Fuel Processing Technology 174 (2018) 69–77

Contents lists available at ScienceDirect

Fuel Processing Technology

journal homepage: www.elsevier.com/locate/fuproc

Research article

The determination of the activation energy of diesel and biodiesel fuels and T
the analysis of engine performance and soot emissions
⁎
Felipe Sotoa, , Márcio Alvesb, Juan Carlos Valdésc, Octavio Armasd, Paula Crnkovice,
Gustavo Rodriguesa, André Lacerdaa, Laura Meloa
a
Universidade Federal de São João del-Rei, Praça Frei Orlando, 170, São João del-Rei, Minas Gerais CEP: 36307-352, Brazil
b
Instituto Federal do Tocantins, AE 310 Sul, Av. LO 05, Palmas, Tocantins CEP: 77.021-090, Brazil
c
Universidade Federal do Tocantins, Av. NS 15, 109 Norte, Palmas, Tocantins CEP: 77001-090, Brazil
d
Universidad de Castilla-La Mancha, Escuela de Ingeniería Industrial, Edif. Sabatini, Av. Carlos III, s/n, 45071 Toledo, Spain
e
Universidade de São Paulo – Escola de Engenharia de São Carlos – Núcleo de Engenharia Térmica e Fluidos, Av. Trabalhador Sancarlense, 400, SP CEP: 13566-970,
Brazil

A R T I C L E I N F O A B S T R A C T

Keywords: The behaviors of a biofuel and a fossil fuel were evaluated by different techniques, namely thermal analysis
Biodiesel fuel (Thermogravimetry-TG and Differential Thermal Analysis-DTA) and engine performance tests. From TG and
Thermogravimetric analysis DTA curves, two decomposition phases were identified for diesel fuel with ignition temperature (IT) at 250 °C
Engine performance and three phases for biodiesel fuel with IT at 300 °C. Combining conversion (α) versus temperature and a model-
Soot emissions
free kinetics method, the range of activation energies (Ea) values were determined for both diesel (from
48.5 kJ mol−1 up to 61.0 kJ mol−1) and biodiesel (from 58.6 kJ mol−1 up to 55.0 kJ mol−1) fuels. At the initial
phase of combustion, diesel fuel presents lower Ea and IT than biodiesel fuel. At the final phase of combustion, an
opposite behavior is observed. These results provide subsides to predict that the behavior of diesel-biodiesel
blends is the best option due to the lowering of Ea along the entire combustion process. As regards the engine
tests, the B20 blend showed improvements compared with diesel fuel. In average, B20 presented increases of
1.2% in power, 1.0% in torque and 1.2% in thermal efficiency. In terms of soot, B20 presented 8.9% lower
emissions than diesel fuel. Although thermal analysis and engine performance tests are executed under very
different conditions, the results from both techniques showed the same trend, i.e., the best option for the
combustion performance is a diesel-biodiesel blend.

1. Introduction limiting its performance. Biofuels can positively contribute to this

equilibrium showing its sustainable nature. The use of alternative fuels
The global energy matrix is mainly based on non-renewable re- is increasingly attracting scientific interest [5], being biofuels widely
sources and extremely dependent on petroleum. The possible alter- used in diesel engines [6] due to its advantages and especially due to its
native energy sources are biofuels originated from numerous oleaginous environmental qualities [7,8].
plants. Normally, this fuel is produced by transesterification, which Cardenas et al. [9] investigated the emissions and performance of
consists of a chemical reaction of vegetable oils or animal fats with rapeseed, soybean and sunflower biodiesel fuels and their 30% blends
ethyl or methyl alcohol, stimulated by a catalyst [1,2]. with diesel, comparing them with a reference mineral diesel fuel. The
Biodiesel fuel obtained from soybean is widely significant in the experimental tests were done according to the NEDC system. In terms of
substitution of mineral diesel fuel. In Brazil, it represents 64.84% of the performance results, the biodiesel fuels displayed higher consumption
national production of biodiesel fuel according to ANP (the Brazilian over time than diesel fuel. Referring to pollutants, the biodiesel fuels
National Agency of Petroleum, Natural Gas and Biofuels) [3]. The diesel presented higher emissions of CO, THC and NOx, and lower values of
engine emits high levels of particles and NOx due to high injection smoke opacity. The accelerations during urban cycles showed con-
pressures and increased air-fuel ratio [4]. It is, in this sense, critical to siderable emissions peaks. An explanation for these emissions data is
diesel engines, the reason why its emissions must be reduced without that biodiesel and its blends have a lower exhaust gas recirculation

⁎
Corresponding author.
E-mail addresses: felipesp@ufsj.edu.br (F. Soto), marcio-alves@ifto.edu.br (M. Alves), juancs@uft.edu.br (J.C. Valdés), octavio.armas@uclm.es (O. Armas),
paulam@sc.usp.br (P. Crnkovic), souzagr@ufsj.edu.br (G. Rodrigues), almlacerda@ufsj.edu.br (A. Lacerda).

https://doi.org/10.1016/j.fuproc.2018.02.008
Received 6 November 2017; Received in revised form 31 January 2018; Accepted 8 February 2018
Available online 22 February 2018
0378-3820/ © 2018 Elsevier B.V. All rights reserved.
F. Soto et al. Fuel Processing Technology 174 (2018) 69–77

valve opening, which decreases opacity and soot emissions, but in- Table 1
creases other pollutants. Fuel properties.
Bermudez et al. [5] investigated consumption and emissions beha-
Property Unit Diesel Soybean B100
vior of soybean, rapeseed and palm biodiesel and an alternative Fischer
Tropsch (FT) fuel, comparing them with a mineral diesel fuel in an Mass fractions
engine under NEDC. The fuel consumption of all biodiesel fuels showed Carbon (−) 0.870 0.769
Hydrogen 0.126 0.118
higher values compared to diesel and FT fuels. They also led to higher
Oxygen 0.003 0.113
emissions of NOx, CO and HC. By analyzing soybean biodiesel in Sulfur 0.001 (−)
comparison to the other biodiesel fuels, higher concentrations of HC Cetane number (−) 50.5 50.6
and CO and lower of NOx have been found. Because of the presence of Viscosity at 40 °C (mm2/s) 2.68 4.24
aromatics in their formula, mineral diesel and FT fuels created benzene Density at 15 °C (kg/L) 0.84 0.88
Lower heating value (MJ/kg) 42.9 36.8
and toluene in HC emissions, which are cancer-causing compounds.
Stoichiometric air-fuel ratio (−) 14.66 12.49
Bermudez et al. [10] studied the pollution rate and particle sizes of Distillation 10 vol% (°C) 195 316
particulate matter from the burning of soybean, rapeseed and palm Distillation 50 vol% (°C) 262 328
biodiesel, S10 and S50 mineral diesel and Fischer Tropsch fuels. As Distillation 90 vol% (°C) 339 334
regards the total particulate matter production, the lowest emissions
were from palm biodiesel, while soybean presented the third best re-
2.2. TG and DTA experiments
sults. It was observed that at low speed and load, the biodiesel fuels
produced more PM than mineral diesel fuels. However, at medium
Thermogravimetric (TG) and Differential Thermal Analysis (DTA)
speed and load, the biodiesel fuels showed lower total concentrations of
experiments were carried out in the Shimadzy 51H and 50H analyzers,
particles on emissions. In terms of particle size from biodiesel fuels
respectively. Experimental conditions were: heating rates of 5, 10 and
emissions, the particles are smaller than those produced by mineral
20 °C min−1, temperature range from room temperature up to 600 °C,
diesel fuels. According to the author, the presence of sulfur in mineral
sample mass of 10 ± 0.5 mg of each fuel for each test and a
diesel fuels is the reason why the particle size and geometry increase.
100 mL min−1 flow rate of synthetic air atmosphere (80% nitrogen and
There are plenty of results from experimental tests of biodiesel-
20% oxygen).
fueled internal combustion engines. Most of them present data on
performance and emissions. However, detailed analyses of the com-
bustion are needed to better understand the biodiesel burning process. 2.3. Activation energy determination. Mathematical model
A technique which has been extensively used to determine the activa-
tion energy of various solid and liquid fuels is the thermal degradation The model-free kinetics is based on the calculation of the activation
assessed by thermogravimetry. These studies have been performed energy as a function of the conversion (α). In several reactive processes,
based on fuels to be used under different combustion conditions, i.e., the determination of the reaction rate depends on conversion α, tem-
for different kinds of thermal machines. Kok et al. [11] studied the perature (T) and time (t). Conversion is given by:
nature of the combustion of agricultural solid residues using thermal α = (m 0 − m)/(m 0 − m∞), (1)
analysis techniques. Luo et al. [12] investigated the combustion che-
mical kinetics of coke on deactivated catalysts with the use of ther- in which m is the sample mass which is a function of time, m0 is the
mogravimetry. Borsato et al. [13] adopted the thermogravimetric initial mass of the sample and m∞ is the final mass of the sample.
analysis to determine the activation energy of the soybean biodiesel The reaction rate given as a function of conversion, ƒ(α), is different
fuel mixed with three kinds of synthetic antioxidants. Crnkovic et al. for each process and the reaction model must be determined experi-
[14] determined the activation energies of the crude glycerin and of the mentally. For a simple reaction, the estimated ƒ(α) has the value of the
beef tallow for their application in engines. Also for that purpose, reaction order n. For complex reactions, ƒ(α) is complex and generally
Conconi et al. [15] analyzed the behavior of the activation energies of unknown, in which case the algorithm of order n results in inaccurate
three kinds of fuels produced in Brazil: mineral diesel, sugarcane far- chemical kinetics data.
nesane and soybean biodiesel. The use of model-free kinetics based on the Vyazovkin theory
The objective of the present study is to determine the activation [16–18] makes it possible to retrieve kinetic information about complex
energy for the thermal decomposition of diesel and soybean biodiesel reactions. This model is based on isoconvertional techniques for the
fuels, to specify which fuel or fuel blend leads to the best combustion calculation of the effective activation energy (Ea) as a function of the
quality under atmospheric pressure on a thermogravimetric balance. conversion (α) of the chemical reaction, i.e. Ea = ƒ(α). Hence, this ap-
Additionally, it tries to make the same estimations from engine per- proach follows all the conversion points collected from the numerous
formance and soot emissions experimental test, i.e., under high pressure experimental tests, avoiding uncertainties that can result from a single
and high heating rates. The fuels tested were: Brazilian commercial experiment. The theory states that:
diesel fuel, pure soybean biodiesel (100% biodiesel, B100) and blends dα / dt = k (T ) f (α ), (2)
of 20% (B20) and 50% (B50) biodiesel in diesel.
in which t is the time, T is the temperature and α is the conversion, ƒ(α)
2. Experimental tests represents the reaction model and k(T) is the Arrhenius reaction rate
coefficient, so:
2.1. Materials
−E
dα / dt = A exp ⎛ a ⎞ f (α ),
⎝ RT ⎠ (3)
Two different fuel samples, diesel and biodiesel, were used in this
study. The diesel fuel is commercialized by Petrobras as S10 which is a in which R is the universal gas constant, A is a pre-exponential factor, Ea
blend of 93% pure diesel and 7% biodiesel fuels. The biodiesel fuel is the activation energy and dα/dt is the reaction rate [s−1].
sample for the tests was obtained from the Brazilian company Granol, The global relations of a multi-step process can be represented by
processed by transesterification with methanol. Properties of the pure combining a single-step equation, so Eq. (3) is representative of the
fuels are shown in Table 1. In the engine experimental tests, in addition whole process. In effect, the global relation of the process is approxi-
to pure fuels, two blends with different proportions of diesel and bio- mated by the combination of a few (commonly two or three) relations
diesel fuels were also used: B20 and B50. of equations and each one of them influences the whole temperature

70
F. Soto et al. Fuel Processing Technology 174 (2018) 69–77

region (T region) and the extent of the reaction. the calculation of an FSN (Filter Smoke Number) with a resolution of
Dividing Eq. (3) by the heating rate β = dT/dt, the following 0.001 FSN within measurement limits from 0 to 10 FSN.
equation is obtained:
3. Results
1 A −E
dα = exp ⎛ a ⎞ dT .
f (α ) β ⎝ RT ⎠ (4)
3.1. TG, DTG and DTA results and discussion
Integrating Eq. (4) up to conversion α (at the temperature T):
Fig. 2 shows the TG and DTG curves at a 10 °C min−1 heating rate of
α 1 A T −E biodiesel and mineral diesel fuels. These figures reveal curves for a
∫0 f (α )
dα = g (α ) =
β
∫T
0
exp ⎛ a ⎞ dT .
⎝ RT ⎠ (5) single heating rate for the sake of displaying the different thermal de-
For Ea/RT ≫ 1, the integral of the temperature can be approximated composition phases in a more visually accessible manner. However,
by: similar curves were obtained for the other heating rates applied (5.0
and 20.0 °C min−1).
T −E R 2 −E Two and three phases were identified as significant for diesel and
∫T exp ⎛ a ⎞ dT ≈
⎝ RT ⎠ Ea
T exp ⎛ a ⎞.
⎝ RT ⎠ (6)
0 biodiesel fuels, respectively, represented by rectangles in Fig. 2. The
boundaries of each phase were determined based on the inflections and
Substituting the integral of the temperature and representing in
peaks of the DTG curves, as previously reported by Crnkovic et al. [14].
logarithmic form:
For biodiesel fuel (Fig.2a), mass loss on the first phase (I), also called
β RA ⎤ E 1 low-temperature oxidation region, is approximately 8% over the
ln = ln ⎡ − a ,
Tα2 ⎢
⎣ Ea g (α ) ⎥
⎦ R α Tα (7) 27–181 °C temperature range. Water removal and some low-tempera-
ture volatiles, such as the alcohol used in the transesterification process,
in which index α represents values related to a specific conversion and are considered responsible for the loss of mass in this first step [20].
α 1 The second decomposition phase (II) of biodiesel fuel, the main
g (α ) = ∫0 f (α )
.
(8) region, is developed over the 181–330 °C temperature range with a
mass loss of 86%. This phase was considered the potential of the fuel.
One of the main advantages of this method is the possibility of Therefore it was chosen for the calculation of the activation energy Ea,
isolating function g(α) from the linear coefficient. Determining this since it is characterized by the largest mass loss related to the main
function in complex processes is rather difficult [17]. released compounds.
Eq. (6) is described as a dynamic equation, which is used for de- The third region (III) of the biodiesel fuel sample lasts over the
termining the activation energy for all the conversion values. 330–660 °C temperature range with a minimal mass loss of approxi-
Eq. (7) assures the calculation of the activation energy Ea and of the mately 2%. According to Dou et al. [20], the degradation of impurities,
pre-exponential factor A from the Arrhenius curve (linear regression), such as alcoholic esters, fatty acids and degradation residuals from the
plotting y = ln (β/Tα2) and x = 1/Tα. Knowing that y = a0 + a1x is the previous phases, are responsible for this third phase of degradation.
straight line equation, then a0 = ln [RA/Eag(α)] and a1 = − Ea/R, thus Among the degradation residuals from the previous phases (I and II) are
Ea = Ra1. carboxylic acids, aldehydes, ketones and hydroperoxides, i.e., by-
products of the fuel oxidation process [21]. The final temperature for
2.4. Engine performance and emissions tests biodiesel fuel was limited to 600 °C, since, at this point, the phase
corresponding to the largest mass loss had already been defined and
Fig. 1 supplies a general scheme of the engine test bench and of the there was no need to carry on with the tests.
experimental equipment used in this work, assembled at the Engines Fig. 2b shows TG and DTG curves of diesel fuel at the same heating
and Fuels Laboratory of the National Institute of Metrology, Quality and rate used for biodiesel fuel (10 °C min−1). Two phases are highlighted
Technology – Rio de Janeiro (INMETRO-RJ). In the performance tests, a for the thermal decomposition. The first region (I) is related to the
turbocharged four-cylinder four-stroke stationary agricultural Diesel oxidation of lighter compounds of diesel fuel, which is a mixture of
engine was used. It has a combustion pre-chamber, displaces a total of numerous hydrocarbons. It shows 10% mass loss over the 30−100 °C
4485 cm3, its nominal maximum power is 97 kW at 2400 min−1 and its temperature range.
nominal maximum torque is 500 Nm at 1600 min−1. It has a mechan- For diesel fuel, the second region is also considered one main re-
ical fuel supply system with a piston injector pre-pump and a pump. gion, as its mass loss corresponds to 89% of the total mass loss. This
An eddy-current air-cooled dynamic dynamometer AVL Schneider region extended over the 100–285 °C temperature range. It is important
Electric Drives, Serial 130408, was employed. Puma AVL Open 1.2.1 to notice that, due to the location and the characteristics of this event,
was the software used for analyzing the performance of the diesel en- there was almost no mass loss above 285 °C, the reason why this tem-
gine in terms of power, torque, fuel consumption and other parameters. perature was considered the end of the process. The final temperature
The tests for the assessment of the performance of the engine were for diesel fuel was limited to 400 °C. Over the 285–400 °C temperature
carried out in accordance to the standard NBR ISO 1585. This standard range the DTG curve was linear.
determines that data regarding power, torque and fuel consumption Combining TG with DTA curves, the thermal events become more
should be measured at different engine speeds, from the highest to the apparent. From the DTA curves (Fig. 3), which show endothermic and
lowest speed, in a non-transient regime and at full load for each one of exothermic enthalpy transitions, it can be observed that the region
the three approaches. Hence, the experimental tests were performed around 170 °C for the diesel fuel is related to an endothermic event.
with the engine idling at 800 min−1, for warming up, gradually in- After 240 °C, a large peak can be seen. For the biodiesel fuel, the region
creasing the rotational speed and the accelerator pedal needed for the after 300 °C is related to two noticeable exothermic events.
stabilization of the engine up to the maximum speed of 2300 min−1 and For the biodiesel fuel, the first exothermic event is caused by the
accelerator pedal at 100%. combustion of released volatile matter and the second is caused by the
To measure soot emissions, the equipment AVL 415S Smoke Meter burning of residuals. Moreover, the biodiesel fuel DTA curve in Fig. 3
was employed. According to Souza et al. [19], this equipment is a filter shows that the ignition temperatures of both volatile matter and re-
meter that measures the soot content of gases from a Diesel engine. The siduals shift to higher temperatures when the same event is observed in
filter gets blackened by soot and then the measurements are made by a the TG curves. The ignition temperature is considered as the point at
photoelectric head. The results are evaluated by a microprocessor for which the slope of the DTA curve changes due to the heat release rate

71
F. Soto et al. Fuel Processing Technology 174 (2018) 69–77

Fig. 1. Scheme of the experimental setup.

Fig. 2. TG and DTG curves of the biodiesel and diesel fuels.

[22]. second decomposition phase shows that the thermal degradation pro-
For both fuels, the exothermic peaks of the DTA curves correspond cess is different for each sample. Therefore, the phase II activation
to the remarkable mass losses in the TG curves. While for biodiesel fuel energy can be linked to the ease of igniting the fuel in a combustion
the exothermic events are divided into two phases, corresponding to process.
phases (II) and (III), for diesel fuel the exothermic event correspond to
phase (II). A similar behavior was reported by Conceição et al. [23]. 3.2. Determination of activation energy and discussion
In addition, from the DTA curves, it is also found that diesel and
biodiesel fuels have different types and temperatures of ignition. For Fig. 4 shows the conversion of the diesel and biodiesel fuels samples
the diesel fuel the ignition starts at a lower temperature (250 °C) on phase II plotted as a function of temperature for every tested heating
compared to the biodiesel fuel (300 °C), proving that biodiesel fuel al- rate (5, 10 and 20 °C min−1). It can be noticed that the rise on the
lows more security in terms of storage and handling than diesel fuel. conversion curve is shifted to a higher temperature as the heating rate is
There is an indication that the exothermic peak extending from 240 raised. Qualitatively, the conversion of biodiesel fuel is similar to that
to 400 °C in the DTA curve of biodiesel fuel is a reflection of the com- of diesel fuel, but it happens at higher temperatures. Each conversion
bustion of the released volatiles. For diesel fuel, the endothermic peak level is related to a different temperature at different heating rates. By
extending from 100 to 270 °C is a reflection of the decomposition or plotting ln(β/Tα2) versus 1/Tα, a curve is obtained with a slope equal to
dissociation reaction. This difference in the enthalpy transitions in the −Ea/R, which is different for each conversion level. This principle

72
F. Soto et al. Fuel Processing Technology 174 (2018) 69–77

notably rising to the Ea value for biodiesel fuel corresponding to a 24%

mass conversion. After that, it rises with a less steep positive slope until
it reaches the maximum activation energy of 61.0 kJ mol−1 at 0.9
conversion.
From the model-free kinetics model, the activation energies corre-
sponding to the lowest and highest conversion values obtained for the
biodiesel fuel sample were respectively 58.6 kJ mol−1 and
55.0 kJ mol−1. For the diesel fuel, these values were 48.5 kJ mol−1 and
61.0 kJ mol−1, respectively.
Comparing with results for diesel fuel from other works, Crnkovic
et al. [21] have studied the activation energies of different diesel fuels
produced in Brazil and have concluded that these values range from
43.1 kJ mol−1 to 66.0 kJ mol−1. On average, the activation energy
values for biodiesel and diesel fuels studied in this paper are close to
this range. However, due to assumptions made by model-free kinetics,
this parameter cannot be considered constant during the reaction. In
addition, as indicated by the determined ranges of both fuels, the ac-
Fig. 3. DTA curve of diesel and biodiesel fuels at a 10 °C min−1 heating rate. tivation energy should not be considered a single average value, but
rather a set of values. Thus, the combustion can be considered a two
phase process: the pre-ignition and the ignition phases [14,15]. This is
valuable information for the prediction of the behavior of new fuels in
combustion processes at atmospheric pressure.
Fig. 5 also shows that, from the beginning of the pre-ignition up to
0.24 conversion, the activation energy is higher for biodiesel fuel
compared to diesel fuel. From this point on, an inversion in this be-
havior is observed, as the diesel fuel activation energy values become
substantially higher than those for biodiesel fuel. This inversion might
find an explanation in the alkyl ester chemical structure (the main
compound of biodiesel), which is simpler than those of heavier hy-
drocarbons, the main compounds of the diesel fuel. These results can
lead to the interpretation that the best combustion quality at atmo-
spheric pressure and low heating rates are from a blend of both fuels.
The activation energy values obtained in this study are typical of the
oxidation reaction on the thermogravimetric balance that was used.
These values are considerably lower than the activation energies used
by Arkhangelsky et al. [24] and Heywood [25] to determine the igni-
tion delay and by Payri and Desantes [26] to determine the combustion
Fig. 4. Conversion as a function of temperature for phase II of the diesel and biodiesel
fuels.
reaction rate, all in engine combustion chambers.

3.3. Performance results and discussion

Fig. 6 displays the maximum effective power and torque and the
minimum specific fuel consumption values. The highest maximum
power observed from the fuels was for B20, followed by diesel fuel.
Diesel fuel shows a better response on maximum torque, followed by
B20, with a small difference of 2.0 Nm (0.5%). Analyzing the minimum
values of specific fuel consumption, the best result was for mineral

Fig. 5. Activation energy versus conversion over phase II for both samples: diesel and
biodiesel fuels.

indicates that the activation energy can be plotted as a function of

conversion for both cases, as shown in Fig. 5.
Fig. 5 shows that the runs of the curves of both fuel samples are
qualitatively very distinct. The run of the biodiesel curve reaches the
maximum activation energy (58.6 kJ mol−1) at 0.1 conversion and then
slightly decays, reaching a 0.7 conversion at the minimum activation
energy (50.7 kJ mol−1). After this point, a positive slope takes the ac-
tivation energy to 55.0 kJ mol−1.
Contrastingly, the diesel fuel sample curve starts at 48.5 kJ mol−1,
Fig. 6. Performance results.

73
F. Soto et al. Fuel Processing Technology 174 (2018) 69–77

Fig. 7. Engine performance at full load.

diesel, followed by B20, at the same engine speed (1700 min−1). larger diameters due to the lower pulverization. As follows, the high
Fig. 7 shows engine performance results at full load as a function of kinematic viscosity leads to low atomization in the burn, i.e., in-
the engine speed. complete combustion [31]. Viscosity is one of the most important
These results can lead to a wide discussion, starting with the fuel properties, since it plays a dominant role in the spraying, in the mixing
consumption. In mechanical diesel fuel injection systems (such as the and in the combustion processes [32].
one used in this work), the fuel quantity injected is volumetrically It can be observed that the increase in engine speed leads to an
controlled. This causes an increase in the fuel mass injected due to the increase in power, for all the fuels tested, and the maximum values of
higher density of the biodiesel [27]. Simultaneously, the high viscosity power at low engine speed show a slight difference (see Fig. 7a). From
of the fuel causes an increase in the pressure in the fuel delivery system 1400 to 2300 min−1, B20 clearly delivers higher power compared to
and in the fuel volume injected due to the decrease in the delivery B50 and B100. Compared to diesel fuel, B20 led to a modest increase in
losses [28]. Thus, as the density and the viscosity of the diesel-biodiesel power.
blends increase as a consequence of the progressive addition of bio- There is a continuous decrease in torque for all the fuels tested from
diesel, the fuel quantity injected in the cylinder is higher [29]. In this 1000 to 1400 min−1. Comparing diesel fuel to B20 reveals similarities
study, according to Table 1, the biodiesel and diesel fuels show different and alternations in torque, whereas B50 shows the lowest torque. In the
viscosity and density values. This justifies the behavior of the fuel 1400 to 1800 min−1 range, all fuels lead to an increase in torque
consumption results shown in Fig. 7d. (Fig. 7b). In the 1000 to 2300 min−1 range, all the fuels tested show
Although B20 has insignificant differences compared to diesel fuel similarities concerning gradual deviations in torque values, with small
in terms of torque and power, as the biodiesel fuel proportion in the oscillations, i.e., climbs and plunges, though the best behavior in terms
blend is increased, a decrease in performance and increase in fuel of torque is noted from the B20.
consumption is observed, due to the rise in the viscosity index and the The specific fuel consumption (Fig. 7c) shows a similar behavior to
density. For some higher concentrations of biodiesel in diesel fuel, the the fuel consumption regarding the differences between the reference
density can shift the engine performance. Therefore, it influences the diesel fuel and the biodiesel blends, since the differences in power are
power developed by the engine, regardless of the higher molecular small between them. Up to 1600 min−1, the differences in fuel con-
weight of the oleaginous energy source. It is an important property for sumption between the reference diesel fuel and the B20 are significant.
the analysis of the performance of an engine because the difference in Following on from 1700 min−1, these differences are negligible. To
the mass of the different types of fuel influences the injection of these complete this analysis, the engine thermal efficiency was calculated,
substances and alters the delivered power [30]. which is presented on Fig. 8. It can be observed that the B20 blend,
Higher biofuel concentrations tend to increase the fuel consumption following on from 1700 min−1, leads to a slightly higher thermal effi-
and one of the causes for that is their large viscosity indexes, especially ciency at most engine speeds.
for B100. Blends richer in biodiesel show higher viscosity. This impairs The comparative values between the reference mineral diesel and
the mixture between oxygen and the fuel droplets, which will have the B20 become more significant for higher engine speeds, following

74
F. Soto et al. Fuel Processing Technology 174 (2018) 69–77

higher gas temperature in the cylinder [35]. These authors also ana-
lyzed reductions in soot emissions for biodiesel fuels at full load.
They point out that the main cause of this decrease is the oxygen
content of biodiesel: B20 has approximately 2.5% of oxygen by weight.
The oxygen atoms present on the richest zone of the fuel jet, corre-
sponding to a region of diffusive flame, create OH radicals, acting as
soot nucleation suppressors. These OH radicals allow the oxidation of
soot precursors. As the content of oxygen on fuel rises, soot con-
centrations decrease [35–37]. These authors also claimed that other
studies highlighted the tendency of soot production by the presence of
aromatics and sulfur in the fuel. The fuel blends are less likely to pro-
duce soot because they have smaller fractions of long-chain hydro-
carbons than petroleum diesel fuel.

3.5. Analysis of results

The differences on the profiles of the DTA curves can be understood

Fig. 8. Engine brake thermal efficiency at full load. based on the chemical composition of the fuels. Biodiesel has higher
amount of oxygen (11% by weight - Table 1) than diesel. The heat
release increase observed in the DTA curve for biodiesel may be asso-
from 1700 min−1. In this operating condition, the combustion is im-
ciated with its oxygen content. Since the fuel oxygen may provide ad-
paired by the shorter time available for the: (i) penetration of the fuel
ditional oxygen to the combustion process, it facilitates both the reac-
spray in the combustion pre-chamber, (ii) formation of the air-fuel
tion and heat release. Such behavior can be related to the lowering of
mixture in the combustion pre-chamber, (iii) vaporization of the fuel,
activation energy at the end of the thermal process, as can be observed
(iv) flow of gases from the pre-chamber to the main chamber, (v)
in the Fig. 5. Furthermore, from the DTA curves, it is found that the
burning of the fuel. It is under these adverse conditions that the engine
diesel and biodiesel fuels have different ignition temperatures. For the
performance with the B20 blend shows a 1.2% increase in power, 1.0%
diesel fuel, ignition starts at lower temperatures (250 °C) compared to
in torque and 1.2% in thermal efficiency.
the biodiesel fuel (300 °C), which means that this biofuel allows more
security in terms of storage and handling than diesel fuel.
3.4. Soot emissions experiment and discussion The differences in the performance behavior of the engine should be
proportional to the lower heating values (Table 1). This has been ob-
With the use of higher proportions of biodiesel in the blended fuel, a served for the B100 and for the B50 in comparison to the mineral diesel.
considerable reduction in emissions, mainly in sulfur, is expected, de- However, the performance behavior of the B20 was not proportional to
spite the higher specific fuel consumption. This justifies the application the difference in the heating value of this fuel blend. From the engine
of diesel-biodiesel fuels blends, including biodiesel rich blends [33]. thermal efficiency results for B20, following from 1700 min−1, it is
Fig. 9 shows the results for soot concentrations expressed as FSN. At clear that a slightly superior behavior is predominant in relation to the
full load, biodiesel fuels led to lower soot emissions at all the engine reference diesel fuel. Definitely, the 20% soybean biodiesel fuel content
speeds tested. This is due to the higher oxygen concentrations and in mineral diesel fuel favors the engine performance at higher engine
lower boiling points of biofuels compared to mineral fuels, which result speeds, because of changes in both physical and chemical properties.
in a better evaporation that favors a more complete combustion of the The physical properties of the fuels, such as viscosity and volatility,
biodiesel fuel [34]. Thus, diesel fuel and B20 caused higher soot affect the combustion process. According to Table 1, these properties
emissions, with small differences between them. have different values for each of the fuels tested. The viscosity influ-
Soot emissions for B20 are predominantly lower than for mineral ences the quality of the fuel spray and the volatility influences the fuel-
diesel fuel within the range of engine speeds tested. Unexpectedly, as air mixture formation. Due to the differences in viscosity, diesel fuel has
the engine speed increases this difference is reduced, with a slight in- a finer spray which reduces the ignition delay, whereas the biodiesel
crease in B20 soot emissions at 1900, 2200 and 2300 min−1. Other fuel spray is thicker but can achieve a larger jet penetration length. This
authors have found larger differences between B20 and mineral diesel favors the distribution of the biodiesel fuel on air but causes the fuel
fuel as the engine speed increases, claiming the reason for that is the droplets to be larger in diameter. The distillation temperatures shown
in Table 1 indicate that diesel fuel has a higher volatility, which also
reduces the ignition delay and benefits the pre-ignition and the start of
the combustion. However, the flammability of the biodiesel fuel is
better during the remaining stages of the combustion. Therefore, a
diesel-biodiesel blend in adequate proportions is the best fuel to ensure
the proper ignition delay and the maximum completeness of the com-
bustion.
Diesel engines have the almost unlimited capability of burning lean
air-fuel mixtures, i.e., to increase the coefficient of excess air. This al-
lows the engine power to vary just by reducing the fuel mass injected
without varying the intake air mass. Diversely, the operation with a
non-homogenous charge is the main disadvantage of Diesel engines
because it impairs a sufficiently complete combustion with low soot
formation when the air-fuel mixture is made rich, i.e., the coefficient of
excess air is reduced. Therefore, by analyzing the tested biodiesel and
mineral diesel fuels, it is observed that the stoichiometric air-to-fuel
ratio of the biodiesel fuel is lower compared to diesel fuel, according to
Fig. 9. Soot concentrations expressed as FSN.
Table 1. This means that the engine can operate with higher coefficients

75
F. Soto et al. Fuel Processing Technology 174 (2018) 69–77

of excess air when burning biodiesel fuel. This is the cause related to the and biodiesel fuel, Atmos. Environ. 50 (2012) 307–313, http://dx.doi.org/10.1016/
results shown in Fig. 9, which reveal that the soot emissions of the j.atmosenv.2011.12.007.
[9] M.D. Cárdenas, O. Armas, C. Mata, F. Soto, Performance and pollutant emissions
engine fueled by biodiesel are lower. from transient operation of a common rail diesel engine fueled with different bio-
diesel fuels, Fuel 185 (2016) 743–762, http://dx.doi.org/10.1016/j.fuel.2016.08.
4. Conclusions 002.
[10] V. Bermúdez, J.M. Luján, S. Ruiz, D. Campos, W.G. Linares, New European Driving
Cycle assessment by means of particle size distributions in a light-duty diesel engine
In this study, we observed that from two different techniques fuelled with different fuel formulations, Fuel 140 (2015) 649–659, http://dx.doi.
(Thermal analysis and engine experiments) it is possible to evaluate org/10.1016/j.fuel.2014.10.016.
[11] M.V. Kok, E. Özgür, Thermal analysis and kinetics of biomass samples, Fuel Process.
fuels to be used in engine. The findings of thermal analysis (TG and Technol. 106 (2013) 739–743, http://dx.doi.org/10.1016/j.fuproc.2012.10.010.
DTA) experiments suggest that, under atmospheric pressure and low [12] S. Luo, S. He, X. Li, J. Li, W. Bi, C. Sun, Combustion kinetics of the coke on deac-
heating rate conditions, the mixture of diesel and biodiesel fuels is the tivated dehydrogenation catalysts, Fuel Process. Technol. 129 (2015) 156–161,
http://dx.doi.org/10.1016/j.fuproc.2014.09.002.
most beneficial for the combustion process rather than using either pure
[13] D. Borsato, J.R. de M. Cini, H.C. da Silva, R.L. Coppo, K.G. Angilelli, I. Moreira,
diesel fuel or pure biodiesel fuel. E.C.R. Maia, Oxidation kinetics of biodiesel from soybean mixed with synthetic
From the engine performance and emissions tests, the B20 blend antioxidants BHA, BHT and TBHQ: determination of activation energy, Fuel
proved to be the best option. Such conclusion was possible due to the Process. Technol. 127 (2014) 111–116, http://dx.doi.org/10.1016/j.fuproc.2014.
05.033.
results obtained from the burning conditions, such as the increases of [14] P.M. Crnkovic, C. Koch, I. Ávila, D.A. Mortari, A.M. Cordoba, A. Moreira dos Santos,
1.2% in power, 1.0% in torque and 1.2% in thermal efficiency and the Determination of the activation energies of beef tallow and crude glycerin com-
8.9% lower soot emissions, compared to diesel fuel. bustion using thermogravimetry, Biomass Bioenergy 44 (2012) 8–16, http://dx.doi.
org/10.1016/j.biombioe.2012.04.013.
The advantage of using thermal analysis to predict the behavior of [15] C.C. Conconi, P.M. Crnkovic, Thermal behavior of renewable diesel from sugar
fuels is that it can be performed in a shorter time in comparison with cane, biodiesel, fossil diesel and their blends, Fuel Process. Technol. 114 (2013)
experiments performed in engines. It can be applied to different kinds 6–11, http://dx.doi.org/10.1016/j.fuproc.2013.03.037.
[16] S. Vyazovkin, D. Dollimore, Linear and nonlinear procedures in isoconversional
of fuels. The methodology provides safe and easy-to-operate conditions computations of the activation energy of nonisothermal reactions in solids, J. Chem.
and a very small fuel sample amount is required for each test (≫10 mg). Inf. Comput. Sci. 36 (1996) 42–45, http://dx.doi.org/10.1021/ci950062m.
Although thermal analysis and engine performance tests are exe- [17] S. Vyazovkin, N. Sbirrazzuoli, Confidence intervals for the activation energy esti-
mated by few experiments, Anal. Chim. Acta 355 (1997) 175–180, http://dx.doi.
cuted under very different conditions, the results from both techniques org/10.1016/S0003-2670(97)00505-9.
showed that the best option for the combustion performance is a diesel- [18] S. Vyazovkin, C.A. Wight, Model-free and model-fitting approaches to kinetic
biodiesel blend. In addition, the results of the thermal analysis guided analysis of isothermal and nonisothermal data, Thermochim. Acta 340–341 (1999)
53–68, http://dx.doi.org/10.1016/S0040-6031(99)00253-1.
the engine performance tests to determine the superior results of the
[19] P.G.M. de Souza, M.L. Alves, Comparação da fuligem emitida pelo motor Diesel, An.
diesel-biodiesel blend over both the pure diesel fuel and pure biodiesel Do 8o Congr. Bras. Metrol., Bento Gonçalves, Rio Grande do Sul, Brasil, 2015,
fuel. pp. 1–4.
[20] B. Dou, V. Dupont, P.T. Williams, H. Chen, Y. Ding, Thermogravimetric kinetics of
crude glycerol, Bioresour. Technol. 100 (2009) 2613–2620, http://dx.doi.org/10.
Funding 1016/j.biortech.2008.11.037.
[21] P.M. Crnkovic, C.R.M. Leiva, A.M. dos Santos, F.E. Milioli, Kinetic study of the
This research did not receive any specific grant from funding oxidative degradation of Brazilian fuel oils, Energy Fuel 21 (2007) 3415–3419,
http://dx.doi.org/10.1021/ef700219u.
agencies in the public, commercial, or not-for-profit sectors. [22] M. Canakci, J.H. Van Gerpen, Comparison of engine performance and emissions for
petroleum diesel fuel, yellow grease biodiesel, and soybean oil biodiesel, Trans.
Acknowledgements ASAE 46 (2003) 964–975, http://dx.doi.org/10.13031/2013.13948.
[23] M.M. Conceição, R.A. Candeia, F.C. Silva, A.F. Bezerra, V.J. Fernandes, A.G. Souza,
Thermoanalytical characterization of castor oil biodiesel, Renew. Sust. Energ. Rev.
The authors want to thank: (a) INMETRO-RJ for the engine ex- 11 (2007) 964–975, http://dx.doi.org/10.1016/j.rser.2005.10.001.
perimental tests, (b) Granol for donating the soybean biodiesel fuel, (c) [24] V. Arkhangelsky, M. Khovakh, Y. Stepanov, V. Trusov, M. Vikhert, A. Voinov, Motor
Vehicle Engines, MIR Publishers, Moscow, 1971.
Department of Agriculture and Livestock of the State of Tocantins
[25] J.B. Heywood, Internal Combustion Engine Fundamentals, McGraw-Hill, New York,
(SEAGRO-TO), for integrating the government, the companies and the 1988.
education and research institutions, allowing this work, (d) Rômulo [26] F. Payri, J.M. Desantes, Motores de combustión interna alternativos, Editorial
Reverté, València, 2011.
Neves for reviewing the English writing of this paper.
[27] R. Behçet, Performance and emission study of waste anchovy fish biodiesel in a
diesel engine, Fuel Process. Technol. 92 (2011) 1187–1194, http://dx.doi.org/10.
References 1016/j.fuproc.2011.01.012.
[28] M. Lapuerta, O. Armas, J. Fernandez, Effect of biodiesel fuels on diesel engine
emissions, Prog. Energy Combust. Sci. 34 (2008) 198–223, http://dx.doi.org/10.
[1] M.M. Pedroza, J.F. Sousa, G.E.G. Vieira, M.B.D. Bezerra, Characterization of the 1016/j.pecs.2007.07.001.
products from the pyrolysis of sewage sludge in 1 kg/h rotating cylinder reactor, J. [29] E. Öztürk, Performance, emissions, combustion and injection characteristics of a
Anal. Appl. Pyrolysis 105 (2014) 108–115, http://dx.doi.org/10.1016/j.jaap.2013. diesel engine fuelled with canola oil–hazelnut soapstock biodiesel mixture, Fuel
10.009. Process. Technol. 129 (2015) 183–191, http://dx.doi.org/10.1016/j.fuproc.2014.
[2] G.E.G. Vieira, G.A. Romeiro, S.M. Sella, R.N. Damasceno, R.G. Pereira, Low tem- 09.016.
perature conversion (LTC) – an alternative method to treat sludge generated in an [30] A.M. Ashraful, H.H. Masjuki, M.A. Kalam, I.M. Rizwanul Fattah, S. Imtenan,
industrial wastewater treatment station – batch and continuous process compar- S.A. Shahir, H.M. Mobarak, Production and comparison of fuel properties, engine
ison, Bioresour. Technol. 100 (2009) 1544–1547, http://dx.doi.org/10.1016/j. performance, and emission characteristics of biodiesel from various non-edible
biortech.2008.08.040. vegetable oils: a review, Energy Convers. Manag. 80 (2014) 202–228, http://dx.doi.
[3] Agência Nacional do Petróleo Gás Natural e Biocombustíveis, Boletim Mensal do org/10.1016/j.enconman.2014.01.037.
Biodiesel, https://goo.gl/EepQmk, (2017) , Accessed date: 2 February 2017. [31] O.M. Ali, R. Mamat, N.R. Abdullah, A.A. Abdullah, Analysis of blended fuel prop-
[4] C. Arcoumanis, T. (Take) Kamimoto, Flow and Combustion in Reciprocating erties and engine performance with palm biodiesel–diesel blended fuel, Renew.
Engines, Springer, 2009. Energy 86 (2016) 59–67, http://dx.doi.org/10.1016/j.renene.2015.07.103.
[5] V. Bermúdez, J.M. Lujan, B. Pla, W.G. Linares, Comparative study of regulated and [32] E. Alptekin, M. Canakci, Determination of the density and the viscosities of bio-
unregulated gaseous emissions during NEDC in a light-duty diesel engine fuelled diesel–diesel fuel blends, Renew. Energy 33 (2008) 2623–2630, http://dx.doi.org/
with Fischer Tropsch and biodiesel fuels, Biomass Bioenergy 35 (2011) 789–798, 10.1016/j.renene.2008.02.020.
http://dx.doi.org/10.1016/j.biombioe.2010.10.034. [33] M.M.K. Bhuiya, M.G. Rasul, M.M.K. Khan, N. Ashwath, A.K. Azad, M.A. Hazrat,
[6] L. Ryan, F. Convery, S. Ferreira, Stimulating the use of biofuels in the European Prospects of 2nd generation biodiesel as a sustainable fuel – part 2: properties,
Union: implications for climate change policy, Energ Policy 34 (2006) 3184–3194, performance and emission characteristics, Renew. Sust. Energ. Rev. 55 (2016)
http://dx.doi.org/10.1016/j.enpol.2005.06.010. 1129–1146, http://dx.doi.org/10.1016/j.rser.2015.09.086.
[7] R.E. Morris, Y. Jia, Impact of Biodiesel Fuels on Air Quality and Human Health: [34] M. Lapuerta, O. Armas, R. Ballesteros, J. Fernandez, Diesel emissions from biofuels
Task 4 Report Impacts of Biodiesel Fuel Use on PM, Novato, California, (2003). derived from Spanish potential vegetable oils, Fuel 84 (2005) 773–780, http://dx.
[8] K.T. Magara- Gomez, M.R. Olson, T. Okuda, K.A. Walz, J.J. Schauer, Sensitivity of doi.org/10.1016/j.fuel.2004.11.010.
hazardous air pollutant emissions to the combustion of blends of petroleum diesel [35] R. Chiriac, N. Apostolescu, Emissions of a diesel engine using B20 and effects of

76
F. Soto et al. Fuel Processing Technology 174 (2018) 69–77

hydrogen addition, Int. J. Hydrog. Energy 38 (2013) 13453–13462, http://dx.doi. [37] M. Aldhaidhawi, R. Chiriac, V. Bădescu, G. Descombes, P. Podevin, Investigation on
org/10.1016/j.ijhydene.2013.07.095. the mixture formation, combustion characteristics and performance of a diesel
[36] C.Y. Choi, R.D. Reitz, An experimental study on the effects of oxygenated fuel engine fueled with diesel, biodiesel B20 and hydrogen addition, Int. J. Hydrog.
blends and multiple injection strategies on DI diesel engine emissions, Fuel 78 Energy 42 (2017) 16793–16807, http://dx.doi.org/10.1016/j.ijhydene.2017.01.
(1999) 1303–1317, http://dx.doi.org/10.1016/S0016-2361(99)00058-7. 222.

77