Evaluation of a glycerol derived biofuel by thermal analysis

Article abstract of DOI:10.1007/s10973-011-1558-x

This work describes thermal analysis evaluation of a glycerol derived compound (fatty acid esters of (2,2-dimethyl-1,3-dioxolan-4-yl) methanol) developed to work as a biofuel. Mixtures of these ketal-glyceryl esters with fatty acid methyl esters typical of soybean biodiesel were prepared and evaluated in relation to biodiesel critical thermal properties such as temperature of crystallization, thermal stability and volatilization measured by differential scanning calorimetry and thermogravimetric analysis. The volatility of the products containing fatty acid methyl esters and (2,2-dimethyl-1,3-dioxolan-4-yl) methyl esters could be predicted by thermogravimetric analyses conducted in nitrogen that avoided time consuming distillation and greatly reduced material expenditure.

Full text of DOI:10.1007/s10973-011-1558-x

J Therm Anal Calorim (2011) 106:697–702
DOI 10.1007/s10973-011-1558-x
Evaluation of a glycerol derived biofuel by thermal analysis
•
Luciano N. Batista Rosane A. S. San Gil
•
•
Luciana B. R. Lopes Vera L. P. Soares
CBRATEC7 Conference Special Issue
´
´
ꢀ Akademiai Kiado, Budapest, Hungary 2011
Abstract This work describes thermal analysis evalua-
tion of a glycerol derived compound (fatty acid esters of
(2,2-dimethyl-1,3-dioxolan-4-yl) methanol) developed to
work as a biofuel. Mixtures of these ketal–glyceryl esters
with fatty acid methyl esters typical of soybean biodiesel
were prepared and evaluated in relation to biodiesel critical
thermal properties such as temperature of crystallization,
thermal stability and volatilization measured by differential
scanning calorimetry and thermogravimetric analysis. The
volatility of the products containing fatty acid methyl esters
and (2,2-dimethyl-1,3-dioxolan-4-yl) methyl esters could
be predicted by thermogravimetric analyses conducted in
nitrogen that avoided time consuming distillation and
greatly reduced material expenditure.
requirements for a compound to act as a fuel is that it must
present a high cetane number which reflects the delay in
ignition time and also indicates a probable low emission of
CO, NOx and particulates [1]. Biodiesel from soybean oil
presents a high cetane number comparable to that of pet-
rodiesel due to its high O/C ratio and to the predominance
of linear alkyl chains (high CH2/CH3). A second structural
feature that can influence the cetane number is the presence
of oxygen. Oxygen being an electronegative element leaves
a C–H bond at the same carbon atom as oxygen weaker and
easier to be broken, which can contribute to the onset of
combustion [1]. The better lubricity performance of bio-
diesel when compared to that of petrodiesel is attributed to
the presence of carboxyl groups in alkyl esters [2]. The
ketalization of glycerol hydroxyl groups which keeps the
oxygen atoms in the molecular structure increases its vol-
atility and reduces viscosity. This is a well known reaction
to protect reactive groups or prepare volatile products in
organic syntheses. The preparation of glyceryl ketals and
glyceryl esters to be used as biofuels has been claimed by
some authors; however, neither the fuel performance nor
their critical properties have been discussed by them [3, 4].
This paper describes the evaluation of thermal properties of
ketal–glyceryl esters developed to work as a biofuel, the study
of the influence of these glyceryl derivatives on the thermal
properties of a soybean biodiesel and a method to predict
products distillation behaviour by thermal gravimetric analysis.
Keywords Glycerol Á Thermal analysis Á Biofuel Á
Ketal–glyceryl esters Á Dioxolanyl esters Á NMR
Introduction
Due to the success of the implementation of biofuel in
Brazil, to its increase in production and to the gradual
alteration of the ratio of biofuel in diesel oil, it became
urgent to exploit the glycerol by-product. As the amount of
glycerol predicted would be more than could be absorbed
in a single industrial segment, a molecule derived from
glycerol with biofuel features was tailored. One of the main
Experimental
L. N. Batista Á R. A. S. San Gil Á L. B. R. Lopes Á
V. L. P. Soares (&)
Materials
´
Instituto de Quımica, Universidade Federal do Rio de Janeiro,
Rua Athos da Silveira Ramos, 149, bloco A,
s/616, Rio de Janeiro 21949-909, Brazil
e-mail: veralps@iq.ufrj.br
The reagents and solvents used were ethanol, methanol, glyc-
erol, chloroform, methyl palmitate (methyl hexadecanoate),
123

698
L. N. Batista et al.
methyl oleate (methyl 9Z-octadecenoate), methyl stearate
(methyl octadecanoate) and acetone from VETEC S.A.,
Brazil; methyl linoleate (methyl 9Z, 12Z-octadedienoate)
and p-toluenesulfonic acid from Aldrich, USA; a commer-
cial soybean oil (SO) and a biodiesel from soybean oil
transesterified with methanol (SB). SB was composed of
16% of methyl palmitate, 3% of methyl stearate, 29% of
methyl oleate and 52% of methyl linoleate according to gas
chromatographic analysis. The purity of all reagents was
checked using infrared and nuclear magnetic resonance
spectrometry and gas chromatography when appropriate,
and they were purified by distillation or recrystallization
when necessary.
linked (5% phenyl)-methyl-silicone; the column tempera-
ture was programmed from 75 to 250 ꢁC at 10 ꢁC min-1
and the detector voltage was 75 eV.
Two products batches containing ketal–glyceryl esters
and methyl esters in the proportions 25:75 (BDM25) and
65:35 (BDM65), respectively, were evaluated in relation to
the following thermal properties: temperature of crystalli-
zation at a cooling rate of 10 ꢁC min-1, in N2 atmosphere,
and oxidation stability from 30 to 300 ꢁC in air atmosphere
in a differential scanning calorimeter (Perkin-Elmer model
7); thermogravimetric analysis in N2 atmosphere, at
20 ꢁC min-1, up to 800 ꢁC (Perkin-Elmer model 7e) and
distillation behaviour according to ASTM D1160 [5]. A
sample of biodiesel was analysed in these same conditions
and a sample of (2,2-dimethyl-1,3-dioxolan-4-yl) methyl
palmitate (PDM) was submitted to thermal analyses.
Methods
The preparation of the product of interest II (Fig. 1)
involved in a first step, the ketalization of glycerol to give
product I (Fig. 1) and, in a second step, the esterification of
I to furnish the desired product II. For the first step, a
mixture of glycerol, propanone, added in excess, and
p-toluenesulfonic acid (catalyst) in chloroform were
refluxed for 10 h while water was being removed with
a Dean–Stark apparatus. The ketal-glycerol (I), that is
(2,2-dimethyl-1,3-dioxolan-4-yl) methanol, was isolated by
vacuum fractional distillation after adding sodium car-
bonate to neutralize the catalyst and filtrating the catalyst.
The second step was the transesterification of a mixture of
methyl esters, consisting of palmitate, oleate, linoleate and
stearate in the same proportion as found for biodiesel SB,
with the product I. This reaction was conducted in the
presence of anhydrous sodium carbonate, for 6 h at
125–140 ꢁC, generating the product II that was purified by
filtration of the catalyst and distillation of the excess of
product I. A transesterification of methyl palmitate with
product I was also conducted and produced (2,2-dimethyl-
1,3-dioxolan-4-yl) methyl palmitate (PDM).
Results and discussion
Products characterization
Characterization of (2,2-dimethyl-1,3-dioxolan-4-yl)
methanol (I)
The FTIR spectrum (Fig. 2) of the ketal (I) after distillation
presented: the typical C–O band at 1072 cm-1, an OH band
at 3404 cm-1 and a C–O band at 1042 cm-1 relative to the
non-reacted hydroxyl, as expected, and the typical CH3
bend vibrations from the methyl groups at the dioxolanyl
radical at 1375–1372 cm-1. No carbonyl bands from ace-
tone were detected.
1
The structure of product I was also confirmed by H
NMR (Fig. 3) through the CH3 signals at 1.36 and
1.43 ppm.
The products were characterized by solution 1H and 13
C
nuclear magnetic resonance spectroscopy in an NMR
Bruker DPX-200 (4.7 T), by Fourier-transform infrared
spectrometry in a Nicolet FT-IR, model Magna 740, and by
high resolution gas chromatography with mass spectrom-
etry detector (Agilent Technologies model CG6850 and
EM5973), provided with a 30 m capillary column of cross-
80
70
60
O
50
O
40
30
HO
20
R2
R2
10
0
R1
O
R1
O
O
O
HO
3500
3000
2500
2000
1500
1000
500
O
O
Wavenumber/cm–1
R
Fig. 1 Structure of intermediate I and product II (R = 15–17
carbons; R1, R2 = CH3)
Fig. 2 FTIR spectrum of (2,2-dimethyl-1,3-dioxolan-4-yl) methanol
(I)
123

Evaluation of a glycerol derived biofuel by thermal analysis
699
7.20
a'
a
f
H
e
H
–
H3C
OH
O
Esters of (2,2-dimethyl-1,3-dioxolan-4-yl)
b
Hg
methanol
Methyl esters
Hd
Hc
O
CH3
a'
11.29
d
e,f,c
6.45
g
9.49
b
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0 ppm
11.35
11.45
7.45
8.00
Fig. 3 1H NMR spectrum of (2,2-dimethyl-1,3-dioxolan-4-yl) meth-
anol (I)
4.00
6.00
10.00
12.00
14.00
Time/min
Fig. 6 GC analysis of BDM65 product
80
70
60
50
40
30
20
acid residue type, and also at 178 ppm due to methyl esters,
besides the signal corresponding to the ketal carbon at
109.9 ppm.
It was observed that conversion times longer than 10 h
resulted in darkened products, so that it was decided to let
some methyl esters in the final mixture. The chromato-
graphic analysis with mass detector (Figs. 6, 7) also con-
firmed the presence of ketal–glyceryl esters (peaks with
retention times between 9.45 and 11.45 min) and methyl
esters (peaks between 6.45 and 7.40 min). The interpreta-
tions of mass spectra fragments were based on the literature
[6–9]. Both spectra have in common the fragments m/z 43
and 101 derived from the 2,2-dimethyl-1,3-dioxolan-4-yl
group, as well as the [M - 15]? peaks at m/z 355, 379 and
381 corresponding, respectively, to the dioxolanyl-methyl
esters of palmitate (PDM), linoleate (LDM) and oleate
(ODM). The fragments m/z 312, 336 and 338 correspond to
[M - 15 - 43]?• from PDM, LDM and ODM, respec-
tively, and, although not always abundant, again confirmed
that the dioxolanyl-methyl esters were successfully
prepared.
4000
3500
3000
2500
2000
1500
1000
500
–1
Wavenumber/cm
Fig. 4 FTIR spectrum of ketal–glyceryl esters
Characterization of esters of (2,2-dimethyl-1,3-dioxolan-4-
yl) methanol (II)
In Fig. 4 a typical FTIR spectrum of ketal–glyceryl esters
(II) is depicted, showing ester bands at 1742 and
1213 cm-1, and the disappearance of OH bands, while 13
C
NMR spectrum (Fig. 5) showed chemical shifts at 173.7
and/or 174.3 ppm for ester carbonyls, depending on the
Fig. 5 13C NMR spectrum
of ketal–glyceryl esters
29.4
22.7
109.9
14.2
31.9
173.7
178.2
73.7
66.4
34.2
64.6
ppm
180 170 160 150 140 130 120 110 100 90 80 70 60 50 40 30 20
123

700
L. N. Batista et al.
Fig. 7 Mass spectra of
chromatographic peaks of
BDM65 interpreted as: a PDM
at 9.49 min; b LDM ? ODM at
11.29 min
355
(a)
43
101
57
110
71
171
185
85
80
269
312
239
213
283
297
255
199
130
157
40
60
100 120 140 160 180 200 220 240 260 280 300 320 340 360
m/z
43
129
(b)
379
381
101
57
336
185
83
230
295
265
149
207
165
223
311
396
355
40
60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400
m/z
Thus, the results of FTIR, 1H and 13C NMR and GC–MS
analyses confirmed the formation of the ketals (I) in the
first step and of ketal-esters (II) in the second step. The
NMR and GC–MS quantitative analyses characterized two
product batches as containing ketal–glyceryl esters and
methyl esters in the proportions 25:75 (BDM25) and 65:35
(BDM65), respectively. Considering that the methyl esters
present in BDM products were constituted of biodiesel
main components these products were evaluated in relation
to biodiesel critical thermal properties considering them as
mixtures of biodiesel and ketal–glyceryl esters.
biodiesel it depends on methyl esters composition [10],
specifically, on the percentages of methyl esters of stearic
and palmitic acids. These are both solids at ambient tem-
perature, and, on cooling their solution, as in biodiesel,
these components precipitate or form a gel [11]. The DSC
curve for biodiesel (SB) showed a crystallization onset at
-7.9 ꢁC with two peak maxima at about -8.7 and -15 ꢁC,
which is typical of soybean methyl biodiesel at the present
DSC experimental conditions. The products BDM65 and
BDM25 presented a similar qualitative behaviour, but with
less intense peaks, that is, a smaller variation of enthalpy
per gram, and a small increase in Tc as shown in Fig. 8. A
higher crystallization temperature would be expected for
the products containing ketal–glyceryl esters, because of its
greater average molecular mass. Corroborating with this
expectation, the (2,2-dimethyl-1,3-dioxolan-4-yl) methyl
palmitate (PDM), prepared in this work, presented a higher
Temperature of crystallization
Temperature of crystallization (Tc) is one of the most
critical properties of a biodiesel: as crystallization starts,
viscosity increases leading to a higher pour point. In
123

Evaluation of a glycerol derived biofuel by thermal analysis
701
Onset = 128.2 °C
Biodiesel
Biodiesel
Peak = 156.0 °C
Onset: –7.9 °C
Peak: –8.7 °C
Onset: –11.2 °C
Peak: –15.5 °C
BDM25
Onset: –6.8 °C
Onset: –10.1 °C
Peak: –7.2 °C
Peak: –12.4 °C
Onset = 126.7 °C
BDM25
BDM65
BDM65
Onset: –5.3 °C
Onset: –14.6 °C
Peak = 153.5 °C
Peak: –6.3 °C
Peak: –16.5 °C
–30 –25 –20 –15 –10
–5
0
5
10
15
20
25 30
Onset = 143.1 °C
Temperature/°C
Peak = 177.7 °C
160 180
Fig. 8 Crystallization behaviour of biodiesel, BDM25 and BDM65
by DSC
60
80
100
120
140
200
Temperature/°C
Fig. 10 Oxidation behaviour of BDM 25, BDM 65 products and
biodiesel
Peak = 68.00 °C
Onset = 59.72 °C
10
8
Peak = 48.84 °C
Onset = 53.50 °C
6
4
(a)
(b)
step can either react with other radicals, resulting in high
molecular weight insoluble sediments and gums, or break
apart to form carboxylic acids [13, 14]. In the present study
this process would be a concern as unsaturated chains are
present. From DSC analysis (Fig. 10), the BDM65 product
presented oxidation onset at 143 ꢁC, BDM25 at 127 ꢁC and
SB at 128 ꢁC. This is a significant result indicating that the
glycerol part of the structure did not enhance the oxidation
process. Its presence in a higher amount, as in BDM65,
seemed to work in the opposite direction helping to delay
oxidation. It could also be an effect due to the presence of
saturated derivatives such as palmitic ester (PDM) that do
not oxidize before 200 ꢁC.
2
PDM
0
Methyl palmitate
–2
–4
–6
–8
Onset = 19.12 °C
Peak = 16.97 °C
–10
–12
–14
Onset = 49.63 °C
Peak = 51.58 °C
10
20
30
40
50
60
70
80
90
Temperature/°C
Fig. 9 DSC of PDM showing a melting and b crystallization
temperatures influenced by the presence of methyl palmitate
Distillation range determined by TGA and ASTM
melting point (66 ꢁC) than methyl palmitate (30 ꢁC)
[11, 12] and crystallized at 51 ꢁC, under the present DSC
experimental conditions (Fig. 9). However, the fact that
BDM products presented crystallization temperatures near
that of biodiesel (SB) can be explained by its high content
in unsaturated fatty acid chains that would bring down the
Tc of the mixture relatively to derivatives with saturated
fatty acid chains, as it is observed for the methyl esters
mixtures that constitute soybean biodiesel and, as well, by
the fact that BDM contains methyl esters.
The distillation curve gives a pattern of the volatility of the
components providing important information. In the case
of diesel, the distillation curve following ASTM 1160 is
used to determine cetane index (ASTM D4737) [5].
However, this essay requires a great amount of product
which is not convenient at an early stage of product
development. Instead, a test for volatilization was con-
ducted in a thermogravimetric analyser, in N2 atmosphere,
and compared to the results obtained by ASTM distillation.
According to the Brazilian agency for fuel regulation
(ANP), 90% of biodiesel B100 must distillate below
360 ꢁC. Table 1 gives the temperature of distillation
determined by ASTM method for the biodiesel SB and
product BDM25 at 10, 50 and 90% volume cuts established
by ASTM. The product BDM25 and the biodiesel SB were
inside the 90% limit.
Oxidation stability
An important characteristic to be considered in soybean oil
derived biofuel is its stability with respect to oxidation
during storage. The process of oxidation of soybean oil or
of its methyl esters, by the action of atmospheric oxygen,
starts at the allyl carbon present in the oleate, linoleate and
linolenate chains, which constitute about 80% of biodiesel
from soybean oil. Hydroperoxides formed at the initiation
The TG analysis in N2 atmosphere (Fig. 11) showed the
following mass losses at 360 ꢁC: 98% for biodiesel SB,
123

702
L. N. Batista et al.
to be added to biodiesel could be a solution to the problem
of utilizing the excess of the by-product glycerol.
Table 1 Distillation behaviour of biodiesel and BDM25, following
ASTM
Sample
Temperature of distillation/ꢁC
Acknowledgements The authors gratefully acknowledge CAPES
(Coordenac¸a˜o de Aperfeic¸oamento de Pessoal de Nivel Superior) for
providing a scholarship for Luciano N. Batista.
Distilled volume/%
10
50
90
SB
188.6
199.0
192.9
214.8
357.9
356.8
BDM25
References
1. Aboud-rachid H, Bonneviot L, Xu G, Kaliaguine S. On the cor-
relation between kinetic rate constants in the autoignition process
of some oxygenates and their cetane number: a quantum chemical
study. J Mol Struct. 2003;621:293–304.
2. Knothe G, Steidley KR. Lubricity of components of biodiesel and
petrodiesel. The origin of biodiesel lubricity. Energy Fuels. 2005;
19:1192–200.
3. Hillion G, Delford B, Durand I. Method for producing biofuels,
transforming triglycerides into at least two biofuel families: fatty
acid monoesters and ethers and/or soluble glycerol acetals. WO
2005/093015 A1. http://v3.espacenet.com/publicationDetails/bib
lio?DB=EPODOC&adjacent=true&locale=en_EP&FT=D&date=
20051006&CC=WO&NR=2005093015A1&KC=A1.
4. Dubois JL, Iborra-Chornet S, Canos AC, Velty AI. Diesel fuel
containing long chain glycerol acetals. EP application EP209482,
sep. 2009. http://v3.espacenet.com/publicationDetails/biblio?DB=
EPODOC&adjacent=true&locale=en_EP&FT=D&date=20080502
&CC=WO&NR=2008050030A2&KC=A2.
100.0
50.00
360 °C (limit)
(d)
(c)
(b)
(a)
(a) Soybean methyl biodiesel
(b) BDM25
(c) BDM65
(d) Soybean oil
0.00
0.00
200.00
400.00
Temperature/°C
600.00
800.00
Fig. 11 Thermogravimetric curves in N2 atmosphere for (a) biodie-
sel; (b) BDM25; (c) BDM65; (d) soybean oil—showing their different
volatilization behaviour
5. ASTM Standard D-1160, Standard Test Method for Distillation
of Petroleum Products at Reduced Pressure; ASTM Standard
D-4737-04, Standard Test Method for Calculated Cetane Index
by Four Variable Equation. ASTM International, West Cons-
hohocken, PA, 2003.
6. McCloskey JA, McClelland MJ. Mass spectra of O-isopropyli-
dene derivatives of unsaturated fatty ester. J Am Chem Soc.
1965;87:5090–3.
7. Kyeremateng SO, Amado E, Kressler J. Synthesis and charac-
terization of random copolymers of (2,2-dimethyl-1,3-dioxolan-
4-yl) methyl methacrylate and 2,3-dihydroxypropyl methacrylate.
Eur Polym J. 2007;43:3380–91.
93% for BDM25, 87% for BDM65 and 27% for soybean
oil that were interpreted as volatilization. The TG curve for
(2,2-dimethyl-1,3-dioxolan-4-yl)-methyl palmitate (PDM)
was very close to the BDM65 curve.
These results showed that dioxolanyl-methyl esters
promoted an increase in the final distillation temperature of
the products. However, considering the TG analysis and the
ASTM distillation curve for BDM25, it can be inferred that
as much as 25% of dioxolanyl esters could be added to
biodiesel methyl esters without interfering with its vola-
tilization behaviour. Most interesting, this technique
allowed a prediction of the distillation behaviour of the
product BDM65 without being necessary to perform a
distillation according to ASTM.
´
8. Batista LN. Desenvolvimento de um biocombustıvel a partir de
glycerol. MSC Thesis. Instituto de Quımica, Universidade Fed-
´
eral do Rio de Janeiro (May 2008). http://teses2.ufrj.br/Teses/
IQ_M/LucianoDoNascimentoBatista.pdf.
´
9. Garcia E, Laca M, Perez E, Garrido A, Peinado J. New class of
acetal derived from glycerol as biodiesel fuel component. Energy
Fuels. 2008;22:4274–80.
10. Knothe G. Dependence of biofuel fuel properties on the structure
of fatty acid alkyl esters. Fuel Process Technol. 2005;86:
1059–70.
11. Nascimento RSV, Soares VLP, Albinante SR, Barreto LR. Effect
of ester-additives on the crystallization temperature of methyl
hexadecanoate. J Therm Anal Calorim. 2005;79:249–54.
12. Soares VLP, Nascimento RSV, Albinante SR. Ester-additives as
inhibitors of gelification of soybean oil methyl esters in biodiesel.
J Therm Anal Calorim. 2009;97:621–6.
13. McCormick RL, Ratcliff M, Moens L, Lawrence R. Several
factors affecting the stability of biodiesel in standard accelerated
tests. Fuel Process Technol. 2007;88:651–7.
14. Candeia RA, Sinfronio FSM, Bicudo TC, Queiroz N, Barros
Filho AKD, Soledad LEB, Santos IMG, Souza AL, Souza AG.
Influence of the storage on the thermo-oxidative stability of
methyl and ethyl esters by PDSC. J Therm Anal Calorim. 2011.
doi:10.1007/s10973-011-1410-3.
Conclusions
The thermal properties measured for the BDM products,
the (2,2-dimethyl-1,3-dioxolan-4-yl)-methyl esters of fatty
acids, presented values close to those obtained for a soy-
bean methyl biodiesel. Thus the substitution of methyl
alcohol by a glycerol derivative in the structure of fatty
esters resulted in a product with good potential to be used
as a biofuel according to the measured thermal properties.
The conversion of such a glycerol derivative into a product
123