Microsoft Word - 66iervolino.docx
CHEMICAL ENGINEERING TRANSACTIONS
VOL. 65, 2018
A publication of
The Italian Association
of Chemical Engineering
Online at www.aidic.it/cet
Guest Editors: Eliseo Ranzi, Mario Costa
Copyright © 2018, AIDIC Servizi S.r.l.
ISBN 978-88-95608-62-4; ISSN 2283-9216
Elutriation and Sedimentation Process for the
Characterization of Sugarcane Bagasse and Straw
Deyber A. Ramirez-Quintero*, Waldir A. Bizzo
University of Campinas, School of Mechanical Engineering, R. Mendeleyev, 200, 13083-960, Campinas, SP, Brazil
deyber@fem.unicamp.br
In gas-solid interaction, particle characteristics such as shape, size and density, interacting simultaneously,
have a strong influence on the terminal velocity. The behaviour of fluid-dynamics systems acquires great
uncertainty when the bed is composed by irregular particles, such as biomasses. Biomass, usually with a wide
particle size distribution, is characterized by analyses that do not consider the irregular shape of the particles.
Thus, these parameters are introduced in the different theoretical correlations, empirical or combinations of
both to represent the biomass behaviour. However, the results obtained in the modelling are generally not
satisfactory, because they provide results with considerable deviations. This work aims to compare the drag
velocity measured experimentally in biomass (bagasse and straw) with the estimates found through
correlations for the determination of the terminal velocity of spherical and non-spherical particles and sieving
opening. It was developed appropriated and flexible system to this separation, the superficial air velocity can
be regulated to the expected conditions. The material was separated at different air velocities and analysed by
known methods for material characterization. The use of elutriation and sedimentation could contribute to the
understanding the granulometric and morphological separation of biomass, obtaining separate fractions with
homogeneity of size and shape in each fraction obtained, which are not satisfactorily predicted by the
correlations used for the design of the fluidized systems for energy generation from biomasses.
1. Introduction
The aim to optimize the process of combustion, pyrolysis and gasification of biomass for energy generation,
reduce its impact on the environment and make these more attractive economically, requires a better
understanding of the influence of shape, size and density on fluid dynamics processes (Kunni, Levenspiel,
1991; Yang, 2003; Nikky et al. 2014). Given the diversity of biomass, it is difficult to characterize them in a
general way through correlations or mathematical models. However, different configurations of fluidized bed
reactors are being used in several processes, the biomass particles are mixtures and used in different
proportions to obtain their greatest potential.
In the 1970s, the world faced an economic and geopolitical crisis that gave a first hit to the growing, and even
unconscionable, energy consumption, therefore one of its main resources, fossil fuels. In the faced with the
possibility of the oil depletion, increase in the price of a barrel of oil and the need to use clean and renewable
energy, some dependent countries on the importation of this fuel sought energy alternatives. In Brazil, a boost
was given to the path of energy generation using sugarcane with the creation of the National Alcohol Program
(PROALCOOL), 1975, which received a certain reception by the increase of distilleries and the
commercialization of cars fuelled by ethanol (Coelho-Carvalho et al, 2013). Thus, bagasse and straw left over
from the sugarcane mill were considered as waste and, like any residue, it was necessary to discard them.
The burning of these "residues" for the generation of steam to produce electricity was an alternative that
appeared to give a solution to the problem of waste treatment and the creation of new energy resources.
Currently, only bagasse is used as fuel in boilers to produce steam for process and electricity generation.
Some mills add sugarcane straw to bagasse, no more than 15 % in weight, but reliable data about this is not
available. Until a few years ago, the leaves of sugarcane were burned directly on the plant, before the harvest,
in order to facilitate manual harvest. Due to the environment impact that the pre-burning caused, this practice
was banned and is being replacing by mechanized harvesting. Sugarcane straw (green and dry leaves and
607
DOI: 10.3303/CET1865102
Please cite this article as: Ramirez Quintero D., Bizzo W., 2018, Elutriation and sedimentation process for the characterization of sugarcane
bagasse and straw, Chemical Engineering Transactions, 65, 607-612 DOI: 10.3303/CET1865102
tops) is left on the field after mechanized harvesting. The collecting methods for sugarcane straw still are
being studied and developed, but sugarcane straw is already available through mechanized harvesting and
the energy contained in this by-product is roughly of some order of magnitude as the bagasse and the ethanol
produced (Carvalho, Veiga and Bizzo, 2017)
Bagasse and straw by-product of sugarcane also have great potential as a raw material for the production of
fuels derived from processes such as gasification, pyrolysis or hydrolysis followed by fermentation. For the
development of these technologies, detailed knowledge of the physical and chemical characteristics of
sugarcane bagasse and straw is necessary (Barbosa-Cortez, Silva-Lora, Olivares-Gomez, 2008; Rendeiro et
al, 2008).
The use of software in measuring the particles physical characteristics usually requires advanced language
programing such as Visual C,Visual Basicand Matlab with specialized image processing toolboxes
(Igathinathane et al, 2008). The plugins developed for the ImageJ program become a striking option for
particle size measurement. ImageJ is a public domain Java processing program, freely available, open
source, platform independent and analysis program developed at the National Institutes of Health (NIH)
(Bailer, 2006).
The particles entrainment in fluidized bed can be considered as a predictable consequence in fluidized beds
(Colakyan, Levenspiel, 1984; Kunni, Levenspiel, 1991), this 'lost material’ can be utilized not only to create
techniques, also to optimize combustion, gasification and pyrolysis processes (Yang, 2003).
Haider and Levenspiel (1989), from a revision of the existing correlations until that moment to calculate the
terminal velocity of the falling particles, , Eq(1), proposed a dimensionless particle diameter, *, (Eq(2)) and
a dimensionless terminal velocity, *, to spherical particles, Eq(3) and non-spherical, Eq(4)
= * ⁄ (1)
* = ⁄ (2)
* = * + . * , (3)
* = * + . . ∅* , 0,5 ≤ ∅ ≤ 1 (4)
Where denote the equivalent spherical diameter, is the acceleration due to gravity, denote the
viscosity of fluid, and are density of particle and density of fluid, respectively; and ∅ is particle sphericity.
According to Anderson (1988), Grobart (1973) presented an extensive study on the air entrainment velocity of
sugarcane bagasse in various sizes and moisture content (0-48% wb) and developed an equation to calculate
the pneumatic entrainment velocity of bagasse, Eq(5). = 0.115 + 0.819 − 0.0517 + 0.00293 + 0.00116 (5)
Where, denote the entrainment velocity (m/s), is the maximum sieve aperture (mm), denote the
moisture content.
Aiming to present an introduction of bagasse dryers connected to the energy recovery system of the boilers
operating with exhaust gases, Sosa-Arnao and Nebra (2009) proposed an optimized low cost design for
energy recovery configurations to be applied in boilers of water pipes, having as heating fluid the boiler
exhaust gas itself. They obtained Eq(6) to determine the terminal velocity of the particle, . = 31.699 . (6)
Where is the mean particle diameter.
Nikky et al. (2014) describes a characterization method of an average drag force between gas and different
regular and irregular particles, they performed an experimental investigation with a fluidization characterization
test device and presented that the shape and size of the particles have an important effect on the fluidization
behavior of the particles by the drag force.
The aim of this work is to aerodynamically characterize the sugarcane bagasse and straw obtained from a
sugar mill and compare it with the traditional method of characterization, sieving, to determine the drag
velocity.
608
2. Experimental study
The collection of bagasse and straw was carried out from sugarcane mill, located in São Paulo State. The
material was dried in the sun for one week to equilibrate moisture to the environment, samples were taken for
the experiments. For the measurement of each measured quantity (including separation) three tests were
made, the value presented is the average value. The characterization of the basic properties (density,
humidity, equivalent spherical diameter and sphericity) of the previous biomass was carried out in order to
estimate the terminal velocity of the particles. The separation experiments by elutriation were executed in the
SESY -sedimentation elutriator separator system, illustrated in Figure 1. SESY is composed of a tube riser (R)
with a glass section for visual inspection of the fluid dynamic behavior, internal diameter of 0.1 m and 1.8 m in
height. The riser is followed by a U-bend downward to end in a sedimentation chamber (SC). Air suction is
performed by an air exhaustor (AE). A filter is placed between the sedimentation chamber and the air
connection to avoid the output of the material. Air velocity was controlled by valve (V). The material to be
separated by elutriation is spread in the vibrating feeder system (VS) to dose the biomass into the riser. The
dragged material is collected in the sedimentation chamber and the sedimented material is collected at the
bottom of the riser to be feed in the vibrating system at the next speed. The air flow rate was measured with
an orifice plate system (OP) constructed in accordance with ASME MFC-14M/2003 standard and pressure
was measured with Smar LD301 transducers (P). The air temperature of the system was measured to
determine the air density. The device is able of producing superficial air velocities between 0.5 and 6.5 m/s.
Using atmospheric air at ambient pressure and temperature as the working fluid, an airflow was generated in
the riser at different velocities, where biomass particles were feed in the airflow. The feeding of the particles
was done continuously and in small quantities, preventing a large amount from being simultaneously arranged
in the riser, in order to minimize the agglomeration of particles.
The density of biomass samples was measured with water pycnometry. The Sauter diameter was calculated
based on weight fraction of the separated material by sieving. The projected area of the particles was obtained
by image scanning. The ImageJ program was used to measure this dimensions, as well as major and minor
axis length for the calculation of sphericity, Eq(7), based on Wadell’s original definition. These dimensions are
described by Baxes (1994). ∅ = (7)
Where is the surface area of a sphere of the same volume as the particle, and is the actual
surface area of the particle.
Figure 1: Experimental bench SESY. Air exhaustor (AE), elutriated material collector (Ce), sedimentary
material collector (Cs), pressure gauge (P), riser (R), sedimentation chamber (SC), temperature gauge (T),
valve (V) and vibratory feed system (VS).
609
The bagasse was separated with velocities of 0.65 to 6.53 m/s and straw at 0.65 to 5.50 m/s. For the terminal
velocities predicted by correlations the size distribution obtained by sieving was used. Three correlations were
used: the equation presented by Grobart (1973) and the correlations of Haider and Levenspiel (1989) for
spherical and non-spherical particles.
3. Results and analysis
Figure 2 shows the results of the preliminary sieving for bagasse and straw. From Figure 2, it can be seen that
the biomass is concentrated in the larger sieves (51% for bagasse and 32% for straw). The biomass
properties are presented in Table 1.
Table 1: Biomass sample properties.
Biomass
Density
(kg/m³)
Sauter mean diameter
(µm)
Φ
W
(%)
Bagasse 617 ± 174 684 ± 50 0.67 5 %
Straw 333 ± 82 1036 ± 118 0.76 3 %
Figure 2: Sieving analysis on particle size distribution.
Figure 3: Sieving analysis on particle size distribution
610
Due to velocity limitations of the measuring system, it was able to separate by elutriation 98% of the bagasse
and 97% of the straw. Figure 3 shows the accumulated mass distribution for both bagasse and straw for
separation, by elutriation and sieving. Clearly, Figure 3 presents that due to the higher concentration of
bagasse particles in the sieves with higher mesh opening, it need higher velocity than the straw to be dragged.
From Figure 3, we notice that the behavior of particles separated by elutriation and the particles separated by
sieving is very different, showing the influence of the irregular shape and size of particles.
Figures 4 and 5 show the accumulated mass fraction as a function of the terminal velocity of the particle both
measured in the elutriator and predicted by correlations for the bagasse and the straw, respectively. In the
case of bagasse, Figure 4, a good agreement was found between the measured elutriation velocity and the
one obtained by the Haider and Levenspiel (1989) equation for non-spherical particles up to 2.52 m/s. For
straw, a slight approximation was found between the measured elutriation velocity and the one obtained by
Haider and Levenspiel (1989) equation for non-spherical particles ut to 2.00 m/s. It should be noted that the
velocities provided by the correlations are lower than those measured experimentally, e.g., for the estimation
of the terminal velocity of the coarse fraction of bagasse, sieved with higher mesh opening and calculated with
the equation Haider and Levenspiel (1989) for non-spherical particles, is 55 % (59 % for straw) less than the
measured value for the coarse particles of bagasse separated by elutriation at the highest velocity.
Figure 4: Mass fraction accumulated versus terminal velocity of bagasse particles, measured and predicted by
correlations of Grotart (1973), Haider and Levenspiel (1989), and Sosa-Arnao and Nebra (2009).
Figure 5: Mass fraction accumulated versus terminal velocity of straw particles, measured and predicted by
correlations of Grotart (1973), Haider and Levenspiel (1989), and Sosa-Arnao and Nebra (2009).
611
4. Conclusions
The results showed great difference between separation techniques by sieving and elutriation for
characterization of size particles. Terminal velocity values obtained experimentally differ, in large part, of the
value calculated by classical correlations of literature, due to the variety and irregularity of the particle shape
and size distribution from biomass.
Reference
Anderson C.N., 1988, Aerodynamics of bagasse particles, Proceedings of the Australian Society of Sugar
Cane Technologists, Cairns, Queensland, Watson Ferguson and Company, 235-239.
Bailer W., 2006, Writing ImageJ Plugins – A Tutorial, Version 1.71
<http://rsb.info.nih.gov/ij/download/docs/tutorial171.pdf> accessed July 29.07.2010
Barbosa-Cortez L.A., Silva-Lora E.E., Olivares-Gómez E., 2008, Biomass for energy, Eds. Unicamp,
Campinas, Brazil (in Portuguesse).
Baxes G.A., 1994, Digital Image Processing: Principles and Applications. John Wiley & Sons, New York,
United States of America.
Carvalho D.J., Veiga J.P.S., Bizzo W.A., 2017, Analysis of energy consumption in three systems for collecting
sugarcane straw for use in power generation, Energy, 119, 178-187,
DOI: doi.org/10.1016/j.energy.2016.12.067
Coelho-Carvalho L., Freitas-Bueno R.C., Carvalho M.M., Favoreto A.L. Godoy A.F., 2013, Sugarcane and fuel
alcohol: history, sustainability and energy security. <www.conhecer.org.br> accessed 01.07.2013 (in
Portuguesse).
Colakyan M., Levenspiel O., 1984, Elutriation for fluidized beds, Powder Technology, 38, 223-323,
DOI:org/10.1016/0032-5910(84)85005-6.
Haider A., Levenspiel O., 1989, Drag Coefficient and terminal velocity of spherical and nosplherical particles,
Podwer Technology, 58, 63-70, DOI: doi.org/10.1016/0032-5910(89)80008-7
Igathinathane C., Pordesimo L.O., Columbus E.P., Batchelor W.D., Methuku S.R., 2008, Shape identification
and particles size distribution from basic shape parameters using ImageJ, Computers and Electronics in
Agriculture, 63, 168-172, DOI: 10.1016/j.compag.2008.02.007
Kunii D., Levenspiel O., 1991, Fluidizationg engineering, Second Edition, Butterworth-Heinemann, Stoneham,
United States of America.
Nikky M., Jalali P., Hyppanen T., 2014, Characterization method of average gas-solid drag for regular and
irregular particle groups, Powder Technology, 253, 284-294, DOI: 10.1016/j.powtec.2013.11.035
Rendeiro G., Martins-Nogueira M.F., Almeida Cruz D.O., Silva-Guerra D.R., Negrão-Macedo E., Araújo-
Ichíhara J., 2008, Solid biomass combustion and gasification, Ministry of Mines and Energy, Brasilia, Brazil
(in Portuguesse).
Sosa-Arnao J. H., Nebra S. A., 2009, Bagasse Dryer Role in the Energy Recovery of Water Tube Boilers,
Drying Technology, 19, 587-594, DOI: 10.1080/07373930802716326
Wadell H., 1932, Volume, shape and roundness of rock particles, Journal of Geogoly, 40, 443-451.
Yang W.-C., Eds., 2003, Handbook of fluidization and fluid-particles systems. Marcel Dekker, Pennsylvania,
United States of America.
612