process conditions (Srokol et al. 2004; Brunner 2009, 2009;
Kabyemela et al. 1997; Chornet and Overend 1985; Russell
et al. 1983).
Biller and Ross used compounds additive approach to
obtain a linear prediction model for the bio-crude yield (Biller
and Ross 2011). This proposed model was only limited to
microalgae species like Chlorella and Nannochloropsis to
accurately predict their bio-crude yield. However, it did not
work well for other biomass species. Teri et al. obtained a
quantitative approach by incorporating interaction terms and
studied liquefaction of binary components (Teri et al. 2014).
The model constituents with interaction terms showed less
accuracy for the investigation of bio-crude yield than linear
model. Ky Vo et al. examined HTL of high-lipid microalgal
at varied temperatures ranging from 250 to 400 °C and
retention times (10–60 min) (Vo et al. 2016). In this study,
they used HTL mechanism for kinetic modeling developed
(Valdez et al. 2014). In this model, lipid, protein and carbohydrate fractions reacted independently to yield aqueous
phase and bio-crude products. Consequently, reversible
interconversion between these products resulted in further
transformation to gaseous products (Fig. 5).
They suggested ten reaction rate constants and assumed
all reactions follow first-order kinetics. By following rate
law equation, they proposed rate of each reaction according
to the following pathways:
Proteins:
dx 1;p
dt
¼ À k 1;p þ k 2;p
À
Á
x 1;p
Lipids:
dx 1;l
dt
¼ À k 1;l þ k 2;l
À
Á
x 1;l
Carbohydrates:
dx 1;c
dt
¼ À k 1;c þ k 2;c
À
Á
x 1;c
Aqueous À phase product:
dx 2
dt
¼ À k 4 þ k 5
ð
Þx 2 þ k 1;p x 1;p þ k 1;l x 1;l þ k 1;c x 1;c þ k 3 x 3
Biocrude product:
dx 3
dt
¼ À k 3 þ k 6
ð
Þ x 3 þ k 2;p x 2;p þ k 2;l x 2;l þ k 2;c x 2;c þ k 4 x 2
Gaseous product:
dx 4
dt
¼ k 5 x 2 þ k 6 x 3
Gai et al. and Chen et al. examined the results produced
from HTL of microalgae with low-lipid contents under
subcritical condition (Gai et al. 2015; Chen et al. 2014). The
general reaction pathways for complete HTL process were
inferred from experimental outcomes of previous studies
reported in the literature (Fig. 6).
Chen et al. proposed a potential HTL reaction scheme
based on possible reaction pathways and GC-MS data
(Fig. 7) (Chen et al. 2014). The thickness of the arrow shows
relative amount of product distribution into different phases.
Currently linear additive models are available by assuming
that each component behaved independently during the HTL
process. Thus, it is crucial to design prediction models by
using more illustrative model compounds and considering
the influence of interaction between them. Such model
derivatives can be more advantageous to predict product
yields accurately and assess the viability of co-liquefaction
for use of various biomass sources to enhance energy production. Additionally, the chemical reactions among the
biomass components can also be explored to mimic their
conversion pathways. It would provide the basic information
for significant alteration of product supply and enhanced
knowledge of synergistic phenomenon, when different biomass feedstocks are mixed.
Fig. 5 HTL reaction network
(adapted from Vo et al. 2016)
32
K. Sharma et al.
Kabyemela et al. 1997; Chornet and Overend 1985; Russell
et al. 1983).
Biller and Ross used compounds additive approach to
obtain a linear prediction model for the bio-crude yield (Biller
and Ross 2011). This proposed model was only limited to
microalgae species like Chlorella and Nannochloropsis to
accurately predict their bio-crude yield. However, it did not
work well for other biomass species. Teri et al. obtained a
quantitative approach by incorporating interaction terms and
studied liquefaction of binary components (Teri et al. 2014).
The model constituents with interaction terms showed less
accuracy for the investigation of bio-crude yield than linear
model. Ky Vo et al. examined HTL of high-lipid microalgal
at varied temperatures ranging from 250 to 400 °C and
retention times (10–60 min) (Vo et al. 2016). In this study,
they used HTL mechanism for kinetic modeling developed
(Valdez et al. 2014). In this model, lipid, protein and carbohydrate fractions reacted independently to yield aqueous
phase and bio-crude products. Consequently, reversible
interconversion between these products resulted in further
transformation to gaseous products (Fig. 5).
They suggested ten reaction rate constants and assumed
all reactions follow first-order kinetics. By following rate
law equation, they proposed rate of each reaction according
to the following pathways:
Proteins:
dx 1;p
dt
¼ À k 1;p þ k 2;p
À
Á
x 1;p
Lipids:
dx 1;l
dt
¼ À k 1;l þ k 2;l
À
Á
x 1;l
Carbohydrates:
dx 1;c
dt
¼ À k 1;c þ k 2;c
À
Á
x 1;c
Aqueous À phase product:
dx 2
dt
¼ À k 4 þ k 5
ð
Þx 2 þ k 1;p x 1;p þ k 1;l x 1;l þ k 1;c x 1;c þ k 3 x 3
Biocrude product:
dx 3
dt
¼ À k 3 þ k 6
ð
Þ x 3 þ k 2;p x 2;p þ k 2;l x 2;l þ k 2;c x 2;c þ k 4 x 2
Gaseous product:
dx 4
dt
¼ k 5 x 2 þ k 6 x 3
Gai et al. and Chen et al. examined the results produced
from HTL of microalgae with low-lipid contents under
subcritical condition (Gai et al. 2015; Chen et al. 2014). The
general reaction pathways for complete HTL process were
inferred from experimental outcomes of previous studies
reported in the literature (Fig. 6).
Chen et al. proposed a potential HTL reaction scheme
based on possible reaction pathways and GC-MS data
(Fig. 7) (Chen et al. 2014). The thickness of the arrow shows
relative amount of product distribution into different phases.
Currently linear additive models are available by assuming
that each component behaved independently during the HTL
process. Thus, it is crucial to design prediction models by
using more illustrative model compounds and considering
the influence of interaction between them. Such model
derivatives can be more advantageous to predict product
yields accurately and assess the viability of co-liquefaction
for use of various biomass sources to enhance energy production. Additionally, the chemical reactions among the
biomass components can also be explored to mimic their
conversion pathways. It would provide the basic information
for significant alteration of product supply and enhanced
knowledge of synergistic phenomenon, when different biomass feedstocks are mixed.
Fig. 5 HTL reaction network
(adapted from Vo et al. 2016)
32
K. Sharma et al.