FACULTY OF SCIENCES

DEPARTMENT OF PHYSICS

CONTRIBUTION TO THE GEOTHERMAL ENERGY STUDY

IN RWANDA

Dissertation presented for the award

of a Bachelor of Sciences (BSc).

By: NDASHIMYE Maurice

Supervisor : Prof. MAREMBO KAREMERA Claver

Huye, March 2007

 i

DEDICATION

To my mother,
I dedicate this work.
 ii

ACKNOWLEDGEMENT

Many people were involved with this study, We are grateful for the effort they made,
many helped us, many taught us or inspired us. We can’t name all to whom a debt of
gratitude is owed, but We will try.

First of all, let us express our deepest gratitude to our project supervisor
Pr Marembo Karemera who tirelessly went the extramile to improve the quality of this
work.
We would like to express our heart-felt gratitude to the late Hitimana Anastase
our beloved father who initiated our first steps to school and encouraged us to learn science
subjects from our tender age, even though he is physically gone, without his inspiration,
We would never have managed to realize this dream.

Furthermore, We express our sincere gratitude to our teachers from primary school
up to the National University of Rwanda for their intellectual guidance during our studies.
Particularly, We would like to express our special thanks to Pr Marembo Karemera,
Dr Safari Bonfils, Dr Ndahayo Fidele and Mr Mageza Celestin who made this all happen.

We’re highly grateful to the National University of Rwanda for providing financial
support to carry out this study, our thanks goes to the staff of the chemistry departement for
their help in our laboratory work, great appreciation to Mr Namugize J. Nepo, who spent so
many of his days guiding us in our laboratory work.

We also express our gratitude to our fellow students with who We share our student
life, for their cooperation and moral support. We appreciate the warmth and friendship that
existed between us and people We interacted with during our study at the National
University of Rwanda.

Finally, We’re especially indebted to our mother and our sisters who care and gave
this work a lot of guidance and encouragement.
 iii

TABLE OF CONTENTS

DEDICATION..................................................................................................i
AKNOWLEDGEMENT..................................................................................ii
TABLE OF CONTENTS................................................................................iii
TABLE OF FIGURES....................................................................................vi
LIST OF TABLES..........................................................................................vi
SYMBOLS AND ABBREVIATIONS..........................................................vii
NOMENCLATURE FOR LANT FLOW DIAGRAMS................................vii
SOMMAIRE.................................................................................................viii
ABSTRACT....................................................................................................ix

CHAPTER I: INTRODUCTION....................................................................1
I.1. STUDY INTEREST...............................................................................................................................1
I.2. PROBLEMATICS...................................................................................................................................1
I.3. HYPOTHESIS.........................................................................................................................................1
I.4. OBJECTIVES.........................................................................................................................................1
I.5. METHODOLOGY..................................................................................................................................2
I.5.1. Documentary research..........................................................................................2
I.5.2. Samples and datas collection...............................................................................2
I.5.3. Research in laboratory.........................................................................................2
I.5.4. Learning about GIS technology...........................................................................2
I.6. WORK SUBDIVISION..........................................................................................................................2

CHAPTER II. GEOTHERMAL OVERVIEW................................................3

II.1. GEOLOGICAL BACKGROUND.........................................................................................................3
II.1.1. The Earth’s structure...........................................................................................3
II.1.2. The plate tectonic theory....................................................................................4
II.1.3. The Earth’s heat..................................................................................................6
II.1.3.1. Heat transfer within the Earth.......................................................................7
II.1.3.2. The Earth’s geothermal gradient and the thermal conductivity of rocks......9
 iv

II.2. GEOTHERMAL RESOURCES..........................................................................................................10

II.2.1. General aspects.................................................................................................10
II.2.2. Hydrothermal systems......................................................................................11
II.2.3. Water-dominated fields....................................................................................12
II.2.4. Wet steam fields...............................................................................................13
II.2.5. Vapour-dominated fields..................................................................................13
II.2.6.Hot Dry Rock systems.......................................................................................14
II.2.7. Geopressured reservoirs....................................................................................15
II.3. UTILISATION OF GEOTHERMAL RESOURCES..........................................................................16
II.3.1. Electricity from geothermal fluids....................................................................16
II.3.2. Direct uses of geothermal energy.....................................................................17

CHAPTER III: GEOTHERMAL PROSPECTS AND ELECTRICITY

GENERATION..................................................................18
III.1. GEOTHERMAL RESERVOIR EXPLORATION.............................................................................18
III.1.1. Inventory and survey of surface manifestations..............................................18
III.1.2. Geological and hydrogeological surveys........................................................19
III.1.3. Geochemical surveys.......................................................................................19
III.1.4. Geothermometers............................................................................................20
III.1.5. Geophysical surveys........................................................................................21
III.1.5.1. Seismic surveys........................................................................................21
III.1.5.2. Gravity surveys.........................................................................................21
III.1.5.3. Magnetic surveys......................................................................................21
III.1.5.4. Electrical-resistivity surveys....................................................................22
III.1.5.5. Electromagnetic surveys...........................................................................22
III.1.5.6. Thermal-measurement surveys.................................................................23
III.1.6. Exploratory wells............................................................................................24
III.2. GEOTHERMAL ELECTRICITY GENERATION...........................................................................24
III.2.1. Thermodynamic background...........................................................................24
III.2.1.1. Thermodynamic laws...............................................................................24
III.2.1.2.Thermodynamic cycles..............................................................................29
A. The Carnot cycle.............................................................................................29
B. The Rankine cycle..........................................................................................31
C. The Kalina cycle.............................................................................................33
 v

III.2.2. Geothermal electricity generating cycles........................................................34

III.2.2.1.Direct intake cycles...................................................................................34
III.2.2.2.Flash steam cycles.....................................................................................35
III.2.2.3. Binary cycle..............................................................................................36
III.2.2.4.Combined or hybrid plants........................................................................37
III.2.2.5. Efficiency of generation...........................................................................38

CHAPTER IV: RWANDA GEOTHERMAL SYSTEM...............................39

IV.1. GEOTHERMAL IN AFRICA............................................................................................................39
IV.2. RWANDA GEOTHERMAL RESOURCES LOCATION................................................................40
IV.2.1. Southern region...............................................................................................40
IV.2.2. Western region................................................................................................40
IV.2.3. Northern region...............................................................................................40
IV.3. GEOTHERMAL FLUIDS ORIGIN...................................................................................................42
IV.3.1. Deep waters origin.........................................................................................42
IV.3.2. CO2 origin........................................................................................................42
IV.4.CALCULATION OF DEEP TEMPERATURES...............................................................................42
IV.4.1. Introduction on geothermometers...................................................................42
IV.4.1.1. Geothermometer with Silica...................................................................42
IV.4.1.2. Geothermometers with Na/K...................................................................43
IV.4.1.3. Geothermometer with Na/Li....................................................................44
IV.4.1.4. Adjustment considering Mg.....................................................................45
IV.4.2. Analysis of natural waters with Atomic Absorption Spectroscopy................45
IV.4.3. Deep temperatures estimated with geothermometers.....................................45
IV.4.3.1 Mashyuza prospects..................................................................................46
IV.4.3.2 Ruganji prospects.....................................................................................46
IV.4.3.3. Ntaresi prospect.......................................................................................47
IV.4.3.4. Amakera and Gihugu prospects...............................................................47
IV.4.3.5. Gishyita geothermal prospect..................................................................47
IV.5. CONCLUSION..................................................................................................................................47
IV.6. RECOMMENDATIONS...................................................................................................................48

REFERERENCES..........................................................................................49

APPENDIX....................................................................................................50
 vi

TABLE OF FIGURES

Figure II.1: The earth structure..............................................................................................3

Figure II.2: The basic concept of plate tectonics...................................................................5
Figure II.3: World pattern of plates.......................................................................................6
Figure II.4: Heat conduction.................................................................................................8
Figure II.5: An hydrothermal field......................................................................................11
Figure II.6: Schematic representation of a hot dry rock......................................................15
Figure III.1: A reversible cyclic process.............................................................................26
Figure III.2: Irreversible process.........................................................................................27
Figure III.3: T-S diagram for a Carnot heat engine.............................................................29
Figure III.4: T-S diagram for a simple Rankine cycle........................................................31
Figure III.5: Simplified Kalina cycle..................................................................................33
Figure III.6: Simplified flow diagram for a direct-steam geothermal power plant.............35
Figure III.7: Simplified flow diagram for a single-flash geothermal power plant..............36
Figure III.8: Simplified flow diagram for a basic binary geothermal power plant.............37
Figure III.9: Simplified flow diagram for a Kalina binary geothermal power plant...........37
Figure IV.1: East African rift system..................................................................................39
Figure IV.2: Rwanda Geothermal resources visited during our study................................41
Abacus I: Estimates of DMg...............................................................................................55

LIST OF TABLES

Table I : GPS data...............................................................................................................51

Table II : Chemical analyses results (NUR 2007)...............................................................51
Table III : Chemical analyses results (Chevron 2006)........................................................52
Table IV : Chemical analyses results (BRGM 1982)..........................................................52
Table V : Chemical geothermometry results ( NUR 2007).................................................53
Table VI : Chemical geothermometry results (Chevron 2006)...........................................53
Table VII : Chemical geothermometry results (BRGM 1982)............................................54
Table VIII: Estimated installation costs for a 10 MWe geothermal development..............54
Table IX : Estimated installation costs for a 50 MWe geothermal development................55
Standard atomic absorption conditions for Mg, Na, K and Ca......................................56-59
 vii

SYMBOLS AND ABBREVIATIONS

HDR: Hot Dry Rock Mg: Magnesium

HWR: Hot Wet Rock Na: Sodium
GPS: Global Positioning System Ca: Calcium
GIS: Geographic Information System K: Potassium
GEA: Geothermal Energy Association MT: Magnetotelluric
DC: Direct Current

NOMENCLATURE FOR PLANT FLOW DIAGRAMS

BCV: Ball Check Valve M: Make-up water

C: Condenser P: Well Pump
CP: Condensate Pump PH: Preheater
CS: Cyclone Separator S: Silencer
CSV: Control and Stop Valves SH: Superheater
CT: Cooling Tower SP: Steam piping
CW: Cooling Water SR: Sand Remover
CWP: Cooling Water Pump T/G: Turbine/Generator
F: Flasher Th: Thottle valve
FF: Final Filter WP: Water Piping
IP: Injection Pump WH: Welhead valves
IW: Injection Wells
 viii

SOMMAIRE

Le but principal de cet etude est de prouver qu’il est possible de produire de
l’électricite à partir des resources géothermales du Rwanda.
Pour réaliser notre travail, il était requis de se baser sur plusieurs paramètres de ces
resources, à savoir la temperature et le volume du réservoir geothermal, la composition
chimique du fluide geothermal, la pression de sortie du fluide geothermal, la géologie des
régions abritant ces ressources.
Etant donné que toutes ces données n’étaient pas disponibles, nous avons procédé à
un prélevement d’échantillons sur huit sources chaudes qui s’avèrent susceptibles de
produire de l’électricite, ces sources sont: Mashyuza I et II, Ntaresi, Gishyita, Cyabararika,
Gihugu et Ruganji I et II.
Une localisation de ces sources par GPS a été faite ainsi qu’une analyse au
laboratoire de ces échantillons par géothermometres chimiques, afin d’éstimer les
temperatures en profondeur de ces sources.
Après ces analyses, nous avons constaté que seules les sources de Mashyuza, Ntaresi et
Ruganji sont susceptibles de produire de l’électricite en utilisant la technologie binaire.
 ix

ABSTRACT

With this study, we intend to proof that it is possible to produce electricity from
Rwanda geothermal resources.
In order to fulfill this aim, several parameters on these resources are required, among
them we can say deep temperatures and volume of the geothermal reservoirs, the geothermal
fluid chemical composition, geology of the reservoirs, the geothermal fluid up flow
pressure.
Given that all of these datas were not available, samples were gathered from eight
hot springs judged to have a high potential. These springs are Mashyuza I and II, Ntaresi,
Cyabararika, Gihugu, Gishyita and Ruganji I and II.
A GPS location of these spring has been done and geothermometric analyses of the
gathered samples has been done in order to estimate deep temperatures of these springs.
From these analyses, we ended up with a conclusion that only Mashyuza, Ntaresi
and Ruganji can generate electricity by the use of binary technology.
 1

Chapter I: INTRODUCTION

I.1. STUDY INTEREST

Our study has been chosen in order to show that geothermal energy can be one of the
solutions to the energy crisis in our country.
In fact we have chosen geothermal energy because it is:
• Present in our country in a considerable quantity.
• Renewable
• Stable because it doesn’t depend on weathers or times .

I.2. PROBLEMATICS
Actually, Rwandan population is estimated at 8.2 million and the annual growth rate
at 2.5%. As consequence, we have an increasing energy consumption and if we consider the
economic activities development in our country, we realize that energy is really a big problem
for our country.
The current status is given by:
• A weak electrification (4% of the population).
• Most of energy comes from firewood which is dangerous for the ecosystem.
• A huge potential in solar energy, organic-gas and bio-gas but not used efficiently.
• Other forms of energy like Aeolian, geothermal… not exploited at all.

I.3. HYPOTHESIS
Can geothermal energy be economically exploited in Rwanda?

I.4. OBJECTIVES
Our study has three objectives:
• Location of all of the geothermal resources of Rwanda and establishment and
mapping the latter using GIS technology.
• Calculation of deep temperatures using chemical geothermometers
• Estimate the geothermal potential in Rwanda using calculations and simulations.
 2

I.5. METHODOLOGY
I.5.1. Documentary research
In order to gather enough information on geothermal energy we have conducted
a documentary research in research centers, Infrastructure ministry and in different libraries.
We have also visited several websites.

I.5.2. Samples and data collection

We went in different regions where geothermal resources are located in order to collect
GPS data on these resources and take samples for laboratory analyses.

I.5.3. Research in laboratory

Our research has been conducted in laboratory whereby we intended to measure
different chemical elements concentrations from our samples.

I.5.4. Learning about GIS technology

We went in a GIS centre and learnt about this technology in order to establish a map of
Rwanda geothermal system.

I.6. WORK SUBDIVISION

This study is subdivided into four chapters:
I. INTRODUCTION
II. GEOTHERMAL OVERVIEW
III. GEOTHERMAL PROSPECTS AND ELECTRICITY GENERATION
IV. RWANDA GEOTHERMAL SYSTEM
 3

Chapter II. GEOTHERMAL OVERVIEW

II.1. GEOLOGICAL BACKGROUND

II.1.1. The Earth’s structure
The Earth is formed by three concentric zones, crust, mantle and core (Fig. II.1).
The crust: The Earth’s crust is analogous to the skin of an apple. The thickness of the
crust (7 km on average under the ocean basins, 20–65 km under the continents) is insignificant
compared to the rest of the Earth which has an average radius of 6370 km.

Figure II.1: The Earth structure

Wells give us direct access only to the crust, and to depths not much beyond
10 km. Studies of seismic waves have shown that the Earth’s crust is thinner beneath the
oceans than beneath the continents, and that seismic waves travel faster in oceanic crust than in
continental crust. In part because of this difference in velocity, it is assumed that the two types
of crust are made up of different kinds of rock. The denser, oceanic crust is made of basalt,
whereas the continental crust is made of granite.
 4

The mantle: The mantle lies closer to the Earth’s surface beneath the ocean (at a depth
of 7 km), than it does beneath the continents (20–65 km). It extends from the base of the crust
for about 2900 km. The most accepted hypothesis about the composition of the mantle is that it
consists of ultrabasic rock (very rich in Fe and Mg) such as peridotite, which is a heavy
igneous rock made up chiefly of ferromagnesian minerals. The Earth’s crust and uppermost
mantle together form the lithosphere, the outer shell of the Earth that is relatively rigid and
brittle (Fig. I.1). The lithosphere is split into a number of large blocks at continental scale or
more, which are called lithospheric plates in the plate tectonic theory.
The lithosphere (crust and upper mantle) is about 70 km thick beneath the oceans and
100–125 km thick beneath the continents. Its lower boundary inside the mantle is marked by a
particular layer, known as the low-velocity zone, in which seismic waves slow down. This
zone, extending to a depth of perhaps 200 km, or more, from the surface, is called the
asthenosphere.
Rocks in the asthenosphere may be closer to their melting point than rocks above or
below this zone. Mantle rocks in the asthenosphere are weaker than they are in the overlying
lithosphere, then the asthenosphere can deform easily by plastic flow, and convection can take
place within the asthenosphere as well as within the lower mantle.
The lithosphere seems to be in continual movement, probably as a result of the
underlying mantle convection, and plates of brittle lithosphere probably move easily over the
asthenosphere, which may act as a lubricating layer below.
The core: The Earth’s core extends from 2900 to 6370 km (the Earth’s centre): its
thickness, or radius, is 3470 km. The temperature in the core should be around 4000°C and the
pressure at the Earth’s centre is estimated at 3.6 million bar (360,000 MPa).

II.1.2. The plate tectonic theory

The plate tectonic theory, is a unifying theory that accounts for many apparently
unrelated geological phenomena. According to this theory the rigid outer shell of the Earth, or
lithosphere is divided into separate blocks or plates, termed lithospheric plates (Fig. II.3).
These plates move slowly across the Earth’s surface, at a speed of a few centimeters per year.
As the plates comprise continents and sea floors, the plate tectonics concept means that the
continents and sea floors are moving, sliding on top of the underlying plastic asthenosphere.
 5

These plates either pull away from each other, slide past each other, or move towards
each other. The boundaries between plates are of three types (Figure II.2):
• Diverging plate boundaries (or spreading centres, or ocean ridges): These occur where two
plates are moving apart, thus permitting the upwelling of magma from the asthenosphere to
form new lithosphere. Most spreading centres coincide with the crest of submarine
mountain ranges, called mid-oceanic ridges which rarely rise above sea level . (e.g.
Iceland)
• Converging plate boundaries: These correspond to oceanic trenches, where two plates
converge and collide so that one plate slips and sinks below the other and is eventually
reabsorbed into the mantle and “destroyed” (for example, the Nazca plate in the eastern
Pacific Ocean). Convergence occurs when one plate is made of oceanic crust and the other
of continental crust. The less dense, more buoyant, continental plate will override the
denser oceanic plate. The oceanic plate sinks into the mantle beneath an overriding plate, it
becomes hotter as it descends deeper into the Earth’s interior and melts down.
• Conservative plate boundaries. These are faults where two plates slide past each other, so
that no lithosphere is either created or destroyed. In this case, the direction of relative
motion of the two plates is parallel to the fault. Conservative plate boundaries occur within
both the oceanic and continental lithosphere.

Figure II.2: The basic concept of plate tectonics.

 6

The different types of plate boundaries were originally distinguished on the basis of
their seismicity. Earthquakes commonly occur at the boundaries of plates, and only
occasionally in the middle of a plate. There is a close correspondence between plate boundaries
and earthquake belts. We have already seen that plate tectonic processes are linked to mass
movement in the mantle. Because such movement can only occur within materials that have
the properties of fluids, this implies that the mantle shows some form of fluid behaviour.

Figure II.3: World pattern of plates. (Arrows show the direction of movement of the plates)

II.1.3. The Earth’s heat

The philosopher and mathematician Pitagoras (580-500 B.C.) declared that there is a
great fire in the Earth’s interior, and that this fire is manifest at the surface in the form of
volcanoes.
The first systematic measurements of temperature underground are accredited to the
British chemist Robert Boyle, well-known for his gas law. In 1671 Boyle wrote of the heat,
sometimes very strong, noted in the mines of Britain, and stated that temperature increases
with depth.
 7

This phenomenon was reappraised in 1846 by young William Thomson, later Lord
Kelvin, one of the fathers of thermodynamics, who, in his PhD thesis at the University of
Glasgow, tried to estimate the age of the Earth from the “distribution and movement of heat
within it”.
However, it was not until 1882 that the values obtained for the geothermal gradient and
the thermal conductivity of rocks were combined by Lord Kelvin (their product gives the heat
flow).
The Earth’s heat flow is the amount of heat that is released into space from the interior
through a unit area in a unit of time. It is measured in milliwatt per square meter. It varies from
place to place on the surface, and it has varied with time at any particular place during the
history of our planet.
The Earth’s heat flow originates from the primordial heat, which is the heat generated
during the Earth’s formation, and from the heat generated since the Earth’s formation by the
decay of long-lived radioactive isotopes.
Although all radioactive isotopes generate heat as they decay, only isotopes that are
relatively abundant and have half-lives comparable to the age of the Earth (4.5 billion years)
have been significant heat producers throughout geological time and remain so at present. Four
long-lived radioactive isotopes are important heat producers:40K,232Th,235U and 238U.
The average heat flow from the continental crust (granite) is 57 mW/m 2, and through
the oceanic crust (basalt) is 99 mW/m2. The Earth’s average heat flow is 82 mW/m2, and the
total global output is over 4x1013W, four times more than the present world energy
consumption which is 1013W. Continental heat flow appears to be derived from radiogenic
decay within the upper crust, together with the heat generated in the most recent magmatic
episode and the heat coming from the mantle. In the oceanic crust, the concentration of
radioactive isotopes is so low that radiogenic heating is negligible, and the heat flow largely
derives from heat flowing from the mantle below the lithosphere.
 8

II.1.3.1. Heat transfer within the Earth

Two forms of heat transfer occur within the Earth: conduction and convection.
Conduction is a heat transfer process caused by a difference in temperatures between two areas
of a same medium or between two media in contact but without displacement of matter. This
process involves a transfer of random kinetic energy between molecules without the overall
transfer of material. Moving molecules strike neighbouring molecules, causing them to vibrate
faster and thus transfer heat energy. Conduction is the primary heat transfer mode in solids.
Metals are very good conductors of heat, whereas most rocks are relatively poor conductors.
This transfer of heat spontaneous from an area of high temperature towards an area of
lower temperature obeys the Fourier’s law : “The flux density of heat is proportional to the
variation in temperature.”
Let’s consider a quantity Q of heat s transferred from a point x with a temperature T to
a point x+dx of the same matter with a temperature T+dT

Figure II.4: Heat conduction

If we take dSx as a transversal isothermal surface, in the direction (0, X), we can
express the variation of the heat according to X:
δT
dQ x = −k x dS x dt (II.1)
δx
The thermal flux variation is then:

dQx δ T
dΦ = = − k x dS x (II.2)
dt δx
The flux density is then:
dΦ δT
ϕ= = −k (II.3)
dS x δx
 9

k is the thermal conductivity constant, it depends on the nature of the material and its
temperature.
Convection is the common heat transfer process in liquids or gases and consists of the
movement of hot fluid (that is, a liquid or a gas) from one place to another. Because motion of
material occurs, convection is a vastly more efficient process of heat transfer than conduction.
In fact if heat is generated in a layer of a liquid, it spreads in all the liquid by
conduction but if the variation in temperature exceeds a limit called “the adiabatic gradient”,
the liquid becomes unstable and the hot liquid goes up on the surface where it cools and returns
in the depths, this movement is called thermal convection. The adiabatic gradient is given by
the formula:
T
γ = gα (II.4)
c
Where α is the coefficient of expansion of the liquid, g the constant of gravity, T the
temperature and c the specific heat. The coefficient of expansion of a rock is lower in the
mantle than under the atmospheric pressure by experiments under high pressure, we estimate it
at 10-6/oC with this value we obtain γ=1 oC/Km.

II.1.3.2. The Earth’s geothermal gradient and the thermal conductivity of rocks
Studies of the thermal behaviour of the Earth imply the determination of how
temperature varies with depth, and how such temperature variations may have changed
throughout geological time. However, studies of this kind are based entirely on measurements
made on, or within, a few km of the Earth’s surface.
The average gradient near the surface, say within a few km, is about 30°C/km,but
values as low as about 10°C/km are found in ancient continental crust and very high values
(100°C/km) are found in areas of active volcanism. Once the gradient has been measured, it
can be used to determine the rate at which heat is moving upwards through a particular part of
the Earth’s crust.
As the heat generally moves upwards through solid impermeable rock, the principal
mechanism of heat transfer must be conduction. The amount of heat flowing by conduction
through a unit area of 1m2 of solid rock in a given time, that is the rate of heat flow, is
proportional to the geothermal gradient and to a constant of proportionality which is known as
the thermal conductivity of rocks defined as the amount of heat conducted per second through
 10

an area of 1m2, when the temperature gradient is 1°C/m perpendicular to that area. The unit of
thermal conductivity is the W/m K.
The gradient is measured in wells with electrical (platinum-resistance) thermometers.
Temperature logging is quick and relatively inexpensive.
The thermal conductivity of rock samples is best measured in laboratory, as there are no
reliable downhole methods. If the gradient is expressed in °C/km and conductivity in W/(m°C),
heat flow will be in mW/m2 (milliwatt per square metre).

II.2. GEOTHERMAL RESOURCES

II.2.1. General aspects
Geothermal resources are the thermal energy that could reasonably be extracted at costs
competitive with other forms of energy at some specified future time.
Geothermal resources are generally confined to areas of the Earth’s crust where heat
flow higher than in surrounding areas heats the water contained in permeable rocks (reservoirs)
at depth. The resources with the highest energy potential are mainly concentrated on the
boundaries between plates, where visible geothermal activity frequently exists. By geothermal
activity we mean hot springs, fumaroles, steam vents, and geysers. Active volcanoes are also a
kind of geothermal activity, on a particularly and more spectacular large scale.
Geothermal activity in an area is certainly the first significant indication that subsurface
rocks in the area are warmer than the norm. The local heat source could be a magma body at
600–1000°C, intruded within a few kilometers of the surface. However, geothermal fields can
also form in regions unaffected by recent shallow magmatic intrusions. The anomalous higher
heat flow may be due to particular tectonic situations, for example to thinning of the
continental crust, which implies the upwelling of the crust-mantle boundary and consequently
higher temperatures at shallower depths.
However, we need more than a thermal anomaly to have a productive geothermal
resource. We also need a reservoir, which is a sufficiently large body of permeable rocks at a
depth accessible by drilling. This body of rock must contain large amounts of fluids, water or
steam, which carry the heat to the surface.
The reservoir is bounded by cooler rocks hydraulically connected to the hot reservoir
by fractures and fissures, which provide channels for rainwater to penetrate underground.
 11

These cooler rocks come out at the surface where they represent the so-called recharge areas of
the geothermal reservoir.

Fig. II.5. An hydrothermal field

Water moves inside the reservoir by convection, due to density variations caused by
temperature, transferring heat from the lowest parts of the reservoir to its upper parts. The
result of the convection process is that temperature in the upper parts of the reservoir is not
much lower that the temperature of its deeper parts, so that the lowest values of the geothermal
gradient are actually found inside the reservoir.
Convection, implying a real transfer of matter, is therefore a more efficient process of
heat transfer than conduction. Heat is transferred by conduction from the magma body towards
the permeable reservoir rocks, the reservoir, filled with fluids. Hot fluids often escape from the
reservoir and reach the surface, producing the visible geothermal activity described above.

II.2.2. Hydrothermal systems

The heat source, the reservoir, the recharge area and the connecting paths through
which cool superficial water penetrates the reservoir and, in most cases, escapes back to the
surface, compose the hydrothermal system.
The reservoir is the most important part of the system, from the point of view of energy
utilisation, and in fact we define the reservoir “as the hot part of the geothermal system that can
 12

be exploited either by extracting the fluid contained (water, steam, or various gases), or using
anyhow its heat”.
The existence of a hydrothermal system will not necessarily ensure production at
industrial levels, Only a part of its rocks may be permeable, constituting a fluid reservoir, so
that the system will be able to produce industrially from that part only.
Four types of geothermal fluid reservoir have been identified:
• Hydrothermal
• hot dry rock
• geopressured
• magmatic
The reservoirs that can be exploited at present are the hydrothermal reservoirs. The
other three may be exploited industrially in the future after more technological development.
We have two kinds of hydrothermal reservoirs :
• water-dominated
• vapour-dominated
The latter having a higher energy content per unit fluid mass.
Water-dominated fields are further divided into hot water fields, producing hot water, and wet
steam fields producing mixtures of water and steam.

II.2.3. Water-dominated fields

They are capable of producing hot water at the surface at temperatures up to 100°C.
They are the geothermal fields with the lowest temperature, and the reservoir contains water in
liquid phase. The reservoir may not have a cover of impermeable rock acting as a lid, however
some of these thermal aquifers are overlain by confining layers that keep the hot water under
pressure. Temperatures in the reservoir remain below the boiling point of water at any pressure
because the heat source is not large enough. Surface temperature is not higher than boiling
temperature of water at atmospheric pressure.
These fields may also occur in areas with normal heat flow. On the surface there are
often thermal springs whose temperatures are, in some cases, near the boiling point of water. A
hot water field is of economic interest if the reservoir is found at a depth of less than 2 km, if
the salt content of the water is lower than 60 g/kg, and if the wells have high flow-rates (above
150 t/h).
 13

II.2.4. Wet steam fields

They contain pressurised water at temperatures exceeding 100°C and small quantities
of steam in the shallower, lower pressure parts of the reservoir. The dominant phase in the
reservoir is the liquid one, and it is this phase that controls the pressure inside the reservoir.
Steam is not uniformly present, occurring in the form of bubbles surrounded by liquid water,
and does not noticeably affect fluid pressure.
An impermeable cap-rock generally exists to prevent the fluid from escaping to the
surface, thus keeping it under pressure. This is common, but not absolutely necessary. In
fact, at any depth below the water table, water bears its own hydrostatic pressure. When the
fluid is brought to the surface and its pressure decreases, a fraction of fluid is flashed into
steam, while the greater part remains as boiling water. Once a well penetrates a reservoir of
this type, the pressurised water rises into the well because pressure is lower there. The
consequence of the pressure drop is the vaporisation of part of the water, with the result that
the well eventually produces hot water and steam, with water as the predominant phase. The
water-steam ratio varies from field to field, and even from one well to the next within the same
field. As in many cases only steam is used to produce electrical energy, liquid water must be
removed at the surface in special separators.
The surface manifestations of these fields include boiling springs and geysers. The heat
source is large and generally of magmatic origin. The water produced often contains large
quantities of chemicals (from 1 to over 100 g/kg of fluid, in some fields up to 350 g/kg). These
chemicals may cause severe scaling problems to pipelines and plants. They are mainly
chlorides, bicarbonates, sulfates, borates, fluorides and silica. More than 90% of the
hydrothermal reservoirs exploited on an industrial scale are of the wet steam type. Electricity
generation is their optimal utilisation.
One typical exemple of wet steam fields producing electricity is Olkaria in Kenya.

II.2.5. Vapour-dominated fields

Vapour-dominated reservoirs (fields) produce dry saturated, or slightly superheated
steam at pressures above atmospheric. They are geologically similar to wet steam fields, but
the heat transfer from depth is certainly much higher. Research suggests that their permeability
 14

is lower than in wet steam fields, and the presence of the caprock is of fundamental importance
here.
Water and steam co-exist, but steam is the continuous predominant phase, regulating
the pressure in the reservoir: the pressure is practically constant throughout the reservoir. These
fields are called dry or superheated fields. Produced steam is in fact generally superheated,
with small quantities of other gases, mainly CO2 and H2S.
The mechanism governing production in these fields is believed to be the following:
When a well penetrates the reservoir and production begins, a depressurized zone forms at
well-bottom.
This pressure drop produces boiling and vaporisation of the liquid water in the
surrounding rock mass. A dry area, i.e. without liquid water, forms near the well-bottom and
steam flows through this zone. Steam crossing the dry area starts to expand and cool, but the
addition of heat from the very hot surrounding rocks keeps steam temperature above the
vaporisation value for the pressure existing at that point. As a result, the well produces
superheated steam, for example with wellhead pressures of 5-10 bar (0.5–1 MPa) and a steam
outlet temperature of more than 200°C. Surface geothermal activity associated with vapour-
dominated fields, whether dry or superheated, is similar to the activity present in wet steam
fields.
About half of the geothermal electric energy generated in the world comes from six
vapour-dominated fields: Larderello (Italy) , Mt. Amiata, (Italy), The Geysers (California),
Matsukawa (Japan), Kamojang and Darajat (Indonesia) Of the approximately 100
hydrothermal systems that have been investigated, less than 10% are vapour-dominated, 60%
are wet steam fields (water-dominated), and 30% produce hot water.

II.2.6.Hot Dry Rock systems

Hot Dry Rock (HDR) geothermal reservoirs differ significantly from conventional
geothermal reservoirs, which probably exist only in the geologically favoured regions of the
world shown. In these regions, nature provides not only the hot rock, but also the hot water or
steam. HDR reservoirs are, instead, man-made reservoirs in rocks that are artificially fractured,
and thus any convenient volume of hot dry rock in the earth’s crust, at accessible depth, can
become an artificial reservoir. A pair of wells is drilled into the rock, terminating several
hundred meters apart.
 15

Water is circulated down the injection well and through the HDR reservoir, which acts
as a heat exchanger. The fluid then returns to the surface through the production well, and thus
transfers the heat to the surface as steam or hot water.
In the other hand we have Hot Wet Rock systems (HWR) as enhancement of low-
permeability fracture systems in high temperature rocks exploiting the techniques developed in
the HDR experiments.
The most common stimulation techniques are hydraulic fracturing, chemical fracturing
and explosive fracturing.

Fig. II.6. Schematic representation of a hot dry rock (HDR) reservoir.

II.2.7. Geopressured reservoirs

Geopressured reservoirs are deep reservoirs (4–6 km) in large sedimentary basins
containing pressurised hot water that remained trapped at the time of deposition of the
sediment, and at pressures of up to 100% in excess of the hydrostatic pressure corresponding to
that depth, these fields could produce not only the thermal energy of the pressurized hot water,
but also hydraulic energy, by virtue of the very high pressure, and methane gas.
These three forms of energy can also be converted to higher value forms of energy
using available technologies.
Thermal energy can be converted to electricity in a geothermal turbine. Hydraulic
energy can be converted to electricity using a hydraulic turbine.
 16

Dissolved methane gas can be separated and sold, burned, compressed, liquefied,
converted to methanol, or converted to electricity by fuelling a turbine.

II.3. UTILISATION OF GEOTHERMAL RESOURCES

Geothermal utilisation is divided into two categories, electric energy production and
direct uses. Conventional electric power production is limited to fluid temperatures above
150°C, but considerably lower temperatures can be used in binary cycle systems.

II.3.1. Electricity from geothermal fluids

According to the GEA report, the world geothermal electrical capacity installed in the
year 2006 was 8900MWe with the generation in that year of 56.5billion kWh.
The total electricity produced world-wide from all sources in that year was 14,855.8
billion kWh of which 3,282.5 billion kWh were generated by renewable sources (2,939.3
billions by hydropower alone, and 343.3 billions altogether by biomass, geothermal, wind,
solar and tidal). These values show that renewables with the exception of hydro play a very
minor role on the world energy scene.
The contribution of geothermal energy is entirely different if we distinguish between
industrialised and developing countries.
In the industrialised countries, where the installed electrical capacity are very high (tens
or even hundreds of thousands GWe), geothermal energy represents one percent, at most, of
the total.
In developing countries, with a limited electrical consumption but good geothermal
prospects, electrical energy of geothermal origin could, on the contrary, make quite a
significant contribution to the total: at the moment, for instance, 21% of the electricity in the
Philippines comes from geothermal sources, 20% in El Salvador, 17% in Nicaragua, 10% in
Costa Rica and 8% in Kenya.
 17

II.3.2. Direct uses of geothermal energy

The utilisation of natural steam for electricity generation is not the only possible
application of geothermal energy. Hot waters, that appear to be present in large parts of all the
continents can also be exploited and offer interesting prospects, especially in space heating,
agriculture and industrial processes.According to the GEA, the thermal power installed for
non-electrical uses of geothermal energy worldwide in the year 2005 has been estimated at
16,000 MWt, distributed among 72 countries.
 18

Chapter III: GEOTHERMAL PROSPECTS AND ELECTRICITY

GENERATION
III.1. GEOTHERMAL RESERVOIR EXPLORATION
Present technology and economic factors restrict extraction of geothermal energy to the
upper few kilometres of the Earth’s crust. Geothermal wells, to date, are drilled to less than 5
km depth.
As in the search for any natural resource, a strategy for geothermal energy exploration
must be defined and followed. Once a geothermal region has been identified, the next step is to
use various exploration techniques to locate the most interesting geothermal areas and identify
suitable targets for fluid production.
It is necessary to estimate temperature, reservoir volume and permeability at depth, as
well as to predict whether wells will produce steam or just hot water. Ideally weshould also
estimate the chemical composition of the fluid to be produced. To obtain this varied
information, there are a number of exploration techniques available:
• Inventory and survey of surface manifestations,
• Geological and hydrogeological surveys,
• Geochemical surveys,
• Geophysical surveys, and finally
• Exploratory wells.
To reduce the cost of exploration, it is normally approached in a prescribed sequence of
steps, altering the order from time to time depending on our prior knowledge of the area in
question. In some cases, high costs will lead to the elimination of some steps in the sequence.

III.1.1. Inventory and survey of surface manifestations

The knowledge of surface thermal manifestations (hot springs, steam vents, fumaroles,
etc.) and their physical and chemical characteristics is of fundamental importance.
This information, which can usually be obtained simply and at relatively low cost, is
extremely useful for subsequent planning of exploration.
 19

The surface survey is conducted in two consecutive phases:

• The collation, processing and standardisation of published and recorded data relative to
local manifestations (chemistry, temperature, flow-rates, etc.),
• The collection of new data, water samples, gas samples, temperature measurements, etc.

III.1.2. Geological and hydrogeological surveys

These are not limited to studies of groundwaters, but also include geological surveys
that provide information on the stratigraphic and structural framework of the area. Geothermal
reservoirs are often associated with volcanoes and volcanic regions and therefore volcanology
also offers many examples of how geological field data give evidence of the location, nature
and size of a geothermal resource.
An hydrogeological surveys permit us to correlate the hydrothermal manifestations
with faults, fractures and other tectonic features. These surveys are aimed at identifying the
distribution of confined and unconfined aquifers that will permit us to reconstruct the
underground pattern of water circulation.
Fluid inclusions may give information on the temperature of deposition of inclusions
and therefore determine the temperature and salinity of geothermal fluids. Fluid inclusions are
defects in crystals, formed during or after deposition. All crystals have inclusions. Some
inclusions are solid, others empty, a few contain fluids. Inclusions need to be multi-phase
(liquid and vapour) to be most useful.

III.1.3. Geochemical surveys

Geochemical exploration can start simultaneously with geologic and hydrogeologic
reconnaissance, provided that springs and other geothermal manifestations are available for
fluid sampling. Geochemical studies of geothermal fluids involve three main steps:
• Sample collection,
• Chemical analysis
• Data interpretation.
The types of samples collected are water samples from hot springs, steam samples from
fumaroles, gas samples from hot pools.
Geothermometers enable to estimate the temperature of deep reservoirs by analyzing
hot fluids samples and calculating the ratios of certain chemical elements.
 20

14
The content of tritium and C radioisotopes permit us to evaluate the age of the
geothermal fluids, that is, the time lapsed since their infiltration into the ground.
Geochemical surveys with the use of tracers can also offer information on the direction of
movement of subsurface groundwater and of reinjected fluids, and also the type of corrosion
and scaling problems that could be encountered during the operation of wells and a plant.
Hydrogen and oxygen isotopes can be used to identify the recharge areas of the
geothermal reservoirs, in fact the isotopic composition of a rain precipitation is dependent on
its formation temperature .

III.1.4. Geothermometers
Geothermometers enable to estimate the temperature of deep reservoirs by analyzing
hot fluids samples and calculating the ratios of certain chemical elements (i.e., Na, K, Mg, Ca,
etc.), and making some adjustment for the degree of mixing of the hot geothermal reservoir
water with cooler groundwater in the shallow part of the hydrothermal system.
In fact, at particular temperatures, common assemblages of minerals will tend towards
equilibrium with a given water chemistry. It has been noted that for certain parameters or ratios
of parameters, the relationship between temperature and chemical composition will be stable
and predictable.
These parameters or ratios of parameters are known as geothermometers. In order for
these to work, we have to assume that effects of dilution are insignificant and that
thermodynamic equilibrium has been attained.
In general geothermometers based on ratios will be more resistant to dilution effects
than those based on absolute concentrations.
In addition, it should be realised that the temperature indicated by the geothermometer
is not necessarily the maximum temperature of the water, but the temperature at which mineral
and water phases were last in equilibrium with respect to the phases in question.
The three principal indicators of deep reservoir temperatures, to be sought in hot spring
chemistry, are silica, magnesium and sodium/potassium ratios.
Silica concentrations are more reliable for hot springs of high discharge than those of
low discharge. Magnesium is of limited value as a temperature indicator, but its total absence
could be suggestive of economically useful temperatures (at least 200°C), as magnesium is
retained in clay rocks which are stable at high temperatures.
 21

III.1.5. Geophysical surveys

Classical geophysical techniques, namely, seismic, gravity and magnetic surveys as
applied to geothermal research, can be defined as indirect methods. These methods, in fact, are
not directly associated with the properties of the hot fluids that are being sought. Rather, they
yield information about the attitude and nature of the host rocks. However, there are other
geophysical methods that may directly reveal variations in the physical properties of the rocks
caused by the presence of hot and saline fluids. These techniques include electrical-resistivity,
electromagnetic and thermal-measurement methods.

III.1.5.1. Seismic surveys

Elastic waves are transmitted through rocks, and their velocities can be used to help
determine the structure and properties of rock bodies. Seismic waves are introduced into the
Earth by detonating an explosive charge in a shallow borehole or by using a large mass to
thump the surface. Returns of seismic waves are measured at the surface.
Seismic waves also originate naturally from earthquakes and microearthquakes, and
these waves can also be detected at the surface. Interpretation of the seismic information can
provide data on the location of active faults that can channel hot fluids towards the surface.

III.1.5.2. Gravity surveys

Variations in the Earth’s gravity field are caused by changes in the density of
subsurface rocks. Gravity surveys are rather simple and inexpensive. Gravity anomalies alone
are not necessarily indicative of a geothermal region, but they do give valuable information on
the type of rocks at depth and their distribution and geometric characteristics.

III.1.5.3. Magnetic surveys

The Earth has a primary magnetic field which induces a magnetic response in certain
minerals at and near the Earth’s surface. By detecting spatial changes of the magnetic field, the
variations in distribution of magnetic minerals may be deduced and related to geologic
structure.
 22

However, each magnetic mineral has a Curie temperature, above which it loses its
magnetic properties. For iron, the Curie temperature is 760°C.
Aeromagnetic surveys are much more commonly used in geothermal exploration than
ground based surveys. The basic principle is to detect zones which are magnetically
featureless, due to destruction of magnetite in near-surface rocks by hydrothermal alteration.

III.1.5.4. Electrical-resistivity surveys

Most electrical methods are based on measurement of the electrical resistivity of the
subsurface. Resistivity in the Earth is often largely affected by electrical conduction within
waters occupying the pore spaces in the rock. Consequently, resistivity varies considerably
with porosity. Temperature and salinity of interstitial fluids tend to be higher in geothermal
reservoirs than in the surrounding rocks. Consequently, the resistivity of geothermal reservoirs
is generally relatively low. It is this contrast in resistivity between hot water-saturated rocks
and the surrounding colder rocks that is used in resistivity surveys.
These techniques are based on injection of current into the ground and measurement of
voltage differences produced as a consequence at the ground surface.
One of the major drawbacks with electrical methods is the shallow depth of penetration.

III.1.5.5. Electromagnetic surveys

Induction or electromagnetic methods are a tool for determining the electrical resistivity
distribution in the Earth by means of surface measurements of transient electric and magnetic
fields. These fields can be naturally or artificially generated.
These methods are more suitable for measuring the low resistivities of geothermal
reservoirs than the above-mentioned electrical-resistivity methods. Furthermore, in geothermal
areas the surface resistivity is sometimes so high as to prevent current from entering the
ground, and the electromagnetic methods, with a much deeper penetration, help eliminate the
screening effect of very resistive surface rocks.
Currents of varying frequency (generally from a few to several tens of thousands of Hz)
are transmitted into the ground, either via the electrodes as in the electrical methods, or by
induction loops. Mobile stations measure, at several points, the electrical and magnetic fields
created by this transmission.
 23

Comparison between these fields enable the resistivities of the underlying formations to
be obtained, as a function of the frequency used, that is as a function of the depth, as in the
magnetotelluric soundings (MT).
Magnetotelluric soundings use natural oscillations of the Earth’s electromagnetic field
to determine the resistivity structure of the sub-surface. The electromagnetic waves are
assumed to be planar and nearly vertically incident on the surface of the Earth where they are
detected. The lower the frequency, the deeper the penetration is taken to be, but the longer it
takes to collect a signal with a satisfactory signal to noise ratio. The depth of penetration of MT
surveys is also much greater than of during current (DC) resistivity measurements, and it is
often possible to achieve a penetration as great as 3–5 km with a reasonable degree of precision
and in a reasonable period of time, this can be compared to a maximum depth of penetration
from most DC methods of less than 2 km. MT does not require a current source.

III.1.5.6. Thermal-measurement surveys

In geothermal research the traditional geophysical methods mentioned above, which
originally had been developed for the oil industry, are used side-by-side with more specific
techniques.
Geothermal prospecting provides information on the thermal conditions of the
subsurface, the area distribution of the Earth’s heat flow, and the location and intensity of
thermal anomalies. To be more specific, geothermal prospecting allows us:
• To verify the existence of high-temperature fluids in areas without surface manifestations,
but in which the geostructural and hydrogeological situation is favourable to hydrothermal
circulation;
• To more precisely site deep drilling in areas that are considered potentially productive;
• To delineate the boundaries of geothermal fields that have been identified, so as to avoid
drilling of dry holes in non-productive marginal areas;
• To acquire data for evaluation of the geothermal potential of the field.
Heat flow measurements are made by drilling small diameter (10 cm), shallow wells
(generally <300 m), the number of which depends on local conditions and on the results one
wants to achieve. The geothermal gradient is obtained from temperatures measured with
electric thermometers at various depths along a well. Temperature logging is quick and
relatively inexpensive.
 24

Thermal conductivity of the rocks in the interval in which the gradient has been
measured is usually determined by laboratory measurements on core samples. The product of
the gradient and conductivity gives the heat flow.
Sometimes gradient values alone are sufficient to give the information required.
However, this is possible only if the survey is carried out in areas that are lithologically
homogeneous at depth, in which the thermal conductivity can be considered constant.
In geothermal areas the heat flow is higher than the general background level, so that high heat
flow values are a good indicator of underlying geothermal resources.

III.1.6. Exploratory wells

The final stage of an exploration survey is exploratory well drilling. Usually the final
diameters of these wells are on the order of 20 cm or less, allowing the insertion of special
logging tools to measure various parameters from the surface to total depth, and sometimes to
carry out fluid production tests.
Since most geothermal reservoirs are made up of fluid-filled fractures, it is essential
that an exploratory well intersects as many fractures as possible. In some cases it may be
necessary to redrill the well at an angle in order to intersect the natural fracture pattern. Since
natural fractures are related to tectonic activity (folding and faulting), the siting of exploratory
wells is greatly dependent on our geologic interpretation of the local structural conditions.

III.2. GEOTHERMAL ELECTRICITY GENERATION

III.2.1. Thermodynamic background
III.2.1.1. Thermodynamic laws
The first law states that energy cannot be crated or destroyed, it can only change forms.
This law cannot be proved mathematically, it is based on experimental observations and
nothing has yet been found in nature that violates this law. The law can be written for a closed
system in an equation form in the following manner:
E s − E r = Ei (V.1)
Where Es is the energy supplied to the system, Er energy removed from the system and
Ei increase in the energy level of the system.
For a constant mass system, the only forms of energy that can be supplied or removed
from a system are heat energy and work energy.
 25

Adopting a sign convention such as heat energy entering a system is positive and the
work energy leaving the system is also positive. We can now write the first principle for a
constant mass system as:
Q − W = ∆E = E 2 − E1 (V.2)
Where Q is the energy of the system before the transformation, W the work done by the
system and the energy variation.
The enthalpy is a thermodynamic property of substance and is defined as the sum of its
internal energy and the product of its pressure and volume. The enthalpy is given by:
H = U + pV (V.3)
For a body undergoing a constant pressure process, its work is given by

W = ∫ pdV = p(V2 −V1 )

2

1
(V.4)

Application of the first law gives

Q − p(V2 − V1 ) = U 2 − U 1 (V.5)
Thus
Q = (U 2 − U 1 ) + p (V2 − V1 ) = (U 2 + pV2 ) − (U 1 + pV1 ) = H 2 − H 1
Then we have:
Q = ∆H
Thus, for a constant pressure process, the amount of heat supplied to a body equals the
change in its enthalpy.
The first law of thermodynamics is the principle of conservation of energy. It forbids
creation or destruction of energy, but permits transformation of one form of energy into
another, it doesn’t put any restriction on the direction in which the transformation of energy
may occur, that is why we need another law which governs the direction of transformations.
The second law of thermodynamics specifies in what direction a process may proceed.
There are two classical statements called second law of thermodynamics. One is due to
Clausius and the other one to Kelvin.
• The Clausius statement: It is impossible to construct a device that executes a
thermodynamic cycle so that the sole effect is to produce a transfer of heat energy from a
body at a law temperature to a body at a high temperature.
 26

• The Kelvin-Planck statement: It is impossible to construct a device that executes a

thermodynamic cycle, exchanges heat energy with a single reservoir, and produces an
equivalent amount of work.
There are two important corollaries of the second law.
1. Any heat engine operating between two constant temperature reservoirs cannot have an
efficiency that is greater than the efficiency of a reversible engine operating between the
same two reservoirs.
2. All reversible heat engines operating between two constant temperature reservoirs have the
same efficiency regardless of the operating media used by these engines. The efficiency is
solely a function of temperature of the reservoirs.
Let’s consider a system undergoing a reversible cyclic process, path 1-2-3-4-1 and
interacting with a number of energy reservoirs, Inasmuch as the cyclic process is reversible, the
system and the reservoirs have essentially identical temperatures at every point in the cycle.

Figure III.1: A reversible cyclic process

If we consider the Clausius statement i.e. that heat transfer always takes place from
high temperatures to low temperatures we can introduce the Clausius inequality which states
that the algebraic sum of the ratios of heat energy extracted to temperature for each of the
various reservoirs exchanging heat energy is less than or equal to zero.
We can then write the Clausius inequality for this system as:
δQ
∫ 1−2 −3−4 −1
T
≤0 (V.9)
 27

Now let the system reverse the cycle, that is, follow the path 1-4-3-2-1, for this path we can
write:
δQ
∫ 1−4 −3−2 −1
T
≤0 (V.10)

The only way to satisfy both these equations is to have

δQ δQ δQ
∫ 1−2 −3−4 −1
T
= ∫ 1−2 −3
T
+ ∫ 3−4 −1
T
=0 (V.12)

δQ δQ δQ δQ
Then: ∫ 1−2 −3
T
− ∫ 1−4−3
T
=0 or ∫ 1−2 −3
T
= ∫ 1−4 −3
T
(V.13)

This result means that if we have a reversible process taking a system from point 1 to
point 3, then the integral of the quantity, δQ/T, is the same regardless of the path chosen from
point 1 to point 3. If a function when integrated, has the same value regardless of the path of
integration so long as the two end points are the same, this function is a point function, the
integrand in such case is an exact differential. We represent the quantity δQ/T by an exact
differential dS of a thermodynamic property, called entropy S. Which is a function of the
thermodynamic state of a system and it is defined only for equilibrium states. Only changes in
entropy can be calculated. Then we have:

2 δQ
 δ Q S 2 − S1 = ∫
dS =   and 1 T (V.14)
 T  reversible reversible

Now suppose that the system is brought back from state 3 to 1 by some irreversible
process. This is represented by dotted line 3-α-1 in figure III.2.
 28

Figure III.2: Reversible and Irreversible processes

The Clausius inequality for the irreversible cycle 1-2-3- α-1 becomes
δQ δQ δQ
∫ 1−2 −3−α−1
Treeservoir
<0 Or ∫ 1−2 −3
T
+ ∫ 3−α−1
Treservoir
<0 (V.17)

If we subtract the equation (V.15) from the above inequality we obtain:

δQ δQ
∫ 3 −α−1
Treservoir
− ∫ 3−4−1
T
<0 (V.18)

The second integral in the above inequality is for reversible process 3-4-1 and therefore
can be replaced by ∆S or S1-S3 note that the starting point is 3 and the end point 1. Thus we
obtain the inequality :
δQ δQ
∫ 3−α −1
Treservoir
− ( S1 − S 3 ) < 0 or ( S1 − S 3 ) > ∫ 3−α−1
Treservoir
(V.19)

This result means that in any irreversible process, the change in entropy is always
greater than the integral of the quantity δQ/T .
δQ
Thus for an irreversible process we have dS > (V.20)
T
For a system undergoing an irreversible process from state 1 to state 2,

2 δQ
S 2 − S1 > ∫
1 T (V.21)
irreversible

If we consider an isolated system i.e. which doesn’t interact with its surrounding in any
manner; that is, δQ=0.
Combining these results we can write, for an isolated system:

( S 2 − S1 ) ≥ 0 or ( ∆ S) isystsolaetmed ≥ 0 (V.22)

Which is another expression of the second law of thermodynamics which states that:
The entropy of an isolated system not at equilibrium will tend to increase over time,
approaching a maximum value.
A mathematical statement of the Second Law is:
 29

dS
≥ 0 (V.23)
dt
Where S is the entropy and t the time.

III.2.1.2.Thermodynamic cycles
A thermodynamic cycle is a combination of processes taking a system through a
succession of states and returning the system to its initial state.

A. The Carnot cycle

The carnot cycle was historically the first to be used for thermodynamic analysis. It is a
cycle that exchanges heat energy with only two reservoirs. It receives heat energy from a high-
temperature reservoir and rejects heat energy to a low temperature reservoir in a reversible
quasistatic manner. We are going to discuss the Carnot cycle operated as a heat engine but it
can also be reversed and function as a refrigerator or as a heat pump.
The carnot cycle consists of two reversible constant temperature processes and two
reversible adiabatic, or isentropic, processes. The T-s diagram for the Carnot cycle is shown in
figure III.2.
Consider a piston cylinder system containing a gas for execution of the Carnot heat
engine. The thermodynamic state of this gas is represented by point 1 on the T-s diagram. Let
an energy reservoir at temperature T h supply heat to the gas reversibly so that the gas has now
a new state, state 2. Then allow the gas to undergo an isentropic expansion (Q=0) to state 3.
Now let the gas reject heat to another reservoir at temperature T l in a reversible constant
temperature process until state 4 is reached . Finally, we return the gas to its initial state 1 by
subjecting the gas to isentropic compression (Q=0).
 30

Figure III.3: T-s diagram for a Carnot heat engine

The thermal efficiency of power cycle is the ratio of the net work of the cycle to the
heat energy supplied to the cycle. In view of the first law, the net work output of a cycle equals
the net heat input to the cycle.

Wcycle = Qin −Qout

Where Qin and Qout are the magnitudes of the heat energy supplied to and rejected by the cycle,
respectively.

 δ Q
Equation dS =   is written on a unit of mass as δq = Tds
 T  reversible

And considering the figure III.2 we have:

Qin = Area 1 − 2 − B − A = Th ( S 2 − S1 ) (V.24)
Qout = Area 3 − B − A − 3 = Tl ( S 3 − S 4 ) (V.25)
Since process 2-3 and 4-1 are isentropic, we can write S3 = S2 and S4 = S1 so that:
Qout = Tl ( S2 – S1 ) (V.26)
 31

The net work of the cycle is given by

Wcycle = Qin − Qout

= Th ( S 2 − S1 ) − Tl ( S 2 − S1 )

= ( Th − Tl )( S 2 − S1 ) (V.27)
The thermal efficiency of the cycle, ηt, is
net work output Wcycle ( Th − Tl )( S 2 − S1 )
ηt = = = (V.28)
gross heat sup ply Qin Th ( S 2 − S1 )
Or
Th − Tl T
ηcarnot = =1− l (V.29)
Th Th

According to the second law there cannot be a device which converts thermal energy into work
with 100% efficiency. For a carnot cycle, this means
Th − Tl T
ηcarnot = =1− l <1 (V.30)
Th Th

Or Tlow > 0 (V.31)

In other words it is impossible to have an energy reservoir with its temperature equal to
absolute zero. The unattainability of absolute zero is sometimes called the third law of
thermodynamics. Then, since the Carnot cycle is reversible, the two corollaries of the second
law stated earlier mean that it is impossible to have an engine operating between two energy
reservoirs whose thermal efficiency is greater than that of a carnot engine operating between
the same two reservoirs.

B. The Rankine cycle

 32

Figure III.4: T-s diagram for a simple Rankine cycle.

We begin the cycle with the saturated liquid state of the working medium, denoted by
point 1 in figure V.3. Process 1-2 is the isentropic compression of the working fluid from a
saturated liquid state to a compressed liquid state. This process is carried out in a feed water
pump. Usually, the temperature rise in this process is very small. Process 2-α-3 involves
constant pressure heating from a compressed liquid state to a saturated vapor state.This
operation takes place in a steam generator or in a boiler. In process 2- α the temperature
increases and the process requires heat transfer from a high temperature heat energy source,
while in the process α -3 there is no temperature change as the process involves a change of
phase. High-pressure saturated steam is admitted into the turbine at point 3, and it is expanded
isentropically, delivering work. Finally the wet steam at state 4 is condensed to a saturated
liquid state, state 1, by using a cooling agent as water.

In the analysis of the simple Rankine cycle, it is assumed that the changes in the kinetic
energy and the potential energy of the working medium are negligible. This assumption is quite
realistic. Another assumption made is that all processes are reversible, and that there are no
pressure drops or heat losses in the pipes. In reality there are irreversibilities and pressure
drops. For the reversible Rankine cycle, area 6-2-α-3-4-5-6 represents the heat energy supplied
to a unit mass of the working medium, and area 4-1-6-5-4 represents the heat energy removed
to the unit of mass.
Consequently, the difference in the two areas, namely area 1-2- α -3-4 represents the
net work obtained from the cycle.
Appling the laws of thermodynamics, for the various processes in the simple Rankine
cycle, we obtain the following for a unit mass flow rate:
 Process 1-2: Isentropic compression; hense
q1−2 = 0 and − ws ,1−2 = h2 − h1 (V.32)
 Process 2- α-3: Constant pressure heating and
ws , 2 −3 = 0 and q 2−3 = qin = h3 − h2 (V.33)
 Process 3-4: Isentropic expansion; hence
 33

q3−4 = 0 and ws , 3−4 = h3 − h4 (V.34)

 Process 4-1: Constant pressure heat removal and
ws , 4 −1 =0 and − q 4 −1 = h4 − h1 (V.35)

The net work of the cycle is

wcycle = ws ,3−4 + ws ,1−2 = ( h3 − h4 ) − ( h2 − h1 ) (V.36)
The thermal efficiency of the Rankine cycle is
wcycle ( h3 − h4 ) − ( h2 − h1 )
η Rankine = = (V.37)
qin h3 − h2

With the same upper and lower limits of the temperature for the Carnot and the Rankine
cycles, the Rankine cycle efficiency is less than the Carnot cycle efficiency. We may reason
that the average temperature at which heat is supplied in a Rankine cycle is less than the
temperature at which heat is supplied in a Carnot cycle, thus leading to the lower efficiency of
the Rankine cycle.

C. The Kalina cycle

The use of mixtures as working fluids has opened new possibilities to improve the
efficiency of power cycles with less costly equipment.
The Kalina cycle, which use a ammonia-water mixture, may show 10 to 20% higher
efficiency than the conventional Rankine cycle. The ammonia-water mixture boils at a variable
temperature unlike pure water which boils at a constant temperature.

Variable temperature boiling permits the working fluid to maintain a temperature closer
to that of the hot combustion gases in the boiler, thus, improving the exergy efficiency, a fact
which has been well known among scientist and engineers. But there was no practical, efficient
way to condense the mixture back to a fluid for recycling until the Kalina cycle was
introduced.
 34

Figure III.5. Simplified Kalina cycle

By circulating the mixture at different compositions in different parts of the cycle,

condensation (absorption) can be done at slightly above atmospheric pressure with a low
concentration of ammonia, while heat input is at a higher concentration for optimum cycle
performance.
Figure III.5 shows the simplified Kalina cycle assumed in this study. This is a
bottoming cycle feed by exhaust gases (1, 2) to the boiler. Superheated ammonia-water vapor
(3) is expanded in a turbine to generate work (4).
The turbine exhaust (5) is cooled (6, 7, 8), diluted with ammonia-poor liquid (9, 10) and
condensed (11) in the absorber by cooling water (12, 13). The saturated liquid leaving the
absorber is compressed (14) to an intermediate pressure and heated (15, 16, 17, 18). The
saturated mixture is separated into an ammonia-poor liquid (19) which is cooled (20, 21) and
depressurized in a throttle and ammonia-rich vapor (22) is cooled (23) and some of the original
condensate (24) is added to the nearly pure ammonia vapor to obtain an ammonia
concentration of about 70% in the working fluid (25). The mixture is then cooled (26),
condensed (27) by cooling water (28, 29), compressed (30), and sent to the boiler via
regenerative feedwater heater (31).
The mass flow circulating between the separator and the absorber is about 4 times that
of the turbine, thus, causing some additional condensate pump work. However, this loop makes
possible the changes in composition between initial condensation in the absorber and heat
addition in the boiler. By changing the dew point of the mixture, the waste heat from the
 35

turbine exhaust, which is lost in a Rankine cycle, can be used to dilute the ammoniawater
vapor with a stream of water, thus, producing a mixture with a substantially lower
concentration of ammonia which allows condensation at a much higher temperature.
Usually thermodynamic properties of pure fluids and information of the departure from
ideal-solution is sufficient to derive mixture properties. Stability, secondary reactions, safety,
etc must, of course, also be considered.
III.2.2. Geothermal electricity generating cycles
III.2.2.1.Direct intake cycles
The simplest and cheapest of the geothermal cycle used to generate electricity is the
direct-intake non-condensing cycle. Steam from the geothermal well is simply passed through
a turbine and exhausted to the atmosphere: there are no condensers at the outlet of the turbine
(Figure V.5). Such cycles consume about 15–25 kg of steam per kWh generated.
Non-condensing systems must be used if the content of noncondensible gases (CO2,
H2S, NH3, CH4, N2 and H2) in the steam is very high (greater than 50% in weight), and will
generally be used in preference to the condensing cycles for gas contents exceeding 15%,
because of the high power that would be required to extract these gases from the condenser.

Direct intake condensing plants, with condensers at the outlet of the turbine and
conventional cooling towers, show a much lower consumption, only 6–10 kg of steam per kWh
generated, but the gas content of the steam must be less than 15%. The specific consumption of
steam of these units is greatly influenced by the turbine inlet pressure.
For pressures ranging from 15 to 20 bar (1.5–2.0 MPa), the consumption is close to 6
kg/kWh. For pressures ranging from 5 to 15 bar (0.5–1.5 MPa) the consumption is from 9 to 7
kg/kWh, and it becomes much greater for even lower pressures.
 36

Fig….: Simplified flow diagram for a direct-steam geothermal power plant.

In power plants where electricity is produced from dry or superheated steam (vapour-
dominated reservoirs), steam is piped directly from the wells to the steam turbine. But vapour-
dominated systems are less common in the world, steam from these fields has the highest
enthalpy (energy content), generally close to 670 kcal/kg (2800 kJ/kg), at present these systems
have been found only in Indonesia, Italy, Japan and the USA. These fields produce about half
of the geothermal electrical energy of the world.
Flash steam plants are used to produce energy from these fields that are not hot enough
to flash a large proportion of the water to steam in surface equipment, either at one or two
pressure stages.

III.2.2.2.Flash steam cycles

Hydrothermal fluids above 150°C can be used in flash plants to make electricity.
Fluid is sprayed into a tank held at a much lower pressure than the fluid, causing some of the
fluid to rapidly vaporize, or "flash." The vapor then drives a turbine, which drives a generator.
If any liquid remains in the tank, it can be flashed again in a second tank to extract even more
energy.

Figure III.7: Simplified flow diagram for a single-flash geothermal power plant

III.2.2.3. Binary cycle

If the geothermal well produces hot water instead of steam, electricity can still be
generated, provided the water temperature is above 85°C, by means of binary cycle plants.
These plants operate with a secondary, low boiling-point working fluid (freon,
isobutane, ammonia, etc.) in a thermodynamic cycle. The working fluid is vaporised by the
 37

geothermal heat in the vaporiser. The vapour expands as it passes through the organic vapour
turbine, which is coupled to the generator. The exhaust vapour is subsequently condensed in a
watercooled condenser or air cooler and is recycled to the vaporiser by the motive fluid cycle
pump (Figure III.7) The efficiency of these cycles is even lower: between 2.8 and 12%.
However, the binary power plant technology has emerged as the most cost-effective
and reliable way to convert large amounts of low temperature geothermal resources into
electricity, and it is now well known that large low-temperature reservoirs exist at accessible
depths almost anywhere in the world.
The most used binary cycle in geothermal power generation is the Rankine cycle which
uses organic working fluid. The Kalina cycle is being developed as a competitive improvement
of the Rankine steam cycle. By using an ammonia-water mixture as the working fluid and a
condensing system based on absorption refrigeration principles.

Figure III.8 : Simplified flow diagram for a Rankine binary geothermal power plant

Figure III.9: Simplified diagram for a Kalina binary geothermal power plant
 38

III.2.2.4.Combined or hybrid plants

Since geothermal fluids are found with a wide range of physical and chemical
properties such as temperature, pression, noncondensable gases, scaling and corrosion
potential,…
A variety of energy conversion systems have been developed to suit any set of conditions. The
basic systems described in the earlier sections can be combined to achieve more effective
systems for particular applications. Thus, the following hybrid plants can be designed: Direct –
steam/Binary plant; Single flash/Binary plant; Integrated single and double flash plant.
Properly designed hybrid systems have a higher overall efficiency compared with using the
two systems.

III.2.2.5. Efficiency of generation

The efficiency of generation ηshows how well a plant converts the exergy of the
resource into useful output.
For a geothermal plant, it is found as follows: η=W/me
Where W is the net electricity delivered to the grid, m the required total geofluid mass flow
rate, and e is the specific energy of the geofluid under reservoir conditions. The latter is given
by: e=h(Pl-Tl)-h(Po-To)-To[s(Pl-Tl)-s(Po-To)]
The specific enthalpy h, entropy s, are evaluated at reservoir conditions, P l and Tl and at the so
called dead-state, Po and To which correspond to the local ambient conditions.
 39

Chapter IV: RWANDA GEOTHERMAL SYSTEM

IV.1. GEOTHERMAL IN AFRICA

The geothermal potential of Africa is mostly centered within the East African Rift,
which provides both the extensional rifting environment associated with forced convective
systems and the volcanic environment that favors the development of large, high temperature
geothermal systems. The East African Rift System is a major geologic structure that extends
from Mozambique north to the Red Sea rift (Figure IV.1). The rift consists of two main
braches.
• The eastern branch, which extends from Mozambique north through Tanzania into Kenya,
Ethiopia, and Djibouti, has the best geologic expression.
 40

• The western branch, which extends along the western borders of Tanzania, Burundi,
Rwanda and Uganda, is less developed and less active seismically and volcanically.
Active volcanoes are associated with the rift, two active volcanoes occur along the
Western Branch in Democratic Republic of the Congo, Nyiragongo and Nyamuragira. The
highest volcano, Mt. Kalisimbi, which is considered dormant, is in Rwanda.
These volcanoes are all part of the Virunga Volcanoes.
Despite the widely recognized geothermal potential of the East African Rift, the region
remains largely undeveloped. Except for one field in Kenya and another in Ethiopia,
geothermal exploration has not progressed beyond the reconnaissance exploration stage, even
though the geothermal potential of most East African countries was inventoried in the 1970’s
and 1980’s. To date, only one high temperature geothermal system associated with a volcanic
center has been developed. This system in Kenya, known as Olkaria and hosts a total of 129
MWe of installed generating capacity.

Figure IV.1: East African rift system

IV.2. RWANDA GEOTHERMAL RESOURCES LOCATION

Geothermal resources that we are going to discuss in this part are geothermal resources
with surface manifestations, generally we observe a temperature rise as we approach the axis of
the rift located at the west of the country and from an assessment made in 1982 we have the
following geothermal resources countrywide.

IV.2.1. Southern region

• Ntaresi geothermal resources : A few of hot spring emergences, in the left side of the
Lukarara valley(North of Nyamagabe). With a low emergence temperature, that is, around
20.9oC.
 41

• Akanyaru geothermal resources : Tepid water emerges with a low flow (≈ 1l/s) and some
gaseous emissions.

IV.2.2. Western region

• Mashyuza geothemal resources : Many hot springs flow up from the Mashyuza graben
bordering faults. Their emergence temperature is about 54.2oC.
• Gishyita geothermal resources :Located in the south-western part of Karongi, we have
an emergence of hot water at about 31.7 oC, with a little gaseous emission.
• Ruganji geothermal resources : We have two hot springs located at 500m from
BRALIRWA on the peninsula of Kivaga in the lake Kivu. They are the hottest sources of
the country 73.5oC, the liquid flow is about 1l/s for each source and there is a weak gaseous
.
IV.2.3. Northern region
Amakera geothermal resources : Located in the East of Musanze, this source is cold 18.7
o
C, presents a strong gaseous emission and deposits of Iron hydroxide.
Gihugu and Buseruka geothermal resources : Located in the west of Ntaruka electric power
plant. Characteristics similar to the source of Amakera.
Gitagata geothermal resources : Two emergences on left side bank of a small valley located
in the North-West of Base, the flow is of approximately 1l/s at a temperature of 37 oC.
 42

Figure IV.2: Rwanda Geothermal resources visited during our study

IV.3. GEOTHERMAL FLUIDS ORIGIN

IV.3.1. Deep waters origin
The water molecule is made up of two types of atoms, Hydrogen and Oxygen, these
atoms can have different masses, the ratio between Oxygen atoms of mass 18 and those of
mass 16 as well as the number of the Hydrogen atoms of mass 1 compared to those of mass 2
will allow to characterize and distinguish water from different origins.
An analysis made on geothermal waters of Rwanda showed that all of them have a meteoric
origin.
 43

IV.3.2. CO2 origin

The isotopic ratio of Carbon 13 allow to know the origin of CO 2 dissolved in water, the
atmospheric CO2 has a Carbon 13 ratio ranging between 0.6% and 0.8% .
An analysis made on Nyamyumba and Cyabararika gave a ratio of 0.47% and 0.2% for
Mashyuza. This means that CO2 from these sources has a deep origin.

IV.4.CALCULATION OF DEEP TEMPERATURES

IV.4.1. Introduction on geothermometers
Concentrations of certain chemical elements dissolved in geothermal waters are controlled by
their solubilities and the latter are function of the temperature.
The relation Solubility-temperature allowed the establishment of equations that make possible
the calculation of the temperature in geothermal reservoirs.
These equations are called geothermometers, some of the are based on absolute concentrations
and others on concentrations ratios.

IV.4.1.1. Geothermometer with Silica

Silica is one of the major chemical constituent of the Earth crust. The various allotropic
varieties like Quartz, Chalcedony and amorphous Silica can be dissolved in water, this reaction
of solubilization is mainly function of the temperature and doesn’t depend on the pH or the
ionic force of the solution.
In a laboratory analysis, the dissolved Silica is obtained by precipitation of an allotropic variety
of Silica, this latter depend on the temperature of the solubilization.

If the temperature is higher than 160oC the precipitated form is quartz.

Between 120 oC and160oC we have quartz and Chalcedony.
Under 120 oC, only Chalcedony.
Amorphous Silica is obtained for low temperatures.
Based on the solubility curves of these different allotropic varieties, the following equations
has been established:
 44

1309
For Quartz: T 0C = − 273.15
5.19 − log SiO2

1032
For Chalcedony: T 0C = − 273.15
4.69 − log SiO2

731
For amorphous Silica: T C = − 273.15
0

0.26 + log SiO2

Concentrations are expressed in mg/l

All that is valid only if the concentration of the dissolved Silica were not modified during the
transfer from the reservoir to the surface.
The modification can be consequence of:
A mixture with surface water.
Boiling or evaporation .
To avoid errors due to the modifications of concentrations, we use geothermometers based on
concentration ratios.

IV.4.1.2. Geothermometers with Na/K

From the reaction K Al Si3 O8 + Na + ⇔ Na Al Si3 O8
We have two non-miscible feldspars, moreover, this reaction depends only on the pressure and
the temperature but in our case the pressure influence is weak compared to that of the
temperature.
Thus the Na/K ratio depends only on the temperature, and the corresponding equation is:

933
T 0C = − 273.15
0.993 − log Na K

Concentrations in mg/l
Based on a concentrations ratio, the process of dilution particularly with surface waters
doesn’t affect this geothermometer.
The most serious disadvantage of this geothermometer is the supposition of the
equilibrium of the reaction that governs this geothermometer, it is difficult to know if this
equilibrium had been really attained.
Most of the time, the ration Na/K ranges between 3 and 30 thus, giving temperatures
between 150oC and 500oC.
 45

To overcome these difficulties, a geothermometer Na-K-Ca has been proposed as follows:

1647
T 0C = − 273.15
2.24 − log Na K + β log Ca Na

Concentrations in mol/l
We first try β=4/3, if we obtain a temperature higher than 100 oC, we change for β=1/3.
If not β=4/3 gives a good estimate. This geothermal is like a limitation, if we take β=4/3 and
we obtain a temperature lower than 100oC, better to reject the temperature calculated by Na/K,
but the reverse is not true.

IV.4.1.3. Geothermometer with Na/Li

This geothermometer is empirical, there is no scientific proof to its operation, it has been
established from a statistical study.
For waters with a moderate salinity (Cl<10g/l) we have the following equation:
1000
T 0C = − 273.15
0.38 − log Na Li

For brines (Cl>10g/l) :

1195
T 0C = − 273.15
log Na Li − 0.13

Concentrations are expressed in mol/l

This geothermometer has an advantage of being obeyed by cold surface water, this is not the
case for Silica nor Na/K geothermometers.

IV.4.1.4. Adjustment considering Mg

• If T(Ca,Na,K) < 70oC there is no adjustement.
• If this temperature exceeds 70oC, we calculate the expression:
Mg
R= x100
Mg + Ca + K

Concentrations expressed in equivalents/l

 46

• If R>50, We are in presence of a shallow acquifer with a temperature close to that of

the spring. In this case there is no need to consider any geothermometer.
• If T(Ca,Na,K)>70 and 1.5<R<5, we use the abacus I (Appendix) to estimate DMg,
which is the temperature correction to take from T(Ca,Na,K).

IV.4.2. Analysis of natural waters with Atomic Absorption Spectroscopy

This method describes the determination of Calcium,Magnesium, Potassium and
Sodium in natural waters and may be applicable to other elements.
Prepare all standard solutions except Ca and Mg by suitable dilutions of the stock
solutions described under the Standard Conditions for each element presented in the appendix.
For Ca and Mg, dilute the the stock solutions with 5% Lanthanum solution and HCl to give
dilute standards wich contain 0.25% La and 5% HCl.
Filter each sample through a 0.45 micron micropore membrane filter, if necessary to
avoid clogging of the burner capillary. Aspirate each sample directly, except for Ca and Mg.
For Ca and Mg you have to dilute with La and HCl. Read the concentration of the element of
interest directly against the appropriate standards given in the appendix.
Ca and Mg results should be corrected by using a reagent blank.

IV.4.3. Deep temperatures estimated with geothermometers

Hot springs had been sampled in 1982. The chemical analyses from these samples are
reported table IV in the appendix, because the sampling conditions and preservation techniques
were not well understood for the samples collected in 1982, in order to verify we visited eight
most interesting geothermal springs namely Mashyuza I and II, Ntaresi, Gishyita, Cyabararika,
Gihugu and Ruganji I and II, we took new samples from these hot springs for analyses.
(table II)
We also have located these hot springs with a GPS, geographic coordinates of these
sources (table I )and a map from these GPS data is provided as figure IV.2.
From our Atomic Absorption analyses (Table II) and geothermometric calculations
(table VI) we concluded that the AAS machine we used was not well calibrated, but
considering previous prospects we ended up with the following estimates for the deep
temperatures.
 47

IV.4.3.1 Mashyuza prospects

The calculated temperatures are very similar for the two points of emergence.
The cations geothermometers indicate a high temperatures (200-250 oC), let us note that the
temperature calculated by the Na/Li ratio geothermometer is 150oC.
This geothermometer often gives lower values than others and thus makes it possible to
fix the minimal value of the estimate.
The temperature calculated by the silica concentration geothermometer is low. Indeed,
this geothermometer is directly influenced by dilution with surface cold water. In spite of the
heterogeneity of the computation results, one can retain that the temperature is higher than
150oC, and could approach 200oC. We have also a Cesium concentration, which is in favour of
a high temperature of the reservoir.
We note that there exists on the south-eastern edge of the graben (in Burundi), a thermal
spring, with chemical characteristics very similar to that of Mashyuza. This shows the
homogeneity and the possible extension of the reservoir which feeds these sources and then,
the existence of a geothermal field in this region.

IV.4.3.2 Ruganji prospects

The two springs provide very similar results. The cations geothermometers give values
around 180oC and low values for Na/Li (approximately 60oC).
The temperature calculated with the silica concentration to in equilibrium with quartz is
of approximately 140oC. In addition, these sources have relatively high Cs and Li
concentrations compared to the other sources and a week Mg concentration.
We concluded that this water reaches a deep temperature of approximately 160oC.

IV.4.3.3. Ntaresi prospect

The results obtained with the various geothermometers are homogeneous, which
increases the reliability of the results. The Li content is higher, the content in As and Cs
relatively high and the Mg is present in small quantity. Count held of these various results, The
depth temperature is of approximately 180oC.

IV.4.3.4. Amakera and Gihugu prospects

 48

For these sources, the Cs, As and Li concentrations are very weak whereas the Mg has a
very high concentration. It appears that these sources are probably originating from a cold
surface reservoir.

IV.4.3.5. Gishyita geothermal prospect

The results from the cations geothermometers are heterogeneous, The Cs, As and Li
concentrations are relatively low as that of Mg. But with only one source we cannot have a
reliable estimate of depth temperature. However the results are not in favor of high
temperatures.

IV.5. CONCLUSION
From our study, we realized that Mashyuza, Ruganji and Ntaresi are geothermal fields
that can be economically exploited by the use of binary technology, estimates for 50MWe and
10MWe are reported in tables IX and X.
The most likely geologic model for Ruganji geothermal field is that the reservoir waters
are rising vertically along a normal fault near the eastern boundary of the East African Rift.
The fluids could be mixing along their flow path from the reservoir with low temperature
ground water or water from Lake Kivu. The reservoir waters may also be degassing during
their ascent as suggested by the relatively small amount of gas being vented from the hot
springs. An alternative and less likely model is that the hot springs represent a distal outflow of
fluids from a high temperature reservoir associated with the Virunga Volcanoes to the north.
For Mashyuza geothermal field, the most likely geologic model for this prospect is that
warm waters are rising vertically along the fault forming the western margin of the graben. The
upside potential is limited.
However, additional geologic mapping could locate extensions of the fault, geologist
from the cement factory noted that travertine deposits also occur in Burundi to the south and
that they may occur along the same geologic structure.
Tertiary volcanics, primarily basaltic lavas, are noted both south and west of the hot
spring area. Although no age dates are available for these lavas, they appear to be too old to
indicate that a viable magmatic heat source underlies the geothermal system.
For Ntaresi, there is no available geologic model, but all of the geothermometers
 49

showed that probably there is a geothermal field in that region further research have to be
conducted in order to locate and evaluate the potential of this field.

IV.6. RECOMMENDATIONS
Our recommendations for the three regions suspected to be geothermal fields are
provided as it follows:
Additionnal geothermometric analyses should be done in order to verify and complete
available datas on these fields, if possible, a complete set of gas samples should be obtained
from the hot springs. At the very least, samples for Helium isotope analysis. The He isotopes
should help reveal if the heat source contains a component of magmatic gas, or whether the
fluids are being conductively heated while circulating in Precambrian metamorphic rocks.
A detailed geologic map should be created through additional field work. The
distribution of hydrothermal alteration, hot spring deposits, and thermal springs should be
mapped within a five to ten kilometer radius of the hot springs. Aerial photography and remote
sensing images would assist with the field mapping and the location of geologic structures.
Geophysical surveys, such as magnetotelluric, electromagnetic and regional gravity surveys
should be conducted to define the extensions of the prospective area. Consideration should also
be given to conducting resistivity surveys, although these may be cost prohibitive.
Ultimately, a well will need to be drilled in each region in order to assess the temperature of the
resources. The surface location for the first well would probably be located near the hot spring.

Refererences
1. Paul Kruger and Caren Otte, Geothermal energy:Resources, Production, Simulation,
Stanton University, June 1973
2. H. Christopher H. Armstead, Geothermal energy: Review of research and development,
Unesco 1974
3. John W. Twidell & Anthony D. Weir, Renewable energy resources, Spon 1986
 50

4. R. Fabriol & P. Verzier, Reconnaissance geothermique de la republique du Rwanda,

BRGM 1983
5. Enrico Barbier, Geothermal energy technology and current status, an overview,
Pergamon 2002
6. Cornel Otieno Ofwana, Olkaria east geothermal system, Kenya, The United Nations
University, September 2002
7. Ronald DiPippo, Small Geothermal Power Plants: Design, Perfomance and Economics,
GHC Bulletin, June 1999
8. Bureau de recherches géologiques et minières, Ministere de l’Economie, des Finances
et de l’Industrie (France), Les géothermomètres chimiques, Note technique nº 13,
Février 1999
9. Francis F. Huang, Engineering thermodynamics: Fundamentals and applications,
Collier Macmillan, 1976
10. Mac Naughton, Edgar, Elementary steam power engineering, Wiley 1966
11. http://macservcart.uncc.edu/faculty/haas/geol13190/termpap/harris/ visited on August
10, 2006
12. http://www.cc.utah.edu/~ptt25660/geo.html visited on Dec 18, 2006
13. http://www.geothermalenergyassociation.org Visited on Feb 20, 2007
 51

APPENDIX

Table I: GPS data

Location Latitude Longitude Elevation

Mashyuza 1 S 02º 35.102' E 029º 01.191' 1136m
Mashyuza 2 S 02º 34.956' E 029º 00.938' 1173m
Ntaresi
Cyabararika S 01º 30.369' E 029º 39.270' 1833m
Gihugu S 01º 30.837' E 029º 38.223' 1801m
 52

Ruganji 1 S 01º 44.396' E 029º 16.439' 1470m

Ruganji 2 S 01º 44.322' E 029º 16.421' 1470m
Gishyita 1 S 02º 10.058' E 029º 18.362' 1683m
Gishyita 2 S 02º 04.626' E 029º 22.939' 1638m

Table II: Chemical analyses results (NUR 2007)

Site/Parameter K Na Ca Mg
Wavelength 766.49 589.00 422.67 285.21
Mashyuza 1 66.7 171.5 14.3 273.5
Mashyuza 2 97.8 164.9 2.5 11.5
Ntaresi 108.1 173.5 6.4 189.5
Cyabararika 204.3 172.9 6.7 266.5
Gihugu 220.3 164.4 29.7 341.5
Ruganji 1 70.7 167 14.8 112
Ruganji 2 75.8 164 11.6 118.5

Concentrations in mg/l

Table III: Chemical analyses results (Chevron 2006)

Site/Parameter ToC pH Na K Ca Mg
Mashyuza 1 33 7.0 291.2 45.3 89.5 51.8
Mashyuza 2 47 6.5 307.8 48 76 55
Ruganji 1 70.6 7 518.8 39.8 36.4 11.1
Lake Kivu - 8 110.9 87.5 8.1 80.1

Site/Parameter B SiO2 Cl SO4 HCO3

Mashyuza 1 1.18 50.2 141 50.5 1115.8
Mashyuza 2 1.07 48.3 137.9 55.3 1122.7
Ruganji 1 0.55 58.5 236.8 62.1 1137.3
Lake Kivu 0.10 7.9 32 22.3 796.1
 53

Concentrations in mg/l

Table IV: Chemical analysis results (BRGM 1982)

Site/Parameter ToC pH Cond. Ca Na Mg

Mashyuza 1 41.8 6.45 2.9 76.95 287.4 51.77
Mashyuza 2 54.2 6.26 3.5 72.94 298.8 53.96
Ntaresi 20.9 6.28 2.8 220.03 321.8 96
Gishyita 31.7 6.09 2.1 149.89 224 17.6
Cyabararika 18.7 6.22 2.5 121.04 174 205.86
Gihugu 72.4 6.2 2.7 121.84 174 226.03
Ruganji 1 70.6 6.69 5.2 37.99 531 11.39
Ruganji 2 31.7 6.47 5.2 37.79 528.7 11.12

Concentrations in mg/l

Table IV: Chemical analyses results (BRGM 1982)

Site/Paramete Cl K Li SiO2 HCO3 SO4

r
Mashyuza 1 120.89 45.36 0.9 84.72 1049.5 48.03
Mashyuza 2 127.98 47.31 0.95 75.1 1061.7 46.01
Ntaresi 155.99 26.59 1.66 186.86 1690.2 99.9
Gishyita 48.92 24.91 2.8 155.02 1037.2 27.9
Cyabararika 209.88 179.87 0.05 156.22 2038 42
Gihugu 180.1 179.87 0.05 153.21 2123.4 50
Ruganji 1 233.99 41.06 0.41 109.95 1122.7 54
Ruganji 2 233.99 40.67 0.41 105.75 1122.7 44

Concentrations in mg/l
 54

Table V: Chemical geothermometry results (NUR 2007)

Prospect R Na-K Na-K-Ca(β=4/3) Na-K-Ca(β=1/3)

Mashyuza 1 90.3 1327.6 57.9 400.960302
Mashyuza 2 26.5 944.7 233.5 493.0808995
Ntaresi 83.49 911.6 134.9 424.1406831
Cyabararika 79.78 602.5 103.4 347.2217986
Gihugu 79.8 559.8 -44.8 210.5012566
Ruganji 1 78.35 1232.4 45.8 380.985298
Ruganji 2 79.48 1145.2 66.2 400.960302

Table VI: Chemical geothermometry results (Chevron 2006)

Prospect Spring ToC Quartz Na-K Na-K-Ca Na-K-Ca-Mg

Gisenyi 75 110 161 to 212 181 72
Mashyuza 54 101 to 102 241 to 272 204 to 205 21 to 27

Table VII: Chemical geothermometry results (BRGM 1982)

Prospect Spring ToC Quartz Chalc. Amorphous SiO2

Mashyuza 1 42 128 99 9
Mashyuza 2 54 122 92 4
Ntaresi 21 176 153 52
Gisenyi 1 72 143 116 22
Gisenyi 2 71 141 113 20

Prospect Na-K Na-Li Na-K-Ca

Mashyuza 1 259 151 204
Mashyuza 2 259 151 205
Ntaresi 201 193 92
Gisenyi 1 195 63 181
Gisenyi 2 195 64 181

Table VIII: Estimated installation costs for a 10MWe geothermal development

Item Cost in US$ (Million) Cost in Rwf(Billion)

 55

2 production wells 7 3.85

1 Injection well 3.5 1.925
4 Km of 36" brine pipeline 7.1 3.905
10 MW power plant 12 6.6
Office and Salaries 2 1.1
Total 31.6 17.38

Table IX: Estimated installation costs for a 50MWe geothermal development

Item Cost in US$ (Million) Cost in Rwf(Billion)

7 production wells 24.5 13.475
4 Injection well 14 7.7
5Km of 36" steam pipeline 8.8 4.84
5Km of 36" brine pipeline 10.6 5.83
2Km of condensate pipeline 0.16 0.088
50 MW power plant 50 27.5
Office and Salaries 6 3.3
Total 114.06 62.733
 56

Abacus I: Estimates for DMg

Standard Atomic Absorption Conditions for Mg (12)

Wavelength Slit Relative Characteristic Characteristic Linear

Noise Concentration Concentration Range
(nm) (nm)
(mg/l) Check (mg/l) (mg/l)
285.2 0.7 1.0 0.0078 0.30 0.50
202.6 0.7 3.9 0.19 9.0 10.0

1. Recommended Flame: air-acetylene, oxidizing (lean, blue)

2. Data obtained with a standard nebulizer and flow spoiler. Operation with a High
Sensitivity nebulizer or impact bead will typically provide a 2-3 × sensitivity
improvement.
3. Characteristic Concentration with a N2O-C2H2 flame at 285.2 nm: 0.036 mg/L

Standard Flame Emission Conditions for Mg

 57

Wavelength Slit Flame

(nm) (nm)
285.2 0.2 Nitrous oxide-acetylene

Stock standard solution: MAGNESIUM, 1000 mg/L. CAUTIOUSLY dissolve 1.000

g of magnesium ribbon in a minimum volume of (1+1) HCl. Dilute to 1 liter with 1% (v/v)
HCl.
Inteferences: Aluminum, silicon, titanium, and phosphorus depress the magnesium
signal. This effect can be controlled by the addition of lanthanum
(0.1% as chloride) to samples and standards.
The use of the nitrous oxideacetylene flame will also overcome the effect, but
ionization should be controlled by the addition of an alkali salt (0.1% or more potassium as
chloride) to samples and standards.

Standard Atomic Absorption Conditions for Na (11)

Wavelength Slit Relative Characteristic Characteristic Linear

Noise Concentration Concentration Range
(nm) (nm)
(mg/l) Check (mg/l) (mg/l)
0.2/0.4 1.0 0.012 0.50 1.0
589.0 0.7 0.63 1.7 80.0 ---
330.2

1. Recommended Flame: air-acetylene, oxidizing (lean, blue)

2. Data obtained with a standard nebulizer and flow spoiler. Operation with a High Sensitivity
nebulizer or impact bead will typically provide a 2-3 × sensitivity improvement.
3. Data collected with an alkali salt (0.1% or more) added to control ionization.

Standard Flame Emission Conditions for Na

 58

Wavelength Slit Flame

(nm) (nm)
285.2 0.2 Nitrous oxide-acetylene

Stock standard solution: SODIUM, 1000 mg/L. Dissolve 2.542 g of sodium chloride,
NaCl, in deionized water and dilute to 1 liter with deionized water. Preparation of
uncontaminated standards for this element is difficult.
Inteferenes: Ionization should be controlled by the addition of an alkali salt (0.1% or
more potassium or cesium as chloride) to samples and standards. In the
presence of high concentrations of mineral acids, the sodium signal is reduced.
Doublets: The 589.0 nm and the 303.2 nm sodium lines are actually doublets
(589.0 nm/589.6 nm, 303.2 nm/303.3 nm)

Standard Atomic Absorption Conditions for K (19)

Wavelength Slit Relative Characteristic Characteristic Linear

Noise Concentration Concentration Range
(nm) (nm)
(mg/l) Check (mg/l) (mg/l)
776.5 0.7/1.4 1.0 0.043 2.0 2.0
769.9 0.7/1.4 1.4 0.083 4.0 20.0
404.4 0.7 1.9 7.8 350.0 600.0

1. Recommended Flame: air-acetylene, oxidizing (lean, blue)

2. Data obtained with a standard nebulizer and flow spoiler. Operation with a High
Sensitivity nebulizer or impact bead will typically provide a 2-3 × sensitivity
improvement.
3. Data collected with an alkali salt (0.1% or more) added to control ionization.
4. A red filter which absorbs radiation below 650 nm should be used.

Standard Flame Emission Conditions for K

 59

Wavelength Slit Flame

(nm) (nm)
766.5 0.2/04 Air-acetylene

Stock standard solution: POTASSIUM, 1000 mg/L. Dissolve 1.907 g of potassium

chloride, KCl, in deionized water and dilute to 1 liter with deionized water.
Interferences: Ionization can be controlled by the addition of an alkali salt (0.1% or
more cesium or lanthanum as chloride) to samples and standards. Strong concentrations of
mineral acids may cause the potassium signal to be depressed.
Doublets: The 404.4 nm potassium line is actually a doublet (404.41nm/404.72 nm)

Standard Atomic Absorption Conditions for Ca (20)

Wavelength Slit Relative Characteristic Characteristic Linear

Noise Concentration Concentration Range
(nm) (nm)
(mg/l) Check (mg/l) (mg/l)
422.7 0.7 1.0 0.092 4.0 5.0
239.9 0.7 14.0 13.0 600.0 800.0

1. Recommended Flame: air-acetylene, oxidizing (lean, blue)

2. Data obtained with a standard nebulizer and flow spoiler. Operation with a High
Sensitivity nebulizer or impact bead will typically provide a 2-3 × sensitivity
improvement.
3. Characteristic Concentration with a N2O-C2H2 flame at 422.7 nm: 0.048 mg/L

Standard Flame Emission Conditions for Ca

Wavelength Slit Flame

 60

(nm) (nm)
766.5 0.2/04 Air-acetylene

Stock standard solution:CALCIUM, 500 mg/L. To 1.249 g of calcium carbonate,

CaCO3, add 50 mL of deionized water. Dissolve by adding dropwise 10 mL of HCl. Dilute to
1 liter with deionized water.
Flame Adjustement: The absorption of calcium is dependent on the fuel/air ratio and
the height of the light beam above the burner. Although maximum sensitivity is obtained with
a reducing (fuel-rich) flame, an oxidizing(fuel-lean) flame is recommended for optimum
precision.
Other Flames: Calcium determination appears to be free from chemical interferences
in the nitrous oxide-acetylene flame. Ionization interferences should be
controlled by the addition of alkali salt (0.1% or more K as KCl).