Available online at www.sciencedirect.

com

ScienceDirect
Energy Procedia 61 (2014) 1852 – 1859

The 6th International Conference on Applied Energy – ICAE2014

Comparison of biodiesel production scenarios with

coproduction of triacetin according to energy and GHG
emissions
Stefanie Van Dammea *, Svend Brama, Francesco Continoa
a
Vrije Universiteit Brussel, Department of Mechanical Engineering Pleinlaan 2, 1040 Brussels, Belgium

Abstract

A solution for the limited fossil oil resources could be biodiesel. However, the future of biodiesel is threatened by
the use of vegetable oil as biodiesel feedstock and the low economic value of the byproduct glycerol.
The use of algal oil as feedstock and triacetin instead of glycerol as byproduct could be a solution. Triacetin can
be used as a biodiesel additive, so more biofuel will be produced. However the net energy gain and the greenhouse
gas (GHG) reduction for this changed biodiesel production process needs to be analyzed.
To this end, we compared conventional biodiesel production from algal oil with three scenarios to produce
biodiesel and triacetin: catalytic and enzymatic interesterification, and catalytic esterification of glycerol.
The energy content of the produced biofuel from the same mass of algal oil increases, when triacetin is formed.
The increase compared with conventional biodiesel production is 11.0, 2.8 and 5.6% for the catalytic and enzymatic
interesterification reactions and the catalytic esterification of glycerol respectively.
However, the energy required for the production increases with a similar, even higher, value, leading to no impro-
vement in the net energy. The energy demand is mostly depending of the reactant production from fossil fuel.
Therefore the influence of changing the reactant production process to an alternative process from biomass was
assessed. The results showed that the net energy values of all the scenarios increase and that interesterification
scenarios and the biodiesel scenario have a similar net energy value. Regarding the GHG emissions, conventional
biodiesel scores best. Here, too, the production of the reactants plays an important role.
In conclusion, the reactant production from fossil sources leads to a lower net energy value for the scenarios with
triacetin coproduction in comparison with the conventional biodiesel production. However, when the production
process is changed to a biomass based process the interesterification scenarios and conventional biodiesel scenarios
have similar results.
© 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
© 2014 The Authors. Published by Elsevier Ltd.
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Selection
Peer-reviewand/or peer-reviewofunder
under responsibility responsibility
the Organizing of ICAE
Committee of ICAE2014
Keywords: Biodiesel; Triacetin; Energy balance; GHG emissions

* Corresponding author. Tel.: +32 (0)2 629 31 82

E-mail address: [email protected].

1876-6102 © 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Peer-review under responsibility of the Organizing Committee of ICAE2014
doi:10.1016/j.egypro.2014.12.228
 Stefanie Van Damme et al. / Energy Procedia 61 (2014) 1852 – 1859 1853

Nomenclature
AA acetic acid IR interesterification reaction R reactant
BD biodiesel LHV lower heating value RT reaction time
ER esterification reaction MA methyl acetate T temperature
GHG greenhouse gas ME methanol TRI triglycerides

1. Introduction

Many countries promote biofuels produced from non-edible oils. Hence a switch in biodiesel
feedstock will occur from edible vegetable oils to non-edible vegetable, waste or algal oils.
After the switch glycerol will still be formed as a byproduct, since edible, non-edible, waste and algal
oils consist of the same chemical molecules, triglycerides, and no radical changes in the biodiesel
production reaction are expected. The crude glycerol produced leads to an excess supply on the market,
which negatively affects the price of glycerol and the overall economic feasibility of the process.
Therefore new applications for, or the elimination of glycerol are needed.
The conventional biodiesel process can be changed so that not glycerol is formed as byproduct but
triacetin. Triacetin is a triglyceride, consisting of a glycerol backbone, esterified with three acetate
groups. Since it is an ester, it is miscible with biodiesel and can be used as a biofuel additive. The biofuel
production can increase with maximal 20 wt% when the glycerol is fully converted to triacetin. However,
the European standard about biofuels (EN 14214) restrains triacetin to 10 wt%. No limitation is found for
the American Society for Testing and Materials guidelines [1, 2].
At the moment, biodiesel is formed by the transesterification reaction of triglycerides and an alcohol,
mostly methanol (Appendix A, a). An alkaline catalyst like NaOH/KOH or CH3ONa is used in the
industry [3].
Biodiesel with triacetin as co-product can be produced in two different ways. If the alcohol (methanol,
ME) is substituted by an ester (methyl acetate, MA) the transesterification reaction changes in an
interesterification reaction (Appendix A, b) and triacetin is formed directly as a co-product. The
interesterification reaction has already been investigated with a chemical or enzymatic catalyst [4, 5, 6].
The second option is to perform the conventional transesterfication reaction and add an extra
esterification step of the glycerol with acetic acid (AA) to form triacetin (Appendix A, c). This additional
reaction has been investigated by Zhou et al. [7].
To determine if the coproduction of triacetin is positive in terms of energy and GHG, we examined
four scenarios: (1) the conventional biodiesel production (BD), the direct (2) catalytic and (3) enzymatic
interesterification (IR) with methyl acetate and (4) the conventional biodiesel production followed by
catalytic esterification (ER) of glycerol with acetic acid. This paper presents the comparison of these
scenarios.

2. Methodology

To establish the energy balance and calculate the GHG emissions we conducted three steps. In the first
step, we collected literature data about the assessed reactions. Next we determined what the energy
analysis comprises and finally, we compared the four alternative production processes.

2.1. Data from the literature

Data to create the energy balance for the four assessed scenarios was derived from the literature. We
selected for each scenario a reference paper from which six reaction parameters were extracted. The
1854 Stefanie Van Damme et al. / Energy Procedia 61 (2014) 1852 – 1859

Table 1. Experimental data derived from literature and used to calculate the energy balance

Process T RT Reactant Excess ratio Catalyst/Enzyme Yield BD Yield T Ref.

(°C) (h) (mol TRI: (wt%) (mol%) (mol%)
mol R)
1. BD Catalytic 60 2 ME 1:6 CH3ONa 1 97 0 [3]
2. IR Catalytic 50 0.25 MA 1:50 MeOK 0.8 97 81 [6]
3. IR Enzymatic 40 94 MA 1:20 C. antarctica 23 82 77 [4]
4. BD Catalytic 60 2 ME 1:6 CH3ONa 1 97 0 [3]
+ ER Catalytic 110 5 + AA 1:9 Amberlyst-15 0.5 0 45 [7]

parameters were the reaction temperature and time, the amount of catalyst needed, the excess ratio and
the derived yields for biodiesel and triacetin. The values of the parameters can be found in Table 1.
Each scenario contains an equilibrium reaction. To produce biodiesel out of one triglyceride molecule,
three molecules of reactant are needed. However to force the reaction to the biofuel side, an excess
amount of reactant is added. The exact quantity is expressed by the excess ratio.
The added catalyst/enzyme is expressed by wt% of the amount of oil.

2.2. Energy analysis

For each scenario the studied system includes the assessed equilibrium reaction and the combustion of
the produced fuels. Only the reaction to form the biofuel and the energy content of the derived biofuel
differ between the scenarios. Other production steps, like transport and storage of the obtained biofuel,
are assumed the same for the different scenarios and have not been assessed. The energy to store and
transport biodiesel from rapeseed to a service station is 0.02 and 0.04 GJ per GJ biodiesel produced
respectively, according to the EcoInvent database [8].
The energy content of the algal oil does not influence the result, since the chemical composition is
chosen to be the same for each scenario. Therefore, the algal oil and production is not taken into account.
Jorquera et al. made a life-cycle analysis of algal biomass and oil production and found that 0.35 GJ of
energy is needed to produce 1 GJ of biodiesel [9]. The numbers found in Slade et al. stated that the energy
needed for algal biomass production is 0.24 GJ.GJ-1 biodiesel produced [10].
The assessed reactions are divided in three energy and GHG contributions, which are (1) the
production of the used reactant, (2) heating during the reaction of the reaction mixture and (3) the
recovery of the excess reactant. The energy needed to produce the reactants and the GHG emitted by the
production are extracted from the EcoInvent database [11, 12]. The other energy consumptions are
estimated with the use of thermodynamic properties of the assessed reactants derived from literature [13,
14, 15, 16, 17] and no heat recovery or insulation is taken into account as a conservative choice. The only
outgoing energy flow is the produced fuel. The energy content of the fuel is calculated from the LHV of
biodiesel and triacetin. We considered that carbon dioxide from biogenic origin should not be taken into
account when the GHG emissions of biofuels are calculated.
The energy balance and the GHG emissions are expressed per GJ biofuel produced.
Other assumptions are:
x The catalyst or enzyme is reusable, so the production is not taken into account. Also the
recovery of the catalyst or enzyme is not assessed.
x The resources to build the plant for the equilibrium reaction are not considered. To build an
esterification plant for conventional biodiesel production, 9.11ή10-4 GJ per GJ of biodiesel
produced is used according to EcoInvent [8].
x No allocation for energy or GHG is made to glycerol in the BD scenario, since glycerol is
not soluble in biodiesel and can not be directly used as a fuel additive.
 Stefanie Van Damme et al. / Energy Procedia 61 (2014) 1852 – 1859 1855

3. Results and discussion

In this section, we calculated the energy balance and the GHG emissions for each scenario. We also
assessed the influence on the results of an alternative production process for the reactants from biomass.

3.1. Energy analysis

All the scenarios where triacetin is coproduced, consume more energy than the BD scenario to
produce 1 GJ of biofuel (Fig. 1a). The net energy, which is the energy content of the produced biofuel
minus the energy needed for production normalised with the energy content of the biofuel, varies from
0.94 to 0.76 for the different scenarios. The BD scenario has with 0.94 the highest value, which means
that the energy consumption for the production of the biodiesel is only 6% of the energy content of the
produced biodiesel. The energy consumption can be divided in energy needed for the reaction (Fig. 1b)
and energy needed for the production of the reactants (Fig. 1c). Both figures are expressed in GJ
consumed for each GJ produced as biofuel. For both contributions the BD scenario scores best, resulting
in the highest net energy.
The energy consumed in the equilibrium reaction is heating to the reaction temperature and
afterwards the recovery of the reactant (Fig. 1b). The catalytic IR scenario has the highest energy
consumption due to the high value for the recovery. The related excess ratio of 50 is the explanation for
this. Despite the low reaction temperature of 40°C for the enzymatic IR scenario, the heating contribution
is the highest. This can be explained by the reaction time of 4 days.
The production of the used reactants is an important factor (Fig. 1c). The difference between the BD
scenario and the scenarios with triacetin coproduction is explained by the used reactant. The production
of methyl acetate and acetic acid consumes more energy to produce the same mass of reactant compared
to the production of methanol.
On the other hand, less algal oil is needed to produce 1GJ biofuel if triacetin is co-produced. In Fig. 1d
the consumption of algal oil is expressed in kg algal oil needed to produce 1 GJ of biofuel. If we convert
1 kg of algal oil to biofuel, the energy content of the produced biofuel increases with 11.0, 2.8 and 5.6%
for the catalytic and enzymatic IR and the catalytic ER respectively in comparison with the BD scenario.
This aspect can be important as in further research the algal oil production is taken into account.
Although the scenarios with triacetin production produce more biofuel from the same amount of algal
oil, their net energy is lower than the value for the BD scenario. The reactant production consumes more
energy than that the extra triacetin fraction can deliver as fuel.

3.2. GHG emissions

The GHG emitted during the production and the combustion of 1 GJ biofuel, are higher for the
scenarios where triacetin is coproduced. Again the emissions can be subdivided in emissions during the
equilibrium reaction (Fig. 1e) and emissions released by the production of the reactants and the
combustion of the biofuels (Fig. 1f). The combustion of the biofuels emits GHG gasses, since it is
assumed that the reactants are made from fossil fuels.
The GHG emissions for the equilibrium reaction follows the same trend as the related energy
consumption (Fig. 1e vs 1b). The production of the reactant and the combustion are the main GHG
emission sources (Fig. 1f), the emissions are one order of magnitude higher than the emission of the
equilibrium reaction (Fig. 1e). The production of methyl acetate and acetic acid emits more GHG than the
methanol production and more carbon atoms of fossil fuel origin ends up in the biofuel, so the CO2
emissions by combustion increase, are the two main explanations.
1856 Stefanie Van Damme et al. / Energy Procedia 61 (2014) 1852 – 1859

Figuur 1: The net energy of the 4 scenarios is shown in (a). For the BD scenario the highest value of 0.94 is found meaning that
6% of the energy content of the produced biofuel is needed for production. However when the reactants are produced from biomass
the net energy values raise to the italic values and the BD and IR scenarios have similar results. The energy consumption for
production is the sum of the energy needed during the equilibrium reaction (b) and to produce the reactants out of fossil resources
(c). Both figures are expressed in GJ consumed.GJ-1 biofuel produced. In (c) the italic values are the possible decrease in energy
demand if the reactants are produced from biomass. The algal oil consumption to produce 1 GJ of fuel is lower for the scenarios
where triacetin is co-produced (d). The GHG emissions are calculated separately for the equilibrium reaction (e) en the reactant
production and biofuel combustion (f). For both contributions the BD scenario has the lowest emission values of respectively 0.65
and 6.8 kg CO2,eq.GJ-1 biofuel produced

3.3. Influence of reactants derived from biomass

The production process of the reactants is the main contribution to the energy consumption and GHG
emission. Therefore an alternative production process from biomass for each reactant was searched in the
literature. The main energy input for these processes is biomass that is converted to syngas and finally to
the reactant. Since the used biomass is a renewable source, it can be discarded from the total energy
consumption, and only the energy to process the biomass remains. A second advantage is that the reactant
is now also derived from biomass, so no GHG contribution for combustion has to be taken into account.
Data for methanol production out of syngas is included in the EcoInvent database [8]. The alternatives
for methyl acetate and acetic acid are based on literature data. Haro et al. describes a possible route to
produce methyl acetate from syngas with methanol and dimethylether as an intermediate [18] and acetic
acid can be directly produced by anaerobic CO2 fixation using acetogenic bacteria [19]. Both reference
where used to estimate the energy consumption for production.
An improvement is found in the net energy value, especially for the IR scenarios (Fig. 1 a, c, italic
values), if we apply the new values for the energy consumption for reagent production. The BD scenario
and the IR scenarios have now a similar value for the net energy.
 Stefanie Van Damme et al. / Energy Procedia 61 (2014) 1852 – 1859 1857

4. Conclusion

Four scenarios to produce biodiesel were studied to identify the main differences and dis/advantages of
each option. Conventional biodiesel production was compared with three possibilities to produce
biodiesel and triacetin. To keep the comparison simple, only the main differences in energy consumption
and GHG emissions between the scenarios were taken into account. These differences are the equilibrium
reaction, the used reactant and the compound and energy content of the derived biofuel.
For the equilibrium reaction the temperature, reaction time and excess ratio differ, leading to different
energy needs. The conventional biodiesel scenario requires the least energy, mainly due to the lower
excess ratio and the related lower energy need for the recovery of the reactant.
The energy to produce the used reactant (ME, MA and AA) from fossil fuel was deduced from the
EcoInvent database. To produce methanol less fossil fuels is needed and less GHG were emitted.
Therefore the conventional biodiesel production is again the most positive option.
The amount of the produced biofuel increases when triacetin is coproduced, however when the net
energy is calculated it seems that the extra energy produced does not meet the extra energy needs for
reactant production and recovery. Therefore to make the coproduction of triacetin economically feasible
another production process for methyl acetate and acetic acid is needed that is not based on fossil, non-
renewable fuels.
If the alternative reactant production processes from syngas are used, the energy needed to produce the
reactants decreases since the renewable biomass contribution is discarded. The net value of all the
scenarios improves and the BD scenario and IR scenarios have a similar net energy value. The influence
on GHG is that the combustion term can be removed. The GHG of the alternative reactant production
process is not assessed.
In conclusion, the coproduction of triacetin is worse than the conventional biodiesel production when
the reactants are produced from fossil fuels. The main advantage is that instead of the valueless glycerol,
triacetin is formed that can be used as fuel additive. However when the production process is changed to a
biomass based process the scenarios that coproduce triacetin via interesterification and the conventional
biodiesel process have an equivalent net energy value.

5. Further research

Only one reference paper was selected to base the calculations upon for each scenario and alternative
production processes of the reactants. A sensitivity analysis to assess the influence of the selected
parameters on the results is in progress.

References
[1] A. Casas et al. „Effects of triacetin on biodiesel quality,” Energy Fuels, vol. 24, pp. 4481-
4489, 2010.
[2] J. Melero et al., „Oxygenated compounds derived from glycerol for biodiesel formulation:
Influence on EN 14214 quality parameters,” Fuel, vol. 89, pp. 2011-2018, 2010.
[3] G. Santori et al., „A review analyzing the industrial biodiesel production practice starting
from vegetable oil refining,” Applied Energy, vol. 92, pp. 109-132, 2012.
[4] E. Usai et al., „Simultaneous enzymatic synthesis of FAME and triacetyl glycerol from
triglycerides and methyl acetate,” Bioresource Technology, vol. 101, pp. 7707-7712, 2010.
[5] W. Du et al., „Comparative study on lipase-catalyzed transformation of soybean oil for
biodiesel production with different acyl acceptors,” Journal of Molecular Catalysis B:
Enzymatic , vol. 30, pp. 125-129, 2004.
1858 Stefanie Van Damme et al. / Energy Procedia 61 (2014) 1852 – 1859

[6] A. Casas et al., „New trends in biodiesel production: chemical interesterification of

sunflower oil with methyl acetate,” Biomass and bioenergy, vol. 35, pp. 1702-1709, 2011.
[7] L. Zhou et al., „Catalytic characteristics and parameters optimization of the glycerol
acetylation over solid acid catalysts,” Fuel, vol. 103, pp. 617-625, 2013.
[8] N. Jungbluth et al. , „Life Cycle Inventories of Bioenergy,” Final report ecoinvent data
v2.0, vol. 17, 2007.
[9] O. Jorquera et al., „Comparative energy life-cycle analyses of microalgal biomass
production in open ponds and photobioreactors,” Bioresource Technology , vol. 101, pp.
1406-1413, 2010.
[10] R. Slade et al., „Micro-algae cultivation for biofuels: Costs, energy balance, environmental
impacts and future propects”.
[11] H.-J. Althaus et al., „Life Cycle Inventories of Chemicals,” Final report ecoinvent data
v2.0, vol. 8, 2007.
[12] J. Sutter, „Life Cycle Inventories of Petrochemical Solvents,” Final report ecoinvent Data
v2.0, vol. 22, 2007.
[13] R. Ceriani et al., „Prediction of heat capacities and heats of vaporization of organic liquids
by group contribution methods,” Fluid Phase Equilibria, vol. 283, pp. 49-55, 2009.
[14] A. Casas et al. , „Product separation after chemical interesterification of vegetable oils with
methyl acetate. Part I: Vapor-Liquid Equilibrium,” Industrial & Engineering Chemistry
Research, pp. 8087-8094, 2012.
[15] C. Reid, J. Prausnitz en B. Poling, The properties of gases & liquids.
[16] McCabe en Smith, Unit Operations of Chemical Engineering.
[17] National Institute of Standards and Technology, „NIST Chemistry WebBook,” 13 jan 2013.
[Online]. Available: http://webbook.nist.gov/chemistry/.
[18] P. Haro et al., „Thermochemical biorefinery based on dimethyl ether as intermediate:
Technoeconomic assessment,” Applied Energy 102 , vol. 102, pp. 950-961, 2013.
[19] P. Hu et al., „Anaerobic CO_2 fixation by the acetogeni bacterium Moorella
thermoacetica,” American Institute of Chemical Engineers, vol. 59, nr. 9, pp. 3176-3183,
2013.

Biography
Stefanie Van Damme finished her bio-engineering studies at the University of
Ghent in 2012. She works as an assistant and PhD student at the Fluid Mechanics and
Thermodynamics Research group at Vrije Universiteit Brussel. Her research focusses
on the use of algae biomass for biofuel production and more in particular on LCA of
the conversion processes.
 Stefanie Van Damme et al. / Energy Procedia 61 (2014) 1852 – 1859 1859

Appendix A. Chemical equilibrium reactions

a) Transesterification reaction of triglycerides with an alcohol to form biodiesel and glycerol

b) Interesterification reaction of triglycerides with an ester to biodiesel and triacetin
c) Transesterification reaction followed by esterification of glycerol to form biodiesel and triacetin