Available online at [Link].

com

ScienceDirect
Availableonline
Available [Link]
[Link]
Energy Procedia 00 (2017) 000–000

ScienceDirect
ScienceDirect
[Link]/locate/procedia

Energy
EnergyProcedia 142
Procedia 00(2017)
(2017)1699–1705
000–000
[Link]/locate/procedia

9th International Conference on Applied Energy, ICAE2017, 21-24 August 2017, Cardiff, UK

Thermogravimetric Kinetics and High Fidelity Analysis of Crude

The 15th International Symposium on District Heating and Cooling
Glycerol
Assessing the feasibility of using the heat demand-outdoor
Manar Almazroueia, Tala El Samada, Isam Janajreha*
temperature functionMechanical
for aandlong-term district
Materials Engineering Department heat demand forecast
Khalifa University of Science and Technology, Masdar Institute
a,b,c
I. Andrić *, A. Pina , P. Ferrão , J. Fournier ., B. Lacarrièrec, O. Le Correc
aMasdar City, AbuaDhabi, UAE, [Link]
*
ijanajreh@[Link]
b 54224

a
IN+ Center for Innovation, Technology and Policy Research - Instituto Superior Técnico, Av. Rovisco Pais 1, 1049-001 Lisbon, Portugal
b
Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, 78520 Limay, France
Abstract c
Département Systèmes Énergétiques et Environnement - IMT Atlantique, 4 rue Alfred Kastler, 44300 Nantes, France

This work undertakes the determination of the glycerol thermochemical conversion kinetics in one hand, and the utilize of these
quantities in high fidelity reactive flow model on the other. The kinetic parameters of glycerol are determined by using
thermogravimetric
Abstract analyzer (TGA) in the temperature range of 50-700 ͦC. Due to the distinct behavior of the crude glycerol, the
TG curves of the thermal decomposition process were divided into several events, moisture release and two-subsequent
devolatalization
District heatingand gasification
networks phases. The
are commonly kinetic parameters,
addressed are calculated
in the literature as one ofusing four effective
the most different methods:
solutions Arrhenius, Coats-
for decreasing the
Redfern, Ingraham Marrier and Differential Method. Results of the kinetics study were nearly consistent
greenhouse gas emissions from the building sector. These systems require high investments which are returned through the heat within the individual
kinetic
sales. model
Due tobut discrepancy
thethechanged was conditions
climate more obvious andamongst
buildingthe different models.
renovation policies,Theheatresulted
demand Coat-Redfirm
in the future kinetics
could were the
decrease,
best fit amongst
prolonging the other four considered
investment methods demonstrated by its highest correlation R2. It resulted with activation energy of 62
return period.
kJ/mol
The mainand scope
138 kJ/mol
of thisand pre-exponential
paper is to assess thefactor of 4.3e5
feasibility using and
of min
-1 2.25e9
the heat min-1 –for
demand phase temperature
outdoor 2 and phase function
3, respectively.
for heat A high
demand
fidelity
forecast. The district of Alvalade, located in Lisbon (Portugal), was used as a case study. The district is consisted of by
CFD model is developed to assess the performance of glycerol gasification in a tubular reactor. The model is governed 665
the Navier-stokes
buildings equations
that vary in bothcoupled to species
construction transport
period equation for
and typology. reactive
Three flowscenarios
weather under non-isothermal
(low, medium, conditions.
high) andResults
three of the
district
syngas conversion
renovation are in
scenarios agreement
were developedwith(shallow,
relevant literature and theoretical
intermediate, deep). To molar yield
estimate theoferror,
0.56 and 0.42 mole
obtained fractions values
heat demand and
for H2 were
CO, respectively.
compared The developed
with results reactive
from a dynamic heatflow model
demand has brought
model, more
previously insight into
developed and the glycerol
validated by gasification,
the authors. enabling further
parametric
The resultsstudies
showedinthat the when
pursuit of weather
only the optimal
changereactor design configuration
is considered, the margin ofanderroroperating
could beconditions
acceptableforfor potential process
some applications
integration.
(the error in annual demand was lower than 20% for all weather scenarios considered). However, after introducing renovation
scenarios, the error value increased up to 59.5% (depending on the weather and renovation scenarios combination considered).
The value of slope coefficient increased on average within the range of 3.8% up to 8% per decade, that corresponds to the
©decrease
2017 The inAuthors.
the number Published by Elsevier
of heating hours ofLtd.
22-139h during the heating season (depending on the combination of weather and
Peer-review
renovation under
scenariosresponsibility
considered).of the
On scientific
the other committee of theintercept
hand, function 9th International
increasedConference
for 7.8-12.7%on Applied Energy.
per decade (depending on the
coupled scenarios). The values suggested could be used to modify the function parameters for the scenarios considered, and
Keywords:
improve Glycerol,
the accuracyPyrolysis, TGA
of heat analysis,
demand Activation Energy, Arrhenius equation, Kinetic parametrs, CFD simulation.
estimations.

© 2017 The Authors. Published by Elsevier Ltd.

Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and
Cooling.

Keywords: Heat demand; Forecast; Climate change

1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the scientific committee of the 9th International Conference on Applied Energy.

1876-6102 © 2017 The Authors. Published by Elsevier Ltd.

Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the scientific committee of the 9th International Conference on Applied Energy .
10.1016/[Link].2017.12.552
1700 Manar Almazrouei et al. / Energy Procedia 142 (2017) 1699–1705
2 Almazrouei, Elsamad and Janajreh* / Energy Procedia 00 (2017) 000–000

1. Introduction

The design of glycerol gasifier requires an in-depth understanding of the gasification conversion process which
depends on the overall kinetics and species distributions. Glycerol is the main byproduct of transesterification process
which is comonly used for biodiesel production. It mounts nearly 10% of the process yield. Fig. 1 presents the
simplified transesterification process of triglyceride reaction to produce biodiesel and the byproduct glycerol, it also
dipicts the recently developed continuous transesterfication reactor [1,2]. The staggering amount of crude glycerol
continue to add an environmental stress as waste disposal which madates proper treatement to negate their negative
impact [3,4].

Fig. 1. Transesterification reaction of triglyceride and the geometry of the continues reactor [1,2].
Glycerol is a viscous organic liquid of simple sugar alcohol which is colorless, odorless, with low toxicity index. It
has a chemical formula of C3H8O3 and is synonymous to glycerin, propane-1, 2,3-triol,1,2,3-propanetriol,1,2,3-
trihydroxypropane,glyceritol, and glycol alcohol. Glycerol starts to decompose at around 250 oC and requires 6.3
KJ/mol as activation energy to break the hydrogen bonds. It catches attention due to its relatively high heat of
combustion (18MJ/kg) which makes it a viable source of energy and fuel [5]. Thermo-chemical conversion is the most
suitable method for glycerol utilization, particularly when pure glycerol is in surplus in the market [4]. As per
stoichiometric, one mole of glycerol can produce four moles of H2 and three moles of CO (𝐶𝐶3 𝐻𝐻8 𝑂𝑂3 → 4𝐻𝐻2 + 3𝐶𝐶𝐶𝐶).
H2 is notably used in refinery hydrotreating, ammonia production, or as in direct fuel cells or gas turbines whereas CO
can be further shifted through steam-reforming to more H2 and easy to separate CO2. Practically, the Syngas is
cogenerated with CH4, CO2, C2H2 and heavy tars which may be used in a utility burner or co-fired as fuel gas to
produce electricity. A few authors have investigated the numerical simulation of the hydrogen production from
glycerol. Dou et al [6] investigated the numerical simulation of hydrogen production from steam reforming of glycerol
in fluidized bed reactor. In their model the chemical reactions were modeled by laminar finite-rate model. It computes
the chemical sources using Arrhenius expressions. Jin et al [7,8] conducted numerical simulation of glycerol
gasification in supercritical water (SCWG) with a tubular reactor. Their kinetic model was based on a proposed kinetics
model for describing the gaseous products of SCWG of glycerol. The reaction rate constant was obtained for each
reaction using the nonlinear least square-fitting method while relying on the experimental data. In this work, the
kinetics parameters of crude glycerol degradation from thermogravimetric analyzer were determined following
different kinetic methods. These parameters then are used in high fidelity reactive flow to investigate the thermo-
chemical conversion of glycerol into syngas

2. Material and method

2.1. Sample and Thermogravimetric analysis

Multiple samples of crude glycerol are used, those generated from transesterification of WCO at Masdar Institute
Waste to Energy lab and those of medical grade glycerol traded under Druggists LTD, England. Thermogravimetric
analyses are performed using simultaneous DSC/TGA Q600 thermal analyzer at three different heating rates 5 oC/min,
 Manar Almazrouei et al. / Energy Procedia 142 (2017) 1699–1705 1701
Almazrouei, Elsamad and Janajreh* / Energy Procedia 00 (2017) 000–000 3

10 oC/min and 15 oC/min. The samples were heated from ambient temperature until 700 oC in nitrogen environment
flow of 100 ml/min. To ensure the uniformity of the samples temperature and results reproducibility, three samples of
25 ±5 mg were analyzed. Fig. 2 depicts the temperature profile under which all glycerol samples were tested. Initially,
the sample is placed in the device and equilibrates at 50 oC for 2 minutes. Then the sample is heated till 110 oC with
the stipulated constant heating rate and held at 110 oC for 5 minutes to ensure the release of the sample moisture. In
the next stage, the sample is heated from 110 to 700 oC at the same prescribed heating rate.

Fig. 2. (L) Temperature profile in DSC/TGA and TGA/DSC for (M) Crude glycerol (R) pure glycerol pyrolysis at 10 ͦC/min.

The first stage of decomposition of the crude glycerol is due to the presence of methanol and water which are
characterized by low boiling temperatures of 78 and 100 oC, respectively [9]. There is distinct feature between pure
and crude glycerol in the subsequent events in which only one event apparent for the former whereas two events for
the latter. Due et al [10] reported four sub-events/phases incurred during glycerol biodegradation. They also reported,
the mass loss, initial-, maximum- and final-mass loss temperatures of each of these phases. It was further observed
that crude glycerol is catalyzed by the presence of impurities such as water and methanol. The role of the heating rate
increase is observed to be like those have been seen in the degradation of plastic and coal which shifts the degradation
to a higher temperature [11,12]. These events are due to the reaction of multiple components including the catalyzed
glycerol decomposition, fatty acid methyl esters cracking and the residual salts transformation.

2.2. Kinetic evaluation

Because pyrolysis of glycerol is a complex reaction, a kinetic analysis is used to unravel this complex reaction through
the gathered thermogravimetric data. These kinetic parameters include activation energy (E) and Pre-exponential
factor (A). The pyrolysis encompasses the last two degradation events. The data gathered during these complex
reactions and under specific assumptions the evaluation of kinetic parameters by Arrhenius theory can be pursued
following four common methods: i) Arrhenius, ii) Coats-Redfern, iii) Ingraham Marrier and iv) Differential Method
[13, 14]. The 1st order reaction is applied to determine the kinetics parameters from the reaction evolution as:
dX
= kf(X)n (1)
dt
Where dX
dt is the conversion from the TGA data, n is the reaction order and is assumed unity, and  is the heating
rate and after substituting results as:
dX dX w0 −w
= β = kf(X) where x is defined as X= (2)
dt dT w0 −wf
Where w is the weight of the sample at a given time t, w0 and wf are referred to values at the beginning and the end
of the weight event of interest. f(x) and k(T) are functions of conversion and temperature, respectively. k(T) is the
temperature dependence of the rate of weight loss and is written as:
EA
K(T) = Ae−RT and by substituting eq. 1 in eq. 2 yields:
E
dX − A
β = Ae RT f(X) (3)
dT
Where E is the activation energy (kJ/mole), A is the pre-exponential factor (min-1) and R is the universal gas constant,
(8.314 J/mole K) while T is the absolute temperature (K). In 1st order model eq. 3 is expressed as:
E
dX A
= e−RT dT (4)
1−X β
1702 Manar Almazrouei et al. / Energy Procedia 142 (2017) 1699–1705
4 Almazrouei, Elsamad and Janajreh* / Energy Procedia 00 (2017) 000–000

The solution of the final form of the used methods and the parameters used in the kinetic models are shown in Table
1. From the linear plot of x and y axes presented in table 1, E and A are obtained from the line slope and interception.

Table 1. Parameters used for kinetic models

Coats- Redfern Arrhenius Ingraham -Marrier Differential
ln(1 − X) AE E 1 XT1 − XT2 E 1 A dw E dX
ln⁡(− ) = ln − ln ( ) = − ( ) + ln⁡( ) log ( ) = − log T − log β + log A − ( ) E
Form T2 βR RT 1 − X T2 − T1 R T β dT 2.303RT ln⁡( dt ) = lnA −
1−X RT
ln(1 − X) 1 XT1 − XT2 dw
log ( dT )⁡+ log T + log β dX
ln (− ) ln ( )⁡ ( )
y-axis T2 1 − X T2 − T1 ln ( dt )
1−X

x-axis 1 1 1 1
T T T T

2.3. High fidelity model setup

Accurate modeling of the gasification phenomenon requires the application of the four conservative laws including,
mass, momentum, energy, and species in the chemically none-isothermal reacting two-phase flow. In axisymmetric
(cylindrical reactor), the conservation of mass applied to the differential elements in steady state is written as:
𝜕𝜕(𝜌𝜌𝑣𝑣𝑥𝑥 ) 𝜕𝜕(𝜌𝜌𝑣𝑣𝑟𝑟 ) 𝜌𝜌𝑣𝑣𝑟𝑟
𝜕𝜕𝜕𝜕
+
𝑟𝑟
= 𝑆𝑆𝑚𝑚
𝜕𝜕𝜕𝜕
+ (5)
Where ρ is the density and Sm is the source terms due to the dispersed/discrete phase interaction. The conservation of
the axial and radial momentum are described as:
1 𝜕𝜕 1 𝜕𝜕 𝜕𝜕𝜕𝜕 1 𝜕𝜕 𝜕𝜕𝑣𝑣𝑥𝑥 2 1 𝜕𝜕 𝜕𝜕𝑣𝑣𝑥𝑥 𝜕𝜕𝑣𝑣𝑟𝑟
𝑟𝑟 𝜕𝜕𝜕𝜕
(𝑟𝑟𝑟𝑟𝑣𝑣𝑥𝑥 𝑣𝑣𝑥𝑥 ) +
𝑟𝑟 𝜕𝜕𝜕𝜕
(𝑟𝑟𝑟𝑟𝑣𝑣𝑟𝑟 𝑣𝑣𝑥𝑥 ) = −
𝜕𝜕𝜕𝜕
+
𝑟𝑟 𝜕𝜕𝜕𝜕
[rμ (2
𝜕𝜕𝜕𝜕
− (∇ ∙ 𝑣𝑣⃗))] +
3 𝑟𝑟 𝜕𝜕𝜕𝜕
[rμ (
𝜕𝜕𝜕𝜕
+
𝜕𝜕𝜕𝜕
)] + 𝐹𝐹𝑥𝑥 (6)
1 𝜕𝜕 1 𝜕𝜕 𝜕𝜕𝜕𝜕 1 𝜕𝜕 𝜕𝜕𝑣𝑣𝑥𝑥 𝜕𝜕𝑣𝑣𝑟𝑟 1 𝜕𝜕 𝜕𝜕𝑣𝑣𝑟𝑟 2 𝑣𝑣𝑟𝑟 2 𝜇𝜇 𝑣𝑣𝑧𝑧 2
(𝑟𝑟𝑟𝑟𝑣𝑣𝑥𝑥 𝑣𝑣𝑟𝑟 ) + (𝑟𝑟𝑟𝑟𝑣𝑣𝑟𝑟 𝑣𝑣𝑟𝑟 ) = − + [rμ( + )] + [rμ (2 − (∇ ∙ 𝑣𝑣⃗))] − 2𝜇𝜇 + (∇ ∙ 𝑣𝑣⃗) + 𝜌𝜌 +𝐹𝐹𝑟𝑟 (7)
𝑟𝑟 𝜕𝜕𝜕𝜕 𝑟𝑟 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝑟𝑟 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝑟𝑟 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 3 𝑟𝑟 2 3 𝑟𝑟 𝑟𝑟
Where p is the pressure, µ is the fluid viscosity, and Fx is the present axial body forces in the form of gravitational
force. The conservation of energy (E) in steady state system, is written as:
𝑝𝑝 𝑣𝑣 2
∇ ∙ (𝑣𝑣⃗ (𝜌𝜌𝐸𝐸 + 𝑝𝑝)) = ∇ ∙ (𝑘𝑘𝑒𝑒𝑒𝑒𝑒𝑒 ∇𝑇𝑇 − ∑𝑗𝑗 ℎ𝑗𝑗 𝐽𝐽⃗𝑗𝑗 + (𝜏𝜏̿𝑒𝑒𝑒𝑒𝑒𝑒 ∙ 𝑣𝑣⃗)) + 𝑆𝑆ℎ with 𝐸𝐸 = ℎ − + and ℎ = ∑𝑗𝑗 𝑌𝑌𝑗𝑗 ℎ𝑗𝑗 (8)
𝜌𝜌 2
Where E is the internal energy, Keff is the effective conductivity, h is the enthalpy ( h  C p h where Cp is the medium
specific heat) and Yi is the mass fraction whereas Sh is any external energy source. The steady state specie conservation
is written as:
∇ ∙ (𝜌𝜌𝑣𝑣⃗𝑌𝑌𝑖𝑖 ) = −∇ ∙ 𝐽𝐽⃗𝑖𝑖 + 𝑅𝑅𝑖𝑖 + 𝑆𝑆𝑖𝑖 (9)
𝑅𝑅𝑗𝑗,𝑟𝑟
𝑅𝑅𝑗𝑗,𝑟𝑟 = 𝑅𝑅𝑘𝑘𝑘𝑘𝑘𝑘,𝑟𝑟 (𝑝𝑝𝑛𝑛 − )𝑁𝑁 with the Arrhenius reaction rate 𝑅𝑅𝑘𝑘𝑘𝑘𝑘𝑘,𝑟𝑟 = 𝐴𝐴𝑟𝑟 𝑇𝑇𝑝𝑝 𝛽𝛽𝛽𝛽 𝑒𝑒 −(𝐸𝐸𝑟𝑟/𝑅𝑅𝑇𝑇𝑝𝑝 ) (10)
𝐷𝐷 0,𝑟𝑟
Where Do is the effective surface area. The Discrete Phase Equations are written as:
𝑑𝑑𝑢𝑢𝑝𝑝 𝑔𝑔𝑥𝑥 (𝜌𝜌𝑝𝑝 −𝜌𝜌) 18𝜇𝜇 𝐶𝐶𝐷𝐷 𝑅𝑅𝑅𝑅 𝜌𝜌𝑑𝑑𝑝𝑝 |𝑢𝑢𝑝𝑝 −𝑢𝑢|
⁡ = 𝐹𝐹𝐷𝐷 (𝑢𝑢 − 𝑢𝑢𝑝𝑝 ) + + 𝐹𝐹𝑥𝑥 where ⁡𝐹𝐹𝐷𝐷 = and 𝑅𝑅𝑅𝑅 = (11)
𝑑𝑑𝑑𝑑 𝜌𝜌𝑝𝑝 𝜌𝜌𝑝𝑝 𝑑𝑑𝑝𝑝 2 24 𝜇𝜇
Where FD (u - up) is the drag force acceleration; u is the fluid phase velocity; up is the particle/droplet velocity; ρ is the
fluid density and ρp is the density of the particle/droplet. Fx are the particle force balance including Thermophoretic and
Brownian forces which can be used for additional model accuracy. The numerical solution approach [15], the detailed
geometry dimensions and its numerical baseline mesh of a bench-scale gasifier are depicted in Fig. 3. The gasifier is
subjected to 0.5g/sec glycerol mass flux and a fixed wall temperature of 627.15 oC in the subsequent analyses.

Fig. 3. Solution approach, the gasifier geometry and its baseline mesh setup
 Manar Almazrouei et al. / Energy Procedia 142 (2017) 1699–1705 1703
Almazrouei, Elsamad and Janajreh* / Energy Procedia 00 (2017) 000–000 5

The numerical model of the gasification resembles a drop tube gasifier and is pursued in Ansys Fluent [16]. It follows
the modelling path presented in Fig. 3 which combines the Eulerian scheme for the resolution of the conservation of
mass, species, momentum, and energy for the temperature and velocity in the gas phase while using the Lagrangian
scheme to obtain glycerol droplets/mist trajectory. The particle-source-in cell approach is used to couple both the
Eulerian and the Lagrangian schemes and the mist/particle dispersion is solved stochastically. The steady state solution
is pursued in which all the spatial derivative are 2nd order accurate. Glycerol reaction proceed in two events one
governs the volatilization, and the second is the gasification. The kinetics of each event is based on the best fit of these
two events at the highest heating rate to better match the gasifier environment.

3. Results and Discussion

3.1. Kinetic analysis

The thermograph of crude glycerol under the three different heating rates (5, 10, 15 oC/min) is depicted in Fig. 4. The
observed mass loss of the first three events are nearly 13%, 68%, 10% that corresponds to temperature ranges {58-
154 oC }, {154-342 oC }, and {245-505 oC }, respectively. It should be noted that pure glycerol is characterized with
a single event representing the pyrolysis that correspond to temperature range {127-255 oC}. These event values are
in agreement of those reported by Due et al. [10]. For the selected phases of pyrolysis of crude glycerol, the results of
activation energy, pre-exponent factor and R2 calculated by different methods are summarized in Table 2.

Fig. 4. TGA for crude and pure glycerol pyrolysis at different heating rates and their DTG plots.

Fig. 4. Thermographs of crude and pure glycerol under the three different heating rates (5, 10, 15 oC/min)

Table 2. Kinetic analysis results at 15 ͦC/min heating rate

Method Phase Equation R2 E(J/mole) A(min-1 )
  ʹ ›ൌǦ͹ͷͲͷǤͷš൅ͳǤ͵Ͷ͵ͳ ͲǤͻͺ ͸ʹǡͶͲͳ Ͷ͵ͳǡʹͻʹ
‘ƒ–•Ǧ‡†ˆ‡”‡–Š‘†  ͵ ›ൌǦͳ͸ͷͶ͹š൅ͻǤͳͳʹʹ ͲǤͻ͸ʹͳ ͳ͵͹ǡͷ͹ʹ ʹǡʹͷͲǡͲ͵ʹǡͲͶͺ
  ʹ ›ൌǦ͹Ͳͷ͸Ǥ͸š൅ͻǤͷͺʹ ͲǤͻ͹Ͳ͹ ʹͳ͹ǡͷʹͳ ʹͳ͹ǡͷʹͲ
””Š‡‹—•‡–Š‘†  ͵ ›ൌǦͳͶͺ͵͵š൅ͳ͸ǤͶͻͺ ͲǤͺͻ͸ͺ ͳʹ͵ǡ͵ʹʹ ʹͳͻǡ͵ʹͳǡ͹Ͳͻ
  ʹ ›ൌǦʹ͹͹ͺǤʹš൅ͺǤͷʹͶͳ ͲǤͻ͹ͷͶ ͷ͵ǡͳͻͷ ͵͵Ͷǡʹ͹ʹǡͲͲͲ
‰”ƒŠƒǦƒ””‹‡”‡–Š‘†  ͵ ›ൌǦͶʹͺͳš൅ͺǤ͵ͷͶͶ ͲǤͻ͵Ͷͳ ͺͳǡͻ͸ͻ ʹʹ͸ǡͳͷͳǡ͹͹Ͷ
  ʹ ›ൌǦͺ͸ͳͺǤͶš൅ͳͷǤͷʹͺ ͲǤͻͺʹͳ ͹ͳǡ͸ͷ͵ ͷǡͷͶʹǡ͹Ͷʹ
‹ˆˆ‡”‡–‹ƒŽ‡–Š‘†  ͵ ›ൌǦͳͶͺ͵͵š൅ͳͻǤʹͲ͸ ͲǤͺͻ͸ͺ ͳʹ͵ǡ͵ʹʹ ʹͳͻǡ͵ͳͲǡ͸ͻͻ
In general, the activation energies are to some extent consistent within the same group of the kinetic model. However,
more deviations incur between the different modelling approaches. The kinetic data at 15 oC/min heating rate under
Coats- Redfern Method model is utilized in the high fidelity that resulted in the best R2 values amongst other models.

3.2. High fidelity analysis

The temperature distribution along the discrete phase model (DPM) mass source of glycerol profile when the reactor
is set at wall temperature of 626.85 ̊C is shown in Fig. 5. The results illustrate that the glycerol droplets convert to
syngas when the temperature within the reactor reaches the devolatalization temperature. The H2 and CO along the
reactor are shown in Fig. 5. As soon as the glycerol is sprayed into the inlet, it converts to syngas. The conversion of
glycerol to syngas takes place closer to the inlet and results in a mole fraction yield of 0.56 and 0.42 for each of the
1704 Manar Almazrouei et al. / Energy Procedia 142 (2017) 1699–1705
6 Almazrouei, Elsamad and Janajreh* / Energy Procedia 00 (2017) 000–000

H2 and CO, respectively. These results are in a good comparison to the theoretical yield of 0.57 and 0. 43. Therefore,
this tool can be utilized to carry out further parametrical and sensitivity study on the operation conditions and the
reactor design to arrive to optimal configuration for glycerol gasification.

Fig. 5. Profiles of the temperature, DPM mass, and the mole fractions of the syngas (H2 and CO)

4. Conclusion

Thermal devolatalization and gasification of glycerol is a complex process which involves numerous species and
radicals of numerous reactions that nearly impossible to capture. Overall kinetics can be a useful tool to gain a more
insightful picture of the undergoing thermochemical conversion. It is one level up more accurate than the equilibrium
based analysis, i.e. energy minimization or rates constant approach and offer the advantage observing the spatial
distribution of the flow parameters and species This work uses the Thermogravimetric (TG) data for the crude glycerol
to obtain the multiple events kinetics. As initial event (moisture or alcohol evaporation) takes place at a low
temperature range this event is ignored, and the focus was on the two subsequent events, i.e. devolatalization and
gasification. Glycerol gasification proceeds without oxidizer following the devolatalization phase driven by the
presence of high temperature. The resulted Coat-Redfirm kinetics were the best fit amongst other four considered
methods demonstrated by its highest correlation R2. It resulted with activation energy of 62 kJ/mol and 138 kJ/mol
and pre-exponential factor of 4.3e5 min-1 and 2.25e9 min-1 for phase 2 and phase 3, respectively. This best fit was
utilized in high fidelity reactive flow model in a simplified reactor geometry. The calculated kinetic data showed some
consistency within the single model at different heating values but with some discrepancies between the different
models. The best fitted kinetic data was utilized in the high fidelity model and results of CO and H2 fractions
respectively of 0.42 and 0.56 were in a good agreement to the theoretical yield (0.43 and 0.57). These results constitute
the basis to conduct sensitivity study accounting to different flow and geometrical conditions in an attempt to design
robust glycerol gasifier.

References

[1].M. Ardi, M. Aroua, and N. A. Hashim, “Progress, prospect and challenges in glycerol purification process: A review,” Renewable and
Sustainable Energy Reviews, 42, 1164–1173, 2015.
[2] I. Janajreh, R. Abd Rabu. System and method for continuous transesterification oils, Oct 18 2016, US Patent 9,468,899 B2.
[3] R. Mangayil, M. Karp, and V. Santala, “Bioconversion of crude glycerol from biodiesel production to hydrogen”, Int. J. of Hydrogen Energy,
37, 12198–12204, 2012.
[4] S. Andhikari, S. Fernando, S. Gwaltney, S. To, R. Bricka, P. Steele, A. Haryanto,” A thermodynamic Analysis of Hydrogen Production by
Steam Reforming of Glycerol, Int. J. of Hydrogen Energy, 32, 2875-2880, 2007.
[5] H. Tan, M. Aroua, and A. Abdul Aziz, “Progress, prospect and challenges in glycerol purification process: A review,” Renewable and
Sustainable Energy Reviews, pp. 118–127, 2013.
[6] B. Dou and Y. Song, “A CFD approach on simulation of hydrogen production from steam reforming of glycerol in a fluidized bed reactor,” Int.
J. of Hydrogen Energy, 35, 10271–10284, 2010.
[7] H. Jin, S. Guo, L. Guo, and C. Cao, “A mathematical model and numerical investigation for glycerol gasification in supercritical water with a
tubular reactor,” The Journal of Supercritical Fluids, 107, 526–533, 2016.
 Manar Almazrouei et al. / Energy Procedia 142 (2017) 1699–1705 1705
Almazrouei, Elsamad and Janajreh* / Energy Procedia 00 (2017) 000–000 7

[8] S. Guo, L. Guo, J. Yin, and H. Jin, “Supercritical water gasification of glycerol: Intermediates and kinetics”, The Journal of Supercritical Fluids,
vol. 78, pp. 95–102, 2013.
[9] A. Y. Maturana and J. D. Pagliuso, “Thermal Decomposition Behavior Of Crude Glycerin,” 21st Brazilian Congress of Mech. Eng., 9, 461–
484, 2011.
[10] B. Dou, V. Dupont, P. Williams, H. Chen, Y. Ding, Thermogravimetic Kinetics of Crude Glycerol, Bioresources, 100 (2009), 2613-2620
[11] W Zeng, W. Chow, R. Huo, B. Yao, and Y. Li,, “The Degradation Kinetic Study of Polystyrene by a Combination of Non-isothermal
Differential and Integral Methods”, Polymer Materials Science and Engineering (Chinese), 22:5, 162-165, 2006.
[12] S. Shabbar, I. Janajreh, Thermodynamic equilibrium analysis of coal gasification using Gibbs energy minimization method, Energy Conversion
and Management, 65, pp755-763, 2013
[13].S. Syed, R. Qudaih, I. Talab, and I. Janajreh, “Kinetics of pyrolysis and combustion of oil shale sample from thermogravimetric data,” Fuel,
90, 1631–1637, 2011.
[14].S. Vyazovkin and C. A. Wight, “Model-free and model-fitting approaches to kinetic analysis of isothermal and nonisothermal data,”
Thermochimica Acta, 340, 53–68, 1999.
[15] I. Adeyemi, I. Janajreh, T. Arink, T., C. Ghenai, “Gasification behavior of coal and woody biomass: Validation and parametrical study”.
Applied Energy, 185, 1007-1018, 2017.
[16] ANSYS Fluent, ANSYS Release Version 15.0, ANSYS Inc. 2013