EFFECT OF IMPURITIES ON IGNITION AND COMBUSTION
CHARACTERISTICS OF CRUDE GLYCEROL AS A BY-
PRODUCT OF BIODIESEL MANUFACTURING
Hendrix Yulis Setyawan
This thesis is presented for the degree of Doctor of Philosophy of
The University of Western Australia
School of Engineering
Centre for Energy
2018
iii
ABSTRACT
Biodiesel manufacturing produces crude glycerol as a by-product, through
transesterification of fatty acids derived from vegetable oils and animal tallows. Crude
glycerol is produced in large quantities and has low market value, and can be
environmentally and financially burdensome for the biodiesel manufacturing industry.
A potential means of dealing with the large quantities of crude glycerol produced could
be direct combustion to generate heat and possibly power. This could serve as a simple
and effective means of addressing the problems, but little is known about the
combustion of this by-product. In particular, the effects of glycerol impurities such as
water, methanol, soap, and biodiesel on the ignition characteristics and combustion
performance of crude glycerol have rarely been reported.
The overall aim of this research was to investigate the fundamental ignition and
combustion characteristics of glycerol to establish a scientific basis for utilisation of
crude glycerol. The specific objectives were to (i) use a single droplet combustion
technique to study the ignition and combustion characteristics of crude glycerol
compared with those of pure glycerol, biodiesel, diesel, and ethanol; (ii) investigate the
effects of water, methanol, biodiesel, and soap on the ignition and combustion
characteristics of glycerol; and (iii) use Chemkin Pro-based kinetic modelling to further
explore the effects of methanol and biodiesel impurities on the combustion
characteristics of crude glycerol.
Ignition and combustion characteristics of crude glycerol were measured using a single
droplet combustion technique. This technique included use of a horizontal furnace, a
step motor, a CCD camera, a computer, and a flame emission spectrometer. The single
droplet combustion allowed approximation of ignition delay time, burnout time, and
burning rate of the droplet. The results were compared to those of pure glycerol,
iv
biodiesel, diesel and ethanol. Specific attention was paid to the effect of water,
methanol, biodiesel and soap. Chemical effects of the impurities were investigated using
Chemkin Pro-based kinetic modelling. Methanol and biodiesel were selected for
modelling work since these have well-known combustion mechanisms. Combinations
of the combustion mechanisms of glycerol-methanol and glycerol-biodiesel were
proposed separately.
Crude glycerol had a shorter ignition delay time, more rapid burnout, and a higher
burning rate than pure glycerol. These differences were attributed to impurities that
changed the physical and thermochemical properties of crude glycerol, such as
increasing the vapour pressure, reducing the latent heat, and decreasing the boiling
point. The ignition delay time and burnout time of pure glycerol were longer than that
of biodiesel, diesel, and ethanol, leading to a lower burning rate. This occurred because
glycerol has a low Cetane number, high boiling point, and high auto-ignition
temperature. Impurities decreased the burnout time and increased the burning rate for
crude glycerol when compared with pure glycerol, biodiesel, diesel, and ethanol.
An investigation into the effect of each of the major impurities on the ignition and
combustion characteristics of crude glycerol was then carried out on pure glycerol
mixed with water, methanol, biodiesel, or soap, respectively. The addition of water
increased the ignition delay time, and decreased the burnout time and burning rate of
glycerol due to the formation of steam bubbles and subsequent micro-explosions. Water
evaporated at the beginning of the combustion process to form steam and bubbles inside
the droplet. Due to the high heat capacity and latent heat of water, steam formation
delayed heating of the droplet, decreased droplet temperature, and slowed the ignition
process. However, the burnout time decreased due to the micro-explosions of steam
v
bubbles, resulting in the formation of many smaller droplets and consequently an
increased burning rate.
The addition of methanol enhanced the combustion performance of glycerol by
decreasing the ignition delay time and burnout time, and increasing the burning rate.
This effect became more pronounced as the amount of methanol increased. Micro-
explosion phenomena were also observed during combustion of the droplets, which in
turn improved the combustion performance of the glycerol.
The addition of biodiesel decreased the ignition delay time and burnout time, and
increased the burning rate. The flame colour changed during combustion, and the
ignition delay time of the mixture decreased due to the early ignition of biodiesel. A
reddish flame appeared at the beginning of the combustion process due to early ignition
of biodiesel. The flame colour became greenish towards the end of the process due to
glycerol combustion. The burnout time of the glycerol droplets decreased, and the
burning rate increased with increasing addition of biodiesel.
The addition of soap decreased the ignition delay time slightly, decreased the burnout
time, and increased the burning rate. The mixture exhibited two-stage combustion
behaviour: a glycerol-dominated combustion stage followed by a soap-dominated
combustion stage. The addition of soap decreased the ignition delay time slightly and
decreased the overall burnout time of the droplets. The burning rate increased with
increasing soap content due to sodium ions in the flames that promoted the ignition and
subsequent combustion of fuel vapours. The two-stage combustion resembled
evaporation behaviour of a binary mixture of two components with vastly differing
volatilities.
Kinetic modelling results showed that the addition of methanol or biodiesel decreased
the ignition delay time of glycerol. The ignition delay time decreased with increasing
vi
concentration of methanol or biodiesel. The reaction pathway analysis showed that the
ignition was promoted due to an early injection of OH radicals. The presence of O
atoms in the system contributed to the increase in rates of the key reactions. In the
glycerol-methanol mixture, O atoms were released from the formation of combustion
products H2O and CO2, while in the glycerol-biodiesel mixture, O atoms were released
from consumption of CO.
This research has resulted in significant new experimental data regarding the ignition
and combustion characteristics of crude glycerol. Understanding of the effects of the
main impurities has also improved, providing a sound basis for future development of
glycerol combustion theories and technological applications.
vii
TABLE OF CONTENTS
Abstract ............................................................................................................................... iii
Table of contents ................................................................................................................. vii
Acknowledgements ............................................................................................................. xii
Authorship declaration: co-authored publications .............................................................. xiii
List of figures ...................................................................................................................... xviii
List of tables ........................................................................................................................ xxiii
Chapter 1 Introduction .................................................................................................... 1
1.1 Background and motivation ......................................................................................... 1
1.2 Scope and aims ............................................................................................................ 2
1.3 Thesis outline ............................................................................................................... 3
Chapter 2 Literature review ............................................................................................ 6
2.1 Pure glycerol and crude glycerol ................................................................................. 6
2.1.1 Glycerol chemistry and properties ...................................................................... 6
2.1.2 Crude glycerol from biodiesel manufacturing .................................................... 7
2.1.3 Impurities in crude glycerol ................................................................................ 9
2.1.4 Purification of crude glycerol ............................................................................. 11
2.1.5 Market size and current use of glycerol.............................................................. 14
2.1.6 Limitations of glycerol utilisation ...................................................................... 15
2.2 Advance glycerol utilisation Technologies .................................................................. 16
2.2.1 Glycerol pyrolysis and steam reforming ............................................................ 16
2.2.2 Glycerol gasification........................................................................................... 16
2.2.3 Conversion of glycerol into high-value chemicals ............................................. 17
2.3 Combustion Fundamentals and glycerol combustion .................................................. 20
2.3.1 Fundamentals of liquid fuel combustion ............................................................. 20
viii
2.3.2 Crude glycerol as a fuel ...................................................................................... 25
2.3.3 Glycerol combustion in CI engines .................................................................... 29
2.3.4 Glycerol combustion using various burners for utility boilers ........................... 30
2.3.5 Glycerol in fuel blends ....................................................................................... 31
2.4 Summary and specific research objectives .................................................................. 31
Chapter 3 Research methodology, approach and techniques ...................................... 34
3.1 Overview of research strategies ................................................................................... 34
3.2 Single droplet combustion ........................................................................................... 35
3.2.1 Principles of single droplet combustion ............................................................. 35
3.2.2 Single droplet combustion apparatus.................................................................. 36
3.2.3 Analysis of ignition and combustion characteristics .......................................... 41
3.2.4 Limitations and errors......................................................................................... 49
3.3 Single droplet combustion of glycerol ......................................................................... 53
3.3.1 Comparison between ignition and combustion of crude glycerol and that of
biodiesel, diesel, and ethanol .............................................................................. 53
3.3.2 Effect of water on glycerol combustion ............................................................. 53
3.3.3 Effect of methanol on glycerol combustion ....................................................... 55
3.3.4 Effect of biodiesel on glycerol combustion ........................................................ 55
3.3.5 Effect of soap on glycerol combustion ............................................................... 55
3.4 Kinetic studies of mixtures of glycerol with methanol and biodiesel .......................... 56
3.4.1 Chemkin Pro and kinetic model ........................................................................ 57
3.4.2 Reaction mechanisms and kinetic simulations for glycerol-methanol
mixtures .............................................................................................................. 58
3.4.3 Reaction mechanisms and kinetic simulations for glycerol-biodiesel
mixtures .............................................................................................................. 59
ix
Chapter 4 Properties and combustion characteristics of crude glycerol .................... 61
4.1 Properties of crude glycerol ......................................................................................... 61
4.1.1 Water in crude glycerol ...................................................................................... 62
4.1.2 Methanol in crude glycerol ................................................................................. 63
4.1.3 Soap in crude glycerol ........................................................................................ 63
4.1.4 Biodiesel in crude glycerol ................................................................................. 64
4.2 Ignition and combustion characteristics: comparison of crude glycerol against
pure glycerol, petroleum diesel, biodiesel, and ethanol ............................................... 64
4.2.1 Combustion phenomena of crude glycerol ......................................................... 64
4.2.2 Ignition delay time ............................................................................................. 67
4.2.3 Burnout time ....................................................................................................... 69
4.2.4 Burning rate ........................................................................................................ 70
4.3 Summary ...................................................................................................................... 72
Chapter 5 Effect of water................................................................................................. 74
5.1 Properties of the glycerol-water mixtures .................................................................... 74
5.2 The role of water in combustion .................................................................................. 75
5.3 Single droplet combustion study of a pure glycerol-water mixture ............................. 75
5.3.1 Combustion phenomena of pure glycerol-water droplets................................... 75
5.3.2 Ignition delay time .............................................................................................. 77
5.3.3 Burnout time ....................................................................................................... 79
5.3.4 Burning rate ........................................................................................................ 80
5.4 Summary ...................................................................................................................... 81
Chapter 6 Effect of methanol ......................................................................................... 83
6.1 Properties of glycerol-methanol mixtures .................................................................... 83
6.2 Single droplet combustion study of a pure glycerol-methanol mixture ....................... 85
x
6.2.1 Combustion phenomena of pure glycerol-methanol droplets ............................ 85
6.2.2 Ignition delay time ............................................................................................. 87
6.2.3 Burnout time ....................................................................................................... 88
6.2.4 Burning rate ........................................................................................................ 89
6.3 Summary ...................................................................................................................... 91
Chapter 7 Effect of biodiesel ........................................................................................... 92
7.1 Single droplet combustion study of a pure glycerol-biodiesel mixture ....................... 92
7.1.1 Combustion phenomena of pure glycerol-biodiesel droplets ............................. 92
7.1.2 Ignition delay time .............................................................................................. 96
7.1.3 Burnout time ....................................................................................................... 97
7.1.4 Burning rate ........................................................................................................ 98
7.2 Summary ...................................................................................................................... 99
Chapter 8 Effect of soap .................................................................................................. 100
8.1 Single droplet combustion study of a pure glycerol-soap mixture .............................. 100
8.1.1 Ignition and combustion phenomena of pure glycerol-soap droplets ................ 100
8.1.2 Ignition delay time .............................................................................................. 103
8.1.3 Burnout time ....................................................................................................... 104
8.1.4 Burning rate ........................................................................................................ 105
8.2 Sodium ions in the flame of pure glycerol-soap droplets ............................................ 106
8.3 Sodium ions in the solid residue ................................................................................. 108
8.4 Summary ...................................................................................................................... 110
Chapter 9 Kinetic modelling of glycerol-methanol and glycerol-biodiesel
mixtures ........................................................................................................... 111
9.1 Effect of addition of methanol on glycerol combustion kinetics ................................. 112
9.2 Effect of addition of biodiesel on glycerol combustion kinetics ................................. 116
xi
9.3 Summary ...................................................................................................................... 120
Chapter 10 Evaluation and practical implications ...................................................... 122
10.1 Integration and evaluation of the effects of impurities in crude glycerol
combustion ................................................................................................................. 122
10.1.1 Impurities that enhance the ignition and combustion of glycerol .................. 122
10.1.2 Impurities that partly enhance the ignition and combustion of glycerol ....... 124
10.1.3 Damaging effects of soap on combustion ...................................................... 125
10.2 Effect of combined impurities in glycerol combustion ............................................. 126
10.3 Practical implication ................................................................................................. 129
10.4 Identification of the new gaps ................................................................................... 130
Chapter 11 Conclusions and recommendations ............................................................. 131
11.1 Conclusions ............................................................................................................... 131
11.2 Recommendations...................................................................................................... 133
References .......................................................................................................................... 135
xii
ACKNOWLEDGEMENTS
I am sincerely grateful to the many people who contributed to the completion of my
PhD thesis. I would like to express my sincere and utmost gratitude to my supervisor
and academic father, Professor Dongke Zhang, for giving me a precious opportunity to
undertake this PhD study at the Centre for Energy at The University of Western
Australia. It has been an honour to learn from the best combustion engineer in the
world. Professor Zhang inspired and guided me with great patience, provided me with
continuous support and encouragement, graciously shared his phenomenal experience,
knowledge and passion for science with me, and provided me with invaluable advice on
academic and life philosophy. His passion for science has set a wonderful role model
for me to follow in the future. Without his supervision, continuous support and
encouragement, I would not have reached where I am.
I am grateful for the postgraduate research scholarships provided by The Australia
Awards Scholarship and specially thank to Mrs. Debra Basanovic. I am also sincerely
thankful to my home university, Universitas Brawijaya - Indonesia for the time and
support provided.
I am sincerely grateful to my academic family at the Centre for Energy, especially my
co-supervisors, Dr Mingming Zhu and Dr Yang Zhang, and colleagues Mr Zhezi Zhang,
Dr Zhijian Wan, Ms Yii Leng Chan, and other postgraduate students at the Centre for
Energy. I am grateful to Dr. Jo Edmonston, and Dr. Krys Haqq for their support on the
English writing of my thesis. I express my deepest love and gratitude to my parents,
Bapak Tamsir and Ibu Triasmiaisyah, for their continuous love and spiritual support,
and my wife, Juwita Ratna Dewi, for her consistent love and support during my PhD
journey. I am thankful to the Indonesian community in Perth for making me feels at
home, especially Maroonah, Warneds, Aipssa, IBE, and also the Indovarsity photo club.
xiii
AUTHORSHIP DECLARATION: CO-AUTHORED PUBLICATION
This thesis contains work that has been published and/or prepared for publication.
Details of the work:
Setyawan, H. Y., Zhu, M., Zhang, Z. Zhang, D. 2016. Ignition and combustion
characteristics of single droplets of crude glycerol in comparison with pure glycerol,
petroleum diesel, biodiesel and ethanol. Energy, 113, 153–159.
Location in thesis: Chapter 4
Student contribution to work:
The work was supported by and a part of an ARC Linkage Project grant held by
Professor Dongke Zhang
The candidate planned and conducted experiments, and analysed results with
assistance from Mr Zhezi Zhang
The manuscript was drafted by the candidate under the supervision of Professor
Dongke Zhang, and critically reviewed by Professor Dongke Zhang and Dr
Mingming Zhu
Details of the work:
Setyawan, H. Y., Zhu, M., Zhang, Z. Zhang, D. 2015. An Experimental Study of the
Effect of Water on Ignition and Combustion Characteristics of Single Droplets of
Glycerol. Proceedings of the 7th International Conference on Applied Energy, 2015,
Abu Dhabi.
Location in thesis: Chapter 5
Student contribution to work:
The work was supported by an ARC Linkage Project grant held by Professor Dongke
Zhang
xiv
The candidate planned and carried out experiments, and analysed results with
assistance from Dr Mingming Zhu and Mr Zhezi Zhang
The candidate drafted the manuscript under the supervision of Professor Dongke
Zhang and Dr Mingming Zhu
The candidate worked closely with Professor Dongke Zhang, Mr Zhezi Zhang and
Dr Mingming Zhu to critically review the manuscript
The candidate presented the findings at the 7th International Conference on
Applied Energy
Details of the work:
Setyawan, H. Y., Zhu, M., Zhang, Z. Zhang, D. 2015. An Experimental Study of the
Effects of Water on Ignition and Combustion Characteristics of Single Droplets of
Glycerol. Energy Procedia, 75, 578–583.
Location in thesis: Chapter 5
Student contribution to work:
The work was supported by an ARC Linkage Project grant held by Professor
Dongke Zhang
The candidate planned and carried out experiments, and analysed results with
assistance from Dr Mingming Zhu and Mr Zhezi Zhang
The candidate drafted the manuscript under the supervision of Professor Dongke
Zhang and Dr Mingming Zhu
The candidate worked closely with Professor Dongke Zhang, Mr Zhezi Zhang and
Dr Mingming Zhu to critically review the manuscript
xv
Details of the work:
Setyawan, H. Y., Zhu, M., Zhou, W. Zhang, D. 2013. Effects of Methanol Addition
on Combustion Characteristics of Single Droplets of Glycerol. Proceedings of the
Australian Combustion Symposium, 2013, The University of Western Australia, pp.
332–335.
Location in thesis: Chapter 6
Student contribution to work:
The work was supported by an ARC Linkage Project grant held by Professor
Dongke Zhang
The candidate planned and carried out experiments, and analysed results with
assistance from Dr Mingming Zhu
The candidate drafted the manuscript under the supervision of Professor Dongke
Zhang, Dr Mingming Zhu, and Dr Wenxu Zhou
The candidate presented the findings at the Australian Combustion Symposium
Details of the work:
Setyawan, H. Y., Zhu, M., Zhang, Z. Zhang, D. 2015. An Experimental Study of the
Effects of Biodiesel on Ignition and Combustion Characteristics of Single Droplets of
Glycerol. Proceedings of The 11th Asia-Pacific Conference on Combustion, 2017.
Location in thesis: Chapter 7
Candidate contribution to work:
The work was supported by an ARC Linkage Project grant held by Professor
Dongke Zhang
The candidate planned and carried out experiments, and analysed results with
xvi
assistance from Dr Mingming Zhu
The candidate drafted the manuscript under the supervision of Professor Dongke
Zhang
The candidate worked closely with Professor Dongke Zhang and Dr Mingming
Zhu to critically review the manuscript
The candidate presented the findings at the Asia-Pacific Conference on
Combustion
Details of the work:
Setyawan, H. Y., Zhu, M., Zhang, Z. Zhang, D. An Experimental Study of the Effects
of Soap on Ignition and Combustion Characteristics of Single Droplets of Glycerol.
Proceedings of the Australian Combustion Symposium, 2015, University of
Melbourne, pp. 376 - 379.
Location in thesis: Chapter 8
Student contribution to work:
The work was supported by an ARC Linkage Project grant held by Professor
Dongke Zhang
The candidate planned and carried out experiments, and analysed results with
assistance from Mr Zhezi Zhang and Dr Mingming Zhu
The manuscript was drafted by the candidate under the supervision of Professor
Dongke Zhang, and critically reviewed by Professor Zhang and Dr Mingming Zhu
The candidate presented the findings at the Australian Combustion Symposium
xvii
Details of the work:
Setyawan, H. Y., Zhu, M., Zhang, Z. Zhang, D. An Experimental Investigation into
the Effect of Soap on Ignition and Combustion Characteristics of Single Droplets of
Glycerol. Combustion Science and Technology, 189, 1540–1550.
Location in thesis: Chapter 8
Student contribution to work:
The work was supported by an ARC Linkage Project grant held by Professor
Dongke Zhang
The candidate planned and carried out experiments, and analysed results with
assistance from Dr Mingming Zhu
The candidate analysed results and discussed findings with guidance from
Professor Dongke Zhang and Dr Mingming Zhu
The candidate drafted the manuscript under the supervision of Professor Dongke
Zhang and Dr Mingming Zhu
The candidate worked closely with Professor Dongke Zhang and Dr Mingming
Zhu to critically review the manuscript
Student, Principle Supervisor,
Hendrix Yulis Setyawan Dongke Zhang, FTSE
Date: 31 July 2018
xviii
LIST OF FIGURES
1.1 Thesis map ................................................................................................................... 5
2.1 Chemical structure of glycerol ..................................................................................... 6
2.2 Industrial applications of glycerol ............................................................................... 14
2.3 Alternatives pathways and secondary products of glycerol conversion to
chemicals...................................................................................................................... 19
2.4 Flue gas composition of glycerol combustion ............................................................. 26
2.5 Adiabatic flame temperatures of glycerol, biodiesel, diesel, methanol, and
methane ....................................................................................................................... 28
2.6 The 7 kW refractory burners (left) and 82 kW refractory-lined furnaces (right) ........ 30
3.1 Detailed experimental design and network .................................................................. 35
3.2 Schematic of single droplet combustion indicating the radial distribution of fuel
mass and flame ............................................................................................................. 37
3.3 Experimental set-up: (a) a schematic illustration of the single droplet combustion
apparatus; (b) A photograph of the experimental system. ........................................... 38
3.4 The schematic of the furnace heating system .............................................................. 38
3.5 The temperature gradient along the furnace ................................................................ 39
3.6 The difference between actual and measured temperature as a function of
measured gas temperature ............................................................................................ 39
3.7 Schematic representation of a burning droplet ............................................................ 41
3.8 Pixel greyscale intensity .............................................................................................. 42
3.9 Binary image of droplet ............................................................................................... 43
3.10 Image of binary droplet separated from fibre ............................................................ 44
3.11 Image of filament separated from droplet ................................................................ 44
3.12 Major and minor axis of the droplet .......................................................................... 45
xix
3.13 Micro-explosion of droplet with numbering for user input ....................................... 46
3.14 Typical combustion of a droplet ................................................................................ 46
3.15 The d2 behaviour of droplet vapourisation and combustion ...................................... 48
3.16 A schematic representation of single droplet combustion experimentation
apparatus equipped with flame emission spectrometer ............................................. 48
3.17 Shape irregularities from droplet swelling ................................................................ 51
3.18 The reflected lights occur on the droplet surface ....................................................... 52
3.19 Schematic of the ignition delay time as simulated using Chemkin Pro .................... 58
3.20 Validation of the ignition of the combined mechanism of glycerol-methanol
model ......................................................................................................................... 59
3.21 Validation of the ignition of the combined mechanism of glycerol-biodiesel
model ......................................................................................................................... 60
4.1 Typical time-sequenced images of burning droplets for crude glycerol (CG),
pure glycerol (PG), biodiesel (BD), diesel (DS) and ethanol (ET). ............................. 65
4.2 Temporal variation of the square of the normalised droplet diameter of fuels
during combustion ....................................................................................................... 66
4.3 The ignition delay times at different temperatures for droplets of different fuels .... 67
4.4 Burnout times of various fuels tested at different temperature .................................... 69
4.5 Comparison of the burning rates of crude glycerol, pure glycerol, biodiesel,
diesel, and ethanol ........................................................................................................ 71
5.1 Typical images of burning glycerol droplets with various water concentrations ........ 76
5.2 Temporal variations of the square of the normalised droplet diameters of
glycerol with different water concentrations ............................................................... 77
5.3 Ignition delay time of glycerol droplets with and without water addition, and
crude glycerol ............................................................................................................... 78
xx
5.4 Burnout time of glycerol droplets with addition of different concentrations of
water ............................................................................................................................. 79
5.5 Burning rates of glycerol droplets with addition of different concentrations of
water ............................................................................................................................. 80
6.1 The viscosity of the glycerol-methanol mixture at various temperatures .................... 84
6.2 The density of the glycerol-methanol mixture at various temperatures ...................... 85
6.3 Typical images of burning glycerol droplets with various methanol
concentrations: pure glycerol (PG), glycerol-methanol mixture (GM) ....................... 86
6.4 Measured temporal variation of the square of the normalised droplet diameter of
glycerol with different concentration of methanol (GM) ............................................ 87
6.5 The effect of the addition of methanol on the ignition delay time of glycerol
combustion ................................................................................................................... 88
6.6 Burnout time of glycerol with different concentration of methanol ............................ 89
6.7 Burning rates of glycerol with different concentration of methanol............................ 90
7.1 Typical time-sequenced images of burning glycerol droplets and pure glycerol-
biodiesel additions (GB) at 1023 K ............................................................................. 94
7.2 Temporal variations of the square of the normalised droplet diameters of pure
glycerol (PG) and glycerol-biodiesel (GB) .................................................................. 95
7.3 Effect of the addition of biodiesel on the ignition delay time of glycerol droplets .... 96
7.4 Burnout time of glycerol droplets at different concentration of biodiesel ................... 97
7.5 Effect of the biodiesel addition on burning rate of glycerol droplets .......................... 98
8.1 Typical time-sequenced images of burning glycerol droplets and pure glycerol-
soap droplets ................................................................................................................ 101
8.2 Normalised temporal evolution of the squared diameter for pure glycerol, GS5,
and pure soap at 1023K................................................................................................ 103
xxi
8.3 Ignition delay time of glycerol droplets with soap addition at different
concentrations at 1023K .............................................................................................. 104
8.4 Burnout time of glycerol droplets with soap addition at different concentrations
at 1023K ....................................................................................................................... 105
8.5 The burning rates of glycerol droplets with the addition of different amounts of
soap: (a) Stage 1 and (b) Stage 2.................................................................................. 106
8.6 Changes in signals of sodium in flames of droplets with the addition of various
concentrations of soap in glycerol ............................................................................... 107
8.7 The SEM images of the solid residue surface at various soap concentration .............. 109
9.1 Mole fractions of glycerol, methanol, and OH of combustion of glycerol-
methanol mixture ......................................................................................................... 112
9.2 Effect of various methanol addition on ignition delay time of pure glycerol-
methanol mixture ......................................................................................................... 113
9.3 Relative rates (indicated by the horizontal bar) of the consumption of pure
glycerol (PG) and glycerol-methanol (GM) by various elementary reactions at
residence time 0.1, initial temperature 1023 K and the equivalence ratio 1.0 ............. 114
9.4 Comparison of mole fractions of OH, H2O, CO2, O, CO and H of pure glycerol
(PG) and 40v% glycerol-methanol mixture (GM40) ................................................... 115
9.5 Mole fractions of glycerol, methanol, and OH of combustion of glycerol-
methanol mixture ......................................................................................................... 117
9.6 Effect of various biodiesel additions on ignition delay time of pure glycerol-
biodiesel mixture .......................................................................................................... 118
9.7 Relative rates (indicated by the horizontal bar) of the consumption of pure
glycerol (PG) and 10v% glycerol-biodiesel (GB) by various elementary
xxii
reactions at residence time 0.1s, initial temperature 1023K and the equivalence
ratio 1.0 ........................................................................................................................ 119
9.8 Comparison of mole fractions of OH, CO2, H2O, CO, O, and H of pure glycerol
(PG) and 10v% glycerol-biodiesel mixture (GB10) .................................................... 120
10.1 Fuel atomisation with addition of water in the fuel ................................................... 125
10.2 Ignition delay time of pure glycerol mixed with various concentration of
methanol and biodiesel with addition of: a) 5% water and 1% soap; b) 20v%
water and 1v% soap; c) 5v% water and 5v% soap .................................................... 127
10.3 Burnout time of pure glycerol mixed with various concentration of methanol
and biodiesel with addition of: a) 5% water and 1% soap; b) 20v% water and
1v% soap; c) 5v% water and 5v% soap ..................................................................... 128
10.4 Burning rate of pure glycerol mixed with various concentration of methanol
and biodiesel with addition of: a) 5% water and 1% soap; b) 20v% water and
1v% soap; c) 5v% water and 5v% soap ..................................................................... 129
xxiii
LIST OF TABLES
2.1 Comparison of biodiesel production technologies ...................................................... 9
2.2 Various properties of glycerol ..................................................................................... 11
2.3 Projection of biodiesel production 2014–2023 ............................................................ 13
2.4 Alternative technological applications and products of glycerol conversion .............. 18
2.5 Properties of crude glycerol and pure glycerol ............................................................ 27
3.1 Properties of silicon carbide fibre ................................................................................ 40
3.2 Thermodynamic properties of crude glycerol compared with those of other fuels ..... 54
3.3 Compositions of olive oil used in the present study .................................................... 56
3.4 Simulation condition .................................................................................................... 58
4.1 Properties of crude glycerol ......................................................................................... 62
5.1 Key physical and thermodynamic properties of glycerol, water, and glycerol-
water mixtures (GW) ................................................................................................... 74
6.1 Chemical components and physical properties of glycerol and methanol ................... 84
7.1 Thermodynamic properties of pure glycerol and biodiesel ......................................... 93
8.1 Ash and sodium in ash produced by combustion of various concentrations of
soap in glycerol ............................................................................................................ 109
10.1 Research on sodium related to combustion ............................................................... 126
1
Chapter 1 Introduction
1.1 Background and motivation
Biodiesel is an alternative liquid transport fuel derived from various sources of fatty
acids including vegetable oils and animal tallow [1]. Crude glycerol is a by-product of
the production of biodiesel and contains impurities from the manufacturing process. The
production of every 100 tonnes of biodiesel yields approximately 10 tonnes of crude
glycerol as a by-product. [2]. In 2015, ~40 million tonnes of biodiesel were produced
globally [3], generating ~4 million tonnes of crude glycerol, twice the amount the global
market could absorb. Remaining crude glycerol causes significant environmental and
economic challenges for the biodiesel manufacturing industry, particularly regarding
disposal and further utilisation [2].
A number of processes have been investigated for the effective utilisation of crude
glycerol, including pyrolysis, fermentation, and steam reforming. While these processes
have been used to convert crude glycerol into valuable products such as hydrogen [4],
acrolein [5], syngas [6], and surfactant [7], the processes can be expensive or the
product volume and value too low to be economically viable. Utilisation of crude
glycerol is likely to require large throughputs to be economically viable [8], and
biodiesel plants need to dispose of large quantities of glycerol. Therefore, combustion
for heat and power production (CHP) in biodiesel plants could serve to effectively
utilise excess crude glycerol.
The combustion of crude glycerol is inherently challenging due to the nature of glycerol
and the impurities in crude glycerol. By nature, glycerol has high viscosity, low energy
density, high auto-ignition temperature, and low heating value. Combustion is further
2
complicated by impurities introduced during biodiesel production, including water,
methanol, biodiesel, and soap [9], and these are costly to remove.
Pure glycerol has been tested for direct combustion using utility boilers and diesel
engines [8, 10]. In a heating boiler equipped with a high-swirl refractory burner, pure
glycerol has been burned and combusted efficiently [10]. A diesel engine generator
fuelled with pure glycerol has also generated continuous power [8]. In contrast to
combustion research of pure glycerol, studies of crude glycerol are limited. Combustion
of crude glycerol is challenging due to the presence of impurities, but the role of these
impurities in combustion is not clear [11]. Therefore, to utilise crude glycerol as a fuel,
a better understanding of the fundamental combustion characteristics is required,
particularly regarding the effects of the impurities in combustion.
1.2 Scope and aims
This research aimed to investigate the utilisation of crude glycerol as a fuel for diesel
engines. Combustion characteristics of crude glycerol were compared with those of pure
glycerol and other fuels using the single droplet combustion technique. The effects of a
range of impurities on crude glycerol combustion were also assessed.
Four significant impurities of crude glycerol, namely water, methanol, biodiesel, and
soap, were investigated. The single droplet combustion technique was used to determine
the effects of these impurities on ignition delay times, burnout times and burning rates
of glycerol droplets. Investigation of the effects of soap included exploration of sodium
released in the flame and in the deposit, using a flame spectrometer and SEM-EDS,
respectively. The kinetic effects of methanol and biodiesel in glycerol combustion were
also examined.
3
1.3 Thesis outline
There are a total of 11 chapters in this thesis. Figure 1.1 presents a schematic map of the
11 chapters in this thesis, and each chapter is outlined below:
Chapter 1 Glycerol production and utilisation are identified, providing reasoning for
the alternative solution via combustion for the abundance production of
crude glycerol. Scope, aims and thesis structure are defined.
Chapter 2 Existing knowledge of glycerol production, consumption, and utilisation is
reviewed and combustion of glycerol is introduced. Knowledge gaps and
specific objectives for the research are identified.
Chapter 3 Methodology employed to achieve the research objectives identified in
Chapter 2 is presented. The experimental procedure using the single droplet
combustion technique is described. Kinetic modelling of the glycerol-
methanol and glycerol-biodiesel mixtures is also presented.
Chapter 4 Comparisons between the ignition and combustion of crude glycerol, pure
glycerol, biodiesel, diesel, and ethanol are discussed. An investigation of
whether the ignition and combustion of crude glycerol differs from that of
pure glycerol and the other fuels, and whether impurities change the
combustion characteristics of crude glycerol, is presented.
Chapter 5 The effect of water on the ignition and combustion characteristics of the
single droplet of crude glycerol, including combustion phenomena, ignition
delay, burnout time, and burning rate, are investigated.
Chapter 6 The effects of methanol on the ignition and combustion characteristics of
the single droplet of crude glycerol, including combustion phenomena,
ignition delay, burnout time, and burning rate, are investigated.
4
Chapter 7 The effect of biodiesel on the ignition and combustion characteristics of the
single droplet of crude glycerol is investigated.
Chapter 8 The effect of soap on the ignition and combustion characteristics of the
single droplet of crude glycerol is investigated. Sodium released in the
flame and the solid deposit is discussed.
Chapter 9 Simulation of the kinetic effect of the addition of either methanol or
biodiesel on the ignition and combustion of crude glycerol is described.
Chapter 10 Results are evaluated against those in the literature and specific research
objectives. The implications of these findings in functional processes, along
with new knowledge gaps identified for future studies, are discussed.
Chapter 11 Conclusions and recommendations for future investigations are presented.
5
Figure 1.1 Thesis map
Chapter 1: Introduction
• Background and motivation O Crude glycerol: production, utilisation, and limitation O Combustion of crude glycerol as a simple means of utilisation• The scope of thesis O Glycerol combustion with emphasis on the effect of impurities: water, methanol, biodiesel, and soap• Overall aims• Thesis structure
Chapter 2: Literature Review
• Pure and crude glycerol: properties, impurities, production, utilisation• Glycerol utilization technologies• Combustion of liquid fuels, glycerol combustion• Crude glycerol as a fuel• Conclusions from literature review• Specific research objectives of this thesis work
Chapter 3: Methodology, Approach and Experimental Techniques
• An outline of research strategies• Single droplet combustion technique O Flame emission spectroscopy O SEM/EDS for investigation of ash from crude glycerol combustion• Kinetic Modelling• Data analysis O Error analysis
Chapter 4: Properties and Combustion Characteristics of Crude Glycerol
• Crude glycerol properties• Ignition characteristics and combustion behavior of single droplets, comparison with pure glycerol, petroleum diesel, biodiesel and ethanol• Summary
Chapter 5: Effects of Water
• Ignition and combustion mechanisms• Ignition delay time• Burnout time • Droplet size• Burning rate • Summary
Chapter 6: Effects of Methanol
• Ignition and combustion mechanisms• Ignition delay time• Burnout time • Droplet size• Burning rate • Summary
Chapter 7: Effects of Biodiesel
• Ignition and combustion mechanisms• Ignition delay time• Burnout time • Droplet size• Burning rate • Summary
Chapter 8: Effects of Soap
• Ignition and combustion mechanisms• Ignition delay time• Burnout time • Droplet size• Flame spectroscopy• Ash morphology • Burning rate • Summary
Chapter 9: Kinetic Modelling of Ignition of Glycerol-Methanol and Glycerol- Biodiesel Mixtures
• Kinetic modelling of ignition of mixtures of glycerol – methanol• Kinetic modelling of ignition of mixtures of glycerol – biodiesel• Summary
Chapter 10: Evaluation and Practical Implications
• Integration of findings from present thesis work• Evaluation against the literature• Evaluation against the specific objectives• Practical implications.• Identification of new gaps.
Chapter 11: Conclusions and Recommendations
• Summarise new, significant findings as conclusions of the thesis research• Recommendations for future work based on new gaps identified
6
Chapter 2 Literature review
In this chapter, the production, and physical, and thermodynamic properties of
glycerol, are reviewed. The characteristics of crude glycerol as a by-product of
biodiesel manufacturing, and the current status of glycerol utilisation, including use
for fuel and combustion purposes, are also reviewed. The characteristics and
mechanism of glycerol combustion are discussed, and technical and scientific issues
regarding combustion of crude glycerol, particularly the effects of impurities, are
also highlighted.
2.1 Pure glycerol and crude glycerol
2.1.1 Glycerol chemistry and properties
Glycerol is also known as 1,2,3-propanetriol or glycerine [12] and has the chemical
formula C3H8O3. Glycerol is a stable alcohol under most conditions, but can easily
react to form derivatives. Glycerol is a non-toxic, odourless, colourless viscous
liquid with a high boiling point, and is miscible with both water and alcohol [13]. At
low ambient temperatures, glycerol remains in liquid form rather than forming
crystals.
Figure 2.1 Chemical structure of glycerol.
Glycerol is an important intermediate chemical compound in the chemical industry.
Glycerol consists of three hydrophilic alcoholic hydroxyl groups (Figure 2.1), that
cause glycerol to be hygroscopic in nature [14], and able to easily absorb moisture.
7
In the aqueous phase, intramolecular hydrogen bonds and intermolecular solvation of
the hydroxyl group serve to stabilise the glycerol molecule [14].
2.1.2 Crude Glycerol from biodiesel manufacturing
Glycerol was first discovered in the late sixteenth century [15]. In 1779, Scheele, a
Swiss pharmacist, isolated glycerol when heating a mixture of litharge (PbO) with
olive oil [15]. In 1811, a French chemist, Chevrel, patented the first industrial
method for obtaining glycerol that involved reacting fatty material with lime and
alkali [15]. In 1836, another French scientist, Pelouze, determined the empirical
formula of glycerol [15]. The earliest method of glycerol production used microbial
fermentation, and involved manipulation and selection of particular strains of fungi
and algae [16], such as Saccharomyces cerevisiae [17-19], Candida glycerinogenes
[20], and Dunaliella tertiolecta [21]. More recently, biodiesel industries have
become the largest producers of glycerol. Transesterification, a process converting
oil and fat into biodiesel, produces glycerol as a by-product.
Transesterification is a reaction of oils or fats with alcohol to form esters and
glycerol (Reaction 2.1). Various oils and fats are known as biodiesel feedstock,
including palm oil, rapeseed oil, canola oil, soybean oil, crambe oil, waste cooking
oil, and animal fat [22, 23]. Feedstock quality, particularly the free fatty acid (FFA)
content, dictates the process and type of catalysts used [24]. The triglyceride
feedstock, stimulated by a catalyst, reacts with alcohol to produce biodiesel and
glycerol (Reaction 2.2).
RCOOR′ + ROH → RCOOR + R′OH .............................................................. (R 2.1)
C3H5(OCOR)3 + 3CH3OH Catalyst↔ 3COOR + C3H5(OH)3 ............................. (R 2.2)
8
Methanol or ethanol is often used for transesterification. In a stoichiometric reaction,
each triglyceride requires three moles of alcohol (Reaction 2.2) but an excess of
alcohol, up to 100%, is needed to accelerate the reaction and obtain a higher yield of
biodiesel [22].
Catalysts commonly used for transesterification include NaOH, KOH, and sodium
methoxide (NaOMe). The productivity of transesterification varies according to the
chosen catalyst [25]. Base catalysts are commonly used in transesterification since
these can achieve a higher reaction rate than the acid counterparts. The average yield
for biodiesel in a transesterification process is approximately 90%. The remaining
10% consists of the by-product, crude glycerol [2].
Many studies have investigated modifying the transesterification process to increase
the yield of biodiesel, decrease that of crude glycerol and simplify the purification
process. Novel catalysts and processes investigated include biocatalysts [26, 27] and
the supercritical methanol method, neither of which are economically viable.
Biocatalysts are expensive and have low reaction rates for enzymatic biodiesel
production [28]. The supercritical methanol method, that involves simultaneous
etherification of fatty acids and rapid transesterification, does not require a catalyst
[29] but does require a high methanol-to-triglyceride molar ratio of 42:1, and this is
not feasible in chemical industry [30]. The enzymatic and chemical processes of
biodiesel production are compared in Table 2.1.
Research into methods of biodiesel production, particularly regarding the
development of catalysts, is ongoing [24]. Transesterification using a homogeneous
base catalyst is a kinetically rapid and economically feasible method [31]. One
drawback of this method, however, is the production of low-grade glycerol (crude
glycerol), that contains impurities.
9
Table 2.1 Comparison of biodiesel production technologies
Parameter Chemical process Enzymatic process
Alkaline process Acid process
FFA content in the
raw material
Saponified
product
Converted to
biodiesel
Converted to
biodiesel
Water content in the
raw material
Soap formation Catalyst
deactivation
Not deleterious
Biodiesel yield (%) >90 >90 ±90
Reaction rate High Slower than
alkaline process
Low
Glycerol recovery Complex, low-
grade glycerol
Complex, low-
grade glycerol
Easy, high-grade
glycerol
Reaction
temperature (oC)
60–80 >100 20–50
Catalyst cost Cheap Cheap Expensive
Environmental
impact
High, waste water
treatment needed
High, wastewater
treatment needed
Low, wastewater
treatment not
needed
2.1.3 Impurities in crude glycerol
Crude glycerol contains impurities such as water, soap, excess of transesterification
catalyst (NaOH or KOH), methanol, and fatty acid methyl ester (FAME) are
commonly present [9]. These impurities come from the feedstock, reactants, method
used, and the secondary reaction that produces soap and water.
Different types of feedstock and transesterification method affect the composition of
crude glycerol [32, 33]. The effects of feedstock on biodiesel and crude glycerol
10
have shown, for example, that feedstocks with high water content can increase the
proportion of water in crude glycerol [28]. The batch, or continued transesterification
process, affects the ratio of alcohol-to-catalyst in the process. The type of catalyst
used in the transesterification process has significant effects on properties of crude
glycerol, e.g. NaOH or KOH can determine the metal content in crude glycerol.
NaOH contributes 1.2–1.8wt% of sodium-to-crude glycerol as excess catalyst [9].
Variations on transesterification methods can be used to increase the degree of
biodiesel conversion [22]. However, excess reactants (catalyst and alcohol) may be
stored in crude glycerol once transesterification has been completed. These reactants,
e.g. methanol, can be undesirable due to toxicity [34].
Other impurities are also formed in crude glycerol during transesterification. Water
is consistently present in crude glycerol because glycerol is highly hygroscopic.
Impurities can also be derived from elemental reactions during transesterification, as
in Reactions 2.3–2.5 [35]. Reaction 2.5, the transesterification process that uses an
alkaline catalyst, produces soap and water.
CH3OH + NaOH → CH3ONa + H2 ................................................................ (R 2.3)
C3H5(OCOR)3 + H2O → C3H5OH(OCOR)2 + RCOOH .................................. (R 2.4)
RCOOH + NaOH → RCHCOONa + H2O ......................................................... (R 2.5)
Table 2.2 shows impurities known to occur in crude glycerol. The glycerol content
can vary from 27%wt to 92%wt, depending on the feedstock and transesterification
process. At lower glycerol concentrations (<50%), impurities are mostly methanol,
biodiesel, water, and soap. The transesterification process can be improved by
recovering the methanol contained in crude glycerol for use in the next round of
transesterification.
11
Table 2.2 Various glycerol properties
Glycerol
(wt%)
Na and K
salt (wt%)
Methanol
(wt%)
Water
(wt%)
Soap
(wt%)
Ash
(wt%) Reference
27 1 8.6 4.1 20.5 2.7 [36]
33 1 12.6 6.5 26.1 2.8 [36]
48.7 n/a 22.7 25.6 3.0 n/a [37]
50 4–5 1.3 36 n/a n/a [38]
56.5–62.4 n/a 12.8–28.3 n/a 15.3–25.2 n/a [39]
62.5–76.6 n/a 23.4–37.5 n/a n/a 0.25–5.5 [33]
65 4–5 3 26 n/a n/a [40]
65 4–5 1 28 n/a n/a [41]
70 12 2 14 n/a n/a [42]
80 4 1 13.5 n/a n/a [43]
80.1 5.6 <0.2 12.9 1.2 n/a [42]
83.4 n/a n/a 11.6 1.3 2.7 [44]
85.0 4.5 1.5 6.0 3.0 n/a [42]
86 6.5 <0.2 n/a n/a n/a [45]
92 2–3 0.01 6 n/a n/a [46]
2.1.4 Purification of crude glycerol
Since 2006, crude glycerol has been produced on a large scale as the by-product of
biodiesel [47]. Global production of glycerol is expected to increase as biodiesel
production increases. The average production of biodiesel in 2011–2013 was
approximately 31 million tonnes (Table 2.3), with a concomitant production of over
3 million tonnes of crude glycerol. Biodiesel production is projected to increase to 49
12
million tonnes by 2023, and this will significantly increase crude glycerol
production.
Since crude glycerol is continually produced as a by-product of biodiesel,
exploration of the economic utilisation of this by-product in biodiesel production is
essential. Despite being produced in large quantities, crude glycerol from
commercial biodiesel production is poor in quality and requires purification for
further utilisation [48]. Crude glycerol can be purified using a number of methods
including distillation, chemical extraction, adsorption using activated carbon, and
ion- exchange [23].
Glycerol purification has been achieved by molecular distillation at pH 3.5,
producing 96.6% glycerol, 0.03% ash, 1% water, and 2.4% matter organic non-
glycerol (MONG) [29]. However, this method is not feasible as the cost per yield is
too expensive to be economically viable [29].
Chemical extraction [26] neutralises alkaline substances with a strong acid and
remaining esters are saponified with an alkali, producing approximately 86wt% of
glycerol [29]. This method is unfavourable due to the resulting chemical waste.
Glycerol purification using sewage sludge-derived activated carbons (ACs) has
produced 93wt% glycerol [49]. However, the reaction rate is low, requiring two
hours of adsorption time. Ion exchange has been used to separate sodium from
glycerol [50]. Salt needs to be removed before separation of the sodium.
While purification of crude glycerol has been achieved using a number of
techniques, none of these can be considered cost-effective at present [51].
13
Table 2.3 Projection of biodiesel production for 2014–2023 [52]
Country or Union of Countries
Production (× 103 Tonnes)
Growth (%)
Domestic use (× 103 Tonnes)
Growth (%)
Average 2011–2013
Projection 2023
Projection 2014–2023
Average 2011–2013
Projection 2023
Projection 2014–2023
Canada 384 759 -0.01 667 1050 2.4
USA 5167 8245 2.3 4666 7981 2.2
EU 12,905 19,978 5.1 16325 24008 4.7
Australia 827 934 1.1 827 934 1.1
South Africa 94 136 3.36 94 136 3.4
Mozambique 88 121 2.92 24 59 5.1
Tanzania 78 164 7.29 0 72 64.2
Argentina 3282 4595 3.29 1270 2194 3.8
Brazil 3455 4903 1.9 3460 4830 1.8
Colombia 760 1191 3.4 760 1190 3.4
Peru 117 220 6.3 342 426 2.3
India 365 919 10.0 504 1115 7.5
Indonesia 2247 4098 4.4 715 2306 7.0
Malaysia 200 1054 11.5 112 816 15.0
Philippines 196 528 8.1 196 528 8.1
Thailand 1036 1462 1.96 1036 1462 2.0
Turkey 15 34 8.3 15 35 8.5
Vietnam 30 123 10.4 30 122 10.4
TOTAL 31248 49464 4.0 31046 49263 4.1
14
2.1.5 Market size and current use of glycerol
Industries that use glycerol as raw material are shown in Figure 2.2. In 2009,
approximately 0.9 million tonnes of refined glycerol were produced in the global
market and chiefly used for pharmaceutical and cosmetic purposes, animal nutrition,
and glycol substitution. In the pharmaceutical industry, glycerol is used to increase
humidity and viscosity of liquid drugs. Glycerol can also act as a humectant and
moistener in skin and hair care products by drawing water from its surroundings.
Glycerol is non-toxic and therefore useful in the food industry as a lubricant, solvent,
sweetener, preservative, and also serves to keep adhesives and glues from drying too
quickly [23].
Figure 2.2 Industrial applications of glycerol [21].
In 2015, approximately 2 million tonnes [2] of glycerol were produced in the global
market. The global market for refined glycerine increased with an increase in
products derived from glycerol. Examples include conversion of acrylic acid to
acrolein, propylene glycol, and methanol through glycerol dehydration [2].
15
2.1.6 Limitations of glycerol utilisation
Factors limiting development of the glycerol industry include geographically diverse
locations of glycerol production, problems with glycerol purification, and market
saturation. Most glycerol resulting from biodiesel production is produced in
countries of the European Union, the USA, Argentina, Brazil, Indonesia, and
Thailand (Table 2.3). Given the low price and high transportation cost of glycerol,
transport of crude glycerol from biodiesel production plants to glycerol purification
industries is generally not economically viable. Glycerol is a viscous liquid and
therefore difficult to transport by pipeline, and hence more expensive to transport
than other liquid fuels, such as gasoline or diesel.
Another limiting factor in the industrial glycerol market is the requirement for
glycerol feedstock with a high level of purity. However, purification of crude
glycerol is expensive and not economically viable. Purification of crude glycerol is
also potentially detrimental to the environment through the production of chemical
waste as by-product and this further restricts the use of crude glycerol in industry.
In addition to the limitations of geographical locations and purity, market saturation
is also a problem for development of the glycerol production industry. The
production of crude glycerol increased to 3.1 million tonnes from 2003 to 2013, but
less than 2 million tonnes could be absorbed by the market [2]. As a result, the
excess glycerol caused a drop in the prices of both refined and crude glycerol [2]. In
2000, the price for refined glycerol was 4000 Euro/tonne, but fell dramatically to 450
Euro/tonne early in 2010. By mid-2014, glycerol had fallen even further to 224
Euro/tonne [2]. Since the utilisation of glycerol is limited by geographical locations
of production sites and requirement for purity, technological applications need to be
implemented for the utilisation of crude glycerol.
16
2.2 Advanced glycerol utilisation technologies
Many efforts have been made to use glycerol as a raw material for industrial
products [13, 22], and are described below.
2.2.1 Glycerol pyrolysis and steam reforming
Glycerol can be converted into syngas through pyrolysis. Research into this method
is developing rapidly and has resulted in kinetic modelling of glycerol pyrolysis [53],
and a better understanding of the pathways of glycerol dehydration via glycidol [54].
The development of catalytic glycerol processes also suggests that glycerol could be
a potential source for syngas production [55]. However, current technological
applications are limited to the laboratory and have not been tested in industry.
The steam reforming method has potential for converting glycerol into hydrogen.
The optimum processing temperature for this conversion has been determined by
thermodynamic analysis of hydrogen production from glycerol, using oxidative
steam reforming [56]. Preliminary studies of co-steam reforming on glycerol mixed
with biomass [57], and the effect of catalysts (Ni/CeZrO2/Al2O3) have also been
undertaken [58]. Hydrogen production from glycerol steam reforming has been
highly selective, resulting in 88% hydrogen and 99.7 v/v% purity [59].
2.2.2 Glycerol gasification
Gasification is an advanced, stable technological application of glycerol combustion,
and involves gasification of glycerol-biomass and glycerol-liquid fuel mixtures.
Gasification of a glycerol-biomass (hardwood) mixture has been used to produce
syngas [60]. Gasification of 20wt% glycerol absorbed into hardwood chips showed
insignificant effects on emissions (e.g. H2 and CO2), LHV, and particles, but as
expected, increased CO, CH4, and tar concentrations [60]. While these results
17
suggest that glycerol could be used as a substitute for hardwood chips in syngas
production, further work is required to determine how glycerol should be loaded into
the mixture.
Glycerol can be added to fuel to improve consumption and reduce emissions [61].
Processes that convert glycerol into a fuel additive are etherification, acetylation, and
acetalation. Etherification of glycerol with isobutylene produces glycerol ethers that
have a high octane number, improving the combustion performance of gasoline.
Etherification of glycerol using tertbutyl alcohol produces an oxygenated additive
suitable for diesel fuels [62]. Catalytic glycerol decomposition produces acetol and
1.2 propanediol, that have been used to decrease the formation of particulate matter
in diesel engines [63]. However, the use of glycerol as a fuel additive requires pure
glycerol.
The acetylation of glycerol generates diacetylglycerol and triacetylglycerol, both of
which are miscible with gasoline and diesel [64], and results in over 90wt% glycerol
conversion [64]. However, in the acetylation process, glycerol reacts with acetone
and formaldehyde in the presence of ethanol [65], suggesting that glycerol acetal can
only be blended with gasoline in the presence of ethanol.
2.2.3 Conversion of glycerol into high-value chemicals
Glycerol has been used to create a variety of valuable derivative products. Table 2.4
shows the conversion of glycerol to high value products using microbial and
chemical processes. Propanediol, used for the production of polyesters,
polycarbonates, and polyurethanes, has been produced from glycerol using
fermentation.
18
Table 2.4 Alternative technological applications and products of glycerol
conversion
Product name Process method/Nature Researchers
1,3-
propanediol
Continuous and batch microbial fermentations by
Clostridium butyricum and Klebsiella
pneumoniae.
Xiu et al. (2004)
Menzel et al. (1997)
Wang et al. (2001)
Hydrogen
Continuous microbial fermentation by
Enterobacter aerogenes HU-101. Ito et al. (2005)
Catalytic reforming functioning at moderate
temperatures and pressures. Wood (2002)
Steam reforming of glycerol Hirai et al. (2005)
Pyrolysis and steam gasification of glycerol. Valliyappan (2001)
Succinic Acid Microbial fermentation by Anaerobiospirillum
succiniciproducens. Lee et al. (2001)
1,2-
propanediol
Low-pressure hydrogenolysis. Dasari et al. (2005)
Selective hydrogenolysis with Raney nickel
catalyst with hydrogen.
Perosa and Tundo
(2005)
Dihydroxy-
acetone
Chemoselective catalytic oxidation with
platinum metals. Garcia et al. (1994)
Selective oxidation of glycerol with the
platinum-bismuth catalyst. Kimura (2001)
Microbial fermentation by Gluconobacter
oxidants in a semi-continuous process. Bauer et al. (2005)
Polyesters
The reaction of glycerol and adipic acid in the
presence of tin catalysts.
Stumbe and
Bruchmann (2004)
Polycondensation of oxalic acid and glycerol. Alkanis et al. (1976)
The reaction of glycerol and aliphatic
dicarboxylic acids. Nagata et al. (1996)
Polyglycerols Selective etherification of glycerol Clacens et al. (2002)
Polyhydroxy
alkanoates
Fermentation of hydrolysed liquid phase of
glycerol and whey permeate by the osmophilic
organism.
Koller et al. (2005)
20
Figure 2.3 shows alternative pathways for the conversion of glycerol into valuable
chemicals. One of the products derived from glycerol is acrolein, that has the
potential to produce acrylic acid (essential monomers for polyester industries),
acrylic acid esters, absorber polymers, and detergents [66-68]. While the dehydration
of acrolein from glycerol has increased to nearly 96% [69], increasing the selectivity
of acrolein, by avoiding the formation of hydrogenated products or coke through
minimising secondary reactions, remains a challenge [68]. Among the catalysts used,
super acid [70] and smaller catalysts provide better yields [71, 72]. The optimal
temperature for acrolein production is approximately 300 ºC [73]. The addition of
oxygen has also been shown to reduce catalyst deactivation and enhance yield [74].
While efforts have been made to utilise glycerol using pyrolysis, steam reforming,
gasification, and chemical conversion, these technological applications often require
pure glycerol, are under development, or not economically viable and have limited
usefulness. Considering the increasing amount of crude glycerol produced, a
technological application that can be used directly, such as use as a liquid fuel for
generating heat and power in biodiesel plants, would be beneficial.
2.3 Combustion fundamentals and glycerol combustion
2.3.1 Fundamentals of liquid fuel combustion
Combustion is a physical and chemical process that occurs when fuel reacts with
oxygen at high temperatures to produce heat and combustion products [75]. In a
typical combustion process, the fuel (generally hydrocarbon) is oxidised to form
carbon dioxide (CO2) and water (H2O). However, since combustion occurs in air
rather than pure oxygen, the presence of nitrogen can induce the formation of NO×.
Also, since many fuels are more complex than simple hydrocarbon, unburned and
partially burned products can be produced.
21
In the combustion of liquid fuel, information on the fuel composition is generally
limited. Liquid fuel is a mixture of hydrocarbon species. The most common
properties of a liquid fuel can be determined by ultimate analysis and assessment of
the physical properties of the fuel. Ultimate analysis can determine the fractions of
carbon, hydrogen, sulphur, oxygen, nitrogen, and ash that constitute the fuel. Useful
physical properties include specific gravity, viscosity, flash point, and distillation
profiles. Selected important physiochemical properties of liquid fuels are described
below [75]:
Density is the mass-to-volume ratio of fuel at 15ºC and is useful for measuring
the volume of fuel delivered into a combustion chamber. Density is an important
measurement of the caloric value of liquid fuel injected into a combustion
chamber to provide the required output of power.
Caloric value is the amount of heat emitted by complete combustion of a fuel
and is measured in MJkg-1.
Specific heat is the amount of energy required to raise the temperature of 1 kg of
oil by 1ºC. Light oils have low specific heats and heavier oils have higher
specific heats.
Specific gravity is the ratio of fuel density to that of a reference substance,
normally water at 4ºC. Fuels with low specific gravity evaporate easily, and
have lighter fractions and burn more rapidly compared to fuels with high
specific gravity.
Viscosity is the internal resistance of a liquid to flow, depending on temperature.
Fuel with high viscosity is difficult to flow and requires a preheating system for
handling, storage, and satisfactory atomisation during combustion.
22
Vapour pressure is the pressure developed by a vapour in the condensed phases
(solid or liquid) in thermodynamic equilibrium at a given temperature in a
closed system. Types of molecules involved and temperature affect vapour
pressure. The intermolecular force between the molecules determines the vapour
pressure. If the bond is strong, vapour pressure is low and if the bond is weak,
vapour pressure is high. At high temperatures, molecules have enough energy
with which to escape from the liquid or solid, but at a lower temperature, fewer
molecules can escape.
Flashpoint is the temperature to which a fuel must be heated for the fume can be
ignited with an igniter.
Auto-ignition temperature is the minimum reactor/environmental temperature
required to ignite a vapour or gas in air without a flame or igniter.
The Cetane number measures the ignition delay of fuel, ranging from ignition of
methylnaphthalene (standard value of 0) to that of cetane (standard value of
100).
The heat of evaporation is the amount of energy that must be added to transform
a quantity of a liquid into a gas.
During combustion, liquid fuel is sprayed into the combustion chamber.
Atomisation, a process that separates the liquid fuel into droplets through external
force [76], increases the surface area of the fuel and this in turn increases the
evaporation rate and rate at which fuel blends with air [77]. The atomisation of liquid
fuel produces fuel droplets at various sizes, and the size and volatility of the droplets
govern the combustion. Droplet size is determined by fluid mechanical forces and
surface tension that pull and hold the liquid during atomisation [78].
23
Droplets require time to vapourise, in which a raised temperature must be introduced
and latent heat of vapourisation must be supplied to convert the liquid droplet into
the gas phase. Droplet lifetime is the time required for the droplets to vapourise and
varies according to the square of its initial radius. Long droplet lifetime may cause
incomplete combustion or unreacted fuel which leads to inefficiency of the
combustion. Maximum droplet size must therefore be controlled to minimise these
effects [79].
When fuel droplets are mixed with oxygen and injected into a combustion chamber
in the presence of heat, ignition occurs [80]. Ignition can be achieved by elevated
temperature (auto-ignition) or with an external igniter. Heat and pressure
subsequently increase and a flame develops [81]. Two important parameters before
and after ignition are ignition delay and burnout time.
2.3.1.1 Ignition delay
Ignition delay is the time between fuel injection and ignition and a key
physicochemical property of a combustible air mixture. During ignition delay, fuel
vapourises and mixes with air before ignition occurs. The delay links the pre-flame
process and the formation of flame.
The duration of ignition delay is determined by physical and chemical delays [81].
Physical delays occur during atomisation of the liquid fuel, vapourisation and mixing
with air. The quality of fuel atomisation is determined by the properties of the fuel,
such as density, viscosity, and surface tension. These properties are inversely
proportional to the Cetane number, H:C ratio, pour point, and specific combustion
heat [82]. Other physical factors that delay ignition include air temperature and
pressure, engine speed, and combustion chamber design.
24
Chemical delays are caused by slow chemical reactions at the beginning of
combustion [83]. Given that chemical reactions occur more rapidly at higher
temperatures, physical delays become longer than the chemical delays, resulting in
the mixing process determining the ignition delay. During ignition delay, radicals
and molecular products of combustion are formed, e.g water, hydrogen peroxide, and
carbon oxides [82]. Duration of ignition delay can therefore be determined based on
formation of these products. Fuel properties also affect ignition delay. The Cetane
number can be used to determine ignition quality of the fuel. The higher the Cetane
number, the shorter the delay of fuel ignition [81].
Ignition delay can determine the overall course of the combustion process. Prolonged
ignition delay increases fuel quantity, and the vapourised fuel injected into the
combustion chamber, and promotes more efficient mixture of fuel vapour and air,
resulting in explosive combustion followed by a rapid increase in pressure. However,
this type of combustion can cause a jet engine to fail or result in a knock in a diesel
engine. An ignition delay that is too short can lead to incomplete combustion,
formation of smoke, loss of power, and reduction of engine efficiency [82, 84].
2.3.1.2 Burnout time
Liquid fuels, particularly atomised droplets, experience burnout after
ignition. Burnout time is the time required for a droplet to burn from the
moment ignition begins to the end of combustion when the droplets are
completely evaporated and burnt. During combustion of a single droplet,
temperature of the heat absorbed by the droplet changes from the ambient
temperature to the temperature of the flame when the droplet is surrounded
by the flame, showing that the flame is the main heat source for evaporation
and combustion of the droplet [81]. Based on d2-law of droplet burning,
25
burnout time decreases with decreasing droplet size [81]. The smaller the
droplet, the more rapidly combustion occurs because there is less fuel to be
evaporated. Oxygen also affects burnout time, since the more oxygen
available, the shorter the burnout time [85].
2.3.1.3 Burning rate
The burning rate is a measure of the linear combustion rate of a fuel over time [81].
The rate is controlled by the interaction between the rate of loss of fuel mass and the
oxygen. In liquid combustion, mass refers to the droplet sprayed in the combustion
chamber. The loss of mass of the droplet is affected by the availability of oxygen
[86], that is determined by the rate of oxygen diffusion into the fuel surface, as
opposed to the chemical reaction between fuel and oxygen [87]. The burning rate is
also affected by thermal enhancement around the droplets. The higher the air
temperature in the combustion chamber, the higher the burning rate of the droplets
[81]. Fuel properties also affect the burning rate significantly. A simulation result
showed that the burning rate is most sensitive to liquid fuel density, vapour pressure,
and surface tension, due to effects of these properties on atomisation processes [88].
Overall, liquid fuel is advantageous due to its physical properties, but atomisation is
required for combustion. Atomisation can influence combustion through droplet
lifetime and uniformity of droplet size. Key parameters for studying the fundamental
aspects of liquid combustion include ignition delay time, burnout time, and burning
rate, all of which are useful for studying glycerol combustion.
2.3.2 Crude glycerol as a fuel
In the combustion stoichiometry of glycerol, glycerol and oxygen molecules are
broken down and rearranged to form molecules of carbon dioxide and water
(Reaction 2.6).
26
2C3H8O3 + 7(O2) → 6CO2 + 8H2O ............................................................... (R 2.6)
The presence of inert diluent nitrogen in the air would not change the chemistry, and
the coefficients are determined by considering atom conservation. Figure 2.4 shows
the calculated combustion product of glycerol considering the excess of oxygen. The
excess oxygen, related to percentage excess air, is the amount of oxygen that added
during combustion to ensure completion of combustion.
Figure 2.4 Flue gas composition of glycerol combustion.
Glycerol is a medium quality fuel for combustion in an engine [89]. Table 2.5 shows
that glycerol has low energy density (18 MJkg-1), one-third that of diesel (42.6 MJkg-
1) or gasoline (43 MJkg-1). Glycerol is difficult to pump and atomise due to high
viscosity, and tends to form larger droplets than diesel in practical combustion
systems [10]. Glycerol is difficult to ignite due to high auto-ignition temperature,
vapour pressure, and flash point. Despite these difficulties, pure glycerol has been
used as an alternative fuel for furnaces [90] and has been successfully burned in
modified compression ignition engines [11] and burners [91].
0 20 40 60 80 100
0
20
40
60
80
CO2
H2O N2
O2
Mas
s fra
ctio
n of
flue
gas
(% m
ole)
Excess of oxygen (%)
27
Table 2.5 Properties of crude glycerol and pure glycerol
Fuel property Crude glycerol Pure glycerol
Formula n/a C3H8O3
Molecular weight (gmole-1) n/a 92.09
Density (kgm-3) 1205 1261
High heating value (MJkg-1) 14.84 18
Boiling point (K) 403 563
Cetane number ~6.7 5
Heat of Vaporisation (MJkg-1) n/a 0.67
Viscosity (mPa.s) 1100-1300 1500
Auto-ignition temperature (K) 673 796
Flash point (K) 463 450
Specific heat capacity (Jmol-1kg-1) ~2.8 2.43
Thermal conductivity (Wm-1K-1) n/a 0.29
Adiabatic flame temperature (K) n/a 2201
Stoichiometric fuel/air ratio(w/w) n/a 5
Spalding number n/a 4
Combustion of glycerol at low temperatures (270–370ºC) produces the toxic
substance acrolein [11]. However, the amount of acrolein produced in glycerol
combustion is only slightly higher than that produced in natural gas combustion
(13.3ppb). Acrolein concentrations of the combustion of methylated glycerol,
demethylated glycerol, and technical glycerol are 16.5 ppb, <18 ppb, and 20.7 ppb,
respectively. The low acrolein concentration in glycerol combustion is due to the
high oxygen content of glycerol. Glycerol has 1.33 oxygen/carbon ratio, and the high
28
oxygen content suppressed soot formation, and therefore complete combustion can
be expected from glycerol.
Figure 2.5 shows the calculated adiabatic flame temperature of glycerol compared
with that of biodiesel, diesel, methanol, and methane. The maximum adiabatic
temperature that can be achieved by glycerol is lower than that for biodiesel or diesel
due to the low heating value.
Figure 2.5 Adiabatic flame temperatures of glycerol, biodiesel, diesel, methanol,
and methane.
Crude glycerol has the potential to be used as a fuel as it has average energy content
and is inexpensive. Using crude glycerol as fuel could be beneficial for generating
heat and power in biodiesel plants, but economic benefit as a fuel source for industry
is low due to the high cost of transportation. Using crude glycerol as a supporting
fuel for the biodiesel industry would optimise energy integration and save costs.
Crude glycerol is relatively inexpensive compared with other commonly used fuels
[51]. In 2011, the price of crude glycerol was 40–110 US$/tonne, ~20% of the price
of diesel (~$500 US$/tonne) [22]. Biodiesel production requires energy for
29
generating power and steam, and a portion of this energy could be provided by crude
glycerol combustion. The cost of steam generation for an 8000 tonnes/year biodiesel
plant is ~$30,000–$170,000 /year, depending on the feedstock and process used [22].
2.3.3 Glycerol combustion in compression ignition engines
Utilisation of glycerol in both spark and compression ignition engines has been
investigated. The low volatility and high viscosity of glycerol can cause serious
problems in spark ignition engines, but a mixture of glycerol and methanol can
improve performance in these engines [92].
Glycerol is commonly used in compression-ignition (CI) engines (diesel engines),
mostly as an additional fuel, such as oxyfuel and fatty acid glycerol formal ester
(FAGE). Oxyfuel derived from etherification of glycerol with tertbutyl alcohol and
isobutylene is a mixture predominantly comprising higher glycerol ethers [93].
Similarly, the reaction of glycerol with dimethoxymethane and triglyceride converts
glycerol into FAGE in the presence of an acid [93]. The oxyfuel and FAGE are
subsequently blended to form a combination diesel/biodiesel fuel.
Glycerol is an uncommon diesel engine fuel as it potentially causes clogging and
produces acrolein [11]. Stenhede (2008) believes that glycerol cannot be ignited or
burned in petrol or diesel engines [8]. However, more recent studies suggest that
pure glycerol can be successfully used in diesel engines by creating a unique engine
cycle [8]. In the proposed cycle, gas is used to preheat the glycerol, and once the
engine heats up, glycerol can be used exclusively as the fuel [8]. Resulting output of
emissions is low due to the high oxygen content of the glycerol [8]. By contrast, the
combustion of crude glycerol in a diesel engine produces fly ash, possibly derived
from impurities. Little information regarding these impurities is available in the
literature.
30
2.3.4 Glycerol combustion using various burners for utility boilers
Glycerol can be burned in burners or boilers as an exclusive or additional fuel for
power production. As an exclusive fuel, the glycerol flash point should preferably be
maintained below 373K to stabilise the combustion. As an additional fuel, glycerol
can be mixed with another fuel with a viscosity below 200 mm s-1 or alcohol at 3–
40wt%. In both cases, vapour pressure and flash point should be reduced [90].
Figure 2.6 shows the recently developed high-swirl refractory burner for glycerol
combustion [91]. The burner overcomes the challenges related to flame ignition and
glycerol stability during combustion [91] and therefore effectively burns pure
glycerol. Modification of the burner nozzle, notably the flow-blurring liquid fuel
injector, enables glycerol to be burned without the need for preheating [90].
Emission output from glycerol combustion using this burner is low, with low
amounts of NO× and CO, undetectable acrolein, and negligible amounts of
acetaldehyde [94].
Figure 2.6 The 7 kW refractory burner (left) and 82 kW refractory-lined furnace
(right) [91].
31
2.3.5 Glycerol in fuel blends
A fuel can be mixed with a liquid to create an alternative fuel to enhance fuel
performance, e.g. diesel has been blended with jatropha, karanja, mahua, linseed,
rubber seed, cottonseed, and neem oils [95]. Twenty percent of these vegetable oil
blends did not diminish engine performance [96]. Diesel has also been blended with
butanol, methanol, or ethanol to reduce emission [97].
Mixing glycerol with other fuels can improve the combustion characteristics of
glycerol [98]. Combustion of a pure glycerol-propanol mixture was clean, creating
little soot [99]. However, the addition of crude glycerol to fuel is impracticable due
to the resulting decomposition and polymerisation reactions [100]. Further research
into the use of crude glycerol blends as starters as supporting fuels is needed,
particularly regarding performance, emissions, and process optimisation. Research
directed towards improving burner design is also needed to meet the challenges
associated with glycerol combustion.
2.4 Summary and specific research objectives
An increase in biodiesel production has led to a concomitant increase in the
production of crude glycerol. Currently, biodiesel production relies on
transesterification with a homogeneous base catalyst since this process is
economically viable and kinetically efficient. However, this process produces low-
grade glycerol (crude glycerol) that contains impurities. The presence of these
impurities renders the crude glycerol unsuitable for use as feedstock in industrial
applications. Crude glycerol cannot be appropriately absorbed by the current market,
which requires refined glycerol. This has resulted in a drop in the price of glycerol
and environmental problems, e.g. poisonous impurities (methanol). Efforts to utilise
crude glycerol, such as pyrolysis, thermal cracking, gasification, and chemical
32
conversion may be expensive, produce unwanted secondary products, have low
product volume and value, or require large throughputs and may therefore not be
economically viable or practical.
Crude glycerol may potentially be used as cheap liquid fuel for heat and power
generation in biodiesel plants. Combustion of glycerol in compression ignition
engines and burners for boilers, and the blending of glycerol with other fuels, have
been investigated. Though glycerol can be used as a fuel, pure product is needed, and
the presence of impurities limits the utilisation of crude glycerol. Little is known
about the influence of these impurities on crude glycerol combustion, therefore
understanding the combustion characteristics of crude glycerol is essential if this by-
product is to be utilised as fuel in practical combustion systems.
The objective of this research is to investigate the effects of the main impurities,
including water, methanol, soap, and biodiesel, on the ignition and combustion
characteristics of crude glycerol. This research will provide a basis for the
development of technological applications for the combustion of crude glycerol and
therefore aims to:
Investigate the differences between the ignition and combustion characteristics
of crude glycerol and pure glycerol.
Compare combustion of crude glycerol with that of well-known fuels, e.g
biodiesel, diesel, and ethanol.
Systematically investigate the effects of the main impurities, including water,
methanol, soap, and biodiesel, on the ignition and combustion of crude glycerol.
Develop an understanding of the chemical kinetics of glycerol combustion in
both the presence and absence of impurities, using the biodiesel and methanol
models since these have both been widely investigated. It is anticipated that the
33
results of this research will enhance the understanding of the behaviour of the
impurities in the ignition and combustion of glycerol and provide a kinetic study
of combustion of a mixture of methanol-glycerol and biodiesel-glycerol.
34
Chapter 3 Research methodology, approach and
techniques
In this chapter, details of research methodologies, approaches and techniques
employed to achieve the objectives are provided, including descriptions of
experimental and modelling techniques, and experimental facilities.
3.1 Overview of research strategies
The experimental program focused on the effects of impurities on ignition and
combustion of a single droplet of glycerol and kinetic modelling of glycerol
impurities. Interrelated components of the research strategies employed in the
experimental study are represented schematically in Figure 3.1.
The presence of impurities differentiates crude glycerol from glycerol. Although
analysis of crude glycerol has provided information regarding inherent impurities,
evidence that these impurities influence combustion characteristics of glycerol is
lacking. To determine potential effects, a series of investigations was conducted on
crude glycerol combustion and comparisons were made with combustion of diesel,
biodiesel, and ethanol.
Analysis of crude glycerol showed the main impurities to be water, methanol,
biodiesel, and soap. The effects of each of these impurities were investigated further
using the single droplet combustion method, to determine the fundamental properties
of ignition and combustion characteristics of glycerol. The single droplet combustion
method is well-established method [101] but has limitations and these are discussed
below. The kinetic study provides a theoretical approach to confirm the experimental
35
data on ignition. The kinetic study was done for the well-established kinetic model,
including glycerol, methanol, and biodiesel.
Figure 3.1 Detailed experimental design and network
3.2 Single droplet combustion
3.2.1 Principles of single droplet combustion
Single droplet experimentation is a well-established method for studying the
fundamental aspects of combustion [102]. The technique evolved from transient
combustion to microgravity single droplet combustion to reduce buoyancy [103,
104]. This method has been applied to combustion of liquid/liquid fuel, i.e. single
ethanol/octane droplet [105], combustion of alkaline and kerosene [106], and
combustion of solid/liquid fuel [107].
Pure glycerol Crude glycerol
Glycerol utilisation
CombustionWater Methanol Biodiesel Soap
Mechanism of impurities effecting glycerol
combustion
Kinetic model Single droplet combustion
Glycerol, methanol, biodiesel combustion model
Kinetic of binary mixture of glycerol-methanol and glycerol-biodiesel
Kinetic of mixture of glycerol-methanol-biodiesel
Ignition and combustion mechanism
Ignition delay
Burnout time
Droplet size
Combustion rate
Determination of sodium in flame using flame spectrometry
Determination of solid residue
Crude glycerol properties
Crude glycerol impurities
Comparison of crude glycerol combustion to pure glycerol and other fuels
36
The single droplet combustion technique simulates the behaviour of a fuel droplet in
an engine during combustion by burning a single droplet in a controlled
environment. The technique was developed from observation of the phenomenon in
a combustion chamber, e.g., diesel engine, where fuel sprayed into the combustion
chamber formed small droplets that later reacted with oxygen. The assumption was
that the droplets behaved uniformly during combustion, and that investigation of a
single droplet would be beneficial for understanding the combustion properties of the
fuel.
The evaporation and combustion mechanism of a single droplet of fuel in a furnace
is illustrated in Figure 3.2 [81], showing that the droplet (Rd) evaporates radially
outwards. In the evaporation zone, the mass fraction of the droplet (YF,S) decreases
from the surface of the droplet to the reaction zone, and conversely, the fuel vapour
temperature increases from the droplet surface (Ts) to the flame temperature (Tf).
These phenomena occur because of the difference in temperature between the
droplet and the air inside the furnace. As the droplet heats up, the surface is
gradually evaporated, forming the evaporation zone. The vapours mix with air from
the air at a relatively high temperature. When conditions are ideal, ignition occurs,
and this is indicated by formation of the flame enveloping the droplet (Rf). Outside
the reaction zone (the flame), the oxygen mass fraction (Yo) gradually increases to
the level of air, and the temperature (Yf) decreases to furnace temperature.
3.2.2 Single droplet combustion apparatus
The different stages of single droplet combustion were recorded chronologically by
capturing images with a high-speed camera, and this enabled investigation of the
fundamental phenomena of combustion. Apparatus of the single droplet combustion
system included a furnace, a droplet suspension system, a step motor for delivering
37
the droplet, and a CCD camera equipped with a computer for capturing the images
(Figure 3.3).
Figure 3.2 Schematic a single droplet combustion indicating the radial distribution
of fuel mass and flame.
3.2.2.1 Furnace
The horizontal tube furnace, 600 mm long and 40 mm wide, was equipped with
heating elements wrapped around the tube and covered with fibreglass insulation. At
the centre of the furnace, a thermocouple was attached to the tube to estimate the
temperature. The digital temperature control device was placed on the furnace body
to regulate the heating element according to the thermocouple reading.
Figure 3.4 shows the construction of the furnace. The isothermal zone within the
tube was in the centre of the furnace. The isothermal zone was approximately 300
mm long, and this was long enough to cover the required area for droplet ignition
and combustion. Temperature decreased at the front and back of the tube due to loss
of heat to the environment.
39
Figure 3.6 shows the differences between the actual temperatures and measured
temperatures of the gas as a function of measured gas temperature. The temperature
difference is increase with furnace temperature. However, in the present research, the
temperature used was 1023K and has insignificant temperature different.
Figure 3.5 The temperature gradient of along the furnace
Figure 3.6 Differences between actual and measured temperatures as a function of
measured gas temperature, Tg is gas temperature and Tp is measured
temperature
0 500 1000 1500 2000 2500 30000
50
100
150
200
250
300
350
400
T g-T
p(K)
Tp(K)
41
capture images of the burning droplet to record changes in droplet size. Aided by a
step motor, the droplet was delivered to the centre of the furnace at a linear velocity
of 1 ms-1.
3.2.3 Analysis of ignition and combustion characteristics
Methods used to investigate combustion phenomena, droplet size, ignition delay
time, burnout time, and burning rate are described below.
3.2.3.1 Measurement of droplet size
Droplet size (ds) was measured using a backlit image technique that showed a clear
edge of the burning droplet during combustion. This technique is schematically
represented in Figure 3.7, where the burning droplet is supported by a silicon carbide
fibre. When a droplet (d0) is heated to the furnace set temperature (T), vapourisation
occurs, causing the vapour to diffuse and mix with air. Once the fuel mixed with
oxygen in the present of heat, ignition occurs, and a flame forms around the droplet.
Heat generated by the flame induces evaporation and the droplet continually
decreases in size [81].
Figure 3.7 Schematic representation of a burning droplet
The decrease in droplet size was measured from the droplet at the centre of the
furnace (to) to the end of combustion (tec), by calculating droplet size per frame from
42
the images captured during combustion. Droplet size can be calculated manually by
measuring the diameter of the droplet. However, since approximately 1000 useful
images are produced per experiment, manually plotting droplet diameter is time-
consuming and impractical. Therefore, a Matlab code was developed to measure
droplet size.
Figure 3.8 shows typical data collected by Matlab to measure droplet size. The
greyscale intensity shows the relevant peaks that correspond to the droplet and
flame. The equivalent droplet diameter was calculated using the following equation
[77]:
d = √(dmax)(dmin)23 ........................................................................................... (Eq. 3.1)
Following Eq. 3.1, the equivalent spherical diameter d was equated by measuring the
height of droplet 𝑑max and width of droplet 𝑑min.
The script analyses each frame of the images captured during the single droplet
combustion. The steps taken to approximate the droplet size are described below.
Figure 3.8 Pixel greyscale intensity
1. Selection of the area of the droplet
43
The captured image occasionally has inconsistent background colour, particularly in
the corners, and this may lead to a false reading being taken of the droplet and
background. However, the script allows users to select the droplet area, the relevant
area in which the droplet experiences combustion. When selecting this area,
choosing the working area provided in the script is important, as is reducing the error
of the calculated area.
2. Conversion of the selected area into greyscale
Each image was converted to greyscale and the threshold value technique was
applied to separate the droplet from the background, creating a binary image.
Threshold value refers to the intensity of the image. Figure 3.9 shows that the area
above a selected threshold value will be black, and areas below the value will be
white. Threshold value was set to 200, allowing droplet leeway from the surrounding
light. Modification of the threshold value provides for areas of higher or lower
intensities. When background exposure is low, threshold value has to drop below the
intensity of the droplet.
Figure 3.9 Binary image of droplet
3. Fibre removal
A shape detection tool was used for removing the fibre (Figure 3.10), leaving only
the droplet. By measuring the pixel of the remaining area, the droplet size could be
determined by using an equivalent circular diameter (ECD) and the inbuilt
44
MATLAB tools that measure the properties of the remaining regions in the binary
image.
Figure 3.10 Image of binary droplet separated from fibre
4. Measurement of fibre width
The width of the fibre, 0.142 mm, is a key to obtaining the equivalent unit of
conversion. Measurements found in pixels of the fibre can be converted to mm by
using the ratio of fibre width in pixels to width in mm. Figure 3.11 shows the
separated fibre.
Figure 3.11 Image of filament separated from droplet
5. Droplet size approximation
The droplet becomes ellipsoidal in shape on the fibre. Approximation of the
spherical diameter is shown in Equations 3.2 to 3.4.
Vsphere = Vellipsoid ............................................................................................... (Eq. 3.2)
43π (D
2)3= 43π (L
2× W2× H2) .................................................................................. (Eq. 3.3)
45
where L, W, and H are the length, width and height of the ellipsoid. The droplet
changes shape in the direction of the filament and gravity, therefore L and W are
equal and Equation 3.4 can be reduced to the following:
D = (L2H)13 ............................................................................................................ (Eq. 3.4)
The length and height of the droplet correspond to lengths of the minor and major
axes respectively (Figure 3.12). Droplet size in mm is converted from pixels to mm
using the ratio of the measured filament size (Equation 3.5).
DmmDpixels
= FilmmFilpixels
= 0.142Filpixels
................................................................................... (Eq. 3.5)
Figure 3.12 Major and minor axes of the droplet
6. Detection of the micro-explosion
In the hot air environment, the droplet may swell, experience a micro-explosion and
separate into multiple droplets (Figure 3.13). Since the code cannot distinguish
which droplet to calculate, the user is required to provide the corresponding number
for the droplet. Important properties of the region are subsequently measured and
stored.
46
Figure 3.13 Micro-explosion of the droplet with numbering for user input
3.2.3.2 Ignition delay time, burnout time and total combustion time
The single droplet combustion apparatus recorded the arrival of the droplet at the
centre of the furnace (to), the ignition (ti), and the end of combustion (tec). Figure
3.14 shows a post-process glycerol droplet at the centre of the furnace (a), an ignited
glycerol droplet surrounded by a flame (b), and the end of droplet combustion (c).
The approximation of ignition delay time (ti), burnout time (tb), and total combustion
time will be discussed in the next section.
Figure 3.14 Typical combustion of a droplet showing: a) glycerol droplet at the
centre of the furnace, (b) an ignited glycerol droplet surrounded with a
flame, (c) the end of droplet combustion
Ignition delay time, burnout time, and total combustion time of a burning droplet
were calculated from the images taken by the CCD camera that had an exposure time
1
2
47
of 1000 µs (60 fps) (Figure 3.14). Ignition delay time (ti) was the difference in time
from the moment the droplet arrived at the centre of the furnace to the moment the
flame first appeared. Burnout time (tb) was the time from ignition until completion of
the combustion process. Total combustion time (tc) was the period from the moment
the droplet entered the furnace until the combustion process was completed.
To investigate the ignition delay time of each mixture, a controlled starting point was
defined as the point at which the droplet first reached the centre of the furnace. The
frames captured until ignition were subsequently counted and converted to seconds
using the appropriate capture rate. The data were normalised against d02 to find an
ignition delay time in smm-2. The size of the droplet at t0 was used to normalise
ignition delay data. The normalised ignition delay time was calculated using
Equation 3.6.
Ignition delayNorm =Frames
Capture Rate×Do2 .............................................................. (Eq. 3.6)
3.2.3.3 Burning Rate
The images of backlit burning droplets were used to determine the burning rate.
Droplets were backlit by the halogen lamp, and image capturing speed was 200
frames per second during the combustion. Combustion time was determined from
the image frame rate. The burning rate (𝑘) was determined using the d2-Law of
droplet combustion by measuring the rate of droplet size reduction during the
combustion process [80]:
2( )sd dkdt
............................................................................................................. (Eq. 3.7)
48
Figure 3.15 The d2 behaviour of droplet vapourisation and combustion [81]
3.2.3.4 Flame emission spectroscopy
Flame emission spectroscopy, that uses single droplet combustion apparatus and a
spectrometer, quickly and efficiently determines the presence of sodium ions in the
flame (Figure 3.16). The furnace and droplet suspension system are described in the
previous section.
Figure 3.16 A schematic representation of single droplet combustion
experimentation apparatus equipped with flame emission spectrometer
The Black-Comet-XR spectrometer, produced by StellarNet, measured spectra in the
wavelength range from 400 to 750 nm and the resolution was 1.5 nm. Spectral
energy was transmitted to the monochromators through a fibre optic cable terminated
by a collimator. A 2048 pixel CCD sensor served as the detector of the spectrometer.
49
Sodium signals were recorded by the spectrometer from the arrival of the droplet at
the centre of the furnace until completion of combustion. Software used to control
the spectrometer was Spectra Wiz, with Episodic Data Capture (EDC) acquisition
mode. The EDC saved spectral over time, allowing some episodes/runs to be saved
continuously. The episodes were saved per 10 µs to ensure high precision spectral
imaging. At least ten experiments were carried out for each sample to ensure
repeatability.
3.2.3.5 SEM-EDS analysis
The SEM-EDS was used to analyse the solid deposit left on the tip of the silicon
fibre during single droplet combustion. The sample for SEM-EDS analysis was
prepared by loading a small quantity of the solid deposit onto an aluminium stud.
The sample was coated with 4 nm of platinum to reduce charging. The SEM-EDS
images were captured using a Zeiss 1555 VP-FESEM scanning electron microscope
operating at 15 kV.
3.2.4 Limitations and errors
Two main limitations of the experimental work include physical properties of the
single droplet combustion method and the potential error in the calculation.
3.2.4.1 Limitation of the single droplet combustion method
In a combustion chamber, i.e. diesel engine, fuel is sprayed to create smaller droplets
to increase the relative surface area. However, physical properties of the single
droplet combustion apparatus limit the ability of the apparatus to mimic the original
droplets in the chamber. Limiting physical properties include droplet shape and size,
the fibre, and the pressure. In the single droplet combustion method, droplet shape is
assumed to be spherical through neglecting buoyancy and gravitational force.
50
However, in the experiment, buoyancy occurred, and the droplet shape was oval, not
spherical.
Droplet size and velocity in the combustion chamber of the single droplet
combustion method present a physical limitation. Droplet size in the combustion
chamber is assumed to be constant and therefore in the single droplet combustion
method, a single droplet is collected for behaviour investigation. However, the
droplets undergoing atomisation vary in size. Droplet size was normalised in the
experimentation to reduce this limiting effect. During atomisation in an engine, the
droplet has a velocity inside the combustion chamber. However, in the single droplet
combustion method, the droplet was attached on the fibre.
The furnace is maintained at a normal pressure of 1 atm. This pressure may differ in
a diesel engine, but the condition is necessary to obtain the expected image history.
3.2.4.2 Potential errors in the droplet analysis script
The script is beneficial in allowing users to rapidly analyse images but some
limitations exist and these are described below.
1. Shape irregularity
The code assumes that the droplet is ellipsoidal but this is not always the case. The
heating effect of the air often changes the shape of the droplet, especially if the fuel
is a mixture of molecules with greatly different boiling points. If this is the case, the
liquid with the lowest boiling point will evaporate more readily, and if the liquid is
trapped inside the droplet for a short length of time, the droplet may ‘explode’, as per
the theory of micro-explosions. In this instance, the droplet undergoes a massive
swelling and sudden bursting (Figure 3.17).
51
Figure 3.17 Shape irregularities from droplet swelling
In this scenario, the droplet is clearly not ellipsoidal and derivation of the equivalent
spherical diameter is no longer valid, but this does not present a problem in
determining the burning rate or droplet behaviour. Shape irregularity typically occurs
before ignition, where the trend of the size rather than the exact value of the squared
diameter of the droplet is important, and is only marginally affected and can be read
easily from the displayed data. If shape irregularity does occur in the combustion
stage, then the d2 law of single droplet theory does not apply as the data are no
longer linear, and the burning rate cannot be determined, proving once again that the
trend, that is unaffected, is the only important aspect.
2. Reflected light
The light can reflect off the surface of the droplet and increase the intensity in that
area and hence the binary image does not recognise that part of the droplet. The three
possible outcomes (Figure 3.18) are described below.
a) The droplet is recognised as one droplet, the software fills in any holes, and the
final droplet measurements are accurate as the whole droplet is being measured.
b) The reflected light obscures the edge of the droplet, and a crescent shape is cut
out of the resultant measured shape, thus reducing the calculated droplet size.
1 2 3
4 5 6
52
c) The droplet is identified as two droplets, and the user is asked to identify which
droplet is being measured.
Figure 3.18 The reflected lights occur on the droplet surface where 1) the greyscale
image of the droplet, 2) the binary image from thresholding 3) the
binary image after filling techniques have eradicated holes; and 4) the
final measured droplet
As a result, the script measures the incorrect droplet size. However, this sometimes
reflects on the droplet size plot where there is a significant reduction in droplet size.
A manual test is required to verify the outlier data.
3. Error from filament width calculation
Calculation of the filament is important for determining droplet properties. The
filament calculation was made from the initial image that was in focus. Image
selection is important as all the properties are dependent on the measurement. Any
error resulting from an unstable or blurred image will be transferred to every droplet.
This value is used in the conversion of the measurements from pixels to mm using
Equation 3.5.
53
3.3 Single droplet combustion of glycerol
The ignition and combustion behaviour of glycerol including and excluding
impurities was studied using the single droplet combustion technique. Combustion
behaviour of crude glycerol was compared with that of pure glycerol, diesel,
biodiesel, and ethanol, to provide a clearer perspective of the usability of crude
glycerol in an engine since glycerol is not a common fuel. The effects of each of the
main impurities of crude glycerol were investigated separately, and related ignition
and combustion characteristics including ignition delay time, burnout time, and
burning rate, were investigated.
3.3.1 Comparison between ignition and combustion of crude glycerol and that
of biodiesel, diesel, and ethanol
Crude glycerol combustion was investigated and compared with that of pure
glycerol, biodiesel, diesel, and ethanol. Pure glycerol and ethanol were acquired
from Sigma-Aldrich (Perth Australia), crude glycerol and biodiesel were provided by
Wilmar International (Gresik, Indonesia), and petroleum diesel (Caltex No.2) was
obtained from a local service station. The properties of these fuels are shown in
Table 3.2.
The effects of the main impurities were studied using the single droplet combustion
technique. Crude glycerol contains 22.9–63v% free glycerol, 9.9–19.2v% methanol,
7.2–34.5v% water, 0–2.6wt% soap, 0–2.8wt% biodiesel, 0–7.0wt% glycerides, 0–
3.0wt% free fatty acids (FFA), and 2.7–3wt% ash [9]. Impurities investigated
include water, methanol, biodiesel, and soap.
54
Table 3.2 Thermodynamic properties of crude glycerol compared to those of other
fuels
Fuel property
Fuels
Crude
glycerol*
Pure
glycerol
Biodiese
l Diesel
Ethano
l
Density @293K (kgm-3) 1205 1261 878 820-
850 789
High heating value (MJkg-1) 14.84 18 42.2 44.8 29.7
Boiling point (K) 403 563 588–630 433–
633 350
Cetane number ~6.7 5 46–51.7 47.4–
63.9 12
Heat of vapourisation
(MJkg-1) n/a 0.67 0.30 0.34 0.92
Viscosity @313K (mPa.s) 1100–
1300 1500
3.36–
3.68
2.2–
2.6 1.14
Auto-ignition temperature
(K) 673 796 646 588 638
Flash point (K) 463 450 422 328 282
Specific heat capacity
(J mole-1 kg-1) ~2.8 2.43 2.2 1.8 2.72
Thermal conductivity
(Wm-1K-1) n/a 0.29 0.15 0.12 0.20
Adiabatic flame temperature
(K) n/a 2201 2291 2413 2078
Stoichiometric fuel/air ratio
(w/w) n/a 5 11 12 9
Spalding number n/a 4 9 8 3
Vapour pressure at 323K (Pa) n/a 0.03 0.047 135 54800
55
3.3.2 Effect of water on glycerol combustion
The effect of water on glycerol combustion was investigated by adding pure
glycerol, purchased from Sigma-Aldrich, with distilled water. The distilled water
was added to pure glycerol at 5–20% by volume. The addition of water to the pure
glycerol was based on the proportion of water occurring in crude glycerol.
3.3.3 Effect of methanol on glycerol combustion
The effects of methanol on droplet combustion were investigated by adding
methanol to pure glycerol. The pure glycerol and methanol were purchased from
Sigma-Aldrich. Methanol was added to glycerol at 5%, 10%, and 15% by volume
according to the quantity of methanol in crude glycerol, ranging from 2.2% to 12.6%
by volume [9]. A high proportion of methanol-to-glycerol was intentionally used to
enhance the effects of methanol in glycerol combustion due to difficulties of
capturing the dim flame characteristics of methanol using the camera.
3.3.4 Effect of biodiesel on glycerol combustion
Biodiesel was added to pure glycerol at 1–3% by volume, according to the quantity
of biodiesel occurring in crude glycerol. Since biodiesel and glycerol are immiscible,
Tween 40 was added to the mixture as an additional surfactant at 0.1% v/v in each of
the glycerol-biodiesel samples. The mixture was homogenised at 1500 rpm for 10
minutes. During the single droplet combustion tests, a stirrer which consistently runs
at 300 rpm was also used to maintain the homogenous nature of the pure glycerol-
biodiesel mixture sample.
3.3.5 Effect of soap on glycerol combustion
Sodium hydroxide (NaOH) is used as a base catalyst for biodiesel production and
results in the formation of soap in crude glycerol. In the trans-esterification process,
56
excess NaOH reacts with free fatty acid from oil or fat in the presence of water,
producing soap as an unwanted by-product. The content of soap in the crude glycerol
can be up to 20wt% [9].
Pure glycerol used in the experiment was purchased from Sigma-Aldrich, and soap
was made as follows: 5.08 g NaOH was dissolved in 15 g water and the mixture was
poured into a flask containing 8.5 g olive oil that was heated for 15 minutes to 60 º
C. The general formula of the soap was NaCOOR1, where R1 was the fatty acid from
the olive oil, the compositions of which are listed in Table 3.3. The soap was added
to the pure glycerol at concentrations of 1%, 3% and 5% by weight. The
corresponding sodium contents in the mixtures were 0.1, 0.3, and 0.5wt%,
respectively.
Table 3.3 The compositions of the olive oil used in the present study
Fatty acid in olive oil
Concentration (%)
Lauric acid C12 saturated 0
Myristic acid C14 saturated 0
Palmitic acid C16 saturated 11
Stearic acid C18 saturated 2
Oleic acid C18 monounsaturated 78
Linoleic acid C18 di-unsaturated 10
Linolenic acid C18 tri-unsaturated 0
3.4 Kinetic studies of mixtures of glycerol with methanol and
biodiesel
Limited availability of kinetic modelling narrowed investigation of the kinetics of
selected impurities on glycerol combustion to the glycerol-methanol and glycerol-
57
biodiesel mixtures. The chemical model was based on research from the literature
and modelling was performed using Chemkin Pro software.
3.4.1 Chemkin Pro and kinetic modelling
To understand the kinetic effects of methanol and biodiesel on the ignition and
combustion characteristics of glycerol, detailed kinetic modelling was undertaken,
based on information in the literature. Chemical kinetic modelling was performed
using Chemkin Pro, commercial software developed by ANSYS Release 15151 (18
January 2016) with licence number 428841.
Combustion is initiated by the production of radical species after a short delay, and
followed by reactant consumption, and an increase in temperature [109]. The OH
radical occurs in most of the significant reactions in the model. Therefore, the time at
which the production of OH radicals peaked was considered the ignition time. The
use of maximum production of OH radicals as ignition time is supported by research
[110].
The simulation of the model was based on the combustion condition in the single
droplet combustion (Table 3.4) and ignition delay was calculated. Ignition delay time
was considered the time required for the mixture to achieve the maximum
concentration of OH. Figure 3.19 is a schematic representation of the ignition delay
as simulated by Chemkin Pro. The model used for computing the ignition delay is
adiabatic model. The simulation also identified the dominant reaction rate, that was
calculated from the consumption of the fuel (methanol and biodiesel), and the
sensitivity analysis based on the temperature gradient. The sensitivity analysis
measured the reactions that had a significant impact on the process if the temperature
increased or decreased.
58
Table 3.4 The simulation condition
Simulation operating condition value
Temperature 1023 K
Pressure 1 atm
Simulation time 0.1 s
Equivalence ratio 1.0
Figure 3.19 Schematic representation of the ignition delay time as simulated by
Chemkin Pro
3.4.2 Reaction mechanisms and kinetic simulations for of glycerol-methanol
mixtures
The glycerol-methanol model was built from a combination of the glycerol kinetic
model with either the methanol or biodiesel kinetic models. The glycerol kinetic
mechanism [53] consists of over 300 species and 7000 reactions. The methanol
combustion mechanism consists of 32 species and 41 reactions [111]. The model
calculated the mole fraction evolution of the species involved in the ignition.
61
Chapter 4 Properties and combustion characteristics of
crude glycerol
The combustion of pure glycerol is inherently challenging due to the properties of
glycerol, such as the high density, viscosity, and auto-ignition temperature.
Combustion of crude glycerol is complicated by the presence of impurities, and these
are costly to remove. Properties of crude glycerol, particularly in the sample received
from Wilmar International, are discussed in this chapter. Ignition and combustion
characteristics of crude glycerol are compared with those of several well-studied
liquid fuels, including biodiesel, diesel, and ethanol. Combustion characteristics
investigated include ignition delay time, burnout time, and burning rate.
4.1 Properties of crude glycerol
Table 4.1 shows a comparison between the properties of crude glycerol from the
Wilmar International sample and vegetable oil [9]. The Wilmar International crude
glycerol sample contained nearly 70% pure glycerol, 23% water, and a small
proportion of other impurities.
The properties of crude glycerol vary according to the transesterification feedstocks
used and processes applied, as discussed in Chapter 2. The impurities can be divided
into two groups, including those from the reactants, catalyst, and by-products of
transesterification as well as side reactions. Transesterification requires oil, alcohol,
and a catalyst, the excess of which becomes impurities, including water from the oil.
Reactions occur between the catalyst (particularly the base catalyst) and oil in the
presence of water from soap.
62
Table 4.1 Properties of crude glycerol
Crude glycerol from
Wilmar International
Crude glycerol from
vegetable oil
Composition (wt%) (wt%)
Free glycerol 68.2 22.9–63
Water 22.6 6.5–28.7
Methanol 2.2 6.2–12.6
Soap 1.3 0–26.5
Fatty acid methyl ester
(FAME) 0.5 0–28.8
Free fatty acids (FFA) 1.8 0–3.0
Ash 3 2.7–3
Elemental Analysis
C 31.85 24–54
H 10.25 10–12
N 0.32 0.3–1.2
Na 0.76 1.1–1.9
P 0.11 0.03–0.2
4.1.1 Water in crude glycerol
Water in crude glycerol originates from oil feedstock and the biodiesel separation
process. Biodiesel manufacturers use cheap oils to reduce the cost of biodiesel
production [35]. The low quality raw materials, such as waste cooking oils, animal
fats and non-edible oils, contain high concentrations of water. Water is also added
during purification processes in biodiesel production to wash the biodiesel, removing
63
catalyst, soap, and traces of glycerol, and this water remains in the crude glycerol
[35].
4.1.2 Methanol in crude glycerol
Methanol is an alcohol that is commonly used in transesterification. Methanol reacts
with oils or fats to form alkyl esters (biodiesel) and glycerol with the addition of
catalyst. Large amounts of methanol are used in transesterification, resulting in high
concentrations in crude glycerol. A 6:1 alcohol-to-oil ratio for the batch process [23]
and a 9:1 ratio for the continuous process are considered optimal for
transesterification [30]. A high methanol-to-triglyceride molar ratio of 42:1 is
common in transesterification using the supercritical methanol method, a
simultaneous process of etherification of fatty acids and rapid transesterification in
the absence of a catalyst [30].
4.1.3 Soap in crude glycerol
Excess alkaline catalyst in transesterification reacts with free fatty acid from oil/fat
in the presence of water to produce soap [114]. The soap is stored in crude glycerol
as a by-product of biodiesel production, where the concentration can be up to 20wt%
[9].
Soap causes problems during separation of biodiesel from crude glycerol, by
decreasing the yield of biodiesel [115]. During separation, soap reduces interfacial
tension and stabilises the emulsion of the water-biodiesel mixture. This results in the
dispersion of glycerol molecules into biodiesel due to the miscibility of the water-
glycerol mixture [116].
64
4.1.4 Biodiesel in crude glycerol
Biodiesel is the main product of transesterification, and the production process is
described in detail in Section 2.3.2. Biodiesel is a renowned fuel substitute for diesel
that consists of straight chain fatty acids ranging from C16 to C18, depending on the
oil feedstock used.
Inadequate separation of biodiesel from crude glycerol during transesterification
results in the presence of unrecovered biodiesel in crude glycerol. Crude glycerol
contains 0–20% biodiesel, depending on the quality of the separation process.
Impurities in crude glycerol, including water, methanol, biodiesel, and soap,
originate from oil/fat feedstock and the transesterification process. Ignition and
combustion behaviour of crude glycerol was compared with that of pure glycerol, to
gain an understanding of the combustion of crude glycerol, and this is described in
the following section. Comparisons were also made with other well-known fuels,
including biodiesel, diesel, and ethanol.
4.2 Ignition and combustion characteristics: comparison of crude
glycerol against pure glycerol, petroleum diesel, biodiesel, and
ethanol
4.2.1 Combustion phenomena of crude glycerol
Figure 4.1 shows typical time-sequenced non-backlit images of the combustion of
droplets of crude glycerol and other fuels. The bright colour represents the flame.
Images from left to right show the combustion sequence, from the moment the
droplets arrived at the centre of the furnace until completion of combustion. At
certain times, particularly concerning pure glycerol and ethanol, the flame was too
65
weak to be visible. In these instances, images were post-processed to identify the
formation of the flame.
Figure 4.1 Typical time-sequenced images of burning droplets of crude glycerol
(CG), pure glycerol (PG), biodiesel (BD), diesel (DS), and ethanol (ET).
Image 1 shows the droplet at the centre of the furnace; Image 2 shows the droplet at
ignition; Images 3 and 4 show the droplets during combustion; and Image 5 shows
the end of combustion when the droplet has evaporated. Upon arrival at the centre of
the furnace (Image 1), droplet sizes were similar. Upon ignition (Image 2), flames
were formed around the droplets. The flame of crude glycerol was bright yellow-
greenish, suggesting that soot particles were formed. Conversely, the flame of pure
glycerol was dim, with fewer soot particles visible compared with the crude glycerol
68
Ignition delay time results from pre-combustion reactions in the fuel/air mixture,
comprising a physical delay (droplet heating, evaporation, molecular diffusion, and
mixing with air), and a chemical delay (subsequent chemical reaction of the gas
phase between fuel and oxygen) [119]. The ignition delay can also be correlated with
the Cetane number of the fuel, with higher Cetane number affording a shorter
ignition delay [120]. The Cetane number of pure glycerol is 5 (Table 3.2), which is
lower than that of biodiesel (51.7), diesel (47.4–63.9), and ethanol (12). The Cetane
number for crude glycerol (6.7) was slightly higher than that of pure glycerol. The
Cetane numbers of these fuels were consistent with the result of the ignition delay
times measured with the single droplet combustion method. Glycerol, with a low
Cetane number, required more time to ignite, whereas the other fuels ignited faster.
The ignition delay time of the crude glycerol was slightly longer than that of pure
glycerol, particularly at higher temperatures, possibly due to the influence of
impurities including water, methanol, sodium, and FAME. Water has the ability to
significantly delay the ignition time of glycerol droplets [85]. Water would evaporate
from the surface of the droplet before glycerol would evaporate, and this could be
attributed to the low boiling point. The water vapour (steam) diluted the glycerol/air
mixture in the gas phase. Therefore, more time would be required for the ignition to
occur when more water had accumulated in the boundary layer of the droplet,
resulting in a delay in the ignition of the glycerol droplets. Within the crude glycerol
droplets, water would slow the increase in temperature by increasing the heat
capacity of the droplets, since the heat capacity of water is 4200 Jkg-1K-1, and this is
much higher than that of glycerol (2400 Jkg-1K-1). Also, water in hydrocarbon fuels
is known to suppress the ignition [121]. Similarly, the water content in the crude
glycerol would delay the ignition, as observed in the experimentation. On the
70
was the shortest among the fuels tested. This suggests that the impurities in crude
glycerol changed the heat of vapourisation of glycerol, particularly biodiesel, which
has half the amount of heat of vapourisation of glycerol. A physical factor, such as
better atomisation due to the presence of water and methanol, may shorten the crude
glycerol burnout time.
4.2.4 Burning rate
Temporal variation curves of droplet size (Figure 4.2) were linearly fit to obtain
burning rates of the five fuels following d2-law. Theoretical burning rates (k) of the
droplets at 1023 K were also calculated, based on the classic d2-law [81]:
,
8ln(1 )g
cp g
k Bc
..................................................................................................... (Eq. 4.1)
,, ( )o c
p g s
Y QC T T
BH
............................................................................................ (Eg. 4.2)
where λg is the thermal conductivity of air, cp,g is the specific heat of the air/vapour
mixture, B is the Spalding number, ρ is the fuel density, Yo, is the ambient oxygen
mass fraction, σ is the stoichiometric oxidizer to fuel mass ratio, T is the ambient
temperature, Ts is the droplet surface temperature, QC is the heat of combustion of
the liquid fuel, and H is the effective latent heat of vaporization.Figure 4.5 shows
consistency between the theoretical and experimental burning rates of the five fuels,
and only small deviations observed. These deviations are expected due to
experimental errors, simplification of the model, complexity of droplet combustion
processes, and effects of the fibre [123]. Figure 4.5 shows that the burning rate of
pure glycerol was lower than those of diesel, biodiesel, and ethanol. With reference
to Equation 4.1, the burning rate was mainly influenced by the density and Spalding
transfer number of the fuel, since the thermal conductivity and heat capacity of the
72
methanol and water) have a higher latent heat of vapourisation than glycerol. Due to
the differences in latent heat, the temperature would increase more rapidly in
glycerol than in methanol and water. However, the vapour pressures of methanol and
water are higher than that of glycerol. Consequently, methanol and water would
vapourise more rapidly than glycerol at the surface of the droplet, leading to the high
burning rate of crude glycerol that was observed.
The other impurities, including FAME, sodium, and ash may also affect the burning
rate of glycerol. For example, the burning rate of FAME is slightly higher than that
of glycerol (Figure 4.5). The addition of sodium to glycerol may enhance the
propensity for soot formation, that will, in turn, affect the burning rate [124].
However, the effects of these impurities on the burning rate would be very limited
since these have very low concentrations in crude glycerol (Table 4.1).
4.3 Summary
The ignition behaviour and combustion characteristics of crude glycerol were
investigated. The ignition delay times, burnout times, and burning rates of single
droplets of crude glycerol, pure glycerol, petroleum diesel, biodiesel, and ethanol
were investigated experimentally and compared. The following conclusions can be
made:
The main impurities of crude glycerol are water, methanol, biodiesel, and soap.
For all the fuels investigated, ignition delay and burnout times decreased, while
the burning rates increased with increasing air temperature. At a given
temperature, ignition delay and burnout times decreased in descending order
from pure glycerol, through crude glycerol, ethanol, and biodiesel, to diesel.
73
Crude glycerol had the highest burning rate, while pure glycerol had the lowest
rate among the fuels studied. Impurities had a profound influence on the
combustion characteristics of crude glycerol.
Burnout times of droplets of the fuels tested decreased in descending order from
pure glycerol, through ethanol, biodiesel, and diesel, to crude glycerol.
Differences in the burnout times could be attributed to properties of the fuel,
particularly latent heat, boiling point, vapour pressure, and micro-explosion.
Impurities changed the combustion characteristics of crude glycerol when compared
with those of pure glycerol. However, the roles of each of the main impurities in
crude glycerol combustion require further investigation, and hence the effects of
water, methanol, biodiesel, and soap will be discussed in Chapters 5–8.
74
Chapter 5 Effect of water
Water is present in crude glycerol, not only because of the natural properties of the
oil or fat, but also from the elementary reactions of alcohol and the catalyst during
transesterification. In this chapter, the effect of water on the ignition and combustion
characteristics of glycerol are discussed.
5.1 Properties of glycerol-water mixtures
Glycerol has three hydrophilic alcoholic groups that form intramolecular and
intermolecular hydrogen bonds, and these are responsible for the solubility of
glycerol in water. In Table 5.1, the relevant physical and thermodynamic properties
of glycerol, water, and mixtures between these are listed. The density, heating value,
and boiling point of the mixture decreases with the addition of water, but thermal
conductivity increases with the addition of water. These variations in properties
changed the combustion behaviour of the water-glycerol mixture, and this will be
described in the next section.
Table 5.1 Key physical and thermodynamic properties of glycerol, water, and
glycerol-water mixtures (GW) [127]
Property Pure glycerol Pure water GW 10% GW 20%
Density (kg m-3) 1261 1000 1235 1208
High heating value (MJ kg-1) 18 0 16.2 14.4
Boiling point (K) 563 373 411 394
Heat of vapourisation (kJ mol-1) 91.7 43.99 n/a n/a
Specific heat capacity (J mol-1kg-1) 221.9 4.18 n/a n/a
Thermal conductivity (mW m-1K-1) 289 608 292 326
75
5.2 The role of water in combustion
Water is not only a by-product, but also actively involved in combustion. In classic
combustion theory, combustion of fuel produces carbon monoxide and water [81]. In
liquid fuel droplet combustion, water initiates and drives internal circulation in the
droplet by enhancing the droplet evaporation due to its lower volatility [125]. Water
could also diffuse back to the droplet surface and decrease the flame temperature,
causing the flame to be extinguished [126].
Water has been studied as a fuel blend of liquid fuel, particularly diesel. Water may
not contain energy that can directly contribute to combustion, but the addition of
water can reduce emissions [128]. Water significantly reduces nitrogen oxides (NO
and NO2), carbon monoxide (CO), black smoke, and particulate matter [128].
Adding 15% water to diesel reduces NO× emissions by up to 35% [129]. The
addition of water to a fuel also increases combustion efficiency. In a water-oil
emulsion, water reduces the interfacial tension, resulting in finer atomisation of the
fuel during injection [130]. Fine droplet dispersion during atomisation increases the
surface area of the droplet for contacting air.
5.3 Single droplet combustion study of a pure glycerol-water
mixture
5.3.1 Combustion phenomena of pure glycerol-water droplets
Figure 5.1 shows the time-sequenced images of burning glycerol droplets with and
without the addition of water at initial droplet sizes (t0) of approximately 1 mm.
Initially, the general ignition and combustion processes for the glycerol with and
without the addition of water were similar. Before ignition (ti), bubbles formed
inside the droplet, causing distortion and expansion of the droplets. The bubbles
77
Figure 5.2 illustrates temporal variations of the squares of the normalised droplet
diameters (d/d0)2 for glycerol droplets with various water concentrations after
ignition. Droplet size decreased linearly with time, implying that combustion
complied with the d2-law [81]. The droplet size reduction shown in Figure 5.2 is
caused by evaporation during combustion process. However, droplet size fluctuated,
and the intensity was stronger when the water concentration in the glycerol
increased. This fluctuation can be attributed to formation of bubbles and collapse of
the droplets (Figure 5.1).
Figure 5.2 Temporal variations of the square of the normalised droplet diameters of
glycerol with different water concentrations
5.3.2 Ignition delay time
Figure 5.3 shows the effect of the addition of water to the ignition delay time of the
glycerol droplets. Ignition delay time increased as more water was added to the
glycerol. Ignition delay time is influenced by droplet heating and evaporation,
78
molecular diffusion, and mixing with air and the subsequent chemical reactions in
the gas phase between fuel and oxygen.
Figure 5.3 Ignition delay time of glycerol droplets with and without the addition of
water, and crude glycerol
Water evaporated at the beginning of combustion before ignition, since the boiling
point is lower than that of glycerol. The water vapour can suppress the chemical
reaction rate of the hydrocarbon-air mixture as this absorbs heat under certain
conditions [121], leading to a longer ignition delay of the glycerol-water droplet
combustion. The endothermic thermal resistance of water vapour form an energy
barrier that prevents the penetration of heat to the surface of the droplet [121]. This
causes a delay in evaporation of the fuel, and later reduces the accumulation of fuel
vapour on the surface of the droplet.
Figure 5.3 shows that the ignition delay time of crude glycerol is similar to that of
pure glycerol. Crude glycerol is known to contain water, that would delay the
ignition. However, crude glycerol also contains other impurities, such as biodiesel
and methanol, that can accelerate the ignition.
PG GW5 GW10 GW15 GW20 CG0.0
0.5
1.0
1.5
2.0
2.5
Igni
tion
dela
y tim
e (s
)
Fuel
79
5.3.3 Burnout time
Figure 5.4 shows the burnout time of glycerol droplets with and without the addition
of water. The burnout time decreased with an increase in water concentration in the
glycerol droplets. However, reduction of the burnout time of the glycerol droplets
was less than 0.2 s from pure glycerol to 20% water in glycerol.
Figure 5.4 Burnout time of glycerol droplets with the addition of different
concentrations of water
The small reduction of burnout time of glycerol droplets with the addition of water
was due to the evaporation of water in the early stage of combustion. The
combustion mechanism of glycerol-water droplets, as explained in Section 5.3.1,
showed that most of the water evaporated prior to ignition, leaving glycerol as the
main component of the droplet. After ignition, the combustion was predominantly of
glycerol, with minimal water in the droplet.
The burnout time of crude glycerol is shorter than that of pure glycerol and water-
doped glycerol droplets. As discussed in Section 4.2.3, burnout time of crude
glycerol was shorter than that of pure glycerol and other fuels due to bubble
PG GW5 GW10 GW15 GW20 CG0.0
0.5
1.0
1.5
Bur
nout
tim
e (s
)
Fuel
80
formation and subsequent micro-explosion that assist atomisation of the crude
glycerol droplets.
5.3.4 Burning rate
Figure 5.5 shows the burning rates of glycerol droplets with different water
concentrations. According to the classic combustion theory of droplets [81], the
burning rate of fuel increases as the latent heat of evaporation, fuel boiling point, and
density decrease [81]. Table 5.1 shows that the addition of water to glycerol reduced
the boiling point, the heat of evaporation, and the density of glycerol. Addition of
water to the pure glycerol resulted in higher burning rates of the glycerol droplets,
and the burning rate increased as the water concentration increased. The addition of
water further enhanced the burning rates of the glycerol droplets by promoting
bubble formation and subsequent micro-explosion of the droplets during combustion
(Figure 5.1).
Figure 5.5 Burning rates of glycerol droplets with addition of different
concentrations of water
PG GW5 GW10 GW15 GW20 CG0.0
0.5
1.0
1.5
2.0
2.5
Bur
ning
rate
(mm
2 s-1)
Fuel
81
The addition of 5–20% of water increased the burning rates of the glycerol-water
droplets. Droplet size and burnout time affect the burning rate. At ignition, droplets
in the water-glycerol mixture decrease in size with the addition of water. This is due
to evaporation of water at the beginning of the combustion process, leaving smaller
glycerol droplet during ignition. The burnout time was also decreasing with water
addition.
The burning rate of crude glycerol is similar to that of the glycerol-water droplets at
10–20wt%. This is possibly related to the water concentration of the crude glycerol,
that is 22.6wt% (Table 4.1). Water comprises the highest proportion of impurities in
crude glycerol, and as such, may have a greater influence than the other impurities,
and this could possibly have affected the burning rate of crude glycerol.
5.4 Summary
The effect of water on the ignition and combustion characteristics of single droplets
of glycerol was investigated experimentally using the suspended single droplet
combustion technique. Water promoted the formation of bubbles within the droplets
before ignition. Bubbles led to the occurrence of micro-explosions that contributed to
the increase of the burning rates and the decrease of the burnout times of the glycerol
droplets. Water also increased the ignition delay time of the glycerol droplets.
The presence of water in crude glycerol can be disadvantageous or advantageous in
combustion. Water does not add energy to combustion; therefore, increasing the
proportion of water in crude glycerol reduces the total energy content of the fuel.
Water also increased the ignition delay time that could impair combustion
performance, such as in an engine. Conversely, the presence of water increased the
burning rate of glycerol and assisted atomisation due to bubble formation and micro-
explosion.
82
Further research of the utilisation of glycerol doped with water, particularly
regarding developing an understanding of the acceptable load of water in glycerol
burned in a diesel engine, is required. Such research will determine the trade-off of
ignition delays with burning rates of the glycerol droplets.
83
Chapter 6 Effect of methanol
As the simplest alcohol, methanol is a clean renewable fuel, and a common source of
energy [131]. Combustion of methanol has been widely investigated, and includes
aspects such as the mechanism of combustion [132], methanol extinction during
combustion [133], activation energy during extinction [126], and theories of
methanol droplet combustion [134]. The utilisation of methanol as a fuel blend has
been investigated, where methanol has been used as an additional oxygen additive in
diesel engines [135].
In this chapter, the effect of methanol in glycerol droplet combustion is discussed.
Methanol is used extensively in the transesterification process for producing
biodiesel, and this result in large quantities of methanol being dissolved in crude
glycerol.
6.1 Properties of glycerol-methanol mixtures
Table 6.1 shows a comparison between the properties of glycerol and methanol.
Glycerol is a thick liquid of high density and burns at a higher temperature than
methanol. Conversely, methanol is a light liquid easily burnt at low temperatures and
it has low density. Methanol and glycerol are miscible [136]. An investigation of
glycerol-methanol as a binary mixture shows that the viscosity and density of the
mixture are temperature dependent [137].
Figure 6.1 indicates that the viscosity of the mixture decreases with increasing
temperature. At a set temperature, the viscosity of the mixture decreases with the
addition of methanol due to the viscosity of glycerol being much higher than that of
methanol.
84
Table 6.1 Chemical components and physical properties of glycerol and methanol
Properties Glycerol Methanol
Molecular formula C3H8O3 CH4O
Molecular weight (g/mole) 92.1 32.04
Density (kg/m³) 1260 791.8
Boiling point (K) 563 327.7 °C
Latent heat of evaporation (kJ mol-1) 91.7 1.3
Figure 6.1 Viscosity of the glycerol-methanol mixture at various temperatures: ▴
90% glycerol; ■ 80% glycerol; □ 70% glycerol; Δ 60% glycerol; and ○
50% glycerol [137]
The density of the mixture followed the same pattern as shown in Figure 6.2.
Glycerol and methanol have different densities with maximum values of 1.260 kg m-
3 and 0.791 kg m-3, respectively. As expected, the density of the glycerol-methanol
mixture decreased with increasing temperature. At low temperatures, the density of
the mixture increased due to a motionless molecule that contributed to the compact
nature of the mix. At constant temperature, the density of the mixture decreased with
85
an increasing methanol proportion in the mixture. The specifications of glycerol and
methanol are listed in Table 6.1.
Figure 6.2 Density of the glycerol-methanol mixture at various temperatures: □ 90%
glycerol; ○ 80% glycerol; Δ 70% glycerol; ■ 60% glycerol; and ▴ 50%
glycerol [137]
6.2 Single droplet combustion study of a pure glycerol-methanol
mixture
6.2.1 Combustion phenomena of pure glycerol-methanol droplets
Figure 6.3 shows the time-sequenced images of burning glycerol droplets with and
without the addition of methanol. Image 1 indicates the moment at which the
droplets arrived at the centre of the furnace; Image 2 shows bubble/expansion of the
droplets; Image 3 shows a micro-explosion; Image 4 shows ignition of the droplets;
and Images 5 and 6 show the course of burning droplets towards completion of
droplet combustion.
Glycerol droplets with methanol added were large upon arrival at the centre of the
furnace than the initial droplet size due to the expansion of the methanol during
insertion (Figure 6.3). Upon arrival at the centre of the furnace, the droplet was
86
expected to grow larger with a subsequent micro-explosion (Images 2 and 3). When
ignition occurred (Image 4), expansion of the droplet ceased, and flames formed
around the droplet. Combustion continued until completion (Images 5 and 6).
Figure 6.3 Typical images of burning glycerol droplets with various methanol
concentrations: pure glycerol (PG), and glycerol-methanol mixture (GM)
Figure 6.4 illustrates the temporal difference of the square of the normalised droplet
diameter (d/d0)2 for pure glycerol with and without the addition of methanol. There is
a variation of the normalised square of droplet diameters, particularly for the
glycerol-methanol mixtures. Variations in droplet size were not linear due to the
micro-explosion of the pure glycerol-methanol droplets. The d2-law was followed
approximately after an initial heating period.
The micro-explosion phenomena of pure glycerol-methanol droplets were similar to
those of the pure glycerol-water droplets, as described in the previous Chapter.
Methanol, that is more volatile and has a lower boiling point than that of pure
glycerol, was trapped within the glycerol droplets during the combustion process.
88
enough to reduce the ignition delay of crude glycerol, but more water was present in
the crude glycerol to delay the ignition. This was due to the low concentration of
methanol in crude glycerol, which was only 2.2wt%.
Figure 6.5 Effect of the addition of methanol on the ignition delay time of glycerol
combustion
6.2.3 Burnout time
Burnout time of glycerol decreased substantially with increasing addition of
methanol (Figure 6.6). Pure glycerol required 1.61 s to burn completely, while with a
5% addition of methanol, the burnout time was only 0.75 s, and the burnout time
decreased with increasing addition of methanol. Short burnout times of the
methanol-glycerol droplet mixtures compared to those of pure glycerol droplets
could be attributed to the micro-explosions, that later increase the burning rates.
During combustion of glycerol-methanol mixtures, micro-explosions occur, which
leads to the disintegration of droplets into smaller ones and therefore the total
burning time is reduced.
PG GM5 GM10 GM15 CG-0.5
0.0
0.5
1.0
1.5
Igni
tion
dela
y tim
e (s
)
Fuel
89
The burnout time of crude glycerol was shorter than that of pure glycerol and
similar to that of the glycerol-methanol mixture (Figure 6.6). The presence of
methanol contributed to the reduction of the burnout time of crude glycerol, with
respect to other impurities contained in the crude glycerol. Droplet expansions and
subsequent micro-explosions occurred in the crude glycerol mixture due to the
difference in the boiling points of the impurities, and this could assist atomisation of
the crude glycerol, leading to a shorter burnout time.
Figure 6.6 Burnout time of glycerol with the addition of different concentration of
methanol
6.2.4 Burning rate
The addition of methanol increased the burning rates of glycerol droplets (Figure
6.7), likely due to the micro-explosion and reduction of the latent heat of evaporation
of the fuel mixtures.
The more methanol added, the stronger the micro-explosion, and this reduced droplet
lifetime two to three times more rapidly [139]. According to the classic combustion
theory of droplets, the burning rate of a fuel increases as the latent heat of
PG GM5 GM10 GM15 CG-0.5
0.0
0.5
1.0
1.5
Bur
nout
tim
e (s
)
Fuel
90
evaporation and boiling point decrease. Methanol has a lower latent heat of
evaporation and boiling point than that of glycerol [140]. Methanol also has the
ability to increase the combustion rate by having a close flame position in droplet
combustion compared to other hydrocarbon fuels in similar combustion conditions
[141], that in turn provide more heat for the droplet from the flame.
Figure 6.7 Burning rates of glycerol with additions of different concentration of
methanol
The improved combustion process could be attributed to the following factors: the
lower viscosity of methanol reduces the overall viscosity of the fuel blend, leading to
a more rapid evaporation, better mixing of fuel vapour and air, and more complete
combustion. In addition, the high oxygen content of methanol helps the blended fuel
to burn more efficiently, thereby increasing the combustion efficiency.
In Figure 6.7, there is also a comparison between the burning rate of crude glycerol
with those of pure glycerol and the glycerol-methanol mixture. Though the
concentration of methanol in the crude glycerol is low, the presence of methanol,
PG GM5 GM10 GM15 CG0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
Bur
ning
rate
(mm
2 s-1)
Fuel
91
along with other impurities, can contribute to the increased of burning rate of crude
glycerol compared to that of pure glycerol.
6.3 Summary
Methanol enhanced the combustion performance of glycerol. Methanol decreased the
ignition delay and burnout times and increased the burning rate of glycerol. This
effect became more marked with increasing addition of methanol. The addition of
15% methanol to glycerol reduced the ignition delay and burnout times to one-fifth
and one-third, respectively, and increased the burning rate fivefold. The addition of
methanol to glycerol droplets resulted in micro-explosions during the combustion
process, and this in turn improved the combustion performance of glycerol.
92
Chapter 7 Effect of biodiesel
In this chapter, the effect of biodiesel on the ignition and combustion of a glycerol
droplet are discussed. As the desired product of transesterification, biodiesel is
separated from the by-product, crude glycerol, at the end of the manufacturing
process. However, a small quantity of biodiesel might be trapped in crude glycerol as
a result of an incomplete separation process [25]. As one of the main impurities,
biodiesel influences the combustion characteristics of crude glycerol.
7.1 Single droplet combustion study of a pure glycerol-biodiesel
mixture
The thermochemical properties of pure glycerol and biodiesel are listed in Table 7.1.
Biodiesel has a lower density, experiences a higher heating rate, and evaporates and
ignites more rapidly than that of pure glycerol. These properties can affect the
combustion characteristics of the mixture of pure glycerol and biodiesel.
7.1.1 Combustion phenomena of glycerol-biodiesel droplets
Figure 7.1 shows the time-sequenced images of burning glycerol droplets with and
without the addition of biodiesel at the initial droplet size of approximately 1 mm. In
Figure 7.1, images from left to right show the droplets from the moment of arrival at
the centre of the furnace until completion of combustion. Image 1 indicates the
moment at which the droplets arrived at the centre of the furnace; Image 2 shows the
moments of droplet ignition; Images 3 and 4 show the course of burning droplets;
and Image 5 shows completion of droplet combustion.
Upon arrival at the centre of the furnace, glycerol droplets with and without biodiesel
added were the same size (Figure 7.1, Image 1). However, droplets of different sizes
93
were observed (Image 2), particularly the pure glycerol-biodiesel droplets. At
ignition (Image 3), flames formed around the droplets.
Table 7.1 Thermodynamic properties of pure glycerol and biodiesel [142]
Glycerol Biodiesel
Density at 293 K (kgm-3) 1261 878
High heating value (MJkg-1) 18 42.2
Boiling point (K) 563 491–715
Latent heat of vapourisation at 373 K (kJkg-1) 706 450
Vapour pressure at 298 K (Pa) 0.03 135
Cetane number 5 51.7
Viscosity at 293 K (mPa.s) 1500 3.36–3.68
Auto-ignition temperature (K) 796 646
Flash point (K) 450 422
Specific heat capacity (Jmol-1kg-1) 2.43 2.2
Thermal conductivity (Wm-1K-1) 0.2900 0.1490
The flame of the pure glycerol droplet was greenish, suggesting a clean combustion,
and a very small quantity of soot particles formed. Oxygen moiety in the glycerol
may promote combustion, resulting in little or no soot formation. At the end of
combustion of the pure glycerol, the fibre tip was free of any soot or ash deposit.
Conversely, in the combustion of glycerol droplets with various additions of
biodiesel, two flame colours were observed, a red flame at ignition and a green flame
at the later stage of combustion. For example, in the case of 1% biodiesel in a pure
glycerol droplet, ignition (Image 3) started with reddish flames. The red colour was
mainly due to the presence of biodiesel in the flame, and colour intensity increased
94
as the biodiesel concentration increased (Image 3: GB2 and GB3). The intensity of
the red flame decreased as time elapsed (Image 5: GB1, GB2, and GB3). The red
flame was replaced by a green flame, indicating the combustion of pure glycerol that
continued until completion of combustion of the droplet.
Figure 7.1 Typical time-sequenced images of burning glycerol droplets and
additions of pure glycerol-biodiesel (GB) at 1023 K
Upon heating, biodiesel evaporated first, followed by glycerol. Biodiesel is more
volatile than glycerol due to the high vapour pressure. The latent heat of evaporation
of biodiesel (250 kJ kg-1) is also much lower than that of glycerol (706 kJ kg-1).
When the droplet was exposed to hot air, biodiesel at the surface of the droplet was
expected to evaporate first. Subsequently, biodiesel trapped inside the droplets would
96
7.1.2 Ignition delay time
Figure 7.3 shows the effect of the addition of biodiesel on the ignition delay of
glycerol droplets. Ignition delay time decreased with increasing addition of biodiesel
to the glycerol. As previously described, ignition was initiated by the combustion of
biodiesel. Figure 7.1 shows that when the ignition criterion was met, biodiesel
burned at the beginning of droplet combustion. Biodiesel has a lower latent heat and
much higher Cetane number (51.70) than pure glycerol (5). A high Cetane number
implies more rapid ignition [143] and explains why the ignition delay time decreased
with increasing addition of biodiesel.
Figure 7.3 also shows the ignition delay time of crude glycerol compared with that of
pure glycerol and the glycerol-biodiesel mixture. Though the presence of biodiesel
can reduce the ignition delay of crude glycerol, this delay remains similar to that of
pure glycerol. This is due to the presence of other impurities and the low
concentration of biodiesel in the particular crude glycerol (0.5wt%).
Figure 7.3 Effect of the addition of biodiesel on the ignition delay time of glycerol
droplets
PG GB1 GB2 PGB3 CG0.0
0.5
1.0
1.5
Igni
tion
dela
y tim
e (s
)
Fuel
97
7.1.3 Burnout time
Figure 7.4 shows the burnout time of glycerol doped with different concentrations of
biodiesel. Burnout time decreased with increasing addition of biodiesel to the
glycerol droplets. This suggests that the addition of biodiesel increased the droplet
evaporation rate due to the decrease in the latent heat of evaporation. Table 7.1
shows that the latent heat of evaporation of biodiesel (450 kJ kg-1) is lower than that
of glycerol (706 kJ kg-1), suggesting that the addition of biodiesel decreased the
latent heat of the pure glycerol-biodiesel mixture.
Figure 7.4 Burnout time of glycerol droplets with additions of different
concentration of biodiesel
Figure 7.4 also shows the burnout time of crude glycerol, compared with that of pure
glycerol and the glycerol-biodiesel mixture. The burnout time of crude glycerol is
much lower than that of pure glycerol. Biodiesel, that can decrease the burnout time
of glycerol, clearly played a role.
PG GB1 GB2 PGB3 CG0.0
0.5
1.0
1.5
Bur
nout
tim
e (s
)
Fuel
98
7.1.4 Burning rate
Figure 7.5 shows the burning rates of glycerol droplets with the addition of different
biodiesel concentrations. The addition of biodiesel was expected to increase the
burning rates of the pure glycerol droplets. Biodiesel, by nature, has a higher burning
rate than that of glycerol. According to the classic combustion theory of droplets, the
burning rate of fuel increases as the latent heat of evaporation, boiling point, and
density decrease [81]. The addition of biodiesel in glycerol was postulated to reduce
the boiling point, heat of evaporation, and density of glycerol. As a result, higher
burning rates of glycerol droplets were achieved with the addition of biodiesel to the
pure glycerol, and the burning rate increased as the biodiesel concentration
increased.
.
Figure 7.5 Effect of the addition of biodiesel on the burning rate of glycerol droplets
Figure 7.5 also shows the burning rate of crude glycerol, compared with that of pure
glycerol and the glycerol-biodiesel mixture. The burning rate of crude glycerol was
much higher than that of pure glycerol, and biodiesel contributed to the
augmentation of the burning rate.
PG GB1 GB2 PGB3 CG0.0
0.5
1.0
1.5
2.0
2.5
Bur
ning
rate
(mm
2 s-1)
Fuel
99
7.2 Summary
The effect of biodiesel on the ignition and combustion characteristics of single
droplets of glycerol were investigated experimentally using the suspended single
droplet combustion technique. In the combustion of pure glycerol-biodiesel droplets,
biodiesel would preferentially evaporate first, leading to ignition of the droplet and
formation of a reddish flame. Subsequently, glycerol evaporated and burned with a
greenish flame. The presence of biodiesel decreased the ignition delay and burnout
times, and increased the burning rates of the glycerol droplets.
100
Chapter 8 Effect of soap
In this chapter, the effect of soap on the ignition and combustion of glycerol are
discussed. Ignition and combustion behaviour of pure glycerol and a mixture of pure
glycerol-soap were investigated using the single droplet combustion method. The
main component of soap is sodium. Sodium is also present in crude glycerol (Table
4.1) and therefore, the release of sodium in the flame and remaining sodium in the
combustion residue are also addressed.
8.1 Single droplet combustion study of a pure glycerol-soap
mixture
8.1.1 Ignition and combustion phenomena of pure glycerol-soap droplets
Figure 8.1 shows typical time-sequenced, non-backlit images of combustion of
droplets of glycerol with and without the addition of soap. The bright colour
represents the flame. Images of the droplets from left-to-right show the sequence of
changes from the moment at which the droplets arrived at the centre of the furnace
until completion of combustion. Upon arrival at the centre of the furnace, the
glycerol droplets with or without the addition of soap were the same size (Figure
8.1). Upon ignition, flames formed that surrounded the droplets. The flame of the
pure glycerol droplet was greenish, suggesting a clean combustion and only a very
small quantity of soot particles formed. Oxygen moiety in the glycerol could
promote combustion, resulting in the formation of no soot, or only a small quantity
of soot. At the end of the pure glycerol combustion, the fibre tip was clean and free
of any soot or ash deposit.
101
Figure 8.1 Typical time-sequenced images of burning glycerol droplets and pure
glycerol-soap droplets (GSx, where x is the soap concentration. Image 1
(t=0) indicates the moment when the droplets arrived at the centre of the
furnace; and Image 2 shows the moment of droplet ignition.
The combustion of glycerol droplets with the addition of soap produced soot and a
solid residue. For example, in the case of an addition of 1% soap to pure glycerol
(GS1), the ignition and combustion phenomena were initially similar to those of pure
glycerol, but the flame was a yellowish colour. The change of flame colour was
mainly due to the presence of sodium ions in the flame and this will be discussed in
the next section. No residue was attached to the fibre tip at the end of combustion
(Image 4). Combustion of 3% and 5% soap in pure glycerol started with yellowish
flames (Image 2) that lasted for a period of time (Images 2 and 3) and subsequently
dimmed and decreased in size (Image 4). After a very short time, the flame became
bright yellow flame (Images 5 and 6). At the end of the combustion, a small amount
of solid residue remained on the fibre tip (Image 7). The residue was likely the ash
102
formed mainly from sodium that had not vapourised and other metals that became
mineralised [10].
These phenomena can be explained by the well-known evaporation mechanism of a
binary mixture with sufficiently different volatilities [118]. The vapour pressures of
glycerol and soap differed significantly [144]. When the glycerol-soap mixture was
heated, the more volatile glycerol component evaporated and burned preferentially.
The less volatile soap component became concentrated in a boundary layer near the
droplet surface. Since the temperature of the droplet surface would remain close to
the boiling temperature of the mixture, and as the surface mass fraction of the soap
would increase, the temperature of the droplet surface would also increase. Towards
the end of the glycerol evaporation, there was a short period during which the droplet
temperature would increase rapidly, and the evaporation rate of glycerol was
extremely low, leading to the flame contraction. Subsequent flame growth was
characterised by evaporation of the soap. Combustion the glycerol droplets with the
addition of soap therefore comprised two stages, namely preferential evaporation and
combustion of glycerol, followed by evaporation and combustion of soap.
Figure 8.2 illustrates the temporal variation of the square of the normalised droplet
diameter (d/d0)2 for droplets of pure glycerol (PG), 5% soap in pure glycerol (GS5),
and pure soap (PS) at 1023 K. Droplet size was monitored from the moment that the
droplet reached the centre of the furnace until completion of combustion. Droplet
size remained almost constant for PG and PS droplets during the ignition delay time
and showed an approximately linear decrease in size after ignition, implying that
droplet combustion conformed to the d2-law [81].
103
Figure 8.2 Normalised temporal evolution of the squared diameter (d/d0)2 for
droplets of pure glycerol, 5wt% soap in glycerol (GS5), and pure soap at
1023 K
However, changes in the droplet size for GS5 resembled those of bi-component
mixtures with vastly differing vapour pressures [11], as discussed in the previous
section. The first segment of these curves represents the first stage during which the
glycerol was evaporated and combusted preferentially, while the second segment
represents the second stage during which the soap was evaporated and burned. Due
to the relatively low soap content, the second stage lasted very briefly.
8.1.2 Ignition delay time
Figure 8.3 illustrates the effect of the addition of soap on the ignition delay time of
glycerol droplets. Ignition delay times remained relatively stable with increasing
addition of soap. As discussed in the previous section, glycerol, that is more volatile
than soap, evaporates and burns at the beginning of the glycerol-soap droplet
combustion, while the soap is concentrated in the droplet surface and burns
afterwards. This suggests that before ignition, glycerol was the main fuel vapour that
104
subsequently oxidised and burned, explaining the similarity between the ignition
delay of pure glycerol and that of the glycerol-soap mixture.
Figure 8.3 Ignition delay time of glycerol droplets with the addition of different
concentrations of soap at 1023 K
8.1.3 Burnout time
Figure 8.4 shows the burnout time of glycerol droplets with and without the addition
of soap. Burnout time decreased with the addition of soap in the combustion of
glycerol droplets. In the combustion of pure glycerol-soap droplets, sodium was
found in the flame and combustion residue, as discussed in Section 8.2.
Sodium has been used as an additive in combustion using gasoline fuel. Sodium is
considered to accelerate combustion by the following: increasing oxidation of fuel
vapour; increasing flame temperature; achieving higher reaction rate; decreasing
valve burning; reducing emissions of hydrocarbon from the exhaust; producing a
cleaner combustion chamber and valve; decreasing varnish on pistons; decreasing
sludge and varnish in crankcase parts and valve covers; and obtaining lower surface
ignition [145].
PG GS1 GS3 GS5 CG0.0
0.5
1.0
1.5
Igni
tion
dela
y tim
e (s
)
Fuel
105
Figure 8.4 also shows the burnout time of crude glycerol, compared with that of pure
glycerol and the glycerol-soap mixture. The burnout time of crude glycerol is shorter
than that of pure glycerol, with soap contributing to the reduction. Soap would
change the evaporation rates of glycerol and subsequent chemical reactions of the
fuel vapours due to the sodium content.
Figure 8.4 Burnout time of glycerol droplets with the addition of different
concentrations of soap at 1023 K
8.1.4 Burning rate
Figure 8.5 shows the effect of the addition of soap on the burning rates of glycerol
droplets at 1023 K. Since the droplets with soap additions exhibited two stages of
combustion, the burning rates of both stages were calculated and are shown in Figure
8.5a and b. Figure 8.5a shows the burning rate of pure glycerol and the first stage of
glycerol-soap combustion. It can be seen that the addition of soap increased the
burning rates of glycerol droplets. Figure 8.5b shows the burning rates of glycerol-
soap droplets in the second stage. The average burning rates of the residues of GS3
and GS5 were similar to that of pure soap, confirming that the combustion of
PG GS1 GS3 GS5 CG0.0
0.5
1.0
1.5
Bur
nout
tim
e (s
)
Fuel
106
glycerol-soap droplets in the second stage was characterised by the burning of soap.
It is also evident that the burning rate of soap was much higher than that of glycerol.
Figure 8.5 Burning rates of glycerol droplets with the addition of different
concentration of soap: (a) Stage 1; and (b) Stage 2
8.2 Sodium ions in the flame of pure glycerol-soap droplets
Figure 8.6 shows the emission spectra of the flames of glycerol droplets with the
addition of different concentrations of soap. The spectra showed all possible light
emissions ranging from 400 nm to 750 nm during the combustion process, and a
clear spectrum with a main feature near 689 nm corresponding with the atomic
107
emission lines of sodium [122]. The intensity of the sodium signal from the flames
of soap-doped glycerol droplets increased with increasing concentrations of soap,
showing that a droplet with a higher soap concentration released more sodium in the
flame. This finding is consistent with the combustion phenomenon discussed in
Section 8.1.1.
Figure 8.6 Changes in signals of sodium in flames of droplets with the addition of
various concentrations of soap in glycerol
Based on the evidence presented above, soap promoted glycerol combustion as
follows: soap started to decompose at ~500 K, that is lower than the boiling point of
glycerol (563 K) [146]; the difference between the boiling points caused the soap to
decompose while glycerol was burning; when the surface temperature reached the
decomposition temperature of soap, the soap decomposed, releasing sodium atoms in
the reaction zone, and this promoted the oxidation of the fuel vapour.
Sodium promotes a higher reaction rate and increases the flame temperature of the
droplet. The higher flame temperature enhances heat transfer to the droplet, leading
to a higher burning rate and shorter burnout time. The effect of sodium in increasing
108
the burning rate is consistent with the literature, that describes atomic sodium
promoting ignition and combustion of hydrocarbons [145].
8.3 Sodium ions in the solid residue
In Section 8.1, a description of how solid residue was found attached to the fibre
after the single combustion was completed, was provided. Since the solid residue
sample from the single droplet combustion was small, the residue of mixture of
glycerol-soap combustion was collected from pool-fire combustion, providing a
sufficient sample for further analysis. The solid residue investigation was carried out
using Scanning Electron Microscopy (SEM) coupled with an Energy Dispersive
Spectroscopy (EDS) for element identification.
Table 8.1 shows the ash and sodium concentrations recovered from the pool-fire
combustion of soap in glycerol. Ash was absent from combustion of pure glycerol,
but present in combustion of soap in the glycerol mixture, where ash production
increased with increasing concentrations of soap added. The addition of 5wt% soap
in the 50 ml mixture of soap in glycerol produced 1wt% ash, and the ash increased to
3.6wt% for combustion of 15wt% soap in glycerol. These results verified that the
quantity of ash produced was directly related to the quantity of soap in glycerol.
As determined by the EDS, dominant elements in the ash included oxygen, sodium,
and carbon (Table 8.1). Sodium accounted for 32–42wt% in ash at various soap
concentrations. However, compared to the amount of sodium in the soap, 78–85wt%
sodium was recovered from the ash, confirming that most of the sodium remained in
the solid residue/ash during combustion. The increase in soap concentration in
glycerol reduced the recovery of sodium in the ash, suggesting that at higher soap
concentrations, more sodium vapourised in the flame, and this was consistent with
the release of sodium in the flame (Figure 8.5).
109
Table 8.1 Ash and sodium in ash produced by combustion of various
concentrations of soap in glycerol
Soap in glycerol
(wt%)
Ash
(wt%)
Ash main component (wt%) Sodium recovery
(wt%)* O Na C
0 0 - - - -
5 1.0 47.3 41.2 9.2 85.2
10 2.0 45.4 42.0 11.1 82.8
15 3.6 41.4 32.8 20.0 78.2
* Approximate value of average sodium found in spot tests of SEM-EDS to sodium added to soap in glycerol mixture
Figure 8.7 SEM images of the surface of the solid residue at various concentrations
of soap
110
Figure 8.7 shows SEM images of an ash particle from 5–15% of soap in glycerol.
Morphology of the solid residue particles varied due to agglomeration, and the effect
increased with increasing concentrations of soap. The addition of 5% soap produced
porous material with relatively larger particles. The 10% addition of soap produced
smaller, more compact particles, and the addition of 15% soap resulted in a sintered
solid residue.
8.4 Summary
An experimental study on the effect of the addition of soap on the ignition and
combustion characteristics of glycerol droplets at air temperature 1023 K was
undertaken. The ignition delay time, burnout time, and burning rate of single
droplets of glycerol with and without the addition of soap were determined, along
with the sodium intensity in the flame. The ignition and combustion process of soap-
doped glycerol droplets occurred in a two-stage manner: in the first stage, glycerol
was evaporated and combusted preferentially; and in the second stage, soap was
evaporated and burned. The two-stage evaporation resembled the evaporation
behaviour of binary mixtures with vastly different volatilities. The addition of soap
to glycerol decreased the ignition delay time slightly and reduced the burnout time of
the droplets. The burning rate of glycerol increased with increasing soap content.
Sodium ions were detected in the flames of the soap in glycerol droplets, and this
promoted the ignition and subsequent combustion of the fuel vapours.
111
Chapter 9 Kinetic modelling of combustion of glycerol-
methanol and glycerol-biodiesel mixtures
In experimental studies of droplet combustion of pure glycerol- methanol and pure
glycerol-biodiesel mixtures, the addition of methanol or biodiesel decreased the
ignition time of glycerol. In this chapter, a kinetic study of the combustion of
methanol and biodiesel in pure glycerol is discussed, to address the reasons for the
decrease in ignition time of glycerol due to the addition of methanol or biodiesel.
Kinetic aspects of the combustion of methanol as a single fuel or fuel blend have
been widely investigated [147]. The early development mechanism of methanol
oxidation included the mechanism developed by Westbrook and Dryer (1979), that
was upgraded by Held and Dryer (1994), using shock-tube and a flow reactor. The
latest development, based on the work of Ranzi et al., (2014), was considered to be
sufficient for hydrocarbon and oxygenated fuel, and was used in the current research
[111]. The kinetic study of a blend of methanol with another fuel, such as gasoline
[148], is of interest to the researcher. However, little information is available
regarding the kinetic study of the glycerol-methanol blend.
The kinetic study of biodiesel combustion has also been extensively explored [149].
Direct modelling of biodiesel is not preferred, due to the complexity of this
modelling. Simple components have been used as surrogate molecules that match the
characteristics of biodiesel, such as methyl butanoate, methyl crotonate, and ethyl
propanoate. However, these simple components do not represent biodiesel well. A
kinetic mechanism for larger molecules, such as methyl stearate and methyl
decanoate, based on the work on the smaller molecules, was used in the current
study.
113
investigation of the effect of pure glycerol-methanol on single droplet combustion, as
discussed in Chapter 6.
Figure 9.2 Effect of additions of various concentrations of methanol on ignition
delay time of pure glycerol-methanol mixture
The main reactions contributing to the consumption of glycerol were identified to
understand the effect of the addition of methanol on the ignition of glycerol ignition.
In Figure 9.3, reaction rates of pure glycerol and the glycerol-methanol mixture are
sorted in decreasing order of relative rates, which defined as the sum of the rate of
production and rate of consumption (the width of the bar indicates the proportion of
the reaction rate), and compared with the chemical reactions involved in the
consumption of pure glycerol (PG) and a 40v% glycerol-methanol mixture (GM) at
ignition. Additional methanol was added to the 40v% glycerol to enhance the effects.
Based on the ignition delay plot in Figure 9.2, the ignition delay time of glycerol,
with the addition of 40v% or more methanol, was similar to the ignition delay time
of pure methanol.
OH radicals contributed to all consumption reactions of glycerol and the glycerol-
methanol mixture (Figure 9.3). Ignition has been reported to occur when OH radicals
0 20 40 60 80 1000.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
Igni
tion
dela
y tim
e (m
s)
Methanol concentration (%)
114
accumulate in the reaction mixture, where most of the radicals react with molecules
of fuel, producing water and heat, and accelerating the fuel oxidation rate [150].
Figure 9.3 Relative rates (indicated by horizontal bar) of the consumption of pure
glycerol (PG) and glycerol-methanol (GM) by various elementary
reactions at residence time 0.1, initial temperature 1023 K, and
equivalence ratio 1.0
Figure 9.4 shows the comparison of mole fractions of various species during ignition
of pure glycerol and the glycerol-methanol mixture. Before ignition occurred,
characterised by the instantaneous rise of OH, there was a shift of the peaks of
species of the glycerol-methanol mixture compared with those of pure glycerol.
The OH mole fraction increased significantly with the addition of 40v% methanol,
and the peak occurred five times earlier than that of glycerol. As with OH, the
addition of methanol also increased the mole fraction of CO2 and CO. The mole
fraction of H2O was twice that of pure glycerol, and a sharp early increase of O and
H was also observed. The addition of methanol therefore increased the production of
those species, and this contributed to decreasing the ignition delay of glycerol.
Reaction 9.1 was the controlling reaction of glycerol-methanol consumption based
on Figure 9.3 and Figure 9.4. In the combustion of pure glycerol, the rate of Reaction
116
OH radicals produced H2O and O. Reaction 9.3 had the highest reaction rate for CO2
formation, where CO reacted with OH, producing CO2 and H. H subsequently
reacted with O2 to produce OH and O, adding more O in the system (Reaction 9.4),
and this was the highest rate of reaction for the consumption of O2.
2OH ↔ H2O + O ............................................................................................. (R 9.2)
CO + OH ↔ CO2 + H ...................................................................................... (R 9.3)
O2 + H ↔ O+ OH ........................................................................................... (R 9.4)
Reactions 9.2 to 9.4 show that the production and consumption of OH releases O in
the system, that later reacts with glycerol (Reaction 9.1) to enhance the combustion
of the glycerol-methanol mixture (Figure 9.2).
9.2 Effect of addition of biodiesel on glycerol combustion kinetics
Figure 9.5 shows the mole fractions of glycerol, biodiesel and OH during
combustion of the glycerol-biodiesel mixture. The mole fraction of glycerol and
biodiesel decreased with increasing time, while the mole fraction of OH peaked once
glycerol and biodiesel had been completely degraded. This occurred when OH
peaked, similarly to the ignition of the glycerol-biodiesel mixture.
Figure 9.6 shows the ignition delay time of the pure glycerol-biodiesel mixture at
various concentrations at an initial temperature of 1023 K and equivalence ratio 1.0.
The addition of biodiesel decreased the ignition delay time of the mixture,
suggesting that adding biodiesel promoted ignition. The ignition delay time of the
glycerol-biodiesel mixture decreased significantly with the increasing addition of
biodiesel, and with the addition of biodiesel over 10v%, the ignition delay time was
close to that of pure biodiesel.
118
Figure 9.6 Effect of additions of various concentrations of biodiesel on ignition
delay time of pure glycerol-biodiesel mixture
Figure 9.7 shows that the addition of biodiesel increased the relative rates of
reactions of O with glycerol to produce OH, C3H6O2 and ACETOL, and the
reactions of H with glycerol to produce H2, OH and C3H6O2. The OH radical
contributed to all of the consumption reactions of glycerol and glycerol-biodiesel.
The increase of O and H (Reactions 9.5 to 9.7) added more OH radicals to the
system that later decreased the ignition time of the glycerol-biodiesel mixture,
highlighting these as key reactions in glycerol-biodiesel ignition.
O + GLYCEROL → 2OH + C3H6O2................................................................. (R 9.5)
O + GLYCEROL → 2OH + ACETOL ............................................................... (R 9.6)
H + GLYCEROL → H2 + OH + C3H6O2 ......................................................... (R 9.7)
0 20 40 60 80 1000.00
0.01
0.02
0.03
0.04
0.05
0.06
Ig
nitio
n de
lay
time
(s)
Biodiesel concentration (%)
119
Figure 9.7 Relative rates (indicated by the horizontal bar) of consumption of pure
glycerol (PG) and 10v% glycerol-biodiesel (GB) by various elementary
reactions at residence time 0.1, initial temperature 1023 K and the
equivalence ratio 1.0
Figure 9.8 shows the comparison of mole fractions of various species (OH, CO2,
H2O, CO, O and H) during the ignition of pure glycerol and the glycerol-biodiesel
mixture. The peaks of these species shift with the addition of biodiesel. The OH
mole fraction increased significantly with the addition of biodiesel before ignition
(Figure 9.8). As with OH, the addition of biodiesel also increased the mole fraction
of CO2, H2O, O, and H at the same time (0.02 s). This shows that the mole fraction
of these species increased during ignition. The mole fraction of CO increased early
in combustion, suggesting that the molecule might control the ignition.
Based on formation Reaction 9.8, the O increased due to the reactions of H with O2
to produce O and OH. Reaction 9.9 shows the formation of H, mainly from the
reaction of CO with O2 to produce CO2. This reveals that the addition of biodiesel to
glycerol can decrease ignition time of the mixture due to the presence of CO.
H + O2 ↔ O + OH ........................................................................................... (R 9.8)
CO + OH ↔ CO2 + H ....................................................................................... (R 9.9)
121
significant reactions for glycerol consumption during combustion. OH played an
important role during glycerol consumption in both glycerol-methanol and glycerol-
biodiesel mixtures. The presence of O in the system contributed to increasing
reaction rates of the key reactions. In the glycerol-methanol mixture, O was released
from the formation of combustion products H2O and CO2, while in the glycerol-
biodiesel mixture, O was released from the consumption of CO.
122
Chapter 10 Evaluation and practical implications
In this chapter, results are evaluated against objectives and implications are
discussed. Results are also compared with the literature, leading to the identification
of knowledge gaps and recommendations for future studies.
10.1 Integration and evaluation of effects of impurities in crude
glycerol combustion
Combustion behaviour of crude glycerol was investigated and compared with that of
pure glycerol, petroleum diesel, biodiesel, and ethanol. Impurities changed the
combustion characteristics of crude glycerol. The roles of each of the known
impurities in glycerol combustion required clarification. The major impurities of
crude glycerol include water, methanol, biodiesel, and soap.
Investigation of the combustion characteristics of glycerol doped with water,
methanol, biodiesel, or soap, provided a better understanding of the behaviour of
these impurities in the combustion of glycerol. These main impurities can be divided
into three groups, based on the effect on the ignition and combustion characteristics
of glycerol:
10.1.1 Impurities that enhance the ignition and combustion of glycerol
Methanol enhanced the combustion characteristics of glycerol by decreasing the
ignition delay and burnout time, and increasing the burning rate of glycerol droplets.
These effects increased with the addition of increasing concentrations of methanol.
The addition of methanol also promoted micro-explosions during the combustion
process, and this in turn improved the combustion performance of glycerol.
123
In the literature, methanol is commonly reported to be used as a fuel additive in
diesel. The addition of 10–15% methanol to diesel fuel increases the performance of
diesel engines, due to the lower viscosity and higher oxygen content. Methanol
reduces the viscosity of the blended fuel, and this improves the spray of fuel
droplets, increasing fuel evaporation, and improving the mixing of fuel and air. The
higher oxygen content of methanol increases combustion efficiency of the blended
fuel [151].
Although methanol improves the combustion characteristics of glycerol, the
concentration in crude glycerol is limited, originating only from excess
transesterification reactants. Furthermore, methanol is recovered and fed back to the
transesterification process to reduce biodiesel production costs [24], leaving low
methanol concentrations in crude glycerol.
The addition of biodiesel enhanced the ignition and combustion characteristics of
single droplets of glycerol. Two flame colours were observed during droplet
combustion due to early combustion of biodiesel, as discussed in Chapter 7.
Biodiesel also reduced ignition time, decreased burnout times, and increased burning
rates of the glycerol droplets. The addition of biodiesel to other fuel, e.g. diesel,
reduced the ignition delay time and added an extra oxygen atom to enhance
combustion of the mixture [152, 153].
However, adding biodiesel into crude glycerol to increase the combustion
characteristics of the mixture has to be viewed from the economic viability. In
transesterification, biodiesel is the product and crude glycerol is the waste. It requires
a careful economic analysis to use the production yield (biodiesel) for waste
treatment.
124
Methanol and biodiesel improve the ignition and combustion characteristics of crude
glycerol, but the improvement is limited by the low concentrations in crude glycerol.
The addition of more methanol or biodiesel to further assist the ignition and
combustion characteristic of crude glycerol may require thorough analysis.
10.1.2 Impurities that partly enhance the ignition and combustion of glycerol
Water improved the ignition and combustion characteristics of crude glycerol by
improving atomisation of the glycerol droplets. The addition of water promoted the
formation of bubbles within the droplets before ignition. These bubbles led to the
occurrence of micro-explosions, that contributed to decreasing burnout times and
increasing burning rates of the glycerol droplets. However, the addition of water also
increased ignition delay time of the glycerol droplets, but this was not favourable
during combustion. Water did not add any energy to the combustion and reduced the
energy content at high concentrations.
In combustion, water can be advantageous and disadvantageous. Figure 10.1 shows
typical fuel atomisation with the addition of water in a diesel engine. Water promotes
secondary atomisation by bubble formation, followed by micro-explosion, as found
in glycerol doped with water [154], and hence the addition of water can increase the
burning rate of fuels [85]. The addition of water in a fuel blend is also beneficial as
the water dilution effect increases the mixing of fuel and air, and reduces the fuel
rich region that in turn reduces soot formation due to the increase of OH radicals
[155].
The disadvantage of an increasing proportion of water in the crude glycerol is
reduction of the energy content of the fuel. Water increased the ignition delay time
that could reduce combustion performance. Therefore, further research on glycerol
doped with water is required, particularly regarding the proportion of water related to
125
the decrease of the ignition delay time. Future work would require developing an
understanding of the acceptable load of water in glycerol which would determine the
balance of a sufficient ignition delay and fast burning rate.
Figure 10.1 Fuel atomisation with the addition of water in fuel [76].
10.1.3 Damaging effects of soap on combustion
Soap changed the ignition and combustion characteristics of glycerol, mainly
because of the sodium content. In the experimentation, the ignition and combustion
of soap-doped glycerol droplets, which occurred in a two-stage manner, decreased
the ignition delay time slightly, and shortened burnout time of the droplets, and this
subsequently augmented the burning rate. The burning rate of glycerol increased
with increasing soap content due to the presence of sodium ions in the flames that
promoted the ignition and subsequent combustion of the fuel vapours.
The drawback of soap is that sodium forms solid residue near the completion of
combustion. This phenomenon is documented in the utilisation of crude glycerol in a
swirl burner [10, 91]. Sodium can impede the fuel injector in a diesel engine due to
the reaction with fuel additives in the presence of water [156].
126
Table 10.1 lists the investigations of sodium in liquid fuel, which are mainly
characterisation of sodium in combustion. Further research is required to determine
how to remove sodium from crude glycerol. Unlike solid fuel where washing could
reduce the sodium content [79], liquid fuel requires a different approach. In some
studies of the transesterification process, the sodium content was decreased by using
a non-metal catalyst, such as enzymatic catalyst. However, the enzymatic catalyst is
expensive and has a low reaction rate, and this is not suitable for industrial
application. Another catalytic method to reduce sodium content involves use of an
immobilised oxide of zinc and aluminium to avoid catalyst loss [157]. These
knowledge gaps represent interesting topics for further research.
Table 10.1 Research on sodium related to combustion
Research topic Reference
Sodium as combustion additive [145, 158]
Characteristics of sodium droplet combustion [159, 160]
Immobilised catalyst to reduce sodium content [157]
Determination of sodium in flame [124, 161, 162]
Collison of sodium atom with glycerol [163]
10.2 Effect of combined impurities in glycerol combustion
Investigation of the effects of individual impurities on the ignition and combustion
characteristics of glycerol was undertaken to provide clarity of how each impurity
affected the glycerol combustion. However, impurities are mixed in crude glycerol
and therefore the effects of combined impurities were investigated.
Mixed impurities were investigated using the factorial design for characterising the
interaction of the known factors. The experiment was designed using Design Expert
127
10.0.5 software that creates a fractional-factorial design for multilevel category
factors. A subset of all possible combinations of the factors of the categories was
chosen. Factors include the impurities water, methanol, biodiesel, and soap. An
amount of 10 ml glycerol was used as a base for the samples. Impurities added
corresponded to the impurities used for the current research: 5–20v% water, 5–15v%
methanol, 1–3v% biodiesel and 1–5v% soap. Based on the factorial design, there
were 45 combinations of samples. The single droplet combustion technique was used
to approximate the ignition delay time, burnout time, and burning rate.
Figure 10.2 shows the ignition delay of glycerol droplets mixed with impurities. The
addition of both methanol and biodiesel decreased the ignition delay of the glycerol
droplets. Comparison of Figure 10.2 (a) and (b) shows that the addition of water
from 5v% to 20v% increased the ignition delay time of the mixture, confirming
findings in the current research that water can delay ignition time. Conversely,
comparison of Figure 10.2 (a) and (c) shows that the addition of increasing
concentrations of soap to the mixture reduced the ignition time of the glycerol
droplets.
Figure 10.2 Ignition delay time of pure glycerol mixed with various concentrations
of methanol and biodiesel with addition of: a) 5% water and 1% soap;
b) 20v% water and 1v% soap; and c) 5v% water and 5v% soap
128
Figure 10.3 shows the burnout time of glycerol droplets mixed with the impurities.
The addition of biodiesel and methanol decreased the burnout time of the glycerol
droplets, and this effect became stronger with an increase in biodiesel concentration
in the mixture. Comparison of Figure 10.3 (a) and (b) shows that the addition of
water from 5v% to 20v% decreased the burnout time slightly. Similarly, comparison
of Figure 10.3 (a) and (c) shows that the increase of soap concentration from 1v% to
5v% decreased the ignition time of the glycerol droplets.
Figure 10.3 Burnout time of pure glycerol mixed with various concentrations of
methanol and biodiesel with addition of: a) 5% water and 1% soap; b)
20v% water and 1v% soap; and c) 5v% water and 5v% soap
As discussed in Chapters 5–8, the addition of impurities increased the burning rate of
the glycerol droplets. Figure 10.4 shows the burning rate of glycerol droplets mixed
with the impurities. The mixture of biodiesel and methanol increased the burning
rate of the glycerol droplets. This effect increased further with the increase of
concentrations of both methanol and biodiesel. Comparison of Figure 10.4 (a) and
(b) shows that the increase of water concentration from 5v% to 20v% was further
increased the burning rate. Figure 10.4 (a) and (c) show that the effect of soap was
more significant than that of water.
129
Figure 10.4 Burning rate of pure glycerol mixed with various concentrations of
methanol and biodiesel with addition of: a) 5% water and 1% soap; b)
20v% water and 1v% soap; and c) 5v% water and 5v% soap
10.3 Practical implications
The current research reveals how the main impurities (water, methanol, biodiesel,
and soap) change the ignition and combustion characteristics of glycerol. These
findings have implications in the practical utilisation of crude glycerol as a fuel in
biodiesel plants. Understanding the effects of water, methanol, biodiesel, and soap in
crude glycerol provides insight into how each of these impurities changes the
ignition and combustion characteristics of glycerol. This insight provides a basis for
a strategy for the utilisation of crude glycerol as a fuel by identifying effects of
impurities that enhance and cause problems in glycerol combustion. This will enable
the use of crude glycerol as a fuel and solve problems arising from crude glycerol as
a by-product of biodiesel manufacturing by converting the glycerol into an energy
source that will benefit the biodiesel industry and environment. However, further
research and improvements are possible and necessary.
130
10.4 Identification of the new gaps
Although the objectives of the current research were achieved, various knowledge
gaps were identified following evaluation of the results and these are described
below.
Water may promote atomisation but delays the ignition of glycerol. Further research
is required to determine the maximum water content in crude glycerol tolerable for
the glycerol to be used as a fuel.
Further research is required to overcome the effect of soap/sodium in crude glycerol
combustion. While other impurities contribute to the combustion of glycerol, soap
(sodium) is the main impurity causing a drawback in the utilisation of crude glycerol
as a fuel. Investigation into decreasing soap content during transesterification is
recommended.
Only methanol and biodiesel were investigated in the kinetic modelling. Kinetic
investigations of the effect of water and the more complex effect of soap have not
yet been undertaken. More importantly, kinetic modelling of a mixture of the
impurities is required to mimic crude glycerol combustion.
131
Chapter 11 Conclusions and recommendations
11.1 Conclusions
Water, methanol, biodiesel, and soap affected the ignition and combustion
characteristics of crude glycerol, and each of these impurities behaved differently
during the combustion of glycerol. Methanol and biodiesel improved the combustion
characteristics of glycerol. Water also enhanced the combustion characteristics of
glycerol but delayed the ignition, and soap had a detrimental effect on combustion of
glycerol due to the sodium content. Detailed conclusions are described below.
Crude glycerol had shorter ignition delay times, more rapid burnout, and higher
burning rates than pure glycerol due to the presence of impurities. The
impurities changed the physical and thermochemical properties of crude
glycerol, by increasing the vapour pressure, reducing the latent heat, and
decreasing the boiling point. Ignition delays of pure and crude glycerol were
longer than those of biodiesel, diesel, and ethanol, leading to lower burning
rates. This occurred because glycerol has a low Cetane number, high boiling
point, and high auto-ignition temperature. The impurities decreased the burnout
time and increased the burning rate of glycerol, particularly when compared
with pure glycerol, biodiesel, diesel, and ethanol.
The effect of each of the main impurities on the ignition and combustion
characteristics of crude glycerol were investigated using pure glycerol mixed
with water, methanol, biodiesel, or soap. The addition of water to single droplet
combustion of glycerol increased the ignition delay time, decreased the burnout
time, and increased the burning rate. Water slowed the ignition time by
evaporating to form steam at the beginning of combustion, and this delayed the
132
heating of the droplet, decreased the droplet temperature, and slowed the
ignition process. However, the burnout time decreased due to the formation of
bubbles inside the droplet, leading to the micro-explosions, and this resulted in
the formation of many smaller droplets, and consequently an increased burning
rate.
Methanol enhanced the combustion performance of glycerol by decreasing
ignition delay and burnout time, and increasing the burning rate of the glycerol.
The formation of bubbles and occurrence of micro-explosions during
combustion improved the combustion performance of glycerol.
Biodiesel reduced the ignition delay and burnout times, and increased the
burning rates of the glycerol droplets. A change in the flame colour was also
observed during combustion. A reddish flame identified at the beginning of
combustion resulted from the early ignition of biodiesel. The flame colour
became greenish towards the end of the process due to the combustion of
glycerol. The ignition delay time of the mixture decreased due to the early
ignition of biodiesel. Burnout time of the glycerol droplets decreased, and the
burning rate increased with increasing addition of biodiesel.
Soap did not significantly affect the ignition delay time, but decreased the
burnout time, and increased the burning rate of single droplets of glycerol. The
increasing concentration of soap also increased sodium intensity in the flame.
The addition of soap decreased the ignition delay and burnout time slightly, and
increased the burning rate, and the mixture had a two-stage combustion process:
a glycerol-dominated combustion stage, followed by a soap-dominated
combustion stage. The ignition delay and burnout time decreased due to the
release of sodium ions in the flames that promoted the ignition and subsequent
133
combustion of the fuel vapours. The increase of soap addition also increased the
sodium intensity in the flame. This subsequently increased the burning rate that
increased with increasing soap content. The two-stage combustion of the
mixture resembled the evaporation behaviour of a binary mixture of two
components with vastly different volatilities.
The kinetic modelling results showed that methanol and biodiesel decreased the
ignition delay time of glycerol. The ignition delay time decreased with
increasing concentrations of methanol or biodiesel. The reaction pathway
analysis showed that ignition was promoted by an early injection of OH radicals.
The reaction rate increased with the presence of O in the system, derived from
the formation of H2O and CO2 in glycerol-methanol mixture, and from the
consumption of CO in the glycerol-biodiesel mixture.
11.2 Recommendations
Although the overall objectives of the present research were achieved, various new
knowledge gaps were also identified following evaluation of the findings from the
current research, leading to the recommendations for future research described
below.
In the singe droplet combustion experiment, the buoyancy effect was not
considered in the present research. Further investigation using single droplet
combustion in microgravity environment to study the buoyancy effect is
required.
Effects of the impurities on glycerol were investigated individually. Interactions
among the impurities in the ignition and combustion of glycerol were examined
statistically but remain largely unexplored. Further investigation is required to
134
gain an understanding of the interactions among impurities in the ignition and
combustion of glycerol.
The present of sodium in the flame was detected from the highest peak of
sodium captured by spectrometer during combustion. However, the present
technique could not detect the amount of sodium at various combustion stages,
which is interesting for further study.
The sodium content in crude glycerol promoted the formation of solid residue,
and this could potentially cause engine damage. Further investigation regarding
the removal of sodium from crude glycerol is required.
Kinetic investigation included only glycerol-methanol and glycerol-biodiesel
kinetic models. Kinetic modelling of other impurities would enhance
understanding of the kinetics of crude glycerol.
Combustion of crude glycerol in actual burners or engines has not been studied.
Investigation on the suitable type of burners or engines, whether using the
existing design or new design is interesting.
Investigation on the effects the impurities of crude glycerol in actual combustors is
essential.
135
References
1. Ramírez-Verduzco L.F., J.E. Rodríguez-Rodríguez, A.d.R. Jaramillo-Jacob,
Predicting cetane number, kinematic viscosity, density and higher heating
value of biodiesel from its fatty acid methyl ester composition. Fuel, 2012.
91(1): p. 102-111.
2. Ciriminna R., C. Della-Pina, M. Rossi,M. Pagliaro, Understanding the glycerol
market. European Journal of Lipid Science, 2014. 116: p. 8.
3. British Petroleum Company, 1981. BP statistical review of world energy.
London: British Petroleum Co.
4. Gutiérrez Ortiz F.J., A. Serrera, S. Galera, P. Ollero, Experimental study of the
supercritical water reforming of glycerol without the addition of a catalyst.
Energy, 2013. 56(0): p. 193-206.
5. Massa M., A. Andersson, E. Finocchio, G. Busca, F. Lenrick, L. R.
Wallenberg, Performance of ZrO2-supported Nb- and W-oxide in the gas-
phase dehydration of glycerol to acrolein. Journal of Catalysis, 2013. 297(0):
p. 93-109.
6. Fernández Y., A. Arenillas, J. M. Bermúdez, J. A. Menéndez, Comparative
study of conventional and microwave-assisted pyrolysis, steam and dry
reforming of glycerol for syngas production, using a carbonaceous catalyst.
Journal of Analytical and Applied Pyrolysis, 2010. 88(2): p. 155-159.
7. Posada J.A., C.A. Cardona, Design and analysis of fuel ethanol production
from raw glycerol. Energy, 2010. 35(12): p. 5286-5293.
8. McNeil J., P. Day, F. Sirovski, Glycerine from biodiesel: The perfect diesel
fuel. Process Safety and Environmental Protection, 2012. 90(3): p. 180-188.
136
9. Hu S., X. Luo, C. Wan, Y. Li, Characterization of crude glycerol from
biodiesel plants. Journal of Agricultural and Food Chemistry, 2012. 60(23):
p. 5915-5921.
10. Bohon M.D., B. A. Metzger, W. P. Linak, C. J. King, W. L. Roberts, Glycerol
combustion and emissions. Proceedings of the Combustion Institute, 2011.
33(2): p. 2717-2724.
11. Steinmetz S.A., J. S. Herrington, C. K. Winterrowd, W. L. Roberts, J. O. L.
Wendt, W. P. Linak, Crude glycerol combustion: Particulate, acrolein, and
other volatile organic emissions. Proceedings of the Combustion Institute,
2013. 34(2): p. 2749-2757.
12. Linstrom P.J., W.G. Mallard, Eds., NIST Chemistry WebBook, NIST Standard
Reference Database Number 69, National Institute of Standards and
Technology, Gaithersburg MD, 20899, doi:10.18434/T4D303, (retrieved
January 18, 2018).
13. Pagliaro M., R. Ciriminna, H. Kimura, M. Rossi, C. Della Pina, From glycerol
to value-added products. Angewandte Chemie International Edition, 2007.
46(24): p. 4434-4440.
14. Pagliaro M, M. Rossi, The future of glycerol: new usages for a versatile raw
material, The Royal Society of Chemistry, 2008. Cambridge, UK, p. 124.
15. Gesslein B.W., Humectant in personal care formulation: A practical guide, in
Conditioning Agent for Hair and Skin, R. Schueller and P. Romanowski,
Editors. 1999, Marcel Decker: Basel, Switzerland. p. 95.
16. Wang Z., J. Zhuge, H. Fang, B. A. Prior, Glycerol production by microbial
fermentation: A review. Biotechnology Advances, 2001. 19(3): p. 201-223.
137
17. González Benito G., M. Ozores, M. Peña, Continuous glycerol production in a
packed-bed bioreactor with immobilized cells of Saccharomyces cerevisiae.
Bioresource Technology, 1994. 49(3): p. 209-212.
18. Modig T., K. Granath, L. Adler, G. Lidén, Anaerobic glycerol production by
Saccharomyces cerevisiae strains under hyperosmotic stress. Applied
Microbiology & Biotechnology, 2007. 75(2): p. 289-296.
19. Overkamp K.M., B.M. Bakker, P. Kötter, M.A.H. Luttik, J.P. van Dijken, J.T.
Pronk, Metabolic Engineering of Glycerol Production in Saccharomyces
cerevisiae. Applied and Environmental Microbiology, 2002. 68(6): p. 2814-
2821.
20. Zhuge J., H-Y Fang., Z. X Wang, D-Z Chen, H. R Jin, H. L.Gu, Glycerol
production by a novel osmotolerant yeast Candida glycerinogenes. Applied
Microbiology and Biotechnology, 2001. 55(6): p. 686-92.
21. Chow Y.Y.S., S.J.M. Goh, Z. Su, D.H.P. Ng, C.Y. Lim, N.Y.N. Lim, H. Lin,
L. Fang, Y.K. Lee, Continual production of glycerol from carbon dioxide by
Dunaliella tertiolecta. Bioresource Technology, 2013. 136(0): p. 550-555.
22. Quispe C.A.G., C.J.R. Coronado, J.A. Carvalho Jr, Glycerol: Production,
consumption, prices, characterization and new trends in combustion.
Renewable and Sustainable Energy Reviews, 2013. 27(0): p. 475-493.
23. Tan H.W., A.R. Abdul Aziz, M.K. Aroua, Glycerol production and its
applications as a raw material: A review. Renewable and Sustainable Energy
Reviews, 2013. 27(0): p. 118-127.
24. Tomes D., P. Lakshmanan, D. Songstad, B. Moser, Biodiesel Production,
Properties, and Feedstocks, 2011, Springer New York, pp. 285-347.
138
25. Meher L.C., D. Vidya Sagar, S.N. Naik, Technical aspects of biodiesel
production by transesterification—a review. Renewable and Sustainable
Energy Reviews, 2006. 10(3): p. 248-268.
26. Ma F., M.A. Hanna, Biodiesel production: a review. Bioresource Technology,
1999. 70(1): p. 1-15.
27. Hama S., S. Tamalampudi, A. Yoshida, N. Tamadani, N. Kuratani, H. Noda,
H. Fukuda, A. Kondo, Process engineering and optimization of glycerol
separation in a packed-bed reactor for enzymatic biodiesel production.
Bioresource Technology, 2011. 102(22): p. 10419-10424.
28. Robles-Medina A P., A. González-Moreno, L. Esteban-Cerdán, E. Molina-
Grima, Biocatalysis: Towards ever greener biodiesel production.
Biotechnology Advances, 2009. 27(4): p. 398-408.
29. Pinnarat T., P.E. Savage, Assessment of noncatalytic biodiesel synthesis using
supercritical reaction conditions. Industrial & Engineering Chemistry
Research, 2008. 47(18): p. 6801-6808.
30. Marulanda V.F., Biodiesel production by supercritical methanol
transesterification: process simulation and potential environmental impact
assessment. Journal of Cleaner Production, 2012. 33(0): p. 109-116.
31. Shahbazi M.R., B. Khoshandam, M. Nasiri, M. Ghazvini, Biodiesel production
via alkali-catalyzed transesterification of Malaysian RBD palm oil –
characterization, kinetics model. Journal of the Taiwan Institute of Chemical
Engineers, 2012. 43(4): p. 504-510.
32. Ottenbacher M., H. Hasse, Continuous three-phase distillation: A process for
separating thermally unstable substances. Chemical Engineering Research
and Design, 2007. 85(1): p. 144-148.
139
33. Thompson J.C., Characterization of crude glycerol from biodiesel production
from multiple feedstocks. Applied Engineering in Agriculture, 2006. 22(2): p.
4.
34. Yang F., M. Hanna, R. Sun, Value-added uses for crude glycerol--a byproduct
of biodiesel production. Biotechnology for Biofuels, 2012. 5(1): p. 13.
35. Atadashi I.M., M. K. Aroua, A. R. Abdul Aziz, N. M. N. Sulaiman, The effects
of water on biodiesel production and refining technologies: A review.
Renewable and Sustainable Energy Reviews, 2012. 16(5): p. 3456-3470.
36. Luo X., S. Hu, X. Zhang, Y. Li, Thermochemical conversion of crude glycerol
to biopolyols for the production of polyurethane foams. Bioresource
Technology, 2013. 139(0): p. 323-329.
37. Liang Y.N., Y. Cui, J. Trushenski, J. W. Blackburn, Converting crude glycerol
derived from yellow grease to lipids through yeast fermentation. Bioresource
Technology, 2010. 101: p. 7581 - 7586.
38. Saenge C., B. Cheirsilp, T. T. Suksaroge, T. Bourtoom, Potential use of
oleaginous red yeast Rhodotorula glutini for the bioconversion of crude
glycerol from biodiesel plant to lipids and carotenoids. Process Biochem,
2011. 46: p. 210 - 218.
39. Pyle D.J., R.A. Garcia, Z. Wen, Producing docosahexaenoic acid (DHA)-rich
algae from biodiesel-derived crude glycerol: effects of impurities on DHA
production and algal biomass composition. J Agric Food Chem, 2008. 56: p.
3933 - 3939.
40. Papanikolaou S., G. Aggelis, Biotechnological valorization of biodiesel
derived glycerol waste through production of single cell oil and citric acid by
Yarrowia lipolytic. Lipid Technologyl, 2009. 21: p. 83 - 87.
140
41. Papanikolaou S., S. Fakas, M. Fick, I. Chevalot, M. Galiotou-Panayotou, M.
Komaitis, I. Marc, G. Aggelis, Biotechnological valorisation of raw glycerol
discharged after bio-diesel (fatty acid methyl esters) manufacturing process:
Production of 1, 3-propanediol, citric acid and single cell oil. Biomass
Bioenergy, 2008. 32: p. 60 - 71.
42. Chatzifragkou A., A. Makri, A. Belka, S. Bellou, M. Mavrou, M. Mastoridou,
P. Mystrioti, G. Onjaro, G. Aggelis, S. Papanikolaou, Biotechnological
conversions of biodiesel derived waste glycerol by yeast and fungal species.
Energy, 2011. 36: p. 1097 - 1108.
43. Kamzolova S.V., A. R. Fatykhova, E. G. Dedyukhina, S. G. Anastassiadis, N.
P. Golovchenko, I. G. Morgunov, Citric Acid Production by Yeast Grown on
Glycerol-Containing Waste from Biodiesel Industry. Food Technol
Biotechnol, 2011. 49: p. 65 - 74.
44. Liu Y., C.M.J. Koh, L. Ji, Bioconversion of crude glycerol to glycolipids in
Ustilago maydi. Bioresource Technology, 2011. 102: p. 3927 - 3933.
45. Rymowicz W., A. Rywinska, W. Gladkowski, Simultaneous production of
citric acid and erythritol from crude glycerol by Yarrowia lipolytic Wratislavia
K1. Chem Pap, 2008. 62: p. 239 - 246.
46. Rymowicz W., A. Rywinska, M. Marcinkiewicz, High-yield production of
erythritol from raw glycerol in fed-batch cultures of Yarrowia lipolytic.
Biotechnol Lett, 2009. 31: p. 377 - 380.
47. Karinen R.S., A.O.I. Krause, New biocomponents from glycerol. Applied
Catalysis A: General, 2006. 306: p. 128-133.
48. Pachauri N., B. He, Value-added Utilization of Crude Glycerol from Biodiesel
Production: A Survey of Current Research Activities, in 2006 ASABE Annual
141
International Meeting, A.S.o.A.a.B. Engineers, Editor. 2006, American Society
of Agricultural and Biological Engineers: Portland, Oregon.
49. Hunsom M., C. Autthanit, Adsorptive purification of crude glycerol by sewage
sludge-derived activated carbon prepared by chemical activation with H3PO4,
K2CO3 and KOH. Chemical Engineering Journal, 2013. 229: p. 334-343.
50. Carmona M., J. L. Valverde, A. Pérez, J. Warchol , J. F. Rodriguez,
Purification of glycerol/water solutions from biodiesel synthesis by ion
exchange: sodium removal Part I. Journal of Chemical Technology &
Biotechnology, 2009. 84(5): p. 738-744.
51. Ayoub M., A.Z. Abdullah, Critical review on the current scenario and
significance of crude glycerol resulting from biodiesel industry towards more
sustainable renewable energy industry. Renewable and Sustainable Energy
Reviews, 2012. 16(5): p. 2671-2686.
52. OECD/FAO, OECD-FAO Agricultural Outlook 2014. 2014: OECD
Publishing.
53. Hemings E.B., C. Cavallotti, A. Cuoci, T. Faravelli, E. Ranzi, A Detailed
kinetic study of pyrolysis and oxidation of glycerol (Propane-1,2,3-triol).
Combustion Science and Technology, 2012. 184(7-8): p. 1164-1178.
54. Laino T., C. Tuma, A. Curioni, E. Jochnowitz, S. Stolz, A Revisited picture of
the mechanism of glycerol dehydration. The Journal of Physical Chemistry A,
2011. 115(15): p. 3592-3595.
55. Lin K-H., W-H. Lin, C-H. Hsiao, H-F. Chang, A.C.C. Chang, Hydrogen
production in steam reforming of glycerol by conventional and membrane
reactors. International Journal of Hydrogen Energy, 2012. 37(18): p. 13770-
13776.
142
56. Yang G., H. Yu, F. Peng, H. Wang, J. Yang,D. Xie, Thermodynamic analysis
of hydrogen generation via oxidative steam reforming of glycerol. Renewable
Energy, 2011. 36(8): p. 2120-2127.
57. Chen G., L. Zhao, Preliminary investigation on hydrogen-rich gas production
by co-steam-reforming of biomass and crude glycerin. International Journal
of Hydrogen Energy, 2012. 37(1): p. 765-773.
58. Kamonsuangkasem K., S. Therdthianwong, A. Therdthianwong, Hydrogen
production from yellow glycerol via catalytic oxidative steam reforming. Fuel
Processing Technology, 2013. 106(0): p. 695-703.
59. Fermoso J., L. He, D. Chen, Production of high purity hydrogen by sorption
enhanced steam reforming of crude glycerol. International Journal of
Hydrogen Energy, 2012. 37(19): p. 14047-14054.
60. Wei L., L. O. Pordesimo, A. Haryanto, J. Wooten, Co-gasification of
hardwood chips and crude glycerol in a pilot scale downdraft gasifier.
Bioresource Technology, 2011. 102: p. 6266 - 6272.
61. European Chemical News, Fuel additive, in Focus on Surfactants. 2005. p. 4-5.
62. Frusteri F., G. Bonura, C. Cannilla, L. Spadaro, O. Di Blasi, Catalytic
etherification of glycerol by tert-butyl alcohol to produce oxygenated additives
for diesel fuel. Applied Catalysis A: General, 2009. 367(1–2): p. 77-83.
63. Hernández D., J. J. Fernández, F. Mondragón, D. López, Production and
utilization performance of a glycerol derived additive for diesel engines. Fuel,
2012. 92(1): p. 130-136.
64. Melero J.A., R. van Grieken, G. Morales, M. Paniagua, Acidic Mesoporous
Silica for the Acetylation of Glycerol: Synthesis of Bioadditives to Petrol
Fuel. Energy & Fuels, 2007. 21(3): p. 1782-1791.
143
65. Mota C.J.A., C. X. A. da Silva, N. Rosenbach, J. Costa, F. v. da Silva,
Glycerin Derivatives as Fuel Additives: The Addition of Glycerol/Acetone
Ketal (Solketal) in Gasolines. Energy & Fuels, 2010. 24(4): p. 2733-2736.
66. Shen L., H. Yin, A. Wang, Y. Feng, Y. Shen, Z. Wu, T. Jiang, Liquid phase
dehydration of glycerol to acrolein catalyzed by silicotungstic,
phosphotungstic, and phosphomolybdic acids. Chemical Engineering
Journal, 2012. 180(0): p. 277-283.
67. Cavani F., S. Guidetti, L. Marinelli, M. Piccinini, E. Ghedini, M. Signoretto,
The control of selectivity in gas-phase glycerol dehydration to acrolein
catalysed by sulfated zirconia. Applied Catalysis B: Environmental, 2010.
100(1–2): p. 197-204.
68. Corma A., G. W. Huber, L. Sauvanaud, P. O'Connor, Biomass to chemicals:
Catalytic conversion of glycerol/water mixtures into acrolein, reaction
network. Journal of Catalysis, 2008. 257(1): p. 163-171.
69. Haider M.H., N. F. Dummer, D. Zhang, P. Miedziak, T. E. Davies, S. H.
Taylor, D. J. Willock, D. W. Knight, D. Chadwick, G. J. Hutchings, Rubidium-
and caesium-doped silicotungstic acid catalysts supported on alumina for the
catalytic dehydration of glycerol to acrolein. Journal of Catalysis, 2012.
286(0): p. 206-213.
70. Suprun W., M. Lutecki, T. Haber, H. Papp, Acidic catalysts for the
dehydration of glycerol: Activity and deactivation. Journal of Molecular
Catalysis A: Chemical, 2009. 309(1–2): p. 71-78.
71. Cavani F., S. Guidetti, C. Trevisanut, E. Ghedini, M. Signoretto, Unexpected
events in sulfated zirconia catalyst during glycerol-to-acrolein conversion.
Applied Catalysis A: General, 2011. 409–410(0): p. 267-278.
144
72. Gu Y., N. Cui, Q. Yu, C. Li, Q. Cui, Study on the influence of channel
structure properties in the dehydration of glycerol to acrolein over H-zeolite
catalysts. Applied Catalysis A: General, 2012. 429–430(0): p. 9-16.
73. Chieregato A., F. Basile, P. Concepción, S. Guidetti, G. Liosi, M. D. Soriano,
C. Trevisanut, F. Cavani, J. M. L. Nieto, Glycerol oxidehydration into acrolein
and acrylic acid over W–V–Nb–O bronzes with hexagonal structure. Catalysis
Today, 2012. 197(1): p. 58-65.
74. Wang F., J.-L. Dubois, W. Ueda, Catalytic dehydration of glycerol over
vanadium phosphate oxides in the presence of molecular oxygen. Journal of
Catalysis, 2009. 268(2): p. 260-267.
75. Glassman I., R.A. Yetter, Chapter 1 - Chemical Thermodynamics and Flame
Temperatures, in Combustion (Fourth Edition). 2008, Academic Press:
Burlington. p. 1-41.
76. Hagos F.Y., A.R.A. Aziz, I.M. Tan, Water-in-diesel emulsion and its micro-
explosion phenomenon-review, IEEE 3rd International Conference on
Research Challenges in Information Science. 2011. p. 314-318.
77. Zhu M., Y. Ma, D. Zhang, A theoretical investigation into the effect of a
homogeneous catalyst on combustion characteristics of single droplets of
diesel and biodiesel, in Chemeca 2012: Quality of life through chemical
engineering: 23-26 September 2012. 2012, Engineers Australia: Wellington,
New Zealand. p. 543-553.
78. Hasan M.H., T.M.I. Mahlia, H. Nur, A review on energy scenario and
sustainable energy in Indonesia. Renewable and Sustainable Energy
Reviews, 2012. 16(4): p. 2316-2328.
145
79. Zhang Z., M. Zhu, Y. Zhang, H. Y. Setyawan, J. Li, D. Zhang, Ignition and
combustion characteristics of single particles of Zhundong lignite: Effect of
water and acid washing. Proceedings of the Combustion Institute, 2016. 1(8).
80. Zhu M., Y. Ma, D. Zhang, Effect of a homogeneous combustion catalyst on
combustion characteristics of single droplets of diesel and biodiesel.
Proceedings of the Combustion Institute, 2013. 34(1): p. 1537-1544.
81. Law C.K., Combustion Physics, Cambridge University Press. 2006, New
York, United States of America.
82. Zhu M., Z. Zhang, Y. Zhang, H. Setyawan, P. Liu, D. Zhang, An experimental
study of the ignition and combustion characteristics of single droplets of
biochar-glycerol-water slurry fuels. Proceedings of the Combustion Institute,
2016. 1(8).
83. Glassman I., R.A. Yetter, Chapter 7 - Ignition, in Combustion (Fourth Edition).
2008, Academic Press: Burlington. p. 379-408.
84. Setyawan H.Y., M. Zhu, Z. Zhang, D. Zhang. An experimental study of the
effect of soap on ignition and combustion characteristics of single droplets of
glycerol. in Australian Combustion Symposium. 2015. University of
Melbourne: Australian Combustion Symposium.
85. Setyawan H.Y., M. Zhu, Z. Zhang, D. Zhang, An experimental study of effect
of water on ignition and combustion characteristics of single droplets of
glycerol. Energy Procedia, 2015. 75: p. 578-583.
86. Hallett W.L.H., N.A. Clark, A model for the evaporation of biomass pyrolysis
oil droplets. Fuel, 2006. 85(4): p. 532-544.
87. El-Mahallawy F., S.E.D. Habik, Fundamentals and Technology of
Combustion. 2002. Elsevier science.
146
88. Makino A., H. Fukada, Ignition and combustion of a falling, single sodium
droplet. Proceedings of the Combustion Institute, 2005. 30(2): p. 2047-2054.
89. Valliyappan T., N.N. Bakhshi, A.K. Dalai, Pyrolysis of glycerol for the
production of hydrogen or syn gas. Bioresource Technology, 2008. 99(10): p.
4476-4483.
90. Meyrahn H., The use of glycerol, glycerol-ethanol or glycerol and methanol
mixtures as fuel for furnaces, 2007, RWE Power Aktiengesellschaft German.
91. Bohon M.D., W.L. Roberts, NOx emissions from high swirl turbulent spray
flames with highly oxygenated fuels. Proceedings of the Combustion
Institute, 2013. 34(1): p. 1705-1712.
92. Stelmasiak Z., D. Pietras, Utilization of waste glycerin to fuelling of spark
ignition engines. IOP Conference Series: Materials Science and
Engineering, 2016. 148(1): p. 012087.
93. Beatrice C., G. Di Blasio, M. Lazzaro, C. Cannilla, G. Bonura, F. Frusteri, F.
Asdrubali, G. Baldinelli, A. Presciutti, F. Fantozzi, G. Bidini, P. Bartocci,
Technologies for energetic exploitation of biodiesel chain derived glycerol:
Oxy-fuels production by catalytic conversion. Applied Energy, 2013. 102(0):
p. 63-71.
94. Jiang L., A.K. Agrawal, Combustion of straight glycerol with/without methane
using a fuel-flexible, low-emissions burner. Fuel, 2014. 136(0): p. 177-184.
95. Ramadhas A.S., S. Jayaraj, C. Muraleedharan, Use of vegetable oils as I.C.
engine fuels—A review. Renewable Energy, 2004. 29(5): p. 727-742.
96. No S.Y., Inedible vegetable oils and their derivatives for alternative diesel
fuels in CI engines: A review. Renewable and Sustainable Energy Reviews,
2011. 15(1): p. 131-149.
147
97. Kumar S., J. H. Cho, J. Park, I. Moon, Advances in diesel–alcohol blends and
their effects on the performance and emissions of diesel engines. Renewable
and Sustainable Energy Reviews, 2013. 22(0): p. 46-72.
98. Fernando S., S. Adhikari, K. Kota, R. Bandi, Glycerol based automotive fuels
from future biorefineries. Fuel, 2007. 86(17–18): p. 2806-2809.
99. Dee V., B.D. Shaw, Combustion of propanol–glycerol mixture droplets in
reduced gravity. International Journal of Heat and Mass Transfer, 2004.
47(22): p. 4857-4867.
100. Demirbas A., Glycerol-based fuel oxygenates for biodiesel and diesel fuel
blends. Energy Sources, Part A: Recovery, Utilization, and Environmental
Effects, 2009. 31(19): p. 6.
101. Aggarwal S.K., Single droplet ignition: Theoretical analyses and experimental
findings. Progress in Energy and Combustion Science, 2014. 45(0): p. 79-
107.
102. Zhu M., Z. Zhang, Y. Zhang, H. Setyawan, P. Liu, D. Zhang, An experimental
study of the ignition and combustion characteristics of single droplets of
biochar-glycerol-water slurry fuels. Proceedings of the Combustion Institute,
2016. 1(8): p. 1-8.
103. Ulzama S., E. Specht, An analytical study of droplet combustion under
microgravity: Quasi-steady transient approach. Proceedings of the
Combustion Institute, 2007. 31(2): p. 2301-2308.
104. Gogos G., S. Soh, and D.N. Pope, Effects of gravity and ambient pressure on
liquid fuel droplet evaporation. International Journal of Heat and Mass
Transfer, 2003. 46(2): p. 283-296.
148
105. Banerjee R., Numerical investigation of evaporation of a single ethanol/iso-
octane droplet. Fuel, 2013. 107(0): p. 724-739.
106. Dirbude S., V. Eswaran, A. Kushari, Numerical modelling of droplet
evaporation with convection for n-alkanes and kerosene fuel. Atomization and
Sprays. 2011. 21(9): p. 787-798.
107. Mezhericher M., A. Levy, I. Borde, Theoretical Models of single droplet
drying kinetics: A review. Drying Technology, 2010. 28(2): p. 278-293.
108. Krishnan S., B. M. Kumfer, W. Wu, J. Li, A. Nehorai, R. L. Axelbaum, An
Approach to thermocouple measurements that reduces uncertainties in high-
temperature environments. Energy & Fuels, 2015. 29: p. 3446-3455.
109. Herbinet O., W.J. Pitz, C.K. Westbrook, Detailed chemical kinetic mechanism
for the oxidation of biodiesel fuels blend surrogate. Combustion and Flame,
2010. 157(5): p. 893-908.
110. Westbrook C.K., W.J. Pitz, H.J. Curran, Chemical Kinetic Modeling Study of
the Effects of Oxygenated Hydrocarbons on Soot Emissions from Diesel
Engines. The Journal of Physical Chemistry A, 2006. 110(21): p. 6912-6922.
111. Ranzi E., A. Frassoldati, A. Stagni, M. Pelucchi, A. Cuoci, T. Faravelli,
Reduced Kinetic Schemes of Complex Reaction Systems: Fossil and Biomass-
Derived Transportation Fuels. International Journal of Chemical Kinetics,
2014. 46(9): p. 512-542.
112. Kumar K., C. J. Sung, Autoignition of methanol: Experiments and
computations. International Journal of Chemical Kinetics, 2011. 43(4): p.
175-184.
113. Coniglio L., H. Bennadji, P. A. Glaude, O. Herbinet, F. Billaud, Combustion
chemical kinetics of biodiesel and related compounds (methyl and ethyl
149
esters): Experiments and modeling – Advances and future refinements.
Progress in Energy and Combustion Science, 2013. 39(4): p. 340-382.
114. Gog A., M. Roman, M. Toşa, C. Paizs, F. D. Irimie, Biodiesel production
using enzymatic transesterification – Current state and perspectives.
Renewable Energy, 2012. 39(1): p. 10-16.
115. Gomes M.C.S., P.A. Arroyo, N.C. Pereira, Biodiesel production from
degummed soybean oil and glycerol removal using ceramic membrane.
Journal of Membrane Science, 2011. 378(1–2): p. 453-461.
116. Saleh J., M.A. Dubé, A.Y. Tremblay, Effect of Soap, Methanol, and Water on
Glycerol Particle Size in Biodiesel Purification. Energy & Fuels, 2010.
24(11): p. 6179-6186.
117. Setyawan H.Y., M. Zhu, W. Zhou, D. Zhang, Effect of Methanol Addition on
Combustion Characteristics of Single Droplets of Glycerol. Proceedings of the
Australian Combustion Symposium Nov 6th to 8th, 2013, The University of
Western Australia, 2013: p. 332-335.
118. Botero M.L., Y. Huang, D. L. Zhu, A. Molina, C. K. Law, Synergistic
combustion of droplets of ethanol, diesel and biodiesel mixtures. Fuel, 2012.
94(0): p. 342-347.
119. Sazhin S.S., W. A. Abdelghaffar, E. M. Sazhina, M. R. Heikal, Models for
droplet transient heating: Effects on droplet evaporation, ignition, and break-
up. International Journal of Thermal Sciences, 2005. 44(7): p. 610-622.
120. Laza T., Á. Bereczky, Basic fuel properties of rapeseed oil-higher alcohols
blends. Fuel, 2011. 90(2): p. 803-810.
150
121. Jeong I.C., K.H. Lee, Auto-ignition and micro-explosion behaviors of droplet
arrays of water-in-fuel emulsion. International Journal of Automotive
Technology, 2008. 9(6): p. 735-740.
122. Makino A., H. Fukada, Combustion behavior of a falling sodium droplet:
Burning rate-constant and drag coefficient. Heat Transfer—Asian Research,
2005. 34(7): p. 481-495.
123. Farouk T., F.L. Dryer, Microgravity droplet combustion: effect of tethering
fiber on burning rate and flame structure. Combustion Theory and Modelling,
2011. 15(4): p. 487-515.
124. Lyra F.H., M. T. W. D. Carneiro, G. P. Brandão, H. M. Pessoa, E. V. de
Castro, Determination of Na, K, Ca and Mg in biodiesel samples by flame
atomic absorption spectrometry (F AAS) using microemulsion as sample
preparation. Microchemical Journal, 2010. 96(1): p. 180-185.
125. Awasthi I., G. Gogos, T. Sundararajan, Effects of size on combustion of
isolated methanol droplets. Combustion and Flame, 2013. 160(9): p. 1789-
1802.
126. Nayagam V., Activation energy asymptotics for methanol droplet extinction in
microgravity. Combustion and Flame, 2013. 160(11): p. 2638-2640.
127. Glycerine Producers A., Physical Properties of Glycerine and Its Solutions.
1963: Glycerine Producers' Association.
128. Lif A., K. Holmberg, Water-in-diesel emulsions and related systems. Advances
in Colloid and Interface Science, 2006. 123–126: p. 231-239.
129. Mitchell E.W., Heavy-duty diesel emission control : a review of technology,
ed. E.W. Mitchell. Vol. 36. 1986, Montreal, Quebec: Canadian Institute of
Mining and Metallurgy.
151
130. Walavalkar A.Y., Combustion of water-in-oil emulsions of diesel and fresh
and weathered crude oils floating on water, in Mechanical Engineering. 2001,
Pennsylvania State University: Pennsylvania, US.
131. Negrin I., Y. Parmet, E. Schechtman, Developing a model-based sampling
plan. Quality and Reliability Engineering International, 2011. 27(8): p. 1183-
1198.
132. Zhang B.L., F.A. Williams, Alcohol droplet combustion. Acta Astronautica,
1996. 39(8): p. 599-603.
133. Zhang B.L., J.M. Card, F.A. Williams, Application of rate-ratio asymptotics to
the prediction of extinction for methanol droplet combustion. Combustion and
Flame, 1996. 105(3): p. 267-290.
134. Zhang B.L., F.A. Williams, Theoretical studies of methanol droplet
combustion based on results from the USML-2 mission. Acta Astronautica,
1997. 40(12): p. 829-835.
135. Wang C.H., K. L. Pan, G. J. Ueng, L. J. Kung, J. Y. Yang, Burning behaviors
of collision-merged water/diesel, methanol/diesel, and water + methanol/diesel
droplets. Fuel, 2013. 106(0): p. 204-211.
136. Yang L., S. I. Sandler, D. G. Vlachos, C. Peng, H. Liu and Y. Hu, Adsorption
and diffusion of methanol, glycerol, and their mixtures in a metal organic
framework. Industrial & Engineering Chemistry Research, 2011. 50(24): p.
14084-14089.
137. Ghotli R. A, A. R. Abdul Aziz, I. M. Atadashi, D. B. Hasan, P. S. Kong and
M. K. Aroua, Selected physical properties of binary mixtures of crude glycerol
and methanol at various temperatures. Journal of Industrial and Engineering
Chemistry, 2015. 21: p. 1039-1043.
152
138. Marchese A.J., T. L. Vaughn, K. Kroenlein, F. L. Dryer, Ignition delay of fatty
acid methyl ester fuel droplets: Microgravity experiments and detailed
numerical modeling. Proceedings of the Combustion Institute, 2011. 33(2): p.
2021-2030.
139. Wang C.H., S. Y. Fu, L. J. Kung, C. K. Law, Combustion and microexplosion
of collision-merged methanol/alkane droplets. Proceedings of the Combustion
Institute, 2005. 30(2): p. 1965-1972.
140. Ramos M.J., C. M. Fernández, A. Casas, L. Rodríguez, Á. Pérez, Influence of
fatty acid composition of raw materials on biodiesel properties. Bioresource
Technology, 2009. 100(1): p. 261-268.
141. Marchese A.J., F.L. Dryer, The effect of liquid mass transport on the
combustion and extinction of bicomponent droplets of methanol and water.
Combustion and Flame, 1996. 105(1–2): p. 104-122.
142. Goodrum J.W., Volatility and boiling points of biodiesel from vegetable oils
and tallow. Biomass and Bioenergy, 2002. 22(3): p. 205-211.
143. Lapuerta M., J. Rodríguez-Fernández, C. Estevez, N. Bayarri, Properties of
fatty acid glycerol formal ester (FAGE) for use as a component in blends for
diesel engines. Biomass and Bioenergy, 2015. 76: p. 130-140.
144. Cammenga H.K., F.W. Schulze, W. Theuerl, Vapor pressure and evaporation
coefficient of glycerol. Journal of Chemical & Engineering Data, 1977.
22(2): p. 131-134.
145. Graham J.P., J. Ryer, S.J. Lestz, Gasoline Composition Containing a Sodium
Additive, U.S Patents, Editor. 1976.
153
146. Folarin O.M., I.C. Eromosele, C.O. Eromosele, Relative thermal stability of
metal soaps of Ximenia americana and Balanites aegyptiaca seed oils.
Scientific Research and Essays, 2010. 6(9): p. 1923-1927.
147. Aranda V., Experimental and Kinetic Modeling Study of Methanol Ignition
and Oxidation at High Pressure. Chemical Kinetics, 2013. 45(5): p. 11.
148. Ling X., F. Wu, D. Yao, A reduced combustion kinetic model for the
methanol-gasoline blended fuels on SI engines. Science China Technological
Sciences, 2016. 59(1): p. 81-92.
149. Lai J.Y.W., K.C. Lin, A. Violi, Biodiesel combustion: Advances in chemical
kinetic modeling. Progress in Energy and Combustion Science, 2011. 37(1):
p. 1-14.
150. Westbrook C.K., Chemical kinetics of hydrocarbon ignition in practical
combustion systems. Proceedings of the Combustion Institute, 2000. 28(2): p.
1563-1577.
151. An H., W.M. Yang, J. Li, Numerical modeling on a diesel engine fueled by
biodiesel–methanol blends. Energy Conversion and Management, 2015. 93:
p. 8.
152. An H., W. Yang, J. Li, A. Maghbouli, K. J. Chua, S. K. Chou, A numerical
modeling on the emission characteristics of a diesel engine fueled by diesel
and biodiesel blend fuels. Applied Energy, 2014. 130: p. 458-465.
153. Vallinayagam R., S. Vedharaj, W. M. Yang, P. S. Lee, K. J. E. Chua,S. K.
Chou, Pine oil–biodiesel blends: A double biofuel strategy to completely
eliminate the use of diesel in a diesel engine. Applied Energy, 2014. 130: p.
466-473.
154
154. Hagos F.Y., A.R.A. Aziz, I.M. Tan. Water-in-diesel emulsion and its micro-
explosion phenomenon-review, 2011, IEEE 3rd International Conference on
Research Challenges in Information Science.
155. Nour M., H. Kosaka, S. Sato, M. Bady, A. K. Abdel-Rahman, K. Uchida,
Effect of ethanol/water blends addition on diesel fuel combustion in RCM and
DI diesel engine. Energy Conversion and Management, 2017. 149: p. 228-
243.
156. Barker J., S. Cook, P. Richards, Sodium Contamination of Diesel Fuel, its
Interaction with Fuel Additives and the Resultant Effects on Filter Plugging
and Injector Fouling. SAE Int. J. Fuels Lubr., 2013. 6(3): p. 826-838.
157. Bournay L., D. Casanave, B. Delfort, G. Hillion, J. A. Chodorge, New
heterogeneous process for biodiesel production: A way to improve the quality
and the value of the crude glycerin produced by biodiesel plants. Catalysis
Today, 2005. 106(1–4): p. 190-192.
158. Lissianski V.V., V.M. Zamansky, P.M. Maly Effect of Metal-Containing
Additives on NOx Reduction in Combustion and Reburning. Combustion and
Flame, 2001. 125, 1118-1127 DOI: http://dx.doi.org/10.1016/S0010-
2180(01)00231-0.
159. Okano Y., A. Yamaguchi, Numerical simulation of a free-falling liquid sodium
droplet combustion. Annals of Nuclear Energy, 2003. 30(18): p. 1863-1878.
160. Takata T., A. Yamaguchi, I. Maekawa, Numerical investigation of multi-
dimensional characteristics in sodium combustion. Nuclear Engineering and
Design, 2003. 220(1): p. 37-50.
161. de Oliveira A.P., R. D. Villa, K. C. P. Antunes, A. de Magalhães, E. C. e.
Silva, Determination of sodium in biodiesel by flame atomic emission
155
spectrometry using dry decomposition for the sample preparation. Fuel, 2009.
88(4): p. 764-766.
162. Barros A.I., A. P. de Oliveira, M. R. L. de Magalhães, R. D. Villa,
Determination of sodium and potassium in biodiesel by flame atomic emission
spectrometry, with dissolution in ethanol as a single sample preparation step.
Fuel, 2012. 93(0): p. 381-384.
163. Wiens J.P., G. M. Nathanson, W. A. Alexander, T. K. Minton, S. Lakshmi, G.
C. Schatz, Collisions of Sodium Atoms with Liquid Glycerol: Insights into
Solvation and Ionization. Journal of the American Chemical Society, 2013.
136 (8), pp 3065–3074