REUSE OF
COCONUT RESIDUES AS
ACTIVATED CARBON
FOR
DOMESTIC WATER
PURIFICATION
IN
DEVELOPING REGIONS
by
Paul Indeglia
A Project
Presented to
The Faculty of
In Partial Fulfillment
of the Requirements for the Degree
Master of Science
In Environmental Systems:
International Development Technology
Fall 1998
REUSE OF
COCONUT RESIDUES AS
ACTIVATED CARBON FOR
DOMESTIC WATER PURIFICATION
IN DEVELOPING REGIONS
by
Paul Indeglia
Approved by:
_____________________________________________________________
Robert Gearheart, Major Professor Date
_____________________________________________________________
Peter Lehman, Committee Member Date
_____________________________________________________________
Daniel Ihara, Committee Member Date
_____________________________________________________________
Charles M. Biles, Graduate Coordinator Date
_____________________________________________________________
Dean for Research and Graduate Studies Date
(i) Abstract
This work examines the feasibility of incorporating new technologies and strategies for thermally converting coconut residues into useful products. Drawing upon previous documentation, extensive literature reviews, laboratory experimentation and personal experiences, this work analyzes implementing innovative projects involving uses of coconut residues. Laboratory experimentation includes producing activated carbon from coconut shells using technologies similar to those found in developing regions, and examining capacity to remove coliform through filtration. By converting waste from discarded coconuts into activated carbon, water quality and sanitation can be improved, leading to an enhanced physical environment and better public health.
Acknowledgments
I wish to express my appreciation to Dr. Robert Gearheart for allowing me the opportunity to study at Humboldt State University, Dr. Daniel Ihara for his direction and advice, both academically and personally, Dr. Peter Lehman, Dr. Margaret Lang, Dr. Thomas Borgers and Mr. Steve Salzman for their technical guidance, and Dr. Howard Stauffer for enhancing my statistical capabilities. I thank my colleagues, Mr. Jeff Detrick, Mr. Erik Nagy and Mr. Michael Harig for their assistance and direction in areas in which I was deficient. Special appreciation is given to Mr. Tapley Jordan and Mr. Michael McIntyre, who have inspired this project. For allowing me space to conduct my experiments on secure grounds, I am grateful to the City of Arcata Department of Environmental Services.
I would like to offer special thanks and love to my parents for their continued support of their son as he follows his path to internal peace and spiritual fulfillment.
This work is dedicated to all life that suffers from the deficiencies of human society, to those eternally searching for truth, and to Jim Richards.
Table of Contents
Title Page................................................................................................................... i
Signature Page........................................................................................................... ii
Abstract.................................................................................................................... iii
Acknowledgments..................................................................................................... iv
Table of Contents...................................................................................................... v
List of Figures.......................................................................................................... viii
List of Tables............................................................................................................ ix
I.
Introduction................................................................................................... 1
A. Overview................................................................................................. 1
B. Information Gathering and Methods.......................................................... 2
(2)
Project
Background....................................................................................... 3
III. Identification of Problems............................................................................... 5
A. Integrated Design..................................................................................... 5
B. Solid Waste............................................................................................. 6
C. Water Quantity and Quality...................................................................... 7
D. The Coconut........................................................................................... 9
IV. Waste Analysis –
A. Waste Study.......................................................................................... 10
B. Coconut Consumption and Waste Disposal............................................ 14
C. Problems............................................................................................... 15
D. Coconut Survey..................................................................................... 16
E.
Coconut Quantities Available in
V. Adsorptive Capacity of Coconut Waste-Derived Carbon............................. 18
A. Adsorptive Uses of Carbon................................................................... 18
B. Parent Material...................................................................................... 19
C. Production of Carbon for Adsorption..................................................... 20
D. Adsorption Processes............................................................................ 21
1. Particle Attachment.................................................................... 21
2. Chemical Reactivity................................................................... 22
3. Species Removal from Solution.................................................. 23
VI. Experiment to Determine Adsorptive Capacity of Coconut Waste-Derived Carbon 27
A. Experiment Objectives........................................................................... 27
B. Methods................................................................................................ 28
1. Indicators.................................................................................. 28
2. Filter Design.............................................................................. 29
C. Sample Preparation............................................................................... 31
1. Coconut Waste-Derived Carbon (CWD-C).............................. 31
2. Commercially Activated Carbon (CAC)..................................... 33
3. Sand......................................................................................... 34
D. Experimentation..................................................................................... 35
E. Data Collection...................................................................................... 37
F. Data Analysis........................................................................................ 38
G. Results.................................................................................................. 39
1. Raw Water Quality.................................................................... 39
2. Filtration.................................................................................... 39
a. Dissolved Oxygen.......................................................... 40
b. Turbidity........................................................................ 41
c. Conductivity.................................................................. 42
d. pH ............................................................................... 44
e. Flow Rate..................................................................... 45
f. Coliform........................................................................ 46
H. Confounding Elements........................................................................... 49
I. Conclusion of Testing............................................................................. 51
VII. Alternative: Energy Production..................................................................... 52
VIII. Conclusions................................................................................................. 54
References ....................................................................................................... 55
Appendix A: Analysis of Coconut............................................................................ 59
Appendix B: Production of Activated Carbon........................................................... 66
List of Figures
3-1:
Map of
4-1:
4-2: Decomposition Study for Biodegradable, Non-coconut Waste vs.
Coconut Waste at
6-1: Filter Types Used in Testing.............................................................................. 30
6-2: Sand Sieve Analysis......................................................................................... 35
6-3: Comparison of Dissolved Oxygen..................................................................... 40
6-4: Comparison of Turbidity................................................................................... 41
6-5: Comparison of Conductivity............................................................................. 43
6-6: Comparison of pH............................................................................................ 44
6-7: Comparison of Flow Rate................................................................................. 45
6-8: Comparison of Fecal Coliform......................................................................... 47
6-9: Comparison of Total Coliform......................................................................... 48
A-1: Uses of Coconuts............................................................................................ 61
List of Tables
4-1: Summary of Waste Characteristics.................................................................... 11
5-1: Biological Loading for Activated Carbon Filters................................................ 24
5-2: Total Coliform Removal for Activated Carbon.................................................. 24
5-3: Removal Efficiencies for Activated Carbon Filters............................................. 25
6-1: Coliform Removal Efficiencies........................................................................... 46
6-2: Homogeneity of Variance Test.......................................................................... 47
6-3: Summary of Statistical Analysis for Fecal Coliform............................................ 49
6-4: Summary of Statistical Analysis for Total Coliform............................................. 49
A-1: Chemical Composition of Coconut................................................................... 62
B-1: Chemical Reactions for Gasification.................................................................. 70
B-2: Gaseous Fraction from the Gasification Process................................................ 71
B-3: Gas Composition for Pyrolic Emissions............................................................. 72
B-4: Product Yield for Pine Bark and Sawdust Residuals......................................... 73
I. Introduction
A. Overview
This proposal began as a design project for a solid waste
management course in the spring of 1996. Having spent time on the Kenyan coast
as a United States Peace Corps Volunteer, I remembered the numerous vendors of
coconut water that discarded their waste, leftover shells and husks, on the
ground. Further, I recalled the dump for
My primary intention for returning to graduate school was to pursue research and study in the field of water quality engineering. I did not need to search long to discover coconut shells are often a parent material for activated carbon, a substance used in water purification. Many residents of developing nations do not have access to safe drinking water and must take precautions to protect themselves against disease. I decided to research processes employed in converting coconut shells to activated carbon for use in domestic water filters, enhancing the quality of water and improving public health. The result of this research is presented within.
Research for this project has expanded to include studies in
thermal conversion processes, domestic production of activated carbon from
coconuts, and filtration testing. This work is the culminating document for
this academic experience, yet additional study is still required and,
hopefully, will one day be implemented in
B. Information Gathering and Methodology
Information contained in this work has been derived through a
variety of sources and methods. The primary source for data was the
II.
Project
Background
The world’s population is approximately 5.9 billion persons (U.S. Census Bureau, 1998) with about 2.4 billion people living in coastal countries within the tropics, as calculated by the author using figures from the Central Intelligence Agency (1997). It is estimated by Lester R. Brown of the Worldwatch Institute that at least 60% of the world’s population live within 50 kilometers (30 miles) of the ocean (Brown, 1998). Applying this figure to the population in coastal countries in the tropics, there are, conservatively estimated, 1.4 billion people living in coastal zones in the tropics. As coconuts grow only in these coastal regions (Blaak, 1993; Copeland, 1931), 23.7% of human population has immediate access to coconuts, making them one of the most important resource in the world (Duke, 1989).
Regions in which coconuts flourish are also home to some of the most disadvantaged populations (Okun, 1991). Poor living conditions, inability to access proper water and sanitation, and poverty are very common. Many lack education and employment opportunities due to political and economic circumstances. Furthermore traditional societies have disintegrated primarily due to contact with western civilization.
Public health in developing regions, and specifically
in urban centers, is abject. Although there has been an increase in the number
of people with access to clean water and effective sanitation over the past
three decades, the number of those without these privileges has increased even
greater. This is a result of mass population influx from rural to urban areas,
coupled with the tremendous population growth that has accompanied much of the
last 30 years (Daniels, 1990; Okun, 1991). The recent evolution of
drug-resistant malaria, inaccessibility to effective health care facilities,
and the introduction of industrial by-products to the environment has caused
morbidity to raise (Okun, 1991). Those who escape fatal exposure suffer for
much of their lives. Life expectancy at birth for coastal African nations in
the tropics is 51.63 years, compared to 77.56 years for seven major
industrialized nations –
Like many industrialized nations, developing nations
have done little to curb practices that cause ecological disturbances. Unlike
industrialized nations, developing nations have little capital resource to
expend on projects that promote long-term ecological stability. In fact,
developing nations spend nearly all development funds on industrialization
(Gore, 1992). Most efforts toward ecological accountability have been through
nonprofit organizations from the industrialized West and
III. Identification of Problems
A.
Integrated Design
The reason this project has been
written is to help improve the quality of life for people living in coastal
zones within the tropics where coconut shells are currently being
underutilized. This objective will
be accomplished by reducing the amount of coconut waste being discarded to
local dumps through the production of activated carbon for water treatment. At
the same time public health will be improved and mechanisms for environmental
protection and remediation will be provided. The approach presented in the
following sections promotes complete resource recycling to achieve a higher
level of local sustainability and self-reliance.
Figure 3-1: Map of
B. Solid Waste
In rural parts of developing regions, where incomes are low and animal populations are substantial, most refuse consists of food and other organic wastes, which are left to be consumed by chicken, goats and cattle. However, in urban areas, where animals and land are scarce, and incomes and proximity to goods allow for greater consumption, these practices are not possible. Large amounts of material are thus discarded, creating a need for interactive waste management. Landfills require proper engineering, from siting and design to operation and maintenance, to be effective and safe. Unfortunately, concerted efforts to control waste rarely occur in developing nations.
For example, the present situation
of solid waste disposal in
The severity of the situation is obvious from the anaerobic
conditions that exist and the absence of any waterfowl in the adjacent section
of the waterway that surrounds
The smell from the dump is putrid, noticeable from as far away as three miles. It is possible that air carrying odor can carry disease vectors that cause harm to humans and animals. Although no figures were uncovered to verify statistically the health threat, members of nearby communities claim they suffer from chronic intestinal ailments. Members of smaller communities farther downwind of the dump support this claim. The author recorded all statements.
C. Water Quantity and Quality
As a result of shifting political and economic objectives
after World War II, international development agencies and missionaries
implemented vast modern health care programs (Hancock, 1989). Immunization
campaigns proliferated throughout the developing world, and reduced infant
mortality substantially. In
The need for increased agricultural production to feed the
rising population, coupled with inappropriate farming practices and
overgrazing, has caused a tremendous amount of erosion in
Precipitation is a factor of the amount of vegetation in a region and, since vegetation has decreased, there persists a downward cycle of less precipitation, which diminishes opportunity for natural regrowth. According to Finkel (1986), there is a great need to employ soil conservation methods for the preservation of water.
Water levels throughout much of the developing world are
dropping to near crisis levels. Since there is less availability of water,
people are more willing to compromise quality for quantity. For example, in
drier regions, people will take water from unsafe sources (i.e., puddles where
livestock have defecated or polluted wells) to placate demand. Without
sufficient energy to boil water for disinfection, pathogens are easily passes
from animals to humans, resulting in diarrhea, which in turns leads to
dehydration. Dehydration is the second leading cause of infant death in
Additionally, industrialization in this region has led to
contamination of local water supplies. Recent oil and chemical spills in
D. The Coconut
The coconut can play a role in helping to alleviate both
solid waste and water quality problems in tropical developing countries. First,
according to studies conducted by the author in 1992, coconut waste comprises
7% of material dumped in
IV.
Waste Analysis –
A. Waste Study
According to Jeanes (1997), “in order to gain a proper
understanding of the size and characteristics of the local waste stream, a
waste composition and generation study must be performed.” Using professionally
accepted techniques and information, remote sensing, and the personal
experiences of the project designer, a preliminary waste study has been
formulated for
An examination of the
Sources of waste are not dissimilar to those in the
Table 4-1: Summary of Waste
Characterization (conducted November 1992)
|
Type of Waste |
Volume (% total) |
Total Volume (m3) |
|||
|
Food
Waste |
|
|
69 |
|
5,202,821 |
|
|
Meat |
2 |
|
150,806 |
|
|
|
Vegetables |
28 |
|
2,111,290
|
|
|
|
Agricultural Waste |
32 |
|
2,412,902
|
|
|
|
Coconut Waste |
7 |
|
527,822 |
|
|
Paper |
|
|
7 |
|
527,822 |
|
Cardboard |
|
|
2 |
|
150,806 |
|
Wood |
|
|
1 |
|
150,806 |
|
Plastics |
|
|
6 |
|
75,403 |
|
Textiles |
|
|
2 |
|
452,419 |
|
Glass |
|
|
1 |
|
75,403 |
|
Tin Cans |
|
|
2 |
|
150,806 |
|
Appliances |
|
|
4 |
|
301,613 |
|
Batteries |
|
|
2 |
|
150,806 |
|
Tires |
|
|
4 |
|
301,613 |
|
Total |
|
|
100 |
|
7,540,320 |
The only recycling conducted in
A mathematical model for the volumetric growth of the
VT(t) = [ VI + VB + VC] * ert – [ VBed(b)t + VBed(c)t ], (Equation 4-1)
where, VT(t) = Total volume of waste at time t,
VI = Volume of inorganic waste at t = 1,
VB = Volume of biodegradable, non-coconut waste at t = 1,
VC = Volume of coconut waste at t = 1,
r = Rate of waste disposal growth,
t = Time (years),
d(b) = Rate of decomposition for biodegradable, non-coconut waste,
d(c) = Rate of decomposition for coconut wastes (annual).
Volumes of inorganic, biodegradable, non-coconut waste and
coconuts are based on the preliminary waste study for the landfill and are
301,613, 1,102,345 and 104,106 cubic meters respectively. The growth rate of
waste disposal, assuming no variation in the consumptive pattern, is equal to
the population growth rate for
With an estimated 7.5 million cubic meters of garbage added annually, decomposition of organic material becomes a very important factor in the overall volume of the dump. Figure 4-1 shows the projection of the waste volumes. Although there is 10 times as much biodegradable, non-coconut waste as coconut waste deposited each year, the rate of decomposition for coconut waste is such that after only 7 years, there is more volume occupied by coconut waste than by biodegradable, non-coconut waste (Figure 4-2). After 50 years, there are 2 ½ times as much coconut waste as biodegradable, non-coconut waste.
Figure 4-1:
Figure
4-2: Decomposition Study for Bio-Degradable, Non-Coconut Waste vs. Coconut
Waste at
B. Coconut
Consumption and Waste Disposal
As temperatures rise throughout the day in
The entire quantity of the agricultural waste is loaded onto a flatbed truck, or lorry, transported over the causeway, and disposed with the rest of the city's waste at the dump. Discarded coconuts cause unique solid waste problems, as they are a slowly degrading cellulose body that is hollow in the center. This results in a large amount of empty space due to the large quantity of coconut husks discarded, the lack of degradability and, subsequently, the rapid spread of the footprint of the dump. It is estimated that the volume of 43 Olympic-size swimming pools, over 100,000 cubic meters, are added to the Mombasa dump each year due to discarded coconuts. The coconut is very slow to break down, which introduces a problem for a system that has been self-monitoring for decades.
Figure 4-2 depicts the composition of the
C. Problems
During the rainy seasons, precipitation can accumulate in the hollow center of the shell, acting as a suitable and nutritious breeding habitat for malaria-carrying mosquitoes. This poses a great public health risk to the residents and travelers, as malaria is the cause of more deaths than any other disease in the coastal region. Malaria is the number one cause of death for infants in tropical regions worldwide (WHO, 1990; UNICEF, 1991) and any opportunity to reduce mosquito breeding habitats should be seriously examined. Other public health effects have been discussed in the previous section.
The amount of energy and person-power consumed in the civic activity of waste collection is tremendous. Over 20 individuals were observed in a medium sized market cleaning up agricultural wastes. There are 22 major market places on the island, and it assumed that many more exist in the mainland portion of the city. Extrapolated, there are at least 440 market sweepers employed by the city. Additionally, vehicles and petrol must be purchased to transport the waste to the dump. Reducing the need to discard coconut waste will relieve the municipality of a major resource drain.
D. Coconut Survey
The coconut is the largest indigenous crop in the coastal
strip of
The basic purpose of the survey for this project is to
estimate the amount of coconuts available for use in the 5,000 square
kilometers of
E. Coconut Quantities Available
in
Based on personal experience, it is estimated that there are an average of 10 vendors in each of the 20 major market places. Additionally it is estimated that there are 30 mobile carts vending coconut water on the streets, bring the total to 230 vendors. Examining various juice stands, restaurants and hotels which may utilize coconut water, a figure of 20 additional major sources of coconut waste is reached, for a total of 250 potential coconut-producing locations.
Assuming that vendors sell and discard 15 coconuts per hour
during the hottest hours of the day, from 11 a.m. to 4 p.m., plus another 20
for all non-peak hours brings the total coconuts available to 80 per day per
vendor. For 230 vendors, the total for the vendors is 18,400. The production of
waste by the other sources, juice stand, restaurants and hotels, is estimated
to be 100 per day, or a total of 2,100 coconuts. The amount of coconuts
available for reprocessing in the city of
V.
Adsorptive
Capacity of Coconut Waste-Derived Carbon
There are two media from which carbon is able to adsorb: from gas-phase and from solution. This report examines adsorption from solution.
A. Adsorptive Uses of Carbon
Carbon has been used as an adsorbant for impurities in water
since the
As technology improved throughout the 20th Century, the ability to study and activate, or prepare carbon for highly specialized use, increased. There are many modern applications for activated carbon including tertiary treatment for municipal wastewater (Gearheart et al, 1996), industrial wastewater treatment, removing gold from rivers, heavy metals recovery, and environmental remediation of hydrocarbon contamination (Cheremisinoff et al, 1978; Huang, 1978). Carbon is regarded as the foremost removal substance for odors caused by phenols, surfactants, pesticides and herbicides (Tchobanoglous, 1987). Additionally, activated carbon is highly effective in removing radon-222, organic mercury, and trihalomethane precursors. Unfortunately, production of activated carbon is comparatively expensive, primarily due to the cost of equipment used to activate it, and its use remains limited (Gearheart et al, 1996).
B. Parent Material
As an elemental substance, carbon can be derived from any material that has volatile carbon compounds. Because carbon for adsorption is most commonly produced through thermal distillation, which effectively destroys bonds without volatizing the carbon, naturally dense materials are preferable.
Some of the first materials used for carbon production were
coal, peat and nut shells. During the research of the processes of carbon
production, it was discovered that char remains of wood from the Pacific Lumber,
Scotia (CA) Cogeneration Plant were also use as a source of carbon at the
Calgon Carbon Regeneration Plant in
Widely accepted as one of the best sources of adsorptive
carbon is coconut shell. Coconut shell is a very dense, highly cellulosic
material found in abundance throughout the tropical regions of the world. In
fact, most of the experimentation on carbon adsorption examined for this report
has used at least one sample derived from coconut shells (Arulananatham et al,
1989; Cheremisinoff et al, 1978; Cookson, 1978; Gergova et al, 1993; Huang,
1978; Laine et al, 1989; Mattson et al, 1971; Namasinayan et al, 1994, Rao et
al, 1992). One reasons that coconut shells have not assumed a greater role in
carbon production is that production facilities are primarily in industrialized
countries in Europe and
C. Production of Carbon for Adsorption
Carbon is most commonly formed through a three-step process of dehydration, carbonization and activation (Cheremisinoff et al, 1978; Cookson, 1978). Dehydration is the removal of moisture from the material. By causing excess water to vaporize, temperatures in latter steps of conversion are more easily controlled. Dehydration can either be done mechanically, in a controlled heating environment, or naturally, using favorable climatic conditions. Carbonization is the reduction of organic matter to elemental carbon. This is achieved through breaking carbon bonds and volatilizing the non-carbonaceous portion. Because the process includes additional energy and efficiency translates to a financial consideration, this step is typically executed in a specifically designed furnace.
The final step, activation, is achieved by thermally cracking, or pyrolizing, carbon in the absence of atmospheric oxygen. Cracking of carbon particles increases surface area by increasing the number of small diameter fissures in the material. Heated anaerobically, carbon is exposed to oxygen during a gradual cooling phase, forming oxide groups on the surface of the carbon. Another activation process that is currently being used is the addition of chemical solutions that replaces the final thermal process. Common oxidizing agents are potassium permanganate, nitric acid, sodium hypochlorite, ammonium persulfate, and zinc chloride (Rao et al, 1992).
Reactants effect the portions of char with high oxidization potential, developing an extensive pore structure and a complex internal surface. The fissures, or pores, and the oxide groups are critical in the adsorptive characteristic of activated carbon. Several factors influence the quality of activation, including type and amount of oxidant introduced during the process, heating and cooling temperatures and rates, contact time between carbon and oxidant, and parent material (Rao et al, 1992).
D. Adsorption Processes
Carbon adsorption is a function of two primary processes: particle attachment and chemical reactivity. Pore-size distribution and surface area of the material dictate attachment of particles, while surface oxide groups determine chemical reactivity.
1. Particle Attachment
As mentioned in the previous section, activation increases
pore structure of carbon. Typical carbons will have a surface area of 500 -
1400 m2/g after activation. Surface areas for adsorptive material
are commonly achieved through measuring uptake of a readily adsorbed liquid,
usually iodine. Some studies have shown carbon to achieve surface areas of 2500
m2/g. Commercially available activated carbon produced from coconut
shell has a range from 1100 - 1400 m2/g. This tremendously large
surface area lends to the ability to adsorb large quantities of materials from
solution. Pores in the surface of carbon can be as small as 10 microns in
diameter, making removal of very small organic matter and metals possible
(Huang, 1978).
As a solution travels over carbon, either by gravity or localized movement depending on the process for which the carbon is being used, particles become lodged in the pores of carbon of the solution. These dissolved ‘adsorbates’ must be smaller in diameter than that of the pore in the carbon surface, allowing it to enter and amass. This is usually the case for smaller particles, which will continue to penetrate a pore until its diameter is equal to that of the pore, becoming fixed (Deithorn, 1986).
Often, reactive forces between the particle and the carbon surface aid attachment. Activated carbon consists of layers of crystalline carbon supported by weak van der Wall’s forces. The amount of impurities in the parent material will determine the structural formation of activated carbon, with the most pure carbon sources producing the most authentic crystalline configurations and the strongest localized forces (Cheremisinoff et al, 1978).
2. Chemical Reactivity
Impurities in the parent material, often other organic compounds, result in the chemical bonding of oxygen and hydrogen to the surface of carbon, or the ‘surface oxide’ group mentioned above. It can be assumed that the parent material must be ‘impure’ to form surface oxides, which are a primary factor in chemical reactivity. These oxides groups are most typically CO, CO2, OH and H2O entangled among carbon bonds on the surface (Cookson, 1978; Mattson et al, 1971).
Commonly referred to as ‘double layer’ chemical reactivity, closer scrutiny reveals multiple layers of ions. The surface of the carbon can either be charged ‘positively’ or ‘negatively’, depending on rate of heating, temperatures achieved, length of heating, cooling rates and method of oxidation. If the surface is charged, say positively, the inside layer of attracted matter will consist of negatively charged particles. The next few layers will also be negatively charged as the mass of surface oxide groups on the surface of the carbon will still be attractive. Outside the range of surface forces, designated by the ‘Helmholtz plane’, there will be diffuse forces, attracting both negatively and positively charged particles. In our positively charge carbon example, negative particles will still be under the forces of the carbon surface oxide groups and positive forces will be attracted by negatively charged particles, which are electronically reacting with the surface of the carbon (Mattson et al, 1971).
However, there are additional forces that work against the adsorption process. Since solutions are seldom homogenous in composition, oppositely charged particles will interfere with one another and sometimes inhibit adsorption. Often present in waters are solvents, like surfactants and other chemical compounds, which can alter the charge of the adsorbate, lessening the likelihood of the particle to be adsorbed. As long as the attractive force between the surface of the carbon and the adsorbate exceeds the cumulative effect of distractive forces, the particle will be chemically attracted.
3. Species Removal from Solution
Matter will continue to accumulate until there is no more
surface with which to react. Rates of degradation vary widely, depending on the
nature of the adsorbate and quality of the carbon surface. Organic matter,
which carbon is highly effective in removing, is further reduced and rendered
harmless through reduction, where a biologically active layer on the surface of
the carbon metabolizes more harmful microbes. Biological reduction attenuates
the amount of organic matter present within pores of carbon, restoring surface
area and lengthening the effective removal period. This is important for water
and wastewater treatment processes.
Due to its
irregular external surface and porous structure, carbon makes a suitable
habitat for microorganisms, which consume degradable particles seeking to
attach as well as unconverted impurities in the carbon surface. The biological
removal of bacteria through this attached growth treatment system will likely
improve the quality of the water. Bacterial colonies build up on the carbon
surface and within pores, increasing the removal of organisms over time (Tables
5-1 and 5-2). In their research, Rice et al (1978) show that using granulated
activated carbon filtration improved total coliform removal from 58% to 75%.
Table 5-1: Biological Loading for Activated Carbon Filters
|
Maturation
(time) |
g organisms adsorbed/ kg of carbon |
|
fresh |
4 |
|
2 months |
37 |
|
3 months |
63.5 |
|
7 months |
68.5 |
(Adapted
from Rice et al, 1978)
Table 5-2: Total Coliform Removal Rates for Activated Carbon
|
Processes |
Removal Efficiency |
|
Flocculation,
sedimentation, filtration |
58% |
|
Flocculation,
sedimentation, filtration and granular activated carbon filtration |
75% |
(Adapted
from Rice et al, 1978)
However, effluent from biologically active carbon can begin to
increase bacterial counts if hydraulic loading is too great, pushing out
bacteria without chance for bioreduction. Additionally, if there are soluble
carbon residues in waters with low dissolved oxygen, bacterial growth can
become anaerobic, creating sulfur dioxides and associated odor problems. A
common preventative measure used in
Gearheart et al (1996) note that using granular activated carbon for filtration equals sand in removal of residual coagulants and turbidity, and exceeds sand filtration in lowering total organic carbon. Table 5-3 depicts removal efficiency ranges for total suspended solids and oxygen demand, determined by testing a variety of activated carbon samples.
Another advantage of carbon over sand as a filter medium is that empty bed contact times for a carbon filters is around 10 - 20 minutes, far shorter than slow sand filters, allowing for equal treatment in less time (Gearheart et al, 1996).
Table 5-3: Removal Efficiencies for Activated Carbon Filters
Constituent |
Low % |
High % |
|
Total Suspended Solids |
49.3 |
55.7 |
|
Chemical Oxygen Demand |
20.1 |
35.7 |
|
Biochemical Oxygen Demand |
1.7 |
75.8 |
(Adapted from Cheremisinoff et al, 1978)
A limitation for more focused application is that effectiveness of carbon as an adsorbant depends on the pH of the adsorbate (Gearheart et al, 1996; Mattson et al, 1971). Adsorption increases as pH decreases in neutrality and, according to Rao et al (1992), effectiveness is maximum at pH 1, indicating that the effectiveness of carbon adsorption is greatest when surface oxides are alkaline in nature.
VI. Experiment
to Determine Adsorptive Capacity of Coconut Waste-Derived Carbon
To justify use as a filtration medium in developing regions,
coconut waste-derived carbon (CWD-C) must be shown to improve water quality. An
experiment was designed to examine the performance of CWD-C as a filtration
medium. Described below, the experiment was conducted in conjunction with the
Department of Environmental Resources Engineering at
A. Experiment Objectives
The objectives of the project are:
· to determine the applicability of using coconut waste-derived carbon, produced using intermediate technologies, for water purification;
· to examine removal efficiencies for turbidity and coliform for filters containing coconut waste-derived carbon; and,
· to compare the removal efficiencies of coconut waste-derived carbon with other filtration media.
B. Methods
The process design for the examination of the performance of the CWD-C entailed passing water through filters. Filter materials were standardized to decrease design error. To scrutinize the progressive performance of filter media and to make comparisons between media, tests were conducted over a period of one month. The data has been analyzed using a variety of statistical methods and the results are outlined at the end of this section.
1. Indicators
Typically in water quality testing, indicator species are
chosen and monitored. Coliform and turbidity are two of the primary
constituents used for water quality standards in the
2. Filter Design
Experimental filters were constructed to be similar in composition to filter technology propagated throughout the developing world. This filter design was chosen because it most closely represents the intended use of the material, which is as an enhancement to domestic sand filters in developing countries.
Sand filters propagated by United States Peace Corps Volunteers in coastal region of Kenya consist of a 20-liter, high-density polyethylene (HDPE) bucket with a perforated bottom placed upon another 20-liter bucket used for storing of filtered water (Wright, 1991). In the top bucket is placed 5 to 8 cm of small stones, less than 5 cm in diameter, topped with 5 to 8 cm of smaller aggregate, roughly 1 cm in diameter, upon which 15 cm of sand is placed. This layering is used to prevent the filtration media from passing through the perforated bottom, contaminating the stored drinking water, possibly clogging up the holes in the bucket, and causing the filter to fail prematurely. This design maximizes purification of water at a minimum of cost and maintenance.
These filters are slow sand filters, meaning they utilize biological activity within the top layer of sand, or ‘schmutzedecke’, for further reduction of organic matter. Slow sand filters have been shown to produce an effluent with a turbidity below 1.0 NTU, an ideal level for consumption (Roecklein, 1993). By adding carbon to the top of the filter, the performance of the filter is enhanced through the adsorption of species, and by providing a more suitable habitat for the growth of contaminant-reducing microorganisms.
In theory, since performance of a filter is dependent mainly on the vertical cross-section of filter media, it is not necessary to use a bucket of equal diameter to the 20-liter buckets employed in the Peace Corps sand filter. It was expected that a minimum diameter of 20 cm could be used without biasing the outcome as a result of ‘side-channeling’. A 20-cm low-pressure polyvinylchloride (PVC) drainage pipe was chosen as the housing for the filter media. To achieve the depth required for the modified filters while leaving enough freeboard for infiltration of water, the 20-cm PVC pipe was cut to a minimum of 50 cm in length. A section of wire mesh was secured to the bottom of each pipe, serving as the platform upon which media was placed.
1 2 3 4 5 6
ß ß ß ß ß ß
7 cm
50cm
15 cm
ß ß ß ß ß ß
1 - 15 cm sand and 7 cm commercially-activated carbon - commercially-
2 - 15 cm sand and 7 cm
commercially-activated carbon activated carbon
3 - 22 cm of sand
4 - 22 cm of sand - sand
5 - 15 cm sand and 7 cm coconut waste-derived carbon
6 - 15 cm sand and 7 cm
coconut waste-derived carbon -
coconut waste-
derived carbon
Figure 6-1: Filter Types Used in Testing.
As described above, 1-cm aggregate was loaded on top of 5-cm aggregate lining the bottom of the filter (not shown in the above schematic drawing). On top of the aggregate was placed 15 cm of sand and 7 cm of either CWD-C and commercially-available activated carbon (CAC), or another 7 cm of sand so that all filters had 22 cm of media. Before filter construction, all of the aggregate was rinsed three times with boiled water to remove contaminants and minimize interference with filter performance criteria. Since the water being filtered was not used for consumption, there was no need to replicate the bottom storage bucket of the Peace Corps filter, although a collection system was used to facilitate sample collection and prevent puddling at the test location.
C. Sample Preparation
1. Coconut Waste-Derived Carbon (CWD-C)
Shells of coconuts, imported from
Once removed from the wood-stove, the cookie tin was opened slightly, allowing for oxidation of the carbon surface during cooling. This is the most important step in activating carbon for chemical reactivity adsorption, according to Mattson et al (1971). Because of the relatively low temperature of pyrolization and the method chosen to expose carbon to oxygen, this carbon is considered ‘slow-cooled’, meaning it is designed for removal of acidic substances.
This imperfect method of producing carbon is used to produce a material for testing which will closely resemble the carbon resulting from a project implemented in developing regions. This has been done to avoid biasing results of the experiment should carbon produced through a more controlled thermal environment be used for examination.
The resulting char was rinsed three times with water boiled for 20 minutes to remove surface ash content and char dust, which would otherwise be dislodged during filter usage. For more stringent laboratory experiments, de-ionized or distilled water would be used as a scouring agent. However, using boiled water is consistent with what is expected in villages in developing regions. The 20-minute boiling standard was set by the World Health Organization during the 1980’s Water and Sanitation Decade for the household purification of domestic potable water (WHO, 1987).
Seven coconuts (1.75 kg of coconut shell) were used to produce 0.33 kg, or 1,325 cubic centimeters, of sample carbon used in filter number 5, a CWD-C filter, for testing. The density of the carbon was 248.9 kg/m3. This procedure was replicated, using the same heating times, to produce an equal amount of material for testing of a parallel filter, number 6, replicating natural variation.
As a justification for this informal production of carbon for experimental purposes, Mattson et al (1971) conducted analyses of sugar carbons activated by placing coconut material in a porcelain crucible, with the top left ajar for oxidation, in a muffle furnace at 600°C for 6 hours. Although the temperature obtained for the creation of the experimental batch of coconut waste-derived carbon in this current experiment was not maintained at a constant 600°C for the duration of the heating, the temperature was in the range of 550°C and 700°C for the same amount of time. Further, Rao et al (1992) thermally activated carbon by placing coconut shells inside a small metal box, and immersing the box in sand, which was within a larger metal box. The unit was then heated to 800-850°C in a muffle furnace and allowed to cool over a 10 hour period, yielding activated carbon samples for laboratory experimentation.
2. Commercially-Activated Carbon (CAC)
The original samples of CAC were obtained from the Calgon
Carbon, Corporation located in
3. Sand
When designing a sand filter, it is important to know the size of the sand that is being used. Knowing average particle size allows calculation of hydraulic loading, which controls design and operation of sand filters. For water treatment, it is ideal to have a well graded sand, with a uniformity coefficient in the vicinity of 1.4 to 1.8, and an effective size between 0.7 mm and 1.4 mm (Tchobanoglous, 1987). Sand for this experiment was obtained at the closest possible source, the local hardware store. In developing countries sand would be obtained also at the closest possible site, which might not be a commercial and standardized source.
Results of the sieve analyses for two 500-kg samples of sand used are shown in Figure 6-2. The uniformity coefficient for the means of the two samples is 2.8 with an effective size of 0.65 mm. Engineers designing sand filters for larger projects in more industrialized and regulated countries will utilize this information to calculate head loss, run length, and backwash velocities required to maximize operation. Since these filters are being used where this design is a ‘best alternative’, concern for precision of sand quality is diminished.
Sieve analysis indicates this is a well-graded sand and has an effective size that is slightly smaller than the 0.70 mm threshold suggested by Tchobanoglous. Since each of the six filters is using sand from the same source, expected performances of the filters, amendments notwithstanding, should be nearly identical. This will enable isolation of filter amendments for study.
Figure
6-2: Sand Sieve Analysis. For filter
media in the six filters used in experiment.
Figures are averages for the sieve analyses performed.
D. Experimentation
To determine the effectiveness of CWD-C enhanced filters compared to that of CAC enhanced filters and the Peace Corps sand filter, water samples were taken from each after passing through the filter. To facilitate determining statistical significance by replicating ‘within treatment’ variance, two experimental filters of each type were constructed, resulting in a total of six filters (Figure 6-1).
Since it is known that activated carbon loses ability to adsorb matter once the pores have reached maximum capacity (Rice et al, 1978), the performance of the filters over time was analyzed in an attempt to determine exhaustion. However, due to resource constraints, a limit was placed on the amount of water passed through the filter for the test. The CWD-C was scheduled to be tested over a period equivalent to two month’s use by a typical family in a developing region. Some non-arbitrary factors leading to establishing this limit is that this would require users to replace the media every two months, or six times a year, calling for carbon from 168 coconuts annually per filter, a very attainable amount. Furthermore, when working with families, Peace Corps Volunteers recommend that the media in the sand filters be washed or replaced every two months. Additionally, depending on quality of water being filtered, it has been noticed that the flow rate of a typical Peace Corps sand filter diminished markedly after this amount of time, primarily due to poor quality of raw water (Wright, 1991).
However, the determining factor is not actually time, but quantity of material accumulated in the pores of the activated carbon, which, using this filtration method, is a function of the quality and the volume of water filtered. The quality of drinking water sources varies widely and is essentially uncontrollable. Yet, to emulate the poor quality of water consumed in many situations around the world, it was decided to use water from Widow White Creek in McKinleyville, California, which is a watercourse running through residential, light industrial, commercial and agricultural sections of the town. After this site proved infeasible, an alternative source was located, extracting effluent from Hauser Marsh, the last of the three finishing marshes in the Arcata Wastewater Treatment Facility. Since the limit now focused upon quantity, the time constraint was removed, allowing for accelerated testing.
To reach the limit of two month’s use for a family of eight people, meeting the 7 liter-per-day-per-person standard set by the World Health Organization (1987), 3,360 liters would need to be filtered per unit. However, since the filters being used have one-fourth the surface area of the Peace Corps filter, the actual amount of water to be filtered per unit is reduced to 25%, or 820 liters. This figure was later reduced to 360 liters to expedite completion of the research. Subsequently seven samples, systematically chosen, were taken from every 60th liter for each filter, using unfiltered water as control measurements, thus compiling the sample pool for analysis.
E. Data Collection
Water was loaded into each filter using a 4-liter jar to
transport the Hauser Marsh effluent, accessed through a concrete cistern
adjacent to a fish bioassay at the Arcata Wastewater Treatment Facility. Once
filtered, flow rates were recorded and the desired samples were taken to the Water
Quality Laboratory at
As mentioned above, the primary focus of testing was to measure levels of total and fecal coliform. The ‘filter membrane method’ of detection, as described by Clesceri et al (1989), was used to measure levels of coliform in all seven water samples - one from each of the six filters and one of unfiltered water. The amount of water to be filtered for each test was 0.5 mL for total coliform and 100 mL for fecal coliform, quantities determined through a preliminary testing period of three days.
To improve probability of determining, with statistical significance, whether the sand filters enhanced with coconut-waste derived carbon outperform filters with only sand, a power analysis was implemented to determine number of data points needed. With four samples from each filter type and the unfiltered water, a power of 85% was obtained. Increasing the samples to five resulted in an increase in power of only 3%. It would have required six data points per filter type to obtain 90% power and ten to obtain 95% power. These additional points were determined to be infeasible from a ‘return-on-investment’ perspective, as the cost of laboratory material is prohibitive and funding for the project was provided solely by the author. Therefore, four filter-membrane tests were run for the unfiltered water and for each filter type, with two tests per filter for both total and fecal coliform.
In addition to collecting coliform data, measurements recorded for each filter run included dissolved oxygen (DO), turbidity, conductivity and pH. The purpose of these measurements, besides the relative simplicity of the tests, is to determine if these constituents are affected by interaction with the carbon.
F. Data Analysis
Three methods of analysis were used to examine collected data. The first was to inspect manually the data for each constituent, noting any irregularities and searching processes for sampling or data entry error. Second, data was entered into Microsoft Excel 7.0Ó. Averages and removal efficiencies were calculated, data cross-referenced and ordered to facilitate further analysis. This data was plotted to compare visually the performance of each filter and filter type. Finally, the statistical software package, Minitab 9.1Ó, was used to determine means, confidence intervals, and analyses of variance (ANOVAs).
G. Results
1. Raw
Water Quality
Raw, unfiltered water, had an average dissolved oxygen content (DO) of 4.7 mg/L, an average turbidity of 2.73 NTU, an average conductivity of 376 microMHOS, and an average pH of 6.0 over the test period. Without the spike that occurs in turbidity on the final day of testing, average turbidity is under 1.5 NTU. Results from filter membrane tests showed an average of 150,800 total coliform and 12 fecal coliform colony forming units (CFU) per 100 mL.
As a drinking water supply, this water is considered unsafe by many drinking water standards established in industrialized countries (Tchobanoglous, 1987; World Health Organization, 1987). This is due to relatively low level of dissolved oxygen, a result of biological uptake, and high coliform counts, which should ideally be below 1 CFU per 100 mL, as indicated by the California Department of Public Health (Tchobanoglous, 1987). The high level of coliform in this water proved beneficial to the experiment, as there is high potential for significant reduction in bacteria, especially in terms of total coliform.
2. Filtration
DO, turbidity, conductivity and pH measurements for each filter type over the testing period and in relationship to raw water are shown in Figures 6-3, 6-4, 6-5 and 6-7.
a. Dissolved Oxygen
The rise in dissolved oxygen concentration was not trivial. Filtration through each filter type caused a 60% increase in DO on average, although there was little difference between filter types themselves. This is believed to be caused by aeration when the water passed through the filter but, more importantly, from the mixing that occurred when the droplets of filtered water were collected in the sample bottles for testing. While a substantial improvement in water quality ensued, and certainly worthy of further study, it is likely that it is unrelated to the media used in the filtration process.
Figure 6-3: Comparison of Dissolved Oxygen. Measurements for the filter types are averages.
b. Turbidity
As mentioned above, there is little turbidity in the Hauser Marsh effluent and this is a consequence of extensive tertiary treatment of natural wetland systems. Turbidity measurements for filtered water from the first day, after one liter had been filtered, were quite high, doubtless the result of fine particles of sand and carbon dust still present on the media. After 60 liters had been filtered, and probably much before, turbidity levels were satisfactorily low and, before 200 liters had been filtered, the readings were nearly identical for raw water and filtered samples.
Figure 6-4: Comparison of Turbidity. Measurements for the filter types are averages.
On the final day of testing, a fortunate event occurred. The water in the cistern was discernibly more turbid, as a result of rainfall, allowing the filters an opportunity to remove some visible material. All filters lowered the turbidity in this water, with the CWD-C outperforming the other two filters by 80%. Note there was only one observation where considerable turbidity was present and no significant conclusions should be inferred from a single observation. However, the CWD-C filters, on average, performed comparably to the sand and CAC filters. Future research in this area may choose to use more turbid water to focus on the turbidity removal capacity of CWD-C.
c. Conductivity
Throughout the literature reviewed for this project, many scientists praised the capacity for carbon to adsorb ionic species from concentration (Arulananatham et al, 1989; Collins et al, 1990; Mattson et al, 1971; Rao et al, 1992; Rice, et al, 1978). It was believed that monitoring conductivity levels of the water before and after filtration would indicate whether there was a reduction in ion concentrations, provided substantial ions were present in the raw water.
Average conductivity levels throughout the experiment are such that it is improbable that there were an unusually high level of ions in solution. The treatment the water receives in the marsh system generally leads to high ion removal rates (Gearheart et al, 1996). Average conductivity in the raw water, 376 microMHOs, is not greatly different than the 389 to 394 microMHOs range of the filtered water.
Since levels were higher in the filtered water, not lower as expected, re-evaluation of the processes of adsorption from solution should be conducted. One possible explanation is that some oxide groups on the carbon surface may have been unstable and became soluble, adding to the net ion concentration and an increase in electrical conductivity. Another possibility is that salts contained in the sand material have dissolved in solution and have affected the measurements.
Figure
6-5: Comparison of Conductivity. Measurements for the filter types are
averages.
d. pH
Average pH, or hydrogen ion concentration, of the filtered water was generally higher than the raw water (6.2, 6.3 and 6.2 for CAC, sand and CWD-C, respectively, to 6.0 for the raw water). There was virtually no difference in pH between filter types. In no case was the change in pH greater than 0.6, which occurred between one sand filter and one CWD-C filter on the first day of data collection. Since the first day of testing, the greatest deviation was 0.4, an improvement made by CDW-C filters over the raw water.
Figure 6-6: Comparison of pH. Measurements for the filter types are
averages.
e. Flow Rate
As represented in Figure 6-7, flow rates for each filter gradually decreased in an asymptotic fashion, as is to be expected for the run length of a filter (Gearheart et al, 1996). Clogging actually works in favor of removal efficiencies for two reasons. First, since the pores are decreased in size and number, particulate matter has less opportunity to pass through a larger opening, becoming ‘captured’ in the media. Second, the increased time water takes to pass through the filter allows for increase contact time with the biological layer and a higher level of bio-reduction of pathogens and adsorption of contaminants. A filter that follows this pattern of flow rate can be assumed to be acting properly, increasing treatment until all pores are ‘clogged’ and must be cleaned.
Figure 6-7:
Comparison of Flow Rate. Measurements
for the filter types are averages.
Flow rates are generally consistent with what is expected of slow sand filters, in the range of 80 to 400 L/m2 * min (Tchobanoglous, 1987), although, toward the end of the experiment, flow rates decreased below the suggested lower limit, indicating that the filters may be close to failure.
f.
Coliform
Total average removal efficiencies for the three filter types are shown in Table 6-1. The CWD-C filters were able to remove more of the bacteria than the other two filter types, and considerably more in the case of the total coliform. Figures 6-8 and 6-9 depict average levels of total and fecal coliform for each filter type and raw water.
To determine significance using statistical methods, data was first tested for normality. Data sets proved to be highly normal, passing both Bartlett’s and Levene’s test, with p-values less than 0.1%. A summary of the homogeneity of variance test is included in Table 6-2.
Table 6-1. Coliform Removal Efficiencies
|
Filter Type |
Fecal Coliform |
Total Coliform |
|
CAC |
55.7% |
34.8% |
|
Sand |
60.0% |
37.0% |
|
CWD-C |
61.4% |
49.3% |
Table 6-2: Homogeneity of Variance Test
|
|
Bartlett’s
Test |
Levene’s Test |
||
|
|
Test Statistic |
p-value |
Test Statistic |
p-value |
|
Total Coliform |
25.023 |
0.000 |
4.067 |
0.009 |
|
Fecal Coliform |
39.941 |
0.000 |
11.145 |
0.000 |
Figure 6-8: Comparison of Fecal
Coliform. Measurements for the filter
types are averages.
Figure 6-9: Comparison of Total Coliform. Measurements for the filter types are averages.
A series of ANOVAs was conducted testing difference in the filters, with comparison tests on filter types in relation to one another. Tables 6-3 and 6-4 outline important results from these tests, including p-values for each testing cycle, comparison tests (noting either pass or failure) and the p-value for ANOVAs on the sand and the CWD-C filters. Although the filters themselves noticeably improved the quality of the raw water from the standpoint of coliform removal, variation in the data and the relatively small number of data points obtained for each filter led to an inability to find statistical significance for the hypothesis that CWD-C enhanced filters perform better than filters with just sand.
Table 6-3: Summary of Statistical Analysis for Fecal Coliform
|
Fecal Coliform |
|||||
|
|
|
Comparison
Tests for Sand and CDW-C |
|
||
|
Day |
p-value |
Dunnett’s |
Tukey’s |
Fisher’s |
p-value** |
|
1 |
0.106 |
F |
F |
F |
0.045 |
|
2 |
0.016 |
F |
F |
F |
0.435 |
|
3 |
0.000 |
P |
P |
P |
0.025 |
|
4 |
0.000 |
F |
F |
F |
0.036 |
|
5 |
0.001 |
P |
F |
F |
0.504 |
|
6 |
0.000 |
P |
F |
F |
0.693 |
** Second p-value is for
the ANOVA performed directly between sand and CWD-C filters.
P = pass, F = fail.
Table 6-4: Summary of Statistical Analysis for Total Coliform
|
Total Coliform |
|||||
|
|
|
Comparison
Tests for Sand and CDW-C |
|
||
|
Day |
p-value |
Dunnett’s |
Tukey’s |
Fisher’s |
p-value** |
|
1 |
0.055 |
P |
F |
F |
0.127 |
|
2 |
0.242 |
F |
F |
F |
0.382 |
|
3 |
0.070 |
F |
F |
F |
0.003 |
|
4 |
0.058 |
F |
F |
F |
0.253 |
|
5 |
0.213 |
F |
F |
F |
0.473 |
|
6 |
0.000 |
P |
F |
F |
0.387 |
** Second p-value is for
the ANOVA performed directly between sand and CWD-C filters.
P = pass, F = fail.
H.
Confounding Elements
Possible areas of bias for the experiment are that the filters may have allowed ‘side-channeling,’ or water moving down the side of the filter, eluding the filter media and avoiding treatment. A 20-cm PVC pipe was chosen to decrease the potential for this but, since total coliform rates are very high and the filters are biologically immature, it is difficult to say whether this became a factor in the results.
If bacteria were to grow on the PVC pipe, this could be considered a reflection on the material used for the testing, rather than the HDPE buckets which are typically used in the Peace Corps filters. This nuisance, as with the previous one, is consistent from filter to filter and insignificant, unless there is some reactivity between carbon and the filter material. Any reaction between media and HDPE is presently unknown.
Additionally, even though much consideration was given to
replicate processes as they would be implemented in the developing world, there
are a number of inconsistencies between CWD-C samples used in the experiment
and that expected to be made ‘on-site’. First, the sample carbon is derived
from a coconut variety from
Finally, the data was collected for a period only equivalent to one month’s use and the exhaustion point of neither sand nor carbon filters was attained. In fact, due to an unexplained occurrence in the laboratory on the final day of testing, data points for coliform at 360 liters of water filtered were returned as null. It is obvious that additional data points are required to remedy these shortcomings.
The actual effects that may have occurred due to the above mentioned considerations are presently unclear. Additional study may chose to develop strategies that neutralize any effects these factors may cause.
I. Conclusion of Testing
Though the experiment was unable to determine, with statistical significance, the difference in the function of the filters, there are indications that the CWD-C filters did outperform the sand filters for removal of total coliform, notably for the third, fourth and fifth testing cycles. This shows there is potential for CWD-C enhanced filters to improve water quality over simple sand filters, and that more extensive data should be collected.
Due to the relative ease and low cost of accumulating and processing coconut shells in developing regions, the importance of diverting agricultural waste from landfills, and the marked increase in water quality of the CWD-C filters over the sand filters, it is recommended that this technology be implemented on a test basis until such a time as more significant data collection can be supported.
VII. Alternative
– Energy Production
As mentioned previously, this report offers only one of various projects for which implementation of waste reutilization can occur. This section takes a look at an alternative of producing energy for communal use from coconut residues.
It is advantageous for any
community to thermally convert organic wastes for energy because it reduces the
amount of waste needing to be interred by as much as 95% (Tchobanoglous, 1993).
This is critical in urban areas where overconsumption and incomplete
utilization of exploited resources result in huge quantities of waste. It is
even more beneficial to convert wastes in countries that are economically
disadvantaged. These countries typically cannot increase their energy
production to meet international standards. A large percentage of fossil fuel
consumption in these countries is devoted to vehicles and heavy industry,
leaving light manufacturing and domestic energy demands unfulfilled (Shuler,
1980).
In many regions, agricultural
residues, such as dung, straw and cob, have been used for centuries as fuel
sources. Societies in the Nile, Ganges and
Combusting refuse derived fuels
(RDF) in incinerators for large-scale energy production has been practiced in
the
In addition, alternative and more efficient means were developed (Alter, 1980). A common method of maximizing resource efficiency is to operate a cogeneration facility. One possible scenario could be to utilize the entire load of coconut shells for energy production through combustion, while using the energy to manufacture processed goods from the husk of coconuts.
There are 20,500 coconuts available per day in
Consideration can be given to provide electricity to a
portion of the city. Looking at the population of 1 million, and using the
consumption figure of 117 kW-hr/yr per capita in
VIII.
Conclusion
Visiting the beautiful island-city of
This project has shown removing 20,500 coconuts from the
waste stream in
Reducing solid waste dumped in
References
1. Alter,
Harvey, J. J. Dunn, Jr., 1908. Solid Waste Conversion to Energy. Marcel Dekker, Inc.,
2. Arulananatham,
A., N. Balasubramanian and T.V. Ramakrishna, 1989. “Coconut Shell Carbon for Treatment if
Cadmium and Lead Containing Wastewater.” Metal Finishing, November,
1989. Indian
3. Blaak,
Gustaf, 1993. “FAO Support for Coconut Research and Development between
1981-1991. The Coconut Economy. V.L.
Chopra, ed. International Science Publisher,
4. Brown,
Lester R., Christopher Flavin and Hilary French, 1998. State of the World. Worldwatch Institute. W.W. Norton &
Company,
5. Carr,
Marilyn, 1984. Blacksmith, Baker,
Roofing-sheet Maker. Intermediate Technology Publications,
6. Cheremisinoff,
Paul and Angelo C. Morresi, 1978.
“Carbon Adsorption Applications.” Carbon
Adsorption Handbook. Ann Arbor
Science Publishers, Inc.
7. Central
Intelligence Agency (CIA), 1997. The
World Factbook: 1997-1998. Brassey’s,
8. Clesceri,
Lenore S.,
9. Collins, M. Robin, T. Taylor Eighmy, James P. Malley, Jr., Mike Bauer, and A.J. Rachwal, 1990. “Slow Sand Filter Modifications to Enhance Operational and Treatment Performance.” Presented at AWWA Seminar Titled Emerging Technologies and Practices on June 16, 1990.
10. Cookson,
John T., Jr., 1978. “Adsorption
Mechanisms: The Chemistry of Organic Adsorption on Activated Carbon.” Carbon Adsorption Handbook. Ann Arbor Science Publishers, Inc.
11. Copeland,
Edwin Bingham, 1931. The Coconut.
MacMillan and Co., Limited, St. Martin's Street,
12. Daniels,
David L., Simons N. Cousens, Lucia N. Makoae, and Richard G. Feachem. 1990. ”A
study of the association between improved sanitation facilities and children’s
height in
13. Diethorn, Robert T., and Anthony F. Mazzoni, 1986. “Activated Carbon: What Is It, How It Works.” Water Technology, Vol. 9, No. 8., pp. 26-29.
14. Deivanai,
K., Bai, R.K., 1995. Batch Biomethanation of Banana Trash and Coir Pith. Bioresource
Technology, volume 52, number 1, p. 93-94,
15. Donohue,
William, 1996. Personal conversation with Public Relations Director for Calgon
Carbon, Corporation,
16. Duke, James
A., 1989. CRC Handbook of Nuts. CRC
Press,
17. Edachal,
Muhammad, and T.B Nanda Kumar, 1988. Coconut Economy in
18. Erhlich, Paul
R. and Anne H. Erhlich, 1991. The
Population Explosion. Simon and
19. Festin,
Teodorico F.; Jose,
20. Freeman-Grenville,
G.S.P., 1962. The
21. Gearheart,
Robert A., M. McKee and Brad Finney, 1996.
WAWTTAR: An Algorithm for Water
and Sanitation for the Developing World.
22. Gergova, K., N. Petrov, L. Butuzova, V. Minkova, and L. Isaeva, 1993. “Evolution of the Active Surface of Carbons Produced from Various Raw Materials by Steam Pyrolysis/Activation.” Journal of Chemical Technology and Biotechnology
23. Gore,
Albert, 1992. Earth in the Balance.
Houghton Mifflin Company,
24. Hancock,
Graham, 1989. The Lords of Poverty.
Penguin Press,
25. Huang, C.
P., 1978. “Chemical Interactions Between
Organics and Activated Carbon.” Carbon
Adsorption Handbook. Ann Arbor
Science Publishers, Inc.
26. Jeanes,
Henry, 1997. “Proposal: Solid Waste
Recycling Initiative –
27. Laine, L.,
A. Carafat, and M. Labady, 1989.
“Preparation and Characterization of Activated Carbons from Coconut
Shell Impregnated with Phosphoric Acid.” Carbon, Vol. 27, No. 2, pp.
191-199. Pergamon Press, Inc.,
28. Mattson,
James S. and Harry B. Mark, Jr., 1971. Activated Carbon: Surface Chemistry and
Adsorption from Solution. Marcel
Dekker, Inc.
29. Namasivayam
, C.; Kadirvelu , K., 1994. Coirpith, an Agricultural Waste By-Product, for
the Treatment of Dyeing Wastewater. Bioresource
Technology, volume 48. number1, pp79-81,
30. Okun, Daniel A., 1991. “A Water and Sanitation Strategy for the Developing World.” Environment, Volume 33 Number 8, pp. 16-2, 38-43.
31. Ominde,
S.M., 1968. Land and Population Movements
in
32. Pomeroy,
Roger F. C., 1981. Pyrolytic Methods in Organic Chemistry. Academic Press,
33. Rao, Padaki
Srinivas, Shashikant R. Mise, and G.S. Manjunatha, 1992. “Kinetic Study on Adsorption of Chromium by
Coconut Shell Carbons from Synthetic Effluents.” Journal
of Environmental Science and Health.
Volume A(28), 227-2241. Marcel
Dekker, Inc,
34. Reuters
News Services, 1996. “Tanker Still in
35. Rice, Rip
G., G. Wade Miller, C. Michael Robson, and Wolfgang Kühn, 1978. “A Review of the Status of Preozonation of
Granular Activated Carbon for Removal of Dissolved Organics and Ammonia from
Water and Wastewater.” Carbon Adsorption
Handbook. Ann Arbor Science
Publishers, Inc.
36. Roecklein,
Charles, J., 1993. “Westhaven Water
Treatment System Engineering Report.” SHN Consulting Engineers &
Geologists,
37. Shuler,
Michael L., 1980. Utilization and Reuse
of Agricultural Wastes and Residues. CRC Press,
38. Sofer,
Samir S., Oskar R. Zaborsky, 1981. Biomass Conversion Processes for Energy and
Fuels. Plenum Press,
39. Tchobanoglous,
George, Edward D. Schroeder, 1987. Water
Quality. Addison-Wesley Publishing Company,
40. Tchobanoglous,
George, Hilary Theisen and Samuel Vigil, 1993. Integrated Solid Waste Management: Engineering Principles and
Management Issues. McGraw-Hill, Inc.,
41. Tiffen,
Mary, Michael Mortimore and Francis Gichuki, 1994. More People, Less Erosion: Environmental Recovery in
42. UNICEF Annual Report on the State of
43. World
Health Organization, 1987. Guidelines for
Drinking Water Quality. World Health Organization,
44. Wright,
Charles, 1991. “Domestic Water Filtration.” United State Peace Corps Technical
Paper.
Appendix A: Analysis of Coconut Waste
As evidenced by the waste stream characterization, a considerable amount of dump space is used for coconut waste. To approach the reuse of this waste product, an understanding of the material must be obtained.
A. The Coconut
Cocos nucifera, also known as the
coconut tree, is of uncertain origin. Some historians believe the tree is
native to the
The nut of the tree is encased in a thick cellulose husk, which provides protection from birds and other animals that have the ability to reach treetops. Fruit, consisting of the husk and nut, develops from an oblong shape to nearly spherical, averaging 20 cm in length and equal in width when fully mature. When coconuts are harvested for their water, they are removed from the tree before their fruit is ripened; this way there is the maximum water. If the fruit is ripe, the quantity of water will be diminished, because the fruit uses the water for self-nourishment.
B. Uses of Coconut Materials
The coconut is one of the ten most useful plants in the world
(Duke, 1989). The roots are known to be used for traditional medicines, and
research continues to study additional medicinal applications for all parts of
the tree. Coconut oil is believed to be the least harmful of all agricultural
oils and discoveries have led to the conclusion that it actually improves
physiological well-being (Duke, 1989). The oil is also used locally for
cooking, as a salve and a lubricant. Recent findings have indicated that
coconut oil can be used for the eradication of mosquito populations, thus
reducing the potential for malaria. Additionally, coconut oil has been used for
lamp fuel in the
Uses of carbon made from coconut shells include cigarette filters and air filter charcoal. Recent research has focused on the pyrolization of coconut shell to produce carbon for the absorption of organic matter from water, specifically in removing dyes from the effluent of textile manufacturing (Namasivayam et al, 1994). The wood from the tree is less useful removed from the ground than it is as a coconut producer, however, in the occurrence of fallen trees, wood is often used in the construction of canoes and roofing in traditionally built homes. The wood is not used as a domestic fuel sources as the wood is extremely dense and temperature required for combustion is extremely high. A comprehensive list of uses is given in Figure A-1.