Transformation of sugarcane bagasse into biocrude via organic

solvents
Célia R. R. Fernandesa, M. Filipa G. Ribeiroa, Marcelo M. Pereira b
aDepartamento de Engenharia Química, Instituto Superior Técnico-Alameda, Lisbon 1049-001, Portugal.
bInstituto de Química-Centro de Tecnologia, Universidade Federal do Rio de Janeiro-Cidade Universitária, Rio de Janeiro 21949-909,
Brazil.

ABSTRACT

Nowadays there is a major focus in new biomass-derived fuels development and research, however, there are several
obstacles that delay the introduction of this bio-oils into oil refineries. In the current work a recent concept of bio-
oil, entitled biocrude, is divulged. This oil is produced by hydrolysis and ketalization of the sugarcane bagasse fiber,
second-generation biofuel. With this method is obtained a bio-oil far more stable than the known pyrolysis bio-oil,
and in addition, it has a greater miscibility in crude oil, therefore, its co-processing in an oil refinery is exceptionally
compatible.
Essentially, this work is divided into two main parts. In the first part, the biocrude is produced in two regimens, in a
batch reactor, and in a semi-continuous reactor. The key goal was to reproduce and optimize the tests carried out
previously but currently executed on a larger scale. Thereafter, the physicochemical characterization of the oils was
accomplished, in order to obtain fundamental information for the second part of the project.
In the second part of the study, a semi-continuous industrial process was designed, based on the semi-continuous
laboratory process, with the aim of ascertaining the economic viability of this process which is still very preliminary.
Having defined the location of the pilot plant in Brazil and the project capacity of 1.3 kton/year, it was possible to
size all the main equipment.
Finally, the cost of producing the barrel of biocrude was obtained at a very competitive cost of US$ 50. Yet, is
important to note, that this amount does not account any benefits that can come from a green project, benefits that
can arise as well as tax benefits from the government, or as, tradable credits from the ”new” carbon market.

Keywords: biocrude, sugarcane bagasse, second-generation biofuel, ketalization, hydrolysis.

In the laboratory LaCES – Laboratório de Catálise

1. Introduction e Energia Sustentável was studied and developed a bio-
oil produced from the hydrolysis and ketalization of the
Currently, it is common knowledge that the increase in sugarcane biomass [2][3][4].
pollution, more specifically air pollution, is due to the In summary, through this process it is possible to
use of energy from fossil sources, mainly from the achieve a bio-oil of similar appearance to the pyrolysis
burning of oil and coal. As a result of a global concern bio-oil, however, this one has several advantages that
over the years, there has been a growing focus on would differentiate from this and resemble the crude
renewable energy sources, such as hydro, wind, solar, oil, being therefore called biocrude.
geothermal, tidal and biomass energy. Although this study is a preliminary study, due to
The bio-oils produced from the different sources of the results of several tests [2] [3], it is thought that, with
biomasses have been studied in order to produce an oil the development of a production process of this oil and
that is the most analogous as possible to crude oil. Bio- the determination of its production cost, it will be
oil produced from pyrolysis of biomass has been the possible to introduce it into oil refineries.
most widely studied and developed, however, it In order to validate the insertion of this bio-oil into
represents some problems. The pyrolysis oil is a highly an oil refinery, catalytic cracking tests via an FCC
oxygenated bio-oil, which is subsequently very (fluidized catalytic cracking) were performed with
corrosive, poorly miscible with crude oil, thermally several mixtures of bio-oil and diesel. These studies
unstable and tends to polymerize when exposed to air. have proven that the bio-oil can be converted to
Its pretreatment is essential for the removal of the xylenes, benzenes, and other aromatic hydrocarbons,
oxygen in excess, thus causing great difficulties and compounds already used in high-octane gasoline. In
increased costs for its co-processing in refineries[1]. addition, the co-processing of diesel and bio-oil
improved the performance and quality of gasoline
compared to pure diesel, and also reduced the number balance of 10 mL/min and purge flow of 60 mL/min.
of oxygen compounds and coke compared to pure bio- The samples used for thermogravimetric were ± 10 mg
oil, which demonstrates a synergistic effect [2]. of macerated bagasse.
The mass losses associated with certain temperature
2. Experimental Method ranges are considered as indicative of the amount of
different chemical components present in the
Biocrude is produced by a process composed of 3 sugarcane bagasse and its degradation temperature.
phases: 2.1.3. Determination of Biocrude Viscosity
• Pre-treatment: drying, grinding and A rheometer of the TA Instruments brand, Discovery
granulometric classification of cellulosic hybrid rheometer DHR-3 model was used to determine
wood biomass; the viscosity of the oil. For the measurements, a Peltier
• Reaction of hydrolysis and ketalization of steel plate with a diameter of 40 mm was used and the
the hydrocarbons of the biomass fiber with a upper cone had an angle of 2.0°.
solvent in acid conditions; Five measures were taken in peak hold from 40 to
• Neutralization of the medium and oil 80°C, with a heating rate of 10°C.
drying.
2.1.4. Calorimetry
2.1. Physicochemical Characterization of With the Pyris Diamond DSC calorimeter from Perkin
Sugarcane Bagasse and Biocrude Elmer, it was possible to obtain the specific heat
2.1.1. Qualitative and Quantitative Analysis of capacity of both batch and semi-continuous biocrude.
Ketals in Biocrude by High-Performance The calorimeter coupled with two equipment, a
Liquid Chromatography - Mass Spectrometry cooling accessory, the Intracooler 2P, and a thermal
(HPLC-MS/MS) analysis controller, the TAC 7 / DX, also from the same
HPLC-MS/MS analysis was performed on a high- brand.
performance chromatograph, model 1260, from After determining the temperature change of a
Agilent Technologies, coupled to a QTRAP 5500 mass substance of known heat capacity, a sapphire was then
spectrometer, produced by AB Sciex, which is a possible to measure the temperature change of the
spectrometer of quadrupole-trap hybrid masses (ion biocrude. Specific heat measurement was performed in
trap). a temperature range of 10 to 140°C with a heating rate
A column (100 mm x 2.1 mm, 3 μm particle size), of 10°C/min. This analysis was performed in a nitrogen
model Hypersil GOLD, made by Thermo-Fischer was atmosphere under a flow of 20 mL/min.
used, which was maintained at 35°C. The mobile phase 2.2. Determination of Density
used consisted of (A) aqueous solution of 5 mM The density was measured at room temperature
ammonium formate and (B) acetonitrile. (25 ± 2°C) and for this purpose, the sample was
Chromatographic analysis was performed in gradient weighed in a graduated vessel, previously calibrated.
elution (0-0.5 min, 15% B; 0.5-10 min, 90% B; 10-11 After weighing, the density was determined using the
min, 90% B; 11.1-15 min, 15% B) at a flow rate of ratio of a measured mass (on an analytical balance with
350μL/min. The injection volume was 8.0 μL. four precision houses) and the filled volume of the
Ionization was done by electrospray and operated in vessel used (2 mL).
positive mode. The following parameters and their The procedure was performed twice, and then an
values are described: curtain gas (CUR) = 15 psi; average density value was obtained.
collision activated dissociation (CAD) = mean; spray
voltage (IS) = 5500 V; temperature of the source 2.3. Sugarcane Bagasse Pre-treatment
(TEM) = 400°C; gas source 1 (GS1) = 40 psi; gas Initially, the sugarcane bagasse that had been used in
source 2 (GS2) = 45 psi; decluster potential the tests needed to be pre-treated. The bagasse used
(DP) = 30 V; Input potential (EP) = 10 V; Collision was obtained at the Bonsucesso free market, a
cell output potential (CXP) = 10 V. neighborhood in the city of Rio de Janeiro, in October
2015.
2.1.2. Thermogravimetric Analysis (TG) of The pre-treatment of sugarcane bagasse consists of
Sugarcane Bagasse and Biocrude 5 steps procedures: 1st washing; drying; milling;
Thermogravimetric analyses were performed on the screening; 2nd washing. Thereafter, each process
Netzsch TG 209 F1 Iris equipment, with a temperature methodology will be explained.
range between 35 and 800°C and at a heating rate of In the present work one of the goals was to verify
10°C/min. The analysis was carried out under a the quality of the oil produced only from the sugarcane
nitrogen atmosphere, under a flow protection of the bagasse fiber, 2nd generation biofuel, thus it was

2|P a g e
necessary to remove all the sugars that did not come VETEC), sulfuric acid (95-99%, VETEC), ethyl
from the fiber. It should be noted that this bagasse acetate (99%, VETEC), sodium bicarbonate (99%,
comes from the sugar industry, where the sugarcane is VETEC), sodium chloride (99%, VETEC) and
crushed with the purpose of removing the sugarcane anhydrous sodium sulfate (99%, VETEC).
juice, therefore this high sugar concentrate juice gets
2.4.2. Neutralization and Drying of Biocrude
impregnated in the bagasse fiber.
It should be noted that, in the initial production tests of
In the first washing stage, the bagasse was washed
the discontinuous reactor oils, the neutralization
twice in distilled water. This washing served only to
method was different from the one of the succeeding
remove solid particles and residues from the cane juice
oils. This was a result of smaller-scale tests that were
present in the fiber so that bacteria or another
being conducted in parallel, it was perceptible that
microorganism would not degrade the bagasse, then it
there was a significant loss of oil with the first method.
could be stored for a longer period.
A 100 mL aqueous solution of 5% w/w sodium
After this washing, the bagasse had to be dried. The
bicarbonate (99%, VETEC) was added to the liquid
material was left outside for 12 hours at ambient
portion and stirred to neutralize the oil till pH 7.
temperature and then was dried at 135°C for 30
After filtering all the oil from the precipitate into a
minutes in an oven.
decanting funnel, the aqueous phase was separated
The dried material was milled in a knife mill (V611,
from the organic phase.
Winner), and classified for granulometry in a vibrating
Sodium chloride (99%, VETEC) was added to
sieve (J-1 B, from Thomas Scientific), where the
saturate the aqueous phase. If more oil was retained at
fraction between the sieves with meshes of 0.125 mm
the aqueous phase, 50 ml of ethyl acetate was also
and 0.850 mm was selected.
added to increase the volume of the organic phase
Finally, the second wash was carried out, the sieved
available to retain more oil. The precipitate fluid was
bagasse was again washed to be able to remove the last
filtered off and the same procedure was repeated three
traces of "free" saccharides from its surface. The
more times. Note that the organic phase to be
determination of sucrose in the distilled water after
withdrawn was added to the separated organic phase
each wash was performed on a liquid chromatography
initially.
equipment (1200 Series, Agilent Technologies)
Anhydrous sodium sulfate (99%, VETEC) was
coupled to a mass spectrometer (API 2000
added to the total organic phase, to remove all the
ESI-MS/MS, Applied Biosystems/MDS Sciex) [3].
water present. Subsequently, the organic phase was
Ultimately the bagasse was dried at atmospheric
filtered into a flask, and ethyl acetate was used for the
temperature and stored.
transfer.
2.4. Production of Biocrude in Laboratory Finally, the organic phase was distilled on a rotary
Scale evaporator, and pure neutral pH biocrude was
In this work the scale of biocrude production was separated from the solvents, acetone (99%, VETEC)
increased, this oil had previously been produced by the and ethyl acetate (99.5%, VETEC).
laboratory, LaCES-UFRJ, but on a smaller scale [3]. When the methodology for the definitive
With these tests, it was possible to analyze whether this neutralization was improved, only the way sodium
process was reproducible on a larger scale and whether bicarbonate was added changed. Instead of adding an
similar or more satisfactory results could be replicated. aqueous alkaline solution the solid anhydrous salt was
Therefore, tests were performed in two regimens, added, in this case, sodium bicarbonate (99%,
first in batch and then in semi-continuous regimen. VETEC). This eliminated the requirement to decant
When repeating this reaction for any of the reactors it the aqueous phase of the organic phase, since the
was known, due to previous studies performed in the amount of water produced was insignificant and easily
laboratory, that certain conditions would already be eliminated with a desiccant compound, in this case,
fixed and only some variables would be optimized. For sodium sulfate (99.5%, VETEC).
example, in the case of discontinuous reactors less
stringent temperature conditions had to be used, when
2.4.3. Discontinuous Reactor
compared to the semi-continuous reactor, to ensure
In the discontinuous reactor, a 600 mL Parr reactor,
that there would be no degradation of the product.
model 4563 of Parr Instrument Co, 10 g of sugarcane
Therefore, the temperature was one of the variables
bagasse was reacted with acetone (99%; VETEC) in
fixed for each system.
the presence of sulfuric acid (95-98%; VETEC).
2.4.1. Reagents and Solvents The conditions set for the Parr reactor were 120 psi
The following reagents and solvents were used in (approximately 8.3 bar), mechanical stirring at
addition to the sugarcane bagasse: acetone (99%; 600 rpm (corresponding to maximum system

3|P a g e
agitation) and temperature set at 90°C, with the reactor in the system.
taking about 30 minutes until it reaches the temperature The bio-petroleum produced was neutralized and
set in the controller. The reactor was pressurized at the dried according to the second procedure presented
start of the reaction by a gaseous mixture of 5% O2/He previously.
on a molar basis. At the end of each reaction, the mass of dry bagasse
Then the produced biocrude was neutralized and that was not consumed and the mass of biocrude
dried according to the first procedure already produced was weighed. Later the biocrude samples
presented. The biocrude samples produced were produced were analyzed on the HPLC-MS/MS
analyzed on the HPLC-MS/MS (1260 Series, from equipment (1260 Series from Agilent Technologies,
Agilent Technologies, coupled to QTRAP 5500, from coupled to AB Sciex QTRAP 5500).
AB Sciex).
2.4.4. Semi-Continuous Reactor 3. Results and Discussion
At this point, it becomes clear the need to evolve into
an increasing mechanization of the system, to reduce There are two main objectives in this study, firstly, to
the time and cost of a future industrialized process. carry out the reproduction of biocrude, on a larger
The semi-continuous reactor used was a tubular scale, and subsequently, verify the conditions already
reactor, at the inlet and outlet of this reactor there were used on a small scale. After that, with the obtained
two grids which retained the bagasse inside the reactor, biocrude, perform the maximum physical-chemical
resembling a fixed bed reactor. It should be noted that tests possible, for a better characterization and
the flow of liquid reaction mixture, acetone (99%; understanding of the same.
VETEC) and sulfuric acid (95-98%; VETEC), which 3.1. Experimental Results and Discussion
runs through the bagasse "bed", enters the base of the First, it is important to explain how the productivity of
reactor and has an upward flow until the exit at the top. each reaction was evaluated. Efficiency was measured
The reactor has a volume of approximately 205 mL. by three parameters: the amount of biocrude produced
The reaction, carried out previously in the batch in relation to the bagasse used, the reaction yield; the
reactor, was reproduced in this semi-continuous amount of bagasse reacted as a function of the initial
reactor. Since this reactor is a larger reactor, 20 g of bagasse, the conversion of the reaction; and the number
sugarcane bagasse were reacted, instead of the 10 g of ketals formed.
used in the Parr reactor. A pump (LC-8A, Shimadzu) The percentage of ketals was calculated by analysis
was used to pressurize the entire system (up to 8 bar), on liquid chromatography equipment (1200 Series,
together with a backpressure valve. Agilent Technologies) coupled to a mass spectrometer
The heating system was based on a heating jacket and (API 2000 ESI-MS/MS, Applied Biosystems/MDS
a heating resistance. Sciex) and allowed to know whether acetone had been
After the clarification of the semi-continuous incorporated in the oil and formed these protecting
equipment operation, the semi-continuous process will groups, ketals, and the type of ketal formed [2], [3].
be described. Ketals detected by HPLC-MS/MS are divided into 4
The pump was switched on and recirculated with different molecules, xylose (XDK) and glucose (GDK)
acetone (99%, VETEC). The reaction temperature was di-ketals, the trade name of
expected to be constant (a deviation between 3 and 4°C 1,2:3,5-di-O-isopropylidene-α-D-xylofuranose and
was allowed). 1,2:5,6-Di-O-isopropylidene-α-D-glucofuranose,
The reactor inlet was then linked to an Erlenmeyer, respectively. And the mono-ketals of xylose (XMK)
in an ice bath, containing the reaction liquid mixture, and the mono-ketals of glucose (GMK), the trade name
acetone (99%; VETEC) and sulfuric acid (95-99%; of 1,2-O-isopropylidene-α-D-xylofuranose and 1,2-O-
VETEC). The outlet was placed in a new Erlenmeyer isopropylidene-α-D-glucofuranose, respectively.
where the formed biocrude was collected and allowed Analyzing all these data, it should be noted that the
to react for one hour at a predefined flow rate. It should severity of the conditions applied was also considered,
be noted that the ratio of sulfuric acid/acetone (v/v) is each condition variation would need to be balanced out
higher in this semi-continuous than in the batch by the gains obtained in each test. When speaking of
process, this alteration had previously been studied [3]. severity, this does not only mean high temperature or
After 1 h the inlet piping was changed to an pressure conditions, or even a strong acid
Erlenmeyer flask containing pure acetone (99%; concentration but for example, it could be an excessive
VETEC) and recycled acetone from the system. The increase in the reaction time that only resulted in a
flow rate was increased to shorten the amount of time slight gain.
that would take to remove the still formed oil trapped

4|P a g e
 3.1.1. Sugarcane Bagasse and Biocrude – hydrocarbons, so it could not be introduced into a
Properties refinery as 100%, as would be expected. Considering
Before exposing the results obtained in each reaction it this scale, and regarding the future valorization of
will be presented, in the two tables below, some biocrude, it could be diluted in a lighter hydrocarbon
properties of the bagasse and the product. to produce an oil with a higher grade API.
Looking at this from a different perspective, it is
Table 1 – Physicochemical properties of bagasse. clear that such comparisons and assumptions are very
useful when it comes to different types of oil, but
Bagasse biocrude is a biofuel and not a crude oil. When referred
Physical state at to a heavier oil, it refers to a very viscous oil (higher
Solid
25°C processing costs), which contains more heavy metals,
Bulk density at sulfur, nitrogen, and high acidity, and these are the
130
25°C (kg/m3) characteristics that lead to a low degree API being
Specific heat at associated with a low market value oil.
0.46 (dry)
20°C (KJ/(kg.K)) However, biocrude does not contain these components
(it may contain but in negligible proportions), as it is
produced from lignocellulosic materials.
Table 2 - Physicochemical properties of biocrude, Therefore, in the case of biocrude, the chemical
determined by the tests described in the experimental functions and the composition are different and this
method. cannot be evaluated by this scale. Another scale would
have to be defined in the case of this biofuel in order to
Biocrude have a reliable assessment of it.
Physical state at 25°C Liquid In addition, the viscosity of the biocrude, despite
Density at 25°C being high at room temperature, with a slight increase
1100-1200
(kg/m3) in processing temperature, is drastically reduced, thus
Semi- causing minimum problems in the pumping and
Discontinuous handling of biocrude. At 40°C a viscosity of 0.97 Pa.s
-continuous
Reactor is obtained, as shown in Table 2, while at 80°C it
Reactor
Specific heat at 25°C already has a viscosity of 0.06 Pa.s.
1.351 1.814 It should be noted that, although the viscosity of the
(J/(g.K))
Viscosity 40°C 0.97 - biocrude is on average higher than the viscosity of the
(Pa.s) pyrolysis bio-oil (0.035 to 1 Pa.s at 40°C, depending
80°C 0.06 -
on the raw material), this is justified by a high water
content present in the pyrolysis oil, from 15 to 30% [5]
Although the density at 60°F (15.55°C) was not (note that on average crude oil contains not more than
determined from the following equation, the API grade 0.5% of water in its composition [6]). Of course, this
of this oil was calculated using the lowest density value water content will provide a lower viscosity than the
(1100 kg/m3) determined at approximately 25°C. biocrude, thus improving its transport properties,
141.5 however, it will greatly decrease its energy value.
°𝐴𝑃𝐼 = − 131.5 (1) Although it has been concluded that the use of the
𝜌𝑠𝑎𝑚𝑝𝑙𝑒
API grade would not be very useful in the classification
It should be noted that the relative density of the of biocrude, it is advantageous to compare the grade
sample, in this case, biocrude, is the density measured API with the specific heat of a heavy oil and biocrude,
comparatively to the density of the water, so the ρsample since is currently the only comparative term available.
is 1100 kg /m3/1000 kg/m3, thus obtaining a negative The API grade is directly linked to the calculation
API degree. estimate of the oil specific heat. The lighter the oil, the
In general, the classification of the API grade greater its specific heat.
contributes to the product price categorization in the For the biocrude, it was used the determination of
market. The higher the API grade, the higher the value this value from a calorimeter, obtaining the values of
of the product on the market, however oil with a °API specific heat at 15.56°C of 1.469 and 1.712 J/(g.K) in
between 40 and 45° leads the higher prices. Above 45°, a discontinuous regime and semi-continuous regime,
molecular chains become shorter and less valuable for respectively, values not distinctive from those of light
refineries. and heavy oils.
Oils that have a °API of less than 10° are very heavy Now proceeding to the analysis of the calorific
oils, they consist almost entirely of aromatic power of the pyrolysis bio-oil, resorting to tests

5|P a g e
referenced in the literature [6], it was observed that, on Table 3 - Experimental results of the first four reactions in a
average, this value is less than half of the calorific discontinuous reactor.
value of crude oil, and as already was mentioned, this
is a consequence of the high water content in the bio- 1 2 3 4
oil composition. moil (g) 3.99 4.12 3.08 4.11
Also, remember the importance of the relationship mresidual bagasse (g) 6.85 6.30 6.46 6.20
between specific heat (or calorific value) on a ɳ (%) 39.89 41.23 30.83 41.13
volumetric basis and the density of a material. The fact mconsumed bagasse (g) 3.15 3.70 3.54 3.80
that biocrude (and bio-oil pyrolysis) is a very dense Conversion (%) 31.50 37.03 35.43 37.95
fuel, promotes the reduction of transport costs, saving
the volume needed to transport an oil with a high Table 4 - Results of the quantification of xylose (XDK) and
volumetric energetic potential. glucose (GDK) di-ketals, and xylose (XMK) and
ycosecoseglucose (GMK) mono-ketals by HPLC-MS/MS
3.1.2. Discontinuous Reactor from the first four reactions in a batch reactor.
The initial tests were performed in a discontinuous
reactor under the most favorable conditions that had Ketals (%) 1 2 3 4
already been studied[2]. In addition to the described XMK 6.0 6.8 5.3 5.5
conditions, 8.3 bar, 90°C, and mechanical stirring at GMK 0.2 0.2 0.1 0.1
600 rpm, 200 mL of acetone (99%; VETEC); and 0.8 XDK 20.7 14.7 14.4 11.7
ml of sulfuric acid (95-99%; VETEC) which were GDK 14.5 10.4 9.7 7.6
reacted for 2 hours and 30 minutes with the 10 g of Total 41.4 32.1 29.4 25.0
bagasse. Subsequent to this test, there were made 3
more tests in the batch reactor all based on this standard 3.1.3. Semi-continuous Reactor
test. In the semi-continuous reactor, the tests were started
In the second test, the reaction time was increased by with the most favorable conditions, that had already
90 minutes, the reaction was given a duration of 4 been studied [2] and with direct adaptations to the size
hours. of the reactor.
In the third test the ratio of biomass/acetone (w/v) In addition to the conditions already described, 8 bar
was increased, a volume of acetone (99%; VETEC) of and 20 g of sugarcane bagasse, several conditions were
150 ml was used and consequently a lower volume of used. The conditions were being defined as the results
sulfuric acid (95-99%; VETEC), 0.6 mL, to maintain were obtained on a smaller scale process. The smaller
the sulfuric acid/acetone ratio of 0.4% (v/v). quantity tests were being carried out in parallel with
In the second test the yield was increased, see these tests of a higher scale.
Four tests were carried out in the reactor, but only
Table 3. It was decided to further increase the reaction the most reliable results are presented. In this test, the
time, in this last test the reaction duration was conditions used were 120°C, the ratio of sulfuric acid
increased to 5 hours. and acetone was 1/500 %v/v. A flow rate of 7 ml/min
In these 4 tests, after obtaining the oil, the first per 1 hour was chosen due to test results obtained from
procedure of neutralization and drying of the bio- other work not yet publicly disclosed. Table 5 below
petroleum was used. shows the reaction results:
The results obtained in these first 4 tests, the mass of
oil formed, bagasse mass after the reaction, yield (ɳ), Table 5 - Experimental results of the reaction in the semi-
continuous regime.
bagasse mass consumed in the reaction and conversion
of the reaction are shown in
Semi-continuous Test
Table 3. moil (g) 9.2
Analyzing the results, it is observed that biocrude 2 mresidual bagasse (g) 11.0
is the one with the highest yield, very similar to 4 and ɳ (%) 45.9
1. In terms of conversion, biocrude 4 is the leading, and mconsumed bagasse (g) 9.0
2 and 3 also have similar values. Conversion (%) 45.2
After completion of these 4 tests, the percentage of
ketals formed in each reaction was quantified, these Given that the biocrude was neutralized and dried
values are found in Table 4. according to the second methodology. As was said, the
amount of sodium bicarbonate used, 100.1 g, was
much higher than the value calculated by

6|P a g e
stoichiometric calculations, according to calculations it some conditions from the laboratory scale semi-
should have been approximately 3.2 g. Therefore, it is continuous reactor, in order to determine the cost of
possible to conclude that the agitation was very weak, production of this biocrude. However, since this is an
being necessary to use an excess of 317%, a very industrial scale process, some topics were readjusted.
maladjusted value of what would be necessary for this
4.1. Process of Production of Biocrude in
procedure.
Pilot Scale
The HPLC-MS/MS analyses of the semi-continuous
A flowsheet of the biocrude production process is
biocrude were then executed, see Table 6.
presented in Appendix A and then the whole process is
Table 6 - Results of the quantification of xylose (XDK) and described and substantiated. The biocrude production
glucose (GDK) di-ketals xylose (XMK) and glucose (GMK) process is divided into 5 zones: reagent storage zone
mono-ketals by HPLC-MS/MS of the reaction in the semi- (zone 100), pretreatment zone (zone 200), reaction
continuous regime. zone (zone 300), separation zone (zone 400), and
product storage zone (Zone 500).
Ketals (%) Semi-continuous Test
4.1.1. Capacity and Assumptions of the Mass
XMK 8.7 Balance of the Biocrude Production Process
GMK 0.7 Usually, when determining various data for the
XDK 27.7 evaluation of the production process, such as the
GDK 14.3 market price of the product and manufacturing
Total 51.4 capacity, a market analysis of the price is made. This
analysis will ascertain at which price the product is
As expected, a higher percentage of ketals can be currently sold, and will reveal the existing
obtained resulting in an improved quality of the oil. manufacturing capacities, but since this is a new
The equilibrium of the ketalization reaction is favored product such data is not available. For this reason, it
in basic conditions, probably the shortest time in was decided to take as a starting point for the economic
contact with acid, in the semi-continuous reactor, evaluation, of this plant, a daily production capacity of
increased the yield in ketals and the conversion of the 3 tons of biocrude. A reasonable size for a pilot plant
biomass. and a quantity that is plausible from the point of view
When calculating the incorporated acetone in the of profitability.
reaction (this percentage was equal to the efficiency of From this information, it is possible to reiterate the
the oil mass formed versus bagasse mass consumed), it capacity of the process, for which two data are needed:
was concluded that this value is very different from the the annual production and the occupation rate. If in one
ketals quantification result. This result, clearly, was a day is produced 3 tons/day, in one year of production
consequence of operator errors, probably due to oil 1 kilotons would be produced. This value was obtained
losses resulting from a poorly efficient post-reaction considering that the factory is producing 24 hours a
recovery. As it happened to have a much higher day, 335 days a year since it is necessary to subtract the
amount of bagasse, it became more difficult to clean all days in which the shutdown and startup are made (1
the oil from the bagasse fiber. month). As the average occupancy rate, for the initial
One can also make a very simplistic estimation of years, is 80%, the factory project capacity will be
this percentage, taking into account the quantification around 1.3 kton/year.
of ketals. Knowing that one mole of acetone is required Now it is possible to summarize all the assumptions
to form one mole of mono-ketal. It can be deduced that of this process, some of them already mentioned in the
x% acetone formed 9.4% mono ketals and 2x% previous subchapter:
acetone formed 42% di-ketals, this provision resulting
in a 30.4% incorporation. Although simplistic, this Reagent Storage (zone 100):
calculation resulted in a value equivalent to the already • The reagents were assumed to be 100% pure.
proven value by previous tests [2], this value is also Pretreatment (zone 200):
around 30% of acetone incorporated. • The initial bagasse contained 40% moisture;
• Dry bagasse had 7.4% moisture (determined
4. Process and Cost of Production of by the thermogravimetry, the mass loss till
Biocrude 105°C);
• In the dryer, the air temperature at the
Finally, it became important to outline a process of entrance is 145°C and at the exit is 120°C;
biocrude production in a semi-continuous regimen on • The atmospheric heated air contained 3%
an industrial scale. This was done by implementing water at the entrance to the dryer;

7|P a g e
 • In the burner, the oxidizer used was Vacuum distillation column (zone 400):
atmospheric air and the fuel was methane. • The liquid ring pump system produces a
Main Reactors (zone 300): vacuum of 0.6 bar;
• The reactors operate at 120°C and 8 bar; • In the distillation, an acetone with 100%
• The mass/volume ratio of bagasse/acetone is purity is obtained at the top of the column,
1:25; which is then recycled to the main reactors.
• The volumetric ratio of acetone/sulfuric acid
4.1.2. Location and Climate
feed is 500:1;
This project was designed for the implementation of a
• Bagasse conversion is 45%;
pilot unit in Brazil. It is in Brazil where the cheapest
• The yield of biocrude produced is 46%;
raw material is found, being the largest producer of
• Acetone incorporated into biocrude is
sugar in the world.
equivalent to 20% (w/w) of the biocrude
produced (this percentage was equal to the 4.1.3. Construction Materials
efficiency of the oil formed versus bagasse Since very high operating temperatures or pressures
consumed) [2]; are not used, the equipment building materials choice
• The water consumed in the hydrolysis is is not necessarily dependent on these values.
considered equivalent to that formed in the Sulfuric acid and calcium hydroxide are present, so,
ketalization, so it is assumed that the amount as a conservative measure, the 316 stainless steel was
of water contained in the bagasse is the same chosen as the equipment material in which these were
amount of free water that leaves the reactor present. In the others, carbon steel was chosen.
effluent;
4.1.4. Economic analysis
• The water solubility in acetone was
In order to determine the investment required for the
considered negligible;
process and the production cost of this biocrude, the
• The bagasse after reaction draws 0.1%
data obtained in the semi-continuous test, on a
acetone and 0.05% water.
Neutralization Reactor (zone 300): laboratory scale, were used. However, some
• The reactor operates at atmospheric pressure adaptations to this industrial scale were made.
and at room temperature; 4.1.5. Investment Estimate
• The amount of water formed in the The total investment is composed of the fixed capital,
neutralization is stoichiometric; the working capital, plus the interim interest.
• Calcium oxide is added with 10% excess.
Non-Consumed Bagasse Filtration (zone 400): 4.1.6. Fixed Capital
• The rotating filter operates at 0.8 bar; The fixed capital is the sum of tangible and intangible
• It is possible to separate 100% of the bagasse capital, there are various ways to estimate this cost, but,
from the water and acetone; in general, any method is based on the estimated base
• There is no loss of water or acetone by equipment cost.
evaporation.
Separation of Solids from Neutralization (zone i. Estimated Base Equipment Cost.
400):
• In the thickener it was assumed that 72% of
the suspension is clarified and that this phase
does not contain any solids;
• The rotating filter operates at 0.7 bar;
• In the filter, 100% of the solids (calcium
oxide and calcium sulfate) can be separated
from water, acetone, and biocrude;
• There is no loss of water or acetone by
evaporation.
Separator Gas-liquid (zone 400): Figure 1 – Equipment cost by zone (2.1 M€).
• The separator has a heat exchanger that
precedes it, and which heats the entire mixture From Figure 1, it can be seen, that zones 300 and 400
to 75°C; correspond to the most expensive areas of the process.
• 100% of the biocrude is separated in the base This fact is expected since the reaction zone starts from
of the column and 100% of the acetone and a semi-continuous and not from a continuous process,
water leaves at the top. forcing the existence of two identical reactors. In the

8|P a g e
zone 400, it is necessary numerous equipment to make Miscellaneous
the separation of the oil of the remaining equipment, Solvents Supplies Raw
1% 4%
giving rise to this higher cost of the process. These are Materials
Maintenance 29%
the two areas where the future optimization of the
28%
equipment will be more important.
Manufacturing
Workforce
ii. Detailed Estimate of the Fixed Capital. In 10%
this method for the determination of the
various installments that make up fixed
Control
capital, percentage ranges are allowed on the Utilities and Workforce
cost of the basic equipment (Cb) or on total Services 1%
26%
fixed capital (If).
After estimating these fractions, the Total
Fixed Investment was obtained, resulting in Figure 2- Percentage distribution of the various components
8.6 M€. Where 6.5 M€ are Tangible Fixed of direct costs.
Investment and 2.1 M€ are Intangible Fixed
Investment. It was concluded that the raw material share is the
one that has the greatest weight in the direct costs. The
4.1.7. Working Capital total of direct cost was 2.2 M€ in 2018.
Was determined taking in account the process raw
material consumption, stock of products manufactured, • Indirect Cost: The indirect costs represent
quantitative of products in labor, credit offered to the costs of labor sustenance of the production
costumers and obtained from suppliers, and the activity. It corresponds to about 50 to 70% of
working capital. The total working capital was 0.1 M€. the sum of total labor and maintenance costs.
4.1.8. Interim Interest In this present study, an average percentage of
This portion of the total investment was calculated 60% was admitted. This resulted in an indirect
based on the investment plan, considering that equity cost of 0.4 M€ in 2018.
represents 50% of the total investment. The remaining • Fixed Costs: Comprises the expenditure
capital is financed by external capital with an interest which is independent of productive activities.
rate applied for the financing of 1,05%, corresponding The following are the elements that constitute
to the basic interest rate in Brazil, SELIC (Special this topic: depreciation; insurance and taxes,
Settlement and Custody System), plus the average and rents. This resulted in a fixed cost of
spread, 31.9%. 0.8 M€ in 2018.
4.2.2. General Costs
This resulted in a total interim interest of 1.5 M€, and Overheads include the following installments:
a total investment of 10.2 M€. administrative expenses (40% of the manufacturing
4.2. Estimated Cost of Production workforce); commercial costs (will start at 15% till 7%
It consists of the parts of manufacturing cost, which of the total cost of production, decreases with the time);
relate to all the charges directly required by the expenditure on R&D (it was considered a profit margin
production activity, and the portion of the overhead between 30 to 40%, since it’s a small factory, and since
costs which represents all the costs required to support this is a new product the highest percentage of R&D
the production activity. was chosen, 5% of the total cost of production) and
financial charges (calculated assuming annual
4.2.1. Manufacturing Cost repayments to the bank over the course of the project
Manufacturing costs include the direct costs, indirect and an interest rate of 1.05%, and it was admitted a
costs, and fixed costs. year as grace period).
The direct costs are composed of raw materials, The items included in overheads were calculated
manufacturing workforce, control workforce, utilities based on percentages of other costs, this resulted in a
and services, maintenance, solvents, and general cost of 1.8 M€ in 2018.
miscellaneous supplies.
The following graph shows the weight that each For the interim year under study, direct costs
portion of direct costs represents: represent the largest portion of production costs, this is
mainly due to the cost of raw materials. The next table
presents the total of the manufacturing cost, and the

9|P a g e
latter with the general cost, i.e., the production cost for increase in the incorporation of acetone in the oil. This
10 years: increase can be justified by the higher yield in the semi-
continuous, and consequently resulted in a greater
Table 7 – Estimative of Production Cost[7].

M€/year 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027
Manufacturing Cost 5,3 5,4 5,1 4,8 4,9 4,9 5,1 5,3 5,4 5,6
Production Cost 3,5 3,7 3,1 3,4 3,7 3,9 4,1 4,3 4,5 4,7

Finally, the results of the cost of the biocrude barrel number of xylose and glucose molecules available to
where obtained, in the year of 2018 the barrel of be hydrolyzed.
biocrude would cost US$ 50 and in 2027 would be As this semi-continuous process is analyzed, it is
US$ 53. Being that on July 9, 2018, the price of a barrel concluded that at the present it is only an initial concept
of oil was US$ 73.72[8]. and not the definitive process.
It is hoped that, in the near future the installation of
By analyzing the graph (Figure 3), it can be a pilot plant, of this process, would be feasible and thus
concluded that the price of the solvent greatly the desired improvements would be achieved, both
influences the cost of production of a barrel of technological and energetic. Optimizing the process
biocrude. Therefore, in addition to improvements in with more functional equipment and an integration of
procedural terms, it will also be very important to use utilities where there is maximum reuse of all streams
an alternative solvent to acetone. and effluents. It is also worth noticing that, both bio-oil
and bagasse that are not consumed can be used as
51 energy sources for a more sustainable plant.
50 After obtaining the process flow diagram
Biocrude Production Cost

49 (Appendix A), it was possible to carry out the

(US$/Biocrude bbl)

48
economic evaluation, where the investment (and base
47
equipment cost) and the respective investment plan
46
45 were calculated, it was determined that the investment
44 needed for the proposed installation is about
43 € 10.2 million, of which 50% comes from outside
42 capital.
1328 1210 1093 975 858 740 623 505 Finally, the cost of production was obtained, and
even without an optimization of the process, and even
Solvent Price (US$/ton)
with the oil crisis that we are still experiencing, the cost
of producing a barrel of biocrude is US$ 50.
Figure 3- Solvent price versus Biocrude production cost.
This is a cost that can rival the current cost of oil,
US$ 73.72 (July 9, 2018)[8], although it is thought that
5. Conclusion this drop in oil prices will be temporary, resembling all
crises throughout the oil exploration, this is still a big
As can be seen from the results of the discontinuous price gain. Remembering that, these price fluctuations
and semi-continuous reactor, the conversion and yield are mostly derived from exploratory country policies
(from about 32 and 40% in the batch to 45 and 46% in and global government strategies.
the semi continuous, respectively) are higher in the Finally, it should be noted that no in-depth
semi-continuous. This is probably due to the increase investigation of the incentives inherent to a plant using
of temperature in the reaction of the semi-continuous, green technology had been executed. Even so, it has
this increase is allowed owing to the reduction of the been possible to see that there are some governmental,
residence time in the reactor. This reduction of time fiscal and other funding to boost investments in
also reduced the degradation of the oil produced, and projects that are "environmentally friendly"; and that,
increased the opening of the sugarcane fiber, thus in addition to this added value, there is also the carbon
obtaining a greater breakage of the crystalline market, where green companies negotiate their green
polymers of the cellulose, more difficult to reach. credits for significant monetary amounts.
Concerning the formation of hydrocarbon ketals, we
can also conclude that there was an increase in their
production in the semi-continuous reactor (from 41.4%
in the discontinuous to 51.4%), with a quantitative

10 | P a g e
Acknowledgements
The authors thank the Brazilian science foundation,
COPPETEC, and Petrobras, for their financial support.
To Dr. Carla Barbato (COPPE/PEQ - UFRJ) for
determining the viscosity, the Master Rosana
Mauritius (LABTeR/EQ - UFRJ) for the calorimetry
test and Prof.º Dr. Márcio Nele (COPPE/PEQ - UFRJ)
which enabled the realization of these tests.

MAIN REFERENCES

[1] M. I. Jahirul, M. G. Rasul, A. A. Chowdhury, and

N. Ashwath, ‘Biofuels production through biomass
pyrolysis- A technological review’, Energies, vol.
5, no. 12, pp. 4952–5001, 2012.
[2] N. Batalha et al., ‘Gasoline from biomass through
refinery-friendly carbohydrate-based bio-oil
produced by ketalization’, ChemSusChem[1] N.
Batalha al., «Gasoline from biomass through
refinery-friendly carbohydrate-based bio-oil Prod.
by ketalization», ChemSusChem, vol. 7, n. 6, pp.
1627–1636, 2014., vol. 7, no. 6, pp. 1627–1636,
2014.
[3] R. Garrett, T. G. Barros, M. O. De Souza, B. M. C.
Costa, M. M. Pereira, and L. S. M. Miranda,
‘Unveiling the Chemical Composition of Sugar
Cane Biocrudes by Liquid Chromatography −
Tandem Mass Spectrometry’, pp. 5–10, 2015.
[4] M. M. Pereira, L. Mariz e Miranda, M. Oliveira de
Souza, and M. B. Bezerra de Almeida, ‘Patent -
Processo de obtenção de bio-óleo solúvel em
hidrocarbonetos’, WO2015135046A1, 2015.
[5] S. Czernik and A. V. Bridgwater, ‘Overview of
applications of biomass fast pyrolysis oil’, Energy
and Fuels, vol. 18, no. 2, pp. 590–598, 2004.
[6] J. Lehto, A. Oasmaa, Y. Solantausta, M. Kytö, and
D. Chiaramonti, ‘Fuel oil quality and combustion
of fast pyrolysis bio-oils’, VTT Publ., no. 87, p. 79,
2013.
[7] R. Peters, M., Klaus, T., & West, Plant Design and
Economic for Chemical Engineers, 4th ed.
McGraw-Hill, New York:, 1991.
[8] ‘Crude Oil Prices’. [Online]. Available:
http://www.investing.com/commodities/crude-oil-
historical-data. [Accessed: 09-Jul-2018].

11 | P a g e
1. Appendix A - Process Flow Diagram – Biocrude Production

12 | P a g e