Professional Documents
Culture Documents
Fuel
journal homepage: www.elsevier.com/locate/fuel
Predicting cetane number, kinematic viscosity, density and higher heating value
of biodiesel from its fatty acid methyl ester composition
Luis Felipe Ramrez-Verduzco , Javier Esteban Rodrguez-Rodrguez, Alicia del Rayo Jaramillo-Jacob
Programa de Procesos de Transformacin, Instituto Mexicano del Petrleo, Eje Central Norte Lzaro Crdenas 152, C.P. 07730 Mxico D.F., Mexico
a r t i c l e
i n f o
Article history:
Received 24 June 2010
Received in revised form 30 May 2011
Accepted 27 June 2011
Available online 21 July 2011
Keywords:
Biodiesel
Fatty acid methyl ester
Prediction
Mixing rules
Physical properties
a b s t r a c t
Biodiesel is a renewable bio-fuel derived from natural fats or vegetable oils, and it is considered as a
promising alternative to substitute diesel fuels. Cetane number, viscosity, density, and higher heating
value are important properties to affect the utilization of biodiesel fuels, because they are involved in
the denition of fuel quality and are required as input data for predictive engine combustion models. This
work presents the characterization of two biodiesel samples made from beef tallow and soybean oil
through their fatty acid methyl esters (FAMEs) prole. Empirical equations were developed to estimate
four physical properties of methyl esters; and an average absolute deviation (AAD) of 5.95%, 2.57%,
0.11% and 0.21% for the cetane number, kinematic viscosity, density, and higher heating value were
founded. Cetane number, viscosity, and higher heating value increases because of the increase of molecular weight and these physical properties decrease as the number of double bonds increases. Unlike that
of above properties, density decreases as molecular weight increases and density increases as the degree
of unsaturation increases. Two general mixing rules and ve biodiesel samples were used to study the
inuence of FAMEs over the physical properties of biodiesel. The prediction of the cetane number, kinematic viscosity, density and higher heating value of biodiesel is very close to the experimental values.
2011 Elsevier Ltd. All rights reserved.
1. Introduction
Decline of available oil reserves and more stringent environmental regulations have motivated the global interest in renewable energy sources. Biodiesel is considered as an attractive
alternative to replace diesel fuels. Biodiesel consists of a mixture
of fatty acid alkyl esters that can be obtained from animal fats or
vegetable oils, mainly by trans-esterication reactions. Biodiesel
has several advantages because it is renewable, biodegradable,
non-toxic, and friendly with environment. Currently, the main
disadvantage could be the high costs of some types of raw materials. However, as the petroleum oil reserves become scarce, this
behavior could be changed. The main differences between biodiesel and diesel are that the former usually has higher density, viscosity, cloud point and cetane number; and lower volatility, gross
caloric value and sulfur content. Cetane number (/), viscosity
(g), density (q) and higher heating value (d) are four important
properties of fuels. Cetane number is widely used to evaluate
the ignition quality of fuels. Viscosity and density have a direct
effect on the atomization process during combustion. Higher
heating value also known as the gross caloric value or gross energy is the amount of heat released during the combustion of one
Corresponding author.
E-mail address: lframir@imp.mx (L.F. Ramrez-Verduzco).
0016-2361/$ - see front matter 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.fuel.2011.06.070
103
Nomenclature
AD
AAD
FAME
Mi
n
N
BT
CO
CT
SB
absolute deviation
average absolute deviation
fatty acid methyl ester
molecular weight of the ith FAME
number of FAMEs
number of double bounds
biodiesel made from beef tallow
biodiesel made from corn oil
biodiesel made from cottonseed oil
biodiesel made from soybean oil
SF
Greek symbols
/
cetane number
g
kinematic viscosity
q
density
d
higher heating value
No. 1, heating rate, nal temperature and hold time No. 2 was
100 C, 4 min, 3 C/min, 240 C and 15 min, respectively. In order
to determine the retention times of the fatty acid methyl esters,
a standard mixture of 37 FAMEs solved in methylene chloride
(with a standard concentration of 30 mg/mL) was run (see Table 2
and Fig. 1). The standard sample that was used in this work has a
certicate of RESTEK Co. (ISO 9001 registered quality system).
Fig. 1 shows the chromatogram obtained by GCMS of the standard
with 37 FAMEs.
104
Table 1
Specication of BT and SB biodiesel samples.
Fuel property
Units
BT
SB
Flash point
Water and sediment
Kinematic viscosity, 40 C
Total sulfur
Cupper strip corrosion
Cetane number
Cloud point
Acid number
Free glycerin
Total glycerin
Distillation temperature, 90% recovered (T90)
C
vol.%
mm2/s
ppmw
C
mg KOH/g
wt.%
wt.%
C
D
D
D
D
D
D
D
D
D
D
D
156
<0.05
4.36
17
No. 1a
49.2
3
0.26
0.005
0.197
357.4
92
<0.05
4.07
1.4
No. 1a
51.7
0
0.23
0.004
0.178
355.0
93 min.
0.05 max.
[1.96.0]
15 max.
No. 3 max.
47 min.
Report
0.50 max.
0.020 max.
0.240 max.
360 max.
93-02a
1796-04
445-04e
5453-98
130-04
613-05
2500-05
664-95
6584
6584
1160-06
Table 2
Retention times obtained by GCMS of the C4:0-C22:6 methyl esters from the standard.
No. peak
Fatty acid
Compound
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
9.73
11.74
15.48
20.81
23.76
26.72
29.62
32.44
34.38
35.11
37.03
37.69
39.21
40.12
41.62
42.46
43.31
43.74
44.68
45.63
46.80
46.97
47.79
48.02
48.83
49.83
50.79
51.11
51.88
52.00
52.08
52.72
53.85
54.30
54.74
56.08
60.16
C4:0
C6:0
C8:0
C10:0
C11:0
C12:0
C13:0
C14:0
C14:1
C15:0
C15:1
C16:0
C16:1
C17:0
C17:1
C18:0
C18:1
C18:1
C18:2
C18:2
C20:0
C18:3
C20:1
C18:3
C21:0
C20:2
C22:0
C20:3
C22:1
C20:3
C20:4
C23:0
C22:2
C20:5
C24:0
C24:1
C22:6
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
Methyl
3. Theoretical basis
Four new empirical correlations to calculate the cetane number,
kinematic viscosity, density, and higher heating value of fatty acid
methyl esters were derived in this work. These equations were obtained by correlating a set of experimental data compiled from dif-
Butyrate
Hexanoate
Octanoate
Decanoate
Undecanoate
Laurate
Tridecanoate
Myristate
Myristoleate (cis-9)
Pentadecanoate
Pentadecenoate (cis-10)
Palmitate
Palmitoleate (cis-9)
Heptadecanoate
Heptadecenoate (cis-10)
Stearate
Elaidate (trans-9)
Oleate (cis-9)
Linoelaidate (trans-9.12)
Linoleate (cis-9,12)
Arachidate
c-Linolenate (cis-6,9,12)
Eicosenoate (cis-11)
Linolenate (cis-9,12,15)
Heneicosanoate
Eicosadienoate (cis-11,14)
Behenate
Eicosatrienoate (cis-8,11,14)
Erucate (cis-13)
Eicosatrienoate (cis-11,14,17)
Arachidonate (5,8,11,14)
Tricosanoate
Docosadienoate (cis-13,16)
Eicosapentaenoate (cis-5,8,11,14,17)
Lignocerate
Nervonate (cis-15)
Docosahexaenoate (cis-4,7,10,13,16,19)
ferent sources [1,1425], Eqs. (1)(4) was adjusted using 21, 46, 36,
and 22 experimental values, respectively. The parameters that are
presented in this work were estimated numerically by using the
software Design-Expert, the variation of physical properties with
molecular weight and degree of unsaturation were properly tted
by means of statistical regression. In all of these empirical correlations, Mi represents the molecular weight of the ith FAME, and N is
the number of double bonds in a given FAME, therefore N is equal
to zero, one, two, and three for the methyl ester C18:0, C18:1,
C18:2 and C18:3, respectively.
The cetane number of each fatty acid methyl ester is obtained
from
/i 7:8 0:302 M i 20 N
where /i is the cetane number of the ith FAME.
105
Fig. 1. Chromatogram obtained by GCMS of the C4:0-C22:6 methyl esters from the standard.
qi 0:8463
4:9
0:0118 N
Mi
1794
di 46:19
0:21 N
Mi
fb
n
X
zi fi
i1
where f is a function that represents any physical property (the subscripts b and i refer to the biodiesel and the pure ith FAME, respectively), zi is the mass or mole fraction of the ith FAME. The function
fb must be replaced by the variables /b, ln(gb), qb and db in order to
specify the cetane number, natural logarithm of kinematic viscosity,
density and higher heating value of biodiesel, where as the function
fi must be interchanged by the variables /i, ln(gi), qi and di in order
to specify the properties of the individual ith FAME.
We are postulating that both mass and mole fraction could be
numerically similar for each ith FAME of specic biodiesel due to
the following facts:
xi
wi
Mi
n
P wi
Mi
i1
where xi and wi are the mole and mass fraction, whereas Mi is the
molecular weight of the ith FAME.
Comparisons between experimental and calculated values of
physical properties of FAMEs or biodiesel can be made by means
of the average absolute deviation (AAD) dened by
!
X
n
fj;exp fj;cal 100
1
AAD
n
fj;exp
j1
where n is the number of experimental points that are being compared, whereas fj,exp and fj,cal are functions that represent the experimental and calculated individual FAME or biodiesel physical property.
4. Results and discussion
4.1. Biodiesel characterization
The molecular weight of each FAME (Mi) and concentration of
ve biodiesel samples are shown in Table 3. The composition reported in Table 3 for the rst two biodiesel samples made from
beef tallow (BT) and soybean (SB) was determined experimentally
106
Table 3
Average molecular weight (M) and concentration (mass fraction and mole fraction in parenthesis) of ve biodiesel samples.
ith FAME
Methyl ester
Mi (g/mol)
wi (xi)
1
2
3
4
5
6
7
8
9
10
C14:0
C16:0
C16:1
C18:0
C18:1
C18:2
C18:3
C20:0
C20:1
C22:1
242.3975
270.4507
268.4348
298.5038
296.4879
294.4721
292.4562
326.5570
324.5411
352.5943
P
M ni1 xi M i
BT
SB
SF [2]
CO [2]
CT [2]
0.005(0.006)
0.159(0.171)
0.009(0.010)
0.085(0.083)
0.317(0.311)
0.365(0.361)
0.046(0.046)
0.003(0.003)
0.008(0.007)
0.003(0.002)
0.000(0.000)
0.101(0.109)
0.000(0.000)
0.045(0.044)
0.243(0.240)
0.531(0.528)
0.072(0.072)
0.004(0.004)
0.000(0.000)
0.004(0.003)
0.001(0.001)
0.060(0.065)
0.000(0.000)
0.059(0.059)
0.160(0.158)
0.714(0.711)
0.006(0.006)
0.000(0.000)
0.000(0.000)
0.000(0.000)
0.000(0.000)
0.099(0.107)
0.000(0.000)
0.031(0.031)
0.291(0.287)
0.568(0.564)
0.011(0.011)
0.000(0.000)
0.000(0.000)
0.000(0.000)
0.008(0.010)
0.229(0.244)
0.000(0.000)
0.031(0.030)
0.185(0.180)
0.542(0.531)
0.005(0.005)
0.000(0.000)
0.000(0.000)
0.000(0.000)
291.0834 g/mol
292.6725 g/mol
293.4029 g/mol
292.5832 g/mol
288.5638 g/mol
BT: beef tallow oil methyl ester, SB: soybean oil methyl ester, SF: sunower oil methyl ester, CO: corn oil methyl ester, CT: cottonseed oil methyl ester.
in this work, whereas the composition of the following three biodiesel samples (SF: sunower oil methyl ester, CO: corn oil methyl
ester and CT: cottonseed oil methyl ester) was reported previously
by Allen et al. [2]. The analysis technique for the identication of
FAMEs developed in this work allowed a clear identication of
the cis and trans isomers of C18:1 and C18:2 methyl esters in
the BT and SB biodiesel samples, because one of the design features
of the polysiloxane polymer capillary column (Agilent HP-88) is
the separation of cis:trans FAMEs. The ratio cis/trans was 6.4/
93.6 for the C18:1 methyl ester in the BT sample, and only the
presence of the cis isomer C18:2 methyl ester was detected; in
the case of the SB sample only the presence of the cis isomer for
both C18:1 and C18:2 methyl esters was conrmed (none trace
of the trans isomer was founded), the concentrations of C18:1
and C18:2 methyl esters presented in Table 3 encompass the contribution of both isomers. In both samples BT and SB, only the cis
isomer for the C18:3 methyl ester was identied. A possible explanation for the predominance of the cis isomer in the soybean biodiesel may be found if we consider the natural biosynthetic process
of plants, where cis double bounds are inserted by desaturase enzymes in two carbon units at specic positions of the fatty acids
chains. The FAME that had the highest concentration in all biodiesel samples given in Table 3 was the methyl linoleate (C18:2
methyl ester). The soybean biodiesel reported by Allen et al. [2]
has a very similar composition with their homologous biodiesel
SB measured in this work.
The mole fraction for the ve biodiesel samples reported in
Table 3 was derived using Eq. (6). The mole fraction is numerically
similar to the mass fraction for each methyl ester in the ve biodiesel samples, mainly due the molecular weight of FAMEs is restricted to a narrow range.
Table 4
Deviations between the experimental data reported in the literature and results
obtained in this work for the soybean biodiesel characterization.
Methyl ester
AD1
AD2
AD3
C14:0
C16:0
C18:0
C18:1
C18:2
C18:3
C18:4
C20:0
C22:1
0.000
0.039
0.005
0.003
0.011
0.072
0.060
0.004
0.004
0.001
0.002
0.002
0.018
0.010
0.011
0.000
0.004
0.004
0.000
0.019
0.015
0.013
0.019
0.012
0.000
0.004
0.004
AD1: Comparison between Srivastava and Prasad [26] and this work.
AD2: Comparison between Allen et al. [2] and this work.
AD3: Comparison between Goering et al. [27] and this work.
107
/
(calculated)
g (mm2/s)
g (mm2/s)
q (g/cm3)
q (g/cm3)
(reported at 40 C)
(calculated at 40 C)
(reported at 20 C)
(calculated at 20 C)
C8:0
33.6a
39.7b
40.0
1.20b
1.17e
1.2150f
1.16g
1.15
0.8771e
0.8771f
0.8775g
0.8769k
0.87754l
C10:0
47.2a
51.6b
48.5
1.71b
1.69e
1.7258f
1.69g
1.72h
1.72
C12:0
61.4a
60.8c
56.9
2.43b
2.38e
2.4331f
2.28g
2.43h
C14:0
66.2a
65.4
C16:0
74.5a
74.3c
C16:1
Methyl ester
a
b
c
d
e
f
g
h
i
j
k
l
m
n
o
d (MJ/kg)
(reported)
d (MJ/kg)
(calculated)
0.8773
34.907b
34.72o
34.85
0.8724e
0.8723f
0.8730g
0.8724k
0.87301l
0.8726
36.674b
36.50o
36.56
2.45
0.8691e
0.8698f
0.8695g
0.8694k
0.86952l
0.8692
37.968b
37.87o
37.82
3.3381f
3.23g
3.30h
3.33
0.8671g
0.8671k
0.86713l
0.8665
38.89o
38.79
73.9
4.38b
4.4136f
4.32g
4.38h
4.37
0.8650m
0.8644
39.449b
39.45o
39.56
51.0a
53.3
3.67b
3.67h
3.0642i
3.59
0.8690i
0.8764
39.293b
39.29o
39.30
C18:0
86.9a
75.6c
82.3
5.85b
5.8675f
5.61g
5.85h
5.59
0.8643n
0.8627
40.099b
40.07o
40.18
C18:1
59.3a
56.5c
61.7
4.51b
4.5728f
4.45g
4.51h
4.60
0.8741f
0.875g
0.8740k
0.87362l
0.8746
40.092b
39.91o
39.93
C18:2
38.2a
41.1
3.79
0.8862f
0.890g
0.8866k
0.8865
39.698b
39.70o
39.68
C18:3
22.7b
22.7c
20.6d
20.5
3.65b
3.7028f
3.64g
3.65h
3.14b
3.27g
3.14h
3.2980i
3.11
0.9019i
0.8979k
0.8985
39.342b
39.34o
39.43
C20:0
100.0c
90.8
7.2097j
7.00
0.8632j
0.8633n
0.8613
40.61o
40.70
C20:1
64.8c
70.2
5.77h
5.9108i
5.77
0.8738i
0.8732
40.61o
40.45
C22:1
76.0c
78.7
7.21g
7.33h
6.9556i
7.09
0.8707i
0.8706k
0.87024l
0.8720
40.99o
40.89
Knothe [14].
Knothe [15].
Tong et al. [16].
Papadopoulos et al. [17].
Liew et al. [18].
Pratas et al. [19].
Yuan et al. [20].
Knothe and Steidley [21].
Pratas et al. [22].
Extrapolate value from measurements at higher temperatures given by Pratas et al. [22].
Gouw and Vlugter [1].
Hammomd and Lundberg [23].
Extrapolate value from measurements at higher temperatures given by Bonhorst et al. [24].
Derived from empirical correlation proposed by Gouw and Vlugter [1].
Freedman and Bagby [25].
as simple as possible, and secondly, the experimental results reported by Knothe [14] for saturated FAMEs show a linear behavior
(see Fig. 2). In addition, it is also proposed though Eq. (1) that the
108
Fig. 4. Spatial geometry of the methyl stearate or C18:0 methyl ester molecule.
White, gray and red spheres represent hydrogen, carbon and oxygen atoms,
respectively. (For interpretation of the references to color in this gure legend, the
reader is referred to the web version of this article.)
Fig. 5. Spatial geometry of the methyl elaidate or C18:1 (trans-9) methyl ester
molecule. White, gray and red spheres represent hydrogen, carbon and oxygen
atoms, respectively. (For interpretation of the references to color in this gure
legend, the reader is referred to the web version of this article.)
Fig. 2. Cetane number (/) as a function of molecular weight (M) for methyl esters
from C8:0 to C18:0.
Fig. 6. Spatial geometry of the methyl oleate or C18:1 (cis-9) methyl ester
molecule. White, gray and red spheres represent hydrogen, carbon and oxygen
atoms, respectively. (For interpretation of the references to color in this gure
legend, the reader is referred to the web version of this article.)
Fig. 7. Spatial geometry of the methyl linoleate or C18:2 (cis-9,12) methyl ester
molecule. White, gray and red spheres represent hydrogen, carbon and oxygen
atoms, respectively. (For interpretation of the references to color in this gure
legend, the reader is referred to the web version of this article.)
Fig. 3. Viscosity (g) at 40 C as a function of molecular weight (M) for methyl esters
from C8:0 to C18:0 in a loglog plot.
109
Fig. 8. Spatial geometry of the methyl linolenate or C18:3 (cis-9,12,15) methyl ester
molecule. White, gray and red spheres represent hydrogen, carbon and oxygen
atoms, respectively. (For interpretation of the references to color in this gure
legend, the reader is referred to the web version of this article.)
Fig. 9. Density (q) at 20 C as a function of molecular weight (M) for methyl esters
from C8:0 to C20:0.
Fig. 10. Higher heating value (d) as a function of molecular weight (M) for methyl
esters from C8:0 to C22:0.
110
Table 6
Cetane number and kinematic viscosity at 40 C in ve biodiesel samples. Comparison between experimental and calculated values.
Biodiesel
BT
SB
SF
CO
CT
/b
(reported)
P
/b ni1 wi /i
(calculated)
P
/b ni1 xi /i
(calculated)
gb (mm2/s)
gb exp
Pn
(reported)
(calculated)
(calculated)
53.8a, 57.8b
51.7a, 49.0c
50.0c
53.0c
51.2d
56.2
50.2
48.7
51.4
53.8
56.3
50.3
48.8
51.6
54.2
4.36a
4.07a, 3.97e, 3.99f
4.03e
4.18e
4.06e
4.29
4.10
4.05
4.13
4.12
4.28
4.10
4.05
4.13
4.12
i1 wi
Ingi (mm2/s)
Pn
gb exp
i1 xi
Ingi (mm2/s)
BT: beef tallow oil methyl ester, SB: soybean oil methyl ester, SF: sunower oil methyl ester, CO: corn oil methyl ester, CT: cottonseed oil methyl ester.
a
Experimental results obtained in this work.
b
Zheng and Hanna [28].
c
Ramos et al. [12].
d
Knothe [15].
e
Alptekin and Canakci [29].
f
Tate et al. [30].
Table 7
Density and higher heating value in ve biodiesel samples. Comparison between experimental and calculated values.
Biodiesel
qb
Pn
3
i1 wi qi (g/cm )
(calculated at 20 C)
qb
db (MJ/kg)
(reported)
0.8850B,b,
0.8774
0.8810
0.8772
0.8809
0.8840B,c
0.8840B,c
0.8836B,c
0.8820
0.8803
0.8784
0.8819
0.8802
0.8781
39.57a
39.61a, 39.80d,
41.28e
40.00d, 41.33e
41.14e
40.15d, 41.18e
qb (g/cm3)
(reported)
BT
SB
SF
CO
CT
0.8795A,a
0.8812A,a,
0.8845B,c
0.8840B,b,
0.8840B,b,
0.8840B,b,
Pn
3
i1 xi qi (g/cm )
(calculated at 20 C)
P
db ni1 wi di (MJ/kg)
(calculated)
P
db ni1 xi di (MJ/kg)
(calculated)
39.77
39.74
39.77
39.74
39.74
39.75
39.70
39.74
39.75
39.70
BT: Beef tallow oil methyl ester, SB: soybean oil methyl ester, SF: sunower oil methyl ester, CO: corn oil methyl ester, CT: cottonseed oil methyl ester.
A
Density at 20 C.
B
Density at 15 C.
a
Experimental results obtained in this work.
b
Alptekin and Canakci [29].
c
Alptekin and Canakci [32].
d
Krisnangkura [33].
e
Demirbas [34].
Higher heating value is the amount of heat produced by the complete combustion of a unit quantity of fuel, this property is obtained
when all products of the combustion are cooled down to the temperature before the combustion and the water vapor formed during combustion is condensed. Higher heating value of biodiesel is
approximately 12% less than that of petro-diesel (39.5741.33 MJ/
kg (Table 7) compared to 46 MJ/kg). The higher heating value of a
fuel increases with increasing carbon number in fuel molecules and
also increases as the ratio of carbon and hydrogen to oxygen and
nitrogen increases [36]. In regard to the prediction, the estimated high
caloric values are in a very close range, with a minimum and maximum of 39.70 MJ/kg and 39.77 MJ/kg, respectively. The AAD between
experimental and calculated high caloric values was 1.92%, when
the mass and mole fractions were used in the mixing rule.
In general, it appears that in the calculation of the four physical
properties (cetane number, kinematic viscosity, density, and higher
heating value) for biodiesel given by Eq. (5), the terms related with
the FAMEs C16:0, C18:0, C18:1, C18:2 and C18:3 have a more signicant contribution in the sum, due the concentration of these
FAMEs is greater with respect to the remainder (FAMEs C14:0,
C16:1, C17:0, C20:0, C20:1, and C22:1).
It is important to mention that the methodology that was developed for the prediction of some physical properties has some
restrictions:
a. Because in order to derive the empirical correlations of
this work was not considered the methyl ricinoleate within
our database, the methodology should not be used in
the biodiesel of castor oil, it must remembered that the
castor oil consists mainly (approximately 90%) of the ricinoleic acid (12-hydroxy-9(Z)-octadecenoic acid) according
to Knothe [15].
111