Journal of the Energy Institute 106 (2023) 101146

Contents lists available at ScienceDirect

Journal of the Energy Institute

journal homepage: www.elsevier.com/locate/joei

Microwave pretreatment of biomass for conversion of lignocellulosic

materials into renewable biofuels
A.Z. Fia, J. Amorim *
Department of Physics, Instituto Tecnológico de Aeronáutica, 12228-900, São José dos Campos, Brazil

A R T I C L E I N F O A B S T R A C T

Keywords: For the conversion of lignocellulosic material into renewable biofuels the microwave pretreatment of raw ma­
Microwave simulation terial is a new and promising technology. This work presents a numerical simulation of microwave heating of
Pretreatment of biomass materials derived from sugar cane, african palm and green algae in microwave domestic oven. The power ab­
Renewable energy
sorption of biomasses is studied in cylindrical and spherical shapes, with volumes varying from 10− 5 m3 to 100
Temperature
Electric fields
× 10− 5 m3, as function of humidity (from 20% to 80%). Maxwell equations were employed to calculate the
microwave electromagnetic field distribution in the microwave cavity and in the samples, coupled to heat
equation when the humidity is varied. Penetration depth varied from 2 × 10− 2 to 10 m and tan δ from 2 × 10− 3
to 7 × 10− 1 as function of humidity. Plots of electric field and temperature distribution inside the oven were done
for sugar cane, palm oil and green algae for cylindrical and spherical samples for 30% and 65% of humidity in the
samples. The coefficient of variation (COV) varied between 0.05 and 0.18. The rate of heating efficiency was
evaluated using the degree of thermal runaway ΔT (5 K < ΔT < 75 K) and evolution of average temperature T
(300 K < T < 370 K) and their relationship with the energy absorbed by the samples and COV, respectively. The
comprehensive evaluation coefficient, that describes the usability of a specific lignocellulosic material in in­
dustry, varied between 0.3 and 10.0, was calculated for the biomasses shapes and sizes as function of power and
humidity.

1. Introduction The primary energy consumption in the world increased around 38%
since 1990s [3]. The reserves are finite and in near future absence of
The increasing threats posed by the novel coronavirus (SARS-CoV-2) energy will be a reality if new actions to reduce the fossil fuel con­
pandemic is alerting the world to face the challenges and address an­ sumption are not done. On the other hand, the high demand of fossil
swers fast and based on science. Connections between the COVID-19 fuels causes the increase in greenhouse gases (GHGs) in the atmosphere.
global pandemic crisis and the climate change challenge motivates Electricity, heat generation and transport produced around two-thirds of
many of scientists specifically, and the society in general, to work global CO2 emissions [4]. Carbon dioxide is the primary focus of inter­
collaboratively to face the dare on renewable energy sources once there national debate to reduce GHGs because itself comprises 80% of the total
are many parallels between the defiance in fighting this global pandemic emissions of GHGs, not including land-use change and forestry, from the
and fighting the climate change in the longer term to prevent a climate developed countries, according to the Kyoto Protocol.
crisis [1]. The Intergovernmental Panel on Climate Change (IPCC) alerts Personal vehicles are the major cause of global warming. Cars and
that warming is already happening, and we have around three decades trucks account for nearly one-fifth of all US emissions. Around 24
left to drastically reduce our greenhouse gas emissions and remain in the pounds of CO2 and other global-warming gases is emitted for every
target scenario of 1.5 ◦ C above preindustrial levels, according to the gallon of gas used. About five pounds comes from the extraction, pro­
Paris Agreement [2]. To achieve this goal, measures for increasing the duction, and delivery of the fuel, while the great bulk of heat-trapping
energy efficiency must be done with strategies that remove the carbon emissions, more than 19 pounds per gallon, comes right out of a car’s
dioxide from the atmosphere reducing the consumption of fossil fuels tailpipe [5]. In European Union (EU), transport is responsible for nearly
and promoting renewable energy sources. 30% of the total CO2 emissions, of which 72% comes from road

* Corresponding author.
E-mail address: jayr.de.amorim@gmail.com (J. Amorim).

https://doi.org/10.1016/j.joei.2022.11.006
Received 15 September 2022; Received in revised form 16 November 2022; Accepted 21 November 2022
Available online 6 December 2022
1743-9671/© 2022 Energy Institute. Published by Elsevier Ltd. All rights reserved.
A.Z. Fia and J. Amorim Journal of the Energy Institute 106 (2023) 101146

transportation. As part of efforts to reduce CO2 emissions, the EU has set bounds between these molecules are broken, the energy stored is
a goal of reducing emissions from transport in 60% by 2050 compared to released [7].
1990 levels [6]. Only in transport sector, the CO2 emissions from cars In Europe, the biomass accounts for more than 62% of all renewable,
totalize 60.7%, heavy duty trucks 26.2%, light duty trucks 11.9% and in which wood has approximately 80% of the biomass for the renewable
motorcycles 1.2%. energy [8]. In the transportations sector, the biofuels are already largely
The replacement of fossil fuel is an urgent issue as stated above, and deployed in commercial scale in many countries, such as the
due to price hikes and environmental damage, it motivated the search first-generation bioethanol from sugar cane in Brazil [9] and from corn
for economic and eco-benign alternative of gasoline and diesel fuels. in United States [10].
One of the strategic options to face the concerns of climate change is Biomass is currently converted to liquid fuels by fermentation of
replace the fossil-fuels, and thus the CO2 that is emitted when fossil carbohydrates to ethanol, or by extraction and refining of plant oils.
reserve is burned. Significant challenges remain in the conversion of biomass to liquid
The biomass-based fuels are a possible friendly solution once the fuels once only a fraction of the plant is used in current technologies, and
carbon dioxide released during combustion is sequestered back into significant energy is consumed in producing and processing biologically
plant material by photosynthesis. On the other hand, the energy derived fuels. Overcoming these challenges could improve the overall
required to cultivate, harvest, and process the biomass and its associated efficiency of a biomass energy system and reduce the cost of biomass
CO2 emission must be considered to have a zero net balance in a bio­ fuels. The challenge of land-use remains an important issue if biofuels
refinery context. So, the production and use of biomass requires the have to be produced on a scale comparable to current fossil fuel
evaluation of the available land in assessing the ultimate potential of production.
biomass fuels on a global scale. To modify the matrix-based roles of Creation of biomass requires sunlight, water, nutrients, and land.
petroleum-based fuels, that currently play in transportation, significant Sunlight reaching the Earth’s surface have an average energy density of
conversion of biomass to liquid biofuels would be required. roughly 180 Wm− 2, annually averaged over the entire Earth’s surface
Biofuels, as the name suggests, are the fuels derived from biological [11]. The maximum photosynthesis conversion of sunlight to chemical
sources like plants, animals, microbes, which are biodegradable, non- energy is about 6.7%, and only 0.3% of the energy of sunlight is stored as
toxic, and environmentally safe. Biomass resources can be considered carbon compounds in land plants. There is opportunity in science to
as the organic matter, including algae, trees, and crops, in which the improve the use of sunlight to maximize the energy absorption by plants
solar energy is stored during photosynthesis in the carbon, oxygen, and perhaps making use of genetic engineering or studying intensive
hydrogen chemical bonds of molecules in the raw material. When breeding. The potential to expand the use of biomass energy, either for

Fig. 1. Generations of biofuels.

2
A.Z. Fia and J. Amorim Journal of the Energy Institute 106 (2023) 101146

electricity or liquid fuels, relies on the demand for the energy, the eco­ enhance biodiesel production with aid of higher algae lipid content. The
nomic viability, the technology for conversion, the availability of land, aim is to enhance the photosynthetic efficiency improving light pene­
and public policy factors. The ability of biofuel other than lignocellulosic tration using genetic modification of microalgae, macroalgae and cya­
based fuel, e.g. capture and sequestration of CO2, to minimize GHGs nobacteria. This 4G biofuel technology is new and still under
emission is about 13%, while the cellulose biomasses-based biofuel can development [19].
reduce this up to 88%, and the emission of carbon dioxide almost zero The recalcitrance of LCMs demands a pretreatment to breakdown the
[12]. complex lignocellulose matrix structure [20]. It is a crucial step to
According to the feedstock, the biofuels fall into four groups: first, expose the cellulose and hemicellulose contents to enzymatic or chem­
second, third, and fourth generation biofuels [12], see Fig. 1. In the last ical hydrolysis, to produce simple sugars that are then fermented to
thirty years the first-generation (1G) biofuel has been extensively used. produce biofuels, once the crystalline structure of cellulose is modified,
These biofuels include pure plant oil, biodiesel from esterification of hemicellulose is depolymerized, and lignin may be removed [21]. This
plant oil or waste vegetable oils, bioethanol from sugar or starch crops step is the most expensive stage in 2G bioethanol production, and the
maturation. Examples of important biomass for 1G bioethanol are: sugar consumption of energy is high, accounting up to 48% of the total
cane, maize, and sugar beet. However, these raw material fuels sources operation cost [22]. The pretreatment is also a necessary step in 3G
do not fulfill the rising energy demand of the world and may compete biofuel synthesis for algae cell disruption [23], for extraction or for
with food crops. Among the countries that uses the 1G bioethanol we chemical conversion of intracellular compounds such as proteins, car­
may cite Brazil and USA. Brazil developed the 1G ethanol from bohydrates, lipids, oligosaccharides, pigments etc. [24].
fermentation of the sugarcane juice. USA promoted the known crop The pretreatments may be classified as physical, chemical, and bio­
maize grain for ethanol production, after hydrolyzation and formation logical. The physical pretreatments are milling [25], extrusion [25],
of this starchy source which is subsequently fermented. These primary plasmas [26], and microwaves [25]. Chemical pretreatments are alka­
crops as a feedstock for 1G biofuel requires high number of chemical line dilute acid [25], ionic liquids [25], organosolv [25] and ozonolysis
fertilizers to achieve a desired yield. Concerning the market, the only [27,28]. As biological pretreatments we have for example fungus [25].
matrix feedstock with ability to compete in the petroleum-based fuels In some cases, the combination of the above pretreatments is necessary.
prize is the sugary rich sugar cane. Microwave pretreatment has attracted attention of researchers in
The second-generation (2G) biofuels obtained from the lignocellu­ recent years as a green physical-chemistry alternative. When LCMs are
losic materials (LCMs) have been proposed as an alternative in the heated by microwaves the breakdown of lignocellulosic structure is
replacement of fossil-based fuels [13]. The production of 2G biofuels assured by molecular collisions due to dielectric polarization. Polariz­
was adapted to overcome the important limitation of 1G biofuels, that is able particles movement induces a quick heating of the samples and
the feedstock are generally utilized for food supplements. They are therefore the temperature and processing times are reduced.
originated from non-food crops or non-edible residues, like leaves, Several works on microwave-based pretreatment for biomass con­
bagasse, or straw, with minimum effect in foods. The 2G bioethanol is version have been done recently. Most of reviews available do not
based on sugar contained in the cell wall of bagasse or straw, for comprehensively analyze the microwave physicochemical pretreatment
example. To access the cell wall cellulose and decompose it into simple techniques for biomass conversion [29] which are crucial for sustainable
sugars available for fermentation, a hydrolyzation step is added to the biofuel generation. This kind of material heating do not use solvents
hole process. They do not conflict with food industry and propitiate neither separation products. It presents the advantage of reduced time of
countries to choose their most interesting solution according to the processing, around ten times lower than classical heating, and reduces
biomass availability. Genetic engineering is now being applied to lessen the energy consumption due to its fast energy transfer. Microwaves often
the level of lignin in trees, for example, to change the structure of increases the yield and purity of the extract [30].
hemicelluloses [3]. The goal is to reduce the ethanol production cost This work studies the use of microwaves as a pretreatment method to
competing with fossil fuels. heat samples of materials derived from sugar cane, african palm and
Limitations found in 1G and 2G biofuels such as low returns on in­ green algae when they are placed in a domestic microwave oven. The
vestments have led the establishment of third generation biofuels (3G). aim is to optimize the treatment to avoid the formation of hot spots. It is
The raw material for 3G biofuels is photosynthetic microbes like algae determined the electric field and temperature evolution with time in
and cyanobacteria. samples with different shapes, sizes, and humidity, employing a do­
Although algae have many derived secondary metabolites known for mestic microwave oven as reactor, operating with frequency of 2.45
their skin benefits, which include protection from UV radiations and GHz in TE10 mode feed by a rectangular waveguide. The solution of
prevention of rough texture, wrinkles, and skin flaccidity [14], it is a 3G electromagnetic and heat equations is done in COMSOL Multiphysics
biofuel source capable of yielding more biomass with less resources that platform. It is used the Microwave Oven included in software examples
is attracting the attention of research and industrial community. In with proper modifications which is a multimode non-stirred batch sys­
contrast to crops used for 1G or 2G biofuels, they have an exceptionally tem. This study is useful to determine the conditions for efficient mi­
low-density before harvesting. The multifold advantages of algae like crowave assisted material processing via forecasting of the expected
either low-area requirements and tolerance to harsh conditions makes heating characteristics. It is also important to allow the selection of the
them a good option for biofuel production and mitigation of CO2 [15] or appropriate operating parameters, to maximize the use of microwave
being cultivated in fresh water, seawater or in industrial wastewater. heating process as a pretreatment of biomasses.
The growth rate is around 30 times faster that food yielding crops and It is investigated the temperature and electric field distributions
the oil content is approximately 30 times more than conventional 1G around samples of sugar cane, african palm and green algae biomasses.
and 2G feedstocks [3]. The three major components of algae are lipids Due to non-homogeneous composition of the material and geometry,
(5–60%), proteins (40–60%) and carbohydrates (8–30%) in varying electromagnetic waves reflection results in formation of the standing
proportions [16]. Algaes can be a source for bioethanol production once waves. This resonance phenomenon leads to a local overheating, also
they contain a good amount of carbohydrates in the cell wall. Due to known as hot spot. Hot spots in materials may be controlled with the
high lipid content as high as 60% of its weight algae is also a good source distributed microwave incidence in a closed system, besides controlling
of biodiesel produced by the transesterification of its fatty acids [17]. the proper thickness of the samples [31]. Conditions for hot spot for­
Production of biokerosene is another potential and important fuel ob­ mation should be avoided to prevent burning of the sample which may
tained from algae [18]. However, algae biofuel does not compete today strongly change with temperature and humidity what therefore also
with food crops on arable land due to high price of this technology. cause influence in the microwave absorption [32]. The paper focus on
Fourth generation biofuel (4G) uses genetically modified algae to the importance of humidity (from 20% to 80%) as well as its dimensions

3
A.Z. Fia and J. Amorim Journal of the Energy Institute 106 (2023) 101146

(1 × 10− 5 m3, 2 × 10− 5 m3, 3 × 10− 5 m3, 4 × 10− 5 m3 and 5 × 10− 5 m3) or dissipation factor, tan δ = εε′ . It describes the material capacity to be
′′

and shapes (cylinder and sphere) of samples when the microwave power heated by microwaves. It is dimensionless and reflects that a dielectric
is varied. material with ε′′ = 0 the dissipation factor is zero and do not heat.
According to dielectric loss tangent value the material may be
2. Fundamentals of microwave heating processing of biomass conductive, non-conductive, and dielectric. The conductive material is
not transparent to the microwaves with tan δ < 0 such as metals that
Since the microwaves were used in the Second World War to build reflect the incident microwave beam so microwaves cannot penetrate.
the radar, the microwaves experienced a broad use in many domains The non-conductive or microwave-transparent materials, have
such as in elemental analysis, medicine, in the kitchen for heating and 0.1 < tan δ < 0.5, examples are Teflon, glass, quartz, and air. For ma­
drying foods. Domestic applications in heating and food applications terial with tan δ > 0.5, i.e. dielectric materials, the microwaves are better
were proposed by the engineer Percy L. Spencer [22]. Recently, mi­ absorbed and are ideal material to heat under microwaves irradiation
crowave assisted heating has also been widely used for pretreatment of like water, oils etc. [34]. In most pretreatment of LCMs, water is the
biomass for the production of biofuels and bioenergy [33]. main solvent and ionic conduction of H+ ions have significant contri­
Microwaves interact with matter by the action of the electromagnetic bution to dielectric loss factor, being water the main molecule respon­
field waves with the induced dipoles in the material and with free ions sible for electric polarization [40]. Substances with low pH have higher
present in the medium. These interactions may be represented by H+ density and are easily heated. Protons may be displaced easily in
conductive losses, dielectric, and magnetic losses. One of the mechanism biological materials and could be affected by microwaves. This puts in
responsible to transform the energy stored in electromagnetic field of relief the importance of microwave heating study as function of the
microwaves into heat is basically the change of dipole that try to align humidity. Dissolutions of carbohydrates and salts in water reduce the
the polar molecules of the material with the radiation field. The relative permittivity. In mixtures with lower amount of salt the dielectric
realignment of polar molecules produces a displacement inside of the properties are regulated by water at high frequencies and sugar have a
material which generates heat [34] by friction. The second mechanism, weak effect on polarization and dielectric loss [41]. The reduction of
due to interaction of the electric field of the radiation with ions present relative permittivity with water in solutions or mixtures is the result of
in the sample, produces the movement of ions leading to collisions in the two mechanisms: the water volumetric replacement with a substance
bulk of the sample during their conduction, generating heat. Energy with a lower permittivity and the binding of free water molecules [42].
conversion in microwave irradiation results in volumetric heat genera­ Due to the aforementioned arguments, lignocellulosic are good material
tion within the target material and not through the material surface such to interact with microwaves due to the intrinsic polarities of cellulose,
as in conventional heating. hemicellulose and lignin in the presence of water [22].
Microwaves have frequency in the range from 300 MHz to 300 GHz Other important parameters are the electromagnetic wave propa­
with wavelength between 1 m and 1 mm respectively. The two fre­ gation within the sample, the wavelength (λm ) and penetration depth
quencies employed in the industry for processing of foods and biomass (Dp ), defined as:
are 915 MHz and 2.45 GHz. They are non-ionizing waves to distinguish √̅̅̅
from other forms of ionizing radiation such as X-rays or gamma rays. c 2 1
λm = [√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅ ]1/2 (1)
They have low energy in their photons 0.09 cal/mol for 915 MHz and f ′2
ε + ε′′ 2 + ε′
0.23 cal/mol for 2.45 GHz if compared with typical energy found in
chemical bonds ranging from 50 kcal/mol to 100 kcal/mol.
c 1
Microwave technology has been used in many processes such as Dp = √̅̅̅ [√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅ ]1/2 (2)
2πf
drying, thawing, and heating. In the heating of food by microwaves a ε′ 2 + ε′′ 2 − ε′
quarter of the time is spent if compared with conventional heating,
keeping the quality, and preserving nutrients, flavor, color, and vitamin where c is the velocity of light and f the frequency of the radiation. The
contents thanks to the rapid energy transfer. High-power magnetrons wavelength λm is the wavelength of radiation within the sample, while
have efficiency of 85% at 915 MHz and 80% at 2.45 GHz, leading to an the penetration depth Dp determines the distance at which microwaves
overall remarkably high efficiency of microwave system [35]. The penetrate in the sample. The absorption and transfer of microwaves is
disadvantage is the non-uniform temperature distribution, originating also affected by the chemical composition, thermal diffusivity, thermal
hot and cold spots in the samples. Recent applications try to solve this conductivity, specific heat, particle size and density.
problem focusing the microwaves inside cylindrical tubes to minimize The analysis of microwave absorption by materials samples from
the non-uniform heating according to the diameter, flow rates and sugar cane, palm and algae has been performed based on four dimen­
dielectric properties of material [36,37]. sionless parameters, Nw , Np , fp and fw , where fp and fw capture the effect
To simulate the interaction of microwaves with a biomass placed of dielectric properties of the sample material, while Nw and Np captures
inside an oven, one need to find the numerical solution of Maxwell the effect of the sample dimension:
equations describing the electromagnetic field and the heat equation to
obtain the electric field and temperature distribution in loaded domestic Nw =
2R
; fp =
λm λm
; fw = ; Np = 2πfp Nw =
2R
(3)
microwave oven, determining the hot and cold spots when the cavity is λm 2πDp λ0 Dp
filled with a dielectric material. The parameters characterizing dielectric
λ0 is the wavelength of the microwave within the free space, Nw is the
materials are the relative permittivity ε’ (real part) and the relative loss
relative magnitude of the sample in relation to the wavelength of radi­
factor ε” (imaginary part). The relative permittivity ε’ (or dielectric
ation within the sample λm and Np is a relative magnitude of the sample
constant), representing the ability of molecules to be polarized under the
presence of an electric field and thereby the capacity of the material to dimension with respect to the penetration depth. Parameters fp and fw ,
store the energy. This dielectric constant is affected by the molecular related to the dielectric constants of the material, always remain
weight, shape, and direction of their bonds. The relative loss factor ε” bounded between 0 and 1 irrespective of the magnitude of the dielectric
represents the amount of energy lost in the material that is converted properties. The penetration number or Np = 2R Dp , has sometimes been
into heat, caused by the electron free conductivity [38]. As lower as its preferred in the literature to characterize the microwave power ab­
value lesser is the ability of the material to absorb microwaves. It may be sorption within large samples.
partially self-controlled because this dielectric loss decreases with the The parameters Nw and fw mainly influence the resonating features of
increase of the temperature [39]. the power absorption. Resonances occur due to the constructive in
The ratio of these two parameters is known as dielectric loss tangent

4
A.Z. Fia and J. Amorim Journal of the Energy Institute 106 (2023) 101146

terferences between the traveling waves, which are formed due to the equation as function of the humidity in the samples. In this work the
transmission and reflection of microwaves at the sample boundary, COMSOL Multiphysics (COMSOL Inc., Burlington, MA, USA) platform is
leading to the thermal hot spot during the processing of materials in the used to solve both electromagnetic, and heat transfer equations simul­
presence of the microwave radiation. taneously. COMSOL has already been successfully implemented to
For sugar cane, palm oil and green algae biomasses, in order to model microwave heating and compare it with experimental ones by
determine λm , Dp , Nw , Np , fp , and fw it is necessary to know the param­ several researchers [47,48]. It was used previously to simulate the
eters characterizing dielectric materials, i.e. the relative permittivity ε’ interaction of microwave radiation with empty fruit bunch pellets and
and the relative loss factor ε”. potato samples [49,50]. Because limited simulation works are published
The behavior of relative permittivity and relative loss factor for sugar in the literature on microwave heating of biomass, the effects of sample
cane bagasse [43], palm oil [44] and green algae [45] are shown in loading size and specific heat capacity were not fully understood.
Fig. 2. As can be seen the relative permittivity as function of humidity for Numerical study was done with the following assumptions to
the three biomasses investigated in this work are very distinctive, see simplify the problem and reduce computational time:
Fig. 2 (A). Lower values of relative permittivity are found for the sugar
cane reveling its low capacity to be polarized under the presence of an • Samples are homogeneous and isotropic. The special structure of
electric field, i.e. low capacity of the material to store the energy. The biomasses may lead to uneven temperature distribution during the
green algae and the palm molecules are more polarized by the micro­ microwave heating due to a nonuniform distribution of water inside
wave electric field. The palm’s relative permittivity ε′ is higher than the samples. In this work, due to the absence of experimental results
green algae for humidities higher than 50%, and in this region is more published in the literature, to our knowledge, the samples in the
dipolar, presenting more dielectric losses and therefore more heat is simulations were supposed to be homogeneous and isotropic.
generated inside. • Heat transfer has not been resolved spatially in the domains of air or
The relative loss factor ε” as function of humidity are shown in Fig. 2 glass;
(B). It can be noted that green algae convert lower amount of energy into • The heat transfer equations were solved with variable dielectric
heat than sugar cane bagasse and palm oil. In the same figure one can see constant and loss factor as function of the humidity in the samples.
that palm oil is more effective in the transformation of energy received Thermal conductivity properties are assumed constant, and values
from the electric field into heat, being more susceptible to be treated are taken from the literature.
with microwaves. • The initial temperature of the biomass and air was considered
homogeneous.
3. Physical model of microwave heating • The mass transfer and the change in the process phases were ignored.
The temperature distribution during heating occurred until the
Modelling of microwave heating is an auxiliary tool aside experi­ maximum temperature of the hot spots reached the phase transition
mental results to study the behavior of parameters on the raw material temperature at atmospheric pressure.
that affects the absorption of energy from the electromagnetic fields. It is
helpful to find solutions to problems where nonuniform heating occurs This can be a good assumption because the time considered in these
[35]. The mathematical models predict the temperature and electric simulations are short and the thermal conductivity of the air inside the
field distribution patterns in the heated material. The LCMs, especially oven is smaller than the conductivity of the biomasses under study.
cellulose, can be seen as conductors in the ionic crystalline regions and LCM from sugar cane, palm oil and green algae are the biomasses
non-conductors in amorphous regions [46]. High microwave heating employed in the simulations. Its properties such as thermal conductivity
regions and vibrational motion may result in rupture or explosion of (W/mK), density (kg/m3), specific heat (J/kg.K), relative permeability
some LCMs components, causing the crystalline cellulose relocation and electric conductivity are shown in Table 1.
[22], and LCMs breakdown through molecular collisions due to dielec­ The relative permeability and electric conductivity used in this work
tric polarization. These effects put in evidence the need for microwave were supposed equal to 1 and 0 respectively [56]. It should be
simulations of pretreatment of LCMs, due to temperature increase in mentioned that the dielectric and thermal properties are, in general,
polar parts of the material leading to hot spots formations. functions of the frequency of radiation and the physicochemical prop­
Computational model used in this work computes the complex erties of the material, such as: temperature, chemical composition, and
electromagnetic Maxwell’s equations coupled with heat transfer moisture content, and the values reported here are representative of
each material over a wide range of the parameters employed in the
simulations done in this work.
The simulations of microwave radiation in interaction with the
aforementioned biomasses are realized with the COMSOL Multiphysics
software version 4.3b Academic Class kit License (COMSOL Inc., Bur­
lington, MA, USA). The Maxwell equations are solved and coupled to
heat transfer equations for the Microwave Oven included COMSOL
Multiphysics software. COMSOL Multiphysics software can solve these
models iteratively using finite element method. Simulation work was
carried out on a computer which had Intel Xenon 64 Cores, 2.13 GHz

Table 1
Properties of biomasses.
Biomass property Biomass

Sugar cane bagasse Palm oil Green Algae

Thermal conductivity (W/mK) 0.059 [51] 0.03 [52] 0.61 [53]

Density (kg/m3) 80 [54] 1420 [52] 1000 [53]
Fig. 2. Relative permittivity ε’ (A) and the relative loss factor ε” (B) as function Specific heat (J/kgK) 3648.9 [55] 1150 [49] 4200 [53]
Relative permeability 1 1 1
of humidity in the sample. Sugar cane bagasse ( ), palm oil ( ) and green
Electric conductivity 0 0 0
algae ( ).

5
A.Z. Fia and J. Amorim Journal of the Energy Institute 106 (2023) 101146

processor with 256 GB RAM memory and 64-bit system. The process of
building the numerical model is shown, see Fig. 3.
The schematic of the microwave oven used in the physical model can
be seen in Fia and Amorim [56]. It consists of a microwave oven cavity, a
waveguide, a glass tray, and biomass sample (with different sizes and
shapes other than the one shown). The cavity and rectangular wave­
guide were made of copper and the size of cavity had length W0 = 0.267
m, width D0 = 0.270 m, and height H0 = 0.188 m. The rectangular
waveguide was length Wg = 0.05 m, width Dg = 0.078 m, and height Hg
= 0.018 m. Glass tray (Rp = 0.1135 m) was regarded as transparent
medium for the microwave electromagnetic field, and so it absorbs no
heat.
Mesh generation is an important issue in the simulation problem,
once it allows the discretization of geometry in small units of simple
shapes. The quality value of the mesh elements, guarantee of conver­
gence of the simulation and the achievement of accurate results, is given
between 0 and 1, where 0 represents a degenerate element and 1 rep­
resents a perfectly regular element [57]. In this model, all geometry
Fig. 4. Quality evaluation of mesh for green algae cylinder, volume equal to
configurations presented a mesh quality of 0.6587, which is sufficient to
40.77 10− 5 m3 and 65% of humidity.
obtain satisfactory results, taking into account the computational cost
[58]. The evaluation of the quality of the mesh element is shown in
by:
Fig. 4.
The mesh was controlled by the free mesh generator available at → →
D = ε0 εr E
COMSOL, with free tetrahedral meshes and normal element size. For
example, for green algae shape of cylinder, volume of 40.77 10− 5 m3 and → →
B = μ 0 μr H (5)
65% of humidity the complete mesh consists of 110964 domain ele­
ments, 9689 boundary elements and 618 edge elements. For other bio­ → →
j =σE
masses of varying sizes and shapes, the mesh varied from 23244 to
233863 domain elements.
where εr is the relative permittivity, μr is the relative permeability, σ is
The electromagnetic field distribution inside the microwave cavity,
the conductivity, ε0 is the permittivity of the free space (ε0 = 8.854 ×
sample and waveguide may be described by the Maxwell equations in
10− 12 F/m), and μ0 is the permeability of free space (μ0 = 4π ×
the differential form:
10− 7 H/m).
→
∇. D = ρe The electromagnetic energy is converted into heat when it interacts
with dielectric material. This conversion is governed by the amount of
→
→
∂B power absorbed (Pab ) into the material, which is defined by:
∇× E = −
∂t ωε0 ε′′ |E|2
Pab = (6)
→ 2
∇. B = 0 (4)
where the dielectric permittivity of the material (εr ) may be expressed
→
→ → ∂D as:
∇× H = j −
∂t
(7)
′
εr = ε − jε′′
→ → →
where D is the electric flux density, E is the electric field intensity, B is
where ε is the real part known as the dielectric constant and ε′′ is the
′
→
the magnetic flux density, H is the magnetic field intensity, ρe is the imaginary part known as the loss factor. The former reflects the ability of
→
charge density, and j is the current density. the material to store electric energy, whereas the later influences the
The interaction of the electromagnetic field with the biomass is given conversion of electromagnetic energy into heat.

Fig. 3. The process of building the numerical model by COMSOL Multiphysics.

6
A.Z. Fia and J. Amorim Journal of the Energy Institute 106 (2023) 101146

As no conduction current exists in the microwave cavity of the oven 4. Results

→
and in the sample, ρe = 0 and j = 0. The wave equation in microwave
cavity is given by: The results of simulations were analyzed based on absorption re­
( ) gimes, electric field and the temperature distribution in the cavity when
∇ × μ−r 1 (∇ × E) − k02 εr −
jσ
E=0 (8) the samples of LCM from sugar cane, palm and green algae are placed in
ωε0 a domestic microwave oven. Cylinder and sphere geometry were
investigated as function of heating time and humidity from the quali­
where the wavenumber in free space k0 is defined as: tative and quantitative points of view by aid of images obtained.
ω With the calculated values of λm , Dp , Nw , Np , fw and fp for the bio­
k0 = (9)
c masses, see Table 2, one can find the power absorption regimes [59,60].
Sugar cane, palm oil and green algae were chosen because they have
and c is the speed of light in vacuum.
wide variations of ε (1 .3< ε < 56) andε′′ (1.0 × 10− 2 < ε′′ < 24)
′ ′

The biomass, here treated as solid, interacts with the electric field leading to a large range for fp and fw , and due to their importance as
inside the microwave oven and power is absorbed by the material, see
Eq. (6). This absorbed power generates heat inside the biomass. This
Table 2
heat is then transferred by conduction to the low temperature region of
Calculated λm , Dp , Nw , Np , fw and fp for the biomasses. Sphere radius R1 = 0.017
the sample. Only heat conduction in the sample was considered, and that
m, R2 = 0.031 m. Cylinder radius R1 = 0.0146 m, H = 0.031 m; R2 = 0.0373 m,
in the surrounding was neglected. Besides that, water evaporation of the H = 0.031 m. Humidity of 20% and 80%.
sample was also neglected due to the low water evaporation rate in the
Sugar cane bagasse
period before the phase transition temperature of the hot spot. Thus, the
electromagnetic energy is the heat source and its coupling with the λm (m) Dp (m) Nw Np fw fp
temperature inside the sample is given by: Sphere R1-20% 9.3 × 3.2 × 0.4 1.0 × 0.76 0.05
2
10− 10− 1 10− 1
∂T
ρC p = ∇.(k∇T) + Pab (10) Sphere R1-80% 7.1 × 1.3 × 0.5 2.6 × 0.58 0.09
2
∂t 10− 10− 1 10− 1
Sphere R2-20% 9.3 × 3.2 × 0.7 1.9 × 0.76 0.05
where ρ is the sample density, Cp is the specific heat capacity, k is the 10− 2
10− 1 10− 1
thermal conductivity of the biomass sample with temperature T and the Sphere R2-80% 7.1 × 1.3 × 0.9 4.8 × 0.58 0.09
2
power absorbed Pab is given by Eq. (6). 10− 10− 1 10− 1
Cylinder R1- 9.3 3.2 × 0.3 9.1 × 0.76 0.05
Initial values of electromagnetic field components were zero except
×
2
20% 10− 10− 1 10− 2
the initial value of z electric field component which is Ez = πy/Dg . Cylinder R1- 7.1 × 1.2 × 0.4 2.3 × 0.58 0.09
Initial ambient temperature in the simulations was T = 298 K in the 80% 10− 2
10− 1 10− 1
ambient and sample. Cylinder R2- 9.3 ×
2
3.2 × 0.8 2.3 × 0.76 0.05
20% 10− 10− 1 10− 1
Cylinder R2- 7.1 × 1.2 × 1.0 5.8 × 0.58 0.09
3.1. Boundary condition 80% 10− 2
10− 1 10− 1

Palm oil
As boundary condition to our problem, i.e. microwave oven, the λm (m) Dp (m) Nw Np fw fp
cavity walls and waveguide are supposed to have as a perfect electrical
conductor; boundary condition (n × E = 0) and a perfect magnetic Sphere R1- 4.2 × 4.1 × 0.8 0.8 0.34 0.16
2
20% 10− 10− 2
conductor (n × B = 0) are considered too. There is no heat flux across Sphere R1- 1.2 × 5.6 × 2.9 6.1 0.09 0.33
the boundary n.(k∇T) = 0 and the boundaries are thermally well insu­ 80% 10− 2
10− 3
lated. This condition state that the temperature in one side of the Sphere R2- 4.2 × 4.1 × 1.5 1.5 0.34 0.16
2
boundary is the same to the one on the other side. The cavity and the 20% 10− 10− 2
Sphere R2- 1.2 5.6 × 5.4 11.0 0.09 0.33
wave guide are considered as a metal wall resistance loss. This imped­
×
2
80% 10− 10− 3
ance boundary condition is known to penetrate only a short distance Cylinder R1- 4.2 × 4.1 × 0.7 0.7 0.34 0.16
outside the boundary and is given by: 20% 10− 2
10− 2
√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅ Cylinder R1- 1.1 × 5.6 × 2.5 5.2 0.09 0.33
μ0 μr 80% 10− 2
10− 3
n × H + E − (n.E)n = (n.Es )n − Es (11)
ε0 εr − j ωσ Cylinder R2- 4.2 × 4.1 × 1.8 1.8 0.34 0.16
2
20% 10− 10− 2
where Es is the electric field source. Cylinder R2- 1.1 × 5.6 × 6.5 13.3 0.09 0.33
2
80% 10− 10− 3
The rectangular port of the microwave oven is excited by a transverse
electric wave (TE10 mode) at an excitation frequency of 2.45 GHz. The Green Algae

port condition requires a constant β, which is given by:

λm (m) Dp (m) Nw Np fw fp

√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅ Sphere R1-20% 3.2 × 10.0 1.1 3.4 × 0.26 5.1 ×

2
2π c2 10− 10− 3 10− 4
β= f 2 − ( )2 (12) Sphere R1-80% 1.9 × 5.1 1.6 5.8 × 0.15 5.9 ×
c 4 Dg 2
10− 10− 3 10− 4

Sphere R2-20% 3.2 × 10.0 1.9 6.2 × 0.26 5.1 ×

where f is the excitation frequency, i.e. 2.45 GHz, and 4(Dc g ) is the cut-off 10− 2
10− 3 10− 4

Sphere R2-80% 1.9 5.1 3.3 1.2 × 0.15 5.9

frequency. The simulations were done at atmospheric pressure. ×
2
×
10− 10− 2 10− 4
The next section discusses the results obtained after simulation of Cylinder R1- 3.2 × 10.0 0.9 2.9 × 0.26 5.1 ×
electric field and the temperature distribution of sugar cane bagasse, 20% 10− 2
10− 3 10− 4

palm oil and green algae samples with different shapes, sizes, humidity Cylinder R1- 1.9 × 5.1 1.6 5.8 × 0.15 5.9 ×
2
and using a domestic microwave oven. 80% 10− 10− 3 10− 4

Cylinder R2- 3.2 × 10.0 2.3 7.5 × 0.26 5.1 ×

2
20% 10− 10− 3 10− 4

Cylinder R2- 1.9 × 5.1 4.0 1.5 × 0.15 5.9 ×

2
80% 10− 10− 2 10− 4

7
A.Z. Fia and J. Amorim Journal of the Energy Institute 106 (2023) 101146

biomass for conversion of lignocellulosic materials into renewable bio­ Fig. 6 shows the electric field distribution in the oven and in the
fuels. It may be noted that fp varies from 5.1 × 10− 4 to 3.3 × 10− 1 for the samples as function of geometry, for 30% and 65% of humidity. The
selected materials with different humidity. electric field distribution is directly affected by the sample’s geometry,
On the other hand, fw is found in the range 0 .09 < fw < 0.76 for all biomass type and humidity, as can be seen in Fig. 6. Thus, we can
the selected materials with the sugar cane presenting the highest fw = observe the non-uniformity of electric field distribution, showing re­
0.76 at 20% of humidity and palm oil at 80% the lowest fw = 0.09. For gions of high and low electric field intensity, e.g. in green algae with 65%
the materials with lowest fw ’s the microwave power absorption is ex­ of humidity, where the electric field intensity may be sixty times higher
pected to exhibit strong resonances, such as palm oil and algae, and for (5.9 × 104 V/m) than regions with low electric field intensities (0.098 ×
the one with highest magnitude of fw very weak resonances are ex­ 104 V/m) there are resonances which are not observed in other bio­
pected, e.g. sugar cane bagasse, see Figs. 6 and 7. masses. These factors impact directly in the formations of hot spots and
The penetration depth (Dp) is an extremely important information to in the evolution of non-uniform temperature distribution.
the understanding of electromagnetic heating, being used to design the According to Basak et al. [60], in addition to the classification of
size of microwave cavity and to investigate if the electromagnetic field power absorption regimes, the entire biological materials can be clas­
can uniformly heat a given sample. The Dp value is directly related to sified into four groups: (i) low fp , low fw (fp ≤ 0.1, fw ≤ 0.1), (ii) low fp ,
the dielectric property of the material being heated as well as the fre­ high fw ( fp ≤ 0.1, fw ≫0.1), (iii) high fp , low fw ( fp ≫0.1, fw ≤ 0.1) and
quency of the microwaves. For a homogeneous temperature distribution (iv) high fp , high fw ( fp ≫0.1, fw ≫0.1).
inside the material the penetration depth needs to be of the same order The sugar cane bagasse is an example of the group (ii) with low fp and
of dimension of the material to be heated. high fw , see Table 2, with a heating rate higher than the palm oil which
The humidity influence in Dp and tan δ are shown int Fig. 5. Once the belongs to the (iv) group, high fp , high fw . Green algae belong to group
dielectric constant of water is 78 at room temperature, materials with (ii) with low fp and high fw .
high water content present high values of ε’ reducing the penetration Fig. 7 shows the temperature distribution for green algae, palm oil
depth of microwave radiation. and sugar cane bagasse as function of geometry and heating time for
As can be seen in the figure above, as the humidity increases, the Dp 30% and 65% of humidity. The palm oil has symmetric temperature
decreases for green algae but presents high values, e.g. for 20% of hu­ distributions with respect to the center of the sphere and cylinder
midity, the Dp is 10 m! For sugar cane bagasse the penetration depth is samples. It can be noted the presence of resonances due to high fp with
0.13 m for 80% of humidity. The palm oil has the lowest Dp values, higher temperature in the left side for heating times higher than 30 s for
varying from 0.04 m to 0.006 m as humidity increases from 20% to 80%. 30% of humidity. In this situation, it can be seen in Fig. 2, relative
In the present study the Dp for palm oil has the same order of dimension permittivity ε′ and loss factor ε” have the lowest values. On the other
of materials used in the simulations. The penetration depth for sugar hand, for humidity of 65% no resonances are observed, where ε’ and ε”
cane bagasse may reach 30 times greater than samples used in simula­ have the highest values.
tions while for green algae Dp may have values of 700 times the di­ Palm oil has heating time that is comparable to the one of the sugar
mensions of samples simulated. We can expect high temperature cane bagasse. Formation of hot spots can be observed in the left side
homogeneity for palm oil. center of cylinder and sphere samples for times higher than 18 s and
The tan δ is representative of the energy absorbed by the samples. humidity of 30%. Hot spots are observed for sugar cane bagasse for
Higher values indicate more energy absorption and material heating. In humidity of 65% and time of 12 s on the left side of cylinder and from 9 s
general, the biomasses have low tan δ and for an efficient heating in in the sphere in the bottom of the sample.
industrial applications, addition of materials with high tan δ in the The green algae with low fp and high fw presents resonances for the
samples to be treated is desirable. As can be seen in Fig. 5 the green algae electric field and temperature and humidities of 30% and 65%, respec­
have tan δ ≅ 0.001, value that indicates the material is transparent to tively. However, the heating times are considerably greater than the
microwaves, with need of more time for the material heating. Palm oil is ones of palm oil and sugar cane for the appearance of resonances, see
the biomass with the highest 0.3 < tan δ < 0.7 and so is more efficient Fig. 7, from 260 s for cylinder with 30% of humidity and from 590 s for
for heating. The sugar cane bagasse has tan δ < 0.2 and high values of 65%. For sphere from 270 s to 30% of humidity and from 15 s for hu­
Dp which favors an inhomogeneous heating and formation of hot spots. midity of 65%. This very different behavior may be attributed to the
variation of ε’ and ε” with humidity. Among the biomasses studied in
this work the green algae have the lowest loss factor.
It can be seen in Fig. 8 the time for the sample to achieve 373 K
decreases when power increases for all biomasses as could be expected.
The green algae always presents a longest decay time scale for heating
and stabilizes after 400 W. The cylinder shape of green algae has longer
decay time than sphere. Sugar cane and palm oil have similar decay time
independent of the geometrical configuration and attains a plateau after
200 W.
It can be seen in Fig. 9 the average temperature (T) and absorbed
energy as function of power for cylinder (A) and sphere (B) when the
volume of both shapes is 40.77 × 10− 5 m3 and humidity of 65%. The
average temperature (T) and absorbed energy have the same trend when
power is varied. The green algae has the highest average temperature
and absorbed energy for low treatment power, before 100 W, and de­
creases when power increases. It should be emphasized that the heating
time to attain 373 K decreases with the increase of power which explain
the decrease of T and absorbed energy with power. The shape of the
samples influences mainly for the sugar cane. In sphere geometry it has
higher T than cylindrical but the absorbed energy is low. In fact, it is the
Fig. 5. Penetration depth (Dp, continuous lines) and tan δ (dashed lines) as biomass with the lowest absorbed energy independent of the shape. The
function of humidity in the samples. Sugar cane bagasse ( ), palm oil ( ) and
palm oil has constant absorbed energy as function of power and presents
green algae ( ).

8
A.Z. Fia and J. Amorim Journal of the Energy Institute 106 (2023) 101146

5
Fig. 6. Electric field distribution for green algae, palm oil and sugar cane bagasse. Cylinder and sphere, volume of 100 × 10− m3 with humidity’s of 30% and 65%.

values close to the ones of the sugar cane bagasse. the COV until ~700 W for green algae. For the spherical shape ΔT for
The ratio of internal temperature standard deviation (SD) and its green algae is lower than other biomasses until 50 W of power. For 120
average value, namely, coefficient of variation (COV) is a measure of W the degree of thermal runaway for green algae reaches the value of
temperature uniformity inside the sample. As the standard deviation is palm oil at 75 K. The COV in spheric geometry is the same of sugar cane
expected to increase with the mean it is not an appropriate statistic. bagasse at power of 60 W and presents the highest value after 280 W. It
Considering only the standard deviation is appropriate only if the means should be emphasized that for a good uniformity of temperature inside
are equal. Coefficient of variation is an useful parameter to remove the the samples low COV is desired. Sugar cane bagasse and palm oil have a
effect of the mean variation, being more effective measure of tempera­ constant ΔT and COV after 100 W in both geometries. Sugar cane
ture non-uniformity. Higher COV values would signify a more non- bagasse in the spherical shape has low COV value of 0.08.
uniformity of heating. The COV is given by Ref. [56]: The capability of a given sample to absorb the microwave energy is
√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅ expressed by the ratio of absorbed power (Pab ) and input power (Pin )
√
√1 ∑ N
called power absorption efficiency (PAE):
SD = √ (Ti − T(t))2 (13)
N i=1 Pab
PAE = (15)
Pin
COV = SD/T(t) (14)
The comprehensive evaluation coefficient (CEC) is the ratio of power
absorption efficiency (PAE) to temperature COV value. The CEC value
where N is the number of points in the data set, which along this work can measure the samples’ usability at a certain condition once higher
was considered 200, Ti is the temperature in each instant of time and CEC value presents a sample with a high PAE and/or low COV, which
T(t) is the mean temperature of the object under study. could be chosen according to specific needs of a microwave process.
The degree of thermal runaway, that the difference of the maximum The comprehensive evaluation coefficient (CEC) as function of vol­
and minimum temperature inside the sample, has the same behavior of ume for cylinder and sphere shapes are shown in Fig. 11. In industry an
COV when the power is varied as can be seen in Fig. 10. Both parameters ideal object under treatment should have high power absorption effi­
are important and revel the thermal inhomogeneity inside the sample ciency and low value of coefficient of variation assuring high power
during the heating. Green algae presents different behavior than sugar absorption and uniform temperature distribution inside the sample.
cane bagasse, and palm oil as the power is varied. For low power, cyl­ Thus, high values of comprehensive evaluation coefficient, are desired
inder geometry, the ΔT is always lower than other biomasses as well as

9
A.Z. Fia and J. Amorim Journal of the Energy Institute 106 (2023) 101146

5
Fig. 7. Temperature distribution for green algae, palm oil and sugar cane bagasse. Cylinder and sphere, volume of 100 × 10− m3 with humidity’s of 30% and 65%.

indicating the sample usability in each application. Sugar cane bagasse, humidity but with high COV for cylindrical geometry. The palm oil has
cylinder, and sphere geometry, presents a non-uniform behavior as low absorbed energy and high COV independently of geometry and
volume of samples is varied. For cylinder there is an increase of CEC humidity.
from volume 1 × 10- 5 m3 to 2 × 10− 5 m3, decreases at 3 × 10− 5 m3 and The absorbed energy by the samples in cylindrical and spherical
then increases until 5 × 10− 5 m3. In sphere geometry it increases as shapes, as function of power is shown in Fig. 13 for the three volumes of
function of volume. For palm oil, cylinder geometry, is roughly constant sugar cane bagasse, palm oil, and green algae for humidity of 65%. It can
while for sphere, increases initially and then remains constant with be seen in both geometries the green algae absorbs better the energy for
volume variation. For green algae the CEC presents the lowest value volume 3 and 4 for cylinders and spheres, and for volume 1 the absorbed
among other biomasses but increases when the volume increases inde­ energy is lower than the one absorbed by sugar cane and palm oil in both
pendently of the shape chosen. geometries, what is expected for low powers in case (B). The maximum
When the absorbed energy and COV are plotted against moisture the absorbed energy is lower than 6 × 103 J at 15 W for volume 1 while for
result is Fig. 12. As can be seen in Fig. 12 (A), cylindrical shape, the volume 3 and 4 the energy is around 105 J.
green algae has the highest absorption of energy but with high values of Sugar cane bagasse and palm oil present roughly a constant ab­
COVs, indicating the high inhomogeneity of temperature distribution sorption profile for both geometries and volumes, except at powers
inside the samples, see Fig. 7. When the sample is the sphere the COV of lower than 100 W. For the cylindrical geometry the energy absorbed by
green algae decreases until 50% of humidity and increases after this palm oil is two times higher than energy absorbed by sugar cane bagasse
value while the absorbed energy presents the opposite behavior. For the and for spherical geometry they are roughly the same for volumes 3
cylindrical geometry the COV increases until 50% and then is almost (palm oil) and 4 (sugar cane bagasse). The palm oil, volume 4 is the
constant until 65%. The absorbed energy decreases sharply until 50% biomass with the highest absorbed energy, expected for power lower
and more slowly until 65%. The lowest value of absorbed energy is than 100 W in spherical geometry.
obtained for the sugar cane bagasse independently of geometry or Fig. 14 shows the absorbed power as function of volume and surface

10
A.Z. Fia and J. Amorim Journal of the Energy Institute 106 (2023) 101146

Fig. 8. Time to achieve 373 K as function of power. Cylinder and sphere with
volume of 40.77 10− 5 m3, humidity of 65%. Sugar cane bagasse (cylinder ,
sphere ), palm oil (cylinder , sphere ) and green algae (cylinder ,
sphere ).

Fig. 10. Degree of thermal runaway (ΔT) (continuous lines) and COV (dashed
lines) as function of power for cylinder (A) and sphere (B). Volume of 40.77
10− 5 m3, humidity of 65%. Sugar cane bagasse ( ), palm oil ( ) and green
algae ( ).

over volume ratio (S/V). Palm oil presents the highest absorbed power in
the two cases while green algae the lowest. The power absorbed by the
samples increases when sample volume increases due to more mass
disponible for absorption of microwave radiation. When the ratio S/V is
varied, the decrease of the absorption of power by the samples is
observed, putting in evidence the higher importance of volume increase
than the surface of the samples.

5. Conclusions

Using a model for the kitchen oven, simulations were done to study
the heating of sugar cane bagasse, palm oil and green algae as function
of temperature, humidity, sample volumes and shapes to find the best
conditions for microwave pretreatment of biomasses for conversion of
lignocellulosic materials to obtain renewable biofuels. It is important to
mention that this present work is a simulation of heating of sugar cane
bagasse, African palm, and green algae. The Comsol Multiphysics plat­
form was already validated by comparison of numerical simulations and
Fig. 9. Average temperature (T) (solid curves) and absorbed energy (dashed experiments to predict power and temperature in agar gel, and in
curves) as function of power for cylinder (A) and sphere (B). Volume of 40.77 hamburger with different fat contents [32] and in the determination of
10− 5 m3, humidity of 65%. Sugar cane bagasse ( ), palm oil ( ) and green temperature variation as function of heating time of empty fruit bunch
algae ( ). biomass pellets [49].
The choice of microwave power input is crucial to find the situations

11
A.Z. Fia and J. Amorim Journal of the Energy Institute 106 (2023) 101146

heating condition.
Green algae has tan δ ≅ 0.001, value that indicates the material is
transparent to microwaves, with need of more time for the material
heating. Palm oil is the biomass with the highest tan δ
(0.3 < tan δ < 0.7) and higher efficiency for heating while sugar cane
bagasse with tan δ < 0.2 and high values of Dp favors an inhomogeneous
heating and formation of hot spots.
The electric field distribution is directly affected by the sample’s
geometry, biomass type and humidity. The non-uniformity of electric
field distribution results in regions of high and low electric field in­
tensity. e.g. in green algae with 65% of humidity, where the electric field
intensity may be sixty times higher than regions with low electric field
intensities, there are resonances which are not observed in other bio­
masses. These factors impact directly in the formations of hot spots and
in the evolution of non-uniform temperature distribution.
This work has shown the temperature distribution for green algae,
palm oil and sugar cane bagasse as function of geometry and heating
time for 30% and 65% of humidity. The palm oil presents symmetric
temperature distributions with respect to the center of the sphere and
cylinder samples. Resonances was founded and attributed to high fp with
higher temperature in the left side for heating times higher than 30 s for
Fig. 11. Comprehensive evaluation coefficient (CEC) as function of volume, for 30% of humidity.
cylinder, sphere. Volume of 40.77 10− 5 m3, humidity of 65% and power of 400 Palm oil has heating time is comparable to the one of the sugar cane
W. Sugar cane bagasse (cylinder , sphere ), palm oil (cylinder , sphere ) bagasse. Formation of hot spots can be observed in the left side center of
and green algae (cylinder , sphere ). cylinder and sphere samples for times higher than 18 s and humidity of
30%. Hot spots are observed for sugar cane bagasse for humidity of 65%.
of high temperature distribution and uniformity with low energy con­ The green algae presents resonances for the electric field and for tem­
sumption. For a homogeneous temperature distribution inside the ma­ perature and humidities of 30% and 65%, respectively. Heating times
terial the penetration depth needs to be of the same order of dimension are considerably greater than the ones of palm oil and sugar cane.
of the material to be heated. Materials with high ε’ and ε” will have the Among the biomasses studied in this work the green algae have the
lower penetration depth (Dp) than material with low dielectric constant lowest loss factor.
and loss factor. The energy absorbed by the samples which is repre­ The time for samples to achieve 373 K, decreases when power in­
sented by the tan δ parameter is another important factor. Higher values creases for all biomasses. The green algae always presents a longest time
indicate more energy absorption and material heating. For green algae scale for heating. The cylinder shape of green algae has longer decrease
with 20% of humidity the Dp is 10 m! For sugar cane bagasse the than sphere. Sugar cane and palm oil have similar decrease time, inde­
penetration depth is 0.13 m and for palm oil it varies from 0.04 m to pendently of the geometrical configuration.
0.006 m as humidity increases from 20% to 80% having the same order The average temperature (T) and energy absorbed have the same
of dimension of materials used in the simulations and presenting the best trend when power is varied. Green algae has the highest average tem

5
Fig. 12. Absorbed energy (continuous lines) and coefficient of variation (dashed lines) as function of moisture for cylinder (A) and sphere (B). Volume of 40.77 10−
m3, power of 400 W. Sugar cane bagasse ( ), palm oil ( ) and green algae ( ).

12
A.Z. Fia and J. Amorim Journal of the Energy Institute 106 (2023) 101146

Fig. 13. Absorbed energy as function of power for cylinders (A) and spheres
(B). Volumes of 2 × 10− 5 m3 (1), 13.11 × 10− 5 m3 (3), and 40.77 10− 5 m3 (4). Fig. 14. Absorbed power as function of volume (A) and surface over volume
Humidity of 65%. Sugar cane bagasse volume 1( ), volume 3 ( ), volume 4 ( ); ratio (B). Sugar cane bagasse ( ), palm oil ( ) and green algae ( ).
Palm oil volume 1( ), volume 3 ( ), volume 4 ( ); Green Algae volume 1( ),
volume 3 ( ), volume 4 ( ). algae absorbs better the energy for volume 3 and 2 for cylinders.
The absorbed power may vary as function of volume and surface over
perature and absorbed energy for low treatment power what decreases volume ratio. Palm oil presents the highest absorbed power in the two
when power increases. The palm oil has constant absorbed energy as cases while green algae the lowest. The power absorbed by the samples
function of power and presents values close to the ones of the sugar cane increases when sample volume increases due to more mass disponible
bagasse. for absorption of microwave radiation and decreases as function of the
The degree of thermal runaway presents the same behavior of COV surface over volume ratio.
parameter when the power is varied. Green algae presents different
behavior from sugar cane bagasse, and palm oil as the power is varied. Declaration of competing interest
The comprehensive evaluation coefficient as function of volume in­
dicates that sugar cane bagasse, cylinder and sphere geometry, presents The authors declare that they have no known competing financial
quite different behavior from other biomasses. For palm oil and green interests or personal relationships that could have appeared to influence
algae, the CEC increases when the volume increases independently of the work reported in this paper.
the shape chosen.
The lowest value of absorbed energy is obtained for the sugar cane Acknowledgements
bagasse, independently of geometry or humidity, but high COV values is
observed for cylindrical geometry. When the sample is the sphere the A. Z. Fia would like to thank the CAPES for the financial support,
COV of green algae decreases until 50% of humidity and increases after Proc. Number 88882.446976/209-01. J. Amorim thanks CNPq, Grant
this value while the absorbed energy presents the opposite behavior. The 305904/2019-1. Authors are in debit with Laboratório de Guerra
palm oil has low absorbed energy and high COV independently of ge­ Eletrônica - ITA, for the use of computational facilities.
ometry and humidity.
The absorbed energy by the samples, in cylindrical and spherical
shapes, as function of power, indicates that in both geometries the green

13
A.Z. Fia and J. Amorim Journal of the Energy Institute 106 (2023) 101146

References [29] A.T. Hoang, S. Nižetić, H.C. Ong, M. Mofijur, S.F. Ahmed, B. Ashok, V.T.V. Bui, M.
Q. Chau, Insight into the recent advances of microwave pretreatment technologies
for the conversion of lignocellulosic biomass into sustainable biofuel, Chemosphere
[1] S. Jin, COVID-19, climate change, and renewable energy research: we are all in this
281 (2021), 130878.
together, and the time to act is now, ACS Energy Lett. 5 (2020) 1709–1711.
[30] G. Cravotto, L. Boffa, S. Mantegna, P. Perego, M. Avogadro, P. Cintas, Improved
[2] IPCC Global Warming of 1.5 ◦ C, An IPCC Special Report on the impacts of global
extraction of vegetable oils under high-intensity ultrasound and/or microwaves,
warming of 1.5 ◦ C above preindustrial levels and related global greenhouse gas
Ultrason. Sonochem. 15 (2008) 898–902.
emission pathways, in: The Context of Strengthening the Global Response to the
[31] S. Horikoshi, A. Osawa, S. Sakamoto, N. Serpone, Control of microwave generated
Threat of Climate Change, Sustainable Development, and Efforts to Eradicate
hot spots. Part IV. Control of hot spots on a heterogeneous microwave-absorber
Poverty, 2018. www.ipcc.ch/sr15. (Accessed 18 August 2022). accessed.
catalyst surface by a hybrid internal/external heating method, Chem. Eng. Process
[3] K. Ullah, M. Ahmad, Sofia, V.K. Sharma, P. Lu, A. Harvey, M. Zafar, S. Sultana,
Intensif 69 (2013) 52–56.
Assessing the potential of algal biomass opportunities for bioenergy industry: a
[32] L.A. Campañone, C.A. Paola, R.H. Mascheroni, Modeling and simulation of
review, Fuel 143 (2015) 414–423.
microwave heating of foods under different process schedules, Food Bioprocess
[4] H.S. Kheshgi, C.P. Roger, G. Marland, The potential of biomass fuels in the context
Technol. 5 (2010) 738–749.
of global climate change: focus on transportation fuels, Annu. Rev. Energy Environ.
[33] A.A. Arpia, W.H. Chen, S.S. Lam, P. Rousset, M.D.G. Luna, Sustainable biofuel and
25 (2000) 199–244.
bioenergy production from biomass waste residues using microwave-assisted
[5] www.ucsusa.org/resources/car-emissions-global-warming. (Accessed 25 March
heating: a comprehensive review, Chem. Eng. J. 403 (2021) 1385–8947.
2021) accessed.
[34] F. Motasemi, M.T. Afzal, A review on the microwave-assisted pyrolysis technique,
[6] CO2 Emissions from Cars: Facts and Figures (Infographics) | News | European
Renew. Sustain. Energy Rev. 28 (2013) 317–330.
Parliament (europa.eu) (accessed 2022-08-18).
[35] C.S. Gonzáles, M.F.S.M. Gonzáles, A.L. Malo, M.E. Morales, Recent studies related
[7] P. McKendry, Energy production from biomass (part 1): overview of biomass,
to microwave processing of fluid foods, Food Bioprocess Technol. 5 (2012) 31–46.
Bioresour. Technol. 83 (2002) 37–46.
[36] P. Kumar, P. Coronel, V.D. Truong, J. Simunovic, K.R. Swartzel, K.P. Sandeep,
[8] S. Proskurina, R. Sikkema, J. Heinimö, E. Vakkilainen, Five years left–How are the
G. Cartwright, Overcoming issues associated with the scale-up of a continuous flow
EU member states contributing to the 20% target for EU’s renewable energy
microwave system for aseptic processing of vegetable purees, Food Res. Int. 41
consumption: the role of woody biomass, Biomass Bioenergy 95 (2016) 64–77.
(2008) 454–461.
[9] M.L. Lopes, S.C.L. Paulillo, A. Godoy, R.A. Cherubin, M.S. Lorenzi, F.H.C. Giometti,
[37] J. Li, W. Zhou, Y. Su, S. Wei, Y. Zhao, L. Zhang, Y. Ding, L. Xie, F. Sun, J. Gao,
C.D. Bernardino, H.B. Amorim Neto, H.V. Amorim, Ethanol production in Brazil: a
P. Wang, G. Zhao, Y. Qin, Experimental and numerical studies on the heating
bridge between science and industry, Braz. J. Microbiol. 47S (2016) 64–76.
mechanism of millimeter multi-particle system under microwave irradiation, J.
[10] P.S. Nigam, A. Singh, Production of liquid biofuels from renewable resources, Prog.
Energy Inst. 102 (2022) 216–228.
Energy Combust. Sci. 37 (2011) 52–68.
[38] M.N. Berteli, E. Rodier, A. Marsaioli Jr., Study of the microwave vacuum drying
[11] H.S. Kheshgi, R.C. Prince, G. Marland, The potential of biomass fuels in the context
process for a granulated product, Braz. J. Chem. Eng. 26 (2009) 317–329.
of global climate change: focus on Transportation Fuels, Annu. Rev. Energy
[39] K.G. Ayappa, K. Ayappa, Modelling transport processes during microwave heating:
Environ. 25 (2000) 199–244.
a review, Rev. Chem. Eng. 13 (1997) 1–69.
[12] K. Ullah, V.K. Sharma, M. Ahmad, P. Lv, J. Krahl, Z. Wang, Sofia, the insight views
[40] S. Tsubaki, M. Hiraoka, S. Hadano, H. Nishimura, K. Kashimura, T. Mitani,
of advanced technologies and its application in bio-origin fuel synthesis from
Functional group dependent dielectric properties of sulfated hydrocolloids
lignocellulose biomasses waste, a review, Renew. Sustain. Energy Rev. 82 (2018)
extracted from green macroalgal biomass, Carbohydr. Polym. 107 (2014) 192–197.
3992–4008.
[41] B.D. Roebuck, S.A. Goldblith, W.B. Westphal, Dielectric properties of
[13] F.A. Gonçalves, H.A. Ruiz, E.S. dos Santos, J.A. Teixeira, G.R. de Macedo,
carbohydrate-water mixtures at microwave frequencies, J. Food Sci. 37 (1972)
Bioethanol production by Saccharomyces cerevisiae, Pichia stipitis and
199–204.
Zymomonas mobilis from delignified coconut fibre mature and lignin extraction
[42] S. Ryynänen, The electromagnetic properties of food materials: a review of the
according to biorefinery concept, Renew. Energy 94 (2016) 352–365.
basic principles, J. Food Eng. 26 (1995) 409–429.
[14] M.B. Ariede, T.M. Candido, A.L.M. Jacome, M.V.R. Velascoa, J.C.M. de Carvalho,
[43] C. Ellison, M.S. McKeown, S. Trabelsi, C. Marculescu, D. Boldor, Dielectric
A.R. Baby, Cosmetic attributes of algae - a Review, Algal Res. 25 (2017) 483–487.
characterization of bentonite clay at various moisture contents and with mixtures
[15] X. Liu, K. Ying, G. Chen, C. Zhou, W. Zhang, X. Zhang, Z. Cai, T. Holmes, Y. Tao,
of biomass in the microwave spectrum, J. Microw. Power Electromagn. Energy 52
Growth of Chlorella vulgaris and nutrient removal in the wastewater in response to
(2018) 3–15.
intermittent carbon dioxide, Chemosphere 186 (2017) 977–985.
[44] Z. Abbas, Y.K. Yeow, A.H. Shaari, K. Khalid, J. Hassan, E. Saion, Complex
[16] A. Kumar, Y. Singh Negi, V. Choudhary, N.K. Bhardwaj, Characterization of
permittivity and moisture measurements of oil palm fruits using an open-ended
cellulose nanocrystals produced by acid-hydrolysis from sugarcane bagasse as
coaxial sensor, IEEE Sensor. J. 5 (2005) 1281–1287.
agro-waste "characterization of cellulose nanocrystals produced by acid-hydrolysis
[45] A.B. Itolikar, M.L. Kurtadikar, Microwave dielectric properties of algae, Bionano
from sugarcane bagasse as agro-waste, J. Mater. Phys. Chem. 2 (2014) 1–8.
Front. 8 (2015) 371–374.
[17] S. Anto, S.S. Mukherjee, R. Muthappa, T. Mathimani, G. Deviram, S.S. Kumar, T.
[46] V.L. Budarin, J.H. Clark, B. Lanigan, P. Shuttleworth, D.J. Macquarrie, Microwave
N. Verma, A. Pugazhendhi, Algae as green energy reserve: technological outlook on
assisted decomposition of cellulose: a new thermochemical route for biomass
biofuel production, Chemosphere 242 (2020), 125079.
exploitation, Bioresour. Technol. 101 (2010) 3776–3779.
[18] D. Behrendt, C. Schreiber, C. Pfaff, A. Müller, J. Grobbelaar, L. Nedbal, Algae as a
[47] J. He, Y. Yang, H. Zhu, K. Li, W. Yao, K. Huang, Microwave heating based on two
potential source of biokerosene and diesel – opportunities and challenges,
rotary waveguides to improve efficiency and uniformity by gradient descent
Biokerosene (2017) 303–324.
method, Appl. Therm. Eng. 178 (2020) 1359–4311.
[19] B. Abdullah, A.F.S. Muhammad, Z. Shokravi, S. Ismail, K.A. Kassim, A.
[48] X. Gao, X. Liu, P. Yan, X. Li, H. Li, Numerical analysis, and optimization of the
N. Mahmood, Md M. Aziz, Fourth generation biofuel: a review on risks and
microwave inductive heating performance of water film, Int. J. Heat Mass Tran.
mitigation strategies, Renew. Sustain. Energy Rev. 107 (2019) 37–50.
139 (2019) 17–30.
[20] F.A. Gonçalves, H.A. Ruiz, E.S. dos Santos, J.A. Teixeira, G.R. de Macedo,
[49] A.A. Salema, M.T. Afazal, Numerical simulation and heating behavior in biomass
Bioethanol production from coconuts and cactus pretreated by autohydrolysis, Ind.
bed and pellets under multimode microwave system, Int. J. Therm. Sci. 91 (2015)
Crop. Prod. 77 (2015) 1–12.
12–24.
[21] D.C. Nieves, H.A. Ruiz, L.Z. de Cárdenas, G.M. Alvarez, C.N. Aguilar, A. Ilyna, J.L.
[50] Z. Zhang, T. Su, S. Zhang, Shape effect on the temperature field during microwave
M. Hernández, Enzymatic hydrolysis of chemically pretreated mango stem bark
heating process, J. Food Qual. (2018) 1–24, 2018.
residues at high solid loading, Ind. Crop. Prod. 83 (2016) 500–508.
[51] F. Casciatori, P.C. Laurentino, L.K. Lopes, C.M.A. de Souza, G.J. Thoméo, Stagnant
[22] A. Aguilar-Reynosa, A. Romaní, R. Ma Rodríguez-Jasso, C.N. Aguilar, G. Garrote,
effective thermal conductivity of agro-industrial residues for solid-state
H.A. Ruiz, Microwave heating processing as alternative of pretreatment in second
fermentation, Int. J. Food Prop. 16 (2013) 1578–1593.
generation biorefinery: an overview, Energy Convers. Manag. 136 (2017) 50–65.
[52] R. Omar, A. Idris, R. Yunus, K. Khalid, M. Aida Isma, Characterization of empty
[23] R.V. Kapoore, T.O. Butler, J. Pandhal, S. Vaidyanathan, Microwave-assisted
fruit bunch for microwave-assisted pyrolysis, Fuel 90 (2011) 1536–1544.
extraction for microalgae: from biofuels to biorefinery, Biology 7 (2018) 1–25.
[53] T. Murphy, E.H. Berberoğlu, Temperature Fluctuation and Evaporative loss rate in
[24] J. Velazquez-lucio, R.M. Rodríguez-Jasso, L.M. Colla, A. Saenz-galindo, D.
an algae biofilm photobioreactor, J. Sol. Energy Eng. 134 (2012) 011002-
E. Cervantes, C.N. Aguilar, B.D. Fernandes, H.A. Ruiz, Microalgal biomass
1–011002-9.
pretreatment for bioethanol production: a review, Biofuel Res. J. 17 (2018)
[54] P. Britton, E. Jen The, D. Close, Investigation of the efficiency of bagasse as a
780–791.
thermal insulator, Proc. Aust. Soc. Sugar Cane Technol. 27 (2005) 462–471.
[25] R. Ravindran, A.K. Jaiswal, A comprehensive review on pre-treatment strategy for
[55] Z. Liyana, F. Malek, E. Cheng, M.W. Liu, W.Y. Lee, Measurement of dielectric
lignocellulosic food industry waste: challenges and opportunities, Bioresour.
properties of sugarcane bagasse and rubber tire dust for microwave absorber
Technol. 199 (2016) 92–102.
performance, ARPN J. Eng. and Appl. Sci. 11 (2016) 10877–10881.
[26] J. Loureiro, J. Amorim, Kinetics and Spectroscopy of Low Temperature Plasmas,
[56] A.Z. Fia, J. Amorim, Heating of biomass in microwave household oven - a
Springer International Publishing, Switzerland, 2016.
numerical study, Energy 218 (2021), 119472.
[27] J.A. Souza-Corrêa, C. Oliveira, L.D. Wolf, V.M.G. Nascimento, J.M. Rocha,
[57] H. Gothall, How to Inspect Your Mesh in COMSOL Multiphysics, COMSOL Blog,
J. Amorim, Atmospheric pressure plasma pretreatment of sugarcane bagasse: the
2017. https://www.comsol.com/blogs/how-to-inspect-your-mesh-in-comsol-multi
influence of moisture in the ozonation process, Appl. Biochem. Biotechnol. 171
physics. (Accessed 24 November 2020).
(2013) 104–116.
[28] J.A. Souza-Corrêa, C. Oliveira, L.D. Wolf, V.M.G. Nascimento, J.M. Rocha,
J. Amorim, Atmospheric pressure plasma pretreatment of sugarcane bagasse: the
influence of particle size in the ozonation process, Appl. Biochem. Biotechnol. 171
(2014) 1663–1672.

14
A.Z. Fia and J. Amorim Journal of the Energy Institute 106 (2023) 101146

[58] Yi-du Hong, Bai-quan Lin, Li He, Hua-ming Dai, Chuan-jie Zhu, H. Yao, Three- [60] T. Basak, M. Bhattacharya, S. Panda, A generalized approach on microwave
dimensional simulation of microwave heating coal sample with varying processing for the lateral and radial irradiations of various groups of food
parameters, Appl. Therm. Eng. 93 (2016) 1145–1154. materials, Food Sci. & Emerging Technol. 33 (2016) 333–347.
[59] M. Bhattacharya, T. Basak, S. Sriram, Generalized characterization of microwave
power absorption for processing of circular shaped materials, Chem. Eng. Sci. 118
(2014) 257–279.

15