Part 24 - cd3wd424.zip - Offline - Renewable Energy Research in India-Renewable Energy Resources and Rural Applications in the Developing World

Part 24 - cd3wd424.zip - Offline - Renewable Energy Research in India-Renewable Energy Resources and Rural Applications in the Developing World
A project
of Volunteers
in Asia
RemJe
EnResources and Rural
Applications
in the DevdoDina World
Edited
by: Norman L. Brown
Published by:
Westview Press
5500 Central
Avenue
Boulder, CO 80301
USA
Paper copies
are
$13.50.
Available
from:
Westview Press
5500 Central
Avenue
Boulder, CO 80301
USA
Reproduced by permission
of Westview Press.
Reproduction
of this microfiche
document in any
form is subject to the same restrictions
as those
of the original
document.
S Selected Symposia Series
Published by Westview Press
5500 Central Avenue, Boulder,
Colorado
for the
American Association for the Advancement
of Science
1776 Massachusetts Ave., N. W., Washington,
D.C.
Renewable En eru
Resou ces and Rur al
Applic :ation .sin the
Develo ping wo rld
Edited by Norman
L. Brown
AAAS Selected Symposium
6
All rig:lts
reserved.
No part of this publication
may be
reproduced
or transmitted
in any form or by any means,
electronic
or mechanical,
including
photocopy,
recording,
or any information
storage
and retrieval
system,
without
permission
in writing
from the publisher.
Copyright
01978
by the
Advancement
of Science
Published
in
1978 in
the
American
United
Westvie-w Press,
Inc.
5500 Central
Avenue
Boulder,
Colorado
80301
Frederick
A. Praeger,
Publisher
Library
ISBN:
of Congress
O-89158-433-1
Printed
and bound
Number:
in
the
Association
States
for
of America
and Editorial
77-18549
United
States
of America
the
by
Director
About the Book
in developing countries is desperate.
1”3;:?energy Gtuation
Because these countries are primarily
dependent on fossi
oil--for
industrial!
growth, they have been hard
fue %a--chefly
hi.5 by oil price increases. Ekrther, in the rural areas, where
most of the population lives, there are Zimited supplies of
increaSing.Ty e.xpensiue diesel fuel or kerosene. NoncorunerciaZ
energy sources such as firewood, dung, and agricultural
residues are generally used in rural areas, but under the pressure
of growing popuZations the ,forests are disappearing.
This is
resuZting in a critical
shortage of firewood for cooking and
heating, as weZI as in the destruction
of the environment. In
addition,
when dung and agricuZtura2 residues are burned, vaZu&Ze fertizizers
are destroyed. Thus, the rural areas--the
sokrces of food and fiber-- face a particuZa,-Zy akrming situation.
SmaZZ-scaZe, decentralized
technoZogiea for exploiting
the
sun r s energy, received directZy
or as wind, fZowing water, or
biomass, provide potential
solutions to the problem of rura2
energy needs. These technologies have been the subject of
numerous studies, including two by the NationaZ Academy of
Sciences. In this volume, members of the two academy study
paneZs have jo5wd with other experts to discuss the status
of these technoZogies and to pZace them in a realistic
context.
Content.3
ix
List of Figures
List of Tables
xiii
Foreword
xv
About the Editor and Authors
bk!zoducticm--Norman
xvii
L. Brown
1
1 Requiremnts forI3xergy in the Rural Areas
of Developing Countries--Rogep ReveZZe
11
Enel=gy' in Seven Countries,13;
Agricultura$.
Need for More En.ergy,18; Conservation. and
Storuge,BO; The Hills of NepaZ,21;* More
E'xepgg for RuraZ lizdustries,25;
Referel:ces,
26
2 Solar l3xergy in the Less Ikeloped
Countries--
George 0. G. L0f
3 Photovoltaic
27
Technology--Morton
B. Prince
45
Introduction,45;
Program GotiLs and Objectives,
46; Strategy,47;
Additional
Sbategy Activities,48; Program PZan,49; AppZkatiom
Suitable for Rural Are& of Developing Countries,
50
4 Alternative
bergy Technologies in Brazil--
Jo& M. MiccoZis
Introdxction,53;
54; AZtermtive
53
The Brazilian Energy Picture,
Energy Production Technologies,
vii
Contents
viii
,- a
;‘ 5 ;
Energy Conservation Measures,7i; Con2 hSZ;7?1S,73; References, 74; Bibliogruphy,
(7c
ti
5 Wild berg-y Cunversicm in In&a--Sharat
TewQTi
ii.
3 Small Hydradie Prim Movers for Rural Areas
of Develcping Countries: A hok at the
Past-- cYosephJ. Ermene
77
89
XstoriqaZ
Perspeetive,89;
WaterwheeZs,95;
The Overshot hkeeZ,lOl; The Undershot Vertiea ': &ZaterwheeZ,104; The Undershot Horizontal WaterwheeZ,lO7; The Water Turbine,
109; Hydraulic Power Sites,lll;
Summury,
11%; References,213
7 WoodWaste as an Ibergy Source in GJxma-John W. PoweZZ
115
The Tropical High Forest of Ghana,llS;
The Timber Industry,116;
AvaiZabiZity
ef
Wood Waste and Ghana's En.ergy fleeds,ll?;
Present Uses of Firewood and ChnrecaL,?20;
lSameDevelopments in the Use of Wood Waste,
122; Developments in the Charcoal Burning
Industry,l22;
Greater Utilization
of Wood
Waste,125; Appendix 1: Some Approximate
Statisties,l27;
References,128
8 M&hanefrmHumn,Anbal,andAgricultural
129
Wastes--Raymond C. Loehr
Introduction,129;
Background,130; Factors Affecting Methane Production, 131; SZudge UtiZization,l39;
Gas Uti~ization,l40;
CoZZeetion
and Preyaration,l41;
Equipment,l43;
Operation,
145; Economies and FeasibiZity,146;
Summary,
147; AcknowZedgements,l48; References and
notes,149
S~mnary and Discussion--Roger
ReveZZe
.
151
List of Figures
Chapter
Figure
Figure
2
1
2
Figure
3
Figure
4
Figure
5 '
Figure
6
Figure
Figure
7
8
Figure
9
Figure
10
Figure
11
Figure
12
Figure
13
Figure
14
Figure
15
Figure
16
Solar
Water Heaters--Plastic
Bag Type
30
(Japan)
Solar Water Heater--Collector
and
Storage Tank (Australia)
Solar Water Heaters--Collector
Testing (Australia)
Solar Water Heater--Site-Built
(United States)
Space Heating with Solar Warm Air-Storage
Space Heating with Solar Warm Air-Heating from Collector
Space Heating with Solar Warm Air-Heating from Storage
Solar Heating with Passive Systems
(France)
Space Cooling with Solar Energy-Absorption
System
Solar Heated Gelling
Experimental
and Greenhouse (Colorado State University)
Experimental
Solar Grain Drying Equipment (Colorado State University)
Regeneration
of Batch-Type Refrigeration Unit with Solar Heat
Solar Refrigeration
with Ammonia
*Absorption System (USSR)
Solar Distillation
of Salt Water
(Chile,
circa 1870)
Solar Distillation
of Sea Water (Florida,
circa 1965)
Solar Still
for Desalting
Water--Under
Construction
(Australian
Design, 1972)
iX
30
30
30
32
32
32
33
33
33
34
34
34
36
36
36
X
List
Figure
17
Figure
18
Figure
19
Figure
20
Figure
21
Figure
Figure
22
23
Figure
24
Figure
Figure
25
26
Figure.
27
Figure'28
Figure
Figure
29
30
Chapter
1
2
Figure
3
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Solar Still
for Desalting
Water--Completed (Australian
Design, 1972)
Solar Still
for Desalting
Sea Water
(Potmos, Greece, 1967)
Sloping,
Ledge-Type Solar Still
(USSR,
i.975)
Plastic
Solar Still
(Syme, Greece,
1965)
Reflecting
Solar Cooker (Indian Design,
19553
Solar Oven (United States Design, 1955)
Reflecting
Solar Cooker (USSR Design,
1975)
Solar Cooker, Concentrating
Type
(United States, 1960)
Solar Power Plant (Egypt, 1913)
Cylindrical
Plastic
Film Solar Concentrators
for Small Electric
Power Generator (Israel,
1960)
Solar Power Generator--Augmented
Flat
Plate (Italy,
1955)
Solar Power Generator--Strptched
Plastic Film Concentrator
(USSR, 1960)
Solar Power Gei:orators--Faceted
Glass
Mirror Concentrators
(USSR, 1975)
Solar Power Generator--Faceted
Round
Glass Mirror Concentrator
(USSR, 1975)
38
38
38
38
39
39
39
39
40
40
40
42
42
42
6
Figure
Figure
Figure
Figure
Figure
o-f Figw=+es
4
5
6
7
8
9
10
11
12
13
Gallo-Roman Flour Mill at Barbegal
Mills on the Three-Mile-Long
Newfound
River in New Hampshire
Small (Less than 100 Horsepower) Hydraulic Prime Mover Cycles
Small Hydroelectric
Stations
Typical Layout of Micro Hydroplant
Cost per kw of Small Scale Hydroelectric
Package Units
Undershot Waterwheel
Overshot Waterwheel
Horizontal
Waterwheel
Roman Undershot Waterwheel
Francis Turbine and American/Francis
Turbine
Pelton Turbine
Propellor
Turbine
90
94
96
97
98
99
100
102
103
106
108
110
110
.
List
Chapter
Figure
7
1
Chapter
Figure
Figure
of Figures
Annual Wood Flow in Ghana
118
Biogas Production
as Related to the Temperature of the Digester and the Time
of Digestion
Schematic ,of a Small Scale Biogas
Digester
138
8
1
2
144
List of Tables
Chapter
1
Table
1
Table
2
Table 3
Table. 4
Chapter
Table
I
17
Energy Sources in Brazil
Brazilian
Energy Market in 1970
55
56
12
14
16
4
Table 1
Table 2
Chapter
Estimated Per Capita Use of Energy in
Rural Areas of Seven Developin<
Countries
Energy Uses in Rural India and
Bangladesh
Characteristics
of Energy Use in
Rural Areas of Seven Developing
Countries
Total Commerciai and Noncommercial
Energy in India and Bangladesh
7
1
Chapter
Average Weight of Wood and Charcoal
Cord Run
Per
123
8
Table
1
Tabie
2
Table
3
Agricultural
Residues Having Potential
for Methane Generation
Estimated Manure and Biogas Production
from Animal Wastes
Quantities
of Biogas Required for a
Specific
Application
. . .
X22-2
132
132
142
Foreword
The AAAS SeZected Symposia Series was begun in 1977 to
provide
a means for more permanently
recording
and more
widely
disseminating
some of the valuable
material
which is
discussed
at the AAAS Annual National
Meetings.
The volumes
in this Series are based on symposia held at the Meetings
which address
topics
of current
and continuing
significance,
both within
and among the sciences,
and in the areas in which
science
and technology
impact on public
policy.
The Series
format
is designed
to provide
for rapid
dissemination
of information,
so the papers are not typeset
but are reproduced
directly
from the camera copy submitted
by the authors,
without copy editing.
The papers
are reviewed
and edited
by
the symposia organizers
who then become the editors
of the
various
volumes.
Most papers published
in this Series are
original
contributions
which have not been previously
published,
although
in some cases additional
papers
from other
sources have been added by an editor
to provide
a more comprehensive
view of a particular
topic.
Symposia may be reports
of new research
or reviews
of established
work, particularly
work of an interdisciplinary
nature,
since the AAAS
Annual Meeting
typically
embraces the full
range of the
sciences
and their
societal
implications.
WILLIAM
D. WY
Executhe
Officer
American Association
for
the Advancement of Science
xv
About the Editor and Authors
I.
‘1
1.
!.
I
1:
oI
II
i
1.
I
.P
I.
b
I.
;,.,
,:
,_
,-
Norman L. Brown is country program speciaZist
with the
Office of International
Aff&zrs at the Department of Energy.
As such, he is chief techn<caZ adtriser to DOE’s program of
energy development with Zess developed countries.
His internat<onaZ +zterdiscipZinary
background in science and technology, particutarty
energy and food problems, has invoZved
technoZogicaZ
him in seeking and encourag&g
appropriate
sotutions.
He mzs the staff study director
for Energy for
Rural Development: (National
Academy of ScCences, 1976).
Joseph J. Ermenc, professor
of engineering at Dartmouth
CoZZege, specializes
in me&xnicaZ engineering and the his- 9
tory axd ph<Zosophy of technology
development.
He is the
i
author of an 8-voZume seties,
Dartmough Readings on TechnoCnnovalogy, and 38 vo Fumes of {nterviews
with outstanding
tom.
He was a member of the NAS pane2 that contributed
to
Energy for Rural Development EMS, 1976).
Raymond C. Loehr, director
of the Environmenta Studies
Program and professor
of civit,
envCronmentaZ and agriculturat CorneZZ Urrivers$ty,
has pubZished over 100
aZ engheehag
technical
papers and three books, incZud<ng Agricultural
Waste Management (Academic Press, 1974) and Land as a Waste
Management Alternative,
which he edited.
He is aZso an advCsor to vaz4ous governmenta2’ and private
orgmizatiions.
George 0. G. Mf, director
of the Solar Energy AppZications Labotitory
at ~oZcrado State Un<versity and vCce-president of the SoZamn Corporation
in Denver, Cs the author of
over 100 books and papers on energy conservation,
solar
energy uti Zization,
heat transfer,
and environmental
engCneering.
of the Internat&maZ
Solar
He is former president
Energy Society and in 1976 received the Lyndon BaCnes Johnson Founcktion awcLTd for contribution
to the betterment
of
xvii
About the Editor
He too was a member of
mankind.
Rural Development.
the
and Authors
NAS panel
'*'
.xvzz7/
on Energy
for
Jose'M. IcikmZb
is special assistant to the president
of' the National Council for Scientific
Development of BraziZ
in Rio de Janeiro, and dire&or
of the Brazilian Science PoZiq Project at George Washington University.
He is the author of several publications
concerning energy pcticy in BraZiL.
John W. PoweZZ, director of the Technology ConsuZtuzcy
Centre at the University of Science and Technology in Kumasi,
G?zana, has conducted a six-year study of smaZZ industries
in
Ghana. He is the athor and editor of two books in mechanica 2 engineering.
Morton B. Pr&ce
the Division of Solar
In 2954, he deveZoped
TeZephone Laboratories,
cations of the siZicon
tronics Corporation.
is acting director for photovoltaics
of
Energy at the Department of Energy.
the BeZZ Solar Battery at the BeZZ
and Zater deveZoped commercial appZiphotovoZtaic ceZZ at the Hof.fman Else-
Roger ReveZZe is professor
of science and public polLicy
at the Univers+ty of CaZifomia,
San Diego, and the Richard
SaZtonstatZ Professor of Population PoZiey at Harvard Universi-Q. He is former president of the American Association for
the Advancement of Science, a member of the National Acadeq
of Sciences, and the recipient of numerous honorary degrees.
His numerous pubZications in the areas of geophysics, nationa2 resource development, and popuZation studies incZude
Survival
Equation:
Man, Resources,
and His Environment
Change
(Houghton MiffZin,
19?2) and Population
and Social
(Crane-Russak, 1972), both of which he credited.
Sharat K. Tewari, a scientist
with the Wind Energy Group
at the National Aeronautica
Laboratory in Bangalore, India,
is invoZved in analysis of wind energy as an alternative
energy source, m-d hardware development in wind power. He
has wtitten several papers in this field.
le Ener
Resources and
Applications i
D&eloping World
Introchiction
Norman L. Brown
The developing
countries
that are not fortunate
enough to be oil producers
are faced with energy needs thaw
are growing
more and more difficult
to meet.
Primarily
dependent
on oil for industrial
growth and agricultural
development,
they have been hard hit by oil price
increases.
With little
or no flexibility
to meet these energy needs
with other
energy resources
and generally
unable to compensate for increased
oil prices
by increasing
their
exports,
the less developed
countries
(LDCs) find themselves
in A
progressively
worsening
position
to compete for the
_) limited
supplies
of fossil
fuel.
In most of these countries
only a small proportion
of the population
is served by a power distribution
network.
The rural
areas,
where the majority
of the LDC population
lives,
generally
depend on limited
supplies
of diesel
fuel
transport
---or
or kerosene
-- with more or less uncertain
wnoncommercial"
energy sources
such as firewood,
dung, and
agricultural
residues.
Thus, the LDCs have two distinct
energy needs.
On the
one hand, industrial
growth
is dependent
on conventional
urban energy systems that use commercial
energy sources
and
technologies
that are already
in use or are being developed
in the industrialized
countries.
Agricultural
development
schemes -- the "green revolution,"
for example -- pegged
to mechanization
and manufactured
nitrogenous
fertilizers
also depend on these energy sources
and technologies.
On
the other hand, the majority
of the population
in most IDCs,
living
in the rural
areas,
is isolated
from central
power
distribution.
They would therefore
particularly
benefit
from development
of technologies
to exploit
renewable
energy
resources
of the sun, wind,
and flowing
water.
Most of the energy-related
assistance
that has been provided the less developed
countries
to-date
has focussed
attention
on urban/industrial-sector
requirements.
Only recently
have development-assistance
organizations
focussed
attention
on the needs of the rural
areas.
This shift
stems
from a growing
awareness
of the importance
of the way energy
is gathered
and used, and an increased
understanding,
on the
part of the donor nations,
of the necessity
to focus on the
needs of the rural
populations
of the less developed
countries.
The choice
of energy technology
made by developing
countries
will
have a long-term
impact on their
development
that is more widespread
and significant
than that of any
other
technological
choice
currently
facing
them.
Choosing
conventional
large-scale
capital-intensive
technologies
implies
a priori
decisions,
conscious
or not, about many
important
policies.
These include
the course of urban development,
expanding
industrialization,
environmental
impact,
large-scale
borrowing
(or foreign
investment)
with
long-term
indebtedness
and problems
of debt servicing,
increased
dependence on fossil
fuels
or commitment
to nuclear
energy,
and last but not least,
the foreign
policy
stance
dictated
by these requirements.
On the other hand, the choice
of small-scale
decentralized power systems
(e.g.,
solar
heating,
cooling,
and gen-eration
of electricity;
windmills;
small-scale
hydroelectric
plants)
implies
a different
set of a priori
decisions.
These
include,
for example,
de-emphasis
of western-style
industrialization
as the sole or primary
immediate
goal of development;
dispersal
of industry
and, perhaps,
changes in financial mechanisms;
and a shift
from western
agricultural
techniques
to emphasis on improvement
of indigenous
agricultural
practices,
with consequent
reduced
demand for energy-consuming nitrogenous
fertilizers.
All of these factors
could
contribute
significantly
to a slowing
down of migration
to
the cities
and urban growth,
with important
effects
on the
rate of growth of dependence
on commercial
energy supplies.
In making these choices,
less developed
countries
must
decide
on the relative
importance
of the commercial-sector
energy needs versus
those of the traditional
rural
sector.
Interest
in small-scale
technologies
for exploiting
energy
received
from the sun - whether
directly
or in the' form of
biomass,
wind,
or flowing
water - has been increasing
recently in both the industrialized
and the developing
world.
In
a recent
report
(1) the National
Academy of Sciences
examined
these technologies
in terms of their
near- and long-term
availability,
particularly
for use in rural
areas of the less
..
Introduction
3
The report
concluded
that,
for the most
developed
countries.
these technologies
will
have little
short-term
effect
However, application
on energy-use
patterns
in urban areas.
of these technologies
to improving
the quality
- and producof rural
life
may have significant
long-term
effects
on the trend toward urbanization.
It was the objective
of the
collection
of papers was drawn to
and their
near-term
applicability
and place them in
rural
setting,
symposium from which this
examine these technologies
in the developing
country
a realistic
economic con-
A discussion
of energy uses in developing
countries
by
Roger Revelle provides
an appropriate
introduction
to the
symposium.
An analysis
of estimated
per capita
uses of energy in rural
areas of seven developing
countries
shows dramatically
the overwhelming
portion
that is devoted to domestic
activities
and agriculture.
Furthermore,
the analysis
demonstrates
the importance
of taking
into account the noncommercial energy sources in developing
countries.
In two of the
countries
examined,
no commercial
energy at all is available
in rural
areas, and in four others,
the ratio
of noncommercial to commercial
energy ranges from 6.3 to 157.
Using
Nepal as an example, Professor
Revelle points
out the serious
environmental
consequences of the incessant
search,
by rural
people,
for the most common noncommercial
fuel available
to
them--firewood.
The discussion
serves as an illustration
of.
the need to make renewable
energy resources
available
to
,villagers
in rural'areas
of developing
countries
in order to
ssible
for them to improve the quality
of their
In a down-to-earth
assessment of solar thermal
technologies,
George 0. G. Le)f contrasts
the cost of converting
solar heat to mechanical
and electrical
energy,
based on
current
manufacturing
costs in the United States,
with conventional
means of producing
mechanical
or electrical
power.
The comparison
is discouraging,
and accounts
for the slow
growth of the solar technology
industry
in the industrialized
countries.
Nevertheless,
this pessimism must be tempered
with a realization
that in many developing
countries,
for
a variety
of reasons - often simply transportation
costs
and uncertainties,
and maintenance
problems
- the cost of
electricity
is much higher
than in the U.S.
*The paper presented
by Dr. Ibrahim
Sakr,
"Swary
and Discussion"
by Roger Revelle,
for publication
in this volume.
referred
was not
to in the
available
4
NOY???QiZ
1:. Qm Lm
For example, the cost of producing
electricity
in Mauritania
is over $1.00 per kilowatt
hour.
Thus, in such areas solar
technologies
net yet economically
competitive
in the United
States would certainly
be economically
attractive.
Most
important,
of course,
are those technologies
that could find
immediate
application,
such as crop drying
and distillation.
Morton Prince's
discussion
of the U.S. National
Photovoltaic
Program similarly
describes
the cost factors
that
are the main concern of development
programs in the United
States.
The goal of this program is a cost reduction
to
US $.50 per peak watt by 1985, and there is reason to believe
that this goal may be reached earlier.
Although
some of the
cost reduction
is expected to result
from development
of
thin-film
techniques
and novel devices,
the bulk is expected
to come from automation
techniques
that reduce the amount
of hand labor involved
in fabrication
of solar cells.
The
argument is made that "the market for photovoltaic
systems
is not sufficiently
large to justify
additional
investment
by private
industry"
to accelerate
initiation
cf these costreduction
procedures.
It must be borne in mind, however,
that this argument is made in the context
of appraisal
of
the known domestic market and labor costs in the United
States.
The potential
market in LDCs has yet to be ascertained;
there are parts of the developing
world where electricity
costs from photovoltaics
are almost competitive
now
with solar arrays at US $15 per peak watt.
Second, the history
of the semiconductor
industry
has
shown that labor-intensive
fabrication
can be economically
performed
in developing
areas.
Thus the medium-term
prospects of the use of photovoltaic
technology
for generation
of electricity
in developing
countries
are not necessarily
as dim as would be inferred
from the description
of the
U.S. National
Photovoltaic
Program.
Brazil
is currently
engaged in a coherent,
long-range
program to make use of its vast resources
to supply the
increasing
needs of its long-term
development
goals.
With
oil accounting
for half the total
commercial
energy used,
the government
is emphasizing
the development
of alternative
energy technologies.
Jose Miccolis'
paper details
the current energy picture
of Brazil
and the multitude
of schemes
for conserving
and supplementing
scarce supplies
that are
being considered.
Solar
energy is seen as playing
an important
role in
Brazil's
medium- and long-term
plans,
in view of Brazil's
8.5 million
square kilometers
lying
almost entirely
between
the Equator and the Tropic of Capricorn.
Development
of
Introduct<on
solar thermal
but institutional
this activity.
technologies
changes
has received
substantial
support,
are seen as needed to accelerate
is the subject
of a major efBioconversion,
however,
The production
of ethyl
alcohol
as a fuel for interfort.
by fermentation
of sugar and cassava,
nal-combustion
engines,
Serious
consideration
is
is being substantially
supported.
being given to the use of hydrocarbon-producing
plants,
as
a way of converting
sunlight,
via photosynthesis,
to valuable
and fermentation
processes
to produce
chemical
feedstocks,
fuel
(methane) and valuable
nutrients
(vitamins,
enzymes,
amino acids)
are ixluded
in biomass conversion
plans.
Mr. Kiccolis'
description
of the
resources
of Brazil
is illustrative
of
a country
that has recently
"graduated"
assistance
programs,
but finds itself
rapidly
developing
industrial
sector,
lems of rural
development.
energy needs and
the problems facing
from most foreign
facing
not only a
but continued
prob-
One of the earliest
concerted
programs of development
of wind-energy
conversion
systas
for use in rural
areas of
in India
more than twenty
a developing
country
wa: started
out that the
years ago.
However, S. K. Tewari points
"optimism
of cheap electricity
from the grids, W based on
the assumption
of cheap and plentiful
petroleum
supplies,
The oil
was responsible
for halting
the program in 1966.
price increase
of 1973 triggered
a new and more intense
interest
in renewable
energy resources
in India,
and the
In common
wind-energy
conversion
program was resumed.
is primarily
with many other developing
countries,
India
concerned with the use of windmills
to provide
shaft power
for pumping water and grinding
grain,
replacing
the muscle
Generation
of electricity
by
power of people and animals.
windmill
systems seems of secondary
importance
in the near
Although
the
high
costs
of transterm for 2everal
reasons.
mission
lines would seem to argue for decentralized
elecstatistics
show that in the 20-30% of
tricity
generation,
India's
half-miilion
villages
that are electrified,
the
primary
use of electricity
is to operate
irrigation
pumpsets
Electricity
to provide
shaft power.
and grind grain,
i.e.,
-for lighting
can be afforded
only by the rich villagers.
Thus, India's
wind-energy
conversion
program is aimed
primarily
at providing
mechanical
power.
Mr. Tewari makes a point
that is important
in evaluating
the use of all the so-called
non-conventional
energy techConventional
methods of
nologies
in developing
countries.
cost/benefit
analysis
frequently
fail
to give adequate weight
5
6
Norman L. Brow?
to the social
benefits
that result
from making energy available where it had not been and would not be available
for
some time to come if conventional
sources were relied
on.
The value of supplying
irrigation
water can be estimated
in terms of increased
crop production.
But the long-term
benefits
in improvement
of the quality
of rural
life
are
difficult
to quantify
and are easily
glossed over in economic comparisons.
The unique ability
of solar technologies
heat, wind, biomass,
hydropower
- to provide
power in isolatwithout
the necessity
of building
roads,
ed communities,
or
constructing
transmission
lines
providing
transport,
to say nothing
of avoiding
the burden of increasing
costs
and scarcity
of petroleum-based
fuels
- must be given appropriate
weight in national
planning
schemes.
The issue of using conventional
economic analysis,
developed
for use in the industrialized
countries,
to evaluate energy technologies
in rural
areas of the developing
world arises
again in considering
the potential
of smallscale hydropower.
In a detailed
history
of the development
of water wheels and turbines,
Joseph J. Ermenc describes
the important
role that these devices played in the development of rural
areas in the industrialized
world.
He notes,
however, that this technology
was displaced
by the advent
of large-scale
central
thermal
and hydropower
stations,
rural
distribution
of electricity,
the internal
combustion
engine,
and improved rural
transportation.
But all of these fact0L.s
were - and are - based on the availability
of cheap an? ulentiful
petroleum.
Although
cheap and plentiful
are relative
terms, fossil
fuels are obviously
not cheap and plentiful
in developing
countries.
With a realistic
awareness of the
implications
of this situation
in terms of such things
as
road building,
transportation,
transmission
line construction,
and materials,
and manufacturing
costs for large-scale
power equipment,
the exploitation
of small-scale
hydropower
sites for decentralized
power systems becomes well worth
examining.
In this,
the People's
Republic
of China seems
to be leading
the way, with thousands
of unites
of less than
lOO-kW capacity
installed
in rural
areas.
It is encouraging
to note that interest
in this approach has been developing
rapidly
in the United States,
both in commercial
manufacture
of small turbines
and as part of the national
development
program of the new Department
of Energy.
The deforestation
of many parts of the world resulting
search for cooking fuel is an issue of
from the incessant
world-wide
concern.
Nepal estimates,
for example, that at
current
rates of removal and regrowth
only 12 to 13 years
of accessible
forest
are left
in that country.
For countries
Introductiox
fortunate
enough to have forests
to support a substantial
it behooves them to use their
lumber industry,
however,
forest
waste efficiently
in order to maintain,
or even
their
present
resources.
Dr. John W. Powell deimprove,
scribes
Ghana's efforts
to minimize
the wood waste from its
forests
by a research
and development
program to utilize
sawdust for cooking and to improve the efficiency
of traditional
methods of making charcoal.
There is no doubt,
however,
that one of the most significant
ways to reduce
the denudation
of forests
for firewood
is by improving
the
With
efficiency
of traditional
wood-burning
cookstoves.
efficiencies
of the order of 10 percent,
even small improvements would significantly
reduce the amount of firewood
consumed - whether as wood or as charcoal.
Conversion
of biomass into a particularly
useful
form
of energy by anaerobic
digestion
is discussed
by Raymond C.
In a lucid paper describing
the state of the techLoehr.
the technical
and institutional
requirements,
raw
nology,
and disadvantages,
Professor
material
needs, and advantages
Loehr puts the process in a meaningful
overall
context.
He points
out that the use of alternative
energy technologies
in rural
areas of developing
countries
should be aimed at
minimum capital
investment
reduction
of human drudgery,
and operating
expense, and production
of energy in a form
In addition,
where biomass conconvenient
for storage.
version
is concerned,
agricultural
productivity
should be
the efficiency
of the use of biomass (wood, cattle
increased,
must be increased,
and the use of plant
dung, crop residues)
fuels
in general must be decreased to prevent
deforestation
and maintain
soil tilth.
The production
of methane by
anaerobic
digestion
of human, animal,
and agricultural
wastes comes close to meeting most of these goals by producing a combustible
gas that can be stored and a stabilized
residue
that is a valuable
fertilizer,
in equipment
that
can be built
and maintained
on a village
level.
The experience
that has already
been gained at the
village
level in countries
such as Taiwan, India,
Korea,
and the People's
Republic
of China justifies
optimism
about the prospects
of increasing
the use of this process.
Nevertheless,
competent
technical
guidance and a careful
evaluation
of social
and economic feasibility
are needed
in each situation.
Two things
are apparent
from the papers presented
in
this symposium and from the variety
of developments
reported
in other forums.
the developing
countries
themselves
First,
have begun to awaken to the implications
of the ccmmitment
to fossil
fuels and central-power
station
technologies,
by
7
8
Norman L. Brown
pressing
for technologies
that rely on renewable
resources.
Second, the principal
barrier
to greater
use of these technologies
has been their
high initial
cost per unit output
and the almost universal
access to the power distributions
systems associated
with larger-scale
alternatives
in the
industrialized
countries.
The existence
of these alternatives
has inhibited
the development
and mass production
of
small-scale
devices
that would simplify
their
design and
lower their
cost.
Without these alternatives,
therefore,
the developing
countries
will
constitute
the major market
for small-scale
decentralized
power systems for the next
five to ten years.
Thus, it is encouraging
to note that the U.S. Energy
Research and Development
Administration,
recently
incorporated in the Department
of Energy, has stated
that it is
taking
steps to "initiate
research,
development,
demonstration,
and related
activities
aimed at helping
to meet the
energy needs of developing
countries"
in a program that
"will
seek technological
solutions
appropriate
to the
rescilrces
and the social,
economic,
and political
goals of
the developing
countries.
By providing
a vehicle
for consideration
of alternatives
that combine exploitation
of renewable energy resources
with technologies
of use on a
scale suited to rural
community and single-family
needs,
the activities
work to achieve maximum benefit
for both
the United States and developing
countries."
(2)
Furthermore,
it is not merely a question
of transferring technology
to developing
countries--all
too often technology developed
in the context
of an industrialized
society
is irrelevant,
at best, or harmful,
at worst,
to developing
countries.
What is needed, in the long run, are technologies
suited
to the needs and constraints
of the country where it
is to be used.
And perhaps most important
is the need to
develop technologies
to help mee'L what Roger Revelle has referred
to as our uniquely
human needs.
In a discussion
of
technology
transfer
in another
forum (3), Professor
Revelle
noted that".
. . we human beings havs some uniquely
human
needs, because of our special
characteristics
as human
beings.
One of the ways in which we are special
is that we
are able to remember the past and to think
about the future.
We go beyond satisfying
the veterinary
needs [food, warmth,
shelter,
sexual satisfaction,
and health],
and provide
the
means for satisfying
the uniquely
human needs that depend
upon our character
as time-binding
animals.
Four of these
needs are hope, security,
participation,
and remembrance.
Every human being must have hope, the hope that the future
will
be better
than the present.
Every human being needs
some feeling
of security,
that the future
won't be worse than
Introdmtion
Every human being needs a faith
that he will
the present.
attain
some kind of immortality
or remembrance,
that is,
soemthing
that goes beyond his own short lifespan,
not forever, but for a while.
"Most poor people in most less-developed
countries
satisfy
these needs by having children:
their
hope for the
children
are their
security
in
future
lies in children;
children
give them a sense of partiold age or sickness;
and they give them remembrance.
cipation
in the future;
Unless we can satisfy
these uniquely
human needs in other
ways, the world is heading for a Malthusian
catastrophe,
and the problems of technology
transfer
will
disappear
in
a grand Armageddcn."
References
1.
Energy for Rural Development
- Renewable Resources and
Alternative
Technologies
for Developing
Countries.
Academy of Sciences.
1976. Washington,
D.C.: National
2.
U.S. Energy Research and Development
Administration.
1976. Policy
statement
issues by Robert C. Seamans,
October 26.
Jr.,
3.
Edited
Department
of State.
Revelle,
Roger
R. 1976.
preparation
transcript
of November 17, 1976 Meeting,
for 1979 U.N. Conference
on Science and Technology
for
Development.
Washington,
D.C.
9
Requirements for Energy
in the Rural Areas
of Developing Countries
1
Roger Revelle
There is a paradox of energy in the rural
areas of developing
countries.
doth too little
energy and too much energy are used, and the basic
problems are to find ways to provide more energy
and at the same time to conserve energy.
The economic chasm that divides
the world
separates two vastly different
levels and kinds of
energy use.
More than 516 of all the energy
obtained from fossil
fuels and hydroelectric
and
nuclear power is used by the 12 hundred million
inhabitants
of the rich countries,
and less than
l/6
by the nearly 3 billion
inhabitants
of the
poor countries.
Just the reverse is true of the
traditional
sources of energy:
human and animal
labor,
firewood,
crop residues
and animal wastes.
These were the predominant--indeed
almost the
only-- sources of energy everywhere in the world
until
about 200 years ago. A man was actually
hanged in England during the 15th century for
burning coal.
The total
quantities
of energy obtained from
traditional
sources in the poor countries
today
are larger than their
consumption of fossil
fuels
and they greatly
exceed the uses in the rich
countries.
In the rural
areas of the poor
countries
energy obtained from local,
noncommercial sources by the people themselves is 5 to 10
times that obtained from commercial sources.
Nevertheless,
usable energy is in very short
supply and the needs both for a large increase in
supplies
and for conservation--more
efficient
utilization--are
great.
From an energy standpoint,
the rural
areas of less developed countries
can be
11
cr!
Tl
>
.I-!
cd
Tl
e
3
,Cd
-
l-l
-
0 ul
oalrlul
r-i
c;
v3
r-i
LL
l
;:‘,‘:r:,
.,,
._
’
.\
::
-,
z.:
::,;..,:
1:
1
I
I
:
L;
Requirements
for
Energy
in Rurd
Areas
13
thought of as partly
closed ecosystems in which
energy derived by people and animals from the
photosynthetic
products of plants is used to grow
an essenand prepare food, which in turn provides
to grow more food and .so on in
tial
energy input
These ecosystems are being disan endless cycle.
rupted by rapid population
growth.
Lnergy
in Seven Countries
Table 1 shows the estimated
per capita use of
energy in rural regions of seven developing
To allow an easy comparison with human
countries.
food intake the units used are thousands of kilocalories
per person per day.
In India and Africa
the food energy contained
in the average diet is
about 2100 kcal; in Latin America, about 2400.
Total energy use is much more than the intake of
food energy, ranging from about 2% times in
Bangladesh to close to 15 times in northern
tixcept fcr the latter,
most of the
Mexico.
energy is from noncommercial
sources; it is provided by the people themselves.
,‘.
-.
._
Five di*fferent
kinds of noncommercial
energy
.;
Tw.
are listed
in Table 1: human labor,
animal work,
;:r'r,., fuel
.F
Energy
wood,
crop
residues
and
animal
dung.
:.z
-.
provided by human labor is assumed to be about l/3
:
‘;
of the energy in the food eaten by the entire
pOpl..llatiOIl.
This assumption is based on a
I'
','-.
detailed
study of the human energy required
for
.*,
:,_,I. different
agricultural,
domestic
and
other
tasks
i:-,':
.,','
i,n rural India (11, in which it was concluded that
,:r. between 40 and 45% of the total
food energy intake
.:7., -of adults and older children
is utilized
in, manual
The remaining
food energy is utilized
in
;,.., work.
Energy
basal
metabolism
and
nonwork
activity.
:,f. utilized
in animal work is estimated
to be 40% of
the food energy intake of bullocks
and other draft
",.'.
.- animals (1).
The energy used in human labor and
animal work is relatively
small compared to the
energy obtained
from fuel wood, crop residues
and
I
dung.
:,;
.I
Table 2 shows how energy is used in rural
I,..
."
i-9.
,;‘ India and Bangladesh.
About 20% of the total
d2'. * energy is allocated
This is
to
agriculture.
ibe‘
human and animal labor,
plus relatively
:: mostly
small quantities
of energy from commercial sources,
;; used Car pumping irrigation
water and for
i,'
:_'
14
Roger ReveZZe
Table 2.
Energy
USeS
in rurti.
India
lo TFy”
KcaJ
-
&l?lC!ljltLlre
Domestic Activities
and Food
processing
Lighting
Pottery,
Brickmaking & Metal
work
Transportation
&
other use
Total
India
(1)
and Bangladesh
Bangladesh
Daily
P;=$"W;
10T't,"c',
(61
peflaCZ&t
10 3 Kcal
1.32
2,52
1.57
-333
7.31
0.48
4.55
030
.852
.050
3.38
0.76
-47
.014
.06
0.35
.22
---055
.22
7.11
1.304
5.18
11.42
.20
production
of chemical fertilizer.
Between two
and three times as much energy is used in domestic
activities
as in agriculture,
mainly for cooking,
food processing,
and procuring
L"lAel. In many
rural areas at least one member of the family must
spend most of his or (usually)
her working time
gathering
fuel for cooking,
and for space and
water heating.
Much less energy is used for
lighting
and for the traditional
energy-intensive
village
industries--pottery,
brickmaking,
metal
work, and blacksmithing-than for agriculture
or
domestic activities.
Transportation
represents
a
significant
use, but in India it is the least
energy-consuming
of all rural uses.
in Bangladesh,
energy used for transportation
is about equal to
that used for lighting,
Most of the energy for
transportation
represents
animal work, and to a
lesser extent human labor,
(people are the draft
animals in much rural
transportation)
or modern
transport
in trucks and tractors.
Table 3 summarizes some of the characteristics
of energy uses in the 7 countries
listed
in Table 1.
The total
per capita use varies widely from about
37000 kcal per day in the relatively
modern agricultural
economy of northern
rural Mexico to about
Requirements
for
Energy in RlwaZ Areas
15
5000 kcal in Bangladesh.
pnergy use in the
Chinese province
of Hunan is 2% times that in
rural India.
Except for northern
Mexico, domestic
uses are much larger than nondomestic uses,
ranging from a factor
of slightly
more than 2 in
India and Bangladesh to 18 in northern
Nigeria
and
37 in Tanzania.
In these African
areas hardly any
animal draft power is used in agriculture,
and no
commercial fertilizers.
Nondomestic uses are more than twice as large
as domestic uses in north.ern rural Mexico because
agriculture
has been modernized.
Much of the
energy here is utilized
in the manufacture
of commercial
fertilizer,
in pumping irrigation
water
and in mechanized agricultural
equipment (2).
Domestic uses are highest
in Bolivia
and Tanzania
and relatively
high in northern
Nigeria
and northern Mexico because of the large quantities
of wood
used as fuel.
In northern
Mexico most energy
comes from commercial sources; in the other six
countries
noncoIl*mercial
energy is overwhelmingly
predominant.
One of the energy units tabulated
in United
Nations statistics
are kilograms
or tons of coal
equivalent,
taking 7.5 million
kcal
as the energy
in a ton of "standard"
coal.
Table 3 shows that
on the average in rural areas of developing
countries
about 600 kg of coal equivalent
are used per
person per year, more than twice what is generally
ascribed to these countries
in the United Nations
The difference
lies in our inclusion
statistics.
of noncommercial
energy sources, which are not
tabulated
by the UN. On a per capita basis,
commercial energy is used principally
in urban areas
as may be seen from Table 4 which shows total
commercial and noncommercial
energy use in urban and
rural
areas of India and Bangladesh.
Table 3 indicates
that there is a rough relationship
between nondomestic energy use and cereal
yields
per cultivated
-hectare.
Northern Mexico
and the Chinese province
of Hunan, with the highest nondomestic
energy use, also have the highest
Cereal yields
in India,
Tanzania,
cereal yields.
northern
Nigeria,
Bolivia
and Bangladesh are very
low compared to those obtainable
with modern agriculture.
Typical
yields
in European and Japanese
agriculture
and in the Corn Belt of the United
SrL?
MN
l
.
b&f;
orlo
. . .
t-00
YE2
. .
.
MMCO
EcEis
. . .
%x?
[email protected]
. . .
18
Roger ReveZZe
:States are 5000 to 6000 kg per hectare compared to
700 to 1100 kg per hectare in India,
Tanzania,
northern
Nigeria,
Bolivia
and Bangladesh.
some
developing
countries,
for example Egypt, obtain
yields
close to those of the developed countries.
One of the critical
differences
between the high
yielding
agriculture
of the advanced countries
and
the traditional
agriculture
of the Indian subcontinent
and sub-Saharan Africa
is the much larger
use of energy per hectare
(but not per ton of food
produced) in the advanced countries.
In contrast,
energy use for cooking in the
rural areas of developing
countries
is higher than
in many United States households,
because the
energy is used very inefficiently.
Fuel wood, for
in open fires
and partly
example, is burned partly
in inefficient
mud or stone stoves.
The fire is
started
before the pot of rice (in countries
where
rice is the dietary
staple)
is placed on it,
and
the fire keeps on burning after
cooking has been
completed.
Only a small fraction
of the.'heat of
combustion actually
serves to boil the rice;
most
of it escapes outside the cooking pot.
As a result
the quantity
of energy going into the cooking process is about 10% of the total
energy in the fuel,
in contrast
to about 30% in a modern gas stove (1).
This inefficient
use of fuel is likely
to have
serious consequences for many developing
countries
if their rural populations
continue to increase.
The principal
source of domestic energy is wood,
the "poor man's oil,"
and trees and other vegetation are being destroyed
for fuel faster
than they
can grow.
Agricultural
.
for
Need for
Additional
energy
four purposes:
More Energy
is needed in agriculture
a.
to provide
supply;
b.
to allow increased
application
of chemical fertilizers,
especially
nitrogen
fertilizer;
c.
to allow
d.
for
a larger
more rapid
improved
and more stable
water
seedbed preparation;
transportation.
Requ$rements for
Energy in Rural Areas
19
Taking India as an example, estimates
by the
Indian Ir igation
Comission (3) indicate
that
about 10" kcal should be used for pumping water-4 times the bullock,
diesel and electric
power now
being used for lifting
water.
This is equivalent
to 620 kcal per day *for each person in rural
India in 1971 (4).
Chemical fertilizer
manufacture
is quite
especially
nitrogen
fertilizer,
energy-intensive,
which requires
&lose to 17500 kcal per kilogram
of
nitrogen
(5).
*Yet chemical fertilizers
are essential
if advantage is to be taken of the tropic,1
which allows two or three crops to be
climate,
grown on the same land each year.
Where nitrogen
in chemical fertilizer
is not available,
the land
must be left
fallow for most of the year, to allow
time for the soil bacteria
to fix nitrogen
from
The wasteful
use of fallow
is one of the
the air.
principal
reasons why y-ields in the developing
countries
are so low.
Applications
oZ- nitrogen
fertilizer
should be raised to around 1L)O kg per
hectare per crop.
With 100 million
double cropped
hectares
in India,
20 million
tons of nitrogen
would be required,
corresponding
to 2180 kcal per
rural
inhabitantper
day.
.r
Rapid harve$ting
and seedbed preparation
are
necessary if two or more crops are to be grown per
But the bullo&s
and other farm animals now
year.
used for cultivating
the land are often too small,
or tooweak
from malnutrition,
to be able to pull
' the cultZvating
equipment hard enough and rapidly
enough SC that &he ground can be prepared after
harvesting
one crop.in
time to plant another one.
Makhijani
and Poole (2) estimate that an additianal
5. x 105 kcal per hectare-per
crop are
required
in construction
and operation
of small
tractors.
In rural
India this would amount to
620 kcal per rural,inhabitant
per day, suppos%ng
that 100 millron
hectares
can be double cropped.
Cultivation
of two crops per year-would
greatly
improve farm employment, probably by
at
least 50%, corresponding
to an added human energy
input of 190 kcal per rural
inhabitant
per day (1).
Improved transportation
is needed to bring
off-farm
inputs to the farm--fertilizers,
pesticides, high-yielding
seed varieties,
farm tools,
farm machinery and knowledge--and
to facilitate
-
20
Rogm Revel Ze
exporting
part of the crops to cities
and towns.
It is obvious that farmers will
not be able to
purchase the off-farm
inputs necessary for modern
agriculture
unless they are able to sell a portion
of their
crops to non-farmers.
In general,
modernization
of agriculture
demands improved transportation both to lower the cost of getting
inputs onto
the farm and the costs of tranportation
of farm
products to people in the cities
and towns.
The total
additional
energy requirement
would
be more than twice the energy now used in Indian
agriculture.
It would then be possible,
in principle,
to approximate
the average U.S. yield of
3.3 tons of food grain per hectare
per crop,
instead of the present 0.6 tons.
Lven with an
assumed yield of two tons per hectare per crop the
increased
value of farm production
could be about
10 times the cost at 1976 prices of the fossil
fuels used for fertilizer,
pumping water, transportation
and farm machinery.
However there are
problems in the use of fossil
fuels,
the principal
ones being the difficulties
of obtaining
foreign
exchange in many developing
countries
and problems
of transportation
and distribution
in rural areas.
We need to look for other sources of energy
that can be produced locally.
One promising
possibility
is the use of crop residues.
For every
ton of the ordinary
indigenous
rice grown in India,
Bangladesh,or
Africa,
close to two tons of straw
are produced, with the same energy content per ton
as the rice.
In modern agriculture
about half as
much energy is used (not counting
solar energy) as
the energy in the food produced.
Thus crop residues contain about 4 times as much energy as the
amount needed for modern agriculture.
If these
residues
could be processed to produce the kinds of
fuels needed, viz.,
liquid
or gaseous fuels,
even
at.a 40% loss of energy in the process,
there would
still
be roughly twice as much energy available
as
that required
for modern agriculture.
But this
would be true only if the energy used in domestic
activities
could be greatly
reduced, i.e.,
domestic
energy could be conserved.
Conservation
and Storage
For energy conservation,
important
single development
perhaps the most
would be not a new
~
,::
/_,
Requirements
for
Energy
in Rural Areas
21
source of energy but a better
stove, a better way
easily
of cooking food, a cheap and inexpensive,
used device that would enable the efficiency
of
fuel use to be raised from 10% or less, to 30 or
40%, as it is in a modern gas stove in the United
a pilot
light.
States --at least a gas stove without
Problems of energy storage must be solved if
local energy sources are to be fully
utilized.
The requirement
for storage arises because both
the uses and the energy sources are intermittent.
Energy must be stored to obtain synchronism
between the uses of energy and the sources of
energy.
One way to store energy is in chemical
fertilizers
and other chemical products.
Another
is in hydrogen, methane or alcohol.
Another is by
pump-back storage of water into reservoirs.
Another may be the use of solar energy pools of
hot water.
The Hills
i '-
of Nepal
The problems of energy use in rural areas are
particularly
critical
in mountainous regions.
In
the hills
of Nepal the people have traditionally
terraced
the lower valley
areas and hillsides
to
make them flat
for better
cultivation.
As the
population
has rapidly
grown in recent decades the
efforts
to grow more
people are making desperate
food, with the result
that areas that should not
be cultivated
are being used for agriculture.
Terraces are being constructed
further
and further
up the hillsides
on steeper and steeper slopes,
in
some places all the way to the top of the hillsides.
Rainwaters
during the monsoon season run
off on the backsides
of these terraces
and form a
slippery
surface between the terraces
and the hill.
And as a result
landslides
occur in which the terraces are destrcyed
and the soil is denuded.
Driving
west from Kathmandu for 40 or 50 .kilometers one sees many landslide
scars where the
whole side of the mountain has simply slid down
and destroyed
anljr possibility
of agriculture.
Even where landslides
do not occur, the terraces
are being rapidly
eroded, and the numerous streams
that lace the hill
country are choked with sediment during the monsoon season.
When these
streams debouch on the Ganges Plain of India,
the
sediment load is dropped, and the rivers
are
diverted
from their
courses,
causing devastating
_
22
Roger Reve 2 Ze
floods that are becoming more common and more
severe each year.
At the same time pasture lands
in the hills
are being over-grazed,
and forests
are being rapidly
destroyed
for fuel wood and for
forage for livestock.
.f
‘.
What is happening in Nepal is happening
equally
in the mountainous regions of India and
Because of rapid
Pakistan,
and in Latin America.
population
growth, the need to produce more food
-is becoming greater
each year, and because ,it has
not been possible
to increase the very low yields
cultivation
is being extended to
per hectare,
The resource
unsuitable,
easily
eroded areas.
base for agriculture
is literally
being destroyed
But it may
and the future
looks very grim indeed.
be possible
to solve the problem by introducing
In the Nepalese hills
two
new sources of energy.
small-scale
village
sources might be combined:
hydropower plants and plantations
of fast-growing
trees.
These possibilities
were explored at a
recent conference
in Nepal, sponsored by the
Council on Science and Technology of His Majesty's
Government of Nepal and the Asia Society.
‘.
:.
,i
.:
.
,‘
P.
*
” .
:
,.
:.
jk,;.
.:
/.
From a technological
point of view, farming
should be concentrated
on the lower terraces
near
the valley bottoms.
With irrigation
and sufficient fertilizer
it should be possible
to raise
yields during the monsoon season and to grow a
second --possibly
even a thirdL-crop
on these lower
If hay could be
terraces
during the dry season.
cut from the pastures during the monsoon and dried,
it would provide feed for livestock
in the dry
season, and the over-grazing
of the pastures and
destruction
of trees for forage could be curtailed.
If the livestock
could be kept in pens and fed on
the
nitrogen
in their urine (Z/3 of the total
hay >
excreted)
as well as the nitrogen
in the dung could
be recovered to fertilize
the fields.
of
Jo far as is known, lIepa has no deposits
coal,
oil,
or natural
gas which could serve as a
It does
feed stock for nitrogen
fertilizers.
The relatively
small
possess abundant limestone.
amounts of fertilizers
now used are imported from
India or from world suppliers
through the port of
Calcutta.
Very few roads exist in the hill
country,
and imported fertilizers
must be carried
Even at quite
on human backs to most villages.
Requirements
high production
costs,
locally
in each village
for Energy in Rural
Areas
23
fertilizers
manufactured
would be eccnomical.
For' the average hill
village
of 250 people,
50 tons of milled
rice or corn would provide the
At a yield of
basic staple for an adequate diet.
two tons per cropped hectare,
this quantity
of
cereals could be grown on 12 double-cropped
hecAbout 2 tons of nitrogen
in chemical
tares.
fertilizer
would be required
to obtain these
in addition
to the nitrogen
in animal
yields,
Irrigation
water would need to be
dung and urine.
pumped from the rivers
for the second crop during
the dry season, plus supplemental
irrigation
in
With an average pumping lift
the monsoon season.
for pumping sufof 15 meters, the energy required
ficient
irrigation
water would be around 10,000
kilowatt
hours.
An equal amount of energy would
be needed to lift
domestic water supplies
300
This is estimated
to be the average elemeters.
vation of village
houses above the river bed.
Using the old electric
arc process, which requires
about 45,000 kwh per ton of nitrogen,
2 tons of
nitrogen
in chemical fertilizer
could be produced
for an energy expenditure
of 90,000 kwh. This
process generates large-quantities
of waste heat
which could in principle
be used for drying hay.
The required
amount of energy, about 110,000
kwh, could be provided by a 15 kw hydroelectric
For an instalplant operated for 310 days a year.
lation
utilizing
high-head
(about 100 meters) and
small flow, the estimated
cost of turbines
and
The cost of the
generators
would be $7,500.
"civil
works," utilizing
buried PVC pipe to transport water from a simple "run of the river"
intake
to penstocks giving
100 meters of head would be
$22,500, or a total
cost of $2,000 per kilowatt
The
for turbines,
generators,
and "civil
works."
including
the equipment for
entire
installation,
electric-arc
nitrogen
fixation
and production
of
calcium nitrate
fertilizer,
using the abundant
limestone
to combine with the nitrogen
oxide provided by, the electric
arc process,
could be
operated by one or two retired
Gurkha soldiers.
These retired
soldiers
have gained considerable
familiarity
with simple machinery during their
After
period of service
in the British
army.
retirement,
they often utilize
this experience
to
establish
and operate small industries
in the more
.
24
Roger Reve ZZe
accessible
parts
of the hills.
The problem of energy for cooking and heating
water could probably best be solved by establishing
plantations
of fast-growing
trees in each village.
The existing
forests
used for fuel wood consist
of
very slowly maturingtrees,
which do not produce
new growth rapidly
enough to meet the village
of the
needs --hence the present rapid depletion
Each hill
villager
uses about three
forests.
quarters
of 8 ton of fuel wood per year, or about
With
185 tons per year per village
of 250 people.
fast growing trees,
this quantity
of fuel could be
provided on a sustained
basis by a 30 hectare fuel
wood plantation.
The estimated
cost of a forest
plantation
of this area would be around $20,000,
including
fencing io protect
the young trees from
livestock.
The total
cost per village
of a hydropower and
, plus the forest plantation
fertilizer
installation
would be approximately
$60,000, or $250 per inhabitant of the hills,
a total
of 1.5 billion
dollars
for the present population
of 6 million
people.
This cost appears to be very high in view of
Nepal's extremely
limited
resources,
but it could
be more than justified
by the probable reduction
in flood damage in India's
eastern Ganges Plain,
let alone the conservation
of agricultural
soils
From
Nepal's
standpoint,
in the hills
of Nepal.
the investment
might be considered
a social
cost,
to be borne by the government in the same manner
as the cost of roads and schools.
,.
A serious social problem might be expected,
however.
It is likely
that the lower terraces,
on
which production
should be concentrated,
belong to
,~;;Y~~
the richest
rps in the villages,
while the
higher terraces,
which should be abandoned, are
the source of livelihood
fcr the poorest village
groups.
To ensure social e;Mty,
it might be necessary to supplement provision
of small-scale
hydropower installations
and forest plantations
by strong governmental
action,
such as rigorously
which would allow a sharing
enforced land reforms,
of the benefits
by all the inhabitants
of the
villages.
Requirements
@ore Energy
for
for Energy Cn Rural Areas
Rural
Industries
With the small and decreasing
size of agricultural
land-holdings
per farm family and the
growing numbers of rural
people without
land, it
will probably be impossible
to raise rural average
incomes to a satisfactory
level unless employment
can be increased
through development of small
At the same time,
industries
in the countryside.
the agricultural
modernization
that can result
from increased
energy supplies
for agriculture
will
facilitate
development of agriculture-related
industries
in villages
and small cities
and towns
of rural regions,
provided that addit ional energy
beyond the needs of agriculture
can be made available.
With the coming of electrification
to villages
in Thikriwala
Thana of District
Lyallpur
in
Pakistan,
many villagers
have been able to install
and operate small power looms for production
of
grey goods 3 which are shipped to the city for final
processing.
Elsewhere in Pakistan and India,
diesel engines and pumps for irrigation
are being
manufactured
in small machine shops located in the
cities
and market towns of the countryside.
In
the Kaira district
of the Indian state of Gujarat.
the "Amul" dairies
provide milk and other dairy
products to Ahmadabad and Delhi,
and many farmers
have been able to increase
their
incomes by maintaining
dafry cattle.
In the small city of Comilla in Bangladesh
a foundry and repair
shop for irrigation
pumps and
other agricultural
machinery ha&been established,
together
with a cold storage plant for potatoes,
Other possia meat packing plant and a dairy.
leather
bilities
for rural
industries
include:
and wood working,
sugar and flour mills,
and food
Even electric
lights
and vegetable
Freservation.
can be used to increase productivity
by enabling
people in villages
and small towns to work at
night.
For development of rural
industries
which can
be competitive
with some of the more centralized
industries
of large cities,
an essential
rec;uirement is relatively
low-cost
energy to 0pera:;e
lathes,
looms and other machinery,
and to provide
heat for metal-working,
dairies,
and food-
25
26
Roger Revetle
Here 3 jus
processing.
concerned with increas
supplementing
(but not
with electrical,
chemi
In the long run there
incomes.
t as in agriculture,
we are
ing human productivity
by
supplanting)
human labor
Cal, or mechanical
energy.
is no other way to raise
References
1.
R. Revelle,
2.
A. M&.hijzni
Science 152, 969 (197%).
and A. Poole, Energy and Agriculture
Third World (Ballinger,
and 168.
3. Ministry
Report
Cambridge,
in the
Mass. 19'7'5) PP- xv
and Power, Government of India,
Commission, 1972 (New Delhi,
1, pp. 41-56 and 201-246.
of Irrigation
of the Irrigation
1972) vol.
4.
Office of the Registrar
General, Government of India,
Census of India, 1971 (New Delhi, 1972-75), in various
parts.
5.
D. Pimentel,
Energy Use in World Food Production
(Report
74-1, Department of Entomology and Section of Ecology
and Systematics,
Cornell University,
Ithaca, N.Y., 1974)
tables 6 and 15.
(Harvard Center
raphed report.
2
Solar Energy in the
Less Developed Countries
George O.G. L’df
There are two primary
reasons that solar energy has not
been successfully
applied
in the developing
countries
of the
world.
The first
reason,
applicable
also to the developed
Although
"raw"
world,
is that solar energy is expensive.
solar energy is essentially
free,
when converted
to a useful
form, it is one of the most expensive
sources of energy we
have.
The second reason for lack of success in applying
solar
energy in the developing
countries
is a poor understanding
of
needs in those areas,
particularly
among solar energy specialists.
Since needs have not been identified
and assessed,
assumptions
pertainingto
solar applications
have been faulty.
My remarks are directed
mainly at these two problems
which impede application
of solar energy in the developing
countries.
I shall
discuss
a number of examples that illusI would like to divide
potential
solar
trate
these points.
applications
into two main categories;
first
the heat applications,
and then, secondly
and briefly,
the uses for work
or electric
power.
Why is solar energy expensive?
Let me begin my answer
by providing
a few numbers.
In a reasonably
favorable
climate
for space heating
in the United States,
or in any
other favorable
solar climate
where space heating
is needed
for more than half the year, about one and a half therms
(150,000 Btu) of useful
heat can be supplied
per heating
That is equiseason per square foot of solar collector.
valent
to about a gallon
and a half of fuel oil at normal
efficiency
of combustion.
For solar water heating,
a year
around need, nearly
four therms canbeusefully
delivered
per
square foot of well-designed
collector
per year in a favorable climate.
At a near-future
cost of buying and installing
a solar water heating
system, approaching
$20 per square
foot of solar collection
surface,
and a ten percent annual
cost of czFita1,
solar heat for this application
has a cost
x 100,000)
= 50 cents per therm.
So,
of ((0.10 x 20/400,000)
the cost of solar heat delivered
for this continuous
use in a
27
28
George 0. G. LE(f
sunny climate
should be near 50 cents per therm if the chosen
conditions
apply.
Simpler,
home-made hardware might not cost
as much, but it usually
works badly in the U.S. and it is
going to work even less effectively
in the developing
countries.
Although
a system at $10 per square foot appears
it is no bargain
if it has to be thrown away in
attractive,
three years.
And in the developing
countries
maintenance
may
be difficult.
Borrowing
money at 8 percent
interest
in a developing
country
may also be difficult
-40
percent
is a more typical
rate.
With a $10 per square foot collector,
a 20 percent
interest
loan, and a lo-year
amortiz&-,io.i
period,
the solar
heat again would be about 50 cents per therm.
There does not
seem to be much chance of getting
the cost below 50 cents per
therm for solar heat delivery.
Number 2 fuel oil is about 50 cents per gallon
in the
in most other countries.
Oil
U.S., but its price is higher
at this price provides
heat at about 50 cents per therm.
So
it looks as though solar heat may now be competitive
with
fuel oil under particularly
favorable
solar conditions.
This favorable
comparison
rests on several
assumptions.
A continuous
use for solar energy was selected.
If less than
four therms per year are recovered
and used, the solar heat
cost increases.
Solar cooling
probably
requires
heat supply
less than a fifth
of the time,
so instead
of recovering
four
therms per year,
perhaps half a therm per square foot is
actually
used per year.
Hence the equivalent
cost of the
energy delivered
is five to ten times higher.
A high efficiency system was also assumed, otherwise
less than four
therms will
be delivered.
Durability
is also important,
because the investment
will
have to be amortized
in less
than ten years if the system life
is short.
Electricity
has higher
value than heat, so conversion
of solar heat to electricity
should be evaluated.
If we
convert
one therm (100,000 Btu) of solar heat to electricity
with presently
available
equipment,
we will
be doing very
well if we obtain
five percent conversion.
So one therm
(100,000 Btu) at five percent
conversion
efficiency
will
produce one to two kilowatt-hours.
If four therms of heat
are recoverable
per square foot of collector
per year,
five
to six kilowatt-hours
of electricity
may be obtained
per
The cost of heat for this electric
production
may be'
year.
in the 50 cents range, so the heat cost per kilowatt-hour
of
With the additional
electricity
would be about ten cents.
costs of turbine,
generator,
and other power plant facilities,
solar electricity
costs approach 20 cents per kilowatt-hour.
There are some places in the world where electricity
prices
are as high as 20 cents per kilowatt-hour,
so solar electricity.may
be competitive
in regions
where central
station
electric
power is not available.
Sohr
Energy
in the Less Developed
Countries
29
Solar energy applications
may be subdivided
in several
ways, and for this discussion,
I would like
to divide
them
into the two categories
of heat and electricity.
Most of the
potential
uses for solar energy in the developing
countries
appear to be concerned with heat, so I shall
emphasize those
applications.
Figure 1 shows a simple type of solar water heater.
Residential
water heating
is a year-round
need in the
developed
countries,
and it appears that sanitation,
health,
and comfort
would be improved by water heating
in many parts
of the world where 'it is not now being practiced.
I say,
"appears"
because, as a solar specialist
rather
than an
anthropologist,
I am not qualified
to assess the requirements,
limitations,
social,
economic,
and political
factors
involved
in the introduction
of new concepts
and practices
in very
different
environments.
The solar specialists
and the experts
in such appraisals
in the developing
countries
should jointly
attack
this important
problem.
The photograph
is of a unit developed
in Japan a number
of years ago.
A simple plastic
bag on a flat
roof is filled
in the morning r&ith water.
The bottom surface
of the bag is
black and the top surface
is clear.
The water is heated during the day by absorption
of radiation
on the black surface,
and drained
into the tube in the evening for the family
bath.
It is a practical
unit,
many thousands were made.
The heaters
lasted
one or two years and then were discarded.
Their low
cost compensated for their
short life,
so cost-effectiveness
Now an improved model of greater
durawas reasonably
good.
bility
is available.
Figure 2 shows a solar water heater which is commercially
made in Australia,
obviously
much more durable.
The
solar collector
in the glass-covered
panel and the storage
tank in the insulated
jacket
provide
most of the hot water
requirements
of a typical
family
in Australia.
Another Australian
model, shown in Figure 3, is even more
widely
used.
The photograph
shows a number of panels being
tested
in an experimental
facility.
In typical
installations,
three to six collector
panels are connected
to an elevated
insulated
storage
tank,
in a thermosiphon
circulationarrangement requiring
no pump.
Figure 4 shows a "home-made" solar
water heater on an Indian reservation
on the CaliforniaArizona
border.
Solar water heating
is not a difficult
technology,
as
can be seen, but careful
attention
must be given to numerous
requirements.
The tubing
in the absorber
plates
must be of
corrosion-resistant
material,
or premature
failure
will
occur.
A durable
glazing
material
must be provided
for reduction
of
heat losses,
otherwise
low efficiency
will
result.
Solar
heating
is a very "forgiving"
technology
-- most solar equipment will
deliver
useful
heat, but the attainment
of good
Fikp
t’ I
Figure
Figuw
!!
3
Fig. 1. Solar Water Heaters-Plastic
Bag Type (Japan)
Fig. 2. Solar Water Heater-Storage Tank (Australia)
Fig. 3. Solar Water Heaters-Collector
Testing
(Australia)
Fig. 4. Solar Water Heater-Site-Built
(United
States)
SoZar Energy in the Less Developed
Countries
31
performance
over long periods
with virtually
no maintenance
requires
careful,
capable design.
Another solar heat application
is space heating.
Although the need for space heating'
is not great in most of the
winter
heating
is a requirement
in many
developing
countries,
areas.'
Particularly
in the more developed
sections
of these
countries,
the space heating
of public
buildings,
schools,
and business
enterprises
which have the capability
for underwriting
the cost of the system, appears to be a practical
application
of solar energy.
In a commercial
U.S. system, shown in Figure 5, air is
For overheated by passing i t through a solar collector.
the
hot
air
is
circulated
through
a bin
night heat storage,
of small rocks (pebble-bed)
which are heated by the air.
When
heat is needed in the rooms, Figure 6, the hot &.ir is delivered
by the blower to the living
space and th?n returned
to the collector.
Figure 7 shows that,
when heat is needed
at night,
house air is circulated
through
the warm rocks.
Finally,
an auxiliary
heat source is used when there isn't
enough solar heat in storage.
A simpler
way of heating
a building
is used in this
structure
in southern
France (Figure
8), where a blackpainted,
south-facing,
thick
concrete
wall absorbs solar
energy passing through two layers
of vertical
glass surfaces.
The concrete
is heated,
and room air circulates
through
slots
in the bottom of the wall,
up the space between the wall and
the glass,
and back into the rooms through
slots
at the top.
This passive system provides
partial
solar heating
by warm
air to the rooms.
Space cooling
may be a more important
need than heating
in most of the developing
countries,
but I want to state
is first
again that expert anaiy .s.f; of needs and capabilities
required.
One method for cooling
with solar energy,
illustrated
in Figure 9, is by use of an abso:*ption
refrigeration
The principle
is similar
to that in the gas refrigeracycle.
For commercial
buildings
and larger
structures
in the
tor.
developing
countries,
where cooling
is needed, this system
It is not
may become practical
after
further
development.
free of electrical
requirements,
because power is needed for
circulation
of heat transfer
fluids.
This system is in the
development
stage so far as solar application
is concerned,
and it is going to be some time before we know whether it has
practical
application
even in the U.S.; until
we know that
prospect,
we probably
will
not be able to appraise
its potential
elsewhere.
A structure
combining
a greenhouse with a dwelling
may
be useful
in a developing
country.
Figure 10 shows a new
building
at Colorado
State University,
comprising
a small
dwelling
in the rear with a greenhouse
in front.
There are
about 800 square feet of greenhouse
area and 800 square feet
Figure
5
Figuw
7
Fig. 5. Space Heating
with Solar Warm Air-Storage
Fig. 6. Space Heatihg
with Solar Warm Air-Heating
from Collector
Fig. 7. Space Heating
with Solar Warm Air-Heating
from Storage
Figure
6
George 0. G. L8f
Figure
8
Figure
9
Figure
10
Fig.
8. Solar Heating with Passive Systems
(France)
Fig.
9. Space Cooling with Solar Energy--Absorption
System
Fig. 10. Experimental
Solar Heated Dwelling
and Greenhouse (Colorado
State University)
33
George 3. G. L/?f
Figure
11
Fig. 11. Experimental
Solar Grain Drying
Equipment
(Colorado
State University)
Fig. 12. Regeneration
of
Batch-Type
Refrigeration Unit with Solar
Heat
Fig. 13. Solar Refrigeration with Ammonia Absorption
System (USSR)
Figure
:2
Figure
13
SOLD Energy in the Less DeveZoped Countries
35
of living
area.
The combined use of solar heat in this installation
supplies
most of the heat requirements
of the
dwelling
plus part of the heat requirements
of the greenhouse.
The greenhouse
is double glazed throughout,
thereby minimizing'heat
demand.
Food crops are now being produced in this
building,
and performance
and cost-effectiveness
are being
evaluated.
Drying of crops is another
use of solar heat.
Figure 11
shows an experimental
system at Colorado
State University
The hot air
where air is being heated in a solar collector.
is then used to dry grain in a bin.
Crop drying
is an ancient
solar application,
usually
accomplished
by spreading
the crop
in the field
and letting
the sun dry it naturally.
Losses are
sometimes high by that method and the possibility
of increasing food yield
by effective
drying
is extremely
important.
Major studies
of the potential
and practice
of this technology are needed.
Refrigeration
is another application
which is a possibility
for the developing
countries.
Food refrigeration
to
reduce spoilage
and increase
effective
productivity
appears
Figure 12 shows a small-scale
to be a need in many regions.
experiment
with a household
size refrigerator
in which concentrated
solar energy is used to generate
a refrigerant
in
a simple two-chambered
device.
The refrigerant
is generated
for a few hours by distilling
it out of a solution
in one
vessel into a water-cooled
second vessel;
after
the refrigerant is collected
in the second vessel,
the device is
lifted
from its solar heat source and put inside
the refriSlow evaporation
of the refrigerant
then
gerator
box.
produces cooling
for about a day.
This system was used on
some American farms about 50 years ago, with kerosene
The solar adaptation
supplying
the energy for operation.
is workable
but inconvenience
and cost are serious
impediments to use.
Another solar refrigeration
development
is under
investigation
in Turkmenia,
USSR. Figure 13 shows a flatplate solar collector
supplying
ammonia refrigerant
to an
insulated
food cooler
(refrigerator).
Again, high cost is
Food refrigeraa deterrent
to individual
household
use.
tion with solar energy probably
has a more practical
application
in large central
refrigeration
plants where
the high capital
costs of the equipment
can be more readily
borne.
Food could be preserved
in community-size
refrigeration
plants
by one of these solar-operated
methods.
Solar heat can be used to produce fresh water from
salt water.
Figure 14 shows a solar water distiller
built
in Chile one hundred years ago.
It was used to
desalt
brackish
water at a mining operation
in the high
desert.
For nearly
20 years,
this plant produced distilled
water to supply the mules that worked in the mine.
36
George 0. G. Lbrf
Figure
Figure
14
15
Fig.
14. Solar Distillation
of
Salt Water (Chile,
circa 1870)
Fig. 15. Solar Distillation
of
Sea Water (Florida,
circa
1965)
Fig. 16. Solar Still
for Desalting
Water--Under
Construction (Australian
Design,
1972)
Figure
16
SoZar Energy in the Less DeveZoped Countl+es
37
A hundred years later
this solar still
(Figure
15) was built
in Daytona Beach, Florida.
The design is nearly
the same as
the Chilean
still,
but improved materials
were used.
A
shailow basin of salt water is covered by sloping
sheets of
glass.
The salt water slowly
evaporates
by solar abscrption,
the vapor condenses on the glass,
and the condensed water
runs into troughs
and out to storage.
A portion
of the salt
water is discharged
to waste so that salts will
not be deposited in the still.
This process has been used for supplying
drinking
water in a number of parts of the world.
Figure 16
shows an Australian
design of a slightly
different
type.
Glass covers are supported
by two concrete
curbs, between
which a slab of insulation
(to reduce heat losstothe
ground)
and a waterproof
butyl rubber liner
are installed.
Figure 17
shows the completed
solar still.
L solar distiller
in a
sunny climate
will
produce about 5000 gallons
of water per
day per acre.
The water is expensive,
several
dollars
per
thousand gallons,
due to the amortization
of the relatively
high capital
cost of the still.
But if there is a need for
desalted
water in quantities
up to 25 to 50 thousand gallons
per day, a solar still
can probably
produce it at lower cost
than any other distillation
method using fuel.
Figure 18
shows the dedication
of an 80,000 square foot solar still
on
the island of Patmos in thti Aegean Sea a few years ago.
There is no source of fresh water on this island
except rain
water collected
on roofs.
Although
solar distilled
water is
expensive,
it may be the cheapest
source.
A Russian solar distiller
in Turkmenia
(Figure
19),
contains
a series
of ledges over which the salt water trickles
and evaporates;
the distillate
condenses on the glass cover
and runs into a separate
channel at the bottom.
The sloping
concrete
tray is coated with black asphalt
to absorb the
radiation.
On the Island
of Symi in the Aegean Sea, a solar still
recently
occupied
the only sizable
flat
area in the town
(Figure
20).
Plastic
film supported
by slight
internal
pressure
served as the condensing
cover surfaces.
Although
of lower construction
cost than a glass-covered
still,
this
unit deteriorated
and failed
completely
in less than a year.
This experience
shows the importance
of using sound design
and durable materials
in solar systems,
in the developing
countries
as well as in the industrialized
regions
of the
world.
Solar cooking
has long been a popular
concept among solar
designers.
One of the first,
Figure 21, was developed
by a
group in India.
This unit was on display
at a solar conference
20 years ago.
Another
type, shown in Figure 22, is
an oven with bright
metal vanes to reflect
solar radiation
through a glass cover into an insulated
box.
In Figure 23, a
solar cooker is being demonstrated
in a Soviet laboratory
Figul-t. 18
Figuw
17
Fi,<Ul t’ 19
Fig. 17. Solar Still
for Desalting
Nater--Completed
(Australian
Design,
1972)
Fig. 18. Solar Still
for Desalting
Sea Water (Potmos,
Greece, 1967)
Fig. 19. Sloping,
Ledge-Type
Solar Still
(IJSSR, 1975)
Fig. 20. Plastic
Solar Still
(Syme, Greet?,
1965)
Fig.
21. Reflecting Solar Cooker
(Indian
Design,
1955)
Fiprc
21
Figure
22
Figure
23
Fig. 22. Solar
Oven (United
States Design)
Fig. 23. Reflecting Solar Cooker
(USSR Design,
1975)
Fig. 24. Solar
Cooker, Concentrating
Type
(United
States,
1960)
Figure
24
Fig. 25. Solar Power Plant (Egypt,
1913)
Fig. 26. Cylindrical
Plastic
Film Solar Concentrators for Small Electric
Power Generator
(Israel,
1960)
Fig. 27. Solar Power Generator--Augmented
Flat
Plate
(Italy,
1955)
SoZar Energy in the Less Developed
Countries
41
Several
small aluminum reflectors
focus the
near Tashkent.
sun on the cooking vessel.
Figure 24 shows an effective
cooker developed
by the
The metallized
plastic
reflector
University
of blisconsin.
focuses about a kilowa2t
onto the bottom of the cooking
The photograp"ii
shows a group of people,
mainly
vessel.
farmers in rural
Mexico, watching
a cooking demonstration.
Nearly two hundred cookers were distributed
and evaluated
in
several
Mexican communities,
with remarkably
consistent
The cooker was a technical
success and a social
results.
The idea failed
because people didn't
like to cook
failure.
Even though in some areas people did
in this unfamiliar
way.
not have enough food, they preferred
to buy kerosene for
cooking what little
food they had.
Social customs and many
other factors
clearly
must be reckoned with.
I don't think
solar cooking is going to be used anywhere except as a
novelty
in the United States.
Turning
now to the other large potential
application
of
solar energy, we see in Figure 25 the first
large solar power
plant built
in Egypt in 1913.
Long metal troughs
focused
the sun on axial
tubes in which steam was generated
for use
in a pumping engine.
This 50 horsepower
plant operated
a
year or two, but when maintenance
became expensive/the
project
was abandoned.
Figure 26 is a photograph
of the solar collectors
experimentally
used in 1960 for the supply of heated fluid
to a three kilowatt
electric
turbo-generator
designed
in
Israel.
Cylinders
of plastic
film,
inflated
by a bit of
air pressure,
are about 30 feet long and five or six feet
in diameter.
They focus solar radiation
onto a metal tube
in which fluid
is heated for delivery
to a heat storage
vessel and a small engine.
About 20 years ago, the solar pump shown in Figure 27
was displayed
at an exhibit
in Arizona
by an Italian company.
Sulfur
dioxide
was vaporized
in the 400 square foot flatplate collector
and supplied
to a reciprocating
engine which
drove a one-horsepower
water pump. Although
hoped to be a
commercial
product,
high cost forced cancellation
of manufacturing
plans.
Although
numerous solar power generators
have delivered
energy,
the cost of the electricity
produced has been very
high and certainly
not competitive
with other sources,
if
But if power must be supplied
other sources are available.
in areas hihere no other source exists,
solar might be a
practica';
source.
Most of the solar electric
research
effort
in the Soviet Union has been oriented
to this objective.
If
electricity
is needed hundreds of miles from the nearest
power line,
much higher costs of site-generated
power can be
tolerated.
Figure 28 shows one of several
types of solar
concentrators
focusing
the sun onto small receivers
in which
George 0. G. L8f
Figure
29
Fig. 28. Solar Power
Generator--Stretched
Plastic
Film Concentrator
(USSR, 1960)
Fig. 29. Solar Power
Generators--Faceted
Glass Mirror
Concentrators
(USSR, 1975)
Fig. 30. Solar Power
Generator--Faceted
Round Glass Mirror
Concentrator
(USSR,
1975)
Figurk
30
Solar
Energy
in the Less Deve~opec! Countries
43
steam or some other
fluid
is produced.
The steam is then
supplied
to a small engine for electric
generation.
A thin
metallized
plastic
film
is stretched
acrc,ss.a
dish,
on the
back side of which slight
suction
is applied
by means of a
small
pump.
The suction
pulls
the plastic
film
into a curved
contour
which focuses
the sun on a smal'i boiler.
About one
horsepower
would be developed
by this
unit.
Figure
29 shows another
concentrator,
in Tashkent,
consisting
of an array
of glass mirrors
that
focus the radiation on a boiler
which supplies
steam to an engine
driving
a
three
to five
kilowatt
generator.
Figure
30 showsa variation
on this
design.
Although
the USSR is an industrialized
country,
there are large
regions
that have the characteristics
of undeveloped
countries.
The prospects
for solarpower
supply
in such circumstances
need to be thoroughly
evaluated.
In summary, it is important to note that,
compared to
conventional
energy technologies
in the industrialized
However, the higher
countries,
solar energy is expensive.
energy costs in many parts of the developing world make
some of the solar technologies
currently
economically
attractive
in LDCs. Second, the history of attempts to
introduce solar cooking amply demonstrates that efforts
to
introduce such technologies
must take into account not only
the economics of a given situation,
but also local customs.
3
Photovoltaic
Technology
Morton B. Prince
INTRODUCTION
The U.S. National Photovoltaic
Program, which is the
responsibility
of the Division of Solar Energy of the Energy
Research and Development Administration
(ERDA), will be
described with its goals, objectives,
strategy,
and plans.
Since photovoltaics
today is relatively
expensive compared
to other sources of energy, its terrestrial
uses are limited
today.
Some of these applications
will be reviewed and
shown to be applicable
to the developing world.
As the cost
of photovoltaics
is reduced significantly,
more and more
applications
will become cost effective,
tispecially
in rural
areas in the developing world.
into
Photovoltaics
is the direct conversion of sunlight
electrical
energy with no intermediary
processes.
Photovoltaic
systems offer the potential
of clean,
highly reliable
power. They are capable of operating
efficiently
in a variety of applications
ranging from small,
low power devices such as remote instruments,
to dispersed
systems (residences,
shopping centers, schools, industry),
to large central power stations
for intermediate
and peak
load operations.
Present sales consist primarily
of moderate efficiency,
highly reliable
silicon-based
solar cell arrays, at a median
cost of $25,000 per peak kilowatt
(kW). Due to the high
cost, annual production of solar cell arrays has been limited
to several hundred kilowatts.
The cost of photovoltaic
systems will be reduced by
automation or the use of thin film or novel devices in arrays.
However, none of the cost-reduction
approaches is sufficiently developed for commercial use, and the market for photovoltaic
systems is not sufficiently
large to justify
additional investment by private industry.
46
Morton
B. Prince
Market information
indicates
that photovoltaic
power
can be rapidly
expanded both in volume and type of application as the cost of solar cell arrays is reduced.
It is
expected that photovoltaic
arrays can find broad commercial
applications
if the associated
array costs were reduced to
the order of $lDO-$300 per peak kilowatt.
The U.S. Photovoltaic
Program is designed to overcome
the critical
problem of high initial
cost resulting
in low
array production.
This will
allow for significant
growth
in the use and application
of Solar Photovoltaic
Conversion
Systems (SPCS).
PROGRAMGOALS AND OBJECTIVES
:
The overall
goal of the photovoltaic
program is to
develop reliable,
low-cost
photovoltaic
systems and to
stimulate
the creation
of a viable
industrial
and commercial
capability
to produce and.distribute
these systems for
widespread
use in residential,
industrial,
and commercial
and governmental
applications.
More specifically,
a major
goal of the SPCS program is the development and demonstration
of photovoltaic
systems providing
electric
energy at costs
of 40-60 mills/KWH as compared to present conventional
systems costs of lo-30 mills/KWH for intermediate
and peak
load operations.
In pursuit
of these goals, three primary objectives
should be accomplished
by 1986:
The reduction
of solar array costs to $500 per
l
peak kilowatt
with an annual production
of 500
megawatts (MW) per year.
a
The combined costs of collectors
and cells
for
systems using concentration
to be $250 per peak
kilowatt.
0
The demonstration
of thin-film
array technology
feasibility
leading to array costs of $lOO-$300
per peak kilowatt.
Specific
non-technical
objectives
have also been
identified,
including
studies
on:
a
Environmental,
institutional
and legal issues such
as the 0wnershi.p of on-site
systems, systems
financing,
availability
of investment
capital,
sun
rights,
local building
and safety codes, land use
restrictions
and interfaces
with local utility
networks.
l
Development of performance
standards
for photovoltaic systems and components that can be used as a
basis for warranties,
insurance,
and consumer
protection
legislation.
Pho tovo Ztaic
Techno logy
It is anticipated
that achievement
of the primary
objectives
will
make photovoltaic
systems economically
competitive
(for selective
applications)
with alternative
energy sources for on-site
residential
and industrial
as well as for central
power generation.
applications,
STRATEGY
As indicated,
the cost of photovoltaic
arrays must be
reduced if the market for these systems is to be expanded.
Four concurrent
sets of activities
have been defined
to
assist
in achieving
the programs goals and objectives.
These
activities
relate
to the use of existing
technology
to
expand the market,
development
of large sheet silicon
technology,
thin-film
and new material
development,
and
development
of concentrators
and high intensity
cells.
1, Market Expansion of Existinq
Technology
- Market
stimulation
through
government
purchase of a significant
fraction
of early-annual
cell
production
is planned.
Solar
cell manufacturers
will
thus have an incentive
to use more
automated,
lower cost production
techniques.
Government involvement
will
provide
the industry
with
a large initial
market and the public
with a basis for
technology
comparison.
The solar cells
purchased will
be
carefully
tested,
evaluated,
and used in various
government
and non-government
applications.
It is expected that ERDA
purchases
of approximately
600 kW through
1978, coupled
with purchases
by other Federal
agencies,
should result
in
a factor
of 4 reduction
in the present
cost of silicon-based
solar ref is +fi
by
l.w apprcx1 'P+Q~\/
1bAbL'J $5000 per peak kilowatt
1979.
A total
government
purchase of approximately
11 MW
through
1983 is planned.
Costs for silicon
solar arrays
are expected to drop to $1000 per peak kW by 1984.
2. Develop Large Area Silicon
Sheet Technology
- Most
currently
marketed photovoltaic
systems use silicon-based
solar cells
which are hand-processed
and individually
mounted on a supporting
structure.
Though the technology
and operation
of silicon
solar arrays are well established,
present costs limit
their
application.
One approach to cost reduction
is the development
of
high capacity,
low unit cost production
techniques
for
silicon
cells.
The acceleration
of this activity,
involving
silicon
sheet technology,
will
provide
for:
(a) an early
reduction
in production
costs of solar cell grade silicon
from $65 per kilogram
(KG) to $10 per KG; (b) increased
efficiency
of the solar cell production
fabrication
(over 75
percent
of the silicon
material
is now wasted);
(c) improvement in the ratio
of cell to array area (packing
factor);
(d) deveiopment
of suitable
encapsulation
materials
to
increase
array iifetime;
and (e) automated production
of
48
Morton B. Prince
silicon
solar cell arrays.
Production
processes,
techniques,
equipment
and
experimental
process production
plants will
be developed
to
support
the commercial
adoption
of silicon
sheet technology.
It is expected that this process will
reduce the siliconbased solar cell array costs to $500 or less per peak kilowatt.
Research - Use
3. Conduct Thin Film and Novel Materials
of thin film deposition
techniques
utilizing
silicon,cadmium sulphide,
gallium
arsenide
and other materials
may
allow the production
of solar arrays costing
$lOO-$300 per
peak kilowatt.
However, the physics,
engineering
and
manufacturing
of high efficiency,
reliable,
low-cost
thin
film arrays is poorly
understood
at present.
In particular,
present
thin film arrays have efficiencies
of only 3 to 7
percent.
Studies
are underway to develop a better
understanding
of basic processes,
fabrication
techniques
and costs.
The
objectives
of this research
are to develop and demonstrate
thin-film,
10 percent
efficient
solar cell arrays by 1980,
and to demonstrate
the feasibility
of $lOO-$300 per peak
kilowatt
arrays by 1986.
4. Develop Concentrators
and High Intensity
Solar
Cells - For a given power output,
photovoltaic
system costs
may be reduced by employing
solar concentrators
to increase
the solar energy received
by each cell.
This reduces the
number of relatively
high-priced
cells
required.
The use of
concentrators,
combined with cells
from low-cost
silicon
arrays costing
$500 per peak kW, should lead to combined
collector/cell
costs of $250 per peak kilowatt
by 1986.
Systems using such arrays may be competitive
with conventional power for residential
and central
station
applications.
Analyses
of concentrators,
cells,
cooling
requirements,
and
uses of available
thermal
energy are now being conducted.
ADDITIONAL STRATEGY ACTIVITIES
The program strategy
also includes
studies
to determine
the necessary
costs for photovoltaic
systems for on-site
and
central
station
applications.
The effects
of institutional,
legal,
and socio-economic
factors
on the introduction
of
photovoltaic
systems will
be analyzed.
Parallel
design
studies
of residential,
commercial,
and central
station
applications
of photovoltaic
power will
be carried
on to
define
system, subsystem,
and component specifications
including
cost,
reliability,
performance,
and efficiency.
Definition
planning
activities
through the 1980's will
support
and receive
input from test activities
and the
research
and development
program.
These studies
will
be
instrumental
in guiding
the direction
of the photovoltaic
Photovo 2taic
i
..
in
rI.
,
8'
I
Technology
program and wil? aid in eva!uatl:ng
results
from tests,
and early
applications
of phctovoltaic
systems.
experiments,
Marketing
studies
now in progress
will
estimate
the
size and characteristics
of the market for early applications
of photovoltaic
power in both the government
and commercial
An initial
estimate
for near-term
(1975-1985)
sectors.
applications
indicates
that wide-spread
application
of
photovoltaic
power could include
such items as impressed
current
corrosion
protection
of gas well casings
and gas
as well as providing
power for railroad
and oil pipelines,
crossing
signs,
navigational
buoys, and highway signs.
Applications
such as water sampling
systems,
food and
medicine
refrigeration,
and security
surveillance
systems
are under consideration
for implementation
on a cost-sharing
basis.
New and more detailed
market and mission
studies
are
(a) define
the domestic
and foreign
being initiated
to:
market for photovoltaic
systems as a function
of system
cost,
performance
requirements
and uses; (b) define
the
technical,
financial,
institutional,
legal and local factors
affecting
photovoltaic
system acceptance;
(c) analyze certain
potentially
attractive
applications
to define modifications
in existing
design approaches;
(d) develop and maintain
a
plan that defines
the necessary
activities,
responsibilities,
and actions
for the successful
commercialization
of potentially
attractive
applications;
(e) evaluate
the effects
of
the introduction
of photovoltaic
systems on the size and
characteristics
of the photovoltaic
market;
and (f) assure
effective
implementation
of commercialization
activities.
These marketing
studies
will
develop a description
of
the application
and market potential
of photovoltaic
systems.
Necessary technical
developments,
required
changes in
institutional,
legal or local
conditions,
and required
experimental
pilot
and demonstration
facilities
and activities
will
be delineated.
PROGRAMPLAN
The Photovoltaic
Conversion
Program is designed to
expand the commercial
use of photovoltaic
systems as rapidly
as possible
through
research,
process development,
testing,
The cost an: risk of photovoltaic
systems,
and application.
both to potential
purchasers
and the Federal government,
By encouraging
will
belimited
as a result
of this program.
government
purchase of a significant
fraction
of the early
annual solar cell
production,
ERDA will
provide
industry
with %he incentive
to adopt low cost production
techniques.
Industry will be able to produce solar cells at decreasing
cost per peak kilowatt.
49
50
Morton B. Prince
The development
of automated
processes
and experimental
production
facilities
will
aid industry
in reducing
solar
array production
costs.
To take advantage
of the understanding
of silicon
cell technology
developed
over the past
20 years,
emphasis will
be on the development
of processes
that can provide
low cost silicon
solar arrays.
In addition,
reduction
of solar array costs through the use of thin film
deposition
techniques
and novel materials
and devices,
and
development
of concentrating
systems will
be pursued.
commercial,
and other applications
Early residential,
of photovoltaic
systems will
be implemented
to develop
information
on operational
costs,
reliability,
and performance, and to acquaint
potential
customers with the
characteristics
and feasibility
of such systems.
APPLICATIONS
SUITABLEFOR RURAL AREAS OF DEVELOPING COUNTRIES
ERDA has been pursuing
early
low power applications
for
photovoltaics
that could be useful
for rural
areas in
This work has been done primarily
developing
countries.
In addition
through
the NASA/Lewis Research Center (LeRC).
ERDA has been interacting
with the World Bank and the Agency
for International
Development
(Department
of State).
These
latter
organizations
have been involved
in searching
for
low cost,
reliable
power supplies
to help upgrade the quality
of life
in developing
countries.
The LeRC has been cooperating with these organizations
and with the U.S. Department
of Interior,
Bureau of Indian Affairs
on similar
activities
for our own Indian reservations.
The results
of testing
on
the Indian reservations
will
be applicable
to many other
parts of the rural
world.
One of the earliest
experiments
carried
out by the
LeRC was in India with the Satellite
Instructional
TV
Experiment
(S.I.T.E.).
Panels of photovoltaic
cells
were
used to charge batteries
for powering
village
television
sets for receiving
instructional
material
from a stationery
satellite.
The experiment
was concluded
last year when the
satellite
was moved and the test was considered
successful.
Another useful
application
for rural
villages
is the
solar powered refrigerator.
Two installations
have been
These
made in the United States as part of the ERDA program.
are small (4 ftS) camper size refrigerators
used to hold
One
medicines
such as vaccines
and other perishable
items.
installation
was on a remote island
in the Isle Royale
National
Park in upper Michigan
for supplying
a construction
crew.
The second is operating
in a rural
health
dispensary
on an Inds’an reservation
at Sii Nakya in the south part of
Arizona.
Each is powered by a 200 watt photovoltaic
panel.
Under consideration
for development
are shallow
well
water pumps powered by photovoltaics
to supply water for
Photoudtaic
Techndogy
52
human and animal consumption.
Also under consideration
is
a photovoltaic
powered cereal grinder.
These latter
two
applications
would relieve
the rural villagers
of the
tedious work of drawing water and grinding
meal and allow
their time to be spent on more productive
activities
which
in turn could help improve their quality
of life.
Several developing countries
have indicated
that they
would like to encourage tourism to remote village
areas but
that tourists
are reluctant
to stay at such areas without a
modicum of their normal comforts.
By powering small appliances with photovoltaics
it is expected that tourism could
be increased with the villagers
benefiting
from the resulting
commerce.
All of the above applications
require daily energy in
the 100 watt-hour
to 10 kilowatt-hour
energy range.
Photovoitaic
power systems capable of delivering
such energy
requirements
are available.
However the cost of such systems
leads to problems in most capital
scarce developing countries
since these photovoltaic
systems are associated with up-front
Even though the operating
costs of such systems are
costs.
minimal due to lack of fuel needs, the initial
investment
requirements
does pose a serious problem.
With the cost reduction
program under way at ERDA, this
problem will be reduced significantly
during the next ten
years.
4
Alternative Energy
Technologies in Brazil
Jo&M.
Miccolis
Introduction
It is imperative
that Brazil
consider
alternative
energy production
and conservation
techniques
in order to
cope with one of the most demanding crises
of its modern
history:
the so-called
"energy crisis."
To mention just
one of the most serious
consequences
of the current
situation,
Brazil
imports
about 650,000 barrels
of petroleum
per
day, which accounts
for more than 78 percent
of total
consumption,
dollars.
at an annual cost of more than 3 billion
Oil amounts to 50 percent
of the total
energy utilized
in
Brazil.
Over the past few years several
steps have been
taken in the direction
of reducing
Brazilian
dependence on
external
sources of energy.
energy
production
Alternative
technologies
have been examined.
Some of them are very
promising
medium-term
solutions,
and are mostly in the research
and development
stage.
Others offer
attractive
short-term
possibilities
and are currently
in the initial
stages of implementation.
Energy conservation
measures that have been undertaken
in the past three years have not been very successful.
Despite
sharp rises
in fuel prices,
internal
consumption
is rising
consistently
at a rapid rate.
More recently,
however,
the Brazilian
government
has enacted much tougher
conservation
measures that are expected to change substantially
the energy consumption
pattern
in the country.
This paper is an attempt
to report
the major alternative energy activities
underway in Brazil.
Due to the
range of possibilities
which could be considered,
and the
limited
munt
of time and space available,
relevant
technical
details
will
not be included;
such details
are, however, available
upon request.
53
54
Jo&
M. Miccolis
The Brazilian
Natural
Energy
Picture
Resources
Brazil's
waterpower
resources
are among the greatest
in
the world,
with an estimated
potential
output
equivalent
to that of 4.5 million
barrels
of oil per day.
While the
installed
capacity
of generating
plants
increased
fr:.2m about
7 million
kilowatts
in 1964 to 22 million
kilowatts
in 1977,
the vast majority
of Brazil's
estimated
120 million
kilowatts is still
untapped.
The potential
sites
are distributed all over the country.
The Northern
region,
where
the Amazon basin is located,
accounts
for about 40 percent
of that total;
the densely populated
and impoverished
Northeast has 11.3 percent;
the South and Southeast,
the most
prosperous
and industrialized
regions
in the country,
have
48 percent,
of which most of the economically
feasible
sites
will
be totally
tapped by the year 1985.
The Amazon tributaries
located
on the north border of the river
would
alone be able to meet Brazil's
requirements
over the next
several
decades.
However, transporting
this energy to the
main energy-consuming
centers
about 2000 miles away poses
several
technological
problems which are yet to be resolved.
While water power resources
are abundant,
petroleum
and
natural
gas reserves
are relatively
small.
Total proven
petroleum
reserves
are estimated
at 780 million
barrels,
with additional
investigations
now taking
place on the continental
shelf.
Usable natural
gas reserves
are calculated
at 15 billion
cubic meters (530 billion
cubic feet).
Suitable
coal reserves,
mstly
in Southern Brazil,
are
estimated
at 2.2 billion
tons, and total
reserves
at about
4 billion
tons.
The additional
reserves
consist
mainly of
high ash, high sulfur
coal which yields
a low-grade
fuel.
It has been determined
that shale deposits
extend from
Southern Brazil
right
up to the Amazon basin.
Estimates
indicate
that shale reserves
could be the source of 3.1
billion
barrels
of shale oil.
Uranium reserves
are estimated
at between 7500 tons and
12,000 tons.
Newly discovered
radioactive
anomalies
in
Northern
Brazil
show definite
signs of potentially
large
uranium deposits.
Thorium,
an alternative
material
to be
used as fuel for gas-cooled
high-temperature
nuclear
reactors,
can also be found in large monazite
sand deposits,
as
well as in niobium deposits
that are estimated
to be from
several
hundred thousand to about one million
tons.
Alternative
Energy
Supply
Energy Technologies
55
and Demand
Table 1, below,
shows Brazil's
changing
various
energy sources from 1952 to 1972.
Table 1.
Energy
in BraziZ
Energy
sources
sources
Coal
Petroleum
Gas
Hydroelectric
Power
Other Fuels
Firewood
Sugar Cane Bagasse
Charcoal
TOTAL
in Brazil
reliance
on'
(1).
1952
1972
6.1%
28.0%
0%
11.2%
54.7%
(49.9%)
( 2.1%)
( 2.7%)
100.0%
3.6%
44.8%
0.3%
20.8%
30.5%
(27.0%)
( 2.0%)
( 1.5%)
100.0%
-
Overall,
energy consumption
has tripled
in absolute
Petroleum
consumption
invalue in this 20-year period.
creased from 128.9 million
barrels
in 1964, to 302.4
million
barrels
in 1974, of which 78 percent
was imported.
The petroleum
industry
is under fairly
extensive
government
control
through Petrobras,
the government-owned
oil company
which has almost full
responsibility
for exploration,
production,
and refining
of petroleum
in Brazil.
Private
enterprise
.accounts for 12 percent
of the country's
refinery
output.
The Brazilian
government has recently
signed
"risk
contracts*'
-- in which the government
shares in any
profits,
but does not incur losses -- with foreign
oil
companies for drilling
on the Brazilian
continental
shelf.
1
Electricity
consumption
in Brazil
has grown in the
period
from 1968 to 1975 at an average annual rate of 11.8
percent.
The electricity
generating
capacity,
over the
same period,
has increased
at an average annual rate of 12
percent.
In lQ75, the overall
installed
generating
capacity
was almost 20,WO megawatts,
of which 82.7 percent
was hydroelectric
and the remainder
thermoelectric.
Also in 1975,
total
electricity
consumption
was about 68 million
megawatthours.
Thus, for a population
of little
more than 100
million
people the per capita
consumption
was about 650
kilowatt-hours,
or 14 times smaller
than that of the U.S.
Overall
energy consumption
has been growing at a rate of
6.5 percent,
well above the rates for the U.S. and Europe.
Table 2 shows on a percentage
basis the Brazilian
energy
market in 1970.
56
Josd M. MiccoZis
Table 2.
Uses
Brazilian
energy
market
% of Total
Agriculture
Fishing
Mining
Industrial
Steel
Metallurgy
Minerals
Mechanical
Chemical
Textile
Paper & Pulp
Food
Others
5.79
0.19
0.30
34.35
7.70
1.82
5.43
1.48
3.02
1.82
2.33
8.98
1.75
in
1970
(2).
Uses
% of Total
Transportation
Highway
18.25
Railway
1.08
Oceans and
Rivers
1.57
Air
0.98
0.17
Others
Construction
Commerce
Services
Government Services
Domestic
12.50
Urban
Rural
18.71
22.03
0.67
2.32
0.65
2.48
31.22
In 1975 Brazil
agreed with West Germany to purchase a
complete nuclear
fuel cycle.
The terms of the purchase
include
a joint
effort
to mine Brazil's
uranium and thorium
resources,
and to design,
build,
and test eight nuclear
power plants
totalling
10,200 megawatts.
It also includes
transferring
the technology
for the construction
of both
urarli-lm enrichment
and fuel reprocessing
plants
in Brazil.
The reactors
will
be light
water reactors,
using enriched
uranium as the fuel.
The Brazilian
Atomic Energy Authority
has set a target
of 70,000 nuclear
megawatts to be installed
by the turn of the century.
Alternative
Solar
Energy
Production
Technologies
Energy
Brazil
occupies
a vast region in which solar energy is
conveniently
available.
The country
encompasses an area of
more than 8.5 million
square kilometers,
almost entirely
located
between the Equator and the Tropic of Capricorn.
Solar energy is now considered
capable of playing
an important supporting
role on a medium- and long-term
basis in
the country's
energy matrix,
since the economic comparisons
that a!,ways made it unwise to convert
solar energy into
other forms of energy are no longer valid.
Although
several
scattered
efforts
to develop solar energy applications
in
Brazil
occurred
before
1973, by the end of that year the
first
comprehensive
government
attention
was devoted to
solar
energy with the creation
of a task force,
within
the
government's
planing
structure,
to examine all possible
Atternut-bve
Energy TechnoZogies
in Brazil
57
alternative
energy technologies
that could be applicable
to
the Brazilian
situation.
Solar energy was chosen as one of
the potentially
successful
candidates.
A national
solar
energy research
and development
plan was prepared
by the
task force,
and following
approval
by the proper authorities
resources
were allocated
and the plan's
implementation
started
by the end of 1974.
The underlying
philosophy
of the R&D plan for solar
energy in Brazil
took into account several
different
factors:
socio-economic
and natural
resources
particular
to the country,
which differ
from region
to region;
the
immediately
available
and easily
improvable
capability
of
local
research
institutions;
and what appeared to be the
state-of-the-art
of solar energy conversion
techniques
at
that time.
The plan's
budget for the period
1974-75 was 2.5 million
dollars,
in 1974 prices,
and focused mainly on solar energy
conversion
processes
that could be classified
as "lowtechnology"
applications.
Some basic research
was also
funded.
The main projects
within
the plan are listed
below.
A list
of participating
research
institutions
is
available
upon request.
The first
Brazilian
national
solar energy research
and development
plan consists
of the
following
projects:
Collectors.
Design,
construction
of prototypes,
testing
and evaluation
of flat
plate
collectors.
Determination
best geometry,
materials,
configuration,
and costs.
of
Dryers.
Design construction,
and testing
of small-scale
tropical
fruit
dryers.
more
Identification
of fruits
suitable
to the process.
Establishment
of quality
control
standards
for the end products.
Development
of prototypes
of industrial
scale agricultural
product
solar dryers
(e.g.,
cocoa, coffee,
maniac, wood, beans, and grain)
equipped with air preheaters
and functioning
by natural
05;
forced convection.
Studies
of mass and heat transfer
processes
involved
in solar drying.
Distillation.
Development
of a small-scale
still
for
brackish
underground
water.
Testing
of different
systems
and construction
of modular prototypes.
Feasibility
studies
for the oonstruction
of a middle-sized
solar still
to serve
small co-unities.
Cost-comparison
studies
of solar distillation
versus rain collection
and water purification
methods.
58
Jose' M. Miccozis
Refrigeration.
Construction
of an absorption
cycle solar
icemaker,
for application
in food preserving.
Technical
and economic evaluation.
Construction
of a demonstration
scale solar-cooled
food storage
facility
for semi-perishable
agricultural
products
such as potatoes,
onions,
garlic
and
other vegetables.
Bioconversion.
Experimental
tests
and feasibility
studies
of different
processes
for high water content
bio-mass
fermentation.
Anaerobic
fermentation
of solid
organic
wastes.
Feasibility
studies
of the pyrolysis
of organic
matter.
Solar architecture.
Design,
construction
and testing
of
low-income
housing units,
exploring
locally
available
materials
and architectural
concepts
suitable
to prevailing climate
conditions.
Incorporation
into the architectural
design of solar energy devices like water heaters,
organic
waste recyclers,
and solar stills.
Thermal engines.
Development
of a low-power
Rankine cycle solar engine for water pumping.
commercially
available
systems.
Photovoltaics.
Basic research
on phenomena
actions
of photons with matter.
Investigation
cells
with polycrystalline
structures.
organic
fluid
Testing
of
of the interof photo-
Energy storage.
Study of low temperature
energy storage
systems.
Research on heat transfer
processes
for water,
air,
rocks, plastics,
metals,
and eutectic
salts.
Solarimetry.
Rehabilitation
of existing
solar energy data.
Mapping of monthly
total
solar radiation
for the whole
territory
of Brazil,
by means of both direct
and indirect
data.
Establishment
of a new solar radiation
measurement
network of ground stations.
Ocean Thermal Gradients.
Harvesting
the solar energy
collected
and stored in tropical
waters
(Ocean Thermal
Energy Conversion:
OTEC) has received
considerable
attention
in the U.S. in the past few years.
A heat engine can
operate by means of using warm surface
water as a heat
source,
and the much colder
deep ocean water as a heat sink.
The idea is not technically
new and it seems feasible
with
current
technology.
Some regions
along the Brazilian
coast
present
optimum natural
conditions
for the development
of
this process.
In the Cabo Frio region of the Brazilian
coast,
east of Rio de Janeiro
along the Tropic of Capricorn, the deep sea water,
cold and nutrient-rich,
comes to
AZtemative
Energy Technologies
in Brazil
59
An attempt
will
be made to generate
power
the surface.
related
to an aquaculture
research
prousing this water,
ject which is also using this water for ice making, and as
a basic feed for its marine cultures.
Wind Power.
Although
windmills
have been traditionally
used in some regions
of Brazil,
regional
wind patterns
are
A small-scale
research
project
aimed at
not well known.
developing
a vertical
axis (Savonius)
wind conversion
system
is underway.
Solar Energy Perspectives.
If solar energy is going
to play any role in the Brazilian
energy picture,
then the
Northeast
would have to be considered
as the most suitable
region
for successful
solar energy applications.
occupying a considerable
portion
of the Brazilian
territory,
the
Northeast
includes
all of the country's
semi-arid
region,
where droughts
are common, heavy rains occur within
a very
short period
of time, and the average daily
temperatures
are very high.
Furthermore,
even measured by Brazilian
the Northeast
is an underdeveloped
region.
standards,
The
average annual per capita
income is well below the national
level,
birth
and mortality
rates are rather
high,
the
economy is basically
rural
and based upon an outdated
ownership
and land use model.
The national
solar energy
research
and development
plan took into consideration
all
these factors.
That is why several
projects
were recommended
and funded:
solar stills
for areas where an adequate source
of fresh water is not available,
and where it is not relevant to make economic comparisons
with technologies
that
cannot even be applied;
crop dryers designed
so that a
variety
of crops, maturing
at different
times,
can be dried
in sequence by use of the same equipment;
solar refrigeration intended
primarily
for food preservation
and for the
storage
of seasonal
and perishable
crops; water pumping
with small solar engines for irrigation
purposes;
and
others.
The program has thus far failed
to accomplish
the
development
of most of the prototypes
intended
for an
eventual
industrial
production
and commercialization
in
Brazil.
This failure
cannot be attributed
merely to the
fact that only two years have passed since the initial
projects
were funded and that,
naturally,
some projects
have been more successful
than others.
It was also in part
due to a lack of coordination
among several
government
agencies which could have taken a mud1 more active
role in
the process.
For example,
the federak
housing agency in
charge of financing
thousands
of low-income
units
throughout Brazil
could have made economically
feasible
the
60
Jose’
M.
Micco~is
large-scale
production
of solar water heaters
to replace
the
widely
used, inefficient,
and electricity-hungry
electric
resistance
water heaters.
Simply by changing
current
building regulations
and by incorporating
into construction
costs
andifinancing
the added costs for the solar heater,
an
insured
market would have existed,
thus creating
an incentive for industrial
production
of solar heaters.
The
,
government
could also have provided
incentives
for the replacement
of conventional
fuels with solar energy to provide the low-temperature
process heat used in several
industries.
Bioconversion
is another
very promising
solar energy
conversion
technology.
Due to recent developments
in
Brazil
and to the considerable
attention
that it has recently
received,
bioconversion
possibilities
in Brazil
will
be discussed
as a separate
topic in this paper.
Bioconversion
Green plants
are known to be able to capture
the sunshine and convert
it into stored chemical
energy,
usually
but not necessarily
in the form of carbohydrates.
Currently,
with the stored production
of ancient
photosynthesis
-fossil
fuels
-- running
out, it appears that harnessing
the
renewable
energy stored in green plants
is not only highly
desirable
but increasingly
feasible.
The stored chemical
energy can be Eecovered
by different
processes,
at the end
of which solid,
liquid,
and gaseous fuels,
as well as
chemical
feedstocks
and other products,
can be obtained.
Brazil
is in a very privileged
position
to make use of
this natural
energy collection
process,
not only because
Brazil
occupies
a huge area where the incident
average solar
energy
is abundant,
but also due to particular
local
conditions
such as a satisfactory
combination
of appropriate
soils,
water resources
and rainfall
needed for growing green
plants.
Furthermore,
the labor-intensive
characteristics
of the related
agricultural
processes,
coupled with the
availability
of relatively
inexpensive
and arable land,
could very well match the socio-economic
requirements
of a
developing
nation
like Brazil.
It should also be noted
that due to the extensive
forest
resources
that are available in Brazil,
it is quite probable
that a number of other
possibilities
might exist
in addition
to the ones that are
mentioned
here.
However,. the possibiliti&
described
below
do represent
current
knowledge
and give a fairly
accurate
view of projects
under consideration
or presently
being
carried
out.
Altemati~e
Energy TechnoZogies
in BraxiZ
61
Fuels and Chemical Feedstocks
from Sugar Cane.
Since
early in the sixteenth
century,
sugar cane has been one of
the major Brazilian
agricultural
products.
In 1972 Brazil
became the worl..J.'s largest
producer
of cane sugar (6 million
with about one-third
of domestic
production
tons per year),
going for export.
High prices
in world markets during
19731974 stimulated
f,urther
increases
in production
to a level
of 7.2 million
tqns per year.
--Ethyl
Alcohol--Among
the many by-products
of the sugar
cane industry,
ethyl
alcohol
or ethanol
has been traditionally obtained
in Brazil
by fermentation
of surplus
molasses
from sugar manufacture
and used as an adjlunct
in gasoline.
The percentages
of alcohol
in gasoline
have varied over the
years (2 to 15 percent
alcohol
content);
ethanol
has served
as a regulatory
device to maintain
a reasonable
and profitable level of sugar cane production
in the country.
Whenever there was a surplus
of sugar, the production
of alcohol
was increased
and greater
amounts added to gasoline.
.
Over the past several
years ethyl alcohol
production
in
Brazil
has oscillated
between 570 and 700 million
liters
per year (150 to 185 million
gallons),
of which about half
was anhydrous
(for mixing with gasoline).
Measures recently
taken by the federal
government
have
insured
the sugar cane growers that whenever it is possible
to obtain
ethanol
or sugar from a sugar cane crop, both
commodities
would be worth the same; that is, the price
is
set so that the 90 kilograms
of sugar or the 30 liters
of
ethanol
obtainable
from 1 ton of sugar cane yield
the same
monetary
return.
Those measures are clearly
designed to stimulate
ethanol
production
in the country,
and are tied to an ambitious
plan to have 75 percent
of the total
liquid
fuels
consumed in Brazil
replaced
by ethyl
alcohol.
According
to
one scenario
being considered,
domestic petroleum
production
will
increase
frc& the 1975 level of 200,000 barrels
per
day to 500,000 bpd in 1980, and to 1 million
bpd in 1990,
remaining
at that level
to the year 2000.
In the first
phase (to 1980) ethyl
alcohol
would be blended with gasoline
at a 1:lO ratio,
which does not require
any major adjustments to automobile
engines.
In the following
decade,
engines would be gradually
adapted to use pure alcohol
so
that by 1990, 50 percent
of all liquid
fuels could be replaced by ethanol.
By the year 2000, ethyl
alcohol
would
account for 75 percent
of the country's
consumption
of
liquid
fuels
(2).
62
Jose' M. Miccoitis
In order to achieve the goals set forth
by the plan,
about 400 new distillaries
would have to be operational
by
1980, 1,150 by 1990, and 3,500 by 2000 (4).
This assumes
distillaries
of an average capacity
(12,500 liters/day)
and harvesting
periods
of 160 days.
It should be noted,
however,
that of the approximately
200 sugar mills
that are
responsible
for the current
sugar production
in Brazil,
more than 50 percent
are operating
at only marginal
profits.
Ethyl alcohol
has further
useful
properties.
Blended
with gasoline,
it increases
the octane rating
of the fuel,
thus reducing
the need for cyclic
hydrocarbons.
Also, as
an anti-knock
additive,
ethanol
makes the use of lead unnecessary.
As a pure fuel,
ethanol
does have a lower energy
content
per unit weight than does gasoline;
however, the
energy content
per unit weight is not the determining
factor
for the power generated
by an internal
combustion
engine.
Factors
like the heat of combustion,
thermal
efficiency,
etc.,
are imp,:-ant.
When these factors
are taken into
account for gasoline
and for pure ethanol,
it is clear that
ethanol
generates
about 18 percent
more power.
That is one
of the reasons,
by the way, that several
racing
engines use
ethanol
as a fuel.
Ethanol
consumption,
though,
is about
50 percent
higher
than gasoline
consumption
in a conventional,
unadjusted
automobile
engine.
However, for higher
compression
rates,
about 10:3., and with some pre-heating
of the air-ethanol
mixture
in the carburator,
ethanol
consumption has been experimentally
demonstrated
to be about
15 to 20 percent
higher
than that of gasoline
(5).
It
should also be noted that an ethanol-fueled
engine drastically
reduces the emission
of pollutants
such as nitrogen
oxides,
carbon monoxide,
and cyclic
hydrocarbons.
Brazil
is considering
not only
for ethanol.
Cassava, or maniac,
a
everywhere
in the country,
has been
able government attention,
and will
sugar cane as the source
root crop that grows
the focus of considerbe discussed
later.
Just recently
the Brazilian
government
approved about
US$500 million
for implementation
in 1977 of a National
Alcohol
Program that includes
both sugar cane and maniac as
raw inaterials
for ethanol
production.
--Other
Fuels--Besides
ethanol,
a number of other fuels
could be obtained
from the sugar cane biomass.
By means of
thermal
conversion
processes,
the fiber
could either
be used
as a boiler
feed (combustion),
which is currently
done in
Brazil
and which makes the sugar cane industry
practically
self-sufficient
in energy,
or converted
to synthesis
gas by
means of processes
such as Bailie,
Kellog,
Koppers-Totzek,
1
Alternative
Energy Technologies
in BrmiZ
63
Syngas, Purox, etc.
Synthesis
gas is a mixture
of carbon
monoxide and nitrogen
with smaller
proportions
of water,
carbon dioxide,
ethane,
and propane.
It can be directly
used as a low BTU gas, or processed
into a pipeline
gas
(methanation),
and/or methanol
and a motor fuel by processes
such as Fischer-Tropsch,
Mobil.
Also, by microbiological
conversion
techniques,
similar
to those used for the fermentation
of molasses or cane juice
into ethanol,
substitute
natural
gas (SNG) could be obtained
by fermenting
the
bagasse (5).
--Chemical
Feedstocks--A
number of chemical
feedstocks
are obtainable
from sugar crop biomass.
Through fermentation the sugar cane juice,
or the surplus
molasses,
can be
transformed
into acetone,
isopropanol,
butanol,
butadiene,
ethylene,
acetaldehyde,
acetic
acid, xanthane,
and several
other polymers
(7).
Some of those products
would require
an initial
fermentation
and additional
chemical
synthesis
steps,
for which the technology
is already
available.
Furthermore,
recent advances in the field
of microbial
biosynthesis
that make possible
the production
of a specific
organic
chemical
by microbial
fermentation
could indeed
shorten
the additional
chemical
synthesis
steps that are
currently
necessary,
dramatically
changing the economics
for the production
of those products.
--Other
Products
by Fermentation--Sucrose
is an excellent raw material
for fermentation.
Together
with starch
and starch-derived
sugars it served well as a feedstock
for
the microbial
production
of several
chemicals.
During the
post-World
War II petroleum-glut
period,
many of these fermentation
processes
were discontinued
in lieu of lower-cost
petroleum-based
chemical
syntheses.
However, some products,
such as citric
acid, penicillin
and other antibiotics
not
readily
derived
from petrole-um,
continued
to use fermentation as their
commercial
source.
Since raw materials
account for at least
60 percent
of the total
cost of the
average fermentation
end product,
the low cost of petroleum
led to the demise in 1945-1973 of many fermentation
proFesses.
It
must also be mentioned
that for the most part the
*
current
reconsideration
of fermentation
does not yet take
into account the new capabilities
of present-day
microbiology.
Rapid, automated,
systematic,
genetic
strain
improvement
skills
can now replace
improvement
efforts
that
were formerly
the province
of chance.
Achievements
in the
antibiotic
field
stand as evidence
of such claims.
64
Jose' M. Miecozis
While an in-depth
study is needed to determine
for
local
Brazilian
conditions
the best of the many specific
product
opportunities,
numerous candidates
may be suggested
for consideration:
amino acids,
such as L-lysine,
__
L-methionine,
L-tryptophan;
vitamins
like vitamin
C
(ascorbic
acid),
vitamin
B12 (cyanocobalamin),
vitamin
B2
(riboflavin);
food chemicals
such as acetic
acid, citric
acid, nucleotides,
and microbial
polysaccharides;
antibiotics
like penicillins
G and V, tylosin,
gentamicin,
and the cephalosporins
(s>.
--Ethyl
Alcohol
from Maniac (Cassava) --Maniac
is a
somewhat bushy herb which grows everywhere
in Brazil.
large,
The roots,
which have the appearance
of sweet potatoes,
yield
a starchy
product.
Although
2.1 million
hectares
were planted
in 1973, no commercial-scale
operations
have
been established.
All the maniac is grown in small farming units,
manually
planted
and harvested.
The starch
content
of maniac, about 25 percent,
is far
greater
than the 13 percent
sucrose content
in sugar cane.
The result
is that a ton of maniac yields
considerably
more
ethyl
alcohol
than does a ton of sugar cane.
However, one
hectare
of sugar cane crop yields
about 50 tons of cane,
and one hectare
of maniac yields
only about 15 tons.
Maniac crops for energy purpx.ses present
problems
related to mechanization,
and to the fermentation
process
itself.
Since starch
cannot be directly
fermented,
it
requires
enzymatic
action,
and there is a lack of industrial
experience
with its processing.
Furthermore,
an overall
energy balance would probably
reflect
the fact that the
maniac industry
cannot benefit
from burning
the residues
of
its industrial
processing,
as the sugar cane industry
does
by burning
bagasse.
plans
tillary
gram
ingly
even
Anyway, maniac crops are very much within
the government
for ethyl
alcohol
production.
A pioneer
maniac disis being built,
a fairly
large maniac research
prois under way, and official
estimates
assume an increaslarge contribution
of maniac to alcohol
production,
surpassing
that of sugar cane in less than a decade.
--Socio-economic
Considerations
Brazil
has abundant land and can
tivated
cropland
at present
rates
Thus, a major agricultural
issue
make the land resource
contribute
national
and per capita
incomes.
pendent on foreign
petroleum,
for
of Ethanol
Production-continue
expanding
its culfor most of this century.
consists
of finding
ways to
mOre toward raising
Brazil
is currently
dewhich the country
has to
AZtermative
Energy Technologies
in Brazil
65
pay about 10 million
dollars
per day, and it needs to change
both the pattern
of the farm labor force,
which constitutes
a disproportionately
large component of the low-income
group,
and to modify somewhat land ownership
distribution.
Therefore,
it is not only very welcome, but necessary,
to expand
acreage for energy crops, and thus cash crops.
Estimates
indicate
that more than 200,000 people are
directly
employed by the sugar cane industry
in Brazil,
including
factory,
sugar cane farm, cane and sugar transpoktation,
and other service
workers.
Most of these people
reside
in the areas &n which they work.
Given the government goals for ethyl
alcohol
production,
and the increasingly
larger
contribution
of maniac, by 1980 about 700r000
new hectares
will
be planted,
with the corresponding
settlement of 65,000 farm worker families;
by 1990 about 9 million
new hectares
will
be added to production
and 730,000 families
will
be settled;
and by 2000 there will
be 18 million
new
hectares
and 1.4 million
families
(2).
Considering
that the growth rate of another
cash crop,
soybeans,
has been an amazing 34.8 percent
per year over
the period
1967-1972
(about 5 million
hectares),
the goals
set forth by the plan do not seem unattainable.
Much to the
contrary,
they represent
a unique opportunity
for tackling
the problem of poverty
in the agricultural
sector,
for embarking
on a land redistribution
program,
and for implementing a large-scale
credit
program for the acquisition
of
agricultural
machinery,
fertilizers,
and other modern agricultural
inputs.
Such development
offers
the possibility
of solving
most of the current
problems
caused by petroleum
imports
which are now facing
the Brazilian
economy.
The economics of ethanol
production
in Brazil
are such
that right
now its market price
is highly
competitive
with
that of gasoline.
However, there are many different
and
attractive
technologies
and products
which are alternates
to obtaining
ethanol
from sugar cane, and which should be
considered
together
with ethanol
production
for the establishment
of a more complete and efficient
industrial
complex.
A comprehensive
feasibility
study is very much needed.
Raw
material
considerations,
disposition
of by-products,
marketing of all products,
costs of different
processes,
availability
of appropriate
technologies
and overall
energy
balances
are but a few of the topics
that should be investigated.
Additional
Bio-Conversion
Alternatives.
--Charcoal-Charcoal
produced by carbonization
of wood is widely
used
in Brazil
as a raw material
for the production
of pig iron
66
Jo&M.
Miccolis
Currently
more than 2.5 million
tons of pig iron
and steel.
and more
than 1 million
tons of steel are produced in this
manner in the state of Minas Gerais alone.
Due to the limited
availability
of Brazilian
coal and
its characteristics
(it is mainly high-ash,
high sulfur
charcoal
production
will
have to
content,
low-grade
coal),
be increased
by more than two-fold
in the next five years.
methods still
being used are quite outHowever, production
dated.
By switching
to the more modern technique
of wood
distillation,
considerable
amounts of various
by-products
from the production
of 1
could be obtained.
Folr example,
million
tons of pig iron using charcoal,
it is technically
feasible
to recuperate
146,000 tons of fuel gas, 18,000 tons
of acetic
acid,
12,000 tons of methanol,
and 70,000 tons
of tar, which add up to an additional
value of about 200,000
dollars
(10).
A fairly
comprehensive
feasibility
study encompassing
different
aspects of charcoal
production
has been conducted
Commercial
charcoal
producfor the state of Minas Gerais.
tion utilizing
more appropriate
technology
will
follow.
Furthermore,
considering
the very extensive
forest
areas
that are currently
being cleared
for agricultural
production
(more than 2 million
hectares
in the northern
state of Para
alone),
wood pyrolysis
could be of significant
value in the
overall
Brazilian
energy matrix.
--Hydrocarbon
Producing
Plants--Hydrocarbon
(rubber)
producing
plants
that naturally
grow in the Amazon basin are
being considered
as a potential
source for the direct
photosynthetic
production
of hydrocarbons.
It has been suggested
that proper manipulation
of rubber tree (Hevea) production
could control
the molecular
weight of its principal
hydrowhich has a chemical
composicarbon product,
polyisoprene,
tion similar
to that of petroleum,
thus obtaining
a renewable
HowIIfuel"
tree (11).
This is still
only a research
idea.
ever, other Euphorbia
species
(of which there are more than
3000 different
species in Brazil),
like the Euphorbia
Tirucalli,
are also potential
candidates
for this kind of chemical manipulation
and are being investigated.
--Waste Recycling--Organic
wastes might prove to be of
considerable
significance
on the local
level
for supplying
energy whenever large waste concentrations
occur within
a
relatively
small area.
A‘ project
that includes
the future
construction
of a pilot
plant
for pyrolyzing
the municipal
solid
wastes of the city of Rio de Janeiro
(5 million
habitFts)
is currently
in the stage of feasibility
studies.
Akwna-tive
Energy Technologies
in BraziZ
67
Pyrolysis
consists
of heating
organic
material
at high temperatures
for a prolonged
period
of time in the absence of
Different
combinations
of gas, oil,
and residues
oxygen.
can be obtained,
depending
on the organic
waste composition,
and on the temperature,
pressure,
and duration
of the process.
The anaerobic
fermentation
of sewage is another way of
The methane produced by bacdisposing
of organic
wastes.
teria
(typically
60 percent
methane and 40 percent
carbon
dioxide)
can provide
the energy needs of the sewage treatThe resiment plant
itself,
or be used as a pipeline
gas.
can be used as a fertilizer.
duesc unfermented
sludge,
The
process is simple;
presently
it is being used in several
locations
in the world,
and could contribute
to the overall
economic feasibility
of treating
sewage in Brazil.
Smallscale sewage digestors
are now being built
for future
larger
scale developments.
--Water Hyacinth/Fresh
Water Algae--Water
hyacinth,
or
Eichhornia
Crassipes,
is widespread
in tropical
and subtropical
regions.
Submitted
to anaerobic
digestion
it
yields
methane and carbon dioxide.
Fresh water algae,
like
Chlorella,
can grow in‘a sewage oxidation
pond, and are
readily
subject
to bacterial
attack.
Algae are also a good
protein
source,
and can be grown very efficiently,
with
productivity
greater
than that of agriculture,
although
at
somewhat higher
costs.
Experiments
with water hyacinth
and
fresh water algae anaerobic
digestion
are being conducted
in
Brazil
to determine
both their
possible
use as a fuel and as
a protein
source,
and the possibility
of growing them in
sewage oxidation
ponds.
The Hydrogen
Economy
Although
not exactly
an energy production
technology,
the "hydrogen
economy" was chosen to be included
with other
production
technologies
since its appropriate
uses can resuit in a very interesting,
almost unique,
possibility
for
Brazil.
Due to the huge waterpower
resources
available
in
Brazil,
and due to the large distances
between the major
e1ectrZ.c energy consumer centers,
hydrogen was initially
regarded
as a possible
"energy vector"
that could conveniently
solve the energy transportation
and storage problems
associated
with those conditions
which are almost peculiar
to Brazil.
However, studies
mentioned
task force
concluded
that other
conducted
in 1974 by the aboveon alternative
energy technologies
useful
hydrogen characteristics
should
68
cTose/M. MiecoZis
be regarded on a priority
basis for
and that they would be more readily
the national
needs.
a short-term
applicable
R&D program,
to satisfyiny
The National
Hydrogen R&D Program was completed
in the
beginning
of 1975.
It incorporated
two different
but interrelated
lines
of projects:
research
projects,
and technicaleconomic feasibility
studies.
The former was intended
to
foster
research
both in some frontier
areas and in some
established
areas in which the development
of Brazilian
knowhow was thought
to be necessary.
The latter
was designed
to
answer questions
concerning
the applicability
of already
available
technologies
and processes
to short-term
industrial
developments.
The Program was approved for its implementation
in a
two-year
period.
A list
of projects
follows.
A list
of
participating
research
institutions
is available
upon
The Brazilian
National
Hydrogen R&D Program consists
request.
of the following
projects:
Feasibility
Studies:
Primary Electricity
Production
for Electrolysis:
For already-developed
or under-construction
hydro-electric
power plant
sites.
Data gathering
and estimation
of water
availability
based upon historical
data series.
Availability
of electricit:
qf=nerating
equipment.
Price of electricity
projections
J-c)> n):llogen
production
by electrolysis,
both
for peak and off-peak
conditions.
Feasibility
of using
already-developed
or under-construction
power plants
for
supplying
necessary
electricity.
For non-developed
potential
waterpower
sites.
Site location.
Feasibility
of A.C. and B.C. electricity
generation
at those
sites.
Economic studies
for construction
of a power plant
specifically
designed
for providing
electricity
to a nearby
hydrogen facility.
Hydrogen Production.
duction
techniques
Study of electrolytic
and associated
costs.
hydrogen
Hydrogen Transportation:
For short distances.
Feasibility
studies
for
hydrogen pipeline
of about 200 kilometers
in
hydrogen is currently
used as a feedstock
for
industries
(S~O Paulo).
For long distances.
Development
of a computer
assessing
technical
and economic variables
of
1500 kilometer
hydrogen pipline.
pro-
building
a
a site where
different
model for
a 500 to
ALternative
Energy TechnoZogies
Hydrogen Storage:
For gaseous hydrogen.
Studies
of natural
caves and depleted
oil and gas wells,
for
storage.
For liquid
hydrogen.
Costs and processes
large quantities
of liquid
hydrogen.
in Brazil
69
formations
like
gaseous hydrogen
for
storage
of
Hydrogen Utilization:
As an industrial
feedstock.
Studies
related
to the supplydemand market of hydrogen as an industrial
feedstock.
For axunonia production.
Processes
and cost estimates
for
ammonia production
from electrolytic
hydrogen.
Economic
comparisons
with alternative
processes.
Site location
for
an armnonia plant
using pure, electrolytic
hydrogen as a
raw material.
As a fuel.
Testing
of different
kinds of pipelines
for
urban distribution
of gaseous hydrogen.
Feasibility
of
supplying
a more hydrogen-rich
street
gas mixture,
and of
introducing
exogenous hydrogen into the conventional
street
gas production
processes.
Research Projects:
Production.
Evaluation
of existing
electrolysis
installations.
Optimization
techniques.
Research on electrodes
and catalyzers.
Storage,
Metal
Processes
hydrates.
and containers
Transportation.
Material
for
able for hydrogen pipelines.
compatible
plastic
materials.
Utilization.
Industrial
Catalytic
catalyzers.
for
liquid
pipelines.
Development
storage.
Compressors
of hydrogen-
suit-
burners.
Conventional
burners.
Hydrogen-fueled
engines.
Demonstration-experimental
facility.
Construction
of a
laboratory
scale facility
for testing
hydrogen production,
storage,
utilization,
and transportation,
of different
processes,
methods,
and according
to various
end uses.
Preliminary
Results
and Future Trends of the Hydrogen
Program,
Petroleum
and fertilizer
imports
are causing an
extremely
acute balance-of-payments
problem for Brazil.
As
a natural
consequence,
a major portion
of the hydrogen
program effort
has been dedicated
to studying
the feasibility
of producing
arrmonia from pure electrolytic
hydrogen.
The
results
of several
studies
conducted
by FINEP indicate
that
ananonia production
near a hydroelectric
power plant
is
70
Jos&M.
M<ccoi?is
competitive
with the naphtha reforming
process currently
used in Brazil,
provided
that electricity
could be supplied
at a cost of about 12 mills
per kilowatt-hour
(at 1975
and provided
that the plant would produce at least
prices),
600 tons per day of ammonia (12).
Even for an ammonia
plant
located
up to 400 kilomzers
from the power station,
and thus from the hydrcgen producing
plant,
the studies
found that for an installation
producing
in excess of 1000
tons per day, the ammonia so obtained
would still
be
economically
competitive
with that from other processes.
Those conditions
(12 mills/kwh
electricity,
proximity
to a
power plant,
and ammonia production
of more than 600 tons/
day) are very much feasible
today in Brazil.
Plans are now under way for building
an industrialscale ammonia plant which,
if successful,
would constitute
a major contribution
of the hydrogen program to the country's
energy and economy picture,
and would also serve as a
catalyst
for the several-other
research
activities
being
funded by the program.
Fossil
Fuel
Alternative
Technologies
Coal Gasification.
Coal gasification
processes
are
under consideration
in Brazil.
Different
technologically
established
processes
(Winkler,
Lurgi,
Koppers-Totsek,
Otto) were examined as alternatives
for ammonia production,
and for the direct
reduction
of iron ore in steel production.
However, due to the high-ash
content
of Brazilian
coal (55 percent)
it is not at this point
clear whether any
one of those applications
will
find its way from the
feasibility-study
stage to the pilot
and demonstration
plant
scale.
Shale Oil.
Petrobras,
the state-owned
Brazilian
oil
company, has developed
the technology
for shale oil production.
A pilot
plant
(1000 barrels
per day) has been
operating
for the past several
years in Southern Brazil.
Plans were recently
announced for building
the first
commercial-scale
shale processing
plant.
The plant
is
expected to yield
51,000 barrels
of shale oil,
and 1.8
million
cubic meters (63.5 million
cubic feet)
of pipeline
gas per day.
For that production,
approximately
112
thousand tons of shale will
be necessary.
About 94,000
tons of residue
will
result.
An advanced research
project
is under way for determining
other possible
applications
for Brazilian
shale,
as well as for the residues
of shale
oil distillation
(polymers,
fertilizers,
medicines,
construction
materials,
etc.).
tion,
three
1977)
Those
AZtemative
Energ
Technologies
Energy
Conservation
?:n Brazil
Measures
Forced by the increase
in petroleum
products
consumpdespite
stepwise
price increases
during
the past
years,
the Brazilian
government
most recently
(January
enacted very tough fuel conservation
regulations.
regulations
consist
basically
of:
Gasoline
surtax of about
is to be returned
to the
any value corrections
to
50 percent
in 1976), and
as a result
of the surtax,
63 cents per gallon.
The surtax
consumer after
two years,
without
compensate for inflation
(about
no interest
paid.
Gasoline
prices,
will
be about uSS2.20 per gallon.
Similar
measure
per ton of fuel
oil.
Prohibit
Brazilian
for
oil.
fuel
automobile
cities.
The surtax
circulation
Closing
of gasoline
and holidays.
stations
will
in downtown
all
over
the
be US$21
areas
in major
country
on Sundays
The establishment
of different
working
schedules
for
industry,
coxmnerce, and government
agencies,
including
staggered
hours.
Raising
by 50 percent
the
highways during weekends.
fares
Incentives
for
Incentives
from fossil
for the use of electricity
fuels.
Incentives
systems.
for
Reduction
agencies.
Limitation
to 89 HP,
Enforcing
Fining
the
toll
the
use of alternative
expansion
by 10 percent
of power
of badly
energy
bills
to energy
of government
of government
diesel
sources.
transportation
automobile
maximum speed limit
tuned
and federal
as opposed
of public
of fuel
output
the national
owners
on state
engines.
(50 mph).
engines
71
72
Jo&M.
Surtaxing
generators.
gasoline
Restricting
Raising
MiccoZis
imports
diesel
oil
or diesel
oil-fired
of airplanes
by the
electricity
government.
prices.
Since those measures were taken quite
recently,
there
is no way to evaluate
how effective
they really
are.
Furthermore,
it is not very clear at this stage what the
meaning is of some of them which were phrased more as
recommendations
than as regulations.
However, it is fair
to assume that their
reach and potential
use open up a wide
spectrum of possibilities
for stimulating
the development
of alternative
energy technologies
in Brazil.
Incentives
for the use of alternative
energy sources
include
not only the gradual
replacement
of gasoline
and
other liquid
fuels by ethyl
alcohol,
but also the "more
efficient
use" of other natural
resources.
Notwithstanding
the possibilities
that have been mentioned,
at this point
it is not possible
to determine
definitively
which of them
will
be developed
on a priority
basis.
Some implications
for conservation
can be suggested.
Public
transportation
systems (buses) are responsible
for
75 percent
of the total
number of passengers
in the ten
largest
metropolitan
areas in Brazil.
Increasing
this
number, say to 85 or 90 percent,
could save an estimated
430 to 640 thousand gallons
of fuel per year.
Also,
increasing
the person-per-automobile
average number (now
1.35 in the largest
metropolitan
areas) to 2.0 could save
an estimated
10 percent
of the total
annual gasoline
consumption in the country.
The government
hopes to achieve
these goals not only by restricting
the use of automobiles
in downtown areas, but also by a number of complementary
measures such as creating
special
traffic
lanes for buses
and car pools,
charging
special
toll
fares for cars carrying
only one motorist,
stimulating
the production
of electrically-powered
trolleys
and of small,
low gasoline
consumption
automobiles,
increasing
parking
rates at areas near downtown, and a number of other measures.
Also, the federal
investment
in railroads
and construction
and in electrically
powered railroad
equipment
will
be substantially
increased
and the highway construction
program will
be kept at a
minimal
acceptable
level.
kChmn.ative Energy TechnoZogies
in Brazil
73
Conclusions
The possibilities
for developing
alternative
energy
technologies
in Brazil
are many and varied.
Some of them
are peculiar
to the country;
others are applicable
elseSome are fairly
well-developed
and need only be
where.
adapted or improved,
to take advantage of local
conditions.
Others,
although
very promising,
are still
a few years away
from being used on a commercial
scale.
Nevertheless,
for a country
like Brazil,
as for many
other countries,
there is no simple solution
for meeting
its
energy needs.
Different
possibilities,
though,
are not
mutually
exclusive.
They should contribute
to the overall
solution
by solvrng
some specific,
local problems.
It seems
clear that the country's
energy policy
cannot restrict
itself
to short-term
planning
in this field.
Rese:arch and
development
efforts
have to be considered
as an integral
part of the policies
being formulated
-- which,
in all fairness, cannot be said to have happened in the past.
The variety
of methods,
technologies,
processes,
and
end-products
that appear to be feasible
in the Brazilian
case, require
a very well-coordinated
and executed comprehensive approach toward the problem.
A systems treatment
is .a must.
In the first
few iterations,
a definite
answer
is not going to be found; however,
a trial
and error process
should be conducive
to a much better
solution
than the random approach which results
from focusing
attention
on
isolated
alternatives.
1
74
Josex M. Mice0 Eis
References
1.
J. Goldemberg,
3
-.
Brasil,
Ministhio
de Mir,as e Energia,
Energe'tica
Brasileira
(1970).
3.
Brasil,
Ministgrio
da Indktria
e do Comhcio,
0
Etanol
coma
Combustive1
(1975).
-
4.
J. Gomes da Silva,
in Problemas de Energia no Brasil,
108 (Instituto
de Pesquisas,
Estudbs e Asse%oria
do
Congresso,
BrasTlia,
1976).
5.
U. E. Stumpf,
in Acucar e Alcool
Econ?jmico do Brasil,
155-(Coperflc
1976).
-
6.
Conference
Proceedings:
T-39 (Battelle
Columbus
7.
Ibid.,
8.
W. Amon, Jr.,
personal
Berkeley,
California,
9.
J. Comes da Silva,
Intersciencia
A, 33 (1976).
Matriz
um Grande Projeto
Rio de Janeiro,
"Fuels from
Laboratories,
Sugar Cropsam
Ohio, 1976).
T-15.
interview,
November,
op.cit.,
--
Cetus
1976.
Corporation,
114.
10.
J. I. Vargas
no Brasil,
11.
M. Calvin,
..~ .
(University
12.
Brasil,
Financiadora
de Estudos e Projetos,
uma alternativa
para
a
ProduCzo de Amonia,
-Eltrolise
da agua, 28 (outubro
1975).
e J.
T. Veado,
in - Hydrocarbons
of California,
in Problemas
de Energia
via Photosynthesis,
Berkeley,
1976,).
19,
Estudo
agartir
de
da
Alternative
Energy Technologies
in Brazil
Bibliography
Energy for Rural Development,
Technologies
for
Alternative
National
Academy of Sciences
Renewable Resources and
Developing
Countries.
1976).
(Washington,
D.C.:
"Energia
no Brasil."
Goldemberg,
J., -m
et al.
To be
published
by the Academia de Cisncias
de Sso Paulo
(Sgo Paulo, Brasil:
1977).
Miccolis,
J.M.F.
em Energia Solar
Brazil:
1974) *
"Programa
no Brasil."
Miccolis,
J.M.F.
sobre a Economia
Janeiro,
Brazil:
et al,
"Programa de Estudos e Pesquisas
-m
FINEP (Rio de
de Hidrogsnio
no Brasil."
1975).
Proceedings
conversion,
1976).
de Pesquisas
e Desenvolvimento
FINEP (Rio de Janeiro,
of the International
Symposium on Energy BioJuly 5-9, 1976, Campinas (Sso Paulo, Brazil:
5
Wind Energy Conversion
in India
Sharat K. Tewari
It appears that practically
no serious attempt
was made to utilize
wind energy in India until
the
1950s.
While wind power has been utilized
in
propelling
and guiding
ships and boats, for several reasons no other utilization
of wind energy
could develop.
One reasonprobably
was the ready
availability
of animate energy derived from the
m-uscl~ poser of draft animals.
Rural India has
been utilizing
bullock
work in transportation,
in
grinding
of food grains,
extraction
of oil from
seeds, in preparing
lime mortar used in building
construction,
for drawing water from wells,
etc.
With its large rural population,
India has been
significantly
drawing upon manual labour also in
several domestic and agricultural
tasks.
Only
during the last two to three decades have electricity and oil been able to make an entry into the
rural energy scene.
However, conventional
forms
of energy, both animate and inanimate,
suffer
from
some drawbacks.
Energy obtained
as human and bullock work is expensive and reliance
on this form
of energy decreases the opportunities
for a more
productive
alternative
activity.
Bullock work cannot meet the additional
energy needs simply because
the number of bullocks
have not been increasing.
Electricity
drawn from the grids is available
in
only one-third
of the half a million
villages
and
it will
be quite some time before all the villages
obtain electricity
under the conventional
rural
electrification
schemes.
Utilization
of oil for
energy generation
causes an additional
drain on
foreign
exchange resources
apart from having the
problems of maintenance
and repairs
in rural areas.
From this very brief
description
of the Indian
energy scene, it could be inferred
that there is a
77
78
Shmat
K. Tewari
1
need for deriving
energy from alternative
sources
I will
attempt to estabfor use in rural
India.
lish here that energy derived from winds is highly
appropriate
for use in some applications
and deserves serious consideration.
ing
last
this,
for
the
In this paper the approach adopted for designwind energy utilization
in India during the
From
25 years will
be briefly
mentioned.
we shall proceed on to a systematic
analysis
obtaining
a rightful
place for wind energy in
energy scene of rural
India.
In 1952 a wind power sub-committee
was constituted
under the Council of Scientific
and Industrial
Research, New Delhi.
This committee recommended conducting
wind surveys at several locations,
adapting
a suitable
windmill
originally
taking steps
developed abroad, and simultaneously
The anemoto design a windmill
indigenously.
graphic records of wind velocity
were being obtained at the several installations
of the India
Meterological
Department.
Some of these records
were studied and processed from the wind-energy
point of view.
This analysis
was conducted at the
National
Aeronautical
Laboratory
and helped in
characterizing
locations
in terms of annual and
monthly average wind speeds, velocity-duration
curves, frequencies
of unacceptably
low wind spells,
etc.
About 160 water-pumping
windmills
of the
Southern Cross type were imported in the late
fifties
and installed
at several locations
in the
country.
During the 1959-64 period,
the windmill
WP-2, indigenously
developed at the NAL, was produced in batches and about 80 of these were erected for field
trials
at several places in the
country.
This windmill
has 12 blades mounted on
a 5-metre diameter rotor wheel.
The cut-in
speed
is 8 kph (kilometres
per hour) and the furling
speed is about 40 kph.
A few electricity
generating windmills
such as 'Algaier,'
'Dunlite,'
and
'Elektro'
were also tested for performance
in
Indian winds.
By 1966 wind power activity
was
closed down at NAL.
What were the achievements
of this programme,
which lasted just over a decade and operated at
full
steam for about 3 years?
First
of all,
it
was established
through this programme that windmills
are acceptable
in the tradition-bound
sociif only the wind
ety of rural
India.
Therefore,
.
&k.d Energy Conversion
in India
79
energy could match with other alternatives
in terms
its utilization
would not present
of the economics,
Another achievement to the
any social problem.
credit
of this programme was to map out wind energy
distribution
for some parts of the country.
It
was shown that the annual average wind speeds at
most places in India range from 9 to 17 kph.
However, no serious effort
could be made under this
programme to identify
windy locations
apart from
what had been indicated
from the normal meteorological data.
The optimism of cheap electricity
from the
grids and a lack of proper appreciation
for renewable sources of energy,along
with some other factors had led to the closure of the NAL wind-power
programme by 1966.
The oil price hike after
1973
has been responsible
for the restarting
of the
wind energy studies
and a fresh look is being
taken at this alternative
source of energy.
Since
then the following
studies have been initiated.
of reports
from
1. After the publication
the National
Research Council of Canada, a few
groups in India,
appreciating
the advantages of
the Darrieus
rotor,
began constructing
its prototypes.
At NAL attempts were made to energize
a
commercially
available
centrifugal
water pump
with the help of a 4.5-metre
Darrieus
rotor.
However, even with the addition
of two large Savonius rotors
co-axially
mounted with the Darrieus
rotor,
it was not possible
to obtain the starting
torque required
by the pump. At the Bharat Heavy
Electricals
Ltd.,
Hyderabad and at the Indian
Institute
of Technology,
Madras, projects
have
been initiated
for developing
a Darrieus
rotor for
generating
electricity
at around one- to five-kW
Another programme at NAL is dicapacity
levels.
rected towards s,tudying a parallel-blade
variation
thus &wiating
from the
of the Darrieus
rotor,
usual catenary s.hape selected
earlier.
2. Another line of development has followed
the route of the Savonius rotor.
The Application
of Science and Technology in Rural Areas (ASTRA)
Cell at the Indian Institute
of Science, Bangalore
has developed a modified
design of the Savonius
rotor for field
trials
in a village
near
Bangalore.
A 4 x 3-metre prototype
has been fabricated
from wood, iron'wires,
and cloth and it is
proposed to couple this with an improvised
dia-
80
Sharat
K. Tewxri
phragm water
pump.
line of experiment
has gone in
3. The third
the direction
of the Cretan Sail Windmill.
Such
windmills
utilize
relatively
cheaper materials
for rotor construction
and provide opportunities
for utilizing
low wind speeds, and at the same
time remain cost competetive.
A sail windmill
having a rotor diameter of 10 metres was erected
at Madurai in 1974 to pump water with the help of
a reciprocating
pump. Recently designs utilizing
efficient
sail wings have been finalised
for a
windmill
having a lo-metre
diameter and it is proposed to match it with a rotary
pump in order to
minimize energy losses in transmission
mechanisms.
In addition
to these, a few other design and
development programmes are being carried
out, but
the types mentioned are quite typical.
It should
be appreciated
that while the design and development of hardware is essential,
it is equally
necessary to study all related
aspects of wind energy
in a systematic
manner so that an appropriate
policy
could be formulated
at the national
level
for promoting wind energy utilization.
Answers
to the following
questions
are needed:
where utiliza1. Which are the applications
tion of wind energy could be recommended on the
basis of its bejn: more appropriate
than other
alternatives.
2.
as given
locations
and
Are wind speeds and associated
durations
by a velocity-duration
curve for various
in India adequate for the applications;
3.
energy.
What are the economic
implications
of wind
The most common method of using wind energy
Electricity
is by conversion
into shaft work.
may be derived from shaft work simply by coupling
This
a suitable
generator
to the windmill
shaft.
procedure does not involve
any serious losses in
energy availability,
and overall
efficiencies
as
high as 40% have been obtained
for converting
Therefore,
such of
wind energy into shaft work.
those applications
that depend on shaft work can
be readily
served through wind-energy
conversion.
Applications
such as water pumping, grinding
of
Wind Energy Conversion
food grains,
are typical
utilized.
in India
8i
and other agro-industrial
activities
of those for which wind energ] can be
Wind-energy
conversion
into electricity
through shaft work could be examined on the basis
of the mode of utilization
of electricity
in Indian
villages.
In those villages
where electricity
has
been made available,
it is a common feature
that
over 80% of its consumption
is in energizing
irrigation pumpsets.
The rest of it is consumed in
domestic and street
lighting.
The use of electricity
in cooking,
space, and water heating i:;
practically
non-existent.
From this analysis
it would appear that windenergy conversion
could be considered
primarily
for
providing
mechanical
work, and if electricity
is
generated through this conveirsion,
it is ultimately going to be used mainly for generating
mechanical work also.
We would now examine-the
appropriateness
of other alternatives
vis-a-vis
wind
energy for generating
mechanical work.
Prof. Roger Revelle has calculated
that about
9% of the total
energy used in rural
India is obtained from the muscle work of human beings,
and
another 14% is contributed
from bullock
work.
Even though only l/lOth
of the total
bullock
time
is utilized
in raising
water, nevertheless
their
contribution
amounted to as much as 40% in terms
of water lifted
for irrigation
in 1971. As mentioned earlier,
the bullock
population
is not growing and the incremental
energy available
from this
source is therefore
not likely
to be significant,
except for the fact that there is room for more
efficient
utilization
of the existing
bullock
force.
It
may be stressed here that energy needs are
going to increase,
especially
for irrigation
facilities
required
to support medium and high energyintensive
methods in future,
in the place of the
current
practice
of subsistence
agriculture.
The cost of energy derived from manual labour
in tasks like lifting
water from wells and canals
is estimated
to be about 20 times higher than the
commercial price for electricity.
Perhaps human
beings are not best suited for such of those tasks
82
Shmat
K. Tewari
$hat could be carried
out by utilizing
inanimate
energy.
It is a known fact that a prerequisite
for
improving
the quality
of life
of human beings is
to avoid excessive
energy dissipation
that takes
place in the manual labour deployed,
for instance,
in water lifting.
Therefore,
a reduced dependence on this source of energy should be the focal
point of an energy utilization
study like ours.
India is currently
meeting two thirds
of its
consumption of oil from imports.
After the OPEC
price hike, curbs were introduced
on non-essential
uses in order to control
the demand. However, the
demand is bound to increase in the near future
from the fertilizer
industry
and transportation
sector.
It is estimated
that import requirements
would be somewhat reduced with the availability
of
oil in the near future
from offshore
wells.
Even
then, it would be desirable
to substitute
alternatives such as electricity
for oil in applications
such as irrigation
water pumping.
In fact a trend
indicating
a preferential
use of electricity
in
place of oil for this application
was shown to
exist even before the oil price hike.
The electricity
is preferred
invillages
mainly because
it calls
for less initial
capital
from a consumer.
Use of diesel-powered
pumpsets requires
the availability
of technical
skill
for maintenance and
repairs,
which is not easily
available
in remote
areas.
Also it is estimated
that the unit cost of
electricity
obtained by smaller diesel
generators
is four to five times higher than the price of
electricity
purchased from the grids.
This also
explains
the preference
for centrally
generated
electricity
in comparison with the alternative
of
using diesel pumpsets or generators.
Of the half-million
villages
in India,
about
one third
have been covered under the rural electrification
schemes.
Transmission
and distribution networks have to be set up for providing
electricity
to the remaining
villages
and this will
take time.
The villages
that have been electrified so far are those that happened either
to be
located near the larger
load centres of cities
and
towns or were large enough themselves
to warrant
their own generating
facilities.
This is clearly
indicated
from the statistics
of rural electrirication,
which show that about 92% of the villages
with population
exceeding 10,000 were electrified in 1971 whereas only 12% of the villages
with
Wind Energy Conversion
population
under
villages
constitute
of villages.
It
required
to take
smaller villages
in India
83
500 had this facility.
Smaller
over 60% of the total
number
is likely
that more time would be
electricity
to far-flung
remote
under the conventional
practice.
Apart from the time factor,
supplying
electricity to distant
villages
results
in diseconomies
to the system.
Energy losses in transmission
are
significant
and load factors
are rather
low - just
about 10%. The rural electrification
programmes
have the Goyernment's
support on account of social
priorities
but probably
there are other ways of implementing
this programme, apart from the current
practice
of distributing
electricity
generated at
at large power stations.
Suggestions
have been
made for supplying
electricity
to a group of villages from aerogenerators
installed
at the neighboring windy sites.
We may like to examine the
economics of such a possibility.
The data on rural electrification
in India
show that a 250kw installed
capacity
is adequate
for a village
with a population
under 1000.
Under
the current
practice
a distribution
line has to be
run to each village
and this distance
has been
found to be about 8 kilometers
on the average.
For
remote villages
yet to be electrified
this distance
would surely be longer.
The cost for laying down
these lines over a given distance
and the cost of
transformers
and switchgear
can be calculated.
One
could add to this the bought price of electricity
at the last point on the main network to arrive
at
the actual cost of the electricity
supplied
in a
village.
On the other hand one could assume a
reasonable
capital
cost for a well designed and
mass-produced aerogenerator
and arrive
at its
annual cost by making assumptions
concerning
the
cost of its spares and maintenance.
It has been
shown on the basis of a comparison of the cost
factors
involved
with aerogenerators
on one hand
and the existing
practice
of rural electrification
on the other that the former could be economically
competetive
in many villages
yet to be electrified.
However, the optimised
cost-effective
aerogenerator
is yet to be demonstrated
in actual
practice.
Having examined the conventional
sources of
energy we may now proceed to study the availability
of shaft work or electricity
from non-
84
Sharat
K. Tewari
conventional
alternatives,
including
renewable
ones.
Some of the alternatives
are: (1) Solarenergy utilization
directly
for water pumping and
for generating
electricity
or shaft work; (2)
liquid-piston
pumps operating
on bio-gas combustion,
or internal-combustion
engines operating
with
bio-gas;
(3) water wheels converting
the energy
of river
flow into shaft work.
These are the promising examples.
Out of these the water wheels can
be operated only at those very few villages
that
have the advantage of a rapid flowing
river
or a
stream and we shall not consider
this possibility
in c,ur analysis
here.
Solar pumps have been designed to pump water
from wells with
water levels as deep as 30 metres.
Such pumps utilize
flat-plate
collectors,
an evaporator,
a condenser,
a few valves,
and a couple
of water chambers lowered into a well.
The capital
cost of such pumps is rather high today as compared
to the cost of water-pumping
windmills.
The
energy-conversion
efficiency
of solar pumps is
several times lower than the efficiencies
of some
of the well-designed
windmills.
Solar-energy
conversion
resulting
in shaft
work or electricity
through a turbine
and generator has also been suggested for use in rural
areas.
The costs and efficiencies
of these schemes are
comparatively
less favourable
than windmills
and
aerogenerators.
In addition
the complexity
is
much higher in such conversions
of solar energy
while wind-energy
conversion
remains relatively
simpler.
Cow dung and other animal wastes are currently dried and burnt for deriving
heat mainly for
domestic cooking in villages.
Conversion of such
waste material
into bio-gas
through anaerobic
fermentation
and using the gas so collected
has been
shown to be a more efficient
technique
from the
point of view of energy availability.
It has also
been suggested that bio-gas could be burnt in an
internal-combustion
engine to obtain shaft work.
Alternatively
it could be burnt in a liquid-piston
pump for direct
pumping of water.
Two issues
confront
us here.
The first
concerns the limited
availability
of bio-gas and the other its priorities
in utilization.
It is estimated
that with a
75% efficiency
in dung collection,
the bio-gas
that can be generated in Indian villages
can meet
Wind Energy Convemion
i,z India
85
three fourths
of the domestic energy requirements.
Since bio-gas replaces dung cakes that are being
used currently
in cooking and water heating,
the
bio-gas must first
satisfy
these requirements.
unless additional
availability
of bioTherefore,
gas is assured from sources other than cow dung,
and this is yet to be considered
on a serious
it would not be possible
to recommend biobasis,
gas conversion
into mechanical work.
One must
also consider
the thermodynamic
losses in energy
availability
involved
with the energy conversion
from bio-gas into shaft work or electricity.
it appears that wind-energy
conTherefore,
version
into shaft work or electricity
for use in
rural
India is the main contender provided
the
energy availability
from winds can be matched with
demand for energy in the applications.
We may now
consider
these aspects.
The annual average hourly mean wind speed is
generally
used as the parameter for depicting
the
wind energy potential
for a particular
location.
An analysis
of the available
wind sperG data has
indicated
that at most places in India this figure
falls
in the range of 9-17 kph.
While designing
a
windmill
or aerogenerator
one picks up a rated wind
speed at which full
rated power is generated.
The
rated wind speed is normally
higher than the annual
average speed though th number of hours of operation at full
output arc reduced.
For instance,
if
one selects
a rated windspeed as the one to maximize energy availability
during a year with a
typical
aerogenerator,
then the rated speed would
be found to fall
in the range of 20-25 kph corresponding with the range of 9-17 kph of annual average speed.
The number of hours during any year
when wind speed would equal or exceed this rated
wind speed at a given place would probably
fall
in
the range of 1000-2000 hours.
The question
then
arises whether this many hours of operation
are
acceptable
and also whether wind energy is equitably distributed
among the 12 months, or even
better,
if it is concentrated
in the months from
November to March, as these are the months maximum
irrigation
is required.
Unfortunately,
it is just
the other way about with winds in India.
The
three months of May, June, and July account for
about one half of the annual energy availability.
Wind speeds during November to March are relatively
weaker.
Therefore,
under such circumstances,
the
86
Sharat
K. Tewari
rated wind speeds might turn out to be just about
the same as the annual average wind speed.
Under
these circumstances,
is the utilization
of wind
energy economical or even feasible?
There
speeds for
fact that
tely larger
produce a
is no difficulty
in utilizing
low.wind
wind-energy
conversion,
except for the
low wind speeds require
a proportionadiameter for the rotor in order to
given amount of power.
Now the important thing is to ascertain
the size of windmill
required
and then examine whether this size
happens to be too big to be handled in large
numbers in the villages.
It was mentioned earlier
that about 40% of the water lifted
in 1971 from
wells was based on bullock or human work.
The
statistics
indicate
that on the average an open
well irrigates
an area of over 1 hectare.
Using
the data for water requirements
per season, we
find that about 0.75 hectare-metre
of water is
normally pumped during a four-month
season.
Assuming the number of hours of operation
of a windmill
per season as 500, we find a likely
pumping rate of
70-100 litres
per minute coupled with a typical
head of 10 metres.
This pumping could be obtained
from windmills
having a rotor diameter not exceeding 12 metres.
This is not at all a difficult
situation.
The important
concern then is the economics
of wind-energy
utilization.
It should be noted
that the worth of utilizing
an easily available
source of energy such as winds in a rural commun.ity is much higher than indicated
in terms of
commercial factors.
Availability
of energy at the
right time in desired-quantities
could catalyze
developments that were hitherto
just not possible.
Indirectly
in commercial terms, the high value
and price of agricultural
production
more than
justifies
relatively
insignificant
cost of water
pumping.
It is expected that the total worth of
wind energy would be high despite some diseconomies-in its application,
as this would be adequately compensated by overriding
social benefits.
In conclusion
it may be stated that windenergy conversion
has attractive
possibilities
in
India and for some applications
like water pumping it is perhaps the most appropriate
alternative
among the several choices available
or likely
to
be available
in the future.
On account of low
Wind Energy Conversion
in India
cash surplus in villages
it might be necessary for
the Government to support wind-energy
utilization
in the same manner as the programmes of rural elecand the economics of these two altertrification,
In addition,
the
natives would be comparable.
technology
of wind energy could be disseminated
in
a short period of time owing to its simplicity,
which is not likely
to be surpassed by another alternative.
67
Small Hydraulic Prime
Movers for Rural Areas of
Developing Countries:
A Look at the Past
6
Joseph J. Ermenc
Historical
Perspective
The two hydraulic
prime movers,which
have been important
instruments
of change in rural
areas of developing
nations
in the past.are
the waterwheel
and the small water turbine
which superseded
it during the +19C.
The baric difference
between the two is that there is a
contrived
flow of water with respect
to the energy conversion
elements
(blades/vanes/buckets)
of the water turbine
and not
of the waterwheel.
The waterwheel
was the first
prime mover used extensively(other
than persons and animals).
It dominated
the power
scene for about 2000 years.
It was at first
integrated
with
traditional
hand-craft
methods which led to changing patterns
of making and doing things
and with revolutionary
increases
of goods and services.
But the power limitations
of the
waterwheel
(less than 20 horsepower)
essentially
limited
the
size of these operations.
Its use was therefore
characterized by a diffusion
of mills
throughout
rural
areas.
In Europe
and the United States this was the prelude
to their
Industrial Revolutions;
Mumford has called
this the Eotechnic
Phase
of a developing
country
based on wood as a material
of machine construction,
water as a source of power, and the native
craftsman
as the dominant technological
figure.(&)
During the Roman Empire all species of waterwheels
appear in rural
areas.
They were used primarily
for grain
and saw mills.
A three horsepower
(2.2 kilowatts)
waterwheel powered grain mill
near Cassino,
Italy
was capable of
supplying
the flour
needs of about 400 people.
A conglomerate of 16 waterwheels
at Barbegal
in southern
France supplied
power to produce up to 28 tons of flour
per day for the
90
Joseph
&AteL&KOMAN
W?fNUwG
J.
28
Ermenc
FLOUR MILL AT
TWS OF FLOUR
TWO SENES a= ElGHT WffEELS;
ANO 28 mf. (70 CM) w’Itw-4 -
ToTa-
~~QSEPOWER
.--_- .__
BARBE6:AC
NE/-\R ARMS
PER DAY.’ A.D. 308-16:
EACH
6.9 F7: (ZZOCm)
C 170 ; HEA0 (t=
FIG. I
) 2,
65
F~(aQp+j)
1947-48)
01,4.
Roman Legions. (2) Though these were examples of the revolutionaq
possibilities
of the waterwheel,
vested interests
and social-political
restraints
seem to have prevented
its
wider and more intensive
use. (Fig. 1)
The collzipse
of the Roman Empire and its slave-based
economy swept z,way the restraints
against
the use of the
waterwheel
in the West.
L,zlC
wal
ir!
rural
u--u watav-whenl
..---. . . . --A
,.YI
..--- 1 established
B1; the ~7. La."1P tha
Even
Europe, driving
saw and grist
mills
and pumping water.
with Europe, considin England which was then, in comparison
there were,according
to the
ered as a backward country,
Domesday Book of 1085, 5624 water-powered
mills
operating
in
3000 communities
in southern
and eastern
England.
an escalating
use of
By the +13C Europe was experiencing
Two principal
stimuli
set
the waterwheel
in many handcrafts.
this off:
1. The exemplary
use of the waterwheel
in monasteries.
2. The labor shortages
brought
cn by the Black Plague.
By the +16C, the use of the waterwheel
ed to such traditional
handcrafts
as:
had been extend-
ore transport
and crushing).
1. Mining(drainage,
2. Metallurgy
(blast
furnace blowers,
rolling
and strip
mills).
3. Metal working
(forging,
turning,
grinding,
wire drawtools.
ing) ; agricultural
4. Woodworking.
5. Leather processingctanning,
harness
6. Paper making
7. Oil manufacture
8. Cosmetics
(powder, perfume).
9. Polishing
(lenses,
ivory,
minerals).
)
l
The number of waterwheels
in Europe by this time was
Their site
selecreckoned in the tens of thousands.
tion,
design and construction
was attended
to by a superior
group of craftsmen
called
millwrights.(3)
In the American colonies
the use of waterwheels
begins
within
a few years of the establishment
of settlements.
In
1647, only 20 years after
the Pilgrims
had landed at Plymouth, waterwheels
were used in an iron mill
at Sugus,
Massachusetts;
here bog iron ore (limonite)
was processed
into wrought
iron bars (:nerchant iron)
at a rate of eight
tons per week.
I
92
Joseph J. Ermenc
But
iron mill.
wheels in
taries
of
taries.
brook use
small mills
were far more prevalent
than the Saugus
By the end of +18C there were about 10,000 waterNew England driving
a variety
of mills
on tribumain-line
rivers
and tributaries
of those tribuIt was not uncommon to have a dozen mills
on a
its entire
flow in succession.
In China, the first
reference
to the waterwheel
occurs
in +31.
It was used for the job of operating
an air bellows
.
-I..--: --.l p?XXXSSiiXg (cast. iron arid wrought iron)
2-C m,-.t-b-l
~~r\;~arrurrjr~a~
and
fabrication
(casting
and forging).
By +230, iron tools
became three times as abundant as in pre-waterwheel
times.(+)
By the +13C, in the Chinese province
of Szechwan, "tens
of thousands
of waterwheels
for hulling
and grinding
rice,
and for spinning
and weaving machinery,
were established
along the canals
(in the plain
of Chhengtu)
and operated
throughout
the four seasons."
During 1780, a Korean scholar
passing
through
an area
40 miles east of Peking wrote,
"I saw waterpower
used for
all kinds of things;
blowing
air for furnaces
and forges,
winding
silk
off cocoons, milling
cereals,.
There was
nothing
for which the rushing
force of water to turn wheels
was not employed."
So by the beginning
of the +19C, areas of Europe,
United States,
and China were well prepared
technologically
to receive
the next beneficence
of hydraulic
powerr the
water turbine.
the
(Since World War Two up to a hundred watermills
in succession have been seen on small rivers
in rural
areas of
the provinces
of Shansi,
Kansu, and Yunnan.
Some waterwheels,
made of wood, have been connected
to small electric
generators
(for lighting)
via pulleys,
belts,
or ropes.)
The water turbine
was a French triumph.
It was preceded by the development
of sound turbine
theory by French
scientists
during
the +18C.
This guided the French engineer,
Benoit Fourneyron,
between 1825 and 1833, to produce several
small water turbines
which demonstrated
superiority
over
the waterwheel
in almost all respects.(z)
They could operate
adequately
at higher
heads or lower
heads than the waterwheel
- and even submerged.
They could
replace
the waterwheel
at a site and develop twice the power
at ten times the speed.
They occupied
only a small fraction
of the volume of the waterwheel.
They could also be placed
as high as twenty feet above the tailwater
(using a draft
SrnaZZ BydrauZic
tube)
without
appreciable
Prime Movers for Rural Areas
loss
93
of power.
By 1838, the news of Fourneyron's
turbine
had already
reached the United States.
A few were built
in the Boston
area, strictly
following
Fourneyron's
design.
But in the
hinterlands
it was rapidly
taken over by blacksmiths
and
foundrymen who found it easy to make, in great demand and
an extremely
profitable
business.
Changes came very quickly
and by the middle of the century
the Fourneyron
turbine
had
been so radicaiiy
altered
by rural
craftsmen
that American
turbines
began to take names of their
many improvers.
Changes -continued
throughout
the century;
during
this period
there were more than 60 varieties
of small,
cheap, cast iron
turbines
on the market as replacements
for waterwheels.&)
Efficiencyandcost
were often
turbine
selection
by mill-owners.
small mill
operators
rated turbines
passing bodies of small animals.
not the main criteria
in
Many rural
New England
on their
capability
of
The last quarter
of the +19C was the heyday of the small
turbine.
When a waterwheel
needed to be replaced
or the
power increased
at a millsite,
a turbine
was selected.
An
1885 census reported
over 900 turbine
powered mills
operating on the 110 mile long Merrimac River and its tributaries.
(The Merrimac River flows from central
New Hampshire south,
then east through
Massachusetts.)
The average power of
these turbines
was about 30 h-p,
(22 kilowatts).
The mills
at least
four
their
limited
were highly
specialized
and were duplicated
times in the Merrimac watershed
indicating
areas of service.
They turned out such household
products
as cutlery
and
edge tools,
brooms and brushes,
looking-glasses,
furniture,
paper, buttons,
woodenware,
stone and earthen-ware,
pencil
lead, vinegar,
co&s,
ivory and bone-work,
toys and games,
baskets,
needles and pins, watches and clocks,
and even
washing machines.
For the farm they turned out fertilizers,
gunpowder,
axles,
agricultural
implements,
barrels,
ax handles,
wheels,
carriages.
There were woolen,
cotton,
flax and linen mills;
hosiery,
lace, worsted,
glove and mitten,
mattress,
cordage,
twine,
shoddy and bagging mills.
There were tannery,
boot
and shoe,
and leather-board
mi.lls.
There
were also
mills
turning
out
surgical
appliances,
Joseph J. Ermenc
94
ON THE THREE MILE
LON\IG NEWFOUNO
IN NEW HAMPSHIRE, 1885
PjltLS
KIND
HEAO
m
WOOLEN
I
PAPER
3
C6RPENTERING
2
16
30
40
23-l
36
3
30
8lJKKSCyITHIM
I
I2
I5
WOODT~MWNG
I
16
30
I
I2
50
CLOVES
I
I8
76
TANNEQY
2
38
WWHWERy
I
8
STRAW~OARD
I
12
WOOD PlJLP
I
22
SAW
I
I2
FLOUR
C
GRIST
CARRIAGES
BW~~OhlS
TOTAL
(FROM
* REPORTS
DEP~Rl\‘GNT
ONTYE
130
IS0
2
200
50
\060
19
VVRTER POWER OF TclE UNiTFO
OF INTCRlo~
RIVE’R
,CENSUS OFFICE,
FIG. 2
\885.)
STATES”,
SmaZZ &druuZic
musical
and scientific
Prime Movers for Rural Areas
instruments.(7)
-
95
(Fig.2)
After
1900, in the United States,
the advent of large
central
thermal
and hydro-power
stations,
rural
distribution
of electricity,
the internal
combustion
engine,
and better
rural
transportation
all contributed
to the decline
of the
small turbine
powered mill.
Today there is only one nationally
known manufacturer
of small turbines
in the United States;
and the last manufacturer
of waterwheels
closed shop in 1967.
So in the United States and Western Europe, both Eotechnit prime movers, the waterwheel
and water turbine,
have
gone through
complete
cycles of existence
from the rise of
invention,
development,
and acceptance
to a gradual
slide
into obsolescence
just short of oblivion.
(Fig.3)
In China, it appears that century
of foreign
interference has delziyed the introduction
of the small turbine;
but
this is now in progress
with particular
emphasis on small
scale hydra-power
stations.
..
A BBC broadcast,
during
1970, informed
that thousands
of less than 100 kw were being installed
in rural
areas.
This parallels
the installation
of small,
simple
hydra-electric
stations
in the United States at the beginning of this century.
(10,11,12,13,14~(Pig.4)
----I_
of units
Since World War Two, small hyd.r&lectric
package units
of less than 20 kilowattslhave
appeared in West Germany,
J!?ungary, Canada, RUssia, and the United States.(S)
But the
cost of these units
indicate
tha:'Western
manufacturers
have
priced
themselves
out the rural
market not only in developing countries
but in our own as well.
But beyond this they
esent refinements
necessary
in rural
areas of developing c5untrific
and unvise kn'that
their
complexity
and maintenance is beyond the understwe
msmmze~of.--- -~--- ---the rural
craftsmen.
(Fig. 5,6)
Watervheels
By the +4C in the Roman West,
China, the waterwheel
had evolved
which converted
different
forms of
mechanical
energy (force-velocity
and perhaps earlier
in
into two distinct
species
hydraulic
energy into
combinations).
1. The undershot
wheel receives
of a flowing
stream of water on flat
vanes/paddles.rFig.
7)
energy
(later
from the impact
curved)
radial
96
cl-usepizJ. Emenc
UaS. TURBINJES
\
0.S.WAlERWHEELS
EUROPfA N
WATERWHEELS\
I
,
0
/’
0’
0
.c4
1300
1400 l5bO
17bO
i6bO
I400
2dOO A.O.
TOTAL
HORSEPOWER
CHINA
a------
I
WATERWHEELS -/
/
0
0 /’
---
.
SMALL (LESS THAN 100 HORSEPOWER) HYDRAULIC
PRIME MOVEI? CYCLES.
FIG. 3
-
Smaii! HydrauZic Prime Movers for Rural Areas
b5a6*
‘7
SMhLL
$-&
&JKIEN#
YJVKWANGTUNG
fl
\I
HYDRO-ELECTI\lC
STATIONS
(AVERAWNG
40 KW)
1N
SOUTH CHINA
(v. SMIL, BULL. ATOMIC
SCIENTIETS,
PEB.1977)
97
E
.-c0
r”
-0
.-E
5
SmaZZHydraulic
Prime Movers for Rural Areas
kW
COST &Q
)tyDl&
(140~
KW OF SM4cL &%LE
ELEcTXIC PAcJGM
II~WDI
\IdlTS
UG I~IsTALLATM
OR mM COSTS)
99
-.
.
!
/
:
t
.
UNOERSHOT
[email protected]
AUGUSTINO
n
.
WATERWHEEL
WXRATING
MELLOWS FOR TWO IRON HEARTHS
FQt3M
RAMELL I, “LE DIVERSE ET ARTlflCIOSE
FIG.7
FtACHINE;PAQI9,
I
Sm7,Z [email protected]
Prime Movers for Rural Areas
101
2-. The overshot
wheel has 'buckets'
on its periphery.
These are filled
at the top of the wheel's
rotation
and empty
as they approach the bottom,
(the breast wheel receives
in buckets,
water,
at about radial
height).
The overshot
wheel always rotates
in a vertical
plane.
(Fig. 8)
plane
Tfie undershot
or a vertical
wheel may rotate
plane.
in either
a horizontal
The vertically
rotating
wheel was the dominant
form
used in the West though the horizontally
rotating
wheel
was to be found in the hilly
regions
of the Balkans,
France,
Ireland,
the Shetlands,
and the Faroes.
Italy t Scandinavia,
(Fig. 9)
In China, the horizontal
wheel was dominant.
Until
the +19C, both types of waterwheel
were generally
made of wood with cast and wrought iron used sparingly
where
extra strength
or wear resistance
was required.
The Overshot
Wheel
The overshot
waterwheel
had an efficiency
around 80
percent
which was several
times higher
than that of the
undershot
waterwheel.
Indeed it was often higher
than the
efficiency
of the early small turbines.
Further
its efficiency remained high despite
variations
in load imposed
upon it;
this superior
characteristic
was not obtained
with
the undershot
wheel or with the early turbines.
For the mill
operator
the overshot
wheel was most
economical
of water and this explained
its use well into the
+2OC and particular&y
on small streams.
But on the debit
side,
the overshot
wheel was limited
to speeds of less than
20 r-pm for wheel diameters
of 15 feet (4.6 m) and for horsepowers of less than 20 (15 kw).
These limits
were imposed
by the use of wood as the material
of construction
before
the +19C.
Because
achieved with
units had to
millstones
to
consisted
of
grain milling
required
higher
speeds than
the conventional
wheel, power transmission
be inserted
between the water wheel and the
increase
the rotational
speed.
The units
gears or pulleys
and beltsor
ropes.
When the available
heads were greater
than the usual
wheel diameters,
the overshot
wheel tias used in series.
This
was to be seen in surprising
fashion
at the Gallo-Roman
192
Joseph J. Emenc
r--I .._’
!P
. -Y
-.
OV~QSHOT
W~;ERWHEEL
OPERATING
A GRIST MILL
AND A
PISTON PUMP
(HAMELLI
We\
SmaZZHydrauZic Prime Movers for Rural Areas
HORUONTAL
WATERWHEEL
OPGRATING
A GRIST
MILL
(RAMELLI
rsaq
I c.3
104
Joseph J. Ermenc
flour mills at Barbegal in southern France; they were built
between 308 and 316 AD and represented
the largest
hydraulic
power plant the world was to see for fifteen
hundred years:
the total power developed there was about 170 horsepower
(127 kilowatts).
There were two series of eight overshot wheels, each of
which was geared up to drive two sets of millstones.
Each
series of wheels received water from the aqueduct supplying
the city of Arles.
They used the water in succession
along
a slope of 30 degrees and through a vertical
height of about
61 feet (18.6 m). Each wheel had a diameter of 7.2 feet
(220 cm) and a width of 2.3 feet (70 cm). (Fig. 1)
All together
they produced 28 tons of flour per day,
mainly for the Roman Legions in southern France.
A similar
mass production
flour mill was built
at Prety in Burgundy
for the Legions of Northern Gaul.
During the nineteenth
century the overshot waterwheel,
in responding
to demands for increasing
power, made
increasing
use of iron forgings
for axles and hubs and cast
iron for rims.
In 1854 the largest
diameter Western overshot wheel was
constructed
on the Isle of Man, in the Irish
Sea, to pump
water from a lead mine.
It was 72 ft(22 m) in diameter,
6 ft (1.8 m) in width.
It developed about 125 horsepower
(86 kw) at 2 rpm.
It could pump 250 gallons per minute
from a depth of 1200 feet.
In the United States,
in 1851, the largest
overshot
waterwheel was installed
at the Burden Iron Works in Troy,
New York, for the manufacture
of railroad
spikes.
It was
60 ft (18.3 m) in diameter and 22 ft(6.7
m) in width;
it
delivered
about 200 horsepower at a speed of 24 rpm.
The Undershot
Vertical
Waterwheel
The efficiency
of the undershot vertical
wheel with
flat vanes set in a stream ranged from 10 to 20 percent;
with curved blades this rose to 20-30 percent.
If water
was conveyed to either
a vertical
or horizontal
wheel via
a close fitting
channel or if the wheel was shrouded to
minimize
the effect
of turbulency,
these efficiency
ranges
were doubled.
But these efficiencies
are maximums which
are attained
only when the speed of the wheel is roughly
SmII Hydraulic Prime Movers for Rural Areas
105
half that of the approaching
stream of water: on either
side
of this wheel-speed,
the efficiency
decreases;
this would
happen with variations
of load or water supply.
Except for curved blades, the Pomans had developed the
undershot wheel with shrouds into a standardized
piece of
equipment which often was of better
design and construction
than most rural wheels of the +18C. An impression
of one in
lava was found near Cassino, Italy.
It had a diameter of
6.1 ft (1.85 m), a width of 10 inches (0.25 m) and flat
paddles.
It was estimated
that this wheel developed about
three horsepower at about 10 rpm.
Since millstones
required
a higher velocity
for some of its products,
a set of wooden
peg tooth gears were used which, in Roman waterwheel
pracstepped up the speed by a factor of five.
(Fig. '10)
tice,
The output of this rural Roman flour mill was estimated
to be 300 lbs (136 kg) per hour.
Compared with 20 lbs (9 kg)
per hour which could be done by two men on a quern this was
indeed revolutionary.
A novel use of the undershot water wheel was developed
during a siege of Rome during the +6C. This was the boat/
It consisted
of an undershot wheel suspended
floating
mill.
The;? were widely used
between two rigidly
connected boats.
on swift
flowing
sections
of navigable
streams in many
It has been estimated
that
European countries
and China,
on the River PO, in Italy
at the beginning
of the +19C, there
were about 600 boat [email protected])
The use of boa-t mills
in China begins about the same
time as in the West; on them trip hammers were often used
for such tasks as:
1. Hulling
and grinding
rice.
2. Pulping fibrous
materials
3. Pulverizing
materials
for
for paper.
drugs and perfumes.
In the United States the undershot vertical
wheel was
used where terrains
were unfavorable
to the building
of
In the West, the
dams to impound water and create heads.
the undershot wheel with buckets attached
Roman noria,
At Ellensburg
on
to raise water, was used occasionally.
wooden undershot
the Columbia River, two crudely constructed
one of 42 feet in diameter,
and the other
waterwheels,
30 feet were used to irrigate
40 acres of land for alfalfa
and fruit.
106
Joseph J. Ermenc
&‘DlA.(1.85M)
107
SmaZZHydraulic Prime Movers for RuraZ Areas
And some Alaskan Indians constructed
wooden undershot
wheels of the noria type but with nets instead of buckets.
The nets scooped fish out of the water and dumped them into
a chute.
Undershot vertical
wheels were also used to develop
tidal
power along the Atlantic
coasts in Europe, England,
They were generally
used as adjuncts
and the United States.
to inland water mills
or windmills.
The Undershot
Horizontal
Waterwheel
This wheel rotates
in a horizontal
plane (its shaft
vertical).
Water is conveyed to it by an open or closed
channel from water at some elevation
usually
under 10 ft
(3 4.
is
By the +17C, it had evolved into a high speed elementary turbine
in southern France, the Balkans, and China.
With curved blades it reversed the flow cf water to secure
a reactive
effect
which produced speeds four and five time;
that of vertical
wheels.
It could therefore
be directly
connected to millstones
and saws without
the interrediarv
gears or pulleys
and belts.
They also made use .>f large
diameter shafts for their
'flywheel
effect'
in smocthing
out sudden velocity
variations
caused by vagaries
in w,iter
flow and load.
I
Indeed it was the alleged remarkable performance
of the
flour mills,
near
horizontal
water wheels at theBasacle
Toulouse, France, which, during the +18C attracted
the
attention
of such early French hydro-dynamicists
as Euler,
Bernovilli,
d'Alembert.
Their studies of this wheel found
focus in the invention
of the first
practical
water turbine
by Benoit Fourneyron from 1825 to 1833.
During the +19C, horizontal
blades were used in northeastern
and saw mills.
waterwheels
using
United States for
flat
Lyrist
In China, as late as 1958, wooden horizontal
wheels,
with blades of extreme curvature
exhibiting
high speed
characteristics
were being built
for connection
to small
electric
generators
(under 5 kilowatts)
via pulleys,
belts
or ropes.
From 1910 to 1960, a horizontal
wheel was in operation
in central
New Hampshire grinding
grain for animal and
poultry
feed:.
It is 7 ft (2.1 m) in diameter and received
Joseph J. Emetic
ADJUSTABLE
BEGINS HERE
FRANCIS
(RADIAL-INWARD
FLUW)
[email protected]
ADJUSTABLE
GUIDE VANE
tV’lER(CAN/FRANCIS
(MIYED
FIG. !I
FLOIN)
7-Uf?~H’4E
SmaZZHydraulic Prime
Mowers
for Rural Areas
109
water under a head of 6 ft (1.8 m) through a 2 ft (0.6 m)
square closed channel bound with iron straps.
It probably
at 30 rpm.
Its
developed about 10 h.p. (7.5 kilowatts)
shaft is of steel and 5 inches (12.7 cm) in diameter.
The Water Turbine
The water turbine
differs
from the waterwheel
in that
there is a flow of water relative
to the blades/vanes;
under
design conditions
the water enters the moving blades smoothly and leaves with minimum energy.
There are three well established
classes of turbines
which are generally
known by the names of their
inventors,
They are in historical
order:
developers,
or promoters.
1. Francis
(reaction,
pressure change across blades,
full
admission,
inward mixed-flow).
(Fig.11 )
no pressure change across blades,
2. Pelton (impulse,
partial
admission,
tangential
flow). (Fig. 12.)
3.
Nagler (fixed propellor,
or reaction,
pressure
change across blades,
full
admission,
axial flow).
(Fig. 13.)
4, Kaplan (variable
pitch propellor,
reaction,
full
admission axial flow.
Nagler and Kaplan turbines
are propelled
The Francis,
mainly by reaction
forces and appear to be lineal
descendants of the Alexandrian
(Hero/Ctesibius)
steam turbine
( -lC or -2C), the first
jet engine.
The Pelton turbine/wheel
appears to be a descendant of
the horizontal
waterwheel
but it had an independent
conception and development in California
following
the Cold
Rush of 1849.
The inward flow Francis turbine
was a radical
transformation
of Fourneyron's
outward flow turbine
to keep it
a high rpm machine while responding
to demands for more
This was done by keeping the rotor diameter
increase
power.
minimal while increasing
the rotor
(blade) depth.
This
resulted
in a compact rotor of considerable
complexity
but
not beyond the fabrication
capability
of the village
blacksmith or small town pattern-mAit:and foundryman.
It was this small Franc&,= tllrt;lne
which was the principal successor to the waterwheel
In northeastern
United
States.
Due to its full
admission,
steady flow features
it
was able to approximately
double the power of the overshot
waterwheel
at a given site wilere the flow of water was adequate.
However on small brooks the waterwheel
was preferred
110
Joseph J. Ermenc
CKET
_--__--------------------------__------------_----_----------_ -------------__-----L/i. ///H/H~~W~~
PELTQN
TURBINE
PROPELLOR
FIG. I2
TURBtNE
FIG. 13
Srna~ZHydraulic F!rim~ Movers fcr h'ural Areas
because
bine.
it
was more efficient
at par:
loads
I1
than the tur-
The Nagler fixed propellor
turbine
resembles a ship
ropellor.
This
suggests
it
to
be
an
invention
via the inI.
version method.
It is a simpler structure
than the Francis
turbine
and is particularly
compatible
with the usual waterwheel heads.
It c<a.meon the scene >Ver World War One with
the rise of large central
hydro-electric
stations
and was
built
for this market as a competitor
of the Francis turbine.
Its high speed often permitted
direct
coupling
with
an electrical
generator.
The Kaplan turbine
uses a variable
pitch propellor.
It maintains
a higher efficiency
under load fluctuations
than the other types.
But it is mechanically
more complex
and is 5enerally
used only in large hydro-power,
efficiency
oriented
stations.
The Pelton impulse tLtrLi.rle is
mover in which high pressure water
converted
into a high velocity
jet
onto hemi-spherical
blades.
It is
greater
than 100 ft (30 m). Their
high pressure pipes/penstocks
wo;lld
many developing
countries,
Hydraulic
The development
where the evaluation
able:
.*- a partla,
.\1d.L.~
T.i , .-I -L,.-*>;1c
is received
in a nozzle,
of water and directed
generally
used with heads
:,:?yui.rement of long,
preclude their
use in
Power Sites
of hydraulic
power is limited
to sites
of the following
factors
must be favor-
1. Topography (slopes, water storage,
evaporation).
2. Geology (run-off,
dam sites).
3. Stream flow (rainfall;
water shed area).
In developing
countrieJ
qf the past this was done by
experienced
millwrights;
today it is a formalized
procedure
described
in engineering
textbooks
and handbooks.
The drop in elevation
may be of natural
origin
or
artificially
created by a dam which also serves to store
water.
For small mills,
economy generally
required
that the
connection
(penstock or pentrough)
between the dam and ?*Lbine or waterwheel be kept a minimum.
But with small hyardelectric
stations
of 1000 kw or less, in New England, it was
not uncommon to see penstocks of five and six ft (1.5-1.8
m)
in diameter and several miles in lengths.
212
Joseph J. Ermenc
Log dams of the leave' or 'rafter
prop' type were generally
used in New England.
They were cheap and easy to
maintain
and replace.
Their stability
was obtained by the
weight of water above the 'cave' rather than by the mass
of the dam itself.
The cost of monolithic
concrete or
stone dams was generally
prohibitive
for small New England
mill owners.
Summary
The waterwheel
and small hydraulic
turbine
were of
basic importance
in the development of rural areas of the
United States and Europe.
Indeed their use may be seen, from
an historical
perspective,
as the springboard
from which
hydraulically
endowed areas were transformed
from handcraft
economies into diffused
machine-based
economies with consequential
revolutionary
increases
in the production
of
goods and services.
The major role in this Loveiopment was played by the
simple waterwheel.
The advent of the small hydraulic
turbine provided more power at a given site than was feasible
with the waterwheel.
The introduction
of the small water turbine
appears to
be taking place in China at the present in much the same
way as it did in the United States a century
ago.
Though
large central
thermal and hydro-power
technology
are available to the Chinese, it appears that there is considerable
emphasis, at the present,
on small hand-controlled
machines.
It seems that the history
of small scale hydro-&2wer
development provides
sound suggestions
of how to aid
hydraulically
endowed rural areas of developing
countries
in achieving
an improved standard of life
not only for their
areas but also for less advantaged areas of their country.
Small Hydradie
Prime Movers for Rural Areas
113
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
New
Mumford, L. 1934, "Technics and Civilization".
York: Harcourt and Brace.
of Technology",
Vol.8,
OxSinger, C. 1958, "A History
ford: University
Press.
on Mills
and Millwork".
Fairbairn,
W. 1864, "Treatise
London: Longmans, Green and Co.
in China".
Needham, J. 1971, "Science and Civilization
Vol. 4 Cambridge: University
Press.
of Hydraulics ". State
Rouse, H., Ince S. 1957, "History
University
of Iowa.
Ermenc, J.J. 1976, "Dartmouth Readings in Technology:
Thayer
The Historical
Development of Water Power".
School of Engineering,
Dartmouth College.
U.S.Dept.
of Interior,
Census Office,
1885, "Reports on
the Water Power of the United States",
Government Printing Office.
Bachelli,
R. 1950, "The Mill on the PO". New York:
Pantheon.
Vol. 2,
Mosonji,
E. 1960, 'Water Power Development",
Budapest:
Hungarian Academy of Sciences.
Editorial
Publishing
Group, East China College of Water
Conservancy,
Nanking, 1973, "Rural small scale hydroPeople's
Republic of China: Shanelectric
stations".
chai People's
Press (in Chinese)
"How to Run Small Power Stations
Efficiently".
1971
Shanghai People's Press
People's
Republic of China:
(in Chinese).
Canton Bureau of Water Conservancy and Electric
Power
"Rural Small-Scale
Hydro-electric
Stations."
1973.
Canton People's Press (in
People's
Republic of China:
Chinese).
Editorial
Publishing
Group on Rural Small-Scale
HydroBureau of Water Conservancy and ElzcPower Stations,
Committee of Human Province
tric
Power, Revolutionary
"Rural Hydra-Electric
Stations",
2 volumes.
1974.
People's
Republic of China: Hunan's People's
Press.
(in Chinese).
"BBC Summary of
New China News Agency, August 1970.
Part 3: The Far East Weekly Economic
World Broadcasts",
Iondon: British
Broadcasting
Company.
Report.
7
Wood Waste as an
Energy Source in Ghana
John W. Powell
I
TlBTROPICALHXRFOREST OF GRANA
Kumasi is the capital of the ancient kingdom of Ashanti
which now comprises the Ashanti Region of Ghana. Almost the
whole of Ashanti is covered by tropical foresta and more
than half of the sawmills in Ghana are located in or around
Kumasi. It is therefore an ideal location for the study of
the tropical high forest and the industries which it supports.
Two important institutions
located at Kumaai are
engaged on research connected with the cultivation
and uses
of timber. The Forest Products Research Institute
of the
Council for Scientific
6 Industrial Research is the principal
institution
engaged upon research into the industrial
It is situated
utilisation
of the produce of the forests,
on the Campusof the University of Science &Technology which
is the only taohnological university in Ghana and includes a
eourae in Timber Tsohnology among its curricula.
The
University haa established a Teohnology Consultanoy Centrs
to stimulate the development of small industries through
the introduction of intermediate technologies and several
of its projeote involvo the use of wood as a fuel or as a
The present paper draws mainly upon the
raw material.
experienoe of the Forest Products Resoarch Institute
and
the Teohnology Consultamy Centre.
The tropical forests of Ghana cover perhaps one third
of the land area or about 30,000 square miles. Much of the
mmainder of the Country supports more limited tree cover
of 0avama 0pecie6. It is variously estimated that between
150 and 300 species of trees grow in Ghana. Of these, only
18 am listed aa Prime Speciea with timber that is amaptable on the international
market. Only 8 species account
for 70$ of Ghana's timber exports. These 8 are Mahogany,
115
116
J&n W. PuweZZ
African Mahogany, Wawa, Sapele, Utile, Makore, Kokrodua
and Anthotheoa. Although much effort is expended in
attempting to export the so-called seaondary species, the
situation is changing only very slowly. It is true therefore to sr4~rthat the vast majority of the trees growing at
the prssent time are not ussd as timber. According to
Siwekl , about 80$ of the Ghanaian species are suitable for
use as fuel wood.
!l%ers is no doubt that the land area of the tropical
forest is redwing and has boon reduoing for several
centuries.
!Fhe historical
cause has been the slash and
burn agricultural
practioe but in recent years, the timber
companies havs taksn out trees faster than the reforestation
progrsmme can replace them. !Fhe savanna lands drift southwards pursued by the Sahara across the whole of the
Sahalian region, and this effect can be seen in Northern
Ghana. There is also no doubt that the ecological balance
is being upset in other ways especially by the selective
re-planting of commercial spsciee in the forest area and
fast growing species for firewood in the savanna area.
However, this consideration is beyond the scope of the
present paper.
According to Siwek, a total of 10 million cubic metres
of vood is obtained in Ghana every year. This figure is
assumedby the present author to represent the total
quantity of wood cut or burned from the forest by all
nethods. Of this total, Addo* states that in 1973, 2.075
million m3 were won as logs by the logging companies.
Approximately half of this timber was exported, some after
sawing, snd the reminder went to the sawmills and plywood
and venssr factories.
It is the polioy of the Government
of Ghsna to restrict
the export of logs and to enoourags
the export of sawn timber and wood products. As a result,
the quantity of logs exported has been fslling and the
quantity handled by the sawmills has been rising, although
the latter trend is slower. Ibr the purpose of the present
papa, the 1973 data and soms 1974 data is used.
In 1973, the
located accordiq
Ashanti
Eastern
Western
sawmills of Ghanawere said to number 70
to region as follows*:
Region . .
..
.. 34
Region . .
.
m 17
Region 4m
.
a 8
l
l
l
l
~Jooa’Yaste as an. Energy Source in Ghana
II 7
Brong Ahafo Region . .
.. 6
Central Region 4* . .
*.
3
They utilised a total of 0.85 million m3 of logs. Plywood
and veneer factories utilised a total of 0.14 million m3
of logs. It has been estimated* that the conversion factor
in sawmilling is about 4% while that of plywood manufacture
is 3% and that of veneer manufacture is 28,
Hence it
csn be estimated that waste from the sawmills and factories
totalled some 0.56 million m3 in 1973. This waste was in
the form of off-cuts,
sawdust, shavings and log cores from
the veneer and plywood plants,
Off-outs are used for
firewood and as a principal source of material for the
charcoal burning industry.
Sawdust is little
used except
for sealing charcoal kilns and some sawmills have a
smouldering heap as a means of reducing the accumulation.
It is estimated that 25,500 tons of sawdust sxe produced
annually. At least one plywood fautory in Kumasi has
introduced a wood-fired boiler which burns a proportion
of sawdust with wood off-cuts and this trend can be
expected to increase.
In addition to the sawmills0 there were in Ghana in
1962, (1962 Industrial Census Report Vol.1 - Industry)
1,251 wood sawing establishments.
These would include
traditional
carpenters in informal industrial areas and a
few small furniture manufacturers. Approximately, ens
third to one half of the timber produced by the sawmills
and plywood factories is processed further by these
establishments with the conversion of a proportion of the
sawn timber into additional waste. Much of this waste is
in the form of wood shavings produced by planing and some
is used for mattress filling
and some for land reclamation.
Allowing a oonversion factor of about 80$, it can be
estimated that an additional 0.04 million m3 of waste
results from the local processing of sawn timber and
plywood. Hence the total wood waste from the timber
industry in Ghana csn be estimated to be about 0.6 million m3 .
AVAILABILITY OF WOOD
WASTE AND GHANA’S ENRGY IiEFJX
The quantity of wood waste produced by thexber
industry of Ghana must be considsred to be only the smaller
pact of the total quantity produced by logging, reforestation and farming operations.
Using the available data for
the years 1973 and 1973, the chart shown in Figure 1 has
been derived. The data are only approximate and also
change from year to yesr, It is quite clear that the
overall picture is dominated by the &$which is taken by
informal cutting ancj burning, Inoluded ia this amount is
about 1.5 million m felled in land clearance for
118
John W. PoweZZ
In Ghana
Wood Flow
AnnmI
Total
AnnuaI
Yield
IO Y IO’ m 3
L
Informal Cutting
Loggmg Industry
n
Export
\
12%
\,
3”/e
Sawml I Is
I
reforestation.
Ofei4 estimates that logging operations
leave behind in the forest as much felled timber as they
take out. This accounts for about 2O$ of the annual yield.
The wood w;.*ste available to provide energy as firewood and
charcoal rjpresents a very large resource. The quantity
of wood involved could total some 8.7 million m3 according
to the available data.
Assuming after Siwek that SQ$ of
the Ghanaian species are suitable for use as fuel, the
actual quantity of wood z+vailable to produce energy might
be as low as 7 mUlion m> if one allows that non-fuel
species are cut down when land is cleared for farming,
logging and reforestation.
The heat of combustion of this
quantLty of wood amounts to about 10 '4KJ. Burning steadily
throughout the year ) this combustion is equivalent to an
energy flow of 3500 MW,
The hydra-electric
power station at Akosomboon the
Volta River has an installed capacity of 912 PM and
represents some 9% of Ghana’s electricity
generating
capacity.
The total installed capacity in the Country io
therefore less than 1000 MW. However, the power consumed
by Ghana's population of 10 million is very much less than
this, because the Akosomboplant does not operate at full
capacity, 30-40$ of its output ie exported to the American
Volta Aluminion Companysmelter at Tema and more power is
exported to Togo d Dahomey. The Volta River Authority
supplied 3.9 x 10 f%& in 1975 which is equivalent to a mean
power consumption 0.: 445 MW.
It has been stated 7 tha in 1973, Ghana consumed
energy amounting to 4.3 x 10 13KJ. About 70$ of the total
was imported in the form of fuel oils and some coal.
These data therefore suggest that the power consumedin
Ghana is equivalent to 1500 MWof which only some 450 MW
originate in Ghana. It seems probable that the latter
takes no account of firewood utilisation
and considers
only hydra-electric
power. What is more clear is that the
potential power output of the wood wastes of Ghana is well
in excess of the total national power consumption and about
seven times the domestic electricity
consumption.
MOreliable estimate has yet been made of the total
utilisation
of wood waste for energy production in Ghana.
Someestimates have been made of the size of the chmcoal
burning
industry but these vary widely from 70,.000 tons to
300,000 tons per annum. The most likely figure seems to
be about 100,000 tons which is the conclusion drawn by the
Capital Investment Board of Ghana. Marc?than 9% of this
total is made in traditional
earth kilns with a conversion
123
John W. Powell
factor of from 5 to 1%.
Taking a mid value30f lO$ gives
sn annual utilisation
of about 1.6 million m of wood waste
for the uharcoal burning industry.
About a quarter of this
total may come from the sawmills. Referring again o
Figure 1 suggests that the balance cf 7-1 zil7icn
m4 is
-used ae fire-wood or is lost,
All investigators observe
that a large proportion of timber is left to rot where it
Losses include the branches of trees felled for
falls.
logging, secondary species felled and left lying in
reforested areas and large trees left lying on cleared farm
land. Much timber remains unueed because it ia too large
to be cut into pieces for head loading from the forest to
the village.
However, in savanna -as,
where wood is
scarce and trees sre smaller, all of the felled timber is
used and most is cons-d as firewood.
PRESENT
USESOF FIREMOOD
ANDCHARCOAL
The greatest proportion of firewood and almost all
the charcoal is burned to provide heat for domestic purposes,
mainly cooking. It is easy to form the impression in the
town that every homehas a charcoal stove. The two largest
towns, Accra and Kumasi, consume about 7% of total
charcoal production.
However in the rural areas, large
numbers of womenand children are seen cutting and carrying
wood to their homes. No doubt the cost of charcoal ($7.50
= $6.52 per sack of 721bs.) is too high for most families.
It is apparent that in the Country as a whole, most
families use firewood for cooking. The firewood is burned
in a small mud stove of traditional
design (Bokyea). A
few families have metal stoves, others use old car wheels,
concrete blocks or large stones to support the cooking
pots. Most of the firewood is collected from the bush and
costs nothing in cash terms. The womenand children who
do this vork would not be able to spend the time to earn
a cash income and so their time is regarded as free. Ni3X
sawmills, wood wzmte mey be obtained for a small charge,
If it is assumed that 1 million Ghanaians use charcoal
for cooking in the main towns of Accra, Kumasi, Takoradi,
etc. and that the other 9 million use firewood to produce
an equal energy supply per capita, then the firewood use
for cooking in Ghana would amount to about 2.7 million m4
or 2796of the total annual yield.
Zxperience suggests
that the industrial uses of firewood are unlikely to total
more than half the domestic consumption. Hence one arrives
at a very approximate total firevood utilisation
of 4.0
3
million m o This represents a little
over a half of the
wood available and although the remainder includes some
WeedWaste as an Energ So.wce ix Cha?m
121
species not suitable for firewood, it is clear that a great
abundance remains unused. This conclusion accords with the
observation made by several investigators that the logging
and reforestation
operations leave large quantities CL z:,od
It is also consistent with the fact
to rot in the forest.
that firewood is considered to be a free commodity in the
forest sons; this being a sure indication that the supply
elmeeds the demandby a substantial margin.
Nost rural industries we firewood as their main
source of energy. The wood is burned in furnaces and open
fires of varying degrees of efficiency.
As examples, one
csn consider the industries making soap, glass beads, brass
castings and hand printed cloth (Adinkra).
The soap boiling
tank is made of steel and has a capacity of 5 ton of soap.
It stsnds on a metal frame around which a stove is made from
bricks and mud. A short section of drain pipe serves as a
rent for the smoke. At the Soap Pilot Plant at Kwamo,
Ashsnti, established by the Technology Consultamy Centre,
both electrically
heated and firewood plants were assessed.
The fuel cost for firewood was less than one half of the
The firewood plant was cheaper to
cost of electricity.
make and easier to operate. There was no doubt that the
firewood plant was the appropriate technology even in areas
In most rural areas,
where electrioal
power was available.
there is no choice of fuel at the present time.
The furnaces for glass bead making at Dabaar, Ashanti
and elsewhere are made of brick and clay and use lorry or
csr front tiea for the grid which supports the clay mouldss
The fuxnace operates at 700-750°C.
The furnaces used for
the lost wa brace casting industry at Kurofofrom, Ashsnti
are made of similar materials but they are smaller, use a
hand operated bellows and operate at a higher temperature.
!l!he design has been adversely criticised
and is probably
The Adinkra cloth printers of Ntonso, Ashanti
inefficient,
boil their dye oversnopen fire in the fonnof a trough
formed by large stones or concrete blocks.
?hny rural industries making feeds and drink8 use
are bread baking, fish
firewood for heating. Solne e-lea
smoking, gsri msking from cassava, pito (beer) brewing and
skpeteshi(gin)
brewing.
BlauksaLiths generally use palm kernel charcoal for
their hearths, It ie an excellent fuel for this purpose
and has replaced the use of imported coke at the Teohnology
Consultancy CeEtre Workshop.
122
G’ohrzid. PO~JZ.
2Z
SoQlIE
DEVELOFXEXTS
IN THfZUSE OF WOOD
WASTE
The University of Science & Technology has become
involved in some attempts to increase the utilisation
of
the prospebt of using the large
wood waste. In particular,
quantities of sawdust from the satills
presents a great
Mrs
B,
S.
Nijjhar
of
the
Department
of
challenge.
Mechanical Engineering has devised a domestic stove that
burns sawdust. It appears to be successful and has created
Attempts have been made to use the
SOnIs idSreSt
iocally.
sawdust stove as a heat source for other applications.
The developrPenf of a superheated-steam bread-baking oven
has made some progress. An attempt to link the sawdust
stove to a fan driven dryer for agricultural
produce failed
because the smoke tainted the product.
An application in which smoke is intended to affect
the taste of the product is fish smoking. An improved fish
smoking oven has been designed by Dr. B. A. Ntim of the
Teohnology Consultancy Centre. Trials have been successfully completed at Ktmasi and several ovens are now being
constructed at Elmi~ I on the coast of the Gulf of Guinea.
The projeot is being undertaken for the Food Research
Institute of the Council for Scientific & Industrial
Research.
The soap boiling tank with firewood heating developed
by the T.C.C. haa already been described. It was designed
with help from Mr. G. Frakash a Consultant from India who
visited Kumasi for 3 months in 1975.
llEVELOPMENTS
IN THECEARCOALBURNllJG
INDUSTRY
Ch~ocal is made almost exclusively in the traditional
The wood is piled and thoroughly ignited.
earthkiln*
When judged to be sufficiently
hot, the pile is covered
with leaves or sod and then with soil to exclude practically
all air but leaving openings for the escape of smoke and
controlled ventillation.
When in the judgement of the
charcoal burner the charring is complete, all openings are
tightly closed and the pile is allowed to cool. By this
method, some of the wood is burned to produce the heat
requirad to produce the carbonisation.
The yield is poor
and the charcoal contains quantities of earth and ash.
Somecharcoal burners have established themselves
beside sawmills and use exclusively the waste timber for
the pile and sawdust for the covering.
This system makes
use of all the waste and works to the advantage of the
sawmill in that it is relieved
sswdust.
of the need to burn its
Acoording to Siwek, the traditional
charcoal kilns
used in Ghana have an efficient
of conversion of only 5 to
1% whereas efficient
carboniaation should provide yields
It is clear therefore that the annual
of mound2s
production using the same quantity of raw material could
increase from say, 100,000 tons to 250,000 tons if the
industry could be converted to the use of m3s.m kilns.
The Department of Forestry
has undertaken lasts with
several types of kiln including the Princes Risborough
type of kib,
the Chinese kiln, Unique kiln and Tranchant
The Risborough type kiln waa mads from mud dried
kiln.
It suffered from the disadvantage of a long
bricks.
cooling time, although it achieved yields of up to 26%.
The Chinese kiln gave yields of only up to 1556. The best
results were given by the Tranchsnt kiln.
This kiln was
It consisted of two
made of steel and was portable,
cylindrical
sections and a lid.
Four air inlets and four
flues were arranged at the base of the bottom shell.
The
average volums was about 1 cord of wood.
Velsin 2 reports some comparative tests carried out
with the Tranchant kiln and the traditional
kiln.
The
results
are given in Table 1.
TABLE 1
AVEEUGE
WEIGHTOF WOOD
ANDCRARCOAL
PER CORDRUN
(1 CORD= APPROX.3.6 rn2 OR 128ftj)
WOOD
k
TYPE OF
KILH
SATIOH
&
$ DRY
WOOD'CUD
(EEi!iE)
I
'Franchant 1
Kiln
I
Traditional
Earth
MxandKiln
3*75
42%
3580
4.2
599
I
I
124
John w. PmeZI
Other reports give ths yield for the Tranchant kiln
based 33 dry wood weight as 23% and 26%. It clearly provided
a ccrsiderable advance on the traditional
kiln in terms of
both yield and process time.
out in several places using
the Tranchant kiln and people were given careful training in
its use. However, the kiln has not gained acceptance and
are the
many have been abondoned. The main difficulties
need for skilled operators and the high cost ($SoS.CO in
traditional
earth
kilns
which require no
. compared to
. This fate has befallen other attemp:s to
2)~vestment
Experiments
introduce
were carried
improved kilns.
It may well
be that while
the
availability
of waste wood exceeds the demand, there is no
Also,
economic incentive to seek higher efficiency.
Welsing reports that 98% of all workers in the charcoal
burning industry received no education of any kind. Such a
situation must retard the pace of technological progress.
In 1975, there were reported to'be only 33 dern kilns in
9
use producing a total of 500 tons per snnum
l
The T.C.C. is hoping to collaborate with the Building
& Road Research Institute
(BRRI) of C.S.I.R., also located
at Kumasi, to construct a pyrol tic converter developed at
Georgia Institute
of Technology7 . The pyrolytic converter
would be located at the B.R.R.I. brickworks at E\umesua,
Ashsnti and would supply fuel oil for the brick kilns as
well as charcoal for local sale. The process would be
continuous and the consumption of dry sawdust would be about
6 tons per day.
Siwek maintains that some species of Ghanaian timber
up to 30 or 40$ of charcoal by weight, There is
no shortage of local supplies of good quality wood, the
local demandfor charcoal is great and there are prospects
for exporting any surplus. One estimate of the local demand
for domestic use is 140,000 tons per annum. A proposed
ferro-silicon
plant would require a further 30,000 tons a
year. In addition, Ghana imports over 600 tons per anuuunof
activated charcoal for the sugar and pharmaceutical industries
which could be produced locally by pyrolytic converters.
can yield
An excellent raw material for producing‘activated
charcoal is coconut shell.
The T.C.C. has assisted the
village of Elmina to eqort raw coconut shells to
Yugoslavia for this purpose. Together with the available
waste palm kernels, another excellent raw material, waste
coconut shells could be converted in Ghana into an
estimated7 2,000 tons of activated charcoal per annum,,
This operation could earn or save foreign exchange of over
$1 million evsry year,
I
Siwek expresses the view tha-t large organisations
should be established to take up charcoal production by
modern methods. These should be situated near sawmills or
however, such an approach could have
firewood plantations.
unfortunate social consequences and i:ould lead to the demise
This has happened to some other
of the indigenous industry.
industries and advocates of the intarmediate technology
approach to development prefer to stimulate the gradual
upgrading of the indigenous industry.
The traditional
charcoal burning industry has resisted change and it is to
be hoped that it can be persuaded to accept some technological evQl;itXiard before it is swept away by the power of big
business,
GREXiERUT1LISATICN.F~NOD WASTE
There is much evidence to suggest that the waste from
the sswaills and the informal wood-working industry is
Most wood waste finds its wa~r
reasonably fully ti:Uz.s&,
into the charcoal burning ir;&uue~;-, ilae as fiyewood and
other informal industrial uses. The oiis craa of improtement
is the utilisation
of savdust about which something has been
said. However, there is much which could be done to improve
the utilisation
of the wood waste produced in the forest by
the logging operations and by land clearance for farming and
It may be no exaggeration to suggest that 3
reforestation.
or 4 million J of timber per annum is either burned in land
clearance or left to rot in the forest where it falls.
It is difficult
to envisage a rapid change in farming
methods which would avoid burning useful timber in land
crrlearing opsrations.
One csn only suppose that as firewood
becomes scarcer, it will becoms economically necessary to
rsmove all.useful
firewood before burning the land. This
trend csn already be discerned from the effects of population growth and the diminishing aretr of the forest.
On the
other hand, much could bs done to improve the utilisation
of
non-commercial trees felled in lsnd clearance and logging
operations and the br ches of the trees felled for logs.
It has been estimated Y that logging operations leave behind
40 to 5076of the wood of all trees felled.
Almost no use is
made of this wood because it is too large to be cut by hand
methods, too heavy to carry by head portage and often located
in remote areas. ?t has been suggssted that mobile sawmills
oould answer this problem. The felled timber would then be
uut into panageabls proportions for use as firewood or for
charcoal burning, One proposal combines the mobile sawmill
with a mobile chmcoal kiln.
The tipediment to this type of
development would appear to be that the low value of firewood
does not allow for the cost of sawing. In the populus South
of Ghana, most firewood is regarded as free and a lorry Toad
0f off-cuts
frc3
a aatill
Costa
only g!25.00 ($22.00) for
soft wood and $45.00 for hard wood. Until this situation
changes, it
will
eeems to be unlikely
that a~qy rapid
development
occur in wood waste utilisation.
It cannot be doubted that the woDd waste of Ghana
represents a large reservior of energy which can be tapped
when it is needed, firstly
by making use of present losses
and secondly by introducing more efficient
methods of
utilisation.
Perhaps the most immediate prospect lies in
the introduction of more industries using wood as fuel.
It
is perhaps surprising that no wood burning steam engine is
known in the Country. The railways use steam engines burning
impcrted coal and, increasingly,
diesel locomotives. Unlike
East Africa where wood burntig steam engines were supplied
with fuel from specially prepared plantations, Ghana never
seems to have made use of its wood resources in this wsy.
There ms;ybe muoh resistance to the introduction of steam
engines for transport or industrial use at the present time.
Huuever, the possibility
msy intrigue some practitioners
of
intermediate technology who msy find some applications where
steam power [email protected] be both technically and economically
feasible.
The possibility
of power generation using charcoal,.
residual oil or combustible gss from wood carbonisation or
distillation
might also be explored. At a time when so much
foreign exchange is expended on petroleum products, the
incentive to produce fuels and lubricants from indigenous
raw materials is very @eat.
There csn be little
doubt that the further utilisation
of wood waste could reduce importation, support new industries end even boost exports to the great economic advantage
of the Country,
SOMEAPPROXIMUE STATISTICS
Precise data are not available on many aspects of
energy consumption and wood utilisation
in Ghana. The
following
data should be taken as only very approximate.
In
particular,
no reliable
data is available on the consumption
Estimates of the quantity of charcoal produced
of firewood.
A lower estimate is favoured as a
vary by a factor of 4:l.
higher one could suggest a gross utilisation
of wood as
charcoal and firewood in excess of the total available whereas
all investigators
agree that a large proportion remains unused,
Power Consumption in Ghana
Source
Hem Power Consumption (MW)
Imported oil and coal
Zydro-electricity
..
1050
l
.
..
Firewood (Heat of Combustion)
Charcoal (Heat of Combustion)
TOTALI
Annual 'nood ctilisation
..
1035
.
..
115
2650
in Ghana
Volume (Millions
Use
Export . .
..
..
Local Use as Timber
Charcoal Production
Firewood
Unused . .
450
.*
..
..
..
..
..
..
..
&,.
TOTAL:
..
.*
*a
..
..
..
1.1
0.2
1.6
4.0
3.1
10.0
of m3 )
Job-n 5’. PgweZi!
128
1.
K. SIWEK,
'Developmsnt of Charcoal Production
ii1 Ghan&', Ghana Investment Centre,
Investment Journal Vol.5 No.2,
April/June, 1974.
2.
s. ADDO,
ISoms Problems in the
the WoodResources of
Paper No+10 - Seminar
of the WoodResources
August, W5.
3.
J. ASMA&
'Species Differentiation
and
Manufacturing Techniques as Affecting
Marketing'.
Paper No.3 of above Seminar.
4.
S. A. OFFEI,
wCharcoal Production and the Resources
in Gh&na'.
Paper No.12 of above Seminar.
5.
A. WEXSING,
tPlsnni.ng for Charcoal Burners in
Kumasi'. Thesis in part Fulfilment
Requirements for B.Sc. in Urban
Planning, Faculty of Architecture,
U.&T., Kumasi.
of
Report on Charcoal Production in
Ghana by the Capital Investment Board
(C.I.B.), Accra, November, 1973.
6.
7.
Utilisation
of
Ghana'.
on Utilisation
of Ghana,
A
T.
J.
3.
J.
I. CHIANG,
W. TATCM,
W. S. de GRAFT-JORBSON,
W. KMELL - 'Pyrolytic Conversion of Agricultural
and Forests Wastes in Ghana - A
Feasibility
Study1 Report prepared for
U.S.A.I.D. by the Economic Development
Laboratory of Georgia Institute of
Technology.
Methane from Human, Animal
and Agricultural Wastes
RaymondC.
Loehr
Introduction
In considering
alternatives
to meet the energy needs in
there ze several key items that should
developing
countries,
These arc: a) in general current
be recognized
and achieved.
methods of producing
energy for domestic use, such as burning
of cattle
dung, crop residues,
and wood, are inefficient;
b)
less plant fuels should be used for combustion to avoid denuding af forests
and to maintain
soil tilth;
c) there should
d) human drudgery should be
be higher agricultural
yields;
reduced; e) the energy that is produced should be storable,
expense should be -..-*---..minimllr[l.
In
ad f) cagizal
and firkarating
VTaddition,
the technologies
that are used should be environmentally
sound and adapted to and compatible
with the local
economic, social,
and political
situation.
The most appropriate
technology
would serve the dual purpose of resource
conservation
and environmental
protection.
Although there is no "ideal"
technology
that can meet
all of the above goals, methane generation
is a waste management technology
that comes close to achieving
many of the
goals.
The utilization
of human, animal and agricultural
wastes for methane generation
has many positive
aspects,
including:
a) the production
of an energy resource that can
be stored and is independent
of fossil
fuel supplies;
b) the
creation
of a stabilized
residue
(sludge) that retains
the
fertilizer
value of the original
material;
and c) the saving
of the amount of energy required
to produce an equivalent
amount of nitrogen-containing
fertilizer
by synthetic
processes.
Other positive
but indirect
benefits
of methane
generation
include public
health such as a reduction
of
human pathogens i f human -rastes are used and a decrease of
plant pathogens that may be associated
with the crop
residues.
By using human, animal, and agricultural
wastes
129
130
Faymond C. Loehr
for methane generation,
the additional
value cf the
can be gained while realizing
other benefits.
The
nutrients
in the digested material
can be returned
soil as a fertilizer
and to improve soii structure
organic matter content.
methane
original
to the
and
In many countries
in which methane is produced for
domestic or farm use, the drudgery of collecting
wood, dried
dung, or crop residues and the smoke of burning these
In addition,
when applied
to
materials
has been reduced.
fields,
the stabilized
wastes resulting
from the methane
generation
have helped increase agricultural
yields,
control
and maintain
desirable
soil characteristics.
erosion,
Background
The interest
in non-fossil
fuel energy supplies
has
focused increasing
attention
on the production
of methane
from the anaerobic
fermentation
or digestion
of organic
maWcr such as wastes since methane is a high energy byThe production
02 methane from wastes is centuriss
product.
The largest
old and the general technology
is well known.
application
of methane generation
has been with municipal
sewage sludge.
Anaerobic digestion
has been widely applied
in municipal
and industrial
wastewater treatment
plants
for
biologically
stabilizing
organic solids,
reducing siudge
volumes requiring
ultimate
disposal,
odor control
and recovery
of the resultant
methane for plant heating and energy
requirements.
In countries
with low natural
energy supplies
and/or
wnere there is concern over the loss of fertilizer
value by
alternative
uses of agricultural
wastes such as burning,
wastes to meet the
methane has been generated from available
existing
needs.
Extensive work on methane generation
has
occurred in many countries,
especially
India;
Taiwan, the
United States,
and Europe and several extensive
reports
have
been published
describing
design, construction,
operation,
and results
(l-5).
,
Individual
family methane generating
units have been
used in diverse climatic
and cultural
conditions.
Several
thousand units are operative
in Taiwan using primarily
pig
from individually
owned and operated
manure and by-products
pig fa.zms, Crop residues and cow dung have been used in
family and village
operated units in India for decades.
Over
30,000 small methane generation
units have been installed
in
India using some 66.9 million
tons of manure which otherwise
would be burned.
During 1974-75 alone, some 10,000 new
methane plants were installed
with a goal of 100,000 being
..
Methane from Human, Animal, and. AgYlicuZturaZ Wastes
Units also
installed
by 1980 (6).
ing in Korea, the Peoples Republic
other countries
of Asia and Africa.
131
are reported
to be operatof China, Uganda and in
All organic wastes can be fermented to produce methane.
Examples of animal and agricultural
wastes that can be used
to produce methane are noted in Table 1. Those that have
been studied or utilized
most extensively
are animal wastes,
food processslaughterhouse
wastes, and wastes from certain
Much more information
is available
on the
ing industries.
digestion
of manure and human wastes than on crop residues,
Factors
Affecting
Methane Production
The quantity
and composition
of the
Biodegradability.
gases produced during anaerobic digestion
are a function
of
the fraction
of the total waste that is available
to the
anaerobic bacteria,
i.e.,
the biodegradable
fraction,
and the
The more
operating
environmental
conditions
of the process.
the greater
the quantity
of
biodegradable
the waste material,
methane generated per quantity
of waste added to a digester.
Table 2 indicates
the estimated
gas production
from the
manure of various animals produced under conditions
in the
United States.
As much as 8 to 9 ft3 of gas (containing
60 to 70%
methanej can be produced per ound of volatile
solids added
3
to the digester
(0.5 to 0.6 m /kg) when the organic matter is
highly biodegradable
such as untreated
human wastes or fresh
manure.
Not all wastes are equally
biodegradable
and effective in producing methane.
The biodegradable
fraction
of a
waste will vary being a factor of the characteristics
of the
material,
the food isagested to generate the human or animal
wastes, and how the bastes were handled prior
to digestion.
For examplep only 4 ' to 50% of the volatile
solids of dairy
cattle
manure may b: biodegradable
and thus available
to produce methape.
To u%e anaerobic
digestion
effectively,
the
such as sand and dirt in wastes
inclusion
of inert material
should be minimize d and wastes as fresh as possible
should be
utilized.
When a/uaste is exposed to the natural
environment, such as by .$ying on the ground, natural
degradation
of
the organic fract,iion takes place and the biodegradable
fraction will decrea$e,
In addition,
significant
losses of
nitrogen
will o&r.
The 3ata in Table 2 incorporate
average values obtained
from the literature
for manure production
and biodegradability.
The highest amount of gas produced per 1000 pounds live
weight was from chickens,
indicating
the higher amount of
biodegradable
organics
in that material.
Under U.S.
'.
Table 1. Agricultural
generation
- Animal
litter,
residues
wastes including
and manure
- Crop wastes:
sugar
and spoiled
fodder
- Slaughterho\o;e
meat, t:lshery
having
bedding,
cane trash,
JciS'r.~3, animal
wastes, leathe,,
potential
wasted
weeds,
fcr
feed,
crop
methane
poultry
stubble,
straw,
by-products
such as blood,
and wool wastes
- By-products
of ayricultcral
bat?c; induciries
such as oil
cakes, wastes from fruit
and v etcble processillg,
bagasse
and press-mud from sugar factories,
sawdust, tobacco wastes
and seeds, rice bran, tea waste. and c>tton dust from
textile
industries
- Forest
litter
- Wastes from aquatic
and water hyacinths
Table 2. Estimated
wastes (7j
growths
manure and bio-gas
Manure production
(lb/day/l000
lb live
weight)
Volatile
solids
(lb dry solids/day/1000
live weight)
Digestion
efficiency
manure solids
(%)
x
0.454 = kg:
algae,
production
seaweeds,
from animal
Dairy
Cattle
Beef
Cattle
Swine
Poultry
85
58
50
59
ib
5.9
8.7
5.9
12.8
of the
Bio-gas production
(ft3/lb
VS added)
lb live weight/day)
(ft3/1000
(lb
such as marine
ft3/lb
35
50
55
4;. 7
40.8
6.7
39.5
7.3
43.1
x 0.062 = m3/kg)
65
8.6
110.9
Methane fi-om Humun, Animal-, and Agricultural
Wastes
133
conditions,
the minimum animals to produce the energy equivalent to one liter
of gasoline
is approximately:
2 beef
3.2 dairy cattle,
330 chickens,
and 16 swine.
These
cattle,
values and those in Table 2 depend on the weight and feed
The
ration
of the animals and will
vary over a wide range.
the gross energy possible
from
data in Table 2 estimates
digestion
and does not include
energy required
to operate the
process.
The ratio of carbon to nitrogen
(C:N) in the raw mateNitrogen
rials
is important
for the production
of methane.
is needed for the synthesis
and activity
of the microorganIf there is insufficient
isms in the anaerobic process.
nitrogen
to allow the anaerobic bacteria
to reproduce,
carbon
For optimum
dioxide
will be the principal
gas produced.
methane production,
the C:N ratio
should be below about 30-35.
When largely
cellulosic
material
is being contemplated
for methane generation,
such sawdust or crop residues,
nitronitrogen
rich manures, or urea may
gen in the form of urine,
have to be added to have the C:N ratio
in the feasible
range
for best methane production.
If the waste mate-,-al contains
quantities
of nitrogen
such that not all of the nitrogen
will be used for synthesis
of the anaerobic microorganisms,
the ammonia concentration
If excessive nitrogen
is contained
in the
will
increase.
may reach levels that are
wastes, the ammonia concentration
inhibitory
to the organisms and gas production
will
slow or
even cease.
Should such conditions
occur, the ammonia concentration
should be reduced by removing portions
of the digester contents,
adding water to dilute
the remaining
ammonia concentrations
and adding carbonaceous materials
to
the feed materials
to restore
the C:N balance.
Types of
wastes that have a high nitrogen
content are fish scraps,
blood, and fresh poultry
manure.
.
It sho,lld be realized,
however, that ik organic material
has little
biodegradable
content and is resistant
to microbial action,
it wll not be digested
even if it has a favorable C:N ratio.
sources may
In summary, wastes from similar
yield quite different
quantities
of gas.
In interpreting
the
feasibility
of waste materials
for methane production,
both
the C:N ratio
and the relative
ease of degradation
of the
material
needs to be known.
Caution is advised in examining
results
or data obtained
from systems using residues having
characteristics
different
from those under consideration.
The biodegradable
fraction
may be different
and the residues
may have components that may cause inhibition
of the digestion process.
When wastes for which there is no specific
134
RaymondC. Loehr
data are being considered
for methane generation,
investigations should be made to determine
the operating
conditions
Knowthat are needed and the gas generation
that will
result.
ledge of the gas production
per unit of wastes added to a digester is important
to the entire
feasibility
of methane generation
since it determines
the quantity
of gas available
for
utilization
and the required
size of the system.
The oxidation
of organic carbonMicrobial
Reactions.
containing
compounds represents
the mechanism by which heteroactivity
and for
trophic
organisms obtain the energy for their
In aerobic treatment
systems
the synthesis
of new organisms.
via many steps to synthesized
organic carbon is transformed,
microbial
protoplasm,
C5H,02N, and carbon dioxide.
Organic
The uptake
the effects
carbon
+ O2 + C5H,02N + CO2
of oxygen and formation
of respiration.
of carbon
(1)
dioxide
represent
oxygen is not the terIn anaerobic
systems, molecular
minal electron
acceptor and all of the respired
carbon will
Under anaerobic condinot be transformed
to carbon dioxide.
tions,
organic carbon is converted
to microbial
solids,
carbon
Anaerobic
dioxide,
methane, and other reduced compounds.
metabolism
leading to the formation
of methane occurs in a
series of steps.
For simplicity
these can be sununarized as
the conversion
of complex organics
to simpler compounds:
Organic
carbon + microbial
organic
and the conversion
products:
Organic
acids
acids,
aldehydes,
of the simpler
+ oxidized
microbial
cells
cells
f
alcohols,
etc.
(2)
compounds to gaseous end
organic
carbon +
+ methane + carbon
dioxide
(3)
Little
stabilization
of organic matter occurs in the
Stabilization
of the organic matter occurs in
first
step.
the second step in which the carbon compounds, carbon dioxide
(C02) and methane (CH ), are released to the atmosphere and
The oxygen demand of the waste
removed from the subs 0 rate.
At standard conditions,
the production
of
is thu reduced.
5
5.6 ft
of methane results
in the stabilization
of 1 lb of
ultimate
oxygen demand. Methane generation
is a process that
results
in the stabilization
of the wastes, i.e.,
reduces the
biodegradability
of the residual
material.
Methane frarn &man, An&z!,
and AgricuZturaZ hastes
135
These reactions
must occur simultaneously
since if the
reactions
become seriously
unbalanced,
the process can fail.
referred
to as volatile
acids,,
The organic acids, generally
produced during the first
stage of the fermentation
process
tend to depress the PH. This effect
is counteracted
by the
destruction
of volatile
acids and reformation
of bicarbonate
buffer
during the second stage.
If volatile
acids build up
in the system, the buffer
capacity
may be overcome and a precipitous
drop in pH may occur.
Buffering
with an inexpensive
base such as lime may be necessary to return the system to
equilibrium.
Other bases can be used but care must be exercised to avoid chemicals that may result
in other inhibitory
conditions.
The bacteria
in the anaerobic
systems are sensitive
to
changes in pH. The optimal pH range for methane production
is between 7.0 and 7.2, and gas production
is satisfactory
between 6.6 and 7.6.
When the pH drops below 6.6 there is a
significant
inhibition
of the methanogenic
bacteria,
and
acid conditions
below a pH of 6.2 are toxic to these bacteria.
Under balanced digestion
conditions,
the biochemical
reactions tend to automatically
maintain
the pH in the proper
range.
Methane bacteria
are sensitive
to other environmental
factors.
Because they are obligate
anaerobes, their growth
is inhibited
by small amounts of oxygen and it is essential
that a highly reducing environment
be maintained
to promote
their growth,
Not only oxygen, but any highly oxidized
material,
such as nitrites
or nitrates,
can inhibit
the
methanogenic bacteria.
If chemical additives
are required
to help balance a digester
or provide nutrients,
they should
be added in the most reduced forms.
The concentrations
af any additives
should be maintained
below concentrations
that may be inhibitory.
A number ot
chemicals have been shown to be toxic to the anaerobic bacteria.
The toxicity
can result
in a reduction
of gas production,
an increase in volatile
acid concentration
or both.
Examples of inhibitory
compounds are:
ammonium ion (>3000
mg/R) of total
ammonia nitrogen
at any pH; soluble
sulfites
(>50-200 mg/R); and soluble salts of metals such as copper,
zinc, and nickel.
Alkaline
earth-metal
salts,
such as those
of sodium, potassium,
calcium,
or magnesium, may be either
stimulatory
or inhibitory,
depending on the concentration.
With the exception
of possibly
ammonia, the above inhibitory materials
are not likely
to cause problems with "typical"
human, animal, and agricultural
wastes.
Such wastes rarely
contain the above compounds in amounts likely
to be inhibitory.
136
AQqr~;mdC. Lcehr
One exception
might be when the hJnan wastes are combined
with municipal
wastes that include wastes from industry.
Certain industrial
wastes can add chemicals in amounts that
can be inhibitory
to anaerobic
digestion.
Because methane production
is a microbiological
the factors
affecting
the process must be understood
the selection
of the material
for the process and in
operation.
Conditions
that inhibit
the process must
avoided.
process,
in both
its
be
Time and Temperature.
The micrcbial
reactions
require
time to be completed.
Time is needed to degrade and solubilize
the complex organic compounds in the waste and convert
the soluble compounds into the gaseous and other end products
(Equations
2 and 3).
This time is related
to the rate at
which microorganisms
metabolize
the wastes and that rate,
in
turn, is a function
of the temperature
of the system.
Thus
time and temperature
are interrelated.
The measure of time usually
used for design i)r operation
of biological
waste treatment
systems, including
anaerobic
digestion,
is the solids retention
time (SRT) sometimes
referred
to as MCRT (mean cell residence
time) or 0 by
various authorsSRT is the time that the microbia?
mass is
retained
in the biological
system.
Fundamentally,
SRT should be determined
using the quantity
of active microorganisms
in the system.
However,
measuring the active microbial
mass in biological
treatment
systems is difficult.
Fortunately,
other parameters
can be
used.
Assuming that there is a uniform distribution
of the
active microorganisms
and other solids in the methane unit,
SRT can be determined
by using the quantity
of other solids
in the system, i.e.,
volatile
suspended stilids,
total
suspended solids,
or total
solids.
In practice,
the SRT of the
system can be determined
by:
SRT =
weight of solids
in the system
weight of solids
leaving the system/time
(4)
The actual SRT of a biological
treatment
system must be
greater
than the minimum time it takes for the microorgansims
to reproduce in the system.
If this does not occur, the
microorganisms
will
be removed from the system at a faster
rate than they can multiply
and failure
of the system will
result.
Minimum SRT values for anaerobic
systems have been
estimated
to be in the range of two to six days.
Methane from Hman., Animal,
atzd Agricultural
C;'clstes
137
The temperature
of the anaerobic
digester
will
affect
its performance
since it affects
the activity
of the microThe optimum temperature
of mesophilic
anaerobic
organisms.
treatment
is 30°C to 37*C. Although the time to obtain a
desired degree of treatment
is less with thermophilic
treatthermophilic
conditions
ment than with mesophilic
treatment,
have not been demonstrated
as practical
or economic to date.
It does not follow that anaerobic
treatment
must occur
Anaerobic activity
can
at optimum mesophilic
temperatures.
occur if an adequate mass of active microorga:lisms
and a
sufficiently
long SRT are provided
for the system. i At rethere will
be less active organisms in
duced temperatures,
the system.
Total gas production
may be reduced if the
system is operated at low temperatures.
An example of the
relationship
between SRT, temperature,
and gas production
that was obtained
in cne study is presented
in Figure 1.
Only in very large digesters
is it feasible
to consider
heating to maintain
a controlled
temperature
at all times.
Gas utilization
must be planned with the expectation
of
smaller amounts being generated during the colder periods.
The SRT can be increased by having large enough units
so that the input wastes are held long periods of time or
by separating
the solids from the discharged
material
and
recycling
them to the anaerobic
system (9).
For locations
lacking
skilled
manpower having the knowledge needed to
operate a complex anaerobic
system, solids recycle
is not
likely
to be used.
In practice,
the temperature
range of 25* to 35*C and
a solids retention
time of 10 to 15 days appears to be the
most convenient
and economic.
Reasonably constant temperatures are important
to the process.
A system with a larger
SRT than necessary has only increased
digester
size and
system costs without
obtaining
appreciably
more methane per
unit of biodegradable
solids.
It may be difficult,
especially
in colder climates
and
in the winter,
to maintain
mesophilic
temperatures
in the
anaerobic digester.
Supplemental
heat can be used to maintain the necessary microbial
activity.
Such heat can come
from the methane that is generated but will
obviously
reduce
the amount of methane available
for other purposes.
The
anaerobic digestion
unit should be well insulated,
either
artificially
or by constructing
the unit in the ground.
.
238
RaymondC. Loehr
time (days)
Figure 1. Bio-Gas Production
as Related
of the Digester
and the Time of Digestion
to the Temperature
(7)
Methane from Humax, Animal, and AgrieuZt-ural
Sludge
Wastes
139
Wilizaticn
The resi.Jue from a methane generation
process will
conmaterial
protected
from
bacterial
degratain liqnin,
lipids,
synthesized
microbial
cells,
metabolic
degradation
dation,
products
such as volatile
acids and other soluble
compounds,
inert material
in the original
waste, and water.
The residue
will
be a liquid
with a solids concentration
of 4-8 percent.
Anaerobically
digested
sludges can be stored and spread on
land with less risk of creating
conditions
for odor and
insect breeding problems than exists with similar
handling
or partially
treated
organic waste
procedures
for untreated
materials.
Methane generation
conserves the nutrient
elements
needed for crop production.
The only materiais
removed from
the system, other than in the sludge, are the generated gases.
Practically
all of the nitrogen
present in the waste entering
a digester
is conserved.
If the sludge is properly
stored,
and when applied to soils is immediately
incorporated
to
reduce the loss of nitrogen
by volatilization,
most of the
nitrogen
present in plant residues
can be available
for use
by the growing plants.
To minimize
ammonia nitrogen
losses,
the digested
sludge should be stored in lagoons or tanks
which present a minimum of surface area for ammonia evaporation.
Other chemical elements contained
in the added waste
will be conserved in the digested
sludge.
The end result
of applying
digested
sludge on soils is
the same-as that resulting
from the application
of any other
kind of organi 7 matter.
The humus materials
can improve soil
physical
properties
such as aeration,
moisture
holding
capacity,
increase cation exchange capacity,
and improve
water infiltration
capacity.
The sludge can serve as a
'source of nutrients
for crops grown on the soil.
When human,
animal, and agricultural
wastes are used for methane generation,
there is little
likelihood
that any items in the sludge
will
cause adverse conditions
to the crops or to animals fed
the crops grown on land where digested
sludge is used as a
fertilizer.
The application
of the digested
should be done in an environmentally
at rates consistent
with the need of
Runoff that can contaminate
surface
result
in ground water contamination
sludge to the crop land
sound manner, generally
the crops being grown.
waters and loadings
that
must be avoided.
An important
aspect to be considered
with methane generation is that the total
volume of sludge that must be handled
for final
disposal
is equal to or greater
than the initial
140
i?i’aymoizd
C. Leehr
amoclnt of dry waster that are to be digested
because of the
liquid
addE I to obLTin a solids concentration
that can be
Although cc:.:iderable
solids decomposition
occurs in
mixed,
a digeC+::r,
Lpproximately
50%, little
reduction
of the total
voluIRP LO be handled results.
-
-.A=
* Utilization
Key items in the successful
operation
of a methane generation
system are: (a) acceptance by the potential
user, (b)
ability
to use the gas when produced,
(c) sufficient
demand
for the (3as, (dl enough raw material
available
to meet the
production
requirements,
and (e) sl .&able maintenance
and
operdtron3l
control.
Pure ??thane is a colorless
an4 odorless
gas which
<:r?nerally constitutes
between CO and 70% of the gas produced
1:~ anaerobic digestion.
The ~-:t;zrr constituents
a::e primarily
carbon dioxide
and smaller concentrations
of other gases
such as hydrogen sulfide
and hydrogen.
Digester
gas
(biogas) burns with a bsue flame and has a heat3value
ranging
about 600 to 700 Btu/ft
(22,000 to 26,000 kj/m ) when
its methane content ranges from G3 to 70 percent.
Many options exist for utilizing
the digester
gas.
It
can be used directly
in gas-burning
appliances
for heating,
cooking,
lighting
and refigeration
or it can be used as fuel
for internal
combustion engines having a compression ratio of
8:l or greater.
However, to use digester
gas in internal
combustion engines,
it $s necessary to a) reduce the.hydrogen
sulfide
c&tent
of the gas to less than 0.25 percent to prevent corrosion
of metal surfaces,
'bj provide a system to
remove the carbon dioxide
to increase the heat content per
unit ,volume of the resultant
gas, c) compress the gas to a
volume that would fit on a vehicle
or adjacent
to a stationary
engine, and d? have an internal
combustion engine converted
to use either
gasoline
or methane.
Another possible
use is the production
of electricity
using methane.
However, the cost of converting
the methane
to electricity
currently
results
in an energy cost that is
higher tha:t that available
from other sources.
As the cost
of more traditional
energy sources increases,
the costs of
electricity
production
from methane may hecome more attractive.
For small units,
it is not practical
to compress
store the gas for subsequent use.
Major use of such
be at the production
site.
The gas is generally
and
simply used for heat or light
in appliances
that us~t
directly
from low-pressure
collection
units.
A small
and
gas must
most
the gas
digester
Metnane from Human, AnimaZ, and AgricuZturaZ Wtistes
141
with a floating
cover can provide the needed gas storage.
Another approach is to have gas from a digester
with a fixed
cover piped for collection
in an auxiliary
gas holder with a
A gas delivery
line is connected to the gas
floating
cover.
storage unit with an on-off
control
valve.
This delivery
line must contain a flame trap installed
between the gas
storage unit and the appliance
being used.
Systems of this
type are in current use in India,
Taiwan and other warm
climate
areas.
The size of-the digester
and gas collection
unit is
related
to the rate of gas that is utilized
and the number
of appliances
that are connected.
Consumption rates for gas
burning appliances
(Table 3) are useful in determining
the
necessary sizes.
Although general
digesters
is available,
digestion
of agricultural
ducers can successfully
utilize
the resultant
full
scale testing
in
show the practicality
the anaerobic digestion
agricultural
wastes.
Collection
information
on the design of anaerobic
the state-of-the-art
of anaerobic
wastes is not such that all prooperate an anaerobic digester
nor
gas.
Considerable
investigation
and
specific
situations
may be needed to
and feasibility
of energy recovery from
of the available
human, animal,
and
and Preparation
An important
aspect of methane generation
is the collection and preparation
of the materials
tc be used in the process.
When considering
the feasibility
of a methane production system, all of the components, i.e.,
collection,
preparation,
storage of the raw material,
generation,
storage
of the gas and residue,
and utilization
of the gas and residue
must be considered.
Capital
and labor requirements
and annual
costs must be determined
for all the components and related
to the local labor, material,
and cost conditions.
In labor-intensive
economies, methods that can utilize
the human and animal resources
available
for the handling
and
processing
of these wastes should be considered.
Since the
intent
of anaerobic digestion
is to produce an energy source,
methods that utilize
fossil
fuel or other conventional
sources
of energy for handling,
processing,
heating and mixing these
wastes may not be appropriate
unless there is a significant
net benefit.
The collection
and processing
of raw waste materials
depends on their nature,
which may vary between countries
and
regions,
as well as the quantities
in which they are available.
142
RaymondC. Loehr
Table 3. Quantities
application
(10)
of bio-gas
required
for
a specific
Quantity
Use
Specification
of Gas Required
ft3/hr
m3/hr
Cooking
2" burner
4" burner
6" burner
11.5
16.5
22.5
0.33
0.47
0.64
Gas lighting
per mantle
2 mantle lamp
3 mantle lamp
2.5-3.0
5
6
0.07-0.08
0.14
0.17
16-18
per hp
Refrigerator
Incubator
per ft3
capacity
1
0.45-0.51
per hp
0.028
per ft3
capacity
0.45-0.6
0.013-0.017
Gasoline
1 liter
47-66tb)
0.013-0.017
1 liter
53-73(b)
l.50-2.O7(b'
1 liter
2.2(c)
0.62(')
Diesel
Boiling
fuel
water
(4 Based on 25 percent
(b) Absolute
equivalent
efficiency
volume of bio-gas needed to provide
of 1 liter
of fuel
(cl Absolute volume of bio-gas
water to boiling
needed to heat
energy
1 liter
of
Methane from Human, Animal, and AgricuZturaZ Wastes
143
Hence the method for collecting
and handling
the waste
The materials
available
for methane
materials
may vary.
generation
may be solid,
semi-solid
or liquid.
Thus the
collection
methods that will be needed will be related
to
the characteristics
of the wastes and the socially
acceptable
methods of handling
them.
Crop residues such as spent straw, hay, sugar cane trash,
and bagasse can be used for the qenplant stubble,
grasses,
To enhance gas production,
facilitate
eration
of methane.
their mixing and transport
in and out of the digester,
and to
such material
should be finely
chopped or
avoid clogging,
In rural areas manually operated shredders could
shredded.
be used.
When different
wastes are to be added, they should be
mixed prior to their
addition
to the digester.
The wastes
should be added daily to the digester
to insure a continuous
supply of gas and to avoid a deterioration
of the methane
generation
process.
Accumulated wastes can be stored briefly
before being added to the digseter.
Care should be taken to
avoid losses of nutrients,
fly breeding,
or odors during the
storage.
Equipment
The main components of a methane generation
system are
a raw waste feeding unit,
a digestion
unit,
a gas holder,
a
moisture
trap, and an outlet
for the gas to be utilized
and
the digested
solids.
A sketch of a small scale system for
heating and lighting
use is shown in Figure 2. For larger
systems and other uses, a methane generation
system may include a raw waste storage unit and units for hydrogen sulfide
and carbon dioxide
removal and gas compression.
The pieces of equipment can be made from many types of
material.
Where available,
brick,
cement, concrete,
and
steel pipe can be used.
The gas holder may be a steel drum.
Lime mortar can be used if cement is in short supply.
Other
material
such as glazed pottery
rings cemented together
and
similar
local materials
can be used to construct
the needed
equipment.
Hand plungers
can be used to force the feed
material
into the digester.
Both horizontal
and vertical
digesters
have been used with vertical
units being predominant
in countries
where biogas plants are in operation.
For small
systems and needs, large inner tubes have been used as gas
collectors.
The simplest
system consistent
with the need and
labor available
should be constructed.
Materials
that are
long lasting
and require
minimum maintenance
should be used.
144
Raymnd C. Loehr
GAS
GAS HOLDER
ERFLOW
Figure
2.
Schematic
of a Small
Scale Bio-Gas
Digester
idethane from Human, Animal, and AgricuZturaZ Wastes
145
When the raw wastes are added to the digester,
the
digested
slurry
automatically
overflows
and should be captured
for subsequent use as a fertilizer.
The slurry
can overflow
into a container
filled
with straw, leaves or similar
material and used as a solid or semi-solid
fertilizer.
It can also
be collected
as a liquid
and used in that form.
Operation
In a developing
country and a rural setting
it is unlikely
that a continuous
anaerobic digester
would be used,
because of the necessary feeding and controlling
mechanisms.
A batch or semi-continuous
operation
is the more common
method of operation.
A digester
may operate on a "fill
and
draw" basis with a one-day to one-week cycle, or it may be
designed on an "all-in,
all-out"
basis where the reactor
is
charged and emptied when gas production
is almost completed
or at a very low level.
A combination
of the two systems
also may be considered
depending upon particular
needs.
The
semi-continuous
digesters
are used for waste that is available
on a frequent,
preferably
daily,
basis.
Most digesters
in
developing
countries
operate as semi-continuous
units.
Batch-fed
digesters
can be constructed
where raw waste
materials
are difficult
to obtain on a daily basis.
The
wastes are added to the digesters,
covered and permitted
to
digest.
After an initial
interval
that will
depend upon the
temperature
of the digester
and whether the digester
has been
seeded with active anaerobic organisms,
gas production
will
begin and continue until
all the biodegradable
material
is
When ga? production
ceases the digesters
are opened,
used.
cleaned,
and the slurry
is disposed of on land as a fertilizer.
Because of the batch nature of the system, it is desirable
to
have two or more digesters,
so that one or more can always be
in operation
and gas production
be fairly
continuous.
Periodic
emptying of the digesters
is labor intensive
and can
be unpleasant.
When batch units are emptied, about one
quarter
of the digester
contents
should be left
to seed the
incoming raw material.
Biogas production
is dependent upon the concentration
of
solids
in the influent
material.
Too thick of a material
will
retard natural
mixing in the digester
and may require
mechanical
mixing.
Investigations
indicate
that,
in the
absence of toxic materials,
optimum gas production
is obtained
with a total
solids
concentration
of about 7 to 10% in the
effluent
slurry.
This may require
a dilution
of one part
manure to two or three parts of water depending upon the
initial
condition
of the manure.
146
RaymondC. Loehr
Economics
--
and Feasibility
Although methane generation
has been successful
in
specific
areas such as in India and Asia, economic considerations may preclude
or limit
its adoption in other countries.
The economic feasibility
of methane generation
from wastes
can vary widely and is dependent on factors
that include the
availability
of domestic sources of energy, the cost of
the uses and actual benefits
from methane proimported fuel,
public
and private
costs associated
with the developduction,
ment and utilization
of methane, the availability
of wastes
to produce a consistent
supply of methane, and on the equipment and manpower need to generate methane.
The availability
and accessibility
energy sources and the opportunities
to
important
determinants
of the feasibility
tion.
The opportunity
costs associated
energy sources, such as the use of coal
industrial
expansion,
need to be taken
of other domestic
use the methane are
of methane producwith the use of other
or petroleum
for
into account.
Studies in industrialized
countries
with access to fossil
fuels indicate
that methane generation
is not economical even
for farm and small communities
with the present technology
and
energy costs.
Methane generation
is likely
to have its
greatest
use in areas remote from fossil
fuels or other energy
sources, where available
fuels are better
used for other
purposes,
and for small villages
and activities
where the
methane is used at or close to the production
site.
Generally,
the cost of a methane generation
system and
its operation
has been charged against energy production.
The direct
benefits
are those accruing
from heating,
cooking,
lighting,
refigeration
, pumping irrigation
water, or running
other power units.
However, other benefits
also may accrue
as a result
of methane generation,
such as improved public
health,
agricultural
productivity
increases,
and increased
employment.
The use of village
scale digesters
can provide
a sanitary
means of human waste disposal
that might otherwise
be lacking.
These benefits,
although
less tangible
than
direct
energy use should be taken into account when distributing the costs of methane generation.
In addition,
for environmentally
methane generation
the waste handling
normally
occur.
the wastes would have required
some expense
sound management.
The costs charged to
should be those that are in addition
to
and pollution
control
costs that would
Methayz from Hwmn, Animal, and AgricuZturaZ Wastes
147
In order for a methane generation
program to be successful in most developing
countries,
some financial
and continuing technical
assistance
from central
and local governments
Part of the material
and equipment cost of
will be needed.
individual
methane generators
might be the responsibility
of
Trained technical
personnel
central
and/or local governments.
can be important
to help the local people learn to install,
The importance
of
and maintain
methane digesters.
operate,
Demonstration
technical
assistance
should not be minimized.
and early adopters may need to be
facilities
may be required,
subsidized
as an incentive
to aid the large-scale
adoption of
The particular
educational
means to obtain
this technology.
wide scale use of methane generation
can vary among regions
and countries.
It is technically
feasible
to generate methane from
wastes if the factors
dishuman, animal, and agricultural
The
cussed earlier
are understood and properly
utilized.
economic feasibility
of methane generation
will
depend upon
the particular
set of circumstances
and can only be evaluated
on an individual
basis.
In addition
to the cost of the eqtlipment, manpower, disposal,
and opportunity
costs of alternatives,
of particular
interest
are the needs and goals of the
areas, countries,
and regions,
and the impact on the culture
and social
fabric
of the individuals
and communities
to be
involved.
Methane generation
must be considered
as part of a total
resource conservation
and utilization
system.
Factors such
as availability
of the wastes, collection,
transport,
methane
generation,
gas and sludge use, energy needs, manpower, and
the social and cultural
effect
need to be considered
in an
inter-related
and holistic
manner.
Only in such a manner can
the actual feasibility
of methane generation
in a developing
country be ascertained.
Extensive
experience
with methane generation,
especially
at the village
or smaller
level,
has been obtained
in several
countries
and over several decades.
Methane generating
plants
can be designed either
to process a given amount of waste
material,
or to produce a given quantity
of gas required
for
a specific
use or uses.
Irrespective
of which approach is
taken, an understanding
of the parameters
that govern the
process of anaerobic
digestion
is essential
for proper design
and use of biogas plants.
Digester
design, waste input,
and gas utilization
must
be tailored
to the manpower and other resources,
energy and
148
RaymondC. Lcehr
eZK9l
conditions,
and materials
social needs, climatic
nsideraAll of these factors
must receive close
locale.
system is
tion when the feasibility
of a methane generatic
A technical
assistance
program may i needed in
evaluated.
areas where methane generation
is determined
t
feasible.
It is desirable
to have local involvement
in v
-lanning,
construction,
and operation
of the digester
2
:I the utilization of the gas and the sludge.
Use of s::
.E local
materials
and equipment should be investic
-cu.
Before methane generation
is utili:
in any area or
with specific
wastes, the equipment shotiiil be de,+a.onstrated to
be functional
at the scale of the proposed operation,
the
operation
should be simple, the system should have clear,
preferably
written,
instructions
for operation,
the equipment
and its capacity
should be suitable
for the quantities
and
-;
.
JC
to be handled and compatible
with other
t *rLa of material(s)
components, and users and//or operators
must be capable of
properly
maintaining
and operating
the methane generating
and
utilization
equipment.
Individuals
who have experience
with the methane generation are cautiously
optimistic
&out its prospects.
The
fundamentals
of the process are well known, there is il signilicant
quantity
of human, animal,
and agricultural
wastes
that may be available,
and these wastes can produce large
quantities
of methane gas.
The present technology
can be
utilized
and adapted to local conditons
where appropriate.
Such utilization
should occur only with competent guidance
and after a careful
evaluation
of the technical,
economic,
and social feasibility.
Acknowledgements
The material
that is presented
is a brief
summary of the
important
factors
affecting
,the use and operation
of methane
generation
facilities.
Many comprehensive
reports
and
articles
(l-6,
10-13) are available
on the subject,
some with
specific
emphasis on the use of methane generation
in
developing
countries
and rural areas.
Individuals
wishing
to pursue the subject
in greater
depth are encouraged to read
the noted articles
and reports
(1s well as others that are
available.
M&ham fmm @man, Animal, arzd AgricuZtumL
References
-___ _ .-
Wastes
149
and Notes ._
1.
Tietjen,
C.
"Frcm biodung
to biogas
- historical
review
of European
experience."
Energy,
Agriculture
and Waste
Management,
Jewell,
W. Ed., Ann Arbor Science
Publishers,
gn Arbor,
Michigan.
247-260,
1975.
2.
National
Academy of Sciences,
Human, Animal
and Agricultural
1977.
3.
Singh,
R.B.
Bio-Gas
Plant.
Generating
Methane
Organic
Wastes.
Ajitmal,
Etawah,
(U-P.),
India:
Gas Research
Station,
1971.
4.
Fry', L.J.
Practical
Building
for Rural Energy Independence.
Standard
Printing,
1974.
5.
Chung, PO.
"Production
and use of methane
wastes
in Taiwan."
Proceedings,
International
Energy Conference.
Biomass Energy Institute,
129, Winnipeg,
Manitoba,
Canada.
1973.
from
6.
Srinivasan,
H.R.
"Bio-gas
and manure from
farm animals,"
Khadi and Village
Industries
Bombay, India,
January
1977.
the waste of
Commission,
7.
Morris,
G-R., Jewell,
W-J.,
and Casler,
G.L.
"Alternative
animal
waste anaerobic
fermentation
designs
and
their
costs."
Energy,
Agriculture
and Waste Management,
Jewell,
W. Ed., Ann Arbor
Science
Publishers,
Ann Arbor,
Michigan.
317-336,
1975.
8.
Smith,
R.J.
"The
and the prospects
Dept.,
Iowa State
9.
Loehr,
R.C.
Pollution
Control
for
Academic
Press,
New York.
1977.
10.
Singh,
organic
11.
Patankar,
tries."
12.
New Alchemy
and Fertilizer.
Massachusetts,
R.B.
"The
wastes."
Methane;Generation
Wastes,
Washington,
from
D.C.
From
Gobar
of Methane Power Plants
Santa Barbara,California:
animal
Biomass
P-0. Box
anaerobic
digestion
of livestock
for methane production."
Agric.
Univ.,
Ames, Iowa.
1973.
bio-gac
Compost
G.L.
"Role
Khadigramodyog
of
Agriculture.
giant:
Generating
Science.
20-25,
gobar
(India),
wastes
Eng.
gas plants
20 (April)
methane
Jan-Feb,
in
from
1975.
agro-indus:351-35':;
1974.
Institute,
Methane Digesters
for Fuel,
Newsletter
No. 1 3. Woods Hale,
1973.
Gas,
150
13.
RaymondC. Loch
Compere, A.L. and W.L. Griffith.
for the degradation
of cellulose."
National
Laboratory,
Oak Ridge,
"Anaerobic
mechanisms
ORNL-5056,
Oak Ridge
Tennessee,
1975.
Summary and Discussion
Roger Revelle
We had a wonderfully
interesting
discussion
In different
ways nearly all of it was
today.
about solar energy.
Mr. L8f started
out this
abou t the direct
morning by talking
absorption
of
solar radiation
and its use for heating
and cooliuIr. Prinz discussed
the conversion
of the
ing.
radiant
energy coming from the sun into electricity
by the photovoltaic
method, and the further
utilization of the high temperatures
generated
in this
process as a source of useful
heat.
Ar. Tewari
talked about the uses of wind energy in India,
but
of course the energy of the wind results
from the
differential
heating
of the earth by the sun, so
it too is a form of solar energy.
Xr . Ermenc
talked
about small-scale
hydroelectric
power.
Here, the radiant
energy from the sun is converted
into gravitational
energy through the hydrologic
and the hydroelectric
plant in turn concycle,
verts this gravitational
energy into electrical
energy.
This afternoon
Mr. Sakr discussed
the use
of flat
plate collectors
and concentrators
for the
direct
conversion
of solar energy into heat in
Egypt, and Mr. Loehr talked
about the use of biological
materials
as a means of converting
solar
energy into humanly useful work.
From an energy
point of view, as he pointed
out, the non-edible
portions
of agricultural
crops are just as much a
potential
resource
as the part that can be eaten
by human beings,
because energy for cooking is an
integral
part of the human food system everywhere.
If we were forced to subsist
on raw food, there
simply would not be enough to eat for the large
152
Roger Rem zZe
numbers of human beings who now live on this planet,
not to mention all the other things we do with
energy that make the difference
between living
like
human beings and living
in an inhuman way.
Then
about another biological
device
Dr. Powell talked
for converting
solar energy --the use of wood in
Ghana and the potential
for using a good deal more.
And finally
Mr. Miccolis
talked about the very
impressive
new energy program in Brazil,
in which
sugar cane and other biological
converters
of
solar energy will
be used in an attempt to solve
their
severe energy crisis.
Brazil
apparently
has
very little
oil (at least on land) and not much
usable coal, and the four or five-fold
rise in the
price of imported
fossil
fuels is causing serious
econcmic problems.
These same problems will
be
faced by every country
in about 30 years, when the
oil reserves
start
to run out.
So the Brazilians
are simply pioneers
in a situation
that may become
universal within
our lifetimes.
If
we summarize what we have learned about all
these methods of using solar energy it is clear
that we are really
concerned with the best allocation of resources,
which is the central
theme of
economics.
c)ne characteristic
of resources
is
that they are scarce.
There aren't
enough of them
to go around, and consequently
there are always
competing uses and needs for resources.
Unless we
allocate
resources
as economically,
rationally
and
prudently
as possible
in the face of competing
needs, the processes
of development
in the poor
countries
will
become much more difficult.
The
economic objective
of development
can be quite
simply defined:
to increase
human productivity.
How can resources
be best allocated
to increase
human productivity,
which is the only way that
-real incomes can be raised?
Incidentally,
when we
think about appropriate
technologies
for the
developing
countries,
our criterion
should be
1 whether these technologies
are in fact likely
to
1 increase
productivity.
This can be accomplished
j in two ways, either
by producing
more per man-hour,
; or by increasing
the number of man-hours,
i.e. by
:Y-increasing
employment.
In choosing among energy sources and uses for
the rural
areas of developing
countries,
we are
concerned with scarce resources
and their
most
economical
allocation,
both the natural
resources
Swrunaryand Discussion
153
of land, water,
fertilizer,
metals,
and fuels,
and
the human resources
of capital,
labor,
human energy,
management and organization.
skills,
The scarcest
resource
of all in most developing
countries
is
trained
management ability--the
ability
to organize
and administer,
to create an organization
and to
make it work.
We must also ask what are the needs for new
sources of energy?
One need is to reduce human
drudgery,
as Professor
Loehr pointed
out this
afternoon.
Theodore Schultz,
the great agricultural
economist of the University
of Chicago, tells
a wonderfully
illustrative
story.
He talked to
some women in a village
in Senegal, who were
threshing
millet
by pounding a heavy pestle
into a
large wooden mortar.
They had to do this all day
to separate the millet
grains from the chaff.
It
was very hard, very dull work.
They said "We've
heard about a thresher
in the next village
that
does this mechanically.
If we had one we wouldnft
have to do this task by hand.
Why can't we have
one?"
We tend to think of people in less developed
countries
with high levels
of unemployment and
underemployment
as being eager to work in hard
drudging
tasks if they have the opportunity,
tasks
which will
require
them to use about half their
food energy in work.
We think of such drudgery as
being natural
for them, but we don't think of it as
being natural
for us. In fact,
there are probably
few human beings anywhere who are fond of continuous hard manual labor,
such as the labor involved
in traditional
farming.
A second need for new sources of energy is to
reduce costs.
The most expensive
form of energy
is human energy, as can easily
be seen if we consider how much food it takes to keep a man going at
hard manual labor.
"iron
law" says that
Ricardo's
wages always tend to go down to the subsistence
level,
i.e.
to the level where a working man is
able to buy with his wages barely
enough food to
feed himself
and his family.
But the corollary
to
Ricardo's
law is that wages can't go below the
.
level of subsistence,
and this level is not cheap.
It takes about 15 cents a day to feed a man even on
the least expensive
of diets,
and he is able to put
out somewhat less than a kilowatt
hour per day of
useful work.
Even at present
costs of fossil
fuels
a kilowatt
hour from gasoline
costs about 3 cents.
154
Roger ReveZZe
Animal energy is also expensive.
Consider for
example the opportunity
costs of feeding
a bullock.
A farmer can use part of his land to grow fodder
for his bullock,
but he could grow wheat or rice or
corn on that same land.
Finally,
new sources of energy are needed in
order to increase
the total
amount of energy availIn my talk this morning,
for example, I
able.
showed that by increasing
the quantity
of energy
used in agriculture,
the annual yields
per hectare
could be tripled
or quadrupled.
What are the specifications
or characteristics
to be kept in mind in evaluating
potential
new
energy sources?
To be most useful
in less developed countries,
any potential
energy source should
have a low capital
output ratio,
the income
obtained
each year should be at least half and
preferably
equal to the capital
investment.
The
capital
costs should be low for another reason:
capital
is scarce in developing
countries
and correspondingly
interest
rates are, or should be,very
high.
If the capital
costs for a new energy source
are a large part of the total
costs high interest
rates are likely
to make it uneconomical.
For the
same reason it should be possible
to amortize
the
investment
over a long lifetime,
i.e.
the rate of
depreciation
should be low.
Small unit costs are
desirable
in order to keep the investment
from
being "lumpy.'*
For this reason "micro"
or "mini"
hydropower plants
using *'run-of-the-river"
installations
are attractive.
Any energy source for developing
countries
at
the present time should be manpower-intensive
rather
than capital-intensive.
Its development
should use people rather
than capital
because
people are less expensive
and more abundant than
capital
in most less developed countries.
In other
words, energy investments,
like other investments,
should take account of the factors
of production.
But one scarce kind of manpower is skilled
manpower.
Hence new energy installations
should be as simple
as possible,
because there are so many competing
demands for skilled
manpower.
If possible
there
should not be other competing uses for a new energy
source.
For example wood can be used for industrial
purposes as well as for fuel,
and in mountainous
regions
it is essential
to leave the trees
Swnrnaryand Discussion
where they are to prevent erosion.
Straw, if it
not burned for energy, can be used for animal
feeding and for compost to fertilize
farm fields.
155
is
In the hills
of Nepal the farmers are reluctant to use the new dwarf varieties
of rice because
these new varieties
don't produce much straw.
They
need straw to feed their
livestock.
They need the
livestock
to produce dung to fertilize
their
fields.
During the monsoon season the cattle
and buffalo
gather nitrogen
from pasture
land, and part of this
nitrogen
is excreted
in their
dung.
The farmers
collect
the dung and carryit
to the fields
as
fertilizer.
During the dry season little
nutrient
material
remains in the pastures,
and the livestock must be fed on straw.
here is a closed
farming system which is very hard to break into or
to change.
From an energy standpoint
the farmers
would be unable to convert the energy in the straw
into some other form of energy because they must
feed it to their
livestock.
In many cases, in less developed countries,
local energy sources should be used rather
than
centralized
sources,
and local energy conversion
plants
should be constructed
rather
than central
energy conversion
plants.
For example, the load
factors
for electric
power are usually
quite low,
and consequently
the capital
costs of electric
transmission
systems from central
plants
per unit
of energy are very high.
Problems of erergy storage arise because local
energy sources are ilot available
all the time.
F
,nergy storage is es::ential
if the inedible
parts
of crops such as corn stalk
or rice straw are to be
used as an energy source.
When these inedible
portions are left
standing,in
the fields
after
the end
of the growing season, their
contained
energy
quickly
disappears
into the air.
They must be used
when they are freshly
cut.
Similarly
the contained
energy of sugar cane rapidly
disappears
after
it is
harvested.
If one attempts
to use run-of-the-river
hydropower
in Nepal or elsewhere
in the Indian subcontinent,
there will
be a great excess of river
flow in the monsoon season (from the middle of June
to the middle of September)
and very little
flow
during the other eight months of the year.
Some
way must be found to store the energy that is pres.ent in great abundance during the monsoon season
156
for
Roger ReveZZe
use during
the dry season.
Storage is one of the problems of methane production
from biological
materials.
14ethane is difficult
to store because it is so light.
A large
volume of storage is required,
or else an expensive
high-pressure
tank.
This argues for making liquid
fuels such as methanol or ethanol
from biological
materials,rather
than gaseous fuels.
Even the dire-et conversion
of solar energy by
photocells
or by heat absorption
has the problem
that sunshine varies a lot,
not only between day
and night,
but during the course of the year, particularly
in the monsoon climates
of the tropics.
2uring four months of the year the radiation
from
the sun is obscured by heavy clouds,
and considerably less radiation
is received
at the ground than
during the wintertime.
As Mr. Sakr pointed
out,
this is not true for Egypt or other desert areas,
where there are hardly
ever any clouds.
But the
.
populations
of deserts
arL" sparse, orders of magnitude
smaller
than those of India,
Pakistan,
dangladesh,
Southeast
Asia and Indonesia,
where the
monsoon dominates the climate.
One way to store energy is to make nitrogen
fertilizer.
Lven with the most efficient
modern
processes
for nitrogen
fixation,
about 15000 kilowatt hours are required
to produce a ton of fixed
nitrogen.
With the old electric
arc process,
perhaps more than 45000 kilowatt
hours are required
per ton of nitrogen.
With the Norwegian electrolysis process,
hydropower
can be used to produce
hydrcgen by electrolyzing
water.
The hydrogen is
then used as an electron
acceptor
for nitrogen
in
making ammonia.
Other possibilities
for energy storage are
other chemical products.
With present
fossil
fuel
costs it may actually
be less expensive
to make
many chemical products
out of ethanol
or methanol
than out of petroleum.
This may not be true at
present
in the United States,
because we have sunk
so much capital
in petro6hemical
plants,
but thirty
years ago a good deal of technological
research was
devoted to the development
of methodologies
for
using organic
sources to make what are now called
petrochemicals.
Swnmaryand Discussion
157
So-called
pump-back storage--pumping
water up
into a high reservoir
during off-peak
hours, and
using this water to augment power production
during
peak load periods--is
another way to store energy,
at least over short periods,
in order to minimize
required
generator
capacities.
Careful
studies
are needed of energy systems
in rural
villages.
I have mentioned the farmers in
the hills
of Nepal, whose energy system is tightly
locked into their
food system, with the result
that it is hard to break into either.
Village
energy systems are also likely
to interact
with
village
social
systems.
For example, village
level
biogas plants using human wastes might be constructed
to provide
energy for a large fraction
of
the village
population.
Such centralized
plants
would require that the village
people defecate
in a
central latrine
instead of the shady bamboo groves
they have been using for thousands of years.
For
several reasons --some of them obvious--village
people don't like to use latrines.
It may be necessary to find ways of paying them to change their
ancient habits so that the human wastes can be collected for use in a central village
biogas plant.
Small family bi‘ogas plants will provide enough
energy to be useful only for those farmers who
have five or more dung-producing
cattle.
In this
case, the rich farmers will be better off and the
landless laborers worse off than they are at
present.
The landless laborers and other dependent
people fn a village
now usually have a time-honored
agreement with the landowner to provide them with
straw and dung to cook with, in return for certain
traditional
services.
These agreements will be
broken if the landowner finds a profitable
use for
the dung. This is simply one example of the fact
that the entire energy system of a village
must be
carefully
examined in ways that have not received
much attention
from anthropologists--because
anthre
pologists
are usually uninterested
in economics-to find out what actually
can be done to improve
the quality
of village
life and part3cularly
the
quality
of life of the poorest people in the
village.
If we attempt to apply the criteria
I have
outlined,
it is clear that choices among different
potential
energy sources will depend upon differences in environmental
condition,
including
the
158
Roger ReveZZe
social
environment.
For example, wind energy may
be most useful
in coastal
regions,
where average
wind speeds are likely
to be higher than 5 meters
per second (more than 10 miles per hour).
In
desert areas where there is plenty
of sunshine but
little
or no water,
direct
conversion
of solar
using
photocells
or
other
types
of colenergy,
may prove to be economical,
provided
costs
lectors,
can be sufficiently
lowered,
at least in areas such
as Saudi Arabia,
the Persian Gulf states,
or the
southwestern
United States,
in which investment
capital
is easily
available.
Small-scale,
"run of
the river"
hydropower may be preferable
in hilly
or mountainous
regions with many perennial
streams,
such as Nepal and the Himalayan foothill
regions
of India and Pakistan.
Plantations
of quickgrowing trees or coarse grasses will
be most
promising
in regions with abundant water supply
and large areas of non-arable
land, including
parts of Africa
and South America.
In large,
sparsely
populated
countries
such as Brazil,
where
there is an excess of well-watered,
potentially
arable land, biological
conversion
of solar energy
by production
of methanol or ethanol
from sugar
cane or other high-yielding
crops may be an
optimal
energy source.
The most difficult
problems arise in densely
populated
regions with relatively
small areas such
as Japan, Bangladesh,
and the island
of Java.
Useful supplements
to existing
energy sources can
be obtained
in these regions
from the nol,-edible
portions
of agricultural
crops, but it may be
impossible
to satisfy
their
future
energy needs
without
extensive
reliance
on fossil
fuels,
and
in the long run, on nuclear
energy.
Questions
and Answers
Question:
This morning you discussed
the relationship between nondomestic
consumption
of energy and
productivity.
Later in your talk you talked
about
one way of using energy more efficiently,
viz.,
a
better
stove and a better
cooking pot.
So what
you are referring
to at this point is domestic use
of energy.
IJow when domestic uses of energy become
more efficient
what is going to keep people from
still
channeling
most energy into domestic rather
than nondomestic
uses? Let me give you an example.
In many of the less developed countries
people cook
Swnmaq and Discussion
159
their
meals once a day.
If they have more energy
available,
why not cook twice a day? Or why not
wash their
clothes
in h#Jt water?
How are you going
to keep from widening
the gap between domestic and
nondomestic
uses of energy?
In other words what
would keep people in the \less developed countries
from emulating
their
cousins in more developed
countries
by becoming more consumptuous?
I
I
Mr. Revelle:
1 would not object to that if the
energy were available.
After all,
the purpose of
life
is to live as much as you can.
But the problem is that these poor people don't have the
As I said today, one person in a family
energy.
may spend his (or usually
her) entire
time just
gathering
fuel.
This chore is getting
harder and
harder all the time as populations
grow.
So that
you have a tradeoff
here of resources:
the fuelgatherer
will
work less, gathering
less fuel if he
gets a better
stove.
Or he may decide it's
alright
to continue
to work hard and cook twice a
day.
He would at least have more choice than he
has right
now. He has no choice at all.
He is
just barely making it now.
Comment from the floor:
I think basically
the
problem hinges upon making the nondomestic
uses of
energy and food f or that matter more attractive
economically
than the domestic ones, I can't help
thinking
about the situation
in the north of Ghana
where it is commonly understood
that to increase
the food production
by 20% will
achieve nothing
because the people can eat 20% more food, and in
fact they are hungry for two months of the year.
They call it the hungry season.
They virtually
have nothing
to eat.
So I think we should all
realize
that we have this backlog to make up.
Yes,
people are going to use more energy domestically,
yes they are going to eat more food domestically,
but let's
give them tile chance to make that forward step first.
Then we must think about ways of
using their
time, their
energy and their
resources
for economic activities
which will
be to their
own
personal
benefit.
And then they will
choose for
themselves
how they use their
resources
and their
energy.
duestion:
strategies
resources
Can any of the panel members identify
for development
of energy and other
which are more likely
to encourage
260
Roger ReveZZe
demographic transition
or whatever you choose to
such that the proceeds won't tend to be
call it,
eaten up?
Mr. Hevelle:
I don't know that there is any
simple relationship.
But I guess I should have
amplified
my statement
about increasing
productivity by saying that it should be increased
more
rapidly
than populations
grow.
That of course is
in a sense the definition
of productivity
per man.
We need to remember that people have hands as well
as mouths.
They can produce as well as consume.
In general , -it seems to be true that if the
poorest people see a way in which their
lives
can
be improved and in which their
children's
lives
can be improved,
they are very likely
not to have
so many children.
That seems to be the experience
of many countries
where this kind of improvement
has taken place.
Question:
For example
on the Indian
subcontinent?
Mr. Revelie:
Oh yes.
For example in Sri Lanka,
where there is a fairly
good income distribution
compared to the rest of the subcontinent--everybody in Sri Lanka receives
a free kilogram
of rice
once a week and pays only 2 annas for another kg;
every village
has a dispensary,
every market town
every little
city has a hospital,
has a clinic,
all frlze for everyone in the country;
education
up
to and including
the university
is free,
and there
are about as many educated women as men. And Sri
Lanka has the lowest birthrate
in the subcontinent.
Taiwan, Korea, Singapore,
Hongkong, !ilalaysia,
particularly
West Malaysia,
Costa Rica, countries
where there is a relatively
good income distribution,
where living
conditions
of the poor have
been and are improving,
all have low birth
rates.
Improving
the living
conditions
of the poor may be
a necessary
if not a sufficient
condition
for
reducing
population
growth.
This improvement may
well depend on obtaining
and distributing
more
energy.
Index
absorption
refrigeration,
31
Africa,
13, 15, 18, 30, 158
algae, use in Brazil,
67
alternative
energy sources
in India,
78
(See nZso nonconventional
energy resources)
alternative
energy technologies in Brazil,
53, 56,
57, 67, 70, 72, 73
ammonia, 133, 135, 156
production
in Brazil,
69,
70
anaerobic
fermentation,
84
(See a&o bioconversion,
fermentation)
in Brazil,
67
animal wastes (dung), 11-13
applications
of photovoltaic devices,
45ff.
appropriate
technologies,
152
in Brazil,
65, 66
aquaculture,
59
ASTRA (Application
of
Science and Technology
in Rural Areas),
79
(See also India)
automobile,
use in Brazil,
61, 62
bacteria,
135
obligate
anaerobes,
methanogenic,
135
135
bagasse, 55
Bailie
process,
62
Bangladesh,
12-18, 20, 25,
158
Barbegal,
France (Roman
flour mill),
20, 25,
158
bioconversion,
58, 60,
63-65
(See also anaerobic
fermentation)
biodegradability,
131, 134,
145
biogas, 84, 140, 151
(See aho anaerobic
fermentation)
carbon dioxide
in, 140,
143
composition,
140
digester,
144
plants,
15, 138, 143, 147,
157
production,
132, 133, 145
storage,
141
use, 140, 142
biological
materials,
use
of, 151, 152, 156, 158
biosynthesis,
microbial,
63
boat/floating
mill
in China, 105
in Italy,
105
Bolivia,
12, 16, 18
Brazil,
53ff., 152, 158
bullock
work, in India,
77,
81
162
Index
Canada, hydroelectricity
generation
by smallscale turbines,
95
capital
costs, 154, 155
capital
output ratio,
154
capital-intensive
energy
sources, 154
carbon dioxide,
in biogas,
140, 143
carbon to nitrogen
ratio,
133
cassava (see maniac)
Cassino, Italy,
grain mill,
89, 105
catalytic
burners,
use in
Brazil,
69
centralized
energy sources,
155
charcoal
in Ghana, 119, 120
activated,
124, 125
kilns,
123, 124
palm-kernel,
121
production
and use in
Brazil,
55, 65, 66
chemicai feedstocks,
sources
hydrocarbon-producing
plants,
66
sugar cane, 63
China, People's Republic of,
12, 15, 16
small water turbines
to
generate electricity
in, 95
water wheel, use in, 92,
101, 105, 107
coal, in Brazil
composition,
66, 70
gasification,
70
reserves,
54
use, 55, 66, 70
coastal regions,
energy
resources in, 158
concentrators,
photovoltaic
systems, 48
conservation,
energy, in
Brazil,
53, 71, 72
cost, photovoltaic
devices,
45, 46, 47, 48
cow dung, methane from, 130
Cretan sail windmill,
use
in India,
80
crop drying,
35
crop residues,
11-13, 20
dams, 112
log dams , 112
ea?!e type, 112
rafrilr
prop type, 112
Darrieus rotcW, use in
India,
79
depreciation,
154
desert areas, energy
resources,
,158
digestion
(See aZs0 fermentation,
anaerobic)
aerobic,
134
anaerobic,
130
Domesday Book, 91
draft animals,
in India,
77
efficiency
of fuel use, 18, 21
of thin-film
solar cells,
48
electricity
consumption in Brazil,
55
production
in Brazil,
55,
68
transmission
costs, 155
in villages,
in India,
83,
85
use in Brazil,
electrolysis,
68, 69
energy
in agriculture,
14, 15,
18, 19, 20
animal, 11-13, 15, 155
capital-intensive
sources,
154
centralized
sources,155
commercial sources, 11-13,
15, 17
domestic uses, 14-16, 18,
24
human, 11-13, 19, 153
Index
new sources of, 20, 22
.
(See also alternative
energy resources)
need for, 153, 154
noncommercial
sources,
11-13, 15, 17
nondomestic uses, 15, 16
nuclear,
158
solar,
151, 152, 156, 158
traditional
sources of,
11, 12
energy in Brazil
conservation,
53, 71, 72
consumption,
55
conversion
technology,
69
nuclear,
56
storage,
58
uses, 56
energy storage,
21, 58, 155,
A56
environmental
protection,
129, 146
erosion,
130
soil improvement,
139
Eotechnic Phase, 89
ethanol
(ethyl alcohol),
156, 158
production
and use in
Brazil,
61, 62, 64, 65
with gasoline,
61, 62, 65,
72
factors
of production,
154
family biogas plants,
157
fermentation
ammonia concentration
in,
133, 135
anaerobic,
84, 130ff.
bacteria,
135
fertilizer,
129
lime, use of, 135
microbial
reactions,
134
pH, 135
stabilization
of organic
matter,
134
stabilized
residue,
129
fermentation
processes,
in Brazil,
63, 64
fertilizer,
129, 130, 139,
145
163
agricultural
yields,
130,
146
production
and use in
Brazil,
67, 69, 70
firewood
(fu?l wood), 11-13,
15, 18, 22, 24
use in Brazil,
55
use in Ghana, 119, 120
in industry,
121
Fischer-Tropsch
process,
63
food energy, 13
fossil
fuels,
11, 12, 20,
22, 152, 158
Fourneyron,
Benoit,
92, 93,
107, 109
France,
development of water turbine, 92, 93
Roman flour mill in, 89,
104
use of waterwheels,
89,
104, 107
Francis
(reaction,
fulladmission)
turbine,
109,
111
gas I natural,
54, 55
reserves,
in Brazil,
54
gas, pipeline,
in Brazil,
70
in Brazil,
gas s synthesis,
62, 63
gaseous fuels,
20, 156
gasoline,
use in Brazil
cost, 71
mixed with ethyl alcohol,
61, 62, 65, 74
goals, United States photovoltaic
program, 46
greenhouse,
combined with
dwelling,
31, 35
heatin,; and cooling,
151
human productivity,
152,
158
Hungary
hydroelectricity
generation by small-scale
water turbines,
95
hydraulic
power, 111
164
Index
site evaluation,
111, 112
hydrocarbons,
production
from plants,
66
hydroelectric
power, 11, 22,
23, 24, 151, 155, 156,
158
in Brazil,
54, 55, 68
Irr Ghana, 119
from small-scale
turbines
in Canada, 95
in China, 95
cost, 95
in Hungary, 95
in Russia, 95
in United States,
95
in West Germany, 95
hydrogen
in manufacture
of ammonia,
156
production
and use in
Brazil,
67-69
hydrogen sulfide
(in bio140,
143
gad y
income distribution,
160
India,
12-24, 37, 77, 79,
80, 81, 83, 85, 158
methane generation
in,
130, 141
(See akco ASTRA)
ltaly,
use of waterwheels
in, 89, 105
internal-combustion
engine,
use with ethyl alcohol,
62
irrigation,
81, 82, 85
Japan, 158
Java, 158
Kaplan (variable-pitch,
axial-flow)
turbine,
109, 111
Kellog process, use in
Brazil,
62
Koppers-Totzek
process,
use
in Brazil,
62, 70
Korea, methane generation
in, 131
Latin America, 13, 22
liquid
fuels,
20, 156
load factor,
155
local energy sources, 155
Lurgi process, use in
Brazil,
70
maniac, use in Brazil
conversion
to fuel,
57,
62, 64
production,
65
manpower-intensive
energy
sources, 154
market, for photovoltaic
systems, 45, 46, 49
Merrimac River
products from mills,
93
water turbine
installations,
93
methane, 140
in Brazil,
67
methane generation,
129,
131
(See also fermentation,
anaerobic)
cow dung (manure), use
for, 130
economics, 146, 147
equipment,
143
gas composition,
140
production,
131, 132,
134, 137, 143, 145,
156
storage,
143
utilization,
l3?, 140
in India,
130, 141
in Korea, 131
operation,
145
in People's
Republic of
China, 131
in Taiwan, 130, 141
in Uganda, 131
methanol,
156, 158
in Brazil,
63, 65
Mexico, 12, 13, 15, 16
microbiological
conversion
techniques.
use in
Brazil,
63
microbial
reactions,
134,
136
Index
time
and temperature
relationships,
136138
Mobil process, use in
Brazil,
65
muscle -power, in India,
77
natural
resources,
Brazil,
54
Nepal, 21-23, 155, 157, 158
Nigeria,
12, 15, 16, 18
niobium,
in Brazil,
54
nitrogen,
131, 133, 135, 139
nitrogen
fertilizer,
156
Nagler (fixed-propellor)
turbine,
109, 111
use for large-scale
hydroelectric
installations,
111
nonconventional
energy
resources,
in India,
83 ¶
134
(See also alternative
energy sources)
noria (Roman undershot
waterwheel),
105, 107
nuclear energy, in Brazil,
56
light
water reactors,
56
nuclear power, 11
nutrients
in sludge, 139
in wastes, 143
ocean thermal energy conversion (OTEC), 58
oil,
in Brazil,
53
(See a&o petroleum)
consumption of, in India,
82
shale, in Brazil,
54, 70
Otto process, use in Brazil,
70
overshot waterwheel,
101, 109
efficiency,
101
limitations,
101
uses, 104
Pakistan,
22, 25, 158
165
pathogens,
human, 129
plant,
129
Pelton (impulse)
turbine/
wheel, 109, 111
development in California,
109
People's Republic of China,
131
Persian Gulf, 158
Petrobras,
55, 70
petroleum,
in Brazil,
64,
65
imports,
53, 64, 65, 69
production,
61
reserves,
54
use, 55, 71
photovoltaics,
45, 151, 156,
158
in Brazil,
58
photovoltaic
systems,
concentrators
for, 48
market for, 45, 46, 49
pig iron, in Brazil,
65, 66
public health,
129, 146
pump-back storage,
157
Purox process, use in
Brazil,
63
pyrolysis,
in Brazil,
67
quality
of village
life,
157
quick-growing
trees,
158
Rankine engine, use in
Brazil,
58
refrigeration,
35
in United States,
35
in USSR, 35
reprocessing
(nuclear
fuels),
Brazil,
56
resources,
allocation
of, 15
human, 15
natural,
152, 153
resource conservation,
129,
147
rubber (Hevea),
production
in Brazil,
66
166
Index
rural
electrification,
in
India,
82, 83, 87
rural
(village)
energy
systems, 15
Ruqqia
LL_
hydroelectricity
generation by small-scale
water turbines,
95
Saudi Arabia,
15
Saugus, Massachusetts
(iron
miii),
9i
Savonius rotor,
use in
India,
79
sewage, conversion
to
fertilizer,
Brazil,
67
shaftwork,
80, 81, 83-85
silicon,
large-sheet
technology for photovoltaics,
47, 48
sludge, 129
fertilizer
value, 129,
130, 139
municipal
sewage, 130
utilization,
139, 145
SNG (substitute
natural
in
Brazil,
63
gas) 9
social equity,
24
solar concentrators
(concentrating
collectors),
USSR, 41, 43
solar cooking,
37
in India,
37
in Mexico, 41
in United States,
41
in USSR, 37, 41
solar cooling
(cost and
efficiency),
28
solar energy, use of, in
Brazil,
56, 47, 59
architecture,
58
collectors,
57
crop dryers,
57, 59
distillation,
57
irrigation,
59
refrigeration,
58, 59
still,
58, 59
water heaters,
60
solar neat, conversion
to
electricity
(cost and
efficiency),
28
heating
(cost and
efficiency),
27
solarimetry,
in Brazil,
58
solar power
electrical,
41, 43
mechanical,
41
pumping, 41
steam power plant,
41
solar pumps, 84
solar radiation,
direct
absorption,
151
solar stills,
35
in Australia,
37
in Chile,
35
in Greece, 37
in United States,
37
in USSR, 37
solar water heaters,
manufacture
of, 29
in Australia,
29
in Japan, 29
solar water heating,
(cost
and efficiency),
27, 28
solids retention
time (SRT),
136, 137
South America, 158
Southern Cross windmill,
in
India,
78
soybeans, in Brazil,
65
space cooling,
technology,
31
spade heating,
27, 31
in France, 31
in United States,
31
Sri Lanka, 160
steel,
production
in Brazil,
66, 70
straw, use of, 15
substitute
natural
gas
(SNG), Brazil,
61
sugar cane,
conversion
to fuel,
61-64,
152, 158
production,
65
syngas (sythesis
gas)
process,
use in
Brazil,
63
solar
Taiwan,
130, 141
Index
Tanzania,
12, 15, 16, 18
therm, 27
xhermal conversion,
in
Brazil,
62
thJn-film
technology
(photovoltaics),
47, 48
thorium,
in Brazil,
54, 56
toxicity,
135
of metals,
135
transportation,
public,
in
Brazil,
72
tropical
forest,
Ghana, 115
Uganda, 131
undershot waterwheel,
vertical,
95, 104, 108
efficiency,
104, 107
in Europe, 101
uses, 105
undershot waterwheel,
horizontal,
101, 107, 109
efficiency,
104
in People's Republic of
China, 101, 109
uses, 107
unit costs, 15
United States,
15, 18, 20,
158
log dams, 112
use of turbines
in, 93,
109, 111
use of undershot wheel
for fishing,
107
use of undershot wheel
for tidal
power, 107
use of waterwheels,
106,
107
uranium, in Brazil,
54
enrichment,
56
resources,
56
village
volatile
biogas plants,
15
acids, 135, 139
wastes,
collection,
141, 143
crop residues,
129, 143
human, animal, agricultural,
129, 131, 135,
141
167
municipal,
136
organic,
in Brazil
composition,
67
recycling,
66, 67
preparation,
141
slaughterhouse,
133
storage,
141, 143
water hyacinth,
use in
Brazil,
67
water power, resources in
Brazil,
54, 67
water turbine
classes of, 109
comparison with waterwheel, 92
development in the United
States,
93
decline
in use, 95
Francis,
109
French development,
92,
107
Kaplan, 109
Nagler, 109
Pelton,
109
in People's
Republic of
China, 95
use, 93
for hydroelectricity,
95
waterwheel
comparison with water
turbine,
89, 92
efficiency,
101, 104
escalation
of use, 91, 92
first
prime mover, 89
in India,
84
limitations,
89, 101
types, 95, 104
uses, 91, 104, 105
West Germany, hydroelectricity generation
by
small-scale
water
turbines,
95
wind energy, 151, 158
wind energy conversion
in Brazil,
59
in India,
85, 86
windmills
in India,
78
WP-2, 78
168
Imlex
wind velocity,
records in
India,
78, 85
Winkler process,
use in
Brazil,
70
wood
distillation
of, in
Brazil,
66
use of, 154, 155
wood waste, Ghana, 117, 120-‘
lT2, 125
off-cuts,
117
pyrolysis,
124
sawdust, 117, 122, 124
shavings,
117
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement