no controversy on unethical applications of food sources, but makes it possible to, at least partially, satisfy the demand for renewable energy, thus enabling environmentally friendly disposal of material that is considered to be a waste product. The idea of diversification of energy sources, independence from fossil fuels and environmental considerations contribute to the rising demand for biofuels, encouraging researchers to broaden their knowledge on the potential application of biofuel in engines. This is particularly visible in the case of compression-ignition engines. The researchers agree that compression-ignition engines can be fuelled with mixtures of renewable fuels, without any modification of the engine’s injection system [3,15,16]. Numerous studies have confirmed that fuel consumption in a compressionignition engine increases with the increase of the biocomponent portion in the fuel; at the same time, an improvement in the engine’s thermodynamic performance is observed [17–22]. In the case of minor additions of the biocomponent (<20%), several authors have noted a reduction in fuel consumption [23,24]. A majority of authors have reported significantly lower emissions from engines powered by biofuels [10,12,16]. The characteristics of operation of an engine fuelled by pure esters or biodiesel are determined by a variety factors. Parameters such as: engine displacement, fuel injection system, charging pressure, quality and temperature of air cause the test results for specific fuels to differ considerably, depending on the object or methodology. The previously mentioned characteristics of the fat material, the efficiency of the transesterification process and the quality of obtained biofuels also have had an impact on the results of the analysed studies. This necessitates further research to confirm the impact of the given type of fuel on the engine’s performance, depending on its structure and the choice of control parameters. Leading energy journals frequently publish papers on bio-fuelled engine performance and emissions. It bears noting that most authors focus on the use of biofuels of plant origin [3,4,6,10–13]. Studies on the use of fuels of animal origin are significantly less frequent [15,16,25–28,39]. Table 2 summarizes the results of the most recent research on the latter materials. Cherng et al. [16] have assessed the characteristics of operation of a single-cylinder compression-ignition engine with direct injection running on different fuels. The authors used self-produced marine fish-oil esters and industrially produced waste cookingoil esters. The authors noted an increase of fuel consumption with the increase of the engine’s rotational speed. Diesel fuel was characterized by the highest increments of this parameter as compared to the analysed biofuels. For rotational speeds below 1400 RPM, better performance was observed for biodiesel than for diesel fuel. For higher speeds, engine performance was highest for diesel fuel. Furthermore, biodiesel also considerably reduced the emissions of NOx and CO, as well as the levels of exhaust temperature and smoke. Sakthivel et al. [25] have analysed a single-cylinder, air-cooled compression-ignition engine. Tests were conducted at a constant rotational speed, fuelling the engine with a mixture of fish oil ethyl esters and diesel fuel in different proportions. It was observed that with the increase of the biocomponent portion in the fuel, the self-ignition delay period was reduced. At the same time, a reduction of the combustion process duration was observed. Emissions of THC, CO and NOx decreased, while emissions of CO2 and smoke levels increased. Barrios et al. [26] have analysed mixtures of diesel fuel with biofuel of animal and plant origin in different proportions. Biodiesel of animal origin had been obtained from a mixture of swine lard, beef tallow and chicken fat, while plant biodiesel was obtained from soybean oil. The obtained mixtures were tested on a 2.0 L TDI (Turbo Direct Injection) engine. The tests were conducted

Results "-improvement ; – deterioration

M. Mikulski et al. / Fuel 164 (2016) 206–219

Table 2 Comparison of results of studies on biofuels obtained from animal fats [15,16,25–27].

208

M. Mikulski et al. / Fuel 164 (2016) 206–219

according to the UDC (Urban Driving Cycle) and included analysis of emissions. Following the analysis of results, it has been determined that the mixture that appears to provide the best balance between reduction of emissions and maintaining effective parameters and fuel consumption levels is a 30% mixture of biocomponent of animal origin with diesel fuel. Cenzig et al. [27] have analysed the obtained pure animal tallow methyl esters and mixtures thereof with diesel fuel in 5%, 20% and 50% proportions. Tests were conducted on a single-cylinder, air air-cooled, naturally aspirated compression-ignition engine with direct fuel injection. It was observed that the biodiesel admixture decreased the engine’s thermal efficiency, increased specific fuel consumption and decreased exhaust gas emissions. The lowest emissions of CO2 and NOx were observed for a mixture containing 20% of biocomponent. Selvam and Vadivel [15] have assessed the capability of using methyl esters from animal fats as fuel for compression-ignition engines. Esters obtained by single-step alkaline transesterification from beef tallow and methanol were mixed with diesel fuel in 5%, 25%, 50% and 75% proportions. Tests were carried out on a singlecylinder naturally aspirated diesel engine with direct fuel injection. Following the analysis of results, the authors noted that both for the pure ester and for mixtures, the BSFC was higher than for diesel fuel. Emissions of NOx increased proportionally to the volume of biocomponent admixture, while emissions of CO and HC were inversely proportional. The authors have also confirmed that methyl esters of beef tallow and the biodiesel produced from those esters can be used as fuel for a diesel engine without any alterations of the structure thereof. The development of electronic injection systems in recent years has considerably facilitated precise regulation of injection parameters. Injection characteristics can now be arranged without limitations. In the electronic Common Rail (CR) injection systems that are commonly used nowadays, it is possible to adjust the pressure of the injection and the moment of opening of the injector [29–32]. Furthermore, CR systems also have the option of dividing the fuel into several doses, injected in sequence, depending on the current conditions of engine operation. In such systems, usually a very small portion of diesel (pilot dose) is injected early in the compression phase to improve the conditions for the main dose combustion, which is injected some time after the ignition of the pilot dose. Divided injection enables control over the combustion process, reduces toxic emissions and noise levels [29,33]. The summary in Table 2 indicates that the combustion of biofuels of animal origin with divided fuel injection has not yet been studied and assessed. Given the insufficient data on fuelling modern CRDI engines with biofuels of animal origin, in particular swine lard, the authors of this study have undertaken analysis of the key physicochemical properties of biofuel obtained from swine lard and its mixtures with diesel fuel. The obtained mixtures were analysed in terms of the combustion process in the CRDI engine and emissions of toxic compounds.

2. Materials and methods 2.1. Biodiesel production process Swine lard methyl esters were obtained by single-step transesterification of the fatty material with methyl alcohol in the presence of an alkaline catalyst. The following materials were used in the production process:
Edible quality lard, in packages of 500 g, with an acid value of 0.61 mg KOH/g.
P.A. grade (99.8%) methyl alcohol, supplied by Chempur.
85% purity potassium hydroxide, supplied by POCH BASIC.

209

Transesterification was performed in the laboratory setup illustrated in Fig. 1a. The reactor was set up on the basis of a threeneck, round-bottom flask of 1000 cm3 capacity, placed in a heating unit with a magnetic mixer. The heating power and mixing speed were adjusted. The Alihn’s condenser installed in the flask was used to cool down and return the alcohol vapours. The process was conducted for 60 min. at 60 °C in atmospheric pressure. On the basis of available literature [34] and the results of initial research [35,36], the authors have proposed the following composition of reagents: alcohol in a molar ratio of 6:1 to the fatty material, and catalyst in the amount of 0.9% w/w of the fatty material weight. During the reaction, ca. 600 g of fatty material was fed into the flask. Next, to eliminate humidity, the material was heated up to 105 °C, and the temperature was maintained at this level for 10 min. After cooling down to 60 °C, a mixture of the catalyst and alcohol was added to the fat. When the components were mixed, the process of transesterification was initiated. To produce the desired volume of esters, the transesterification reaction was repeated 18 times, processing ca. 10 kg of pure lard. Costs of production of 1 dm3 of biodiesel produced with laboratory grade reagents and edible quality lard was about three times higher than cost of mineral diesel fuel (according to diesel retail price 1.1 €/dm3), including cost of reagents. It was turn out that the fatty material was the most expensive and represented 55% of the total costs. The costs of media (water, energy, etc) used to biodiesel production process constituted 10% of the total costs. It is believed that if the production was carried on a larger scale, with the use of inedible quality fatty material and alcohol purchased in bulk quantities, the total costs of production could be easily reduced. To purify the obtained esters, the excess alcohol removal process was applied. For this purpose, the apparatus illustrated in Fig. 1b was used. The post-reaction mixture was fed into the flask and heated up to ca. 65 °C, at which alcohol condensate formation was observed in the distillation column. Pressure in the system was then reduced gradually to accelerate the alcohol evaporation process. The pressure was reduced until alcohol condensate was no longer visible in the distillation column condenser. After alcohol was collected, the post-reaction mixture was transferred to a laboratory phase separator, where glycerine was then separated from the ester phase. The ester phase was additionally filtered to remove any solid impurities. The resulting purified bioester was then used to produce mixtures for analysis. The mixtures were obtained by mixing bioester with MD purchased at a gas station, in proportions as provided in Table 3. 2.2. Fuel quality assessment methods The obtained fuels were analysed to assay the physicochemical characteristics. A comprehensive list of established parameters, along with methods and the measurement uncertainty level, are presented in Table 4. Assays were performed in accordance with procedures set forth in the PN-EN 14214 standard for esters and mixtures with MD, and EN 590 for MD. To determine the appropriate method, the volume of catalyst and process parameters for transesterification, the acid value of the fatty material, from which esters had been produced, was also assayed. 2.3. Test stand and methods for engine test bench measurements The engine tests were conducted on a four-cylinder ADCR CI engine, manufactured by Andoria-Mot (Table 5). The engine had a standard CR 2.0 injection system developed by Bosch and was controlled by an EDC16C39 controller. Depending on the engine’s

210

M. Mikulski et al. / Fuel 164 (2016) 206–219

Fig. 1. Experimental set-up used for the transesterification process (a), and for alcohol removal process (b).

Table 3 Volume ratio of Mineral Diesel (MD) and swine lard methyl esters (SLME) in mixtures. Mixture name

Biodiesel share (%)

B 75 B 50 B 25

75 50 25

Partial content MD

SLME

1 1 3

3 1 1

Table 4 List of measured biodiesel parameters, with method and measurement uncertainty level. Property

Test-method

Uncertainty

Unit

Density at 15 °C Viscosity at 40 °C Flash point Sulphur content Total contamination Oxidation stability Acid value Cold filter plug point

Pycnometric method PN-EN ISO 3104 PN-EN ISO 3679 PN-EN ISO 20884 PN-EN 12662 PN-EN 14112 PN-EN 14104 PN-EN 116

±0.12 ±0.008 ±0.5 ±0.4 ±0.03 ±0.1 ±0.003 ±1

kg/m3 mm2/s °C ppm mg/kg h mg KOH/g °C

Table 5 Technical data of the ACDR engine. Engine

ADCR

Type

diesel, 4-stroke, turbocharged with intercooler Common Rail fuel accumulator system 4 cylinder inline, vertical 94/95 mm 2636 cm3 17.5:1 85 kW/3700 RPM 250 Nm/1800–2200 RPM 750 RPM 210 g/kW h Accumulator injection system (Common Rail) CR2.0 Radial, with exhaust extraction valve Pneumatic EGR valve with exhaust cooler

Fuel injection Engine layout Cylinder diameter/piston travel Piston displacement volume Compression ratio Rated Power/rotational speed Max. Torque/rotational speed Min. Idle rotational speed Fuel consumption at torque peak Injection system (Bosch) Turbocharger EGR system

operating conditions, the controller carries out two different fuel injection strategies. At low RPM and under small loads in the medium speed range, divided injection is performed. In the remaining range, a single fuel charge is injected. The engine was installed on a test bed at the Department of Mechatronics and IT Education of the University of Warmia and Mazury in Olsztyn. The test stand in the configuration used in the present research consisted of the following elements:










Eddy-current dynometer AVL DP 240. Engine speed control system THA100. Fuel balance AVL 735S. Air mass flow meter SENSYFLOW P. A set of PT100 sensors (cooling water, lube oil, intake air). In-cylinder pressure measurement system (KISTLER). Opacimeter AVL 439. Emission bench AVL AMA i60. Test stand control and data acquisition system PUMA Open.

A more detailed description of the test stand can be found in a previous report by the authors [37]. Engine tests were conducted for two rotational speeds: 1500 RPM – at which the engine controller injects two diesel fuel charges (pilot and main charge). 3000 RPM – at which the controller injects a single fuel charge regardless of the load. At the given speed, tests were conducted for different loads (50 Nm, 100 Nm, 150 Nm) and proportions of biofuel admixtures to mineral diesel (Table 3). The scope of the performed tests is illustrated in Table 8. Before each test of the given fuel, in order to ensure the reliability of the results, the fuel system, as well as the exhaust end emission paths, were purified and flushed. The test material was changed by connecting the tank with the given test fuel. The engine was then allowed to run on the new fuel with the injector pump return hose disconnected. This continued until approx. 8 L of fuel were purged into the separate container. With the total

211

M. Mikulski et al. / Fuel 164 (2016) 206–219

capacity of the fuel system hosing of 2 L, the volume was more than sufficient to ensure that the appropriate fuel was burned. At the same time, the exhaust and emission paths were flushed with a high flux of pressurized air through the exhaust lines. The flushing was performed through a special procedure of the emission test bench. For each test run, engine rotational speed and torque were stabilized, with an accuracy of ±5 RPM and ±2 Nm, respectively. The cooling water and lube oil temperatures were kept at a constant 85/95 °C ± 1 °C. After stabilization, steady state measurements were performed, including basic operation parameters of the engine, in-cylinder pressure, injector coil current, opacity and emissions. An identical time period of 120 s was set for data acquisition in every test run. A piezoelectric pressure sensor (Type 6056A by Kistler), installed in one of the cylinders instead of the heater plug, was used for recording of pressure. The sensor, combined with a Type 5018A charge amplifier, was connected via a DAQ card to a PC. The software for acquisition of pressure measurement results was compiled in the National Instruments LabVIEW environment. The association of dynamic pressure signal with respective rotation angle values was provided by an optical encoder mounted on the crankshaft. The angle-marking gauge resolution was 720 points/ revolution, which enabled pressure recording every 1degree of crankshaft revolution, in the full range of the engine’s work cycle. For each measurement point, after stabilization of the engine’s operation parameters, pressure versus crank angle (a) for 100 cycles was recorded. Pressure measurement results were then cycle-averaged, giving pavg (a). Standard deviation was used to calculate the average error of pressure measurement for each crank angle Dpavg (a). The average relative error was then calculated:

Dpav g r ðaÞ ¼

Dpav g r ðaÞ  100 pav g r ðaÞ

The injector coil current signal was analysed in the same way as described for pressure signal and was used to determine the start of injection angle. The mass of air aspired by the engine (Gair), fuel consumption (Gfuel) and air temperature at the inlet manifold (Tair) were recorded during the tests, and time averaged for each test run. Standard deviation was used to calculate the accuracy of the measurements results. A similar methodology was used to determine the measurement uncertainty of concentration levels of exhaust components. With the use of the measured values, a number of synthetic parameters were calculated to describe combustion performance. The angle of start of combustion was determined on the basis of analysis of second pressure derivative changes [38] and used to calculate the ignition delay angle (Aid). Maximum pressure rise acceleration indicated the ignition point. Brake-specific fuel consumption and brake-fuel conversion efficiency were calculated according to the following formulas:

BSFC ¼

Gfuel Pe

ð2Þ

BFCE ¼

3600 BSFC  Q fuel

ð3Þ

where Pe – engine power, and Qfuel – gross fuel heating value. For the calculated values, the measurement uncertainty was calculated using Eq. (4):

ð1Þ

as an estimate of the engine’s operation repeatability. For recorded in-cylinder pressure traces, Heat Release Rate was calculated using the first law of thermodynamics and equation of state. The calculations were done using constant specific heat ratio of 1.36 and corrected for cylinder heat loss with standard Hohenberg correlation. Additionally, to determine the moment of injection, a current clamp was mounted on the injector of the indicated cylinder, which allowed recording the current changes on the injector coil.

"


DY ¼

dY Dx1 dx1

2




dY þ  þ Dxn dxn

2 #12 ð4Þ

where Y and DY stand for the calculated variable and the measurement uncertainty thereof, and x1 . . . xn stand for independent (directly measured) inputs used for calculating Y. The list of parameters recorded directly or indirectly during the test runs, along with maximum uncertainty for all measurement points, is presented in Table 6.

Table 6 List of parameters recorded directly and derived indirectly from calculation, along with achieved maximum uncertainty. No

a

Symbol

Measurement device

Uncertainty

Unit

Directly measured 1 Engine rotational speed 2 Torque 3 Generated power 4 Air aspired to the engine 5 Fuel consumption 6 Intake air temperature 7 Start of injection angle 8 Total hydrocarbons a 9 Total nitrogen oxides a 10 Carbon monoxide a 11 Carbon dioxide a 12 Oxygen a 13 Opacity

Parameter

N Te Pe Gair Gfuel Tair SOI THC NOX CO CO2 O2 EGO

AVL DP 240

AVL 439

±5 ±2 ±0.2 ±0.5 ±0.1 ±0.2 ±0.5 ±11 ±19 ±13 ±1 ±1 ±0.9

RPM Nm kW kg/h kg/h K CA ppm ppm ppm ppt ppt %

Calculated 14 15 16 17

HRR SOC BSFC BFCE

First law of thermo-dynamics, equation of state Second derivative of pressure analysis Eq. (1) Eq. (2)

– ±1 ±10 ±1

J/CA CA g/kW h %

Heat release rates Start of combustion angle Brake specific fuel consumption Brake fuel conversion efficiency

Concentration of the compound in the exhaust gases.

SENSYFLOW P AVL 735S PT 100 Current clamp AVL AMA i60

212

M. Mikulski et al. / Fuel 164 (2016) 206–219

Table 7 Characteristics of pure SLME, MD, and mixtures. Sample

Density at 15 °C (kg/m3)

Viscosity at 40 °C (mm2/s)

Flash point (°C)

Sulphur content (mg/kg)

Total contamination (mg/kg)

Oxidation stablility (h)

Acid value (mg KOH/g)

CFPP (°C)

Calorific value (MJ/kg)

B 100 B 75 B 50 B 25 MD

895.49 879.79 869.59 857.09 844.49

4.4528 3.924 3.5937 3.168 2.8798

134 80 71 63 58

1.44 2.31 2.74 3.7 4.09

22.5 17.6 11 8.2 5

1.32 6.64 8.22 >19 >19

0.14 0.13 0.1 0.1 0.11

13 8 4 1 1

40.1 40.75 41.4 42.05 42.6

3. Results and discussion

Table 8 Scope of the performed tests with the basic engine operating parameters.

3.1. Fuel quality measurements results The biodiesel mixtures were tested in regard to the most important physicochemical characteristics, in terms of eligibility for use in fuel CI engines. Table 7 presents the compiled results of the performed analyses for MD, esters and their mixtures. The data presented in Table 7 show that only pure esters fail to meet the requirement set by the PN-EN 14112 standard, concerning oxidation stability, which should be at least 6 h. Also, the cold filter plug point of 13 °C for pure esters may raise objections, as it is a very high value. For commercial fuel, purchased at a filling station, this temperature was 1 °C. Parameters such as density and viscosity for pure ester, though within the normal range, are very high, and it is possible that, had the esters not been prepared under laboratory conditions, they might not have met these requirements. Despite the measures taken to remove impurities from the esters, by sedimentation and filtration, the content of impurities in the pure ester sample remained very high, which forces the authors of the paper to seek more efficient methods of purification of both fatty material and the finished product. The low sulphur content indicates a lack of absorption of this element into the fatty raw material used for ester production. The sulphur content in the mixtures rose gradually with an increasing MD percentage, but it did not exceed the maximum value recommended for the content of this element. The flash point for pure esters at 134 °C indicates that excess alcohol collection was conducted correctly; the flash point decreases with an increasing MD percentage. For pure MD, the flash point was 58 °C, which was consistent with the PN-EN 590 standard that sets out the research methodology and the values of characteristics for MD. The very low acid value, both for pure lard at 0.61 mg KOH/g and for the esters, bears noting, as it may indicate a low degree of fatty material hydrolysis, which would be a considerable advantage. The esters did not undergo thorough purification from catalyst residues after production, which could distort the acid value test result, but because of the low value of this characteristic for pure lard, the authors assume that its value would not have been exceeded for pure esters, even after allowing for possible distortions of the results caused by catalyst residue in the final product. Heating values presented in Table 7 refers to higher heating value parameter. The values for pure esters and mineral diesel were obtained from literature [13,14]. Values for admixtures have been approximated in accordance with percentage share of each component in the fuel mixture. 3.2. Engine test bed measurements results The scope of the performed tests with basic engine operating parameters is provided in Table 8. The recorded cylinder pressure changes and selected injector excitation current changes, and standard deviation changes for

Fig. 2. Average recorded pressure versus CA for 1500 RPM (divided injection), for different loads and biodiesel blends.

the pressure signal under individual cycles, as the measure of engine operation unrepeatability, are presented below. From the analysis of the current changes in the injector coil, it can be concluded that, regardless of the fuel used, the engine controller would employ the same fuelling strategy for individual working points. Sample changes are presented in (Figs. 8 and 9).

M. Mikulski et al. / Fuel 164 (2016) 206–219

213

For a speed of 1500 RPM, a small fuel charge was applied as the preliminary injection to prepare the conditions for the combustion of the main charge injected near the TDC. For a speed of 3000 RPM, only the main injection was performed. The factory-mounted engine controller did not respond to the fuel change with an observable individual charge injection angle displacement for any working point. This fact also provides a comparative assessment of the combustion process itself (Figs. 2 and 5). For all the measurement samples, except for N = 1500 RPM, T = 50 Nm (Fig. 2), increasing the swine lard methyl ester percentage weight in the fuel generated higher maximum pressure values for individual working points. This indicates that an admixture of the tested esters to diesel fuel accelerates fuel combustion, thus generating higher heat release rates and, consequently, higher maximum pressure. The Te = 50 Nm case appears to confirm the general observations made for 1500 RPM, with a visible pressure rise in the main charge combustion phase for fuels with ester percentages of 50% and 75%. In this case, globally lower pressure values recorded for these fuels, compared to pure diesel fuel and a 25% ester admixture, resulted from a reduced degree of supercharging, which could be easily observed in the comparison of the respective air consumption values (Table 8). This was most likely the controller’s response to an excessively rapid combustion of mixtures with high ester content for similar in-cylinder pressure conditions. On a smaller scale, similar engine controller behaviour can be observed for 3000 RPM case, by analysing Fig. 5. The results of pressure plots analysis, for 1500 RPM/50 Nm case, find confirmation in the Heat Release Rate plot (Fig. 3), showing higher maximum values of this parameter and faster ignition for ester-rich fuel blends – despite significantly lower pressures during compression. Fig. 4 shows clearly that the combustion took place in two stages correlated with two subsequent injections. In the first stage only about 5% of total energy was released as a preparation for combustion of the main dose combustion. Higher HRR for biodiesel blends visible in Fig. 4 result in higher in-cylinder pressures. The similar behaviour was observed for 1500 RPM/150 Nm case. For 3000 RPM (single injection) the significantly faster combustion of ester-rich blends was observed for the lowest load conditions (Figs. 5 and 6). For other cases, the same trend was observed, but the differences in combustion processes, between tested fuels, were much less significant (Figs. 5 and 7). The use of animal fat admixtures clearly improves the fuel’s self-ignition capabilities (Fig. 11). For a speed of 1500 RPM and a load of 50 Nm, the fuel self-ignites 1degree of crankshaft revolution earlier than diesel fuel for a 75% animal ester mixture, despite

considerably lower cylinder temperatures and pressure at the time of injection (resulting from the reduced degree of turbocharging). Analogically, when similar pressure and temperature values are reached in the compression process (Fig. 5), the reduction of

Fig. 3. Average Heat Release Rates (HRR) versus CA for 1500 RPM (divided injection), for Te = 50 Nm and different biodiesel blends.

Fig. 6. Average Heat Release Rates (HRR) versus CA for 3000 RPM (single injection), for Te = 50 Nm and different biodiesel blends.

Fig. 4. Average Heat Release Rates (HRR) versus CA for 1500 RPM (divided injection), for Te = 100 Nm and different biodiesel blends.

Fig. 5. Average recorded pressure versus CA for 3000 RPM (single injection), for different loads and biodiesel blends.

214

M. Mikulski et al. / Fuel 164 (2016) 206–219

Fig. 7. Average Heat Release Rates (HRR) versus CA for 3000 RPM (single injection), for Te = 100 Nm and different biodiesel blends.

self-ignition delay relative to pure diesel fuel can reach 5degree of crankshaft revolution for a 75% ester admixture. Operation repeatability was similar for all samples (Fig. 10). No tendency for uneven combustion, mistiming or knocking combustion was observed. Engine operation was stable even when fuels with 75% of animal fat esters were used. 3.3. Engine performance analysis results Performance analysis was discussed on the basis of two calculated parameters, namely brake-specific fuel consumption and brake-fuel conversion efficiency. The obtained results of the parameters are presented in Figs. 12 and 13. The results indicate a general tendency towards reducing fuel consumption and improving fuel efficiency, along with the increasing engine load, for all analysed mixtures and at both rotational speeds. The tendency is understandable given the increased portion of power lost on the engine’s internal friction in the power generated by the engine under smaller loads. For a rotational speed of 1500 RPM, along with the increase of load, the increments of BSFC for specific fuels are reduced, while for a rotational speed of 3000 RPM, the increments are maintained at more or less the same level. For individual measurements, an increase of fuel consumption can be observed, proportional to the percentage of the biocomponent in the fuel mixture. Among the analysed fuels, the lowest fuel consumption was observed for MD, and the highest for the mixture with the highest biocomponent percentage. Brake-specific fuel consumption of MD was lower on average by 3.2%, 8.5% and 13.8% compared to B25, B50 and B75, respectively (Fig. 12). These correlations are visible for all analysed loads and engine rotational speeds. The results confirm the reports by other authors who have researched engines with traditional injection systems [15,27]. The reason for this is the fact that biodiesels’ calorific value are lower than that of MD and the response of the controller that reduced the degree of charging of the engine (Table 8). Fuel efficiency values (BFCE) closest to those of MD are observed for the B25 mixture, with the measurement results of the latter differing from the former by 1.6% on average. Fuel efficiency of the B50 mixture is lower by 4.8% on average compared to MD. The largest difference in comparison to MD is observed for the B75 mixture – 7.8% on average, not exceeding 13% at the outside, for all analysed loads and rotational speeds. Despite the fact that differences in fuel efficiency values between MD and B25 are below the margin of error, a clear downwards trend of BFCE is observable with the increase of biodiesel percentage in the fuel mixture.

In the case of an engine rotational speed of 1500 RPM, BFCE values of the B25 and B50 mixtures approach the value of MD with increasing engine load. For a rotational speed of 3000 RPM, a reverse phenomenon is observed, as fuel efficiency values are more uniform for smaller loads, with the largest discrepancies observed for the highest analysed engine load. In the analysed engine, fuel is injected into the cylinder in a sequence over a single work cycle. Figs. 8 and 9, illustrating the changes of the current on injector coils, clearly indicate that for a rotational speed of 1500 RPM, fuel dosage into the combustion chamber was divided into the pilot dose and the main dose. For rotational speeds of 3000 RPM, fuel is injected into the cylinders in a single dose per work cycle. The concept of dividing the fuel dose into the pilot and main doses is aimed at improving the conditions of combustion of the main fuel dose. This is particularly visible for small engine loads. At higher rotational speeds, the pilot dose is not applied, and the combustion process of the fuel mixture takes less time. The absence of the initial dose and shorter time for release of energy from the fuel result in lower BFCE values for tests conducted at higher rotational speed. The drop of fuel efficiency values as well as increase of BSFC, proportional to the percentage of biocomponent, can be explained by heating value of fuel mixtures. The biodiesel mixtures are characterized by lower heating values compared to MD (Table 7). The other possible explanation of increase of fuel consumption can be reduction of ignition delay of pure biodiesel. Fuel mixtures with higher biocomponent content are characterized by shorter ignition delay, which results in earlier ignition of the fuel mixture causing loss of energy.

Fig. 8. Average injector coil current changes for 1500 RPM (divided injection), for mineral diesel at different engine loads.

Fig. 9. Average injector coil current changes for 3000 RPM (single injection), for mineral diesel at different engine loads.

M. Mikulski et al. / Fuel 164 (2016) 206–219

215

The divided injection strategy introduced by engine’s controller complies with combustion of the tested biofuel under high engine loads. Under a load of 150 Nm, at 1500 RPM, the difference of fuel efficiency for running on B75 fuel and MD did not exceed 13%. For B25 fuel, the differences slightly exceed the margin of measurement error. It appears that optimization of the engine charging strategy and dividing the fuel dose according to the given biofuel would also yield similar results for smaller loads. 3.4. Engine emission analysis results The emission results for subsequent operation points are shown in Figs. 14 and 15. A significant reduction of exhaust gas emissions and opacity was observed (excluding NOx) under all examined operating conditions. Higher levels of reduction were observed under lower RPM range when the fuel dose was divided into a pilot and main dose. This was confirmed by the decrease of CO and HC concentration level in the detailed exhaust analysis. These two components were considerably lower for esters admixtures as compared to mineral diesel. Moreover exhaust gasses opacity decreased along with the increasing percentage of esters in the mixture used to power the engine. This was probably due to the higher Heat Release Rates observed for the admixtures of the tested biodiesel, which lead to more complete combustion. For the B75 mixture, the maximum reduction of exhaust gas opacity, hydrocarbons and carbon monoxide, but also the highest increase of concentration of nitrogen oxides, was observed under different states of operation. Also, the test proved that dividing the fuel dose has a certain improvement in combustion conditions, especially under small engine loads, for fuel mixtures with high biodiesel concentration levels. A detailed analysis of emissions of specific exhaust gases is provided below. 3.4.1. Hydrocarbons emissions Hydrocarbons emissions were reduced with the increasing engine load for all fuels. High HC emissions at small loads were caused by overleaning the fuel/air equivalence ratio in the fuel spray. For low volumes of injected fuel, the volume of fuel in the spray that was mixed below the combustion limit, which was higher than for larger fuel volumes, correlated with the increasing load. In the case of the analysed B25–B75 mixtures, the reduction of THC emissions as compared to MD was observed for all measurement points. It was noted that reduction of THC emissions corresponded proportionally to the percentage of the biocomponent in the mixture. This general tendency can be explained by three parallel phenomena. Firstly, due to the lower calorific value of the mixtures, a larger volume of biofuel must be injected under the same operating conditions than for MD (confirmed by the results of analysis of BSFC). Secondly, with the increase of the percentage of the biocomponent, the self-ignition delay is reduced (Fig. 11). Both the higher volume of injected fuel (in the case of ester admixtures) and shorter time of residence in the cylinder contribute to mitigating the overleaning effect. In terms of the impact of ignition delay, similar conclusions were drawn inter alia by Sakthivel et al. [25]. Thirdly, as some authors report [38], the observed correlation may arise from the presence of molecular oxygen in the biodiesel particles, which improves combustion of the fuel mixture. In terms of reduction of THC emissions for loads of 100 Nm and 150 Nm, a visible convergence of results is observed between specific mixtures, with the difference of results amounting on average to 0.5% and not exceeding 1.5%. The highest reduction level was noted for the B75 mixture at an engine rotational speed of 1500 RPM and under loads of 100 Nm

Fig. 10. Comparison of cylinder pressure signal standard deviation changes for maximum load. B75 and MD have been compiled for different injection strategies.

Fig. 11. Ignition delay angle for different fuels and all test conditions.

and 150 Nm; in these conditions, THC emissions were reduced by more than 72% as compared to diesel fuel. For the same mixture, under identical loads but at a rotational speed of 3000 RPM, the THC emissions reduction was at least 45%. THC emissions reduction by 25% was observed for the B25 mixture at 3000 RPM and under 50 Nm; this was the lowest recorded difference of THC emissions in the study. For these conditions, the B50 and B75 mixtures achieved reductions of 56% and 54%, respectively. An increased reduction of THC emissions was observed for studies conducted at a lower engine rotational speed. This was probably caused by the prolonged process of secondary fuel combustion, arising from the applied strategy of dividing the fuel dose into a pilot dose and main dose. This has been confirmed inter alia by the results obtained by Kousoulidou et al. [40]. 3.4.2. Carbon monoxide emissions In Figs. 14 and 15 above, we can observe a correlation between the operating conditions of the engine and CO emissions. Increase of the engine’s rotational speed results in an increase of CO emissions, while the increasing load of the engine reduces the emissions of this compound; the correlation is visible for all analysed mixtures. The correlation between the reduction of CO emissions and the load increase probably depends on the increasing fuel dose, injected under higher engine loads; a larger fuel dose volume

216

M. Mikulski et al. / Fuel 164 (2016) 206–219

Fig. 12. BSFC results for tested fuels in all test conditions.

Fig. 13. BFCE results for tested fuels in all test conditions.

enables more accurate dispersion in the cylinder space. The reason for an increase of CO emissions with the load drop is thus similar to THC emissions and arises from overleaning of the mixture in the jet of sprayed fuel, which results in imperfect combustion. The significant reduction of CO concentration levels in exhaust gas at a low rotational speed arises from the prolonged combustion (and secondary combustion) process of CO, caused both by the reduction of speed and the applied diversification of fuel injection. In the case of the relation between the biocomponent content in the fuel mixture and CO emissions, an inversely proportional dependence is observed – for mixtures with higher biocomponent content, CO emissions are lower for all analysed states of engine operation. These correlations are in line with the results reported by other authors [15,20]. This phenomenon is explained by the oxygen content in the biofuel particles, which plays a crucial role in CO reduction, facilitating more complete combustion of the fuel mixture. For all conditions, the highest CO reduction with respect to MD (47%) was observed for the B75 mixtures at 1500 RPM and under

100 Nm, while the lowest (16%) was for the B25 mixture at 3000 RPM and under 100 Nm. For an engine rotational speed of 1500 RPM, the lowest reduction level (20%) was observed for the B25 mixture under 50 Nm. On average, a difference of 27%, 38% and 37% was noted for B25, B50 and B75 mixtures, respectively. For a rotational speed of 3000 RPM, the highest reduction of CO emissions was observed for the B75 mixture under 50 Nm, with average reductions of 19%, 25% and 30% for B25, B50 and B75 mixtures, respectively. Reduction of CO and THC levels in exhaust gas can also be explained by the physicochemical properties of the biocomponent. Biodiesel is characterized by higher density and viscosity than MD therefore, the biodiesel particles sprayed during injection are slightly larger and heavier than MD particles. The increased weight of fuel particles causes them to move away from the injector, thus mixing more completely with air and providing sufficient oxygen to burn the fuel. It also bears noting that biodiesel contains minor volumes of water. Ahmad et al. and Ali et al. [41,42] point out that water/oil emulsions favour optimization of the combustion process

M. Mikulski et al. / Fuel 164 (2016) 206–219

217

Fig. 14. Emission measurements results for 1500 RPM at different engine loads, for different fuels; results are given in PPM for THC, NOx and CO; PPT for CO2 and O2; %102 for EGO.

Fig. 15. Emission measurements results for 3000 RPM at different engine loads, for different fuels; results are given in PPM for THC, NOx and CO; PPT for CO2 and O2; %102 for EGO.

and contribute to reducing emissions of imperfect combustion of the fuel mixture. 3.4.3. Carbon dioxide emissions CO2 emissions are related to the efficiency of the combustion process in the combustion chamber. The higher the efficiency of combustion, the higher the number of carbon compounds transformed into CO2. The most prominent changes of CO2 concentration levels were observed for the engine operating under the smallest analysed load. At a rotational speed of 1500 RPM, an increase of CO2 emissions was observed by 0.3%, 18% and 13.5% for B25, B50 and B75 mixtures, respectively. For a rotational speed of 3000 RPM, a reduction of CO2 emissions was observed by 1%, 3% and 6% for B25, B50 and B75 mixtures, respectively. For an engine load of 100 Nm and rotational speed of 1500 RP, a reduction of emissions was observed by 2% 3% and 6% for B25, B50 and B75 mixtures, respectively. The average difference of CO2 emissions for 3000 RPM and the corresponding engine load did not exceed 1%.

For an engine load of 150 Nm and rotational speed of 1500 RPM, the average difference of CO2 emissions is 0.3% for all fuels, while for 3000 RPM, the CO2 emissions were reduced by 2% on average for all mixtures, with a maximum reduction of 4% for B75. 3.4.4. Nitrogen oxides emissions The mechanism of creation of nitrogen oxides intensifies with the increase of temperature in the engine cylinder. The higher the temperature of the combustion process, the higher the concentration levels of the resulting nitrogen oxides. The opening degree of the EGR valve, as well as the volume of oxygen in the cylinder space that may bind to nitrogen, have an impact on NOx emissions [26,28]. Researchers [15,26] also point out the correlation between NOx emissions and the physicochemical properties of fuel used in the engine. The cetane number, the iodine value and the oxygen content in biofuel particles are closely related to NOx emissions during combustion of biofuels. Regardless of the type of engine or test, increased emissions of nitrogen oxides are observed when biofuels are used; this

218

M. Mikulski et al. / Fuel 164 (2016) 206–219

hypothesis is confirmed by the results of the discussed tests, which indicate an increase of NOx emissions proportionally to the percentage of biocomponent for all analysed cases. In the case of the B25 mixture, at 1500 RPM and under 50 Nm, a reduction of emissions by 1.6% was observed; however, such a minor change of value is below the margin of error of measurement. In tests conducted at a rotational speed of 1500 RPM and under 150 Nm, the maximum increase of NOx emissions for the B25 mixture was slightly below 3%. An increase of emissions by 15%, 5.6% and 12.5% were observed for the B50 mixture at 1500 RPM and under 50 Nm, 100 Nm and 150 Nm, respectively. Using the B75 mixture resulted in an increase of NOx emissions by 12%, 10% and 22% at 1500 RPM and under 50 Nm, 100 Nm and 150 Nm, respectively. An increase of emissions by 22% was the highest observed during the tests. For a rotational speed of 3000 RPM, NOx emissions were very similar for all mixtures and did not exceed 5% for the B75 mixture. Despite the fact that the differences at a higher rotational speed are not high, a general upwards tendency in emissions can be observed, depending on the engine load and the percentage of biodiesel in the fuel mixture. The pressure changes in the cylinder are in line with the obtained results of NOx emissions. Higher concentration levels of nitrogen oxide with biofuels are related to the previously observed ignition and accelerated combustion, which creates higher pressure delta and, consequently, higher temperatures in the cylinder. However, this correlation is not the only one, as shown by the results of tests for the operating point of 3000 RPM and 50 Nm. Relatively high differences of maximum combustion pressure, observed for mixtures with higher biocomponent levels, are not reflected by significant delta of nitrogen oxide emissions. This confirms the observation [15,26] that NOx emissions also depend directly on the physicochemical properties of fuel used in the engine. Incidentally, those properties for the tested biodiesel appear to favour reduction of nitrogen oxides, despite generating higher combustion temperatures. 3.4.5. Exhaust gasses opacity As evident from Figs. 14 and 15, reduction of the exhaust gas opacity level correlates with the volume of biocomponent in the fuel mixture. The exhaust gas opacity level is related to PM (particulate matter) emissions; therefore, a reduction of EGO also reduces PM emissions. Ballesteros et al. [43] point out the increase of carbonyl particle emissions when biofuels are burned. Carbonyl particles are considerably smaller than PM and are not included in the analysis of emissions of both PM and EGO - thus, the results of these parameters may not reflect the actual nature of the phenomenon. Selvam and Vadivel [15] suggest that the presence of oxygen and lower carbon content in biodiesel particles compared to MD are factors that determine opacity reduction. Results of studies indicate that using a biocomponent in the fuel mixture reduces the opacity levels by 76% on average. At 1500 RPM, the highest EGO reduction levels were observed for the B75 mixture, which reduced opacity by 73% on average. For a rotational speed of 3000 RPM, using B75 resulted in opacity reduction by 89% on average (max. 91% under 100 Nm). Significant EGO reduction levels with the mixtures used confirm a more complete combustion process of the fuel mixture, which is confirmed by the reduction of other components of exhaust gases. 4. Conclusion In the course of the study, it has been confirmed that animal fat-based biodiesel and diesel mixtures meet the quality standards

for CI engine fuels. It has been demonstrated that it is possible to burn fuels with a very high content (up to 75%) of SLME in modern CRDI engines operating with advanced injection strategies. The engine worked correctly for all working points. No adverse phenomena related to misfiring or knocking combustion were observed. Operating stability was similar to running on standard diesel fuel. It has also been demonstrated that fuel admixed with esters of fatty acids of animal origin is characterized by generally higher heat release rates during combustion. It is expected that this might increase the combustion efficiency for specific fuel compositions, with proper selection of engine control parameters. To confirm this suspicion further, control oriented research are required. It has been shown that the controller of the tested engine responds to the increased combustion rate with a reduced degree of supercharging, which decreases the maximum indicated pressure. This is probably also the reason for lower break-specific efficiency when using biocomponents. For high engine speeds, where the volume of air fed to the engine was similar for all fuels, efficiency differences were negligible. Fuel with the SLME admixture also has better capability to self-ignite, which leads to a considerably reduced self-ignition delay. Furthermore, it has been shown that a significant reduction of harmful exhaust gasses emissions can be obtained (excluding NOx where no improvement or minor increase was noted), while using biodiesel produced from animal material. The general improvement was proportional to the biocomponent share. Particular benefits have been observed for CO and HC emissions, which were reduced at all measurement points. The average reduction of CO and HC concentration levels amounted to 29% and 52%, respectively. Pure esters were not used in this study due to the high level of impurities (Table 7), which might have caused operational issues, leading to damage of the injection system. It appears that, after proper purification processes, it is possible to use 100% diesel fuel substitution with esters of fatty acids of animal origin in modern engines with electronic injection, without modification of the engine’s control programme. Further optimization of injection timing and turbocharger operation for biofuels can lead to engine efficiency on par with diesel fuel, while maintaining the proenvironmental benefits of biocomponent combustion. This bodes well for the possibility of using meat industry waste for biofuel or biocomponent production. The obtained results refer to steady state engine operation. Tests in intermediate states should be conducted for a comprehensive picture of the emission properties of the analysed mixtures. References [1] Sahoo PK, Das LM. Process optimization for biodiesel production from Jatropha, Karanja and Polanga oils. Fuel 2009;88(9):1588–94. http://dx.doi. org/10.1016/j.fuel.2009.02.016. [2] Britto RFJ, Martins CA. Experimental analysis of a diesel engine operating in Diesel–Ethanol Dual-Fuel mode. Fuel 2014;134:140–50. http://dx.doi.org/ 10.1016/j.fuel.2014.05.010. [3] Kleinová A, Vailing I, Lábaj J, Mikulec J, Cvengroš J. Vegetable oils and animal fats as alternative fuels for diesel engines with dual fuel operation. Fuel Process Technol 2011;92(10):1980–6. http://dx.doi.org/10.1016/j.fuproc.2011.05.018. [4] Atabani AE, Silitonga AS, Badruddin IA, Mahlia TMI, Masjuki HH, Mekhilef S. A comprehensive review on biodiesel as an alternative energy resource and its characteristics. Renew Sustain Energy Rev 2012;16(4):2070–93. http://dx.doi. org/10.1016/j.rser.2012.01.003. [5] Lin L, Cunshan Z, Vittayapadung S, Xiangqian S, Mingdong D. Opportunities and challenges for biodiesel fuel. Appl Energy 2011;88(4):1020–31. http://dx. doi.org/10.1016/j.apenergy.2010.09.029. [6] Atadashi IM, Aroua MK, Aziz AA. High quality biodiesel and its diesel engine application: a review. Renew Sustain Energy Rev 2010;14(7):1999–2008. http://dx.doi.org/10.1016/j.rser.2010.03.020. [7] Makareviciene V, Matijosius J, Pukalskas S. The exploitation and environmental characteristics of diesel fuel containing rapeseed butyl esters. Transport 2013;28(2):158–65. http://dx.doi.org/10.3846/16484142.2013.801364.

M. Mikulski et al. / Fuel 164 (2016) 206–219 [8] Abbaszaadeh A, Ghobadian B, Omidkhah MR, Najafi G. Current biodiesel production technologies: a comparative review. Energy Convers Manage 2012;63:138–48. http://dx.doi.org/10.1016/j.enconman.2012.02.027. [9] Aransiola EF, Ojumu TV, Oyekola OO, Madzimbamuto TF, Ikhu-Omoregbe DIO. A review of current technology for biodiesel production: state of the art. Biomass Bioenergy 2014;61:276–97. http://dx.doi.org/10.1016/j. biombioe.2013.11.014. [10] Mofijur M, Atabani AE, Masjuki HH, Kalam MA, Masum BM. A study on the effects of promising edible and non-edible biodiesel feedstocks on engine performance and emissions production: a comparative evaluation. Renew Sustain Energy Rev 2013;23:391–404. http://dx.doi.org/10.1016/j. rser.2013.03.009. [11] Pullen J, Saeed K. Factors affecting biodiesel engine performance and exhaust emissions – Part I: review. Energy 2014;72:1–16. http://dx.doi.org/10.1016/j. energy.2014.04.015. [12] Graboski MS, McCormick RL. Combustion of fat and vegetable oil derived fuels in diesel engine. Progr Energy Combust Sci 1998;24(2):125–64. http://dx.doi. org/10.1016/S0360-1285(97)00034-8. [13] Pullen J, Saeed K. Factors affecting biodiesel engine performance and exhaust emissions – Part II: experimental study. Energy 2014;72:17–34. http://dx.doi. org/10.1016/j.energy.2014.02.034. [14] Bankovic´-Ilic´ IB, Stojkovic´ IJ, Stamenkovic´ OS, Veljkovic VB, Hung Y-T. Waste animal fats as feedstocks for biodiesel production. Renew Sustain Energy Rev 2014;32:238–54. http://dx.doi.org/10.1016/j.rser.2014.01.038. [15] Selvam PDJ, Vadivel K. Performance and emission analysis of DI diesel engine fuelled with methyl esters of beef tallow and diesel blends. Proc Eng 2012;38:342–58. http://dx.doi.org/10.1016/j.proeng.2012.06.043. [16] Lin C-Y, Li R-J. Engine performance and emission characteristics of marine fishoil biodiesel produced from the discarded parts of marine fish. Fuel Process Technol 2009;90(7–8):883–8. http://dx.doi.org/10.1016/j.fuproc.2009.04.009. [17] Roy MM, Wang W, Alawi M. Performance and emissions of a diesel engine fueled by biodiesel–diesel, biodiesel–diesel-additive and kerosene–biodiesel blends. Energy Convers Manage 2014;84:164–73. http://dx.doi.org/10.1016/j. enconman.2014.04.033. [18] Kumar N, Varun, Chauhan SM. Performance and emission characteristics of biodiesel from different origins: a review. Renew Sustain Energy Rev 2013;21:633–58. http://dx.doi.org/10.1016/j.rser.2013.01.006. [19] Koszałka G. Model of operational changes in the combustion chamber tightness of a diesel engine. Eksploatacja i Niezawodnosc – Maintenance and Reliability 2014;16(1):133–9. [20] How HG, Masjuki HH, Kalam MA, Teoh YH. An investigation of the engine performance, emissions and combustion characteristics of coconut biodiesel in a high-pressure common-rail diesel engine. Energy 2014;69:749–59. http:// dx.doi.org/10.1016/j.energy.2014.03.070. [21] Labeckas G, Slavinskas S, Mazeika M. The effect of ethanol–diesel–biodiesel blends on combustion, performance and emissions of a direct injection diesel engine. Energy Convers Manage 2014;79:698–720. http://dx.doi.org/10.1016/ j.enconman.2013.12.064. [22] Atabani AE, Silitonga AS, Ong HC, Mahli TMI, Masjuki HH, Badruddin JA, et al. Non-edible vegetable oils: a critical evaluation of oil extraction, fatty acid compositions, biodiesel production, characteristics, engine performance and emissions production. Renew Sustain Energy Rev 2013;18:211–45. http://dx. doi.org/10.1016/j.rser.2012.10.013. [23] Ashraful AM, Masjuki HH, Kalam MA, Fattah IMR, Imtenan S, Shahir SA, et al. Production and comparison of fuel properties, engine performance, and emission characteristics of biodiesel from various non-edible vegetable oils: a review. Energy Convers Manage 2014;80:202–28. http://dx.doi.org/10.1016/ j.enconman.2014.01.037. [24] Meng X, Chen G, Wang Y. Biodiesel production from waste cooking oil via alkali catalyst and its engine test. Fuel Process Technol 2008;89(9):851–7. http://dx.doi.org/10.1016/j.fuproc.2008.02.006. [25] Sakthivel G, Nagarajan G, Ilangkumaran M, Gaikwad AB. Comparative analysis of performance, emission and combustion parameters of diesel engine fuelled with ethyl ester of fish oil and its diesel blends. Fuel 2014;132:116–24. http:// dx.doi.org/10.1016/j.fuel.2014.04.059.

219

[26] Barrios CC, Domínguez-Sáez A, Martín C, Álvarez P. Effects of animal fat based biodiesel on a TDI diesel engine performance, combustion characteristics and particle number and size distribution emissions. Fuel 2014;117(A):618–23. http://dx.doi.org/10.1016/j.fuel.2013.09.037. [27] Oner C, Altun S. Biodiesel production from inedible animal tallow and an experimental investigation of its use as alternative fuel in a direct injection diesel engine. Appl Energy 2009;86(10):2114–20. http://dx.doi.org/10.1016/j. apenergy.2009.01.005. [28] Armas O, Gómez A, Ramos A. Comparative study of pollutant emissions from engine starting with animal fat biodiesel and GTL fuels. Fuel 2013;113:560–70. http://dx.doi.org/10.1016/j.fuel.2013.06.010. [29] Mohan B, Yang W, Chou SK. Fuel injection strategies for performance improvement and emissions reduction in compression ignition engines – a review. Renew Sustain Energy Rev 2013;28:664–76. http://dx.doi.org/ 10.1016/j.rser.2013.08.051. [30] Agarwal AK, Dhar A, Gupta JG, Kim WI, Choi K, Lee CS, et al. Effect of fuel injection pressure and injection timing of Karanja biodiesel blends on fuel spray, engine performance, emissions and combustion characteristics. Energy Convers Manage 2015;91:302–14. http://dx.doi.org/10.1016/j. enconman.2014.12.004. [31] Agarwal AK, Dhar A, Gupta JG, Kim WI, Lee CS, Park S. Effect of fuel injection pressure and injection timing on spray characteristics and particulate size– number distribution in a biodiesel fuelled common rail direct injection diesel engine. Appl Energy 2014;130:212–21. http://dx.doi.org/10.1016/j. apenergy.2014.05.041. [32] Agarwal AK, Dhar A, Srivastava DK, Maurya RK, Singh AP. Effect of fuel injection pressure on diesel particulate size and number distribution in a CRDI single cylinder research engine. Fuel 2013;107:84–9. http://dx.doi.org/ 10.1016/j.fuel.2013.01.077. [33] Rajkumar S, Sudarshan GS. Multi-zone phenomenological model of combustion and emission characteristics and parametric investigations for split injections and multiple injections in common-rail direct-injection diesel engines. Proc Inst Mech Eng, Part D: J Automob Eng 2014:4. http://dx.doi.org/ 10.1177/0954407014560797. [34] Berrios M, Gutiérrez MC, Martín MA, Martín A. Application of the factorial design of experiments to biodiesel production from lard. Fuel Process Technol 2009;90(12):1447–51. http://dx.doi.org/10.1016/j.fuproc.2009.06.026. [35] Duda K, Mikulski M, Mickevicˇius T. Effect of doping diesel oil with methyl esters on physicochemical properties of the obtained fuel, in the aspect of its exploitation potential. J KONES 2014;21(4):71–8. [36] Mickevicˇius T, Slavinskas S, Wierzbicki S, Duda K. The effect of diesel–biodiesel blends on the performance and exhaust emissions of a direct injection off-road diesel. Transport 2014;29(4):441–8. http://dx.doi.org/10.3846/ 16484142.2014.984331. [37] Mikulski M, Wierzbicki S. The concept and construction of the engine test bed for experiments with a multi-fuel CI engine fed with CNG and liquid fuel as an ignition dose. J KONES 2012;19(3):289–96. [38] Pie˛tak A, Mikulski M. On the modeling of pilot dose ignition delay in a dualfuel, self ignition engine. Combust Eng 2011;3(146):94–102. [39] Behcet R. Performance and emission study of waste anchovy fish biodiesel in a diesel engine. Fuel Process Technol 2011;92(6):1187–94. http://dx.doi.org/ 10.1016/j.fuproc.2011.01.012. [40] Kousoulidou M, Fontaras G, Ntziachristos L, Samaras Z. Biodiesel blend effects on common-rail diesel combustion and emissions. Fuel 2010;89(11):3442–9. http://dx.doi.org/10.1016/j.fuel.2010.06.034. [41] Ithnin AM, Noge H, Kadir HA, Jazair W. An overview of utilizing water-in-diesel emulsion fuel in diesel engine and its potential research study. J Energy Inst 2014;87(4):273–88. http://dx.doi.org/10.1016/j.joei.2014.04.002. [42] Attia AMA, Kulchitskiy AR. Influence of the structure of water-in-fuel emulsion on diesel engine performance. Fuel 2014;116:703–8. http://dx.doi.org/ 10.1016/j.fuel.2013.08.057. [43] Ballesteros R, Guillén-Flores J, Martínez JD. Carbonyl emission and toxicity profile of diesel blends with an animal-fat biodiesel and a tire pyrolysis liquid fuel. Chemosphere 2014;96:155–66.