ethanol production from selecteddigilib.library.usp.ac.fj/gsdl/collect/usplibr1/index/... ·...

ETHANOL PRODUCTION FROM SELECTED

CASSAVA VARIETIES IN FIJI AND TESTING OF

ETHANOL-PETROL FUEL BLENDS

by

Pritika Bijay

A thesis submitted in partial fulfilment of the

requirements for the degree of Master of Science in Physics

Copyright © 2013 by Pritika Bijay

School of Engineering and Physics

Faculty of Science, Technology and Environment

The University of the South Pacific

Fiji

January 2013

i

Declaration of Originality

I, Pritika Bijay, hereby declare that this thesis is my original work and wherever the

work of others has been used, it has been clearly referenced.

Signature

Date

Name Pritika Bijay

Student ID S11022830

Statement by Principal Supervisor

The research in this thesis was performed under my supervision and to my

knowledge is the sole work of Ms Pritika Bijay.

Signature __________________

Date __________________

Name Dr. Anirudh Singh

Designation Principal Supervisor

ii

Acknowledgments

The successful compilation of this thesis would not have been possible through the

assistance, support and guidance of many individuals who have helped me

throughout my research work. Therefore, I would like to take this opportunity to

express my sincere gratitude and heartfelt appreciation to the following individuals

whose help has been very much appreciated.

Firstly, I would like to express profound gratitude to my supervisor, Dr. Anirudh

Singh, for his invaluable support, encouragement, supervision and useful suggestions

throughout this research work. Also my sincere gratitude to Dr. Jagjit Khurma and

Mr. Villimone Vosarogo as co-supervisors for providing their help wherever needed.

My sincere appreciation also goes to the technical staff of School of Engineering and

Physics; Mr. Viti Buadromo, Mr. Amit Deo, Mr. Shanil Deo, Mr. Joape Cawanibuka,

Mr. Neil Singh, Mr. Rohit Lal and Mr. Abhinay Shandil for helping me during the

experimental stages of my thesis especially the technical assistance with operation of

the instruments. My special thanks also go to Mr. Steven Sutcliffe, Mr. Shelvin

Prasad, Ms. Roslyn Lata, Mr. Dinesh Kumar and Mr. Hirdesh Singh of School of

Biological and Chemical Sciences for providing me with the necessary equipment

and instruments. My most sincere gratitude also goes to Koronivia Research Station

and Dobuilevu Research Station for providing me with the different cassava

varieties. Also a special thanks to Mr. Poasa Nauluvula for providing the relevant

information on the cassava varieties. Special thanks also go to my dear friends; Mr.

Shivneel Prasad, Mr. Rajneel Prasad, Mr. Naveendra Reddy, Mr. Pranil Singh, Ms

Monishka Narayan, Ms Esha Chetty, Ms Priti Maharaj, Ms. Pritika Reddy, Mr. Atesh

Gosai, Mr. Sunil Chand, Mr. Aman Deo, Mr. Imraan Jannif and Mr. Malvin Nadan

for the words of encouragement and support during every stage of my thesis.

Finally, I would like to thank my parents, for their continuous support throughout my

life and always encouraging me to further my knowledge and studies. Thanks also to

my sister, Asnita. Furthermore, the backing and continual encouragement given by

my husband was a source of inspiration to my work. He has been a pillar of strength

for me and I am deeply obliged to him for standing by my side.

iii

Abstract

Ethanol production from renewable resources has received worldwide attention due

to increasing petroleum shortage. One such renewable resource that has been

identified is cassava starch, which can be extracted from root crop cassava (Manihot

esculenta (Crantz)) and is readily available in Fiji. Ethanol may be used as a fuel for

spark-ignition engines either in its original form, or as blends with petrol. The most

feasible way of expanding the use of ethanol as a fuel, however, is by using ethanol-

petrol blends in vehicles already on the road, without the need to modify the engines.

The main objective of this study was to produce ethanol from some of the locally

available cassava varieties in Fiji, and to test the operation of spark ignition (SI)

engines on various ethanol-petrol blends.

Starch was extracted from the roots of ten different cassava varieties available at two

different research stations in Fiji using the sedimentation technique, and the yields

were determined. In the case of Koronivia Research Station (KRS) the variety

Nadelei had the highest starch yield (23.1 %) whereas Coci had the highest starch

yield (23.3 %) for Dobuilevu Research Station (DRS). The extracted starch was then

used to produce ethanol via the technique of Simultaneous Saccharification and

Fermentation (SSF), with the yeast Saccharomyces cerevisiae being used as the

fermentation agent. Ethanol yield was in the range of 0.35-0.40 L of ethanol per kg

of starch and 0.35-0.41 L of ethanol per kg of starch for KRS and DRS respectively.

Ethanol-petrol blends, E10, E15 and E20 were prepared from alcohol with varying

degrees of water content. It was found that 96 % ethanol could be used to prepare

blends that did not phase separate at temperatures typical of the tropics.

The prepared blends were tested on a SI engine for engine efficiency, fuel

consumption and exhaust emissions. An increase in fuel consumption was noted as

the engine load was increased and also as the ethanol content in petrol was increased.

The latter effect was identified as being due to the lower gross calorific value (GCV)

of ethanol as compared to petrol. Hence, as the ethanol fraction in ethanol-petrol

iv

blend increased, the GCV decreased, and as a result more fuel was required. At

maximum load, there is a decrease in carbon monoxide (CO) emission by 34, 61 and

78 % with the E10, E15 and E20 blends respectively when compared to petrol.

Reductions in exhaust emissions of hydrocarbons (HC) by approximately 10, 30, and

34 % were noted for E10, E15, and E20 blends respectively at maximum engine

loadings. Reductions were also observed in carbon dioxide (CO2) emissions at

maximum engine load for E10 (7 %), E15 (17 %) and E20 (20 %) when compared to

petrol.

v

List of Abbreviations, Units and Nomenclature

LIST OF ABBREVIATIONS

AFR Air-fuel ratio

B Biomass

B5 5 % biodiesel with 95 % diesel

CF Calibration Factor

CI Compression Ignition

CIAT Centro Internacional de Agricultural Tropical

CO Carbon Monoxide

CO2 Carbon Dioxide

DMC Dry matter content

DNS 3, 5 – dinitrosalicylic acid

DRS Dobuilevu Research Station

E Ethanol

E10 10 % ethanol with 90 % petrol



E22 20-24 % ethanol blended with 80-76 % petrol

E85 85 % ethanol blended with 15 % petrol

E100 100 % ethanol

ECU Electronic Control Unit

FAO Food and Agriculture Organization

FDOE Fiji Department of Energy

FFT Flexible Fuel Technology

GC Gas Chromatography

GCV Gross calorific value

HC Hydrocarbons

KRS Koronivia Research Station

LOS Left over sugar

NOx Nitrogen oxide

PM Particulate matter

vi

S Substrate

SD Standard Deviation

SI Spark Ignition

SFC Specific Fuel Consumption

US United States

WOT Wide Open Throttle

LIST OF UNITS

% Percent

° Degree

°C Degrees Celcius

µm Micrometre

atm Atmosphere

cm Centimetre

div Division

g Gram

g l-1 Gram per litre

g cm-3 Gram per cubic centimetre

hrs Hours

K

kJ g-1

Kelvin

Kilojoules per gram

kJ ml-1 Kilojoules per millilitre

L litre

L t-1 Litre per tonne

kg ha-1 year-1 Kilograms per hectare per year

M Molar

m Metre

ml Millilitre

min Minute

ml min-1 Millilitre per minute

nm Nanometer

ppm Parts per million

vii

rpm Revolutions per minute

t ha-1 Tonne per hectare

t ha-1 year-1 Tonnes per hectare per year

W Watt

LIST OF NOMENCLATURE

inE Energy input to the generator

outE Energy output of the generator

ε Compression ratio

1L Power loss from the engine

2L Power loss from the alternator

E� Engine Efficiency

S� System Efficiency

A� Alternator Efficiency

λ Air-fuel equivalence ratios

FS� Density of fuel sample

outP Power output of the alternator

inP Power input of the engine

1P Power input of alternator

QP Volumetric ethanol productivity

YX/S Conversion rate of starch to biomass

YP/S Yield factor of ethanol on substrate

YP/X Yield factor of ethanol on biomass

viii

Table of Contents

Declaration of Originality i

Acknowledgments ii

Abstract iii

List of Abbreviations, Units and Nomenclature v

Table of Contents viii

List of Figures xi

List of Tables xiii

Chapter 1 Introduction 1

1.1 Ethanol as a Biofuel for Transportation 1

1.2 Objectives 8

1.3 Structure of the Thesis 9

1.4 Relevance of the Thesis 10

Chapter 2 Literature Review 11

2.0 Overview 11

2.1 Cassava 11

2.1.1 Cassava Breeding 12

2.1.2 Effect of Environmental Factors on Cassava 13

2.1.3 Cassava Varieties 14

2.2 Production of Ethanol 15

2.3 Ethanol as Fuel 20

Chapter 3 Methodology 28

3.0 Overview 28

3.1 Cassava Varieties in Fiji 28

3.1.1 Dry Matter Content of Cassava Roots 29

3.1.2 Starch Extraction from Cassava Roots 29

ix

3.1.2.1 Moisture Content of Starch 31

3.1.2.2 Ash Content of Starch 31

3.1.2.3 pH Determination of Starch 31

3.2 Ethanol Production from Cassava Starch 32

3.2.1 List of Equipment and Reagents 32

3.2.2 Preparation of Cassava Starch Solution 33

3.2.3 Treatment of Cassava Starch Solution to Simple Sugars 34

3.2.3.1 Pre-treatment of Gelatinized Starch with α-amylase 35

3.2.4 Simultaneous Saccharification and Fermentation 36

3.2.5 Analytical Analysis 37

3.2.5.1 Reducing Sugar Analysis 37

3.2.5.2 Determination of Ethanol Concentration 39

3.3 Preparation of Ethanol-Petrol Blends 40

3.3.1 Stability Testing of Ethanol-Petrol Blends 41

3.4 Physical Properties 41

3.4.1 Gross Calorific Value 42

3.4.1.1 Calibrating the Bomb Calorimeter 43

3.4.1.2 Determination of Gross Calorific Value 43

3.4.2 Density 44

3.4.2.1 Determine the Volume of the Picnometer 45

3.4.2.2 Determine the Density of Fuel Sample 46

3.5 Engine Efficiency, Fuel Consumption and Emission Testing 46

3.5.1 The Testing Equipment 48

3.5.2 The Testing Procedure 50

3.6 Engine Efficiency 50

3.6.1 Power Loss from the System 53

3.7 Emission Testing 55

3.7.1 Instrument Start-up 55

3.7.2 Measurement of Emission 56

Chapter 4 Results and Discussions 58

4.0 Overview 58

4.1 Cassava Varieties in Fiji 58

x

4.2 Ethanol Production from Cassava Starch 67

4.3 Preparation of Ethanol-Petrol Blends 82

4.4 Physical Properties of Ethanol-Petrol Blends 85

4.4.1 Density 85

4.4.2 Gross Calorific Value 87

4.5 Engine Performance and Emission Characteristics of Ethanol-Petrol 87

Blends

4.5.1 Engine Efficiency 88

4.5.1.2 Engine Losses 90

4.5.2 Fuel Consumption 91

4.5.3 Specific Fuel Consumption 92

4.5.4 Engine Exhausts Emission Analysis 93

4.5.4.1 CO Emission 94

4.5.4.2 HC Emission 95

4.5.4.3 CO2 Emission 96

4.5.4.4 The Effects of Various Fuels on Exhaust Emissions at Constant 97

Load

4.5.5 Prospects and Challenges for Bio-ethanol Use in Vehicles 99

Chapter 5 Conclusions 102

5.1 Recommendations and Suggestions for Future Work 104

REFERENCES 106

APPENDIX A 122

APPENDIX B 123

APPENDIX C 124

APPENDIX D 125

xi

List of Figures

CHAPTER 1

Figure 1.1 Schematic diagram of fuel system of an SI Engine 3

Figure 1.2 A four stroke spark ignition cycle 4

Figure 1.3 Fiji’s fuel import bills as compared with total

import bills

7

CHAPTER 3

Figure 3.1 Map of Fiji showing the collection points of

cassava varieties

28

Figure 3.2 Flowchart for cassava starch production 30

Figure 3.3 Gelatinized cassava starch solution 34

Figure 3.4 Starch hydrolysis using α-amylase 36

Figure 3.5 Setup for Simultaneous Saccharification and

Fermentation

37

Figure 3.6 Setup for Ballistic Bomb Calorimeter 42

Figure 3.7 Picnometer with Fuel Sample 45

Figure 3.8 Equipment for testing engine efficiency, fuel

consumption and emission

49

Figure 3.9 Generalised flowchart of the systems input, output

and losses

52

CHAPTER 4

Figure 4.1 Rainfall and Temperature data at KRS 60

Figure 4.2 Rainfall and Temperature data at DRS 61

Figure 4.3 Ethanol Concentration and remnant reducing sugars

concentration from (a) Niumea, (b) Sokobale, (c)

Beqa, (d) New Guinea, (e) Coci, (f) Vula Tolu, (g)

Yabia Damu, (h) Merelesita, (i) Nadelei, (j)

Navolau cassava variety starch obtainedat KRS

72

Figure 4.4 Ethanol Concentration and remnant reducing sugars

concentration from (a) Niumea, (b) Beqa, (c) New

77

xii

Guinea, (d) Coci, (e) Vula Tolu, (f) Yabia Damu,

(g) Merelesita, (h) Nadelei, (i) Navolau cassava

variety starch obtained at DRS

Figure 4.5 Density of Petrol, E10, E15 and E20 86

Figure 4.6 GCV of Petrol, E10, E15 and E20 on mass and

volume basis

87

Figure 4.7 Engine efficiency using Petrol and Ethanol-Petrol

blends under varying loads

89

Figure 4.8 Fuel consumption of Petrol, E10, E15, E20 under

varying loads

91

Figure 4.9 Specific Fuel Consumption of Petrol, E10, E15,

E20 under varying load

93

Figure 4.10 Effect of varying loads on CO emissions for Petrol,

E10, E15 and E20

94

Figure 4.11 Effect of varying loads on HC emissions for Petrol,

E10, E15 and E20

95

Figure 4.12 Effect of varying loads on CO2 emissions for

Petrol, E10, E15 and E20

97

Figure 4.13 CO emissions for various fuels at maximum load 98

Figure 4.14 HC emissions for various fuels at maximum load 98

Figure 4.15 CO2 emissions for various fuels at maximum load 99

xiii

List of Tables

CHAPTER 1

Table 1.1 Modifications required for different ethanol contents

in ethanol-petrol blends

6

CHAPTER 2

Table 2.1 Kinetics parameters of ethanol production from

starch following growth of C. tropicalis, S.

cerevisiae, and S. occidentalis in the presence and

absence of α-amylase treatment

16

Table 2.2 Ethanol concentration and left over sugar in solid

substrate fermentation from various starchy

substrates using thermotolerant yeast (VS3)

18

CHAPTER 3

Table 3.1 Reagents used for ethanol production 33

Table 3.2 Specification of the petrol engine 47

Table 3.3 Specification of the generator 48 Table 3.4 Specifications of the Horiba Automotive Emission

Gas Analyser

55

CHAPTER 4

Table 4.1 Starch Yield from Cassava Varieties 59

Table 4.2 Dry Matter Content of Cassava Varieties 62

Table 4.3 Ash Content of Cassava Starch from Various

Varieties

63

Table 4.4 pH of Cassava Starch from Various Varieties 64

Table 4.5 Moisture Content of Cassava Starch from Various

Varieties

65

Table 4.6 Comparison of ethanol yield made from various

energy crops

66

Table 4.7 Final ethanol concentration and ethanol yield from 78

xiv

cassava varieties from two different locations

Table 4.8 Cassava required for producing 1 L ethanol from

each cassava variety studied

80

Table 4.9 Composition of Obtained Ethanol Yield with

Literature Results

81

Table 4.10 Composition of ethanol-petrol blended samples used

for analysis

83

Table 4.11 Stability testing using absolute ethanol in ethanol-

petrol blends

83

Table 4.12 Stability testing using 97 % ethanol in ethanol-petrol

blends

83


blends

84


blends

84

1

Chapter 1 Introduction

1.1 Ethanol as a Biofuel for Transportation

With the rapid development of industrialised society, the demand for fossil fuel has

been growing every day. The resulting escalation of oil prices coupled with a

forecast shortage of fossil fuel reserves has been the cause of great anxiety. These

factors and the increasing energy demand for transportation to keep the pace of

economic development are alerting many countries to the need to find alternative

energy sources. Biofuels is amongst the principal candidates for such alternative

fuels, especially for the transport sector.

One such biofuel is bio-ethanol or simply ethanol. Bio-ethanol’s greatest benefit lies

in its potential to reduce greenhouse gas emissions by partial replacement of oil as a

transport fuel (IEA, 2004). It can also reduce the burden of foreign currency

expenditure for poor countries that import petroleum products but have the potential

to produce and use bio-ethanol (WWI, 2006).

Brazil’s success story in using sugarcane based ethanol is quite well known. Many

advocates of biofuel subsidies and mandates frequently cite their experience. Brazil

is the world’s number two ethanol producer and the leading ethanol exporter, using

sugarcane as its feedstock (Hofstrand, 2009). In 2011 Brazil contributed to 24 % of

global ethanol production, when compared to 30 % in 2010 (REN21, 2012). After

being number one exporter for many years, the decline in ethanol production is

attributed to the decline in investment in new sugarcane assets and plantations since

the 2008 financial crisis, poor sugarcane harvests due to unfavorable weather and

high world sugar prices (OECD, 2011; Colitt and Nielson, 2012). Authorities in

Brazil had made it mandatory to have 20-25 % ethanol blended in petrol. However,

since the decline in ethanol production Brazil announced new policies to stimulate

sugar production and to reduce the amount of ethanol required in gasoline to 20 %

(Biofuels Digest, 2012). Bio-ethanol is an excellent substitute for gasoline, the main

Chapter 1: Introduction

2

car fuel used by spark ignition (SI) engines around the globe. Bio-ethanol can be

used in SI engines, either in its pure form or blended with conventional petroleum-

derived fuels.

A typical fuel system of an SI engine consists of the following parts, arranged

sequentially (Stephenson, 1973; Stone, 1999):

� Fuel tank- for storage of fuel (petrol)

� Fuel pump- to supply fuel to the carburetor or fuel injection system

� Fuel filter- to remove particles and impurities from fuel

� Fuel lines- tubes through which fuel passes from the tank to carburetor or fuel

injection system

� Carburetor- a (now outdated) device that atomizes the fuel and mixes it with

the correct amount of air (this device has now been replaced by a modern

electronic fuel injection (EFI) system)

� Intake manifold- where fuel is added to the air either by fuel injectors or the

carburetor

� Fuel injectors- devices that inject precise amounts of fuel into the incoming

air ensuring maintenance of the stoichiometric ratio

� Fuel pressure regulator- a device that controls fuel pressure

A schematic diagram of the fuel system for the SI engine is shown in Figure 1.1:


3

Figure 1.1: Schematic diagram of fuel system of an SI Engine

Recent developments have seen the complete computerization of engine controls and

fuel delivery systems (Fergerson and Kirkpatrick, 2001). In the 1980s conventional

carburetors were replaced by throttle body fuel injectors and later in 1990s these

were replaced by port fuel injectors (Fergerson and Kirkpatrick, 2001).

SI engines are also referred to as Otto cycle engines and these operate as four or two

stroke cycle. In a four stroke SI engine, a charge of premixed fuel-air mixture

(delivered by the fuel system) is drawn into the combustion chamber through intake

valves where the charge is then compressed by the motion of the piston. The

compressed fuel-air mixture is then ignited by one or more spark plugs; a turbulent

flame develops and propagates through the mixture, raising the temperature and

pressure of the cylinder. The flame extinguishes when it reaches the cylinder walls.

The burned gasses exit the engine past the exhaust valves, through the exhaust

manifold and into a central exhaust pipe. Figure 1.2 shows the spark ignition cycle of

a four stroke cylinder.

Unused Fuel Returned to Tank

FUEL TANK

FUEL PUMP

FUEL FILTERS

CARBURETOR OR FUEL

INJECTION SYSTEM

FUEL PRESSURE

REGULATOR

INTAKE MANIFOLD

ENGINE COMBUSTION

CHAMBER


4

Figure 1.2: A four stroke spark ignition cycle (Source: Ferguson an Kirkpatrick,

2001)

Bio-ethanol, a clear, colourless liquid made from renewable sources such as

sugarcane, molasses and starch is being considered for use in SI engines. Bio-

ethanol’s characteristics enable cleaner combustion and better engine performance,

which contribute to reduced pollutant emissions, even when it is mixed with petrol.

Bio-ethanol is considered particularly attractive as an alternative fuel because it is a

renewable bio-based resource and is oxygenated, which provides the potential to

reduce particulate emissions in SI engines (Gravalos et al, 2011).

Bio-ethanol has several advantages as a transportation fuel. It has high octane

number (i.e. has a high anti-knock quality). This enables it to operate at high

compression ratios which improve engine efficiency, power output and fuel

consumption (Lin et al, 2010). It also has high heat of vaporization when compared


5

to petrol. This enables the engine volumetric efficiency to increase (Bayraktar,

2005), which means it freezes the air, allowing more mass (fuel) to be drawn in the

cylinder resulting in increased power output (Lin et al, 2010). In addition, the use of

ethanol can reduce significantly carbon monoxide (CO) and hydrocarbon (HC)

emissions. This is due to the leaning effect caused by ethanol addition (Koç et al,

2009).

The simplest and fastest way of expanding the use of ethanol as a fuel is by using

ethanol-petrol blends in vehicles already on the road, without the need for modifying

engines. Developing countries that currently have a limited capacity to produce cost-

efficient ethanol with good energy and environmental balances can diversify their

liquid fuels options by importing ethanol from regions with favorable conditions for

biofuel production. These developing countries in the meantime can consider suitable

resources to be used for bio-ethanol production in their country which can then be

used locally.

It is important to consider the consequences of adopting ethanol-petrol blends on

engine performance, drivability and durability of vehicles, as well as the associated

environmental impacts. Table 1.1 shows the modifications to vehicle engines

required for different ethanol contents in petrol


6

Table 1.1: Modifications required for different ethanol contents in ethanol-petrol

blends (Source: ANFAVEA, 2005)

% of bio-ethanol in gasoline

Changes to a pure gasoline vehicle

Car

bure

tor

Fuel

inje

ctio

n

Fuel

pum

p

Fuel

filte

r

Igni

tion

syst

em

Fuel

tank

Cat

alyt

ic c

onve

rter

Bas

ic e

ngin

e

Eng

ine

oil

Inta

ke h

eade

r

Exh

aust

syst

em

Col

d-st

art s

yste

m

≤ 5 % Any Vehicle

≤ 10 % Vehicles produced from 1990 onwards

≤ 25 % Brazilian gasoline vehicle

≤ 85 % Flexible vehicle used in the USA and in Canada

≥ 85 % Flexible vehicle used in Brazil

No changes are necessary

Changes are probably necessary

Ethanol is an attractive alternative fuel for SI engine and can be used as a pure fuel or

blended with petrol. Using neat ethanol in SI engines will require modifications of

engine design and fuel systems. However, using ethanol in lower percentages

blended with petrol does not (see Table 1.1). Therefore, using ethanol-petrol blends

in SI engines is generally more expedient than using pure ethanol.

Biofuels offer a realistic option for Fiji to minimize some of its energy challenges

(Cloin et al., 2007) as evident in the fossil fuels import bills of the country. As seen

in Figure 1.3 fuel imports form a major part of the country’s total import bills. With

the successful production and use of biofuel, there will be a reduction in the

country’s fossil fuel import bills. Successful use of biofuels will also see a reduction

in greenhouse gas emissions that is caused by burning fossil fuels.


7

Figure 1.3: Fiji’s fuel import bills as compared with total import bills (Source:

Fiji (Bureau of Statistics, 2011)

Biofuels can play a fundamental role in the transportation sector of Fiji (Singh,

2012). The transportation needs of Fiji can be satisfied by using biodiesel and its

blends with diesel for Compression Ignition (CI) engine vehicles or by using bio-

ethanol and its blends with petrol for SI engine vehicles (Singh, 2012).

Fiji has the necessary resources available for bio-ethanol production (Singh, 2012).

Over the past years interest has been shown in bio-ethanol production in Fiji. The

possible feedstock identified was sugarcane and cassava. When considering

sugarcane, there are two options: either to use sugarcane juice, or molasses for

ethanol production. In a presentation made by the Director of Fiji Department of

Energy (Nakavulevu, 2011) during a workshop organized by International

Renewable Agency indicated that Fiji would require 7, 466, 242 L of ethanol for

blending with petrol for E10. His presentation also indicated that of the 100, 000 t of

molasses that Fiji Sugar Corporation produces, could produce 34 M L of ethanol

annually of which 8 M L will be require for blending to produce E10 while the rest

cold be imported. For developing the biofuel or bio-ethanol industry in Fiji it is

important to have available resources. These include feedstock plantations and also

available land area for the further development. Fiji has about 1.8 million ha of

0

500000

1000000

1500000

2000000

2500000

3000000

3500000

4000000

Fuel

and

tota

l nat

ioin

al im

port

s ($

000)

Year

Fiji's fuel imports (2000-2010)

Total fuel imports ($FJ000)

Total national imports ($FJ000)


8

available land, of which less than 19 % is available for agriculture. Since very small

portion of the land is arable, biofuels development needs careful investigation

beyond the technical investigation.

Fiji has recently made significant strides in developing its human capacity for

development of renewable energy resources, especially in the biofuel area. As

discussed by Singh (2012), the Fiji Department of Energy (FDOE) has established

biofuels units with staff experienced in biofuels research. The local universities have

started producing graduates with scientific training in biofuels, and the FDOE’s

biofuel unit has developed a biofuels standard for Fiji. In addition, legislation is now

in place to enforce these standards. This includes the use of 5 % biodiesel with 95 %

diesel (B5) in CI engine vehicles and 10 % ethanol with 90 % petrol (E10) in SI

engine vehicles. There has been significant interest by the private sector in the

biofuels industry as well.

Bio-ethanol is a promising alternative fuel for Fiji, with the availability of more than

one feedstock to consider and the availability of land to increase production, as well

as the potential for improvement in human capacity and interest by government and

private sector. It is clearly evident that with proper planning and development of the

biofuel industry, Fiji can improve its overall energy situation. The primary goal of

this study was to produce bio-ethanol from a renewable resource available in Fiji

which could be considered for blending with petrol for use in SI engines. An

important secondary goal was the testing of such blends in SI engines with a view to

ascertaining their viability for mitigating Fiji’s fuel supply and fuel import bill

situation.

1.2 Objectives

The main object of this project was to investigate the potential of using cassava as

feedstock for bio-ethanol production and the viability of using ethanol-petrol blends

as automotive fuels. The specific objectives were:


9

� To investigate the ethanol yield from selected cassava varieties obtained from

two different locations in Fiji.

� To determine the fuel characteristics of some ethanol-petrol blends and to

compare their properties with those of neat petrol.

� To compare the performance of an SI engine in terms of engine efficiency,

fuel consumption and exhaust emissions when using ethanol-petrol blends

with that of neat petrol

1.3 Structure of the Thesis

This thesis is organised as follows:

� Chapter 1 begins with an introduction which, amongst other things, gives an

overview of the research background, discusses the objectives and gives an

outline of the thesis.

� Chapter 2 provides an overview of the literature on the root crop cassava, the

breeding techniques, and effects of environmental factors on cassava. It also

provides details of the local cassava varieties in Fiji. This chapter briefly

outlines some techniques and results obtained by various authors that have

used cassava as feedstock for ethanol production. Finally, it describes the use

of ethanol as a fuel.

� Chapter 3 discusses the techniques and methods used in the data collection

for this work. The extraction of starch and determination of some of its

properties is described. Following this, the method for ethanol production

from the different varieties of starch and determining the ethanol content is

outlined. The technique used for blending ethanol-petrol together with the

determination of some fuel properties is presented. Also presented is the

method for the engine performance and emission tests.

� Chapter 4 presents the results and discusses their implications. This includes

tabulated results and discussion for starch yield, dry matter content and

various other properties of starch. The stability testing results and discussion

as well as variation of the density and gross calorific value with the varying

percentage of ethanol is shown. The engine performance in terms of fuel


10

consumption, specific fuel consumption and engine efficiency is analysed.

The emission characteristics of the ethanol-petrol bends are also presented

and discussed.

Chapter 5 summarizes the results of this work and provides suggestion for further

studies.

1.4 Relevance of the Thesis The work done in this research can contribute to developing knowledge on the bio-

ethanol market in Fiji, thus providing an important source of information to

policy/decision makers, academic researchers, businesses/industry and interested

groups. Some of the data presented in this research is completely new. This includes

the data on starch yield and the ethanol yield from the ten different cassava varieties

in Fiji that were studied. Therefore, this research is a start to obtaining important

information needed for the development of the bio-ethanol production industry in Fiji

For academic researchers this study can serve as a platform to provide basic

information on the bio-ethanol development using cassava available in Fiji as

feedstock. It enables potential areas to be identified for further study and research.

Similarly, for business society bio-ethanol development in the country is at an early

stage. Currently, there is no large scale ethanol production in Fiji. However, various

feedstock are being considered and this includes mainly molasses. With the

information contained in the study, important understanding can be obtained about

bio-ethanol in general and cassava as a potential feedstock.

Finally, for policy and decision makers some of the social and environmental issues

as well as barriers discussed in this work can provide important inputs for policy and

decision making to make interventions for sustainable domestic bio-ethanol

production.

11

Chapter 2 Literature Review

2.0 Overview

This chapter reviews relevant research done by various researchers in other countries

on cassava ethanol and the use of ethanol-petrol fuel blends. It discusses the use and

breeding of cassava globally and the various varieties of cassava present in Fiji, and

proceeds to consider various procedures used for producing ethanol from biomass

such as starch. Finally, it reviews the work done on ethanol blends as fuel for spark

ignition (SI) engines, and discusses the advantages as well as drawbacks associated

with the use of ethanol as blended fuels.

2.1 Cassava

Cassava or manioc (Manihot esculenta (Crantz)) which is a perennial shrub of the

New World is currently the sixth world food, in terms of global annual production

(Burns et al., 2010; El-Sharkawy, 2004). Cassava roots are an important staple food

for more than 800 million people in tropical and sub-tropical Africa, Asia and Latin

America (Burns et al., 2010). As Srinivas and Anantharuman (2000) have stated,

after cereals and grain legumes, root and tuber crops are the most important food

crops. Due to cassava’s potential for high matter production per day it stands out

among them.

Cassava is considered as one of the most important calorie-producing crops in the

tropics. It is adapted to a wide range of environments, and tolerant to drought and

acidic soils as well as being an efficient producer of carbohydrates (Jones, 1959;

Rogers and Appan, 1970; Kawano et al., 1978; Cock, 1982).

The importance of cassava is derived from its diverse use for human consumption,

animal feed and industrial application. According to Cock (1982) fresh roots of

cassava contain 30 to 40 % dry matter and approximately 85 % of the dry matter is

Chapter 2: Literature Review

12

starch. This percentage may vary according to cultivars, environment and plant age,

as pointed out by O’Hair et al. (1981), Rodriguez-Sosa et al. (1976) and Wholey and

Booth (1979). These researchers have stated that on fresh weight basis, root starch

concentration ranges from 5 to 40 %.

Starch, which consists 13-21 % amylose (Grace, 1977), is an important source of

carbohydrate for people who consume it. Young cassava leaves are also harvested

and used for human consumption as a vegetable or a constituent in the form of sauce

eaten along with staple meals (Lancaster and Brooks, 1983).

According to El-Sharkawy (2004) “About 70 % of the world cassava root production

(which is estimated to be 45 million metric tons of dry root annually) is used for

human consumption either directly after cooking or in processed forms; the

remaining 30 % is used for animal feed and other industrial products such as starch,

glucose and ethanol.”

2.1.1 Cassava Breeding

In the early 1970s a cassava breeding program was initiated by CIAT (Centro

Internacional de Agricultural Tropical) headquartered in Cali, Columbia. The main

objectives of CIAT were to improve yield potential and tolerance to diseases and

insect pests and adverse soil and environmental conditions (Kawano, 2003).

Cassava is one of the most efficient producers of carbohydrates under poor soil

conditions (Cook, 1982). According to this author, cassava crops are grown between

30° N and 30° S, in areas where mean temperature are greater than 18 °C and the

annual rainfall is greater than 750 ml. However, the growing conditions can differ for

different regions. Cassava can be grown in marginally, low fertility acidic soils under

variable rain-fed conditions ranging from less than 600 ml in semi-arid tropics (De

Tafut et al., 1997) to 1000 ml in sub-humid tropics (Pellet and El-Sharkawy, 1997).

Cassava generally grows in many soil types. However, cassava to some extent is

intolerant of saline or persistent water-logged conditions and it also does not tolerate

temperature at or below 10 °C (O’ Hair, 1990). According to Cock (1985) cassava


13

tolerates soil with a pH of as low as 4.4 provided the aluminium level does not

exceed 80 % saturation and it also tolerates a pH as high as 8.0.

Cassava is propagated vegetatively from mature woody stem cuttings. Usually 15-30

cm long stem cuttings are planted horizontally, vertically, or inclined on flat or

ridged soils at densities ranging from 5000 to 20000 cuttings per hectare depending

on cropping system and purpose of production (Keating et al., 1988).

The harvesting time of roots differ and is dependent on cultivars, purpose of use and

growing conditions. However, generally it is 7-24 months after planting (El-

Sharkawy, 2004). According to El-Sharkawy, fresh roots have to be used

immediately after harvest for consumption, processed for starch extraction, dried for

flour production, roasted for food products and/or used for animals. This is because

roots tend to perish and deteriorate rapidly after harvest. However, pre-harvest

pruning in the weeks before harvest decreases root deterioration because of increases

in total sugar/starch ratio in roots (Van Oirschot et al., 2000).

Although, propagation by stem cuttings is the most common practice, it can also be

propagated from seeds (Ceballos et al., 2004). According to the authors seeds are

generated through crossing in breeding programs and this result in creating new

genetic variation. The use of seeds in commercial cassava production is a promising

option to obviate constraints, particularly diseases associated with vegetative

propagation (Iglesias et al., 1994).

2.1.2 Effect of Environmental Factors on Cassava

Like every other crop, cassava also needs certain favourable environmental

conditions for proper growth and development. According to Sriroth et al. (2001)

although cassava has the reputation of being a drought tolerant crop, when deprived

of water, plant and root development are affected. This eventually translates to an

altered starch synthesis expressed by variation in starch quality. According to these

authors, water stress plays an important role in the starch quality. Usually water

stress in early plant development retards growth and this only resumes after

immature plant receives sufficient water. In mature plants, environmental conditions


14

affect the starch quality prior to root harvest. This is especially noticeable after the

onset of rain after a stress period as indicated by reduced starch paste.

Intensive production of cassava requires favourable climatic and soil conditions.

However, according to Silvestre (1989) in comparison with other plants, cassava

exhibits considerable tolerance to these factors. Silvestre has stated that the ideal

conditions for growing cassava are rain every two or three days, with large amounts

of sunshine in between, without a marked dry season or with a dry season lasting no

longer than two months.

The factors responsible for starch variation in roots are either influenced by genetic

or those influenced by environmental condition including the amount of rain

(Moorthy and Ramanujam, 1986; Asaoka et al., 1991; Asaoka et al., 1992; Defloor

et al., 1998; Sriroth et al., 1999; Santisopasri et al., 2001). Therefore, it becomes

extremely important to determine the suitable environmental condition for a

particular cultivar to be planted.

2.1.3 Cassava Varieties

When choosing cassava variety, the main criterion is of course productivity in terms

of dry matter or starch. But it can be the earliness of maturity as well, as some

varieties mature earlier than the others (Silvestre, 1989). According to Silvestre,

research centers and stations have programmes aimed at creating highly productive,

disease-resistant material and selecting varieties that are adapted to the ecological

situation with which they are concerned.

The classification of cultivars (varieties) is usually based on pigmentation and shape

of the leaves, stems and roots (Rogers and Appan, 1973). The cassava varieties in

Fiji are also identified using the above mentioned classifications.

According to Mason (1956), variety collection for Fiji cassava started in 1950,

resulting in the identification of fourteen Fijian varieties. However, Mason’s paper

has actually described sixteen varieties that were present at that time. These varieties

were; Vulatolu, Vulatolu 2, Merelesita, Merelesita 2, Yabia Damu, Yabia Vula,


15

Niumea, Coci, Sokobale, Aikavitu, Kasaleka, Katafaga, Belesilika, Manioke, Yasawa

and Macuata.

Currently, Koronivia Research Station (KRS) is growing the following twenty-eight

varieties; Vulatolu, Vulatolu 2, Merelesita, Merelesita 2, Yabia Damu, Yabia Vula,

Niumea, Coci, Sokobale, Aikavitu, Kasaleka, Katafaga,, Belesilika, Manioke, Yasawa

Vulatolu, Malaya (Macuata), Ro Tubuanakoro, Coci (selection), Vulatolu (Dalip

Singh), H.165, H.97, Tilomuria No.3, Tavioka Falawa, Navolau, New Guinea,

Lomaivuna, Beqa, Hawaii and Kadavu (Nauluvula, 2009, pers. comm.).

Although, all of these cassava varieties can be grown together in the same ecological

conditions, their yield may differ indicating that different varieties may thrive better

if planted in those environmental conditions more suited. Hence, as Silvestre (1989)

has stated the best varieties for a given ecological situation are those that maintain a

good balance between the production of new leaves, stems and that of tuberous roots

in that particular environment.

2.2 Production of Ethanol

Ethanol (ethyl alcohol) is considered renewable when produced from sustainable

agricultural sources. It has the potential for reducing greenhouse gas emissions,

which is significantly dependent on the feedstock and the technology used in the

production process as well as distribution and blending procedures.

Several studies have been conducted on ethanol production from different feedstock

and production processes. Jamai et al. (2001) compared ethanol production using

glucose from calcium alginate-immobilized as well as free cells of Candida

tropicalis and Saccharomyces cerevisiae. The results indicate identical rate of

ethanol production in free and immobilized Saccharomyces cerevisiae YMES2.

However, immobilized Saccharomyces cerevisiae Σ1278 showed increased ethanol

production when compared to the free cells whereas for immobilized Candida

tropicalis YMECI4 and Y1552 ethanol production rate decreased by 25 % and 27 %

respectively. Ethanol concentration at the end of the fermentation cycle is similar for


16

free and immobilized Candida tropicalis, but in order to reach this concentration

more time is required under immobilized condition. Free and immobilized cells of

Saccharomyces cerevisiae achieve complete fermentation at the same time but the

immobilized Candida tropicalis require 2-3 hours more than free Candida tropicalis

to complete the fermentation process.

Jamai et al. (2007) used a different feedstock, which was corn soluble starch, to

produce ethanol using Candida tropicalis. This is capable of fermenting starch at low

rates. Starch liquefaction was sufficient to drive the fermentation of starch to ethanol

by Candida tropicalis YMEC14, reaching ethanol yields comparable to those

obtained by other groups when using cell surface-engineered Saccharomyces

cerevisiae strains that display both α-amylase and glucoamylase. In order to display

the hydrolytic role of Candida tropicalis glucoamylase, ethanol production from

starch was compared between Candida tropicalis YMECI4, Saccharomyces

cerevisiae YMES2 and the amylolytic yeast Saccharomyces occidentalis ATCC

26077. The collective results are indicated in the table below.

Table 2.1: Kinetics parameters of ethanol production from starch following

growth of C. tropicalis, S. cerevisiae, and S. occidentalis in the

presence and absence of α-amylase treatment (Source: Jamai et al.,

2007)

C. tropicalis S. cerevisiae S. occidentalis

(+) (-) (+) (-) (+) (-)

YP/S

(g E/g S)

0.38 ±

0.02

0.19 ±

0.01

0.13 ±

0.01

0.025 ±

0.0005

0.26 ±

0.03

0.14 ±

0.01

YP/X

(g E/g B)

2.66 ±

0.12

1.42 ±

0.06

1.28 ±

0.04

0.26 ±

0.02

1.69 ±

0.04

0.89 ±

0.03

YX/S

(g B/g S)

0.14 ±

0.01

0.13 ±

0.01

0.10 ±

0.01

0.10 ±

0.01

0.15 ±

0.01

0.16 ±

0.02

QP

(g E/l h)

0.51 ±

0.03

0.23 ±

0.02

0.18 ±

0.02

0.03 ±

0.002

0.36 ±

0.03

0.16 ±

0.02

Notes: YP/S and YP/X: yield factor of ethanol on substrate and biomass respectively; YX/S: conversion rate of starch to biomass; QP: volumetric ethanol productivity; (E): ethanol, (S): substrate, (B): biomass; Mean ± SD (n=3)


17

From this result it is evident that in order to get maximum ethanol yield from starch

fermentation. Starch first needs to undergo hydrolysis, that is, be liquefied as well

subjected to saccharification. Although, there are many organisms or yeast present

such as Candida tropicalis which have the ability to ferment starch directly, the

ethanol yield is low. Also evident is certain organisms have the ability to produce

more ethanol than others.

According to Verma et al. (2000) direct fermentation of starch to ethanol can be

carried out effectively by co-culture of Saccharomyces diastaticus and

Saccharomyces cerevisiae 21 with fermentation efficiency of 93 % and production of

24.8 g l-1 ethanol using raw unhydrolyzed starch in single-step fermentation. When

using only Saccharomyces diastaticus on α-amylase and glucoamylase treated starch

the efficiency was 78 % whereas using Saccharomyces cerevisiae 21 the efficiency

was 85 %. Starch is considered as a renewable agricultural substrate and can be

easily hydrolyzed to fermentable sugars using amylolytic enzymes. The most

commonly used distiller’s yeast Saccharomyces cerevisiae is unable to hydrolyze

starch directly and thus there is a need to use α-amylase and glucoamylase to convert

starch to fermentable sugar. Bioconversion of starch to ethanol can also be achieved

by using recombinant strain of Saccharomyces cerevisiae (Nakamura et al., 1997).

The results indicate a maximum ethanol production of 24.9 g l-1 from 100 g l-1 starch

medium using Saccharomyces cerevisiae SR 93.

The use of thermotolerant yeasts for ethanol production has also been described by

many researchers (Banat et al., 1996; Nigam et al., 1998 and Ryu et al., 1988).

Themotolerant yeast strains are of great advantage especially for tropical countries

and for summer months. These yeasts help reduce cost as well as save energy by

avoiding the need to cool reactors and in having faster fermentation rates which

makes the process economical. Sree et al. (1999) used various starchy substrates

such as sweet sorghum, sweet potato, wheat flour, rice starch, soluble starch and

potato starch to produce ethanol using thermotolerant yeast isolate (VS3) by

simultaneous saccharification and fermentation process. The fermentation process

employed was Solid Substrate Fermentation. The results obtained are indicated in the

table below.


18

Table 2.2: Ethanol concentration and left over sugar in solid substrate

fermentation from various starchy substrates using thermotolerant

yeast (VS3) (Source: Sree et al., 1999)

Substrate Ethanol at

37 ˚C

(g/100 g

substrate)

LOS at 37˚C

(g/100 g

substrate)

Ethanol at

42˚C

(g/100 g

substrate)

LOS at 42˚C

(g/100 g

substrate)

Sweet

Sorghum

8.2 0.25 7.5 0.45

Sweet Potato 5.0 0.5 3.5 1.0

Rice Starch 10.0 1.0 3.5 5.0

Wheat Starch 6.0 0.5 3.2 1.4

Potato Starch 4.0 1.5 2.0 2.5

Soluble Starch 3.5 2.5 1.5 3.0

LOS = Left over sugar

There is tremendous scope for Solid Substrate Fermentation because of the following

potential advantages (Gibbons et al., 1986; Kargi et al., 1985 and Amin, 1992): (i)

less requirement of water, (ii) smaller volumes of fermentation mash, (iii) less

physical energy requirement, (iv) less capital investment, (v) less operating costs and

lower space requirement, (vi) reduced reactor volumes, easier product recovery, less

liquid waste to be disposed of hence less pollution problems. Pandey (1994),

Lonsane and Krishnaiah (1994) and Sree et al. (1999) also agree that Solid Substrate

Fermentation is more suitable and economical than conventional submerged

fermentation for producing a variety of fermentation products.

Ocloo and Ayernor (2008) investigated the physical and chemical changes besides

yeast growth in alcoholic fermentation of sugar syrup from cassava flour. Cassava

flour was hydrolyzed using rice malt to produce sugar syrup which was then

fermented at 28-30 ºC for 1, 2, 3, 4 and 5 days using Saccharomyces cerevisiae

(baker’s yeast). “Results showed that pH values decreased with increased total

acidity with concomitant increase in yeast growth (biomass) and alcohol contents of

the fermenting sugar syrup. There were deceases in soluble solid contents, refractive


19

indices of the fermenting medium and increase in volatile acids (as acetic acid) with

increased alcoholic fermentation” (Ocloo and Ayernor , 2008).

The method for conversion of starch to alcohol is usually classified in three groups:

(i) combined saccharification and fermentation of soluble starch by genetically

engineered yeast, (ii) addition of amylolytic enzymes, mostly glucoamylase, to the

broth and (iii) use of mixed culture, either free or co-immobilized for starch

degradation and fermentation (Ülgen et al., 2002).

Ülgen et al. (2002) had used genetically engineered yeast to produce ethanol from

starch. Single step starch fermentation was investigated using a genetically

engineered recombinant Saccharomyces cerevisiae strain YPG/AB. Their findings

revealed a 34 % increase in ethanol production by YPG/AB in 40 g l-1 initial starch

containing medium supplemented with 4 g l-1 glucose.

Amutha and Gunasekaran (2001) used method (iii) description of conversion of

starch to ethanol. Ethanol was produced by batch fermentation using liquefied

cassava starch and co-immobilized cells of Saccharomyces diastaticus and

Zymomonas mobilis and compared to the ethanol concentration with immobilized

Saccharomyces diastaticus. The results showed that co-immobilized cells produced

46.7 g l-1 ethanol from 150 g l-1 liquefied cassava starch while the immobilized cells

of Saccharomyces distaticus produced 37.5 g l-1 ethanol. Ethanol concentration was

also higher when it was produced using immobilized cells rather than free cells of

Saccharomyces diastaticus and Zymomonas mobilis.

Looking at the above review, it can be well noted that the ethanol yield varies. These

variations are a result of experimental procedures employed, enzymes and catalysts

used, conditions such as temperature and pH at which fermentation has commenced.

All of these factors tend to affect ethanol yield.


20

2.3 Ethanol as Fuel

Ethanol, together with biodiesel provides a suitable substitute for fossil fuels. Ethanol

made from biomass provides unique environmental, economic strategic benefits and

therefore can be considered as a safe and clean liquid fuel alternative to fossil fuels

(Chandel et al., 2007). The advantages and some of the drawbacks associated with

the use of ethanol as fuel as concluded by various researchers are discussed below.

An issue paper prepared by Environment Australia (2002) has provided a basis for

discussion of some of the issues associated with the use of ethanol as automotive

fuels. The first part discusses the background information on the use of ethanol as

automotive fuel and the other part raises issues for comments on vehicle operability

and the effect on engines, environmental performance and any health and safety

implications. Ethanol can be blended with petrol to form ethanol-petrol blends. These

are usually stated as volumetric ratio of ethanol to petrol. Some of the most common

of these blends are as follows:

1. 10 % ethanol blended with 90 % petrol (known as E10)

2. 85 % ethanol blended with 15 % petrol (known as E85); this blend is used in

some states of the US and requires a particular vehicle technology known as

‘Flexible Fuel Technology’ (FFT)

3. 20-24 % ethanol blended with 80-76 % petrol (known as E22); this blend is

used in Brazil and requires specific vehicle optimization (recalibration and

component changes) for 22 % ethanol.

4. 100 % ethanol (E100); this is used in Brazil and requires vehicle technology

dedicated to the fuel.

Ethanol is produced in two forms: hydrated and anhydrous. Hydrous ethanol which is

95 % ethanol with the rest being water is suitable for use as a straight spark ignition

(SI) fuel in warmer climates or for blending as a 15 % emulsion in diesel. Anhydrous

ethanol is 100 % ethanol that is used for blending with petrol. Ethanol blends tend to

result in reduced emissions of carbon monoxide (CO), hydrocarbons (HC),

particulate matter (PM) and certain known carcinogens. However, ethanol blends are

likely to increase emissions of aldehydes, particularly acetaldehyde.


21

Within the Australian context, the use of E10 has been found to result in a decrease

of emissions of CO by 32 %, HC by 12 % and slight increase in nitrogen oxide

(NOx) by 1 %. There were increases in non-regulated toxics such as acetaldehyde

(180 %) and formaldehyde (25 %) and decrease in non-regulated toxics: 1-3

butadiene (19 %), benzene (27 %), toluene (30 %) and xylene (27 %).

An Australian life-cycle analysis work had revealed that E10 blends are greenhouse

neutral (Beer et al., 2001). The same study indicates that using E10 blends lead to

decreased tailpipe emission of hydrocarbons and NOx by 25 % and 15 %

respectively, but there was no change to the particulate matter. Studies by Beer et al.

(2002) have confirmed that renewable fuels such as biodiesel and ethanol emit larger

quantities of CO2 than conventional fuels. However, since most of this is from

renewable carbon stocks, that fraction does not count towards greenhouse gas

emission from the fuel.

Duncan (2002) stated that generally, modern vehicles operate satisfactorily on blends

up to 10 % ethanol in petrol and are designed to have fuel systems materials

compatible with the blends. This is because most modern vehicles are now designed

to use fuels containing some level of blended oxygenate including ethanol. Ethanol

will corrode mechanical components that are made of copper, brass and aluminium

due to the water solubility in ethanol (Wu et al., 2004).

Use of 10 % ethanol in petrol results in 3.6 % increase in volumetric fuel

consumption due to lower energy content of ethanol. The energy efficiency of

modern vehicles remains similar for blends as well as petrol. Also there was no

significant difference in exhaust emissions.

This is because modern vehicles are fitted with functional exhaust sensors (oxygen

sensors) and catalysts which help determine if the air fuel ratio of a combustion

engine is rich or lean. The oxygen sensors send information about the difference in

oxygen concentration in exhaust and the oxygen in air to the Electronic Control Unit

or ECU, which adjusts the amount of fuel injected into the engine to compensate for

excess air or excess fuel. The ECU tries to maintain the air-fuel ratio to

stoichiometric.


22

However, in older vehicles without catalysts and in vehicles without fully functional

emission control systems, use of ethanol blends will likely result in reduced CO

emissions compared to petrol, slight reductions in HC and slight increases in NOx.

This is mostly due to the fact that engine would be running on lean conditions.

According to Duncan’s (2002) result an E10 blend will produce 0.1 % less CO2 from

vehicle tailpipes than petrol when used at an equivalent efficiency. CO2 emitted

during manufacture of ethanol ranges between 30 to 90 % of CO2 generated as

tailpipe emission after combustion depending on process technologies and

accounting protocols used. On this basis and with an increase in energy efficiency of

vehicles by 1 %, net CO2 emissions using an E10 blend will be 1.5 to 5.5 % lower

than that of petrol. Although, these reductions seem low, lowering the ethanol

manufacturing emission levels could result in higher reduction in CO2 emissions.

This could depend considerably on (Duncan, 2002):

� Type of feedstock used for ethanol production, location and yields

� Treatment of ethanol production as a by-product of the crop or the objective

of growing the crop

� The procedures used for greenhouse gas accounting

� CO2 sequestration assumed for process by-products, for example, fertilizers

and stock feed

� Energy products used in processing: renewable fuels from process by-

products or non-renewables such as coal, oil or natural gas

Hence, a 10 % ethanol concentration by volume is appropriate because of vehicle

performance and compatibility and consistency with international experience.

Hsieh et al (2002) used a commercial SI engine running on various ethanol-petrol

blends (0 %, 5 %, 10 %, 20 %, and 30 %) to investigate the engine performance and

pollutant emissions resulting from these various blends. The collective results

showed a 10-90 % reduction in CO and 20-80 % in HC as a result of leaning effect.

There was an increase in CO2 tailpipe emissions by 5-25 %. As expected the fuel

consumption increased with increasing amount of ethanol in the blends.


23

A four stroke, four cylinder SI engine (type TOYOTA, TERCEL-3A) was used by

Al-Hasan (2003) in order to study the effect of using unleaded gasoline-ethanol

blends on SI exhaust emission and engine performance. The results obtained by Al-

Hasan showed increase in fuel consumptions by 5.7 % whereas the CO and HC

emissions decreased by 46.5 % and 24.3 % respectively. The CO2 emissions

resulting from the engine increased by 7.5 %. This is due to the fact that blended

fuels enable the fuel to burn more efficiently. Of the ten test blends ranging from 0 %

to 25 % ethanol in increments of 2.5 %, it was found that 20 % ethanol fuel blend

gave best results for the various parameters that were measured. Ethanol addition

also leads to the increase in brake power, brake thermal efficiency and volumetric

efficiency by about 8.3 %, 9 % and 7 %, respectively.

Pikūnas et al. (2003) studied the engine performance and pollution emission of a SI

engine using ethanol-petrol blended fuel (E10) and pure gasoline. The results for

engine test indicated that when ethanol-petrol blended fuel is used, the engine power

and specific fuel consumption of engine slightly increases; CO emission decrease

dramatically; HC emission decreased in some engine working conditions but

increase at maximum load; and CO2 emissions increases because of the improved

combustion and the amount of incomplete combustion products decreases. The

reduction in CO emission is due to the leaning effect. Addition of ethanol to blended

fuels provides more oxygen for the combustion process and this leads to the so called

“leaning effect”. Quantitatively the emission levels indicate a decrease in CO by 15

% at lower engine output and revolutions in comparison with petrol. However,

increase in engine output and revolutions indicate the difference of CO emission

increased by 30 %. The results have shown slight increases of approximately 1-2 %

in specific fuel consumption of E10 over petrol. This is attributed to the fact that

ethanol’s heating value is 1.6 times less than that of petrol.

Similarly, Butkus and Pukalskas (2004) used engine test facilities to investigate the

effect of 3.5 % and 7 % ethanol in the fuel blends and special additives in the engine

performance and pollution emissions of SI engine. The results indicated a general

trend of reduction in the emissions as well as increase in the octane rating. However,

from the results, conclusions were drawn that the maximum CO emissions results


24

due to the use of pure petrol. They concluded that ethanol used in fuel blend with

petrol had a positive influence on the engine performance and exhaust emission.

Theoretical as well as experimental investigation into the effects ethanol blends on

the performance and exhaust emissions have also been carried out by Bayraktar

(2005). Experimental application was performed on blends containing 1.5 %, 3 %,

4.5 %, 6 %, 7.5 %, 9 %, 10.5 % and 12 % by volume ethanol with petrol. Theoretical

investigation was carried out on a quasi-dimensional SI engine cycle model

developed by the author on blends up to 21 %. Experimental results indicated that

among the various blends, blend of 7.5 % ethanol was the most suitable in terms of

engine performance as well as reduction in CO emissions. However, in comparison

with the theoretical results blend containing 16.5 % ethanol was found to be more

viable. This difference experimental and theoretical can be “attributed to the water

content of ethanol: as known, in actual conditions, water may cause phase

separation, therefore, power loss and this can be observed experimentally, however,

this negative effect of water cannot be considered in theoretical model” (Bayraktar,

2005). Therefore, if purity of ethanol is increased and phase separation prevented, the

ethanol content can be increased towards 16.5 % to yield the best results.

It is well known that addition of ethanol to ethanol-petrol blends tends to reduce the

heating value and as a result the fuel consumption of ethanol-petrol blends tend to

increase. Bayraktar (2005) results showed that the specific fuel consumption

measured at compression ratio 7.75 and 8.25 were lower when operating on ethanol-

petrol blends than on petrol. The specific fuel consumption decreased by 5.59 % and

4.94 % at ε=7.75 and 8.25, respectively, when operating on 7.5 % ethanol blended

with petrol. The CO mole ratio showed a reduction of 44.26 % (at ε=7.75) and 41.67

% (at ε=8.25) whereas theoretical reductions were 61.82 % (at ε=7.75) and 68.16 %

(at ε=8.25). The obvious reasons suggested by the author were that increase in

ethanol content leads to more complete combustion and rise and flame temperature

due to stoichiometric combustion. The fact that the carbon content of ethanol is less

than petrol is another reason for such drastic reduction in CO.

Topgül et al. (2006) in their research studied the engine performance and exhaust

emissions when using unleaded petrol (E0) and ethanol-petrol blends (E10, E20, E40


25

and E60). The experiment was performed on a Hydra single-cylinder, four stroke, SI

engine by varying its compression ratio (8:1, 9:1, 10:1) and ignition timing at a

constant speed of 2000 rpm at wide open throttle (WOT).

The collective results indicated a decrease in CO emissions when using different

ethanol-unleaded petrol blends over the test range 8:1 to 10:1 compression ratio. E40

gave the best results for reduction of CO emissions by 31.8 % at 9.1 compression

ratio. Whereas, E60 showed a decrease in CO emission by 19.8 % and 22.3 % mean

average values at 8:1 and 9:1 compression ratio respectively.

The authors also noted that increasing ignition temperature results combustion

process to occur earlier in the cycle that leads to decrease in exhaust temperature. As

a result of this the HC emission increased. However, retarding the ignition timing

leads to reduced HC emissions. This was seen when E60 caused a 31.45 % reduction

in HC emission at compression ratio 10:1.

Celik’s (2008) work included determination of the suitable ethanol-petrol blend rate

in terms of performance and emissions for small engines. The other aim was to

investigate experimentally the performance as well as the emissions resulting from

the engine with suitable ethanol-petrol blended fuel at high compression ratio

without any knock.

In the initial stage of the test the engine was tested with E0, E25, E50, E75 and E100

fuels with its original compression ratio (6:1), 2000 rpm, full throttle opening and air

excess ratio of 1.0. The results obtained for this test indicated slight power increases

in E25, E50 and E75 when compared to E0. However, the power increase starts to

decrease when ethanol content was raised to more than 50 %. When looking at the

emissions, CO and CO2 emission deceased. CO emission was 3.76 %, 2.65 %, 2.06

%, 1.24 % and 0.73 % for E0, E25, E50, E75 and E100 fuels, respectively. In

addition the CO2 emission were 13.25 %, 12.14 %, 11.62 %, 10.25 % and 9.51 % for

E0, E25, E50, E75 and E100 fuels, respectively. There was decrease in the HC

emission for E0, E25 and E50 but for E75 and E100 the HC emission increased. The

NOx emission decreased as the ethanol content in the fuel was increased.


26

In the second stage the testing was performed with E0 and E50 at compression ratios

6:1, 8:1 and 10:1 with full load in the range of 1500-4000 rpm at intervals of 500

rpm. E0 fuel could only be tested at compression ratio of 6:1 since it caused knock at

a compression ratio 8:1, full throttle opening and low engine speeds. However, E50

enabled the engine to run without knock at high compression ratio (10:1) at full

throttle load and all speeds. E50 fuel lead to the increase in engine power by 29 %

when compared to E0 fuel whereas the specific fuel consumption, CO, CO2, HC and

NOx emissions reduced by 3 %, 53 %, 10 %, 12 % and 19 %, respectively.

Abdel-Rahman and Osman (1997) also used blended fuels in the ratio of 10 %, 20 %,

30 % and 40 % in variable compression ratio engine. Their results showed that under

various compression ratios of engine, the optimum blend rate was found to be 10 %

ethanol with 90 % petrol.

A conventional engine under various air-fuel equivalence ratios (λ) was used by Wu

et al. (2004) to study its performance as well as pollution emissions using ethanol-

gasoline blends (E0, E5, E10, E20 and E30). The result of engine performance tests

showed that torque output improves when using ethanol-gasoline blends. However,

there are no noticeable changes with the brake specific heat consumption. HC and

CO emissions reduced with the increase of ethanol content in the blended fuel. The

maximum CO2 emission was obtained at λ~1 but the smallest amount of CO2

emission was obtained with E30. The study indicated that using 10 % ethanol fuel

could reduce pollution emission efficiently.

Moreover, according to Yüksel and Yüksel (2004) ethanol-petrol mixtures

containing up to 20 % ethanol by volume, can be used safely without causing any

damage to the motor. It is observed that phase separation tends to occur in ethanol-

petrol mixtures if the amount of water present in the mixture is over a certain limit.

Water in blended fuels results in corrosion problems on mechanical components

especially those components made from copper, brass or aluminium. Usually

gasoline containing less than 20 % ethanol by volume and is aromatic in character

and said to be more stable. According, to the authors phase separation depends on the

ethanol and water content of the blends, environmental temperature and composition

of gasoline. However, phase separation temperature can be reduced if higher


27

aliphatic alcohols such as tertiary butyl alcohol, benzyl alcohol, cyclohexanol or

toluene are added to gasoline-alcohol blends (Ferfecki and Sorenson, 1983 and

Karaosmanoglu et al., 1992).

From the above literature review, it is understood, that ethanol-petrol blends can be

used to effectively reduce pollutant emissions in both engines with modification and

without modification depending on the percentage of ethanol blended with petrol.

However, it is seen that with increasing percentage of ethanol there is decrease in

heating value and as a result an increase in fuel consumption.

28

Chapter 3 Methodology

3.0 Overview

This chapter gives the details of the procedure that was employed in order to produce

ethanol from some cassava varieties available in Fiji and the ethanol yield from these

different varieties. It also outlines the methodology for the determination of the fuel

properties of ethanol-petrol blends as compared to neat petrol and the fuel economy

and the emissions resulting from the use of certain ethanol-petrol blends.

3.1 Cassava Varieties in Fiji

The cassava varieties that were used for ethanol production were obtained from two

different research stations of the Ministry of Primary Industries in Fiji. One was

Koronivia Research Station (KRS) situated 18˚ 32’811” S and 178˚ 32’133” E and

the other was Dobuilevu Research Station (DRS) situated 17˚ 33’620” S and 178˚

14’736” E. These two locations are indicated in the map of Fiji in Figure 3.1.

Figure 3.1: Map of Fiji showing the collection points of cassava varieties

Chapter 3: Methodology

29

The ten cassava varieties obtained were; Niumea, Sokobale, Beqa, New Guinea,

Coci, Vula Tolu, Yabia Damu, Merelesita, Nadelei and Navolau. The variety

Sokobale was not available at DRS therefore; only nine varieties were used for

ethanol production from this location. The cassava varieties obtained from KRS and

DRS were approximately 12 months old. Cassava from these two sites were obtained

in order to determine whether, apart from varieties or genetic constituent of plants,

location also played a part in starch yield which will then influence the ethanol yield.

3.1.1 Dry Matter Content of Cassava Roots

Dry matter content (DMC), retention of leaf and starch accumulation in roots has an

important impact on dry matter yield (Pérez et al., 2002). The percentage of starch

and starch yield are closely related to dry matter percentage. Therefore, this is one of

the factors that need to be determined in order to identify the best cassava variety for

starch and ethanol yield.

DMC was determined according to the procedure described by Benesi (2005). The

roots of different cassava varieties were analyzed for DMC within 12 hours of

harvesting. The roots were peeled, cleaned and then shredded into fine slice before

100 g of these were weighed in a Petri dish ( 1w ). The Petri dish was then placed in

an oven at a temperature of 65 ˚C for 72 hours. The samples were removed after 72

hours and weighed immediately ( 2w ). DMC was calculated using the equation 3.1:

DMC (%) = %1001

2 �ww (3.1)

3.1.2 Starch Extraction from Cassava Roots

The extraction of starch from cassava was done according to the method described

by Birse and Cecil (1980) however, some parts of the method was modified. A

flowchart of starch extraction is shown in Figure 3.2. Cassava roots were washed,

peeled then washed again before the roots were chopped into approximately 1 cm

cubes. The weight of the chopped cassava ( 3w ) was taken before pulverizing it in a


30

high speed blender for 5-10 mins. The pulp was then suspended in ten times its

volume of water, stirred for about 5 mins before filtering using a double fold cheese

cloth. The filtrate was left to stand for about 6 hours before the starch settled and the

liquid potion discarded. Water was then added to the sediment and the whole process

was repeated. The starch was then dried at 50 ˚C for 24 hours and its weight

measured ( 4w ).

Cassava Root ↓

Peeling ↓

Washing ↓

Chopping/Blending ↓

Mixing with Water ↓

Filtering ↓

Settling ↓

Starch Washing ↓

Settling/Dewatering ↓

Drying ↓

Cassava Starch Figure 3.2: Flowchart for cassava starch production

The starch yield was determined using the equation 3.2:

Starch Yield (%) = 1003

4 �ww

% (3.2)


31

3.1.2.1 Moisture Content of Starch

Moisture content of the extracted starch was determined according to the method

described by Benesi (2005) however the quantity of cassava starch to be analyzed

was increased.

Approximately 10 g of cassava starch ( 5w ) was dried in an oven at 105 ˚C for 24

hours. After 24 hours the samples were cooled in a desiccator and weighed

immediately ( 6w ). The moisture content was determined using equation 3.3:

Moisture content (%) = 1005

65 ��w

ww% (3.3)

3.1.2.2 Ash Content of Starch

Ash content was determined according to the method described by International

Starch Institute (1999 a). Clean ashing crucibles were heated in the furnace for

approximately half an hour at 900 ˚C. The crucibles were cooled in a dessicator to

room temperature and weighed ( 0w ). Approximately 5 g of the starch sample was

uniformly distributed in the ashing crucible and weighed ( 7w ). The samples were

then incinerated on a bunsen burner until it completely carbonised before placing the

ashing crucibles in the furnace for 5 hours at 900 ˚C. After incineration, the samples

were cooled to room temperature in a dessicator and weighed ( 8w ). Ash content of

starch was determined using equation 3.4:

Ash content (%) = 10007

08 ��

wwww

% (3.4)

3.1.2.3 pH Determination of Starch

The pH of starch was determined according to the method described by International

Starch Institute (1999 b). Approximately 5 g of starch was mixed with 20 ml of


32

distilled water. The starch was then allowed to settle for 15 mins before the pH of the

water phase was measured using a calibrated pH meter. The calibration of the pH

meter was done by turning on the meter at least 30 minutes before use, allowing it to

warm up. The pH electrode was removed from the storage solution of 3.5 M KCl

(potassium chloride) solution and rinsed with distilled water. The pH 4 and 7 buffer

were prepared by dissolving the buffer 4 and buffer 7 standard tablets in 100ml of

distilled water. The electrode was submerged into the pH 7 buffer and

“CALIBRATE” button was pressed. When the pH icon had stopped flashing,

CALIBRATE button was again pressed. The electrode was rinsed with distilled

water. Calibrating steps were repeated with pH 4 buffer. The electrode was removed

and rinsed with distilled water again to remove buffer solution preventing it from

contaminating the sample. The electrode was then submerged into the sample and

MEASURE button was pressed to get the pH of the sample.

3.2 Ethanol Production from Cassava Starch

The cassava starch that was extracted from Niumea, Sokobale, Beqa, New Guinea,

Coci, Vula Tolu, Yabia Damu, Merelesita, Nadelei and Navolau from the two

locations were used as feedstock for ethanol production. This was done to determine

the best ethanol yielding cassava variety from the ten varieties of cassava used.

3.2.1 List of Equipment and Reagents A PerkinElmer’s Lambda 25 UV/VIS spectrophotometer was used for determination

of reducing sugars and a PerkinElmer’s Clarus 500 Gas Chromatography (GC) for

determination of ethanol concentration. The reagents that were used for ethanol

production are given in Table 3.1.


33

Table 3.1: Reagents used for ethanol production

Reagents Source α-amylase Type XII-A From Bacillus

Licheniformis (830 U/mg)

Sigma Aldrich, Australia

Amyloglucosidase from Aspergillus Niger (66.6

U/mg)

Sigma Aldrich, Australia

Yeast (Saccharomyces cerevisiae) Sigma Aldrich, Australia

3, 5-Dinitrosalicylic Acid Sigma Aldrich, Australia

Sodium Hydroxide Anhydrous Pellets Sigma Aldrich, Australia

Sodium Potassium Tartrate Tetrahydrate Sigma Aldrich, Australia

Hydrocloric Acid (HCL) Sigma Aldrich, Australia

Absolute Ethanol Unilab, Ajax Finechem, Australia

95 % Ethanol Unilab, Ajax Finechem, Australia

Glucose Sigma Aldrich, Australia

3.2.2 Preparation of Cassava Starch Solution

The extracted starch from the ten different varieties of Fijian cassava was made into

200 g l-1 concentration. Exactly 100 g of the starch was mixed in 100 ml of cold

water. While stirring the slurry was added to approximately 400 ml of gently boiled

water in a conical flask so that the final volume was 500 ml. This resulted in the

formation of gelatinized starch solution.

Gelatinized starch forms from the weakening of the inter and intra hydrogen bonds

which results from the rise in temperature of the starch solution (Wang, n.d). Starch

granules form hydrogen bonds within the same molecule and with the other

neighbouring molecules and these tend to be quite resistant to penetration of water

and hydrolytic enzymes (Wang, n.d). However, raising the temperature tends to

weaken these hydrogen bonds which results in water being absorbed and swelling of

starch granules. The resulting gelatinized starch is shown in Figure 3.3.


34

Figure 3.3: Gelatinized cassava starch solution

3.2.3 Treatment of Cassava Starch Solution to Simple Sugars

Starch substance comprises the major part of human diet and is synthesized naturally

in plants. Some of these plants are corn, potato, rice, sorghum, wheat and cassava.

Starch molecules are glucose polymers linked together by α-1, 4 and α-1, 6

glucosidic bonds (Kearsley and Dziedzic, 1995). Starch must first be hydrolyzed into

glucose units prior to alcohol fermentation by ethanologenic micro-organisms such

as yeast (Lee et al., 1992).

Starch hydrolysis can be achieved by using two enzymes: α-amylase and

amyloglucosidase or glucoamylase. Starch hydrolysis is achieved in two stages-

liquefaction and saccharification. During the starch liquefaction the α-amylase

enzymes work on the gelatinized starch slurry to partially hydrolyze the starch to

dextrin. Dextrin solutions are less viscous hence, the starch gel is liquefied. Dextrin

is short glucose chains, and small amounts of glucose and maltose (Kearsley and

Dziedzic, 1995). Dextrin can be further hydrolyzed to glucose by adding

amyloglucosidase and this stage is called saccharification.

α-amylase reacts endogenously with α-1, 4 glucosidic linkages of polysaccharides to

produce oligosaccharides whereas amyloglucosidase hydrolyzes exogenously the


35

non-reducing end α-1, 4, α-1, 6 and α-1, 3 glucosidic linkages of oligosaccharides to

produce glucose (Pazur and Ando, 1960).

3.2.3.1 Pre-treatment of Gelatinized Starch with α-amylase

The liquefaction of gelatinized starch was conducted at 65 ºC and the level of α-

amylase used was 200 U g-1 starch. In order for complete liquefaction to occur the

experiment was conducted for 2 hours. Montesinos and Navarro (2000) reported that

2 hours of liquefaction was absolutely necessary for complete starch hydrolysis using

raw wheat flour as substrate. Using this literature the current author also assumed the

same time would be required for the complete liquefaction of cassava starch.

Liquefaction of gelatinized cassava starch was conducted by adding α-amylase from

Bacillus licheniformis (830 U mg-1). One unit (1 U) liberates 1.0 mg of maltose from

starch in 3 mins at pH 6.9 at 20 ºC. The sample was then hydrolyzed at 65 ºC with

mild agitation for 2 hours as shown in Figure 3.4. Starch gel was liquefied to produce

dextrin which consists of shorter glucose chains containing glucose and maltose.

Usually hydrolysis of starch consists of two stages- liquefaction and saccharification.

However, in order to save time and make the process cost effective the

saccharification stage has been incorporated with the ethanol fermentation, a

technique known as simultaneous saccharification and fermentation (SSF).


36

Figure 3.4: Starch hydrolysis using α-amylase

3.2.4 Simultaneous Saccharification and Fermentation

SSF reaction mixtures contained liquefied cassava starch, yeast inoculum of 15 %

concentration and amyloglucosidase at 200 U g-1 starch. The amyloglucosidase used

was obtained from Aspergillus niger and has an activity of 66.6 U mg-1. Here 1 U

corresponds to the amount of enzyme that liberates 1 µmol glucose per min at pH 4.8

and 60°C. The 15 % yeast inoculum was prepared by hydrating 15 g of dry baker’s

yeast (Saccharomyces cerevisiae) in 100 ml of distilled water at 37 ºC for 10 mins.

Ocloo and Ayernor (2008) used a ratio of 1:30 yeast inoculum to cassava sugar syrup

while conducting fermentation. The current author used the same ratio for the

fermentation reaction.

Fermentation reaction mixtures were placed in one litre three necked round bottom

flasks. The flasks were topped with special air locks which allowed carbon dioxide to

escape. A syringe and needle was inserted on one of the other necks to draw out

samples and the last neck of the flask was fully blocked. The flask was then placed in

a water bath at 37 ºC as shown in Figure 3.5 and fermentation was allowed to

continue for five days.


37

Figure 3.5: Setup for Simultaneous Saccharification and Fermentation

3.2.5 Analytical Analysis

The samples obtained from the SSF were centrifuged within one hour of sampling.

This was accomplished by spinning the samples in a centrifuge at 6000 rpm for 5

mins. The liquid portion of the sample was then passed through a 0.45 µm filter

before the samples were analysed for reducing sugars and ethanol concentration.

3.2.5.1 Reducing Sugar Analysis Reducing sugars are classified as sugars that contain aldehyde groups and can be

oxidised to carboxylic acids or carbonyl group (Campbell and Farrell, 2009). All

common monosaccharide such as glucose and disaccharides such as maltose and

lactose are examples of reducing sugars (AUS-e-TUTE, n.d). The reducing sugar

concentration was determined using 3, 5–dinitrosalicylic acid (DNS) method as

described by Miller (1959).


38

3.2.5.1.1 Preparation of Dinitrosalicylic Acid Solution (DNS)

The DNS was prepared by dissolving 10 g of 3, 5-dinitrosalicylic acid in 2 M sodium

hydroxide solution. In a separate beaker, 300 g of sodium potassium tartrate was

dissolved in 300 ml of distilled water. To ensure that all the 3, 5-dinitrosalicylic acid

dissolved in 2 M sodium hydroxide solution the mixture was heated and stirred until

no particles were seen at the bottom of the beaker. The mixture of 3, 5-

dinitrosalicylic acid and 2 M sodium hydroxide solution was then added to potassium

sodium tartrate solution and thoroughly mixed. The final volume of this solution was

made up to 1 litre with distilled water in a volumetric flask. The DNS solution was

stored in a refrigerator at 4 ºC in a brown bottle.

3.2.5.1.2 Standard Curve

The calibration curve was prepared with 1 g l-1 of glucose solution. The dilution of

different concentration of 0.1 g l-1, 0.2 g l-1, 0.4 g l-1, 0.6 g l-1 and 0.8 g l-1 was made

by pipetting 1 ml, 2 ml, 4 ml, 6 ml, and 8 ml of the respective volume of stock

solutions into five different volumetric flasks and making the final volume to 10 ml

by adding distilled water. One volumetric flask contained only distilled water that

acted as the control.

To 1 ml of each of the dilution of the glucose solution 1 ml of DNS was added in a

test tube and capped. The test tube was then placed in a boiling water bath for 15

mins and then cooled in an ice bath. The 1 ml of the cooled sample was taken and 5

ml of distilled water was added. This was done to ensure that when the absorbance

was recorded at 540 nm it ranged from 0 to 1. The respective absorbance recorded

was plotted against the different glucose concentrations to make the standard curve

for glucose.

3.2.5.1.3 Determination of Reducing Sugar

The amount of reducing sugar (as glucose content) was determined by adding 1 ml of

the centrifuged and filtered sample from SSF to 1 ml of DNS reagent in a test tube.

The centrifuged and filtered sample from SSF needed to be diluted by adding


39

distilled water to it before adding DNS reagent to this. This was done to ensure that

the absorbance of the samples was within that of the standard curve. The capped test

tubes containing the sample and the DNS reagent was then placed in a boiling water

bath for 15 mins and then cooled in an ice bath. The 1 ml of the cooled sample was

taken and 5 ml of distilled water was added before absorbance was recorded at 540

nm. The absorbance was then converted to glucose concentration using the standard

curve.

3.2.5.2 Determination of Ethanol Concentration

The ethanol produced from the SSF process was determined using Gas

Chromatography (GC). The GC method utilized propan-1-ol as the internal standard

and was used to determine the ethanol concentration in g l-1 from the fermentation

supernatant.

3.2.5.2.1 Preparation of Internal Standard Spiking Solution, Analytical

Standards and Samples

The internal standard spiking solution used was reagent grade propan-1-ol. The

internal standard spiking solution was prepared by diluting propan-1-ol to 0.9 g l-1

with distilled water. The internal standard spiking solution was added in equal

proportion to the standards and the samples that were analyzed.

Five ethanol standards were prepared by diluting absolute ethanol with distilled

water to cover the range of 0.1 g l-1 to 5 g l-1 ethanol. Pure ethanol and distilled water

were chosen to minimize the possibility of non ethanol components interfering in the

determination of the ethanol content of the standards. To each of the standards 0.9 g

l-1 propan-1-ol (internal standard spiking solution) was added in the same ratio.

The fermentation sample that was obtained from SSF was appropriately filtered so

that liquid portion of the sample was obtained. The samples were then diluted by

adding distilled water. A tenfold dilution was suitable for samples obtained at 24, 28,

32, 48 and 72 hours whereas twenty-five fold dilutions brought the 96 and 120 hours


40

sample within range of the calibration curve. The internal standard spiking solution

was then added to each of the samples in the same proportion as the standards.

3.2.5.2.2 Procedure and Conditions of the Gas chromatography

The chromatograph consisted of a Clarus 500 Gas Chromatography (GC) with flame

ionization detector, and a 15 m X 0.53 mm stainless-steel column packed with 100 %

dimethylpolysiloxane. The column temperature was 175 °C and the detector

temperature was 275 °C. The oven was kept at 30 °C. Nitrogen carrier flow was 2.5

ml min-1. A sample size of 1 µl was injected, and peak area counts were printed out

on a chromatogram by the Agilent 3395 digital integrator.

It was ensured that all the standards and samples were at room temperatures before

any injections were carried out. Before injection with the standards and samples,

several injections were made using deionized water. The standards were then

injected from low concentration to high. Following this, again several injections were

made using deionized water before the batch of samples was injected. The GC was

recalibrated after every 6 hours of analysis.

The chromatograms obtained for the standards were used to plot the ratio of peak

area of ethanol to propan-1-ol against the different concentrations of ethanol.

Knowing the ratio of peak area of ethanol to propan-1-ol for the various samples

enabled the determination of the concentration of ethanol from the calibration curve.

3.3 Preparation of Ethanol-Petrol Blends

The blends that were tested were 10 % ethanol and 90 % unleaded petrol (E10), 15 %

ethanol and 85 % unleaded petrol (E15) and 20 % ethanol and 80 % unleaded petrol

(E20). These blends were prepared by mixing together the required proportion of 96

% ethanol and unleaded petrol.


41

3.3.1 Stability Testing of Ethanol-Petrol Blends

Ethanol and petrol blends tend undergo phase separation at low temperatures

especially if there is water present in the mixture. The level of phase separation that

can occur is determined by a number of variables, including the amount of ethanol,

the composition of the fuel, temperature of the environment and the presence of

contaminants. If phase separation of ethanol-petrol blends occur, this can have

detrimental effect on engines. Therefore, it is of extreme importance to test whether

the ethanol-petrol blends are stable before using in engines

To test the blend stability, 93 %, 95 %, 97 % and absolute ethanol was used to make

E10, E15 and E20 blends by mixing with unleaded petrol. These blends were placed

into 50 ml vials at a temperature of 25 °C. They were then thoroughly shaken and let

to stand for a period of 6 months at room temperature, which varied between 14 °C

and 28 °C.

3.4 Physical Properties

The physical properties of the ethanol-petrol blends determined were the density and

gross calorific value.

The gross calorific value (GCV) is the property that influences the fuel consumption.

In order to produce a specific amount of power output, a definite amount of fuel is

required. Fuels with higher GCV are desired as they have the tendency to produce

more power in the engine. Therefore, the determination of the GCV is of

significance.

The density is a parameter used extensively in the calculations and conversions of

other relevant parameters. A density measurement is needed to convert mass results

to volumetric results and vice-versa. Additionally, density is required in correcting

the volume for different temperatures.

These properties are of importance and these are needed in the determination of the

efficiency and fuel consumption of the test engine and the effect.


42

3.4.1 Gross Calorific Value

The GCV measures the energy content of a material. This property is also referred to

as heat of combustion or heating value. The unit used for measurement of the GCV

was kJ g-1. A conversion to a volume basis of kJ ml-1 was also made for calculation

purposes.

The determination of the GCV is necessary since fuels having high gross calorific

values are desired for better fuel economy. It is generally expected fuel with a high

GCV is likely to produce more power during combustion as a result decrease the fuel

consumption. The GCV was determined using the Gallenkamp Ballistic Bomb

Calorimeter shown in Figure 3.6.

In order to determine the GCV of different fuel samples, a known weight of sample

is ignited electrically and burnt in excess of oxygen in the combustion chamber. The

maximum rise in temperature of the sample is measured with the thermocouple and

galvanometer system. Therefore, by comparing the temperature rise of the weighed

sample of unknown GCV with that obtained when a sample of known calorific value

(standard) is burnt, the calorific value of the sample material can be determined.

Figure 3.6: Set up for Ballistic Bomb Calorimeter

Thermocouple

Combustion Chamber

Galvanometer

Oxygen Supply Pipe

Oxygen Pressure Gauge


43

3.4.1.1 Calibrating the Bomb Calorimeter

The bomb calorimeter needs to be calibrated before any measurements can be taken.

This is to ensure that accurate results are obtained. Calibration of the bomb

calorimeter was done by determining the correction for constant heat gain test and

calibration with standard sample. A correction of 0.2 divisions was made for heat

released due to firing current and a 5 cm length of firing cotton string. This amount

was subtracted from the total deflection of the galvanometer. The standard substance

used for calibrating the bomb calorimeter was benzoic acid. The calorific value of

benzoic acid is 26.44 kJ g-1. Samples of benzoic acid were burnt in the calorimeter

and the maximum deflection on the galvanometer was noted. The results obtained

were used to determine the calibration factor which indicated the relationship

between the galvanometer deflection and the amount of heat released by the

combustion of the sample. The calibration factor was determined using equation 3.5:

CF = 12

1144.26�� mkJg (3.5)

where,

CF = Calibration Factor (kJ/div)

1m = mass of benzoic acid (g)

1� = correction factor of 0.2 (div)

2� = deflection of benzoic acid (div)

3.4.1.2 Determination of Gross Calorific Value

In order to determine the calorific value of the fuel samples, the bomb calorimeter

was set as follows. Samples were weighed into the crucible and placed on the support

pillar in the base of the calorimeter. A 5 cm length of cotton thread was positioned

between the coils of the firing wire with the other end dipped in the centre of sample

in the crucible. The system was then enclosed and oxygen was let into the chamber at

a pressure of 30 atmospheres (atm) to ensure that complete combustion took place.


44

The bomb was then fired and the maximum deflection of the galvanometer was

noted.

The temperature rise of the bomb calorimeter was measured with the calibrated

galvanometer-thermocouple assembly. The GCV of the sample was determined

using the calibration factor as calculated using benzoic acid in kJ per division, the

mass of sample burnt and the deflection of the sample. Equation 3.6 was used to

calculate the energy content orGCV of the sample fuels.

GCV = 2

13 )(m

CF �� (3.6)

where,

GCV = Gross Calorific Value (kJ g-1)

CF = Calibration Factor (kJ/div)

1� = correction factor of 0.2 (div)

3� = deflection of sample (div)

2m = mass of fuel sample (g)

3.4.2 Density

Density is defined as the mass per unit volume of any liquid at a given temperature.

The density of the fuel sample was determined using a picnometer as shown in

Figure 3.7. Picnometer is a special glass flask which is used for determining a

relative density of liquids using the weight of a known volume.


45

Figure 3.7: Picnometer with Fuel Sample

3.4.2.1 Determine the Volume of the Picnometer

The mass of a dry picnometer with stopper was measured on an analytical balance (

1m ). The room temperature was also noted. The picnometer was then filled with

distilled water that was at room temperature, excess water (above the mark) was

removed from the picnometer using a strip of filtering paper. The mass of the

picnometer plus the water was then determined ( 2m ).The temperature of the distilled

water was measured and the density at this temperature was found from Aylward and

Findlay (2003) text book. Therefore, the volume of the picnometer ( pv ) was

determined as shown in equation 3.7:

wp

mmv�

)( 12 �� (3.7)

where,

pv = volume of picnometer (m-3)

1m = mass of dry picnometer (kg)

2m = mass of picnometer and distilled water (kg)

w� = Density of water (kg m-3)


46

3.4.2.2 Determine the Density of Fuel Sample

The picnometer was filled with the fuel sample and the picnometer and fuel sample

was weighed ( 3m ). Since the mass of the dry picnometer was known, the mass of the

sample was calculated. Having calculated the exact volume of the picnometer and the

knowing mass of the sample, the density of the sample was determined by using the

equation 3.8.

pFS v

mm 13 �� (3.8)

where, FS� = Density of fuel sample (kg m-3)

1m = mass of dry picnometer (kg)

3m = mass of picnometer and fuel sample (kg)

pv = volume of picnometer (m-3)

3.5 Engine Efficiency, Fuel Consumption and Emission

Testing

A Yamaha petrol genset was used to test the performance of the engine using petrol

and different blends of ethanol and petrol. The specifications of the petrol engine and

the generator are provided in Tables 3.2 and 3.3 respectively.


47

Table 3.2: Specification of the petrol engine (Source: Instruction manual of

Yamaha EF2600 Petrol Generator)

Engine Specifications

Model Yamaha EF2600FW

Type MZ175: Air cooled 4-stroke gasoline

OHV

Cylinder Arrangement Inclined, 1 cylinder

Operation Hours 10.6 h

Fuel Tank Capacity 12 L

Engine Oil Quantity 0.6 L

Spark Plug Type BPR4ES (NGK)

Spark Plug Gap 0.7-0.8 mm

Ignition System Transistor Controlled Ignition (TCI)

Maximum Output Power 4.1 hp (3.1 kW)

Continuous Output Power 3.8 hp (2.8 kW)

Bore 66 mm

Stroke 50 mm

Displacement 171 cm3

Lubrication System Splash type


48

Table 3.3: Specification of the generator (Source: Instruction manual of Yamaha

EF2600 Petrol Generator)

Generator Specifications

Generator Type Single Phase AC generator

Type Bi-polar revolving field / with damper

winding

Frequency 50 Hz

Rated Power Output 2 kVA

Maximum Power Output 2.3 kVA

Voltage (AC) 220 V

Current (AC) 9.1 A

Voltage (DC) 12 V

Current (DC) 8.3 A

Speed 3000 rpm

Power Factor 1

Phase Number Single phase

The fuel samples that were tested on the Yamaha petrol genset were neat (100 %)

petrol, E10, E15 and E20. The details of the testing done are outlined in the

following sections.

3.5.1 The Testing Equipment

Engine performance testing requires the measurement of engine torque and power,

which is measured using a dynamometer. The generator supplied with a resistive

load-bank, was used as an electric dynamometer to determine engine performance

The engine performance together with fuel consumption and the resulting emission

when using petrol and different blends of ethanol and petrol blends was determined

using the genset load assembly as shown in Figure 3.8.


49

Figure 3.8: Equipment for testing engine efficiency, fuel consumption and

emission

As seen from Figure 3.8 the testing equipment consisted of a petrol generator, load-

bank, power meter, Horiba automotive emission gas analyzer and the two fuel tanks.

The fuel tank consisted of two 2000 ml graduated measuring cylinder which had two

fuel lines connected by a 3 way valve. One of the measuring cylinder contained

petrol and the other ethanol-petrol blend. The valve was used to select the

appropriate fuel. The emissions that resulted from different test fuels were measured

using the Horiba emission gas analyzer. A power meter was connected to the system

between the generator and the load-bank and this was used to determine the actual

power output. The load bank consisted of light bulbs

Fuel Tank

Horiba Automotive Emission Gas Analyser

Petrol Generator

Load Bank

Power Meter


50

3.5.2 The Testing Procedure

The testing began by setting up the equipment as shown in Figure 3.8. The generator

was started on petrol and let to run on this fuel for 5 min so that the engine warmed

up. The fuel was then switched to the test fuel and the system was flushed with this

fuel by letting it run on 200 ml of the fuel before any measurements were taken.

In the initial stages the engine loss from the system was determined. This was

achieved by running the system without any load and recording the time taken for

120 ml of the test fuel to be consumed.

The loads were then varied and the time taken for 120 ml of the test fuel to be

consumed was recorded. Loads were added gradually by switching the appropriate

light bulbs. This procedure was repeated twice in order to determine the average time

taken for 120 ml of test fuel to be consumed. At each change of load, the system was

run until 40 ml of the fuel was used. This was done to ensure that the system

stabilized before data could be recorded.

The actual electrical power output was measured using the power meter and the

emissions resulting from the test fuels were measured using the Horibar Automotive

Gas Emission Analyser. The data for these were recorded after 40 ml, 80 ml and 120

ml of the fuel had been consumed. The results were then averaged. The power input

to the system was calculated by knowing the GCV of the test fuel and the fuel

consumption over the time period.

3.6 Engine Efficiency

The efficiency ( S� ) of the whole system (gen-set) that is the engine and alternator is

given by total power output per unit power input. This can also be expressed in terms

of energy that is energy output per unit energy input. Equation 3.9 shows the

efficiency of the whole system in terms of energy.


51

S� = 100�in

out

EE = 100�

��

VGCVtPout (3.9)

where,

S� = System Efficiency (%)

outE = energy output of the generator (J)

inE = energy input to the generator (J)

outP = power output of the generator (W)

t = time for specified volume of fuel to be consumed (s)

GCV = gross calorific value of the fuel (J L-1)

V = volume of fuel consumed per test run (L)

The energy input of the system is determined by the GCV of the fuel multiplied by

the volume of fuel used. The volume of fuel used was kept constant for all runs,

therefore as the GCV of the fuels increased the energy input of the system increased.

As mentioned above the system (genset), which consists of the engine and the

alternator. Therefore, to determine the efficiency of the engine only, the efficiency of

each component of the system needs to be determined.

A dynamometer is usually used to determine the efficiency of engines. However, as

this was not available, the method outlined below which was also used by Singh

(2009) was considered as an alternative. The efficiency of the alternator was

determined knowing the alternator input ( 1P ) and the alternator output ( outP ). The

alternator input ( 1P ) is the same as the engine output, while the alternator output

( outP ) is the same as the generator output. A flowchart illustrating the system is

shown in Figure. 3.9 where 1L represents the power loss from the engine and 2L the

power loss from the alternator:


52

1L 2L Figure 3.9: Generalised flowchart of the systems input, output and losses (Source:

Adapted from Singh, 2009)

The alternator efficiency was assumed to be constant at the rated speed of the engine.

This is due to the fact that the engine has a governor coupled to it. The purpose of the

governor is to maintain the speed of an engine during varying load conditions

(Phakatkar, 2008). When there is a change in load, the governor comes into

operation. During increase in load on the engine, the speed decreases making it

necessary to increase fuel supply by opening throttle valve. Similarly, when the there

is a decrease in load, the engine speed increases and the fuel supply needs to be

reduced. The governor maintains the fuel supply by means of the throttle valve thus

keeping the speed of the engine within the required limits of the load (Phakatkar,

2008). Although fluctuations in the rpm do occur, the governor stabilises it.

During the experiment, data was taken after 40 ml of the fuel had been used, so that

the fluctuations in the rpm could be stabilised. Therefore, to determine the alternator

efficiency equation 3.10 can be used:

A� = %1001

�P

Pout (3.10)

1P inP

%1001

��P

PoutA�

Engine

Alternator

Load bank

outP

�E� %1001 �inP

P


53

where,

A� = Alternator Efficiency (%)

outP = power output of the alternator (kW)

1P = power input of alternator (kW)

Substituting these values as specified by the manufacturer at 3000 rpm, the efficiency

of the alternator becomes:

%1008.20.2��A�

= 71 %

The efficiency of the alternator is 71% based on manufacturers specifications. The

data used in equation 3.10 was obtained from the manufacturers stated values at the

rated speed, which was 3000 rpm. The engine output was 2.8 kW which as stated

earlier is the input of the alternator, while the alternator output was 2.0 kW at the

same speed.

Therefore, knowing the system efficiency and the alternator efficiency the engine

efficiency can be determined. This is shown in equation 3.11:

E� = %100�A

s

��

(3.11)

where,

A� = Alternator Efficiency (%)

S� = System Efficiency (%)

E� = Engine Efficiency (%)

3.6.1 Power Loss from the System

Any system or engine is not a hundred percent efficient. There are often power losses

from the system. When the system contains a heat engine, the Second Law of


54

Thermodynamics states that its efficiency will never be equal to 100 %. This may be

re-stated in terms of Carnot’s Theorem, which states that no real heat engine

operating between two energy reservoirs can be more efficient than a Carnot engine

operating between the same two reservoirs. That is all real engines are less efficient

than the Carnot engine (Carter, 2003). A typical internal combustion engine remains

only about 20-25 % efficient (Carter, 2003).

The method of analysing the energy flow in the whole system comprising of the

engine and the alternator has been given by Singh (2009). This considers the power

inputs and losses at each stage of the system. The following equations, 3.12, 3.13 and

3.14 represent the system energy budget:

inP = 1P + 1L (3.12)

1P = outP + 2L (3.13)

inP = outP + L2 + L1 (3.14)

Therefore, if the input power of the engine is to be calculated, then equation 3.15 can

be used:

inP =AE

outP��

(3.15)

where,

inP = power input of the engine, which is supplied by the GCV of

the fuel

outP = power output of the alternator, supplied to the load bank

E� = Engine Efficiency

A� = Alternator Efficiency


55

3.7 Emission Testing

The Horiba Automotive Emission Gas Analyzer MEXA-554J was used to analyse

the CO, HC, and CO2 components resulting from various fuel samples used to run

the petrol generator. The specifications of the analyser are given in Table 3.4. This is

a portable analyser used for analysing the exhaust gases from automotives as well as

some non-automotives such as a generator.

Table 3.4: Specifications of the Horiba Automotive Emission Gas Analyser

(Source: Instruction manual of Horiba Automotive Emission Gas

Analyser MEXA-554J series)

Model MEXA-554JA

Gases Measured CO, HC, CO2 and O2

Measurement Range CO: 0-10 % vol

HC: 0-20, 000 ppm vol

CO2: 0-20 % vol

O2: 0-25 % vol

Repeatability within one third of [larger one of ± 0.06 % vol

CO, ± 12 ppm vol HC, ± 0.5 % vol CO2 of

readout] 13 times or more in 20 measurement

times

Response speed TD + T90: Within 10 seconds

TD + T95: Within 15 seconds

Output Analog: 0 to 1 V DC

Ambient Temperature Range 0 °C to 40 °C

Ambient Humidity Range 90 % or less

3.7.1 Instrument Start-up

Horiba automotive emission gas analyser was warmed-up and certain tests were

performed before it was ready to be used. This was done so that measurements made

would be stable and have a greater degree of accuracy.


56

The power supply on the rear of the instrument was turned on. The warm-up screen

was displayed and the instrument automatically began the warm-up procedures. The

warm-up time for the instrument was 5 mins.

After the warm-up HC Hang-up test was performed. This test was made to check

how much hydrocarbon (HC) is absorbed by the sampling unit. During this test clean

air continues to be purged in until the reading becomes 20 ppm or less. The screen

will show a passed sign when the test is complete.

Leak test was performed to check if any filter, probe or the sampling unit needed to

be replaced. This is done to ensure that there is no gas leakage. A seal cap is fitted

over the leading end of the probe to block the gas absorption holes on the probe. The

test is then started and a seal retention time of 5 seconds is counted down. A passed

sign is displayed if the test is successful.

Having completed the warm-up and the other tests the instrument was ready for

measurement. The “M” key located on the front panel is pressed and the “MEAS”

mark blinks on the display for 10 seconds before the instrument entered the

measurement mode.

3.7.2 Measurement of Emission

The Horiba automotive emission gas analyser was ready for measurement after the

warm-up, calibration and the two tests; HC hang-up and leak test were completed.

The instrument was pre-calibrated by the manufacturers. The probe was inserted into

the exhaust pipe of the generator. Measurements were carried out after 40 ml of the

sample had been used. This ensured that the system had stabilised and also that any

residual gas components of the previous fuel sample had been removed. Initially,

when the probe was inserted in the exhaust pipe of the generator, fluctuations in the

reading where observed however, the readings stabilised after 20 s, after which data

was taken. Data was again taken after 80 ml and 120 ml of the fuel had been used.

This was to ensure the emission levels of the sample were consistent throughout the

run-time of the fuel. The averages of the readings were taken and it was noted that

the differences between the data were minimal. The results were within an accuracy


57

of ± 0.06 % vol CO, ± 12 ppm vol HC and ± 0.5 % vol CO2. After the completion of

the testing the instrument was left in the measurement mode for 30 mins so the

sample lines were purged with clean air before switching the instrument off.

58

Chapter 4 Results and Discussions

4.0 Overview

This chapter has five main sections. The first section discusses the viability of

producing ethanol from locally available cassava varieties in Fiji. The second section

presents and discusses the results of ethanol production from cassava starch. The

third and fourth section deals with the preparation and identification of physical

properties of ethanol-petrol blends respectively. The final section presents and

discusses the results of the investigations of engine performance, which include the

efficiency, fuel consumption and emission data.

4.1 Cassava Varieties in Fiji

The starch yield from different cassava varieties obtained from two different

locations in Fiji is shown in Table 4.1 below. The starch yields for cassava varieties

obtained from both the stations ranges from 17 to 23 %.

Chapter 4: Results and Discussions

59

Table 4.1: Starch Yield from Cassava Varieties

Cassava

Variety Koronivia Dobuilevu

Starch Yield

(%)

Starch Yield

(%)

Niumea 18.3 19.9

Sokobale 17.7 *

Beqa 21.9 17.0

New Guinea 20.9 19.5

Coci 20.5 23.3

Vula Tolu 17.3 18.1

Yabia Damu 19.6 17.9

Merelesita 18.6 18.8

Nadelei 23.1 22.1

Navolau 17.0 19.8 * means the variety was not available at this location

The results show that the yields are dependent on the sites of the plantations. They

show that Beqa, New Guinea, Yabia Damu, and Nadelei are more suited to the

Koronivia site with regard to their starch yields. On the other hand Niumea, Coci,

Vula Tolu and Navolau performed better at the Dobuilevu site for their starch yields.

The Merelesita starch yield did not show much variation with site. Therefore, it is

seen that location is one of the factors that influence starch yield.

Benesi’s (2005) result indicated that the genetic constitution of the plant is the most

influential factor. However, sites, rounds of starch extraction and their interaction

also have appreciable influence. Similar observations were also made by

Ngendahayo and Dixion (2001) who found that after six months, the starch content

in plants are influenced by genotype, harvesting time and rainfall pattern.

The annual rainfall for KRS was 3626 mm in 2008/2009, during the growing seasons

and for DRS it was 3069.6 mm in 2008/2009. Both research stations received highest

rainfall in January. When cassava was planted at KRS for the first four months of

planting there was decrease in rainfall as seen in Figure 4.1. After September there


60

was increase in rainfall until January.1. Then from February to harvest there were

monthly fluctuations observed in rainfall.

Figure 4.1: Rainfall and Temperature data at KRS

On the other hand at DRS, a slight increase was observed in rainfall for the first

month of planting as seen in Figure 4.2. Then there was a gradual decrease in rainfall

in September until October. From October to January there was eventual increase in

rainfall. Then after January until harvest again there was a decrease in rainfall.

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34

0

100

200

300

400

500

600

700

Tem

pera

ture

(°C

)

Rai

nfal

l (m

m)

Rainfall and Temperature Data for KRS

Rainfall

Max Temperature Min Temperature


61

Figure 4.2: Rainfall and Temperature data at DRS

Monthly mean maximum temperatures ranged between 26.9 and 31.2 °C for

Koronivia and 31.2 and 34.3 °C for Dobuilevu. Monthly mean minimum

temperatures ranged between 19.6 and 23.3 °C for Koronivia, 25.2 and 29 °C for

Dobuilevu. It is seen that Dobuilevu generally had higher temperatures than

Koronivia.

The dry matter content obtained from two different locations is shown in Table 4.2.

The cassava from Koronivia had a dry matter content as high as 41 % (New Guinea

and Beqa) whereas the ones from Dobuilevu had a maximum of 38 % (Nadelei).

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38

0 100 200 300 400 500 600 700 800 900

1000 1100 1200 1300 1400

Tem

pera

ture

(°C

)

Rai

nfal

l (m

m)

Rainfall and Temperature Data for DRS

Rainfall

Max Temperature Min Temperature


62

Table 4.2: Dry Matter Content of Cassava Varieties

Cassava


Dry Matter

Content (%)

Dry Matter

Content (%)

Niumea 37.9 36.2

Sokobale 36.3 *

Beqa 41.2 32.2

New Guinea 41.5 32.3

Coci 37.0 36.8

Vula Tolu 31.1 35.6

Yabia Damu 35.7 36.0

Merelesita 40.9 35.6

Nadelei 40.8 38.0

Navolau 33.7 33.8

* indicates the variety was not available at this location

Dry matter content is very much related to rainfall during six to eighteen months of

plant growth (Ngendahayo and Dixion, 2001). This suggests that the difference in

dry matter content between the sites could be contributed to the rainfall received

during plant growth. However, Benesi et al. (2004) have reported that the root dry

matter content of cassava in Malawi is in the range 38.24 to 46.48 % and that dry

matter content was not as much influenced by the environment as by the genetic

differences.

Most researchers agree that the optimum time for harvest of cassava roots depend on

varieties and ecological factors (Ashoka et al. 1984, Ngendahayo and Dixion, 2001).

Hence, there is a need to identify optimum harvest time for different cassava

varieties and also identify the most suitable environment for these varieties.

Optimum economic returns in terms of dry matter content and starch yield can be

ensured if suitable varieties and ecological factors are identified.


63

The mineral elements and inorganic salts present in starch are referred to as ash.

Table 4.3 shows that the ash content for cassava varieties obtained from Koronivia is

0.10 to 0.17 % and for Dobuilevu it was 0.1 to 0.21 %.

Table 4.3: Ash Content of Cassava Starch from Various Varieties

Cassava


Ash Content

(%)

Ash Content

(%)

Niumea 0.17±0.01 0.19±0.03

Beqa 0.17±0.03 0.21±0.02

Sokobale 0.11±0.02 *

New Guinea 0.09±0.01 0.15±0.04

Coci 0.14±0.02 0.13±0.01

Vula Tolu 0.10±0.02 0.12±0.02

Yabia Damu 0.10±0.02 0.17±0.03

Merelesita 0.12±0.02 0.13±0.01

Nadelei 0.10±0.02 0.10±0.02

Navolau 0.14±0.02 0.16±0.04

Notes: Means of three replicates (± SD) * indicates the variety was not available at this location

According to Thomas and Atwell (1999) ash content is typically less than 0.5 % of

dry mass and this agrees with the results obtained for all the cassava varieties

obtained from Koronivia and Dobuilevu. Variations in ash content depend upon

source of raw material, agronomic practices, extraction and milling procedures and

types of chemical modifications (Benesi, 2005).

The pH of cassava starch as reported in Table 4.4 ranged from 4.07 to 5.23 for the

varieties obtained from Koronivia and 5.03 to 6.20 for those varieties obtained from

Dobuilevu. Benesi (2005) indicated a pH range of 5.0 to 5.5 for the native starch

obtained from ten elite Malawian cassava genotype. The recommended pH range

stated by the National Starch and Chemical Company (2002) is between 4.5 and 7.0.

Therefore, as the obtained starch pH of the different cassava varieties in Fiji are


64

within the recommended range, they are favourably disposed for industrial and food

applications.

Table 4.4: pH of Cassava Starch from Various Varieties

Cassava


pH pH

Niumea 4.07±0.02 5.98±0.05


Beqa 4.89±0.01 5.81±0.04

New Guinea 5.23±0.02 5.03±0.02

Coci 4.47±0.01 5.04±0.03

Vula Tolu 4.63±0.03 5.17±0.01

Yabia Damu 4.44±0.01 6.20±0.02

Merelesita 4.53±0.03 5.21±0.03

Nadelei 4.53±0.02 5.48±0.03

Navolau 4.37±0.01 5.34±0.02

Notes: Means of three replicates (± SD) * indicates the variety was not available at this location

The moisture content of starch from cassava varieties obtained at Koronivia as

shown in Table 4.5, ranged from 12.4 to 14.7 %, whereas, for Dobuilevu it was from

12.0 to 14.5 %. The results obtained are consistent with the results reported by

Nuwamanya et al. (2008) and Benesi (2005). Nuwamanya et al. (2008) reported

moisture content ranged from 14.09 and 16.49 % for the parental lines and 14.80 to

16.11 % in the progenies. The native cassava starch moisture content that Benesi

(2005) found for the ten varieties investigated ranged from 10.47 to 12.83 %. High

moisture content in cassava starch is not a desired property. It leads to growth of

micro-organisms that are capable of degrading starch (Nanda, n. d). Moorthy (2001)

reported that high moisture content affects the pasting properties of cassava starch

and Willet and Doane (2002) stated that the tensile properties and overall granular

structure of starch are also affected by high moisture content.


65

Table 4.5: Moisture Content of Cassava Starch from Various Varieties

Cassava


Moisture

Content (%)

Moisture

Content (%)

Niumea 12.6±0.5 12.9±0.9


Beqa 14.1±0.8 14.5±0.4

New Guinea 13.9±0.6 13.6±0.9

Coci 13.1±0.5 12.5±1.0

Vula Tolu 13.9±0.6 13.7±1.2

Yabia Damu 12.9±0.3 12.8±0.5

Merelesita 12.4±1.0 13.0±0.4

Nadelei 13.7±0.4 13.6±0.9

Navolau 12.5±1.0 12.0±1.2

Notes: Means of three replicates (± SD) * means the variety was not available at this location

In Fiji, cassava is predominantly grown for human consumption. There is almost no

processing of cassava into dried form for human or animal use. However, apart from

food, cassava is exported to Australia and New Zealand as frozen tubers. Cassava is

grown in most parts of Fiji. As it is tolerant to a range of climatic conditions as well

as growing in marginal land, limited effort is currently being placed in improving

conditions for planting. There is minimum land preparation, weeding is hardly done

and limited fertilizer applied. In 2007, the cassava yield was 13.80 t ha-1 (SOPAC,

2009). This yield can be increased with sustainable cultivation practices and also by

identifying high yielding varieties. Cassava research in Fiji is mostly done by KRS

and other stations of the Ministry of Agriculture, Fisheries and Forest.

Bio-ethanol is produced by fermenting sugars or substances that can be converted to

sugars, such as starch and cellulose. Cassava roots contain starch that can be

converted to sugar. As seen in the starch yield results obtained (Table 4.1), cassava in

Fiji can be used for bio-ethanol production. The Fiji Government has plans to


66

produce bio-ethanol from agricultural sources available in Fiji namely sugarcane,

molasses and cassava. It has been pointed out (Rao, 1997), cassava is one of the best

crops to be used for bio-ethanol production. The ethanol yield of cassava per unit

land area is higher than any other known energy crop as seen in Table 4.6. In

addition, it is much cheaper to set up a cassava ethanol factory because of lower

investment and the processing technology is much simpler due to the special

characteristics of starch (Wang, 2002). The cost of cassava ethanol can be lowered

due to production of useful by-products from different parts of cassava plant (Wang,

2002).

Table 4.6: Comparison of ethanol yield made from various energy crops

Crop Yield

(t ha-1 year-1)

Conversion rate

to sugar or

starch (%)

Conversion rate

to ethanol

(L t-1)

Ethanol Yield

(kg ha-1 year-1)

Sugarcane 70 12.5 70 4,900

Cassava 40 25 150 6,000

Carrot 45 16 100 4,500

Sweet sorghum 35 14 80 2,800

Maize 5 69 410 2,050

Wheat 4 66 390 1,560

Rice 5 75 450 2,250

Source: (Rao, 1997)

However, since cassava is primarily produced in Fiji for food by the people, an

approach needs to be found that would balance out the use of agricultural land for

food and fuel. The use of food crops for fuel usually drives the prices of these crops.

For this reason governments in many countries are now ensuring that biofuels do not

increase the price of staple foods. The Fiji Government has dismissed the threat to

food security on the grounds that more than enough land is idle in Fiji according to

Food and Agriculture Organization (FAO) 2006 figures (SOPAC, 2009). As stated in

the FAO report (SOPAC, 2009), promoting diversification and setting aside land for

food production is one strategy. However, governments need to make a national-

level decision as to what extent staple crops should be used for biofuel production.


67

4.2 Ethanol Production from Cassava Starch

Ethanol can be produced from cassava starch by fermentation using yeast such as

Saccharomyces cerevisiae. However, cassava starch first needs to be hydrolyzed to

sugar before bioconversion to ethanol by Saccharomyces cerevisiae.

The results of the ethanol produced and the amount of reducing sugars remaining as

a function of time from each variety of cassava starch obtained from KRS and DRS

in this study are shown in Figures 4.3 and 4.4 respectively.

As seen in both the figures, from 0 hours to 24 hours there is a rise in amount of

reducing sugars from an approximate average of 70 g l-1 peaking at 24 hours and then

declining after 24 hours with a simultaneous rise in the value of ethanol

concentration. This is explained by observing that during the liquefaction stage small

amounts of reducing sugars were produced by the action of α-amylase. The action of

the amyloglucosidase produced further amount of reducing sugars through the

additional breakdown of the dextrin which was obtained from liquefaction.

From all the figures it is revealed that yeast (Saccharomyces cerevisiae) began

producing ethanol after 24 hours with decline in reducing sugars concentration and a

simultaneous increase in ethanol concentration. These results support published

reports that both α-amylase and amyloglucosidase activities are needed for starch

fermentation by Saccharomyces cerevisiae (Eksteen et al., 2003; Shigenchi et al.,

2004; Jamai et al., 2007).


68

(a)

(b)

0 10 20 30 40 50 60 70

0 20 40 60 80

100 120 140

0 24 48 72 96 120 Eth

anol

Con

cent

ratio

n (g

l-1)

Red

ucin

g Su

gars

Con

cent

ratio

n (g

l-1)

Fermentation Time (hrs)

Reducing Sugars and Ethanol Concentration for Niumea (Koronivia)

Reducing Sugars

Ethanol Concentration

0 10 20 30 40 50 60 70

0 20 40 60 80

100 120 140 160

0 24 48 72 96 120 Eth

anol

Con

cent

ratio

n (g

l-1)

Fermentation Time (hrs) Red

ucin

g Su

gars

Con

cent

ratio

n (g

l-1)

Reducing Sugars and Ethanol Concentration Sokobale (Koronivia)

Reducing Sugars



69

(c)

(d)

0 10 20 30 40 50 60 70

0 20 40 60 80

100 120 140 160

0 24 48 72 96 120 Eth

anol

Con

cent

ratio

n (g

l-1)

Red

ucin

g Su

gars

Con

cent

ratio

n (g

l-1)


Reducing Sugars and Ethanol Concentration Beqa (Koronivia)

Reducing Sugars Ethanol Concentration

0

10

20

30

40

50

60

0

20

40

60

80

100

120

0 24 48 72 96 120 Eth

anol

Con

cent

ratio

n (g

l-1)

Red

ucin

g Su

gars

Con

cent

artio

n (g

l-1)


Reducing Sugars and Ethanol Concentartion New Guinea (Koronivia)



70

(e)

(f)

0 10 20 30 40 50 60 70

0 20 40 60 80

100 120 140 160

0 24 48 72 96 120 Eth

anol

Con

cent

ratio

n (g

l-1)

Red

ucin

g Su

gars

Con

cent

ratio

n (g

l-1)


Reducing Sugars and Ethanol Concentration Coci (Koronivia)


0

10

20

30

40

50

60

70

0

20

40

60

80

100

120

0 24 48 72 96 120

Eth

anol

Con

cent

ratio

n (g

l-1)

Red

ucin

g Su

gars

Con

cent

ratio

n (g

l-1)


Reducing Sugars and Ethanol Concentration Vula Tolu (Koronivia)



71

(g)

(h)

0

10

20

30

40

50

60

70

0

20

40

60

80

100

120

140

0 24 48 72 96 120

Eth

anol

Con

cent

ratio

n (g

l-1)

Red

ucin

g Su

gars

Con

cent

ratio

n (g

l-1)


Reducing Sugars and Ethanol Concentration Yabia Damu (Koronivia)


0

10

20

30

40

50

60

70

0 20 40 60 80

100 120 140 160 180

0 24 48 72 96 120

Eth

anol

Con

cent

ratio

n (g

l-1)

Red

ucin

g Su

gars

Con

cent

ratio

n (g

l-1)


Reducing Sugars and Ethanol Concentration Merelesita (Koronivia)



72

(i)

(j) Figure 4.3: Ethanol Concentration and remnant reducing sugars concentration from

(a) Niumea, (b) Sokobale, (c) Beqa, (d) New Guinea, (e) Coci, (f) Vula

Tolu, (g) Yabia Damu, (h) Merelesita, (i) Nadelei, (j) Navolau cassava

variety starch obtained at KRS

0 10 20 30 40 50 60 70

0 20 40 60 80

100 120 140 160

0 24 48 72 96 120 Eth

anol

Con

cent

ratio

n (g

l-1)

Red

ucin

g Su

gars

Con

cent

ratio

n (g

l-1)


Reducing Sugars and Ethanol Concentration Nadelei (Koronivia)


0 10 20 30 40 50 60 70

0

20

40

60

80

100

120

0 24 48 72 96 120 Eth

anol

Con

cent

ratio

n (g

l-1)

Red

ucin

g Su

gars

Con

cent

ratio

n (g

l-1)


Reducing Sugars and Ethanol Concentration Navolau (Koronivia)



73

(a)

(b)

0

10

20

30

40

50

60

0 20 40 60 80

100 120 140

0 24 48 72 96 120 Eth

anol

Con

cent

ratio

n (g

l-1

)

Red

ucin

g Su

gars

Con

cent

ratio

n (g

l-1)


Reducing Sugars and Ethanol Concentration for Niumea (Dobuilevu)


0 10 20 30 40 50 60 70

0 20 40 60 80

100 120 140 160

0 24 48 72 96 120 Eth

anol

Con

cent

ratio

n (g

l-1)

Red

ucin

g Su

gars

Con

cent

ratio

n (g

l-1)


Reducing Sugars and Ethanol Concentration Beqa (Dobuilevu)



74

(c)

(d)

0 10 20 30 40 50 60 70

0

20

40

60

80

100

120

0 24 48 72 96 120 Eth

anol

Con

cent

ratio

n (g

l-1)

Red

ucin

g Su

gars

Con

cent

artio

n (g

l-1)


Reducing Sugars and Ethanol Concentartion New Guinea (Dobuilevu)

Reducing Sugars


0 10 20 30 40 50 60 70

0 20 40 60 80

100 120 140 160 180

0 24 48 72 96 120 Eth

anol

Con

cent

ratio

n (g

l-1)

Red

ucin

g Su

gars

Con

cent

ratio

n (g

l-1)


Reducing Sugars and Ethanol Concentration Coci (Dobuilevu)



75

(e)

(f)

0 10 20 30 40 50 60 70

0 20 40 60 80

100 120 140 160 180

0 24 48 72 96 120 Eth

anol

Con

cent

ratio

n (g

l-1)

Red

ucin

g Su

gars

Con

cent

ratio

n (g

l-1)


Reducing Sugars and Ethanol Concentration Vula Tolu (Dobuilevu)


0 10 20 30 40 50 60 70

0 20 40 60 80

100 120 140 160

0 24 48 72 96 120 Eth

anol

Con

cent

ratio

n (g

l-1)

Red

ucin

g Su

gars

Con

cent

ratio

n (g

l-1)


Reducing Sugars and Ethanol Concentration Yabia Damu (Dobuilevu)



76

(g)

(h)

0

10

20

30

40

50

60

70

0 20 40 60 80

100 120 140 160 180 200

0 24 48 72 96 120 Eth

anol

Con

cent

ratio

n (g

l-1)

Red

ucin

g Su

gars

Con

cent

ratio

n (g

l-1)


Reducing Sugars and Ethanol Concentration Merelesita (Dobuilevu)


0

10

20

30

40

50

60

70

0 20 40 60 80

100 120 140 160 180 200

0 24 48 72 96 120

Eth

anol

Con

cent

ratio

n (g

l-1)

Red

ucin

g Su

gars

Con

cent

ratio

n (g

l-1)


Reducing Sugars and Ethanol Concentration Nadelei (Dobuilevu)



77

(i)

Figure 4.4: Ethanol Concentration and remnant reducing sugars concentration from

(a) Niumea, (b) Beqa, (c) New Guinea, (d) Coci, (e) Vula Tolu, (f) Yabia

Damu, (g) Merelesita, (h) Nadelei, (i) Navolau cassava variety starch

obtained at DRS

The ethanol concentration obtained after 120 hours of fermentation was in the range

of 55-63 g l-1 for cassava obtained from KRS and 55-65 g l-1 for cassava obtained

from DRS as seen in Table 4.7. Statistical analysis performed on the results showed

that there is no significant difference among the ethanol concentration obtained from

starch of different cassava varieties (p < 0.05). Similarly, there was no significance

found in ethanol concentration obtained from starch at the two locations and

interaction between cassava variety and location at (p < 0.05).

The ethanol yield, which was expressed as litres of ethanol produced per kilogram of

starch used, was in the range of 0.35-0.40 L of ethanol per kg of starch and 0.35-0.41

L of ethanol per kg of starch for KRS and DRS respectively. The details of this are

indicated in Table 4.7. There was no significant difference among the ethanol yield

obtained from starch of different cassava varieties (p < 0.05). Similarly, there was no

significant difference found in ethanol concentration obtained from starch at the two

locations and between cassava varieties from the two locations (p < 0.05).

0

10

20

30

40

50

60

70

0 20 40 60 80

100 120 140 160 180 200

0 24 48 72 96 120

Eth

anol

Con

cent

ratio

n (g

l-1)

Red

ucin

g Su

gars

Con

cent

ratio

n (g

l-1)


Reducing Sugars and Ethanol Concentration Navolau (Dobuilevu)



78

Cassava has been shown to be a good source of ethanol production because it has a

high amylose/amylopectin ratio (Dubey, 2001). Starch is a polymer of glucose. It is

made up of two related but structurally different polymers, amylose and amylopectin.

The basic building block of both amylose and amylopectin is glucose monomer

through the α–1, 4 glucosidic bonds (Kearsley and Dziedzic, 1995). As the there is

no significant difference in the ethanol concentration and ethanol yield resulting from

using starch from the different cassava varieties and obtained from the two locations,

it can be assumed that the starch from these cassava are almost similar in structure.

This indicates that the enzymatic activity and also fermentation by Saccharomyces

cerevisiae on the starch from different cassava varieties were the same.

Table 4.7: Final ethanol concentration and ethanol yield from cassava varieties

from two different locations

KRS DRS

Cassava

Variety

Final Ethanol

Concentration

(g l-1)

Ethanol Yield

(L Ethanol kg-

1 Starch)

Final Ethanol

Concentration

(g l-1)

Ethanol Yield

(L Ethanol kg-

1 Starch)

Niumea 61 ± 7 0.39 ± 0.04 55 ± 7 0.35 ± 0.04

Sokobale 62 ± 7 0.39 ± 0.04 * *

Beqa 59 ± 4 0.38 ± 0.03 61 ± 4 0.39 ± 0.03

New Guinea 55 ± 5 0.35 ± 0.03 59 ± 4 0.38 ± 0.02

Coci 63 ± 5 0.40 ± 0.03 59 ± 5 0.37 ± 0.03

Vula Tolu 60 ± 6 0.38 ± 0.04 63 ± 7 0.40 ± 0.04

Yabia Damu 61 ± 9 0.39 ± 0.06 62 ± 4 0.39 ± 0.02

Merelesita 60 ± 7 0.38 ± 0.05 65 ± 6 0.41 ± 0.04

Nadelei 59 ± 8 0.38 ± 0.05 59 ± 6 0.38 ± 0.04

Navolau 58 ± 6 0.37 ± 0.04 60 ± 4 0.38 ± 0.02



79

Analysis of ethanol production from different cassava varieties as seen in Table 4.8

shows that approximately 11.47-15.85 kg of cassava is required by different varieties

to produce 1 L ethanol. As seen in Table 4.8 the least amount of cassava is required

by the variety Nadelei (11.47 kg) which also has the highest starch yield from

cassava roots. On the other hand, it is noted that variety Navolau and Beqa both have

the lowest starch yield (17 %) but since the ethanol yield when using Beqa (0.39 L

Ethanol kg-1 Starch) variety is slightly more than Navolau (0.37 L Ethanol kg-1

Starch), more cassava is required by the variety Navolau (15.85 kg) to produce 1 L of

ethanol.


80

Table 4.8: Cassava required for producing 1 L ethanol from each cassava variety

studied

KRS

DRS

Cassava

Variety

Starch

Yield

(%)

Ethanol

Yield (L

Ethanol kg-1

Starch)

Starch

require

d to

produce

1 L

ethanol

(kg)

Cassava

require

d to

produce

1 L

ethanol

(kg)

Starch

Yield

(%)

Ethanol

Yield (L

Ethanol

kg-1

Starch)

Starch

require

d to

produce

1 L

Ethanol

(kg)

Cassava

required

to

produce

1 L

ethanol

(kg)

Niumea 18.3 0.39 ±

0.04

2.58 13.71 19.9 0.35 ±

0.04

2.84 14.28

Sokobale 17.7 0.39 ±

0.04

2.55 14.40 * * * *

Beqa 21.9 0.38 ±

0.03

2.65 12.10 17 0.39 ±

0.03

2.56 15.08

New

Guinea 20.9 0.35 ±

0.03

2.84 13.59 19.5 0.38 ±

0.02

2.66 13.66

Coci 20.5 0.40 ±

0.03

2.48 12.11 23.3 0.37 ±

0.03

2.68 11.50

Vula

Tolu 17.3 0.38 ±

0.04

2.62 15.14 18.1 0.40 ±

0.04

2.51 13.86

Yabia

Damu 19.6 0.39 ±

0.06

2.56 13.08 17.9 0.39 ±

0.02

2.55 14.24

Merelesit

a 18.6 0.38 ±

0.05

2.62 14.09 18.8 0.41 ±

0.04

2.42 12.86

Nadelei 23.1 0.38 ±

0.05

2.65 11.47 22.1 0.38 ±

0.04

2.66 12.06

Navolau 17 0.37 ±

0.04

2.69 15.85 19.8 0.38 ±

0.02

2.63 13.31



81

The ethanol yield obtained from the Fiji cassava varieties shows a lot of variation

from literature obtained from cassava at other places. According to Li and Halbrendt

(2009), in China 7.2 t of fresh cassava can yield 1 t of ethanol. This is almost same as

producing 1 L ethanol from approximately 5.7 kg of cassava. The high ethanol yield

from cassava available in China can be attributed to the fact that starch content in

cassava from China is higher than 30 %. Researchers Sriroth et al. (2006) have

indicated that for Thailand cassava at 25 % starch content about 6 kg of fresh cassava

can yield 1 L ethanol. Similarly, Sorapipatana and Yoosin (2011) have reported that

approximately 1 t cassava can yield 160 L ethanol, which is same as 6.25 kg cassava

for 1 L ethanol. Table 4.9 summarizes theses results.

Table 4.9: Comparison of Obtained Ethanol yield with Literature Results Country Cassava needed for producing 1 L

ethanol (kg)

Fiji 11.47-15.85

China 5.7

Thailand 6-6.25

Looking at the literature values it is clear that the ethanol yield obtained from Fiji

cassava is low. The reason for this could be due to the cassava varieties and the

agronomic practice in Fiji. The cassava that was used for experimental purposes were

obtained from the research stations. The varieties obtained were considered to be

high yielding and to have good starch content but no data on this was obtained by the

research stations themselves as the stations were still doing trials. According to them

very little scientific input had been put into growing the different varieties. The

following are some factors that need to be considered in order to ensure high yield of

cassava (ICS-Nigeria, n. d):

� Choice of land

� Choosing cassava variety

� Recommended varieties

� Acquisition of planting materials

� Stem storage

� Stem quality


82

� Preparation of planting materials

� Handling of stakes

� Time of planting

� Method of planting

� Plant population

� Weed control

� Fertilizer rate and time of application

� Harvesting

For high ethanol yield, high starch content in cassava roots is the most important

criterion. Therefore, varieties with the highest starch content should be used for

ethanol production and the development of high starch content varieties through

breeding should be considered. There are other factors that affect the results of

ethanol yield such as fermentation temperature, removal of inhibiting substances,

agitation of fermentation tube and the yeast species (Liu and Liang, 1983).

Nevertheless, feedstock selection and conversion of feedstock to ethanol are very

important consideration that needs to be made when production of bio-ethanol is

considered especially for mass production and sustainable use.

4.3 Preparation of Ethanol-Petrol Blends

Ethanol is a hydrophilic compound, which means it absorbs water from the

atmosphere. Due to ethanol’s high affinity for water, storage of ethanol-petrol blends

needs to be done with due care. Also blends should be made when they are

absolutely required rather than storing for long periods of time.

Ethanol-petrol blends were observed for stability at different temperatures that are

expected in Fiji. The composition of ethanol-petrol blends prepared for testing is

shown in Table 4.10.


83

Table 4.10: Composition of ethanol-petrol blended samples used for analysis

Sample code % Ethanol % Petrol

E10 10 90

E15 15 85

E20 20 80

The preparation of the E10, E15 and the E20 samples were done using 93 %, 95 %,

97 % and absolute ethanol respectively. This was done to see whether apart from

absolute ethanol if any other purity of ethanol could be used for blending without

showing signs of phase separation. Table 4.11 to 4.14 summarizes the observations

of the ethanol-petrol blends.

Table 4.11: Stability testing using absolute ethanol in ethanol-petrol blends

Temperature

(ºC)

28 26 24 22 18 14

Phase Separation (Yes/No)

E10 No No No No No No



Table 4.12: Stability testing using 97 % ethanol in ethanol-petrol blends

Temperature

(ºC)

28 26 24 22 18 14






84


Temperature

(ºC)

28 26 24 22 18 14


E10 No No No No No Yes




Temperature

(ºC)

28 26 24 22 18 14


E10 No No No No Yes Yes

E15 No No No No Yes Yes

E20 No No No Yes Yes Yes

As seen in the tables there is no phase separation when using absolute ethanol in

blends. This is consistent with the absence of water in these samples. With the 97 %,

95 % and 93 % ethanol blends it was observed that, with the exception of the 93%

ethanol, there is no phase separation above 18 °C. Generally, the temperature at

which phase separation occurs increases with the water content. In Fiji the average

ambient temperature is usually between 26 and 29 ºC. The results above show that it

is possible to blend ethanol and petrol with even 93 % pure ethanol in Fiji without

phase separation occurring.

Since ethanol has an affinity for water, phase separation is likely to occur as the

water content of ethanol-petrol blend increases. For this reason, ethanol-petrol blends

should be stored in a secure place where the possibility of water entering the blends

is absolutely minimum.

The results of this study are consistent with those reported in the literature, which

show that phase separation in ethanol-petrol blends is temperature dependent, and


85

that it occurs more readily at lower temperatures with lower ethanol content and at

higher temperatures with higher ethanol content (Guibet, 1999). Together with

temperature, phase separation is also dependent on aromatics and ethanol content

(Environment Australia, 2002).

The blends studied in this work were prepared using 96 % ethanol. This was

considered suitable for blending without phase separation. A lower percentage could

also have been used as well but just to avoid too much water being in the blend 96 %

ethanol was considered suitable. Also absolute ethanol requires additional distillation

to produce which can be quite costly to produce and also require extra energy.

Saving on this additional energy expenditure may help ensure that energy returned

on energy invested on bio-ethanol is positive thus indicating that using bio-ethanol

should be pursued.

4.4 Physical Properties of Ethanol-Petrol Blends

The three different ethanol-petrol blends that were prepared were tested for density

and gross calorific value (GCV) before they were used in engine and emission

testing.

4.4.1 Density

The density of the petrol and ethanol-petrol blends at 25 °C as a function of the

ethanol content in the blends has been plotted in Figure 4.5. It is observed that the

density of blends increase with increasing ethanol content in blends. This is expected

considering the density of ethanol is more than that of petrol.


86

Figure 4.5: Density of Petrol, E10, E15 and E20

There is an increase in density by 0.7 % upon addition of 10 % ethanol in petrol. This

change is bought about due to the fact that petrol has a density of 0.741 g cm-3

whereas 96 % ethanol that used for blending had a density of 0.793 g cm-3. Similarly,

E15 had an increase in density by 1.10 % and E20 by 1.52 % when compared to

petrol.

Density is a fundamental physical property that can be used in conjunction with other

properties. It also gives an indication of the purity of the fuel. Whenever fuel gets

contaminated with another liquid, density will either increase or decrease depending

on the density of the contaminant liquid. Density is also an indicator of how much

fuel is supplied to the engine, because the fuel pump delivers fuel on a volumetric

basis (Puhan and Nagarajan, 2008). As seen from Figure 4.5 the density of the blends

is higher when compared to petrol, this means more of the fuel is supplied to the

engine when compared to petrol. Therefore, E10, E15 and E20 supply more fuel to

the engine when compared to petrol which can result in increased power output.

0.74 0.742 0.744 0.746 0.748

0.75 0.752 0.754

0 5 10 15 20 25

Den

sity

(g c

m-3

)

Ethanol Fraction in Blends (%)

Density of Petrol and Ethanol/Petrol Blends


87

4.4.2 Gross Calorific Value

The gross calorific value (GCV) of petrol, E10, E15 and E20 are shown in Figure

4.6.

Figure 4.6: GCV of Petrol, E10, E15 and E20 on mass and volume basis

It is observed that blends containing ethanol have a lower GCV, i.e. the GCV

decreases as the ethanol content in blends increases. This can be attributed to the fact

that ethanol has a lower GCV than petrol.

The GCV of petrol was found to be 45.4 kJ g-1 whereas for ethanol it was 29.2 kJ g-1.

The ethanol GCV is almost 35.7 % lower than petrol GCV. E10 GCV is

approximately 4.6 % lower than petrol. For E15 and E20 a decrease of 8.4 % and 9.9

% respectively is observed in GCV when compared with petrol.

Lower GCV in the ethanol-petrol blends means that more fuel should be consumed

by vehicles and in non automotive engines as the ethanol content in blends increases.

The increase in fuel consumption as a result of using ethanol-petrol blends is

discussed in later sections.

GCV on mass basis (kJ g-1) GCV on volume basis (kJ ml-1)


88

4.5 Engine Performance and Emission Characteristics of

Ethanol-Petrol Blends

Ethanol contains an oxygen atom hence; it can be treated as a partially oxidized

hydrocarbon. When ethanol is added to petrol to make the blended fuel, it provides

more oxygen for the combustion process and leads to the so-called ‘‘leaning effect’’

(Hsieh et al., 2002). The leaning effect leads to reduction in carbon monoxide (CO)

and the hydrocarbon (HC) emissions will also decrease under some operating

conditions. The carbon dioxide (CO2) increases when using ethanol-petrol blends due

to more complete combustion. In this section, the effects of the ethanol-petrol

blended fuels on the engine performance and pollutant emission are the main issues

and will be discussed in detail. These results and discussions are based in terms of

engine efficiency, fuel consumption and specific fuel consumption. The emission

characteristics of ethanol-petrol blends are represented in terms of CO, HC and CO2

emissions.

4.5.1 Engine Efficiency

Three different ethanol-petrol blends (E10, E15 and E20) were prepared and tested

for engine performance in a spark ignition (SI) engine. The engine efficiency using

the ethanol-petrol blends and petrol under varying load conditions is given in Figure

4.7 (For tabulated data refer to Appendix A).


89

Figure 4.7: Engine efficiency using Petrol and Ethanol-Petrol blends under

varying loads

There is a steady increase in the efficiency as the load increases in all the blends and

petrol. When the load on engines increase more fuel is supplied for burning to cater

for the increased load resulting in more energy input. Therefore, this results in high

efficiency with increased energy output. It is observed that the efficiency is quite

insensitive to the variation of the blend rate of ethanol-petrol blends and petrol

especially at lower loading up to 800 W. This indicates that other fuel properties such

as the laminar flame speed and heat of vaporization can counteract the energy

content in a fuel, as indicated by the heating value, to produce the same fuel

conversion efficiency (Curtis et al., 2008).

However, at high load the efficiency of petrol is slightly lower when compared to

E10, E15 and E20, this was also observed by Pikūnas et al. (2003). This can be

explained by the fact that the addition of ethanol in petrol results in leaning effect

which then increases the air-fuel equivalence ratio to a higher value, making the

burning closer to be stoichiometric (Hsieh et al., 2002). In other words addition of

ethanol results in better combustion which then makes it possible to have high

efficiency. Another observation to be noted is that at high load condition E10 has a

better efficiency followed by E15 and E20. This is probably because the heating

0

5

10

15

20

25

30

0 400 800 1200 1600 2000

Eng

ine

Eff

icie

ncy

(%)

Load (W)

Engine Efficiency Under Varying Load

Petrol E10 E15 E20


90

value or GCV of E10 more than E15 and E20. Similarly, E15 has a higher heating

value then E20.

Finally, it can be stated that apart from some differences obvious in the figures which

are not so radical, the load-efficiency curves for all ethanol-petrol blends generally

lie very close to that for petrol. This clearly indicates that the ethanol-petrol blends

studied under the various load conditions are not very different from petrol in their

power production capacity.

4.5.1.2 Engine Losses

The efficiency of engines that transform energy is never greater than the Carnot

efficiency. The Carnot efficiency depends on the temperature of the reservoirs

between which the engine is operating. A hundred percent efficiency can only be

obtained if the temperature of the cold reservoir is 0 K which is practically not

possible with heat engines (Jewett, 2004). In most cases, the cold reservoir

temperatures are around room temperatures of 300 K. These effects lead to Carnot

losses.

There are other energy losses that also decrease the engine efficiency as some of the

energy used does not perform useful work. This may be wasted in moving the engine

parts in order to overcome friction inherent in all engines.

Friction forces in engines are a consequence of hydrodynamic stress in oil films and

metal to metal contact. Frictional losses are significant to power produced in engines;

therefore minimizing the friction is a major consideration in engine design and

operation. In order to reduce friction, engines are usually lubricated. Frictional

energy is eventually removed as waste heat by the engine cooling system.

The frictional processes in engines are classified into three main components:

(1) the mechanical friction

(2) the pumping work

(3) the accessory work

The mechanical work includes the friction of the internal moving parts like the

crankshaft, piston, rings and the valve train. The accessory work is the work required


91

for the operation of the accessories such as the oil pump, fuel pump, alternator and

fan.

The pumping work is the net work done during the intake and exhaust strokes.

Pumping losses are caused by the way power output from an engine is regulated. It is

regulated by constricting airflow to the engine. This constriction of airflow creates

partial vacuum resulting in low pressure in the inlet manifold. Energy gets wasted

when maintaining this low pressure in the inlet manifold.

Some of the other losses associated in engines include; heat being carried away by

the exhaust gases, and also passing through the cylinder walls or cylinder head into

the engine cooling system, eventually passing to the atmosphere via the cooling

system.

4.5.2 Fuel Consumption

The relationship between the fuel consumption of the engine at different load

conditions for petrol and the ethanol-petrol blends is shown in Figure 4.8 (For

tabulated data refer to Appendix B).

Figure 4.8: Fuel consumption of Petrol, E10, E15, E20 under varying loads

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0 400 800 1200 1600 2000

Fuel

Con

sum

ptio

n (L

hr-1

)

Load (W)

Fuel consumption under varying load

Petrol E10 E15 E20


92

There is an increase in fuel consumption of the engine as the load increases and was

found to be maximum at the maximum load for petrol as well as the blended fuels.

Similar observations were made by Hsieh et al (2002), Al-Hasan (2003), Bayraktar

(2005). When the load increases the engine has to do more work to meet the higher

load conditions as a result more fuel is consumed to meet the power demand.

As seen in the graph the fuel consumption is highest for E20 followed by E15 and

then E10. The reason being that ethanol has a lower GCV than petrol, hence; as the

ethanol fraction in ethanol-petrol blend increases the GCV decreases. Therefore,

more fuel is consumed by increasing ethanol fractions in ethanol-petrol blends to

meet the increasing load demands.

An interesting observation that is made from Figure 4.8 is that the fuel consumption

for petrol at lower loads is almost similar to E10 and then at higher loads (above

1200 W) the fuel consumption is slightly more than E10. This can be explained by

the fact that ethanol addition to petrol makes the engine operation leaner and

improves combustion in the engine as well as performance and efficiency. The high

octane rating of ethanol can also to a certain degree compensate the inevitable drop

in the energy content of the fuel (Wallin et al., 2005).

4.5.3 Specific Fuel Consumption

The specific fuel consumption (SFC) is the amount of fuel consumed against the

work done by the engine over time at a constant load. The relationship between the

SFC under varying loads for petrol, E10, E15 and E20 is depicted in Figure 4.9 (For

tabulated data refer to Appendix B).

.


93

Figure 4.9: Specific Fuel Consumption of Petrol, E10, E15, E20 under varying load

Owing to the fact that the GCV of ethanol is lower than petrol, the SFC increases as

the ethanol content in blends increases. The mass ratio of air to fuel present in an

internal combustion engine is defined as Air Fuel Ratio (AFR). The theoretical AFR

of petrol is 1.6 times more than ethanol (Wu et al., 2004). For this reason the SFC

should increase with increase in ethanol content. However, the fuel injection strategy

tends to operate the engine at fuel-rich condition; as a result the ethanol addition

produces leaning effect to enhance the combustion of fuel (Hsieh et al., 2002). Also

usually at a fixed throttle opening and fixed engine speed, the intake of air is

constant. Therefore, to obtain the same AFR more volume flow rate of ethanol-petrol

blends is needed then petrol as a result compensating for the 1.6 times lower heating

value of ethanol than petrol (Wu et al., 2004). Due to these factors there is almost no

difference in the SFC of petrol and the ethanol-petrol blends.

4.5.4 Engine Exhausts Emission Analysis

An exhaust analyzer was used to measure the exhaust emissions resulting from the

influence of different ethanol-petrol blends at different load conditions. The

following sections will discuss the emissions: CO, HC and CO2.

0.4 0.5 0.6 0.7 0.8 0.9

1 1.1 1.2 1.3 1.4

0 400 800 1200 1600 2000 Spec

ific

fuel

con

sum

ptio

n (k

g kW

-1 h

r-1)

Load (W)

Specific Fuel Consumption under varying load

Petrol E10 E15 E20


94

4.5.4.1 CO Emission

Figure 4.10 shows the effect CO emission on of various blends of ethanol and petrol

under different loads.

Figure 4.10: Effect of varying loads on CO emissions for Petrol, E10, E15 and E20

It was observed that while using E20 to run the engine, the CO emission is less than

other fuels (Petrol, E10 and E15) for each engine loading. At idling, the CO decreses

by 26, 45 and 57 % with E10, E15 and E20, respectively, when compared to petrol.

At maximum load, there is a decrese in CO emission by 34, 61 and 78 % with E10,

E15 and E20, respectively, when compared to petrol. The trend that is observed is

reduction in CO emissions with incresing load and increasing ethanol concentration.

Similar results has also been observed by other researchers.

According to Zervas et al. (2003) the decrease in CO emission is due not only to

dilution of the the fuel but it is also because addition of oxygenated compounds

promotes combustion of CO in the cylinder or during postcombustion processes.

Ethanol is an oygenated compound as it contains an oxygen in its compound. When

ethanol containig the oxygen is mixed with petrol the combustion in the engine

becomes better resulting in reduction of CO.

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

0 400 800 1200 1600 2000

CO

(% v

ol)

Load (W)

CO Emission Under Varying Loads

ULP E10 E15 E20


95

It was noted that as the engine loading incresed the CO emissions decreased. This

can be explained by the fact that increasing loads on engine generation leads to an

increase in the combustion temperature, which when combines with high level of

excess oxygen at these loads results in lower CO emission when compared to low

engine loads (Saiyasitpanich et al., 2005).

In addition, when ethanol is added to petrol it creates the leaning effect which

increases the AFR to a higher value and thus causes the burning closer to

stoichiometric conditions. This results in achieving better combusion and also

increase in combustion temperature (Yücesu et al., 2006).

4.5.4.2 HC Emission

The HC emission characteristics as a function of load for ethanol-petrol blends and

petrol is given in Figure. 4.11.

Figure 4.11: Effect of varying loads on HC emissions for Petrol, E10, E15 and E20

Unburt HCs in exhaust is usually caused by three mechanisms: misfiring or

incomplete combustion , flame quenching effect and deposits or oil membranes (Wu

et al., 2004). As seen from Figure 4.11 HCs decrease as the ethanol content in petrol

increases. E20 has the minimum HC emission followed by E15, E10 and petrol,

0.00

10.00

20.00

30.00

40.00

50.00

60.00

70.00

80.00

0 400 800 1200 1600 2000

HC

(ppm

)

Load (W)

HC Emission Under Varying Loads

ULP E10 E15 E20


96

respectively. The decrease in HCs can be attributted to the fact that addition of

ethanol which is an oxygenated compound improves the combustion of HCs. Ethanol

can be treated as an partially oxidized HC when it is added to blended fuels, hence

HC emissions decrease (Yüksel and Yüksel, 2004). Addition of ethanol in fuel

enhances the increase in engine volumetric efficiency (Bayraktar, 2005). Reduction

in HC emission with incremental increase in ethanol content in blended fuel can also

be related to increase in volumetric efficiency (Zervas and Tazerout, 2000).

From Figure 4.11 it can be noted that initially there is are slight increases in HC

emission at low engine loads followed by decrease in HC emission. This trend is

observed for all the blended fuels that were tested. The reason why HC emissions

increase at lower engine loads is that high latent heat of vaporization of ethanol tends

to produce slow vaporization and mixing of fuel and air (He et al., 2003). However,

the reduction in HC emission for high engine load for blended fuels can be explained

by two factors (Cheung et al., 2008): (1) the high combustion temperature at high

engine loads enhances vaporization of ethanol-petrol blends and mixing with air,

leading to better combustion, (2) higher oxygen concentration due to addition of

ethanol in blended fuels promotes HC oxidation.

4.5.4.3 CO2 Emission

CO2 is one of the major greenhouse gases that contributes to greenhouse effect.

Figure 4.12 shows the effect of varying engine load on CO2 emissions for different

blends of ethanol-petrol.


97

Figure 4.12: Effect of varying loads on CO2 emissions for Petrol, E10, E15 and E20

It is observed from Figure 4.12 that CO2 emission increases with increasing load,

with petrol having the highest CO2 emission at all engine load, followed by E10, E15

and E20, respectively. The reason for CO2 reduction with increasing ethanol content

is low carbon/hydrogen ratio and high oxygen content in blends (He et al., 2003).

High oygen content in ethanol-petrol blends ensures that the combustion is complete

as a result of this CO2 is produced. The reduction in CO2 at maximum engine load in

E10 is 7 %, 17 % for E15 and 20 % for E20 when compared to petrol. Similar trend

is also observed at other engine loading, the reason being that ethanol contains less

carbon atoms than petrol, therefore, as ethanol content in blended fuels increase, it

gives off lower CO2 (Celik, 2008).

Another important relationship to note is that CO and CO2 have a complementary

relationship. When comparing Figure 4.10 and 4.12 it can be noted that as CO

decreses for all engine loads the CO2 increases. According to Wu et al. (2004) CO2

emission is dependent on the AFR and the CO emission concentration.

0

1

2

3

4

5

6

7

8

0 400 800 1200 1600 2000

CO

2 (%

vol

)

Load (W)

CO2 Emission Under Varying Loads

ULP E10 E15 E20


98

4.5.4.4 The Effects of Various Fuels on Exhaust Emissions at Constant

Load

It is more enlightening to compare the emissions under a constant load for different

fuels. Figures 4.13-4.15 show the CO, HC and CO2 emissions for the various fuels

when the engine is operated at maximum load (1863.5 W). In all the graph it is

observed that there is a decrease in emissions when the proportion of ethanol in

petrol is increased. E20 gives the highest reduction in CO, HC and CO2 emissions.

Figure 4.13: CO emissions for various fuels at maximum load

0

0.5

1

1.5

2

2.5

Petrol E10 E15 E20

CO

(% v

ol)

CO emission at maximum load for Petrol, E10, E15 and E20

0

10

20

30

40

50

60

70

Petrol E10 E15 E20

HC

(ppm

)

HC emission at maximum load for Petrol, E10, E15 and E20


99

Figure 4.14: HC emissions for various fuels at maximum load

Figure 4.15: CO2 emissions for various fuels at maximum load

In this study, it was found that using ethanol-petrol blended fuels resulted in

reduction of CO and HC emissions by 21-78 % and 6-67 %, respectively. Also

reductions in CO2 emission were by 6-20 % with respect to petrol depending on

engine loading. This is comparable to results obtained by Hsieh (2002) which

showed a 10-90 % reduction in CO and 20-80 % in HC as a result of leaning effect.

In terms of reduction of emissions E20 was found to be the best blend. Al-Hasan

(2003) used a range of blends from 0 % to 25 % in increments of 2.5 % and found

E20 to give the best results. The tabulated data for emissions is given in Appendix C.

4.5.5 Prospects and Challenges for Bio-ethanol Use in Vehicles

Worldwide liquid biofuels make a small but increasing contribution to fuel usage

(REN21, 2012). In 2011 biofuels provided about 3 % of global road transport fuels

and is expected to rise to 27 % by 2050 (IEA, 2011). Therefore, it is timely for

developing countries like Fiji to consider developing their biofuel industry.

0

1

2

3

4

5

6

7

8

Petrol E10 E15 E20

CO

2 (%

vol

)

CO2 emission at maximum load for Petrol, E10, E15 and E20


100

With the policy for use of biofuels namely bio-diesel and bio-ethanol already in place

by the Fiji Department of Energy (FDOE), it should start looking towards sustainable

production and use of biofuels for transportation. SI engine cars later than the 1990

model in Fiji can use a blend of up to 10 % ethanol in petrol without modification of

engines. For models older than 1990 modification to the carburetor will be required

before ethanol-petrol blends with 5 to 10 % ethanol can be used (ANFAVEA, 2005).

This mainly includes the material of the body and cover of the carburettor which

cannot be aluminium or Zamak and also any component of Nylon, which will have to

be replaced if unprotected. As most fleets in Fiji are currently 1990 models and later,

therefore, compatibility issue should not arise when using blends containing up to a

maximum of 10 % ethanol.

Generally, compatibility issues when using blends of ethanol-petrol are due to

metallic materials of the vehicle (giving rise to corrosion), plastic and rubber

materials of vehicle (which are susceptible to chemical attack), high fuel

consumption (due to low energy content of ethanol), losses in drivability (due to

different AFR for combustion) and cold start difficulties (due to lower vapour

pressure) (Coelho et al., 2005). However, the extent of these compatibility issues

depend on the blends of ethanol-petrol used, quality and specification of ethanol and

the vehicle technology. Many countries such as the United States (US), Bolivia,

China, Colombia, Jamaica, Kenya, Nigeria, Venezuela, Argentina, Canada, Chile,

India, Philippines, and Costa Rica are already using petrol blended with 5-10 %

ethanol (IEA, 2011).

In order to use higher blends of ethanol-petrol in Fiji, certain modifications to

engines will have to be made so that the benefits associated with the higher oxygen

content could be utilized which include improved fuel efficiency and reduced

emissions. The modifications required for using higher blends include engines with

higher compression ratios so that ethanol’s higher octane rating can be used to

achieve higher efficiencies, and modifications to the fuel feed system and ignition in

order to compensate for differences in the AFR. In addition, modification of some

materials that come in contact with the fuel, such as anticorrosive treatment of the

metal surfaces of fuel tanks, fuel filters and pumps, substitution of fuel lines, and use


101

of materials which are more compatible with ethanol are necessary (BNDES and

CGEE, 2008).

Flex-fuel, or flexible fuel, vehicles (FFVs) are a solution to using higher blends of

ethanol-petrol. These have engines that can alternate between two sources of fuel,

including petrol and bio-ethanol or petrol and natural gas (The Royal Society, 2008).

FFVs for petrol and bio-ethanol have been used extensively in Brazil and to some

extent in USA and Sweden and offer many advantages (Joseph 2007). FFVs in Brazil

can operate with up to 100 % hydrated ethanol but in the US flex-fuel models can

use a maximum of 85 % ethanol, due to problems at cold start. FFV vehicles are

equipped with electronic sensors which enable the on-board computer to recognize

the fuel type and adjust the engine combustion parameters, with no interference from

the driver (Joseph, 2007). FFVs may become a solution for greater bio-ethanol usage.

With the development of appropriate market incentives to encourage the

development and supply of FFVs, these vehicles could offer more flexibility to

increase bio-ethanol usage (The Royal Society, 2008).

Currently, Volkswagen, General Motors, Fiat, Ford, Peugeot, and Renault are some

of manufacturers producing FFVs in Brazil (Coelho et al., 2005). In Fiji, currently

there has been no move to introduce FFVs. Therefore, the focus is mostly on

introducing blends up to 10 % ethanol with 90 % petrol in the existing vehicle fleets.

However, this might change should the bio-ethanol industry develop and the

production of bio-ethanol increases. In order to ensure sustainability, with increasing

bio-ethanol use there needs to be parallel and compatible development of engine

technologies, bio-ethanol feedstock and feedstock conversion technology.

102

Chapter 5 Conclusions

The aim of this study was to produce ethanol from a renewable resource available in

Fiji which could be considered for blending with petrol for use in spark ignition (SI)

engines. This was accomplished by producing ethanol from ten selected cassava

varieties available in Fiji at two different locations and determining the ethanol yield

from different varieties. Ethanol-petrol blend testing using E10, E15 and E20 were

also conducted on an SI engine and their performance in terms of engine efficiency,

fuel consumption and exhaust emissions were compared.

Cassava roots have a number of end-uses, such as food and feed processing, the

starch industry, bio-ethanol production and as well as contributing to the food export

industry. The ten varieties of Fiji cassava possessed a range of starch yields and dry

matter content. . Cassava can be grown in poor soil and other marginal conditions,

making it a resilient crop. While the Fijian varieties studied had lower starch yields

than those reported for other countries, with suitable farming practices cassava root

yields as well as starch yields can be increased. Ethanol concentration obtained after

120 hours of fermentation was in the range of 55-63 g l-1 for cassava obtained from

Koronivia Research Station (KRS) and 55-65 g l-1 for cassava obtained from

Dobuilevu Research Station (DRS) with an ethanol yield of 0.35-0.40 L of ethanol

per kg of starch and 0.35-0.41 L of ethanol per kg of starch for KRS and DRS

respectively. For high ethanol yield, high starch content in cassava roots is the most

important criterion. Therefore the development of high starch content varieties

through breeding should be considered.

Furthermore, experimental results indicated that using ethanol-petrol blended fuels,

the engine efficiency at lower loading of up to 800 W was quite insensitive to the

ethanol content in the blends but at high load the efficiency of petrol is slightly lower

when compared to the ethanol blends E10, E15 and E20. Fuel consumption of the

engine increased slightly with increasing ethanol content in the blends. Whereas, CO

and HC emissions decrease dramatically as a result of the leaning effect caused by

Chapter 5: Conclusions

103

the ethanol addition, the CO2 emission increases because of the improved

combustion. In this study, it was found that using ethanol–petrol blended fuels

resulted in the reduction of CO and HC emissions by 21-78 % and 6-67 %,

respectively, while CO2 emission decreases by 6-20 % depending on engine

conditions.

Although, the current research was conducted on a laboratory scale, the results

obtained provide valuable insight into the commercial viability of ethanol-petrol

blends. They suggest that the blends are viable candidates as alternative fuels for SI

engines. The study also reveals that for tropical conditions, it is possible to use 96 %

ethanol for blending, without any deleterious effects on the engine. The results of this

study also show that as far as the engine efficiency is concerned, ethanol-petrol fuels

will produce very similar engine performance as petrol.

As cassava is primarily produced for food in Fiji, a suitable compromise needs to be

established that would balance out the use of agricultural land for food and fuel. The

use of food crops for fuel usually drives up the prices of these crops. Although, the

Fiji Government has dismissed this threat, indicating that there is sufficient unused

land, there is still a need to take necessary precautions to avoid a food versus fuel

crisis in Fiji. Setting aside land for food production is one strategy. However,

governments need to make a national-level decision as to what extent staple crops

should be used for biofuel production. Clearly, there are several measures that need

to be taken in order to develop bio-ethanol production on a sustainable basis. These

should include the proper coordination and integration of national policies and the

conduct of feasibility studies which evaluate the opportunities for bio-ethanol

production from different feedstocks.

The most common reason for biofuels development would be to reduce the

greenhouse gas emissions. However, Pacific Island Countries (PICs) are amongst the

lowest carbon emitters in the world. Fiji has a carbon footprint of 1.7 t and is ranked

131 emitter of the world (Singh, 2012). Therefore, considering such low overall

emission of PICs, it becomes clear that the main reason to replace fosil fuels with

biofuels should be to improve the country’s economy and to reduce the import bills.

It may not be feasible to replace all fossil fuel imports in Fiji with biofuels for


104

transportation. However, reduction in fossil fuel imports would be possible by

incorporating the use of biofuel blends in the energy mix of Fiji (Singh, 2012). In

order for Fiji to reduce its petroleum import, drastic measures may be required.

Having a large agro-based economy, Fiji’s bio-ethanol advantage is very much due

to the rich endowment of natural resources. Ethanol can also be an option for other

larger Pacific Island Countries that can support sufficient amounts of sugary or

starchy crops.

5.1 Recommendations and Suggestions for Future Work

The present research has only looked at ten cassava varieties for starch yield and dry

matter content. Research needs to be carried out on the other varieties of cassava that

are available in Fiji and from various other locations. The Ministry of Agriculture in

Fiji should do research on new and better varieties of cassava that are more suitable

to the climatic condition and are high yielding. They should also monitor the new

varieties released for large scale farming and promote the use of superior varieties of

cassava to farmers. Also other possible root crops such as yams should be considered

for starch and bio-ethanol production. Similarly, ethanol yield from other varieties of

starch should also be considered. As the other varieties might have high starch

content and more efficient in ethanol conversion.

The ethanol-petrol blends in this study were only tested for a short duration of time

on the SI engine. Long term research, such as a 2000 hours material compatibility

testing, needs to be carried out in order to determine effect of blends on various

engine components.

Some more fuel properties could have been tested such as flash point and vapor

pressure. However, due to lack of availability of instruments this could not be done.

Therefore, in future research should consider having these properties to be tested by

sending samples abroad.

Fiji already has a policy in place for the use of 10 % ethanol blended with petrol in

vehicles. However, this has not been made mandatory yet due to lack of production

capacity. Therefore, in order to promote the use of bio-ethanol successfully, it should


105

start considering production of ethanol from improved technology so that production

costs are lowered.

Government research investment in the development of crop technology should

focus on interventions that would lower the high labour and input requirements of

cassava production. Higher budget allocation for research on farm mechanization and

varietal development should be provided by government to increase yield, increase

starch content and consequently increase ethanol productivity of cassava as

feedstock. Also tax exemptions for fertilizers and agro-chemicals as well as fuel for

farm machineries would significantly help reduce farm production costs. This will

ensure the total production cost of bio-ethanol is lower than the price of petrol.

106

References

Abdel-Rahman, A. A. and M. M. Osman, 1997. Experimental investigation on

varying compression ratio of SI engine working under different ethanol-gasoline fuel

blends. International Journal of Energy Research 21: 31-40.

Al-Hasan, M., 2003. Effect of ethanol-unleaded gasoline blends on engine

performance and exhaust emission. Energy Conversion and Management 44: 1547-

1561.

Amin, G., 1992. Conversion of sugar beet particles to ethanol by the bacterium

Zymomonas mobilis in solid state fermentation. Biotechnology Letter 14 (6): 499-504.

Amutha, R. and P. Gunasekaran, 2001. Production of ethanol from liquefied cassava

starch using co-immobilized cells of Zymomonas mobilis and Saccharomyces

diastaticus. Journal of Bioscience and Bioengineering 92 (6): 560-564.

ANFAVEA, 2005. Ethanol Fuel Vehicular Application Technology. Presentation of

Henry Joseph Jr. (Brazilian Automotive Industry Associations Energy & Environment

Commission [email protected]) at CEPAL/CENBIO/USP Seminar São Paulo,

17th August, 2005.

Asaoka, M., J. M. V. Blanshard, and J. E. Rickard, 1991. Seasonal effects on the

physico-chemical properties of starch from four cultivars of cassava. Starch/Starke

43: 455-459.

Asaoka, M., J. M. V. Blanshard, and J. E. Rickard, 1992. Effects of cultivar and

growth season on the gelatinization properties of cassava (Manihot esculenta) starch.

Journal of the Science of Food and Agriculture 59: 53-58.

Ashoka, P. V., S. V. Nair, T. M. Kurian, 1984. Influence of stages of harvest on yield

and quality of cassava, Manihot esculenta Crantz. Madras Agricultural Journal

71:447-449.

References

107

AUS-e-TUTE, n. d., Chemistry Tutorial: Reducing and Non-reducing Sugars.

Available online <http://www.ausetute.com.au/redsugar.html> (date accessed 5th

July, 2012).

Aylward, G. and T. Findlay, 2003. SI Chemical Data, 5th Ed. John Wiley and Sons,

Sussex, UK.

Banat, I. M., P. Singh, and R. Marchant, 1996. The use of thermotolerant fermentative

Kluyveromyces marxianus IMB3 yeast strain for ethanol production. Acta

Biotechnology 16: 215-223.

Bayraktar, H., 2005. Experimental and theoretical investigation of using gasoline–

ethanol blends in spark ignition engines. Renewable Energy 30:1733–1747.

Beer, T., T. Grant, G. Morgan, J. Lapszewicz, P. Anyon, J. Edwards, P. Nelson, H.

Watson and D. Williams, 2001. Comparison of Transport Fuels. Final Report

(EV45A/2/F3C), Australian Greenhouse Office, Victoria, Australia.

Beer, T., T. Grant, D. Williams and H. Watson, 2002. Fuel-cycle greenhouse gas

emissions from alternative fuels in Australian heavy vehicles. Atmospheric

Environment 36: 753-763.

Benesi, I. R. M., M. T. Labuschagne, A. G. O. Dixon, and N. M. Mahungu, 2004.

Stability of native starch quality parameters, starch extraction and root dry matter of

cassava genotypes in different environments. Journal of the Science of Food and

Agriculture 84: 1381-1388.

Benesi, M., 2005, Characterization of Malawian Cassava Germplasm for diversity,

starch extraction and its native and modified properties, PhD thesis, Free State

University, South Africa.

Biofuels Digest, 2012. Brazil sets up $38 billion ethanol subsidy program to stimulate

expansion, Biofuels Digest, 27 February 2012. Available

References

108

<http://www.biofuelsdigest.com/bdigest/2012/02/27/brazil-sets-up-38-billion-ethanol-

subsidy-program-to-stimulate-expansion/> (date accessed 10th December, 2012).

Birse, D. G. and J. E. Cecil, 1980. Starch Extraction: a checklist of commercially

available machinery. Tropical Products Institute, London, England.

BNDES and CGEE, 2008. Sugarcane-based bioethanol : energy for sustainable

Development. Rio de Janeiro. Available online <www.sugarcanebioethanol.org> (date

accessed 9th January 2011).

Bureau of Statistics, 2011. Key Statistics 2010. Bureau of Statistics, Suva, Fiji.

Burns, A., R. Gleadow, J. Cliff, A. Zacarias and T. Cavagnaro, 2010. Cassava: The

drought, war and femine crop in the changing world. Sustainability 2: 3572-3607.

Butkus, A. and S. Pukalskas, 2004. The research into the influence of ecological

petrol additives in the automobile laboratory. Transport X1X (1): 24-27.

Campbell, M. K. and S. O. Farrell, 2009. Biochemistry, 6th Ed. Brooks/Cole Cengage

Learning. Kentucky, USA.

Carter, S., 2003. Engine Efficiency: Good or Bad? Available < http://ffden-

2.phys.uaf.edu/212_fall2003.web.dir/Sarah_Carter/index.html> (date accessed 12th

December, 2012).

Ceballos, H., C. A. Iglesias, J. C. Pèrez, and A. G. O. Dixon, 2004. Cassava breeding

opportunities and challenges. Plant Molecular Biology 56: 503-516.

Celik, M. B., 2008. Experimental determination of suitable ethanol-gasoline blend

rate at high compression ratio for gasoline engine. Applied Thermal Engineering 28:

396-404.

References

109

Chandel, A. K., E. S. Chan, R. Rudravaram, M. L. Narasu, L. V. Rao, and P.

Ravindra, 2007. Economics and environmental impact of bioethanol production

technologies: An appraisal. Biotechnology and Molecular Biology 2 (1): 14-32.

Cheung, C. S., Y. Di and Z. Huang, 2008. Experimental investigation of regulated and

unregulated emissions from a diesel engine fueled with ultralow-sulphur diesel fuel

blended with ethanol and dodecanol. Atmospheric Environment 42: 8843-8851.

Cloin, J., A. Woodruff and D. Furstenwerth, 2007. Liquid Biofuels in Pacific Island

Countries. SOPAC Miscellaneous Report 628.

Colitt, R. and S. Nielsen, 2012. Brazil Ethanol Drive Falters on Domestic Supply

Shortage. Bloomberg Businessweek, 13 March 2012. Available

<http://www.businessweek.com/news/2012-03-13/brazil-ethanol-slows> (date

accessed: 16th December, 2012).

Cock, J., 1985. Cassava- New Potential for a Neglected Crop. Westview Press.

Boulder, Colorado, USA.

Cock, J. H., 1982. Cassava: A Basic Energy Source in the Tropics. Science 218: 755-

762.

Coelho, S. T., J. Goldemberg, O. Lucon and P. Guardabassi, 2005. Brazilian

sugarcane ethanol: lessons learned. Paper prepared for STAP workshop on Liquid

Biofuels, Delhi, 29th August to 2nd September, 2005.

Curtis, S., M. Owen, T. Hess and S. Egan, 2008. Effect of Ethanol Blends on a Spark

Ignition, 4-Stroke, Internal Combustion Engine. Brigham Young University Provo,

Utah. Available online <http://www.et.byu.edu/~curtis5/Ethanol%20Paper.pdf> (date

accessed: 16th August, 2011).

Defloor, I., R. Swennen, M. Bokanga and J. A. Delcour, 1998. Moisture stress during

growth affects the breadmaking and gelatinization properties of cassava (Manihot

esculenta Crantz) flour. Journal of the Science of Food and Agriculture 76: 233-238.

References

110

De Taful, S. M., M. A. El-Sharkawy and F. Calle, 1997. Photosynthesis and yield

performance of cassava in seasonally dry and semiarid environments. Photosynthetica

33: 229-257.

Dubey, R. C., 2001. Textbook of Biotechnology, 1st Ed. S. Chand and Company

Limited, New Delhi, India.

Duncan, J., 2002. Blending ethanol into petrol: An Overview. Report Prepared for

Energy Efficiency and Conservation Authority, Wellington, New Zealand.

Eksteen, J. M., P. van Rensburg, R. R. Cordero Otero, I. S. Pretorius, 2003. Starch

fermentation by recombinant Saccharomyces cerevisiae strains expressing the α-

amylase and glucoamylase from Lipomyces kononenkoae and Saccharomycopsis

fibruligera. Biotechnology and Bioengineering 84: 639-646.

El-Sharkawy, M. A., 2004. Cassava biology and physiology. Plant Molecular Biology

56: 481-501.

Environment Australia, 2002. Setting the Limit of Ethanol in Petrol. Available online

<http://www.environment.gov.au/atmosphere/fuelquality/publications/ethanol-

limit/index.html> (date accessed 25th January 2010).

Ethanol in Petrol. Available online <http://www.raa.com.au/page.asp?TerID=146>

(date accessed: 14th November, 2010).

Ferfecki, F. J. and S. C. Sorenson, 1983. Performance of ethanol blends in gasoline

engines. American Society of Agricultural Engineers 26 (1): 0038-0043.

Ferguson, C. R. and A. T. Kirkpatrick, 2001. Internal Combusion Engines Applied

Thermosciences, 2nd Ed. John Wiley and Sons New Jersey, USA.

Fuel Phase Separation in Ethanol Blended Gasoline? Available online

<http://www.enertechlabs.com/fuel_phase_separation_in_ethanol.php> (date

accessed: 2nd January, 2010).

References

111

Gibbons, W. R., C. A. Westby, and T. L. Dobbs, 1986. Intermediate scale,

semicontinuous solid phase fermentation process for production of fuel ethanol from

Sweet Sorghum. Applied Environment Microbiology 51 (1): 115-122.

Grace, M. R., 1977. Cassava production. Food and Agriculture Organization, United

Nations, Rome, Italy.

Gravalos, I., D. Moshou, T. Gialamas, P. Xyradakis, K. Kateris, and Z. Tsiropoulos,

2011. Performance and emission characteristics of spark ignition engine fuelled with

ethanol and methanol gasoline blended fuels. In: Manzanera, M. (Ed.) Alternative

Fuel, ISBN: 978-953-307-372-9, InTech. Available online

<http://www.intechopen.com/articles/show/title/performance-and-emission

characteristics-of-spark-ignition-engine-fuelled-with-ethanol-and-methanol> (date

accessed 2nd February, 2012).

Guibet, J. C., 1999. Fuels and Engines: Technology, Energy, Environment, Vol. 2.

Enfield Pub & Distribution Co, New Hampshire, USA.

He, B-H., S-J. Shuai, J-X. Wang and H. He, 2003. The effect of ethanol blended

diesel fuels on emissions from a diesel engine. Atmospheric Environment 37: 4965-

4971.

Hofstrand, D., 2009. Brazil’s Ethanol Industry. AgDM Newsletter, January, 2009.

Hsieh, W-D., R-H Chen, T-L Wu, and T-H Lin, 2002. Engine performance and

pollutant emission of an SI engine using ethanol–gasoline blended fuels. Atmospheric

Environment 36: 403–410.

ICS-Nigeria, n. d. Growing Cassava in Nigeria. Available

<http://www.cassavabiz.org/agroenterprise/ent%20images/cassava_02.pdf> (date

accessed 10th December, 2012).

IEA, 2004. Biofuels for transport: an international perspective. International Energy

Agency, Paris, France.

References

112

IEA, 2011. Technology Roadmap:Biofuels for transport. International Energy

Agency, Paris, France.

Iglesias, C., C. Hershey, F. Calle, and A. Bolaños, 1994. Propagating cassava

(Manihot esculenta) by sexual seed. Experimental Agriculture 30: 283-290.

International Starch Institute, 1999 (a). Determination of ash in Starch at 900 ˚C: ISI

02-1e. Available online <http://www.starch.dk/isi/methods/02ash.htm> (date

accesses: 13th May, 2009).

International Starch Institute, 1999 (b). Determination of pH in Starch and Syrup: ISI

26-5e. Available online <http://www.starch.dk/isi/methods/26ph.htm> (date accessed:

13th May 2009).

Jamai, L., K. Ettayebi, J. E. Yamani, and M. Ettayebi, 2007. Production of ethanol

from starch by free and immobilized Candida tropicalis in presence of α-amylase.

Bioresource Technology 98: 2765-2770.

Jamai, L., K. Sendide, K. Ettayebi, F. Errachidi, O. Hamdouni-Alami, M. A. Tahri-

Jouti, T. McDermont and M. Ettayebi, 2001. Physiological difference during ethanol

fermentation between calcium alginate-immobilized Candida tropicalis and

Saccharomyces cerevisiae. FEMS Microbiology Letters 204: 375-379.

Jewett, S., 2004. Physics for Scientists and Engineers with Modern Physics, 6th

Edition. Brooks/Cole-Thomson Learning, Califonia, USA. Jones, W. O., 1959.

Manioc in Africa. Stanford University Press, California, USA.

Joseph H., 2007. The vehicle adaptation to ethanol fuel. Presentation to the Royal

Society International Biofuels Meeting held in London, April 2007. Available online

at< www.royalsoc.ac.uk/downloaddoc.asp?id=4248> (date accessed: 13th May 2012).

Karaosmanoglu, F., A. Isigigur and H. A. Aksoy, 1992. The effects of methanol-

gasoline blends on exhaust emissions. Presentation at the International Conference

References

113

Next Generation Technologies for Efficient Energy, End Uses and Fuel Switching.

Dortmund, Germany.

Kargi, F., J. A. Curme and J. Sheehan, 1985. Solid state fermentation of sweet

sorghum to ethanol. Biotechnology Bioengineering 27 (1): 34-40.

Kawano, K. 2003. Thirty years of cassava breeding for productivity-biological and

social factors for success. Crop Science 43: 1325-1335.

Kawano, K., P. Daza, A. Amaya, M. Rios and W. M. F. Goncalves, 1978. Evaluation

of cassava germplasm for productivity. Crop Science 18: 377-380.

Kearsley, M. W. and S. Z. Dziedzic (Eds.), 1995. Handbook of starch hydrolysis

products and their derivatives. Blackie Academic & Professional, New York, USA.

Keating, B. A., G. L. Wilson and J. P. Evenson, 1988. Effects of length, thickness,

orientation and planting density of cassava (Manihot esculenta Crantz) planting

material on subsequent establishment, growth and yield. East African Agricultural

and Forestry Journal 53: 145-149.

Koç, M., Y. Sekme, T. Topgül and H. S. Yücesu, 2009. The effects of ethanol–

unleaded gasoline blends on engine performance and exhaust emissions in a spark-

ignition engine. Renewable Energy 34: 2101-2106.

Lancaster, P. A. and J. E. Brooks, 1983. Cassava leaves as human food. Economic

Botany 37: 331-348.

Lee, C. G., C. H. Kim and S. K. Rhee, 1992. A kinetic model and simulation of starch

saccharification and simultaneous ethanol fermentation by amyloglucosidase and

Zymomonas mobilis. Bioprocess Engineering 7: 335-341.

Li, S-Z and C. C. Halbrendt, 2009. Ethanol production in (the) People’s Republic of

China: Potential and technologies. Applied Energy 86: S162–S169.

References

114

Lin, W., Y. Chang and Y. Hsieh, 2010. Effect of ethanol-gasoline blends on small

engine generator energy efficiency and exhaust emission. Journal of Air and Waste

Management Association 60: 142-148.

Liu, S-Y. and C-L. Liang, 1983. Studies on the efficiency of alcohol production in

sweet potato, cassava and potato. Journal of Agricultural Resources China 32 (2):

111-121.

Lonsane, B. K. and M. M. Krishnaiah, 1994. Solid state fermentation foods of Indian

origin. In: Pandey, A. (Ed.) Solid-state fermentation: An overview, Wiley Eastern

Publishers, New Delhi, India.

Mason, R. R., 1956. Cassava varieties in Fiji. Fiji Agricultural Journal 27 (3/4): 88-

93.

Miller, G. L., 1959. Use of Dinitrosalicylic Acid Reagent for determination of

reducing sugar. Analytical Chemistry 31 (3): 426-428.

Montesinos, T. and J. Navarro, 2000. Production of alcohol from raw wheat flour by

amyloglucosidase and Saccharomyces cerevisiae. Enzyme and Microbial Technology

27 (6): 362-370.

Moorthy, S., 2001. Tuber crop starches. Technical Bulletin No. 18, Central Tuber

Crops Research Institute, Kerala, India.

Moorthy, S. N. and T. Ramanujam, 1986. Variation in properties of starch in cassava

varieties in relation to age of the crop. Starch/Starke 38: 8-61.

Nakamura, Y., F. Kobayashi, M. Ohnaga and T. Swada, 1997. Alcoholic fermentation

of starch by a genetic recombinant yeast having glucoamylase activity. Biotechnology

Bioengineering 53: 21-25.

Nakavulevu, P, 2011. Promoting Renewable Energy Policies in Fiji. Presentation at

workshop organized by IRENA “Accelerated Renewable Energy Deployment on

References

115

Islands with Emphasis on Pacific Islands”, 26-28th October, Sydney, Australia.

Nanda, S. K., C. Balagopalan, G. Padmaja, S. N. Moorthy and M. S. Sanjeev, n. d.

Post-Harvest Management of Cassava for Industrial Utilization in India, Central

Tuber Crops Research Institute (CTCRI), India. Available online

<http://webapp.ciat.cgiar.org/> (date accessed 8th January 2010).

National Starch and Chemical Company, 2002. 100 Years of Food Starch History.

Available online <http://www.foodstarch.com/about/abo_fhistory.asp> (date accessed

2nd September, 2009).

Ngendahayo, M. and A. G. O. Dixon, 2001. Effect of harvest on tuber yield, dry

matter, starch and harvest index of cassava in two ecological zones in Nigeria. In:

Akoroda, M. O., and J. M. Ngeve (Eds.) Root Crops in the Twenty-first Century,

Proceedings of the seventh Triannual Symposium of the International society for

Tropical Root Crops- Africa Branch (ISTRC-AB), Centre International des

Conférences, Cotonou, Benin.

Nigam, P., I. M. Banat, D. Singh, A. P. McHale and R. Marchant, 1998. Continuous

ethanol production by thermotolerant Kluyveromyces marxianus IMB3 immobilized

on mineral kissiris at 45 ºC. World Journal of Microbiology Biotechnology 13: 283-

288.

Nuwamanya, E., Y. Baguma, R. S. Kawuki and P. R. Rubaihayo, 2008.

Quantification of starch physicochemical characteristics in a cassava segregating

population. African Crop Science Journal 16: 191-202.

OECD, 2011. Brazil’s biofuel sector: What future? OECD Observer, No. 287.

Available

<http://www.oecdobserver.org/m/fullstory.php/aid/3748/Brazil_92s_biofuel_sector:_

What_future_.html> (date accessed 13th December, 2012).

References

116

Ocloo, F. C. K. and G. S. Ayernor, 2008. Physical, chemical and microbiological

changes in alcoholic fermentation of sugar syrup from cassava flour. African Journal

of Biotechnology 7 (2): 164-168.

O’Hair, S. K., 1990. Tropical Root and Tuber Crops. In: Janick, J. (Ed.) Horticultural

Reviews, Vol. 12. Timber Press, Oregon, USA.

O’Hair, S. K., J. M. Dangler, P. H. Everett, R. B. Forbes, L. H. Halsey, S. J. Locascio,

H. Y. Ozaki, J. R. Rich, R. L. Stanley, H. J. Trafford and J. M. White, 1981. Location,

growing season and soil type effects on Florida cassava yields. HortScience 16: 262-

267.

Pandey, A., 1994. Solid-state fermentation: An overview. Wiley Eastern Publishers,

New Delhi, India.

Pazur, J. H. and T. Ando, 1960. The hydrolysis of glucosyl oligosaccharides with α-

D-(1→4) and α-D-(1→6) bonds by fungal amyloglucosidase. Journal of Biological

Chemistry 235: 297-302.Pellet, D. M. and M. A. El-Sharkawy, 1997. Cassava varietal

response to fertilization: growth dynamics and implications for cropping

sustainability. Experimental Agriculture 33: 353-365.

Pèrez J. C., N. Morante, J. López, J. I. Lenis, G. Jaramillo, H. Ceballos, F. Calle,

2002. Advantages of the New Cassava breeding scheme at CIAT. In: Taylor N. J., F.

Ogbe, C. M. Fauquet (Eds.) Cassava, an ancient crop for modern times food, health,

culture. A paper presented at 5th international scientific meeting of the cassava

biotechnology network, Danforth Plant Science Center St. Louis, Missouri, USA.

Phakatkar, H. G., 2005. Theory of Machines and Mechanisms-II, 4th Ed. Nirali

Prakashan, Pune, India.

Pikūnas, A., S. Pukalskas and J. Grabys, 2003. Influence of composition of gasoline

ethanol blends on parameter of internal combustion engines. Journal of Kones

Internal Combustion Engines 10: 205-211.

References

117

Puhan, S. and G. Nagarajan, 2008. NOx reduction in a DI diesel engine using

biodiesel as a renewable fuel. International Journal of Sustainable Energy 27 (3):

143-154.Rao, P. J. M., 1997. Industrial utilization of sugarcane and its co-product,

Indian Commission of Sugar Industry Development, ISPCK Publishers, New Delhi,

India.

REN21, 2012. Renewables 2012 Global Status Report, REN21 Secretariat, Paris,

France.

Rodriguez-Sosa, E. J., O. Parsi-Ros and M. A. Gonzalez, 1976. Composition of

cassava (Manihot esculenta Crantz) and the rheological characteristics of its starch.

Journal of Agriculture of the University of Puerto Rico 61: 32-40.

Rogers, D. J. and S. G. Appan, 1970. Untapped genetic resources for cassava

improvement. In Proceedings of the 2nd International Symposium on Tropical Root

and Tuber Crops, University Hawaii Press, Honolulu. Hawaii.

Rogers, D. J. and S. G. Appan, 1973. Manihot and Manihotoides (Euphorbiaceae), a

computer assisted study. In: Rogerson, C. T. (Ed.) Flora Neotropica, Monograph 13.

Hafner Press, New York, USA.

Ryu, B. H., K. D. Nam, H. S. Kim, D. S. Kim, Y. A. Ji and S. J. Jung, 1988.

Screening of thermotolerant yeast strain for ethanol production. Korean Journal of

Applied Microbiology Bioengineering 16. 265-269.

Saiyasitpanich, P., T. C. Keener, S-J. Khang, M. Lu, F. Linag, 2005. The effect of

diesel fuel sulfur content on particulate matter emissions for a nonroad diesel

generator. Journal of Air & Waste Management. Association 55: 993-998.

Santisopasri, V., K. Kurotjanawong, S. Chotineeranat, K. Piyachomkwan, K. Sriroth

and C. G. Oates, 2001. Impact of water stress on yield and quality of cassava starch.

Industrial Crops and Products 13 (2): 115-129.

References

118

Shigechi, H., K. Uyama, Y. Fujita, T. Matsumoto, M. Ueda, A. Tanaka, H. Fukuda,

A. Kondo, 2002. Efficient ethanol production from starch through development of

novel flocculent yeast strains displaying glucoamylase and co-displaying or secreting

α-amylase. Journal of Molecular Catalysis B: Enzymatic 17: 179-187.

Silvestre, P., 1989. Cassava. The Tropical Agricultural Series. Macmillan with the

Technical Centre for Agriculture and Rural Cooperation (CTA), London, England.

Singh, A., 2012. Biofuels and Fiji’s roadmap to energy self-sufficiency. Biofuels 3:

269-284.

Singh, P. J., 2009. Preparation, characterisation and engine performance of coconut

oil based hybrid fuels. MSc thesis, School of Engineering and Physics, The University

of the South Pacific, Suva, Fiji.

SOPAC, 2009. Potential for liquid Biofuels in Fiji. SOPAC Miscellaneous Report

677.

Sorapipatana, C. and S. Yoosin, 2011. Life cycle cost of ethanol production from

cassava in Thailand. Renewable and Sustainable Energy Reviews 15: 1343–1349.

Sree, N. K., M. Sridhar, K. Suresh and L. V. Rao, 1999. High alcohol production by

solid substrate fermentation from starchy substrates using thermotolerant

Saccharomyces cerevisiae. Bioprocess Engineering 20: 561-563.

Srinivas, T. and M. Anantharuman, 2000. Status of cassava production, processing

and marketing in Andra Pradesh. Central Tuber Crops Research Institute, Kerada,

India.

Sriroth, K., B. Lamchaiyaphum, K. Piyachomkwan, 2006. Present situation and

future potential of cassava in Thailand. Cassava and Starch Technology Research

Unit, Available online <www.cassava.org/doc/presentsituation2.pdf> (date accessed

8th January 2011).

References

119

Sriroth, K., K. Piyachomkwan, V. Santisopasri and C. G. Oates, 2001. Environmental

conditions during root development: Drought constraint on cassava starch quality.

Euphytica 120: 95-101.

Sriroth, K., V. Santisopasri, C. Petchalnuwat, K. Kurotjanawong, K. Piyachomkwan

and C. G. Oates, 1999. Cassava Starch granule structure-function properties: influence

of time and conditions at harvest on four cultivars of cassava starch. Carbohydrate

Polymers 38: 161-170.

Stephenson, G., 1973. Small Gasoline Engines. Delmar Publishers, New York, USA.

Stone, R., 1999. Introduction to Internal Combustion Engines, 3rd Ed. Palgrave, New

York, USA.

Thomas, D. J. and W. A. Atwell, 1999. Starches: Practical guides for the food

industry, Eagan Press, Minnesota, USA.

The Royal Society, 2008. Sustainable Biofuels: Prospects and Challenges. Policy

document 01/08. Available online <royalsociety.org> (date accessed: 9th July, 2012).

Topgül, T., H. S. Yücesu, C. Çinar and A. Koca, 2006. The effects of ethanol-gasoline

blends and ignition timing on engine performance and exhaust emissions. Renewable

Energy 31: 2534-2542.

Ülgen, K. Ö., B. Saygili, Z. Önsan and B. Kirdar, 2002. Bioconversion of starch into

ethanol by a recombinant Saccharomyces cerevisiae strain YPG-AB. Process

Biochemistry 37: 1157-1168.

Van Oirschot, Q. E. A., G. M. O’Brian, D. Dufour, M. A. El-Sharkawy and E. Mesa,

2000. The effect of pre-harvest pruning of cassava upon root deterioration and quality

characteristics. Journal of the Science of Food and Agriculture 80: 1866-1873.

References

120

Verma, G., P. Nigam, D. Singh and K. Chaudhary, 2000. Bioconversion of starch to

ethanol in a single-step process by coculture of amylolytic yeasts and Saccharomyces

cerevisiae 21. Bioresource Technology 72: 261-266.

Wallin, M., R. Westerholm, K. E. Egebäck, B. Rehnlund and M. Henke, 2005.

Blending of ethanol in gasoline for spark ignition engines-Problem Inventory and

Evaporative Measurements. Report No. MTC 5407, AVL MTC Tech Centre,

Haninge, Sweden.

Wang, N. S., nd. Starch Hydrolysis by Amylase. Available online

<http://www.eng.umd.edu/~nsw/ench485/lab5.htm> (date accessed: 2nd October,

2010).

Wang, W., 2002. Cassava production for industrial utilization in China-Present and

future perspectives. In: Howeler, R. H. (Ed.) Cassava Research and Development in

Asia: Exploring new opportunities for an ancient crop, Proceedings of Seventh

Regional Workshop held in Bangkok, Thailand, 28th October-1st November, 2002.

Wholey, D. W. and R. H. Booth, 1979. Influence of variety and planting density of

starch accumulation in cassava roots. Journal of the Science of Food and Agriculture

30: 165-170.

Willett, J. and M. Doane, 2002. Effect of moisture content on tensile properties of

starch/poly(hydroxyester ether) composite materials. Polymer 43: 4413-4420.

World Watch Institute (WWI), 2007. Biofuels for transportation: Global potential

and implications for sustainable agriculture and energy in the 21st century. Available

online <www.worldwatch.org/node/4078> (date accesses 26th November, 2011).

Wu, C-W., R-H. Chen, J-Y. Pu and T-H. Lin, 2004. The influence of air-fuel ratio on

engine performance and pollutant emission of an SI engine using ethanol-gasoline-

blended fuels. Atmospheric Environment 38: 7093-7100.

References

121

Yücesu, H. S., T. Topgül, C. Çinar and M. Okur, 2006. Effects of ethanol-gasoline

blends on engine performance and exhaust emissions in different compression ratios.

Applied Thermal Engineering 26: 2272-2278.

Yüksel, F. and B. Yüksel, 2004. The use of ethanol-gasoline blends as a fuel in an SI

engine. Renewable. Energy 29: 1181-1191.

Zervas, E. and M. Tazerout, 2000. Organic acids emissions from natural-gas-fed

engines. Atmospheric Environment 34: 3921-3929.

Zervas, E., X. Montagne and J. Lahaye, 2003. Emissions of regulated pollutants from

a spark ignition engine. Influence of fuel and air/fuel equivalence ratio.

Environmental science technology 37: 3232-3238.

2005. The Internal Combustion Engine. Available online <http://www.hybrid-

vehicle.org/hybrid-vehicle-ice.html> (date accessed 8th April, 2012)

<http://www.enertechlabs.com/fuel_phase_separation_in_ethanol.php> (date accessed

16th November, 2011)

<http://www.hybrid-vehicle.org date accessed> (date accessed 6th June, 2011)

122

Appendix A Engine Efficiency Data

Table A: Engine Efficiency Data for Petrol, E10, E15 and E20

Fuel

Sample

Power

Output

(W)

Vol. Fuel

Used (ml)

Time

(s)

Calorific

Value

(kJ ml-1)

Power

Input

(W)

Overall

Efficiency

(%)

Engine

Efficiency

(%)

Petrol

0.0 120.0 905.0 33.6 4457.9 0.0 0.0

365.5 120.0 682.5 5911.2 6.2 8.7

714.0 120.0 551.5 7315.3 9.8 13.7

1217.5 120.0 449.0 8985.3 13.5 19.0

1567.0 120.0 401.0 10060.8 15.6 21.8

1863.5 120.0 364.0 11083.5 16.8 23.5

E10

0.0 120.0 845.5 32.3 4582.9 0.0 0.0

365.5 120.0 709.0 5465.2 6.7 9.4

714.0 120.0 568.5 6815.8 10.5 14.7

1217.5 120.0 462.5 8377.9 14.5 20.3

1567.0 120.0 427.5 9063.9 17.3 24.2

1863.5 120.0 392.0 9884.7 18.9 26.4

E15

0.0 120.0 832.5 31.2 4493.0 0.0 0.0

365.5 120.0 687.5 5440.6 6.7 9.4

714.0 120.0 547.0 6838.0 10.4 14.6

1217.5 120.0 431.0 8678.4 14.0 19.6

1567.0 120.0 406.0 9212.8 17.0 23.8

1863.5 120.0 376.5 9934.7 18.8 26.3

E20

0.0 120.0 806.0 30.8 4578.2 0.0 0.0

365.5 120.0 665.5 5544.7 6.6 9.2

714.0 120.0 534.0 6910.1 10.3 14.5

1217.5 120.0 415.5 8880.9 13.7 19.2

1567.0 120.0 382.0 9659.7 16.2 22.7

1863.5 120.0 357.0 10336.1 18.0 25.2

123

Appendix B Fuel and Specific Fuel Consumption Data

Table B: Fuel Consumption and Specific Fuel Consumption Data for Petrol, E10,

E15, E20

Fuel

Sample

Power

Output

(kW)

Vol.

Fuel

Used

(ml)

Fuel

Consumption

(L hr-1)

Density

(g cm-3)

Time

(hr)

Mass of

fuel

consumed

(kg)

Fuel rate

(kg hr-1)

Specific fuel

consumption

(kg kW-1 hr-1)

Petrol 0 120 0.477 0.741 0.251 0.089 0.354

0.371 120 0.633 0.190 0.089 0.469 1.264

0.723 120 0.783 0.153 0.089 0.580 0.803

1.232 120 0.962 0.125 0.089 0.713 0.579

1.593 120 1.077 0.111 0.089 0.798 0.501

1.904 120 1.187 0.101 0.089 0.879 0.462

E10 0 120 0.511 0.746 0.235 0.090 0.381

0.366 120 0.609 0.197 0.090 0.455 1.242

0.714 120 0.760 0.158 0.090 0.567 0.794

1.222 120 0.934 0.128 0.090 0.697 0.570

1.576 120 1.011 0.119 0.090 0.754 0.478

1.859 120 1.111 0.108 0.090 0.828 0.446

E15 0 120 0.519 0.749 0.231 0.090 0.389

0.363 120 0.628 0.191 0.090 0.471 1.297

0.709 120 0.790 0.152 0.090 0.592 0.834

1.209 120 1.002 0.120 0.090 0.751 0.621

1.55 120 1.064 0.113 0.090 0.797 0.514

1.85 120 1.147 0.105 0.090 0.859 0.465

E20

0 120 0.536 0.752 0.224 0.090 0.403

0.362 120 0.649 0.185 0.090 0.488 1.348

0.71 120 0.809 0.148 0.090 0.608 0.857

1.207 120 1.040 0.115 0.090 0.782 0.648

1.549 120 1.131 0.106 0.090 0.850 0.549

1.841 120 1.210 0.099 0.090 0.910 0.494

124

Appendix C Emission Data

Table C: Emission Data for Petrol, E10, E15, E20

Fuel Sample Load (W) CO (% vol) HC (ppm) CO2 (% vol)

Petrol

0 2.97 70.67 3.7

365.5 2.65 66 3.85

714 2.56 64 4.06

1217.5 2.4 62.67 4.86

1567 2.35 60.67 5.71

1863.5 2.1 58 6.88

E10

0 2.19 43.33 3.31

365.5 2.09 51 3.58

714 1.98 60 3.73

1217.5 1.82 56 4.56

1567 1.78 54 5.32

1863.5 1.38 52 6.39

E15

0 1.63 34 3.16

365.5 1.32 38.67 3.55

714 1.29 43.33 3.64

1217.5 1.2 42.67 4.45

1567 1.06 42.67 5.05

1863.5 0.82 40.67 5.72

E20

0 1.29 23.33 3

365.5 1.16 28 3.47

714 1.03 40.67 3.61

1217.5 1.02 40.67 4.28

1567 0.74 40.67 4.85

1863.5 0.47 38 5.53

125

Appendix D

One paper has been presented in conjunction with the above work at the International

Conference on Technology Transfer and Renewable Energy, which was held on 21st

to 22nd June, 2012 in Mauritius. The proceedings will be published in a book

“Technological Approaches in Renewable Energy- An Overview in Small Island

States and Beyond”.

� Bijay, P. and Singh, A. 2012. Viability of using cassava as feedstock for

bioethanol production in Fiji. In: Technological Approaches in Renewable

Energy- An Overview in Small Island States and Beyond, Proceedings of the

International Conference on Technology Transfer and Renewable Energy,

Mauritius, 21st-22nd June, 2012. (Accepted for Publication).

Abstract: Ethanol production from renewable resources has received attention

due to increasing petroleum shortage. One such renewable resource that has

been identified is cassava starch. Cassava starch is extracted from root crop,

cassava (Manihot esculenta (Crantz)) and is readily available in Fiji. Many

countries such as China, Thailand and Philippines are already having success

in producing high starch yielding cassava varieties that can be used for ethanol

production.

The current paper investigates the viability of producing ethanol from locally

available cassava varieties in Fiji. Starch was extracted from the roots of ten

different cassava varieties available at two different research stations in Fiji.

The sedimentation technique was used to extract starch from cassava roots and

some properties of the extracted starch were also determined. In the case of

Koronivia Research Station (KRS) the variety Nadelei had the highest starch

yield (23.1 %) whereas Coci had the highest starch yield (23.3 %) for

Dobuilevu Research Station (DRS). The paper discusses and compares starch

yield obtained from Fiji cassava varieties with some other countries and make

recommendations on how starch yield from Fiji cassava varieties can be

increased.

Appendix D

126

Finally, the paper provides recommendations on enhancing the viability of

cassava as a source for bioethanol production in Fiji. It also assesses the

resources available in Fiji currently to make cassava bioethanol in Fiji a viable

proposition.

ethanol production from selecteddigilib.library.usp.ac.fj/gsdl/collect/usplibr1/index/... ·...

Documents