c

ZURICH UNIVERSITY
OF
APPLIED SCIENCES

RESEARCH ON ANAEROBIC DIGESTION

OF ORGANIC SOLID WASTE

AT HOUSEHOLD LEVEL

IN DAR ES SALAAM, TANZANIA

Bachelor Thesis

by
Lohri Christian Riuji
Bachelor of Science 2005

Institute of Natural Resource Sciences

23.01.2009

Supervisors Switzerland
Prof. Dr. Urs Baier, Zurich University of Applied Sciences, ZHAW

Yvonne Vögeli, Swiss Federal Institute of Aquatic Science and Technology, Eawag
Department of Water and Sanitation in Developing Countries, Sandec

Supervisors Tanzania
Prof. Dr. Gabriel R. Kassenga, ARDHI University, Dar es Salaam

Dr. Shaaban M. Mgana, ARDHI University, Dar es Salaam

Impressum
Keywords
Anaerobic digestion, ARTI Compact biogas system, biogas, solid waste management, organic
solid waste, resource recovery, renewable energy, developing countries, Tanzania

Suggestion for citation

Lohri, C. (2009): Research on anaerobic digestion of organic solid waste at household level in
Dar es Sala am, Tanzani a; Bachelor Thesis at ZHAW (Zurich University of Applied Sciences) in
collaboration with EAWAG (Swiss Federal Institute of Aquatic Science and Technology)

Address
Christian Lohri, Stadelstrasse 23, 6048 Horw, Switzerland, E-mail: [email protected]
Abstract

R a p i d u r b a n i s a t i o n a n d p o p u l a t i o n g r o wt h h a v e m a g n i f i e d t h e n e c e s s i t y f o r a d e q u a t e s o l i d
wa s t e m a n a g e m e n t t h r o u g h o u t t h e wo r l d . I n o r d e r t o m i n i m i s e t h e r i s k t o t h e e n v i r o n m e n t
and human health, economically feasible solutions are sought for the treatment of solid
wa s t e , p a r t i c u l a r l y i n u r b a n a r e a s o f l o w- a n d m i d d l e - i n c o m e c o u n t r i e s .

T h i s p a p e r e v a l u a t e s t h e s u i t a b i l i t y o f t h e A R T I C o m p a c t b i o g a s s ys t e m ( C B S ) a s a d e c e n -
t r a l i s e d l o w- t e c h t r e a t m e n t o p t i o n f o r t h e o r g a n i c f r a c t i o n o f h o u s e h o l d wa s t e i n D a r e s
S a l a a m ( D S M ) , T a n z a n i a . A l i t e r a t u r e r e v i e w a b o u t s o l i d wa s t e a n d e n e r g y m a n a g e m e n t i n
D S M wa s f o l l o we d b y t h e i n s t a l l a t i o n a n d m o n i t o r i n g o f a r e s e a r c h C B S wi t h t h e a i m o f
o b t a i n i n g p r e c i s e d a t a c o n c e r n i n g t h e p e r f o r m a n c e o f t h e d i g e s t e r a s we l l a s q u a n t i t y a n d
quality of the biogas and effluent. The unstirred floating-dome digester (capacity 850l, pH:
6 . 5 , T e m p . : 3 0 ° C ) wa s f e d wi t h t wo s u b s t r a t e s : F o o d wa s t e ( F W : 2 4 % T S , 9 1 % V S ) a n d
m a r k e t wa s t e ( M W : 1 0 % T S , 8 8 % V S ) . T h e d a i l y l o a d o f 2 k g wa s t e ( d i l u t e d wi t h 1 8 l o f wa -
t e r ) wa s b a s e d o n l i t e r a r y f i n d i n g s a n d r e p r e s e n t s a r e a l i s t i c f i g u r e o f g e n e r a t e d wa s t e
q u a n t i t y p e r T a n z a n i a n h o u s e h o l d o f 5 m e m b e r s . T h e r e d u c t i o n o f wa s t e v o l u m e ( F W : T S
reduction by 84.9%, MW: TS reduction by 72.8%) and organic load (FW: VS reduction by
92.2% and COD reduction by 83.1%; MW: VS reduction by 85.3% and COD reduction by
8 4 . 2 % ) we r e s u b s t a n t i a l a n d i l l u s t r a t e t h e e f f e c t i v e n e s s o f t h e d i g e s t e r . T h r o u g h t h e a n -
a e r o b i c p r o c e s s , t h e a m m o n i a c o n c e n t r a t i o n ( N H 4 - N ) wa s r a i s e d b y 1 3 4 . 5 % t o 7 4 . 1 m g / l
( F W ) a n d 2 0 6 . 5 % t o 2 7 . 9 m g / l ( M W ) , wh i c h e n h a n c e s t h e u s e f u l n e s s o f t h e e f f l u e n t a s a n
o r g a n i c f e r t i l i z e r . O n a v e r a g e , t h e a n a e r o b i c d i g e s t i o n o f f o o d wa s t e r e s u l t e d i n 6 4 0 l
( 5 7 4 N l ) b i o g a s p e r k g V S wi t h m e t h a n e c o n t e n t o f 5 6 . 8 % . 7 0 0 l ( 6 2 8 N l ) b i o g a s p e r k g V S
wi t h 6 6 . 4 % C H 4 we r e p r o d u c e d f r o m m a r k e t wa s t e ( h yd r a u l i c r e t e n t i o n t i m e f o r b o t h f e e d -
s t o c k s : 4 2 . 5 d a y s ) . K i t c h e n wa s t e i n D S M c o n s i s t s o f a n e q u a l m i x o f t h e s e t wo s u b s t r a t e s
a n d wa s f o u n d o u t t o p r o d u c e a p p r o x i m a t e l y 2 0 0 l o f b i o g a s wh i c h i s s u f f i c i e n t f o r 4 5 m i n o f
cooking on a gas stove (equivalent to 1/3 of the daily cooking time per average house-
h o l d ) . A f a m i l y wi t h 5 m e m b e r s c a n h e n c e s a v e 2 8 k g o f c h a r c o a l p e r m o n t h s ( 2 3 ’ 0 0 0 T Z S )
wh i c h d e f i n e s t h e a m o r t i s a t i o n p e r i o d o f a C B S ( p r i c e 8 5 0 ’ 0 0 0 T Z S ) a s 3 y e a r s .
An inspection tour to the 12 installed CBS at household level in DSM revealed that only 4
s ys t e m s we r e i n o p e r a t i o n . A l t h o u g h t h e c o n s e q u e n t i a l e f f o r t s b y A R T I - T Z r e s u l t e d i n a n
i n c r e a s e t o 1 0 C B S b e i n g i n o p e r a t i o n 5 we e k s l a t e r , t h e s e f i n d i n g s u n d e r l i n e t h e i m p o r -
t a n c e o f a n a d e q u a t e f o l l o w- u p s t r a t e g y t o g u a r a n t e e a h i g h p e r c e n t a g e o f we l l m a i n t a i n e d
ARTI Compact biogas systems.

The final evaluation regarding the suitability of the CBS (technical, economical, environ-
m e n t a l , s o c i o - c u l t u r a l a n d p o l i t i c a l , a s we l l a s s a f e t y a s p e c t s ) wa s a s s e s s e d a s g o o d .
H o we v e r , a s t h e h i g h i n v e s t m e n t c o s t p r e s e n t s t h e h i g h e s t b a r r i e r f o r wi d e d i s s e m i n a t i o n
of the ARTI technology in Tanzania, economical adjustments should be taken into consid-
eration. This study includes various recommendations regarding technical adaptations,
o p e r a t i o n a l e n h a n c e m e n t s , r e b a t e s ys t e m a n d a f t e r - s a l e s s e r v i c e s .
Zusammenfassung

A n h a l t e n d e V e r s t ä d t e r u n g u n d s t a r k e s B e v ö l k e r u n g s wa c h s t u m v e r s t ä r k e n we l t we i t d i e
D r i n g l i c h k e i t f ü r e i n e a n g e m e s s e n e A b f a l l wi r t s c h a f t ( S o l i d W a s t e M a n a g e m e n t ) . U m d i e
R i s i k e n f ü r U m we l t u n d G e s u n d h e i t z u m i n i m i e r e n , we r d e n s p e z i e l l f ü r u r b a n e Z e n t r e n i n
E n t wi c k l u n g s l ä n d e r n z u n e h m e n d a l t e r n a t i v e L ö s u n g e n z u r s i n n v o l l e n A b f a l l b e h a n d l u n g
gesucht.

D i e v o r l i e g e n d e A r b e i t b e u r t e i l t d i e E i g n u n g d e s A R T I C o m p a c t B i o g a s S ys t e m s a l s d e z e n -
t r a l e l o w- t e c h M ö g l i c h k e i t , d e n o r g a n i s c h e n A n t e i l v o n H a u s h a l t s a b f ä l l e n i n D a r e s S a l a a m
(DSM), Tansania zu behandeln. Einer Literaturstudie über Energieverbrauch und Abfall-
wi r t s c h a f t i n D S M f o l g t e I n s t a l l a t i o n , B e t r i e b u n d Ü b e r wa c h u n g e i n e r A R T I F o r s c h u n g s a n -
l a g e . Z i e l d a b e i wa r , p r ä z i s e D a t e n z u r E f f i z i e n z d e r A n l a g e z u g e n e r i e r e n , A u s s a g e n z u r
Menge und Zusammensetzung des produzierten Biogases zu ermöglichen und die Qualität
des Effluents zu bestimmen. Der ungerührte, floating-dome Fermenter (Kapazität 850l, pH:
6 . 5 , T e m p : 3 0 ° C ) wu r d e m i t z we i S u b s t r a t e n g e f ü t t e r t : S p e i s e - A b f ä l l e ( S A : 2 4 % T R , 9 1 %
oTR) und Markt-Abfälle (MA: 10% TR, 88% oTR). Die täglich zugeführte Abfallmenge von
2kg (verdünnt mit ca.18l Wasser) repräsentiert eine realistische Abfallmenge pro 5-
k ö p f i g e n H a u s h a l t . D i e R e d u k t i o n d e s A b f a l l v o l u m e n s wa r b e t r ä c h t l i c h ( S A : T R - A b n a h m e
um 84.9%, MA: TR-Abnahme um 72.8%), ebenso verdeutlicht die Reduktion des organi-
schen Anteils (SA: oTR-Abnahme um 92.2%, CSB-Abnahme um 83.1%; MA: oTR-Abnahme
um 85.3%, CSB-Abnahme um 84.2%) die Wirksamkeit der Anlage. Aufgrund der anaeroben
Vergärung stieg die Ammoniumkonzentration (NH4-N) um 134.5% auf 74.1mg/l (SA), resp.
u m 2 0 6 . 5 % a u f 2 7 . 9 % ( M A ) , wa s e i n e v e r b e s s e r t e D ü n g e q u a l i t ä t a u f g r u n d r a s c h v e r f ü g b a -
rer Pflanzennährstoffe darstellt. Durchschnittlich resultierte aus der Vergärung der Speise-
Abfälle 640l (574Nl) Biogas/kg oTR, bei einem Methangehalt von 56.8%. Die Vergärung der
Markt-Abfälle ergab eine Menge von 700l (628Nl) Biogas/kg oTR mit einem CH4-Anteil von
6 6 . 4 % . D i e A u f e n t h a l t s z e i t b e t r u g b e i b e i d e n S u b s t r a t e n 4 2 . 5 T a g e . E i n g l e i c h we r t i g e s
G e m i s c h d i e s e r z we i S u b s t r a t e e r g a b e i n e u n g e f ä h r e G a s p r o d u k t i o n v o n 2 0 0 l i n D S M , wa s
e i n e r K o c h d a u e r v o n 4 5 m i n a u f e i n e m G a s h e r d e n t s p r i c h t . D i e s wi e d e r u m e n t s p r i c h t 1 / 3
der durchschnittlichen Kochzeit eines tansanischen Haushalts pro Tag. Eine 5-köpfige Fa-
m i l i e k a n n d a h e r 2 8 k g H o l z k o h l e p r o M o n a t e i n s p a r e n ( 2 3 ' 0 0 0 T Z S ) , wa s e i n e A m o r t i s a t i -
onszeit der ARTI-Anlage (850'000 TZS = ca.550 €) von 3 Jahren ergibt.
Die Besichtigung von 12 installierten ARTI-Haushaltsanlagen in DSM deckte gravierende
U n t e r h a l t s m ä n g e l a u f : N u r 4 d e r A n l a g e n wa r e n i n B e t r i e b . D i e n a c h f o l g e n d e n A n s t r e n g u n -
gen von ARTI-TZ verfehlten die angestrebte Wirkung nicht: 5 Wochen später stieg der An-
teil der funktionierenden Anlagen auf 10. Trotzdem verdeutlichen diese Resultate, dass
eine angemessene Nachverkaufsunterstützung unbedingt erforderlich ist.

Die Schlussbeurteilung der ARTI-CBS Eignung beinhaltet technische, ökonomische, ökolo-

g i s c h e , s o z i o k u l t u r e l l e u n d p o l i t i s c h e K r i t e r i e n s o wi e S i c h e r h e i t s a s p e k t e u n d e r g a b e i n
positives Resultat. Nichtsdestotrotz sollten ökonomische Anpassungen ins Auge gefasst
we r d e n , d a d i e h o h e n I n v e s t i t i o n s k o s t e n d a s H a u p t h i n d e r n i s f ü r e i n e we i t e V e r b r e i t u n g d e r
Technologie darstellen. Die Arbeit listet abschliessend Empfehlungen auf bezüglich tech-
n i s c h e n A n p a s s u n g e n , U n t e r h a l t s b e t r i e b , R a b a t t s ys t e m u n d K u n d e n d i e n s t .
Table of Contents

Abstract ........................................................................................................... 1

Zusammenfassung............................................................................................. 1

Abbreviations ................................................................................................... 1

1 Introduction .............................................................................................. 1

1.1 General Background .................................................................................... 1

1.2 Objectives .................................................................................................. 2

1.3 Methodologies ............................................................................................ 2

2 Background Information ............................................................................. 3

2.1 Tanzania .................................................................................................... 3

2.1.1 General facts and figures ................................................................... 3
2.1.2 Dar es Salaam .................................................................................. 3
2.1.3 Solid waste management in Dar es Salaam ............................................ 4
2.1.4 Energy management........................................................................... 7
2.1.5 Deforestation and Health .................................................................... 8
2.1.6 Fertilizer demand .............................................................................. 9
2.1.7 History of domestic biogas dissemination in Tanzania ............................. 9
2.1.8 Policy and plans in regard to biogas................................................... 10

2.2 Anaerobic Digestion .................................................................................. 11

2.2.1 Biological process ........................................................................... 11
2.2.2 Substrates/Feedstock ....................................................................... 12
2.2.3 Biogas ........................................................................................... 13
2.2.4 Effluent .......................................................................................... 15

2.3 ARTI Compact Biogas System ..................................................................... 16

2.3.1 ARTI-TZ ......................................................................................... 16
2.3.2 Design and Technology of ARTI Compact Biogas System ....................... 16
2.3.3 Dissemination and costs ................................................................... 17

3 Materials and Methods ............................................................................. 19

3.1 Research-plant at ARDHI University ............................................................. 19

3.1.1 Installation ..................................................................................... 19
3.1.2 Feedstock and feeding plan .............................................................. 20
3.1.3 Monitoring of digester activity ........................................................... 24
3.1.4 Gas measurements .......................................................................... 24
3.1.5 Effluent .......................................................................................... 26
3.1.6 Calculations ................................................................................... 27

3.2 Inspection of ARTI plants in DSM ................................................................ 30

3.2.1 Analysed parameters........................................................................ 30
3.2.2 Interviews ...................................................................................... 30
4 Results ................................................................................................... 31

4.1 Research-plant at ARDHI University ............................................................. 31

4.1.1 Influence of enhanced feeding procedure on process stability ................ 31
4.1.2 Stratification inside digester in terms of TS, VS and COD ..................... 33
4.1.3 Intense monitoring of VFA, alkalinity and A/TIC ................................... 35
4.1.4 Operational parameters .................................................................... 37
4.1.5 Reduction of waste volume and organic load ....................................... 39
4.1.6 Gas resulting from AD of food waste and market waste ......................... 41
4.1.7 Effluent quality ............................................................................... 44

4.2 Inspection of installed ARTI-plants in DSM ................................................... 46

4.2.1 Result overview of first inspection tour ............................................... 46
4.2.2 Problems and consequences ............................................................. 48
4.2.3 Results of second inspection tour ...................................................... 49

5 Discussion .............................................................................................. 51

5.1 Assessment .............................................................................................. 51

5.1.1 Technical aspects ............................................................................ 51
5.1.2 Economical aspects ......................................................................... 52
5.1.3 Environmental aspects ..................................................................... 52
5.1.4 Socio-cultural and political aspects .................................................... 53
5.1.5 Safety issues .................................................................................. 53
5.1.6 Overview of suitability assessment..................................................... 54

5.2 Recommendations ..................................................................................... 55

5.2.1 General .......................................................................................... 55
5.2.2 Technical ....................................................................................... 55
5.2.3 Economical ..................................................................................... 57
5.2.4 Environmental ................................................................................. 57
5.2.5 Socio-cultural and political ............................................................... 57
5.2.6 Safety ............................................................................................ 58
5.2.7 Simple monitoring of ARTI-plants ....................................................... 58

References ..................................................................................................... 59

Table of Figures .............................................................................................. 62

List of Tables ................................................................................................. 63

Appendix .......................................................................................................... I

A Laboratory ................................................................................................ I

B ARTI ......................................................................................................... I

C Various ..................................................................................................... I

D Presentation .............................................................................................. I
Abbreviations
AD Anaerobic Digestion
ARTI Appropriate Rural Technology Institute
A/TIC Acids / Total Inorganic Carbon
CBS Compact Biogas System
COD Chemical Oxygen Demand
DCC Dar es Salaam City Council
DSM Dar es Salaam
FW Food Waste
GPR Gas Production Rate
GTZ Deutsche Gesellschaft für Technische Zusammenarbeit GmbH
HH Household
HRT Hydraulic Retention Time
JICA Japan International Cooperation Agency
LPG Liquid Petroleum Gas
MSW Municipal Solid Waste
MW Market Waste
OLR Organic Loading Rate
SGP Specific Gas Production
SRT Solid Retention Time
SW Solid Waste
SWM Solid Waste Management
TS Total Solids
TZ Tanzania
VS Volatile Solids
WW Wet Weight
ZHAW Bachelor Thesis 09 Christian Lohri UI05

1 Introduction
On a worldwide scale, rapid population growth and urbanisation have led to an enor-
mous increase of solid waste generation per unit area. Particularly urban- and peri-
urban areas of low- and middle-income-countries are confronted with great challenges
concerning appropriate Solid Waste Management (SWM) in order to minimise the risk to
human health and avoid environmental degradation. However, most municipalities
struggle to provide sufficient and adequate SWM services (Kassim & Ali, 2006). Con-
sidering the fact that the largest fraction of waste in developing countries is of organic
nature and therefore amendable to anaerobic digestion, it makes environmental and
economic sense to survey this option (Mbuligwe & Kassenga, 2004).

1.1 General Background

Urban solid waste management in general and inadequate disposal in particular are
considered to be of the most immediate and serious environmental problems in urban
areas of developing countries (Zurbrügg, 2002). The present SWM systems are pre-
dominantly based on an “end of pipe” solution, i.e. collection-transport-dumping of
waste with limited recycling of inorganic waste, mainly
done by the informal sector. In most cities of low- and
middle-income countries the physical composition of
solid waste consists mostly of organic, hence biode-
gradable matter (Troschinetz et al., 2008), yet less than
50% of the total waste generated is collected and dis-
posed of in sanitary manner (Parrot et al., 2008). Inade-
quate management like uncontrolled dumping bears sev-
eral adverse consequences: It not only leads to an ugli-
fication of the living area, but also to a high risk of pol-
luting surface and groundwater through leachate and
furthermore promotes the breeding of flies, mosquitoes,
rats and other disease vectors. In addition, it emits un-
pleasant odours and methane, a major greenhouse gas
Figure 1: Uncontrolled dumping
in Dar es Salaam, Tanzania contributing to global warming (Yhdego, 1995).

Besides composting and direct animal feeding, anaerobic digestion (AD) is a promising
method to treat the particular fraction of organic solid waste. While AD for the treat-
ment of animal dung is fairly common in rural areas of developing countries, informa-
tion on technical and operational feasibilities concerning the treatment of organic solid
waste is limited. Regarding the enormous waste problems in the urban areas of these
countries, the question arises whether AD could be an appropriate and sustainable
method to treat organic household waste as well as wastes from markets and restau-
rants.

The ARTI Compact Biogas Plant, developed in India, is a small-scale system based on
anaerobic technology that combines organic solid waste treatment with the production
of cooking energy and liquid fertilizer. Already widespread in India on household (HH)
level, it is now being promoted in Tanzania.

1
ZHAW Bachelor Thesis 09 Christian Lohri UI05

1.2 Objectives
The overall objective of this study is to assess the suitability of the ARTI Compact bio-
gas plant as a treatment option for organic solid waste on household level in urban ar-
eas of developing countries.

The installation, operation and monitoring of an ARTI biogas plant at ARDHI University
in Dar es Salaam (DSM), Tanzania, shall give precise and reliable data regarding the
daily gas production and its composition, the reduction of the organic load and the
quality of the effluent for different feedstock. In addition, the hazard imposed on the
health of the operator through toxic substances produced by the anaerobic process is
assessed as well as the risk of attracting disease vectors by the system. All these in-
formation are evaluated on the socio-cultural, economical and traditional waste man-
agement background of urban households in DSM. Interviews with users of existing
ARTI biogas plants in Tanzania shall furthermore provide information concerning the
convenience of operation and the economic efficiency of the system. The household
survey aims to compare the results from the research plant run under controlled condi-
tions with the performance of plants operating under real conditions in private HHs.

Combining these results with literary findings about the SWM on household level in
DSM will indicate whether the ARTI-technology can be regarded as appropriate to treat
organic household waste or how to devise strategies for enhancing the performance of
ARTI-biogas plants.

1.3 Methodologies
y Literature review
An extensive literature review on solid waste management and energy utilization in
DSM, specifically for cooking purpose, is conducted. Moreover research reports and
general literature on anaerobic digestion of organic materials are reviewed.

y Installation, start-up and operation of ARTI Biogas Plant at ARDHI University DSM
The installation and start-up of a research-plant is followed by continuous measuring
on site of the gas production and gas composition resulting from canteen food waste
and vegetable/fruit market waste. Temperature, pH and Redox-Potential of the effluent
are additionally measured on a daily basis.

y Laboratory analysis
Laboratory analyses are completed for determining various parameters of the feedstock
(TS, VS, CODtotal, CODdissolved, Ntotal, NH4-N, Ptotal) and effluent (TS, VS, CODtotal,
CODdissolved, Ntotal, NH4-N, Ptotal, PO4, Pb, Cu, Cd) in the Laboratory of Environmental
Engineering at ARDHI University DSM.

y Inspection of ARTI biogas plants in Dar es Salaam and interviews with operators
Visits to 17 ARTI biogas units in DSM are conducted to assess the performance of
these plants. Additionally, interviews with the owner and operator about their experi-
ences are accomplished.

2
ZHAW Bachelor Thesis 09 Christian Lohri UI05

2 Background Information

2.1 Tanzania

2.1.1 General facts and figures

Land area: 881.289 km2

Total population: 34.6 Millions (2002)
Population growth: 2.9% (2007)
Urban population: 23% (2007)
Capital: Dodoma
Average urban household size: 4.9 persons (2000)
Figure 2: Map of Tanzania [1]
Life expectancy: 52 years (2006)
Estimated per capita income: 349 US$ (2005)
Below basic needs poverty line: 35.7% (2000/01)
Below food poverty line: 18.7% (2000/01)
Unemployment rate: 12.9% (in urban higher than in rural areas)
Total adult literacy rate: 69% (2000-2005)
Electrification rate: 11% (2007)
Currency: Tanzanian Shilling (TZS)
Exchange rate (November 2008): 1’000 TZS = 0.987 CHF
= 0.649 EUR
= 0.811 US$
Main languages: Swahili, English
Climate: Tropical on the coast, semi-temperate inland
Political system: Multiparty parliamentary democracy

2.1.2 Dar es Salaam

Dar es Salaam (DSM) is the principal city and administrative, commercial and industrial
centre in Tanzania. It is located on the eastern coast of the country along the Indian
2
Ocean and in 2005 had a total area of 1393km . The whole city is politically and admin-
istratively divided into three Municipalities (Temeke, Ilala and Kinondoni). The city area
is growing at a rate of 7% per year and it is estimated that roughly 70% of the popula-
tion in the city live in informal or unplanned settlements (Kaseva & Mbuligwe, 2005).
The estimated population of DSM is between 2.5 and 3.4 million, the population density
2
is 1793 persons per km and the average population growth rate is 4.3% per annum
(Kassim & Ali, 2006).

DSM has a warm, moist monsoon climate: relatively cold and dry from April to October,
hot and humid from November to March. The average annual precipitation in Dar is over
1000mm due to the long rains between March and May, when rainfall is about 150-
300mm. The temperature average is 24°C, whereas the maximum temperature is 35°C
in the afternoon and the minimum 13°C at dawn.

3
ZHAW Bachelor Thesis 09 Christian Lohri UI05

2.1.3 Solid waste management in Dar es Salaam

The rapid increase of population in Dar es Salaam, coupled with the increasing growth
of commerce and trades has increased the solid waste generation at a rapid rate. SWM
in DSM is administered by the three municipalities under the Dar es Salaam City Coun-
cil (DCC). In 1994 the DCC acknowledged the difficulty of providing adequate SWM and
hence decided to involve the private sector as a partner in SW collection services:
While the private sector (community-based organisations and local companies) acts as
a collector, the City Council remains as the principal and overall responsible for provi-
sion of the service (Kassim & Ali, 2006). The estimated annual budget devoted to MSW
in DSM is 10million US$ (Parrot et al., 2008).

SWM in DSM at household level is generally considered to be housekeeping and thus

usually falls to women and house servants as practised in many African countries (Kas-
sim & Ali, 2006).

Solid waste generation

The total quantity of solid waste generated in Dar es Salaam City was found to be be-
tween 1800t/d (Mbuligwe & Kassenga, 2004) and 2500t/d (Kaseva & Mbuligwe, 2005).
However, recent findings of the DCC estimate the total waste to be as high as 3500t/d
(Oral information Mkumba, 26.11.2008). Table 1 reveals that households alone gener-
ate about 56% of this total waste.

Table 1: Solid waste generation in DSM city

(Kaseva & Mbuligwe, 2005)
Total waste
Waste source % of total
generated [t/d]
Households 1360 56.1
Commercial
80 3.3
establishments
Institutions 185 7.6
Markets 375 15.5
Others
425 17.5
including industrial
Total 2425 100

The study on waste generation in DSM by Kaseva & Mbuligwe (2005) concluded aver-
age domestic waste generation in unplanned, low-income areas of 0.34kg/capita/day
(Buguruni), 0.39kg/c/d (Kariakoo) and in planned areas 0.43kg/c/d (Upanga), 0.44 (City
centre). These results are comparable to an older study conducted by Ame in 1993,
which established waste generation rates for high income group of households in DSM
city to be 0.45kg/c/d, for medium-income group 0.38kg/c/d and for low-income group to
be 0.34kg/c/d. In 1996, JICA reported a fairly high domestic waste generation rate of
0.7kg/c/d in DSM (Kaseva & Mbuligwe, 2005). All the information found are quite in line
with World Bank Standard for developing countries which is between 0.3 and 0.6kg/c/d.

4
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Solid Waste composition on household level

The composition of household waste in DSM is similar to that of most developing coun-
tries, with organic waste (in Table 2 listed as kitchen waste, grass and wood) around
67% being the biggest component. Other authors estimated the total organic fraction of
MSW in DSM city to be 71% (Mbuligwe & Kassenga, 1998) and 78% (Chaggu et al.,
1998).

Table 2: Composition of household waste in DSM

(Mbuligwe & Kassenga, 2004; Source JICA, 1997)

Low- Middle- High- Weighted

Waste category income income income Average
[%] [%] [%] [%]
Kitchen waste 39.8 41.9 44.5 42.0
Grass and wood 15.0 30.9 19.8 25.3
Paper 3.3 3.0 3.1 3.1
Plastic 1.9 1.9 3.4 2.2
Glass 1.3 2.4 8.8 3.5
Metal 1.8 2.1 1.8 2.0
Textile 0.6 0.5 3.8 1.2
Leather and rubber 1.2 1.0 0.3 0.9
Ceramic and stone 0.4 0.3 0.5 0.4
Other waste types 34.7 16.0 14.0 19.4
Total 100 100 100 100
Moisture content 30.6 31.1 31.5 31.1

It is worth noting that differences in household waste composition among different in-
come groups with respect to kitchen waste are negligibly small. Institutional solid waste
constitutes 56-64% organic waste (Mbuligwe, 2002).

Existing Solid Waste practices

As shown in Table 3, the majority of waste generated by households in DSM city is ei-
ther self-disposed close to the site where it was produced or illegally dumped in a pro-
hibited area such as at the roadside, in open spaces or drains (Figure 3).

Figure 3: Illegally dumped waste in Dar es Salaam

5
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Table 3: Quantities of waste disposed of [t/d] by different methods in TZ, 1997

(Mbuligwe & Kassenga, 2004)

Illegal
Self- Discharge/
Type of waste dumping Recycling Total
disposal Collection
Household 651.3 172.9 478.1 114 1416.3
Commercial
Restaurant 0 12.7 0 1 13.7
Guesthouse/hotel 0 1.6 0 0.2 1.8
Other 0 11.8 0 0 11.8

Institutional 2.1 8.6 0 0 10.7

Market 0 33.9 0 0 33.9
Street weeping 0 1.3 0 0 1.3
Informal sector 0 56.5 226.2 0 282.7
Total 653.4 299.3 704.3 115.2 1772.2

As the current disposal facility is not environmentally suitable and its capacity ex-
hausted, the most problematic functional element of solid waste management in DSM
has been identified as disposal (Mbuligwe & Kassenga, 2004).

Collection service
Managed in a top-down approach, the design and operation of SW collection in DSM
shows activity at City Council and municipal level. Poor flow of information to house-
holds (HH) and low level of involvement have resulted in low participation in the
scheme. In addition, due to the mismatch between the amount of waste generated and
the capacity of the service providers, approximately 52% of SW generated in DSM still
remains uncollected. Only 44 of 73 wards are currently covered to some extent by col-
lection services (Kassim & Ali, 2006). Generally, the high-income and planned areas
receive better services than others because of the good waste storage, payment habit
and accessibility for the collection trucks (Kaseva & Mbuligwe, 2005).

Figure 4: Waste pushcart, waste truck and waste collection point in Dar es Salaam

The charge for SW collection in DSM was set by DSM City Council (DCC) according to
income level. However, every municipality has its own by-laws and charges for solid
waste collection. The monthly payment for the SW collection range between 200 and
3’000 TZS according to income-areas. Yet many of the contractors manage to collect
less than 50% of the service charge (Kassim & Ali, 2006). A great number of HHs is

6
ZHAW Bachelor Thesis 09 Christian Lohri UI05

reluctant to pay for a variety of reasons including: priority set, low income level and
especially lack of awareness. As many HHs are not aware about their service providers
and privatisation as a whole, they do not know if the service provider is from the pri-
vate sector, public sector or an agent who is already paid by the City Council.

Nevertheless, it is generally acknowledged that privatisation of SW collection in DSM

has brought several advantages like creating employment among DSM residents and by
increasing the collection from 10% in 1994 to 40% of the total waste generated in 2001,
also significantly improving the cleanliness of the city (Kaseva & Mbuligwe, 2005).

2.1.4 Energy management

Table 4: Primary energy consumption in TZ 1999 by source (Schmitz, 2007):

Energy consumption by source [TJ] [%]
Total energy consumption 1999 629’402 100.0
Total fossil fuels 31’903 5.1
Coal and coal products 126 0.0
Crude Oil and natural gas liquids 26’293 4.2
Others 5485 0.9
Hydroelectric 7’829 1.2
Total renewables, excluding hydroelectric 589’460 93.7
Primary solid biomass (incl. fuel wood) 589’446 93.7
Biogas and liquid biomass 14 0.0

The Census 2005 presented biomass still to be 92.1% of total primary energy supply.
Domestic energy accounts to 78.8% of total nation-wide energy consumption. Nation-
wide electrification is about 10% (urban 39% and rural 2%). The electricity consumption
per capita in 2002 was 84kWh (Schmitz, 2007).

Table 5: Household (HH) energy sources and utilization (Mwakaje, 2007)

Type of energy % of total HH

Purpose of energy used
used in HH sector energy consumption
Biomass fuels
97.7 Cooking, heating, lighting
(charcoal&firewood)
Petroleum fuels 2.0 Kerosene for lighting and cooking
Lighting, powering the radio and in a few cases
Electricity 0.3
cooking
Solar, biogas, wind 0 Lighting, cooking

In 2007 the major source of energy for cooking nationwide was wood (60%) followed by
charcoal (35%). In rural areas, firewood with 87% is the major source of cooking-
energy. Urban areas predominantly use wood for cooking (57%) with the exception of
DSM, where mainly charcoal is used (Human Development Report 2007).

The majority of homes in Dar es Salaam use more than one kind of cooking fuel. As
mentioned above and presented in Table 6, charcoal is the most preferred cooking fuel
in DSM, as it is considered to be a modern source, unlike firewood, agricultural resi-
dues and dung (Sanga, 2003). 470’000t of charcoal are consumed in DSM per year
(Costech, 2006).

7
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Table 6: Preferences in the use of cooking fuels in DSM in 2001 (Sanga, 2003)

Kind of fuel First choice Second choice Third choice

[% of homes] [% of homes] [% of homes]
Charcoal 69 25 3
Kerosene 25 53 5
Electrical power 4 6 17
LPG 1 2 0
Others 1 14 75

The production, distribution and sale of charcoal constitute one of the largest indus-
tries in the informal sector, offering employment and income for people in rural and
urban areas. The kilns used in Tanzania have a yield of roughly 1kg of charcoal for
every 6kg of wood used. Due to deforestation in the outlying areas of the city, the dis-
tance from the points of charcoal production to the city increased from 50km in the
1970s, to 200km in the nineties (Sanga, 2003), to 2000km (Costech, 2006). Daily con-
sumption for a family of five persons is around 2.8kg. Thus the per capita consumption
is approximately 0.6kg/d. Considering the average efficiency of the charcoal stoves at
20%, and the lower heating value of charcoal being equal to 30.8 MJ/kg, consumption
per capita of useful energy is approximately 1.35 GJ/year (Sanga, 2003).

Currently, the price in Dar es Salaam for one bag of charcoal (60-70kg, of which only
40-50kg are usable) ranges between TZS 30’000-35’000. For an average family of 5
members, one bag lasts for approximately 2 weeks (Oral information Ndimbo 10.9.08,
Mkumba 26.11.08), which adds up to an annual spending between 720’000 and 840’000
TZS for charcoal only.

2.1.5 Deforestation and Health

2
In 2005 the forest area in Tanzania covered 352’600km which is 39.9% of the total
land. Dependence on fuelwood and charcoal as listed in 2.1.4 lead to an estimated an-
3
nual per capita firewood consumption of 1m per year, resulting in 7kg per rural house-
2
hold per day (Schmitz, 2007). Hence between 1990 and 2005, 62’800km of forest area
have been destroyed, resulting in a nationwide average annual change of forest area of
minus 1% (Human Development Report 2007). The greatest part of wood harvested is
used for the production of charcoal (75%), followed by logging (12%), agricultural ac-
tivities (7%) and others (6%). Only one third of all the woods consumed annually is re-
covered through reforestation, and thus the use of wood for charcoal production is not
sustainable (Sanga, 2003). As a consequence of deforestation, erosion is leading to
diminished fertility of land, which is assumed to result in a 0.5 to 1.5% reduction annu-
ally in the gross national product (Schmitz, 2007).

Smoke emissions in the households stemming from the use of firewood, dung and straw
as energy for cooking on open fireplaces are showing undesirable side-effects on the
health of women and children. Indoor air pollution is taken far more serious than in-
creasing air pollution in cities (Schmitz, 2007). In Dar es Salaam, acute respiratory
infections occupy second place in all doctor’s appointments at 13%, right behind ma-
laria which leads with 50% (Sanga, 2003).

8
ZHAW Bachelor Thesis 09 Christian Lohri UI05

2.1.6 Fertilizer demand

In 1999 Tanzania consumed 21’000t of industrial fertilizer, resulting in an intensity of

5kg per hectare cropland (Schmitz, 2007). Other sources describe the fertilizer demand
for Tanzania as 1.79kg/ha. Between 1970 and 2002, fertilizer demand decreased on
average and annually by 4.18% per year even with regional subsidies. Currently, the
use of artificial fertilizer by smallholders in TZ can be described as “virtual absence”.
The number of households in Tanzania using organic fertilizer in 2002 was 1’270’272,
26% of all crop growing households. The total area of organic fertilizer application was
2’334’188ha, 33% of total planted area. Of all organic fertilizer, 88% is farm yard ma-
nure and 12% compost (Schmitz, 2007). These facts imply that even though organic
fertilizer is used by the producer, they have not been commercialised yet.

2.1.7 History of domestic biogas dissemination in Tanzania

Biogas technology was introduced in Tanzania in 1975 by the Small Industries Devel-
opment Organisation (SIDO). These early biogas plants adopted the floating drum tech-
nology from India and were mainly introduced in primary and secondary schools, rural
health centres and other institutions.

In 1982, the Ministry of Industries and Trade through its parastatal organisation Centre
for Agriculture Mechanisation and Rural Technology (CAMARTEC) initiated a biogas
development programme in the Arusha area which was supported by the German
Agency for Technical Cooperation (GTZ). Under this programme, the Chinese dome
biogas plant was adapted to local conditions and standardised into biogas units of 16,
3
30 and 50m . Technical Cooperation between Tanzania and Germany led in 1983 to the
introduction of the Biogas Extension Service (BES) in which CAMARTEC and GTZ were
in charge of implementing the project in TZ. Dissemination strategy and project struc-
tures underwent decisive changes mainly around 1990. These were chiefly a result of
financial and personnel withdrawal of the GTZ from the BES and the subsequent exten-
sive transfer of the project to the counterpart organisation (Schmitz, 2007). Between
1983 and 2005, a total of 707 biogas units have been constructed by CAMARTEC
(Costech, 2003). Other institutions involved in biogas development/popularisation in-
clude MIGESADO, Tanzanian Traditional Energy Development and Environmental Or-
ganisation (TaTEDO), Evangelical Lutheran Church in Tanzania (ELCT), Ministry of En-
ergy and Minerals (MEM) and Tanzania Commission for Science and Technology
(COSTECH).

Figure 5: CAMARTEC fixed-dome plant [2]

9
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Over the period of 1997 to 2007, some 2900 biogas installations have been reported,
120 of the floating drum design and 430 plastic bag plants. Out of these, some 1900
are expected to still be in operation (SNV, 2008).

As a consequence of the “Biogas for better life, An African Initiative”

(www.biogasafrica.org), which has been initiated in May 2006 in Nairobi, Kenya, vari-
ous Tanzanian biogas stakeholders launched a National Initiative in 2007 under the
supervision of the Netherlands organization SNV. The objective is to further develop
and disseminate domestic biogas in rural areas with the ultimate goal to establish a
sustainable and commercial biogas sector in Tanzania. The Initiative aims to install
12’000 new biogas plants (modified CAMARTEC fix-dome design) during the next five
years.

2.1.8 Policy and plans in regard to biogas

Tanzanian energy policies do not specifically promote biogas, as only “alternative” en-
ergy sources are broadly promoted in policy. Recently, though, the Government estab-
lished the Rural Energy Agency (REA) under the Ministry of Energy and Minerals
(MEM). It is the objective of REA to stimulate access to energy in rural areas, to which
extent a Rural Energy Fund was established.

The National Strategy for Growth and Reduction of Poverty states that by 2010 at least
10% of the population shall be using alternative power to wood fuels for cooking (SNV,
2008).

10
ZHAW Bachelor Thesis 09 Christian Lohri UI05

2.2 Anaerobic Digestion

Anaerobic Digestion (AD), also referred to as biomethanization, is a natural process
that takes place in the absence of oxygen. It involves the biochemical decomposition of
complex organic material by various bacterial processes with the release of an energy
rich biogas and the production of a nutritious effluent. Digesters or reactors are physi-
cal structures that facilitate anaerobic digestion by providing an anaerobic environment
for the organisms responsible for digestion.

2.2.1 Biological process

The biological conversion of organic material under anaerobic conditions can be de-
scribed by the following four stages (see Figure 6):

1 Hydrolysis
The first step involves the extracellular enzyme-mediated transformation of higher-
molecular-mass organic polymers and lipids into basic structural building blocks such
as fatty acids, monosaccharides, amino acids, and related compounds which are suit-
able for use as a source of energy and cell tissue.

2 Acidification
The fermentative bacteria degrade the soluble organic monomers of sugars and amino
acids, producing volatile fatty acids (propionic, butyric and valeric acids), acetate, H2
and CO2. Ammonia is also produced by the degradation of amino acids.

3 Acidogenesis
Both long chain fatty acids and volatile fatty acids (VFA) are degraded generating ace-
tate, carbon dioxide and hydrogen.

4 Methanogenesis
The forth and last step involves the bacterial conversion of hydrogen and acetic acid
formed by the acid formers to methane gas and carbon dioxide. The bacteria responsi-
ble for this conversion are strict anaerobes, called methanogenic. Due to their very
slow growth rates, their metabolism is usually considered rate-limiting in the anaerobic
treatment of organic waste (Mata-Alvarez, 2003).

11
ZHAW Bachelor Thesis 09 Christian Lohri UI05

In one-stage systems, the

above described reac-
tions occur simultane-
ously in a single reactor,
whereas in two- or multi-
stage systems, the reac-
tions take place sequen-
tially in at least two reac-
tors with the second
stage consisting of the
conversion to methane.

Another differentiation in
systems is related to its
dry matter content:
Wet-systems: TS<15%
Dry systems: TS=20-50%

Figure 6: Scheme of the biodegradation steps of complex matter

(Mata-Alvarez, 2003)

A major danger for overall anaerobic conversions is presented when microorganism

populations are not balanced. This can be a result of an overload, which is defined as
either an excess of biodegradable organic matter for the active population capable of
digesting it, or any circumstances that produce a decrease in active microorganism
concentration (abrupt change of temperature, accumulation of toxic substances, flow
rate increase etc.). These disturbances mainly affect methanogenic bacteria, whereas
the much more tolerant acidogenic bacteria continue to work, producing more acids,
which in return inhibit the activity of methane-formers. This imbalance can eventually
result in a digester failure (Mata-Alvarez, 2003).

2.2.2 Substrates/Feedstock

In general, all types of biomass can be used as substrates as long as they contain car-
bohydrates, proteins, fats, cellulose and hemicellulose as main components. Lignin, the
main constituent of wood can not be degraded under anaerobic conditions (Mata-
Alvarez, 2003). The performance of digesters operating on solid wastes is dependent
on the particle size of the influent. Therefore, size reduction of the particles and the
resulting increase of the available specific surface, represents an option for increasing
degradation yields and accelerating the digestion process (Mata-Alvarez, 2003).

12
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Table 7: The maximum gas yield per kg VS of different organic waste substrates (+cow manure)
found in literature.

Substrate TS [%] VS [%] Retention Biogas yield Source

3
time [days] [m /kg VS]
Spent fruits 25-45 90-95 0.4-0.7 Deublein
Vegetable wastes 5-20 76-90 8-20 0.4 “
Market wastes 8-20 75-90 30 0.4-0.6 “
Leftovers (Canteen) 9-37 75-98 0.4-1.0 “
Bio waste (HH) 40-75 30-70 27 0.3-1.0 “
Overstored Food 14-18 81-97 10-40 0.2-0.5 “
Fruit wastes 0.2-0.7 Gunaseelan
Banana peels 86-94 0.2 “
Citrus waste 89-97 0.4-0.5 “
Vegetable wastes 0.2-0.3 “
Mixed food waste (Korea) 26 90-95 0.3-0.5 Lee et al.
Bio waste 60-75 50-70 0.2-0.6 Eder&Schulz
Kitchen waste 9-37 50-70 0.2-0.5 “
Market waste 28-45 50-80 0.45 “

Liquid manure from cattle 6-11 68-85 0.35-0.55 Schilling&

Tijmensen
Excreta from cattle (fresh) 25-30 80 0.6-0.8 Deublein

2.2.3 Biogas

Biogas consists primarily of utilizable methane (CH4) and inert carbon dioxide (CO2),
which are both colourless and odourless. It also contains several impurities as listed in
Table 8. Biogas with methane content higher than 45% is combustible and burns with a
blue flame. The explosive limits of CH4 in air are 6 to 12% by volume (Deublein, 2008).

Table 8: Typical biogas components from organic fraction of MSW (Mata-Alvarez, 2003)

Components Concentration (by volume)

Methane (CH4) 55-60%
Carbon dioxide (CO2) 35-40%
Water 2% (20°C) – 7% (40°C)
Hydrogen Sulphide (H2S) 20-20’000ppm (2%)
Ammonia (NH3) 0-0.05%
Nitrogen (N2) 0-2%
Oxygen (O2) 0-2%
Hydrogen (H2) 0-1%

Table 9: General features of biogas (Deublein, 2008)

3
Energy content 6.0 – 6.5 kWh/m
3
Fuel equivalent 0.60 – 0.65L oil/m biogas
Explosion limits 6 -12% biogas in air
Ignition temperature 650 -750°C
3 3
Theoretical air demand 5.7m air/m burning gas
Critical pressure 75-89bar
Critical temperature -82.5°C
3
Normal density 1.2kg/m
Molar mass 16.043g/mol

Smell Bad eggs (the smell of desulfurized biogas

is hardly noticeable)

13
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Use of biogas
Biogas can principally be used like other fuel gas. For domestic purposes it is most
suitable for cooking, heating water and lighting. The following are some approximate
quantities of gas for these different uses (Koottatep et al., year unknown):

3
- Domestic cooking: 2m per day for a family of five or six people
3
- Water heating: 3m per day for a 100-litre tank or 0.6m3 for a tub bath and 0.35m3 for
a shower bath
3
- Lighting: 0.1-150m per hour per light

The gas demand can be defined on the basis of energy consumed previously. For ex-
ample, 1kg firewood corresponds to 200l biogas, 1kg dried cow dung corresponds to
100l biogas and 1kg charcoal corresponds to 500l biogas. The gas demand can also be
defined using the daily cooking times. The gas consumption per person and meal lies
between 150 and 300l biogas. For one litre water to be cooked 30-40l biogas are re-
quired, for 1/2kg rice 120-140l and for 1/2kg vegetables 160-190l. (Kossmann et al.,
year unknown).

The critical temperature of methane is around -82.5°C; even with very high pressure it
is not possible to liquefy methane at higher temperature, which is probably the most
important bottleneck of biogas utilization. As a consequence, biogas cannot be stored
over long periods at reasonable costs, but has to be used immediately or within a few
hours (Mata-Alvarez, 2003). On a larger scale, biogas can be converted into electricity.
Thus this solution needs a gas storage capacity of about half a day to store the night
production and requires the cleaning of gas.

Hydrogen sulphide
Hydrogen sulphide (H2S) is an aggressive gas mainly responsible for corrosion. As
shown in Table 10, H2S is already hazardous to human health in small quantities.

Table 10: Effects of H2S (adopted from Eder & Schulz, 2006)
Concentration
in the air in ppm Effects
(parts per million)
1ppm=0.0001%
0.03-0.15 Threshold of perception, Odour of rotten eggs
Irritation of eyes and respiratory passages, nausea, vom-
15-75
iting, headache, unconsciousness
150-330 Paralysation of olfactory nerves
>375 Death through intoxication (after several hours)

>750 Unconsciousness and death through still stand of breath-

ing in 30-40 minutes
>1000 Rapid death through respiratory paralysis in few minutes

The hydrogen sulphide content in the biogas is subject to considerable unexplainable

variation in the course of the day (Deublein, 2007). In general, gas cleaning for com-
bustion in burners is only necessary at a H2S content of more than 0.1% (1’000ppm)
(Mata-Alvarez, 2003)

14
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Water vapour
Water vapour creates condensation in the tubing. To prevent blocking and corrosion, it
has to be condensed in water traps, especially when higher amounts of corrosive com-
ponents, such as CO2 and H2S are dissolved, making the gas more aggressive.

2.2.4 Effluent

Due to the decomposition and breakdown of its organic content, the residue of the bio-
methanization process, also called slurry, normally gets rid of smell and provides fast-
acting nutrients (mainly NH4-N) which easily enter into the soil solution, thus becoming
immediately available to the plants. Hence digested sludge can increase agricultural
yields according to its nutrients.

The pathogenic organisms present in the digesters raw influent (or inoculum) from ani-
mal faeces get eliminated during the mesophilic digester process at 35°C. Tests
showed a complete elimination after 3 months of all harmful pathogens like Enteric vi-
rus, Salmonella, Shigellas, Vibrio Cholera, Pathogenic Escheria coli, Trichuris and
Hookworms (Costech, 2006). Even more important for agricultural use is the fact that
all plant-pathogenic germs are completely destroyed during the anaerobic digestion
process (Wellinger et al., 1991).

15
ZHAW Bachelor Thesis 09 Christian Lohri UI05

2 . 3 A R T I C o m p a c t B i o g a s S ys t e m
The ARTI Compact biogas system (CBS) was developed in 2003 by Appropriate Rural
Technology Institute (ARTI) India, an NGO based in Pune (Maharashtra state, South
India) and founded in 1996 by a group of scientists, technologists and social workers.
About 2500 ARTI-plants are currently in use both in urban and rural households in Ma-
harashtra (Müller, 2007). The design and development of this technology using organic
waste rather than manure as feedstock has won the ‘Ashden Award for Sustainable En-
ergy 2006’ in the food security category [3].

2.3.1 ARTI-TZ

Appropriate Rural Technology Institute Tanzania (ARTI-TZ) is a Tanzanian registered

non-political, non-profit organization founded in 2007. Its mission is to serve Tanzania
as an instrument of development through the dissemination and application of scientific
knowledge and sustainable technologies for energy production, environmental protec-
tion, employment and income generating opportunities.
ARTI-TZ derives its technologies from ARTI-India under a technology transfer agree-
ment. It provides promotion and training of the ARTI Compact biogas plant, a charcoal
kiln for producing briquettes made from agricultural waste and the ARTI Sarai Cooker
Systems for reduced Indoor Air Pollution and charcoal consumption.

Address: Appropriate Rural Technology Institute (ARTI-TZ)

Office located near Massana Hospital
Goba Road, Mbezi Beach
Dar es Salaam, Tanzania

Contact: [email protected]
+255 (0) 786 427 949 (Nachiket W. Potnis, Executive Director)
+255 (0) 715 235 126 (Dennis Tessier, Programme Director)
+255 (0) 713 234 604 (Joy Desai, Operation Director)

JET (Joint Environmental Techniques) was founded in 2008 to manufacture, sell and
install the products of ARTI-TZ as well as provide technical support.

2.3.2 Design and Technology of ARTI Compact Biogas System

The ARTI Compact biogas system (CBS) is made from two cut-down standard high den-
sity polyethylene (HDPE) water tanks and standard plumber piping. The larger tank acts
as the digester while the smaller one is inverted and telescoped into the digester and
serves as a floating gas holder, which raises proportional to the produced gas and acts
as a store room of the biogas. The CBS is designed for treating 1-2kg (dry weight) of
kitchen waste per day [4]. The gas can directly be used for cooking on an adjustable
gas stove whereas the liquid effluent can be applied as nutrient fertilizer in the garden.
2
Space of about 2 m and 2.5 m height is needed for a CBS of 1000l.

16
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Figure 7: ARTI Compact biogas plant scheme (gasholder empty and gas-filled)

The effective volume of the digester is approximately 850l, given by the dimension of
the 1000l-water tank (inner radius: 51.5cm) and the position of the overflow-pipe (1.04
2
m above ground level). The total surface area of the digester (0.83m ) is covered by
2
roughly 0.65m (78%), in other words the gas released through 22% of the digester sur-
face is lost to the atmosphere without utilization. The usable gas volume of the 750l-
gasholder is 400l (see Figure 14Error! Reference source not found.).
The Hydraulic Retention Time (HRT) suggest ed by ARTI-TZ, which describes the ratio
3 3
of the reactor volume (0.85m ) to the flow rate of the influent substrate (0.02m /day), is
42.5 days. The seemingly rather long period of time that digester liquid spends in the
reactor is justified by the appearances of sinking and floating layers.

2.3.3 Dissemination and costs

Since ARTI-TZ started disseminating CBS in November 2006 until November 2008, 31
ARTI Compact biogas units have been installed in Tanzania and Uganda (Table 11).

Table 11: ARTI-CBS installed by 30 November 2008

Digester
Country Quantity Location Level
si ze
Tanzania 1.0m3 10 9 in DSM, 1 in Mbeya Household
1.5m3 1 Kyela Household
2.0m3 3 DSM Household

3.0m3 2 Saadani (Safari Lodge), Institutional

Mbagala (Kinasi Lodge)
2.0m3 4 2 on Mafia Island (Kinasi Lodge) Institutional
2 in DSM (army campus)
3.0m3 3 DSM (Azania Secondary School) Institutional
4.0m3 1 DSM (Bethsaida Sec. School) Institutional
1.0m3 1 DSM (ARTI-office) Demonstration
1.0m3 1 DSM (ARDHI-university) Research

Uganda 2.5m3 2 Kitende (St.Mary’s Sec. School) Institutional

5.0m3 1
1.0m3 2 Kampala (JET-office, ARTI-office) Demonstration

17
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Table 12: Costs of the ARTI-biogas plants (November 2008)

Si ze Si ze
Price [TZS] Price [EUR]
of digester [l] of Gasholder [l]
1000 750 850’000 550
1500 1000 1’100’000 715
2500 2000 1’500’000 975
4000 3000 1’850’000 1200

It is worth noting that ARTI-TZ ceases to recommend the 1000l-digester (with 750l gas-
holder) as ideal household plant and will in the near future advise customers to pur-
chase the larger model consisting of a 1500l-digester and 1000l-gasholder (oral infor-
mation Potnis, 18.10.08).

18
ZHAW Bachelor Thesis 09 Christian Lohri UI05

3 Materials and Methods

3.1 Research-plant at ARDHI University

To conduct a detailed monitoring of the ARTI system, an ARTI Compact biogas plant
was installed and operated on the campus of ARDHI University. This allowed great
flexibility in terms of substrate changes and on-site measurements of the gas produc-
tion and composition. The analyses were accomplished in the nearby Environmental
Engineering Laboratory of ARDHI University.

3.1.1 Installation

The installation of the ARTI Compact biogas system (di-

gester size: 1000l, gasholder: 750l) was carried out on
30 July 2008 by an ARTI-Technician at the waste water
treatment experiment-site of ARDHI-University (Figure
8). The fenced and guarded location was chosen due to
security consideration and good accessibility of water.
To guarantee direct sunlight (7.5h from 9:00 to 16:30)
and hence high temperature inside the digester, three
medium-sized trees surrounding the digester had to be
cut down previous to the installation. The digester was
put on a wooden pallet.
Figure 8: Experimental site

Table 13: List of materials and prices used for the installation of ARTI plant
Pieces Product Price [TZS] Price [Euro]
1 HDPE (High-density polyethylene) Water tank
152’000 98.6
1000l (Simtank): Ø1.04m, height 1.25m
1 HDPE-watertank 750l (Simtank)
110’000 71.4
Ø1.30m, height 0.90m
1 T-Connector 3” 3500 2.3
1 Male Socket 3” 4500 2.9
1 Female Socket 3” 4500 2.9
1 Male Socket 2” 3000 1.9
1 Bucknut 2” 1800 1.2
1 Elbow 2” 1500 1.0
1 Tank Connector ¾“ 3500 2.3
1 Bush ¾” * ½” 600 0.4
1 Nipple ½” 600 0.4
1 Elbow ½” 500 0.3
1 Cock ½” 2000 1.3
2 Solid Tape 2000 1.3
2 Tangit Clue 3000 1.9
1 2-Component Epoxy (M-Seal) 3000 1.9
1 Pipe 3” (length 180 cm) [6m = 12'000 TZS] 4000 2.6
1 Pipe 2” (length 100 cm) [6m = 10’500 TZS] 2000 1.3
1 Hose (5m) 5000 3.2
1 Gasstove 50’000 32.5
Labor 100’000 64.9
Total 457’000 295.6

19
ZHAW Bachelor Thesis 09 Christian Lohri UI05

The following additional parts were used exclusively for the research plant:
1 PVC plug 2”, 1 PVC Socket 3” * 2”, 1 Ball Valve 2”, 1 Tank Connector ¾”, 1 Bucknut
¾” * ½“, 1 PVC Cock ½, 1 manometer (0-60mbar) to measure the gas pressure inside
the digester.

Procedure
A detailed description of the installation procedure together with a photo-documentation
is attached in Appendix B2. A DVD produced by ARTI-India includes a visual “do-it-
yourself” instruction and can be ordered from ARTI-India. The whole installation includ-
ing the inoculation of the ARDHI plant was carried out by one person and was com-
pleted within 3.5 hours.

Inoculum and start-up

60kg of dried cow dung (TS: 23%, VS: 74%) were
gradually mixed by hand with water in 25l buckets
and straw material was removed (Figure 9). The
homogenous mass was then poured into the di-
gester. Approximately 300l of effluent (pH: 6.38,
Redox-Potential: -238mV, TS: 0.12%, VS: 12%,
COD: 842mg/l) from the existing plant at ARTI-
office was added to the digester and served as
Figure 9: Preparing of cow dung
part of the inoculum. This effluent had previously
been enriched with a special powder-mix, which includes flour and yeast bacteria. Ac-
cording to the ARTI director, this step is done to create a secret about the start-up and
prevents users from doing the installation without support of ARTI-TZ (Oral information
Potnis, 6.8.08). Following the installation and inoculation, the digester was left without
feeding for 10 days, in which the change in gas composition was recorded and the gas
released on a daily basis.

3.1.2 Feedstock and feeding plan

Organic household waste in DSM mainly consists of kitchen waste, which can be di-
vided into food leftovers and peelings or pieces of vegetables and fruits. It was there-
fore decided to test two different substrates in the research plant, both representing
typical parts of household kitchen waste: Food waste (FW) and waste from the vegeta-
ble/fruit market (MW).

Figure 10: Food waste (FW, left) and market waste (MW, right)

20
ZHAW Bachelor Thesis 09 Christian Lohri UI05

The food remains were collected from the student’s canteen at ARDHI-University.
These leftovers consisted of stiff maize porridge (Ugali), rice, vegetable (beans and
spinach), potato chips, pieces of meat with sauce and fish residue (Figure 10). Orange
and banana peelings were also merged into the food remains. Soda caps, tooth picks
and meat bones were frequently among the food waste and had to be removed.

Fruit- and vegetable waste was obtained from Mwenge market. Its composition was
spoiled fruits (papaya, orange, banana, avocado, pineapple, passion fruit), spoiled
vegetables (tomato, eggplant, pepperoni, carrot, potato, cucumber, onion, spinach,
plantain, okra, broccoli) and vegetable peelings (cabbage, bean, cassava) (Figure 10).
All wastes were stored in closed buckets and used within a maximum of 4 days.

Pre-treatment and dilution of feedstock

The objective of pre-treatment is to reduce the size of substrate particles in order to
meet the basic requirement of fitting into the 3”- inlet-pipe. In addition, increasing the
surface of the feedstock allows better digestion for the bacteria responsible for the hy-
drolysis. The pieces of both substrates were treated previous to the feeding to attain a
particle size of less than 1cm.

Figure 11: Food waste raw, mashed, minced (left to right)

Figure 11 shows that pre-treatment of food waste was only needed for the meat, fish
and fruit pieces. They were either cut up with a kitchen knife or minced with a manual
meat-mincer. The large quantity and fibrous texture of the market waste, combined with
the small size of the meat-mincer and the repeated bluntness of its knives, resulted in
a low convenience for the operator. As a consequence, the fruit and vegetable waste
were cut up manually and then prepared with an electric kitchen blender (Philips,
HR2810/A, 400W) (Figure 12).

Figure 12: Market waste raw, manually cut, blended and diluted (left to right)

21
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Feeding plan
The experimental design in terms of feeding was divided into three phases (Table 14):

y Phase 1: Start-up
The quantity of feedstock during the start-up period (first week) was continuously in-
creased by 150g per feed until 1kg in the morning and 1kg in the evening was reached.
Afterwards, canteen waste was fed for two weeks (2kg per day), according to the in-
structions of ARTI-TZ: The chopped feedstock (1kg twice a day) was given into the inlet
pipe through a funnel (made of a cut-up 6l water bottle and a piece of pipe) and subse-
quently flushed with 10l of water. The effluent (10l) landed in the overflow bucket,
which was emptied daily.
Æ Aim of phase 1: Getting to know the system and finding adjustments of the feeding
procedure to enhance the digestion process.

y Phase 2: Feeding strategy

The following minor adjustments of the feeding method were consequently applied when
first feeding 2kg of canteen waste per day and afterwards 2kg of market waste: The cut
up feedstock was mixed previous to the feeding by giving it into a bucket and filling it
up to the 10l mark with water (stored in a container next to the digester to guarantee
equal temperatures). The well stirred diluted feedstock was then poured as influent
through the inlet-pipe to minimise the risk of blocking. The effluent in the overflow
bucket was two to three times fed back for flushing the inlet pipe (20-30l), hereby mix-
ing the bacteria of the effluent with the fresh influent. The amount of 2kg per day was
chosen as it represents a realistic quantity of organic waste produced by an average
household in Dar es Salaam (see 2.1.3).
Æ Aim of phase 2: Obtaining and comparing the influent, gas production/composition
and effluent over a period of constant feeding (reaching steady state conditions) for
both substrates.

y Phase 3: Maximum load

After a break of one week, the feeding load of both substrates was increased up to
5kg/day to observe if the operational, biological or chemical-physical limit was reached.
Æ Aim of phase 3: Testing the capacity of the system given by its size, construction or
biology involved.

To attain comparable results, the time of feeding and measuring was attempted to be
around 8:30 and 17:00 each day.

22
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Table 14: Feeding plan of the ARTI research biogas plant at ARDHI University
Week Month Phase Topic/feedstock Amount per feed Total daily feed Water/feed
[kg] [kg/d] [l]
installation &
31 July start-up
32 August Phase start-up
33 1 0.15 - 1.0 0.3 - 2.0
34
35 foodwaste 1.0 2.0 ca.9
36 September
37 0.5 1.0
Phase
38 0.5 1.0
2
39 marketwaste ca.9
1.0 2.0
40 October
41 pause
42 0.5 - 1.5 1.0 - 3.0
marketwaste (max) ca.8
43 2.0 - 2.5 4.0 - 5.0
Phase
44 November 0.5 - 1.0 1.0 - 2.0
3
45 foodwaste (max) 1.5 - 2.0 3.0 - 4.0 ca.8
46 2.0 - 2.5 4.0 - 5.0
47
48 gasholder removal: examining sludge, stratification and exact digester volume

Analyzed feedstock parameters

Table 15 presents the analysed parameters of the influent. The objectives of these
analyses were to characterize the raw feedstock and diluted feedstock (influent) re-
spectively and to compare these results afterwards with those of the effluent. This con-
sequently allowed an evaluation regarding the efficiency of the AD process.

Table 15: Analysed feedstock parameters (TS, VS) and diluted influent parameters (all)
Method & Frequency
Parameter Explanation/ Relevance
Instruments (Phase 2)
Residue upon water evaporation
after 48 hours drying at 105°C Oven
Total Solids represents the total solids and (Toschnival) Twice
(TS) presents a raw estimation of all Precision Scale per week
the organic and inorganic matter (Satorius 2355)
content in the original sample.
The fraction of solid matter that
can be oxidised and driven off as
gas at 550°C for 2 hours (constant
Muffel-Furnace
weight) is an approximation of the
Total Volatile (Vecstar, LF3) Twice
organic fraction of the dry matter
Solids (VS) Precision Scale per week
determined at 105°C (TS). The
(Satorius 2355)
residue is the inert (mineral) frac-
tion, mainly due to inorganic mat-
ter.
The oxygen equivalent of the or- Closed Reflux-
Total Chemical
ganic matter that can be oxidised. Method Twice
Oxygen Demand
CODtotal is a measure of all the Photospectrometer per week
(CODtotal)
organic matter in the sample. (Hach DR/2010)
Closed Reflux-
Dissolved Method
The CODdissolved shows the
Chemical Oxygen (Vacuum pump, Twice
amount of quickly digestible feed-
Demand membrane filter) per week
stock in the sample.
(CODdissolved) Photospectrometer
(Hach DR/2010)
Anaerobic digestion of N-
containing compounds release Nessler Method
Ammonium- +
NH4 , which is used by the bacte- (Filter paper) Twice
Nitrogen
ria as their source of nitrogen. Photospectrometer per week
(NH4-N)
Optimal NH4-N concentration in (Hach DR/2010)
digester: <1000mg/l

23
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Essential nutrient to growth of Semi-Micro-

Total Nitrogen
organisms. Organic nitrogen and Kjeldahl Method 3 times per
TKN
ammonia are together referred to (conducted by DSM feedstock
(Ntot)
as Kjeldahl nitrogen. University)
Essential nutrient to growth of Acid Persulfate-
Total Digestion Method 3 times
organisms, include orthophos-
Phosphorus Photospectrometer per feedstock
phates and organically bound
(Ptot) (Hach DR/2010)
phosphates.

The three samples per feedstock for Total Nitrogen analyses were stored in the fridge
and lateron brought to the laboratory of DSM University. The pH value and Temperature
of the diluted feedstock were measured sporadically (Hach-Lange Sension156). For
detailed methodology description see Appendix A1.

3.1.3 Monitoring of digester activity

In order to monitor the biological performance of the

different layers inside the unstirred digester, three
sampling sources were chosen (Figure 13): At the
height of the inlet pipe on the bottom of the digester
(low), on medium height (middle: 0.62m above the
ground) and at the height of the overflow pipe
(high). Each time before feeding, 5l of digester liq-
uid were released through the ball valve on the bot-
tom to analyze the pH and temperature (low). The
same was done with 1l off the tap at medium height
(middle, each afternoon the Redox-Potential was
additionally measured here) and afterwards the pH
and Temperature were also analysed by dipping the
electrode into the digester surface (high). Figure 13: Sampling points

3.1.4 Gas measurements

The gas produced in the digester of the research plant at ARDHI University was re-
leased without utilization.

Gas production
Different methods were taken into consideration to measure the daily gas production
(Liquid displacement, rocking displacement gasmeter, gasometer, flow-meter, Appendix
C1 & C2). However, all of them require fixation of the floating drum and hence alter the
system as a whole, e.g. the counter pressure caused by the attached instrument leads
to a significant reduction of digester liquid before it is overcome. Liquid displacement
was not suitable as it only represents the gas production over a short period of time
and is therefore rather applicable for lab-scale reactors with batch-feeding. The ex-
tremely low gas production rate per time, which resulted in low flow rates was the main
obstacle for flow-meter, rocking displacement gasmeter and gasometer. The gasometer
(Schlumberger) for example requires a minimum flow rate or permanent gas production
of 16l/h, which was not reached.

24
ZHAW Bachelor Thesis 09 Christian Lohri UI05

After numerous experiments, it was

decided to apply a scaling on the
gasholder. Hence the biogas pro-
duced was stored in the closed
gasholder until the drum was lifted
to maximal extent. Controlled by
the Gasometer, 20l of gas was re-
leased. The tap was then closed
and a white line drawn on the gas-
holder just above the surface of the
digester liquid, using a Tipp-Ex
marker. This procedure was re-
peated until all biogas (400l) was
released. The scaling was con-
trolled by calculation (Figure 14). Figure 14: Scaling on gasholder

Starting on 17 August 2008, the amount of gas in the gasholder was recorded daily in
the morning and afternoon before feeding.

Figure 15: Continuous rising of gasholder proportional to gas production

Gas composition
The Dräger X-am 7000 was used to measure the volume percentage of methane (CH4),
carbon dioxide (CO2) and oxygen (O2) in the biogas. Integrated sensors further reveal
the content of hydrogen sulphide (H2S) in the range of 0 – 100ppm and of ammonia
(NH3) between 0 and 200ppm. The gas composition was measured on a daily basis
throughout phase 1 to 3 in the afternoon (before feeding) while releasing the gas.

25
ZHAW Bachelor Thesis 09 Christian Lohri UI05

3.1.5 Effluent

Table 16 presents the effluent parameters, explains their relevance and lists the
method and frequency of analyses.

Table 16: Analyzed effluent parameters

Parameter Explanation / Relevance Method & means Frequency

(Phase 2)
Indicates intensity of acidic
or basic character of the Daily
pH Sension 156
solution at a given temp. before feeding
(Hach-Lange)
Optimal pH-range for one- (low,middle,high)
stage digester: 6.5-7.5
Measure of the average ki-
netic energy of atoms or Sension 156 Daily
Temperature molecules. before feeding
(Hach-Lange)
Optimal temp.-range for (low,middle,high)
mesophilic bacteria: 35°C
Measure of the affinity of a Daily
substance for electrons. Sension 156 before feeding
Redox-Potential
Optimal Redox-Potential for (Hach-Lange) in afternoon
CH4-production: <330mV (middle)
Twice
Total Solids (TS) See 3.1.2 See 3.1.2
per week
Total Volatile Solids Twice
See 3.1.2 See 3.1.2
(VS) per week
Total Chemical
Twice
Oxygen Demand See 3.1.2 See 3.1.2
per week
(CODtotal)
Dissolved Chemical
Twice
Oxygen Demand See 3.1.2 See 3.1.2
per week
(CODdissolved)
Fatty acids with a carbon
chain of six carbons or fewer Nordmann-
are produced in the first Titration
Volatile Fatty Acids Twice
stage of the anaerobic diges- (Phase 2&3)
(VFA) per week
tion process. Kapp-Titration
Inhibiting concentration: (Phase 3)
>3000mg/l (Eder &Schulz)
An increase in acids- (or pro-
portional decrease in carbon- Nordmann-
Ratio of Volatile
ate alkalinity-) concentration Titration
Fatty Acids to Total Twice
is the first practical measur- (Phase 2),
Inorganic Carbon per week
able indication that an an- Kapp-Titration
(A/TIC-ratio)
aerobic treatment system is in (Phase 3)
a state of stress.
Total Nitrogen 4 times per feed-
See 3.1.2 See 3.1.2
(N tot) stock
Ammonium-Nitrogen Twice
See 3.1.2 See 3.1.2
(NH4-N) per week
Total Phosphorus 4 times
See 3.1.2 See 3.1.2 per feedstock
(P tot)
PhosVer3 Method
Fraction of total phosphorus (Filter paper)
Phosphate Twice
which is directly available to Photospectrometer
(PO4) per week
microorganisms (Hach DR/2010)
Toxic effect (for AD):
Pb: 340mg/l Atomic Absorption
Heavy Metals 4 times
Cu: 170-300mg/l Spectrometer
(Cd, Cu, Pb) per feedstock
Cd: 20-600mg/l (AAS)
(Eder&Schulz, 2003)

Detailed methodologies for TS, VS, Ammonium, Total Nitrogen, Phosphate, Total Phos-
phorus and Chemical Oxygen Demand are described in Appendix A1. The titration meth-
odologies for VFA, alkalinity and A/TIC-ratio (according to Nordmann and Kapp) can be
found in Appendix A2.

26
ZHAW Bachelor Thesis 09 Christian Lohri UI05

3.1.6 Calculations

The operational parameters commonly used to describe anaerobic processes are listed
and explained underneath (Mata-Alvarez, 2003).

y Hydraulic Retention Time (HRT)

Describes the ratio of the reactor volume to the flow rate of the feed. It hence ex-
presses the time a fluid element spends in the digester (strictly true for ideal reactors).

HRT = V / Q

HRT = hydraulic retention time [days]

3
V = reactor volume [m ]
Q = flow rate [m3/day]

y Solid Retention Time (SRT)

Describes the ratio between the content of total solids in the reactor and the solids flow
rate of the effluent.

SRT = V*X/W

SRT = solid retention time [days]

3
V = reactor volume [m ]
3
X = volatile solids concentration in the reactor [kg VS/m ]
W = flow rate of the extracted volatile matter from the reactor [kg VS/day]

y Organic Loading Rate (OLR)

Describes the substrate quantity introduced into the reactor volume in a given time,
whereby the substrate can be defined as TS, VS, COD or BOD.

OLR = Q*S/V

3
OLR = organic loading rate [kg feed/m reactor day]
3
Q = substrate flow rate [m /day]
3
S = substrate concentration in the inflow [kg/m ]
3
V = reactor volume [m ]

27
ZHAW Bachelor Thesis 09 Christian Lohri UI05

y Gas Production Rate (GPR)

Describes the ratio between the produced biogas and reactor volume in a given time.

GPR = Qbiogas/V

3
GPR = gas production rate [m gas/m3 reactor day]
3
Qbiogas = biogas flow rate [m /day]
3
V = reactor volume [m ]

y Specific Gas Production (SGP)

Indicates the biogas produced by a unit of mass of substrate, in terms of the total vola-
3
tile solids in the feed, as m biogas/kgsubstrate fed. This index is strictly linked both to the
biodegradability of the fed substrate and to the process attitude. The SGP value is of-
ten used to compare the performances of different anaerobic processes.

SGP = Qbiogas/Q*S

3
SPG = specific gas production [m biogas/kg feed]
3
Qbiogas = biogas flow rate [m /day]
3
Q = inlet flow rate [m /day]
3
S = substrate concentration (VS) in the influent [kg substrate/ m ]

y Substrate removal effectiveness

Also called substrate conversion, this parameter can be expressed in several ways and
the substrate measured in terms of TS, VS or COD. Generally, the simplest and most
used equation is:

η% = (Q*S-Q*Se)/(Q*S)*100

η% = TS, VS or COD removed, as percentage [%]

3
Q = inlet and outletflow rate [m /day]
3
S = TS, VS or COD concentration in the inlet flow rate [kg/m ]
3
Se = TS, VS or COD concentration in the effluent flow rate [kg/m ]

28
ZHAW Bachelor Thesis 09 Christian Lohri UI05

y General gas equation

To be able to compare the gas production of the research plant in DSM with results
from literature, the measured gas quantity needs to be converted into norm litres (norm
conditions: 1.013bar, 0°C). Hence, the general gas equation is applied:

p * V = n * R * T

p [bar = 105 N/m2] Pressure of gas, norm conditions: pN = 1.013 bar

V [l] Volume of gas, quantity per time
n [Mol] Mol per gas, 1 Mol = 22.4 Nl
R [N*m*mol-1K-1] Universal gas constant, 8.31 N m mol-1K-1
T [°K] Absolute temperature, norm conditions: TN = 273 °K

y Conversion of gas quantity measured in Dar es salaam into norm liters:

Conditions TZ = Norm conditions
(p1 * V1)/T1 = n * R = (pN * VN)/TN = n * R

p1 = local pressure in system of which the measured gas V1 origins

V1 = measured gas quantity
T1 = temperature in °K

= (p1 * V1)/T1) = (pN * VN)/TN)

= VN = (p1 * V1)/T1) / (pN)/TN)

VN = (p1 * V1 * TN) / (pN * T1)

y Approximate maximum biogas yield

The following two steps enable the calculation of the approximate biogas potential:

1) COD Æ C

C + O2 Æ CO2

12g + 32g (=COD) Æ CO2

12g = 32 COD
Æ 1g COD = 0.375g C

2) C Æ approximate maximum biogas yield

2C Æ CO2 + CH4

2 * 12g Æ 1Mol + 1 Mol

24g = 24g = 2 * 22.4Nl = 44.8Nl
Æ 1g C = 44.8/24
= 1.87 norm litres ‘gas’ maximum (50% CO2, 50% CH4)

29
ZHAW Bachelor Thesis 09 Christian Lohri UI05

3.2 Inspection of ARTI plants in Dar es Salaam

On 18 and 19 October 2008, 17 ARTI Compact biogas systems in DSM were visited dur-
ing an inspection tour to document their status of functioning (12 at household level, 5
at institutional level). A list of the visited ARTI plants can be found in Appendix B1.
5½ weeks later, on 28 November 2008, a second inspection tour was done to observe
the progress.

3.2.1 Analysed parameters

On the first tour, the on-site measurements were conducted on effluent-samples and
included pH, temperature and Redox-potential. The gas composition was furthermore
analysed using Dräger X-am 7000. 1.5l of effluent was taken from each plant and
brought to the Environmental Engineering Laboratory of ARDHI-University where TS,
VS, CODtotal, NH4-N, Ptotal and A/TIC (according to Nordmann), were analysed the fol-
lowing day. The concentration of heavy metals (Pb, Cd, Cu) was examined from 4 in-
spection sources. Samples which showed high quantity of invertebrate were taken to
the zoology laboratory at DSM University and identified on 27.11.2008 (Identification
key see Appendix C4).

On the second tour, the status of functioning was assessed and the combustibility of
the biogas was tested. pH was furthermore measured using Litmus stripes (Merck “Neu-
tralit” pH-indicator stripes pH 5.0-10.0).

3.2.2 Interview s

An interview with the person responsible for operation of the biogas plant was done on
the first tour to learn more about the quality, quantity, pre-treatment method and dilu-
tion of the daily feedstock, the duration of daily gas use and the utilization of effluent.
Additionally, emphasizes was laid on any problems that had occurred and room was
given to suggest ideas for improvements (Questionnaire in Appendix B6)

30
ZHAW Bachelor Thesis 09 Christian Lohri UI05

4 Results

4.1 Research-plant at ARDHI University

All results from the research CBS at ARDHI University can be found in Appendix B5.

4.1.1 Influence of enhanced feeding procedure on process stability

As an indicator of the anaerobic process stability, Figure 16 presents the development

of the pH inside the digester during phase 1 (neither previous mixing of feedstock with
water nor feedback of effluent). The pH was measured twice per day before feeding in
the morning (8:30) and evening (17:00) on three different reactor heights (low, middle,
high). Generally, pH-values below 6 should be avoided as it restrains the activity of the
methanogenic bacteria. The amount of daily feed (food waste) is shown as well as the
digester temperature at medium height (average 27.8°C).

Figure 16: pH development inside digester during phase 1: start-up

As a comparison, Figure 17 reveals the development of the pH during phase 2 where

emphasis was laid on enhanced mixing of the digester liquid. As a result, the pH values
of all heights have stabilized around pH 6.5 and show less deviations. The average
temperature inside the reactor (middle) was 28.8°C. The dark green spots represent kg
wet weight (ww) of food waste (FW) as feedstock, whereas the light green spots stand
for market waste (MW) of fruit and vegetables.

Figure 17: pH development inside digester during phase 2: feeding strategy

31
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Figure 18 illustrates the stability of the digester during phase 3. Although the daily
amount of feed was increased to 5kg/day, the reactor liquid did not show any alarming
sign of disturbance, with pH only dropping slightly down to 6.2. As the average air tem-
perature during this phase increased, so did the average digester temperature
(30.9°C). It is conspicuous that due to the improved mixing of feedstock with water pre-
vious to the feeding and feedback of effluent, the pH values of low, middle and high
reactor height are noticeably more congruent in phase 2 and 3 compared to phase 1.

Figure 18: pH development inside digester during phase 3: maximum load

As the pH- value reacts with quite some delay to changes in quantity and quality of
feeding substrate, the A/TIC-ratio provides valuable information in terms of immediate
process stability. Figure 19 presents the course of A/TIC-ratio (Nordmann titration)
while feeding both substrates. It is apparent that from the inoculum (cow dung), the
amount of VFA was quite high in the beginning (day 8 & 11) compared to the alkalinity,
which results in a high A/TIC value. After 25.08.08 (day 15), the A/TIC ratio was stabi-
lized around 0.15 (feedstock: 2kg food waste/day). Following the change to 2kg market
waste per day (day 46), the A/TIC-ratio was levelled around 0.08.

Figure 19: Development of A/TIC-ratio (Nordmann) while feeding FW (day 8-32) and
MW (day 43-67)

32
ZHAW Bachelor Thesis 09 Christian Lohri UI05

4.1.2 Stratification inside digester in terms of TS, VS and COD

Figure 20, Figure 21 and Figure 22 present the different concentrations of TS, VS and
COD inside the digester on the reactor bottom, at medium height and on the top layer
(effluent) compared to the influent (FW until day 26, MW starting on day 46). A clear
accumulation of solids and organic material can be observed on the ground of the di-
gester due to sedimentation.

Figure 20: Comparison of TS concentration in influent, digester liquid and effluent

Figure 21: Comparison of VS concentration in influent, digester liquid and effluent

Figure 22: Comparison of COD concentration in influent, digester liquid and effluent

33
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Figure 23 illustrates the stratification of the digester liquid which was analysed on
25.11.2008 after removing the gasholder and emptying the digester stepwise layer after
layer (1 layer = 85l) with a hose using the cohesion force of water.

Figure 23: Stratification of digester liquid in ARDHI research plant

Table 17 reveals the detailed results of the TS and VS analyses. The majority of di-
gester liquid (600l) as well as the effluent show similar TS and VS content, while the
surface level (floating layer) inside the gasholder is characterized by a high TS and VS
concentration.

Table 17: TS and VS of layers inside the digester (26.11.2008)

Height no. TS [g/l] TS [%] VS [g/l] VS [%]
0 effluent (high) 3.65 0.37 1.47 40.26
1 surface 74.11 7.41 60.58 81.75
2 3.33 0.33 1.42 42.74
3 3.37 0.34 1.30 38.71
4 3.62 0.36 1.41 38.85
5 middle 3.33 0.33 1.24 37.24
6 3.29 0.33 1.43 43.44
7 3.47 0.35 1.42 40.88
8 3.24 0.32 1.46 45.00
9 5.28 0.53 3.16 59.76
10 bottom (low) 41.17 4.12 32.68 79.36
Average (1-10) 14.42 1.44 10.61 50.77

34
ZHAW Bachelor Thesis 09 Christian Lohri UI05

4.1.3 Intense monitoring of VF A, alkalinity and A/TIC

Table 18 presents the VFA- and alkalinity concentration of the effluent as well as the
A/TIC ratio over a period of 20 days while the amount of daily feedstock (food waste)
was changed. Different pre-treatment methods and the titration according to Kapp were
applied. Because of its time- and material-intensity, the membrane filtration was only
conducted once per day. The results of both other pre-treatment methods (sieve/textile
filtration and centrifugation) are mean values of two repetitions. The rather slow-acting
pH values are listed as comparison.

Table 18: VFA, alkalinity and A/TIC results applying different pre-treatment methods (titration
according to Kapp)

centrifuged
sieve & textile filtered (10min/5000rpm) membrane filtered

Feedstock VFA Alkalinity VFA Alkalini- VFA Alkalini-

FW pH (Kapp) [mg/l] A/TIC [mg/l] ty [mg/l] A/TIC [mg/l] ty [mg/l] A/TIC
Day [kg/d] (effluent) [mg/l] (Kapp) (Kapp) (Kapp) (Kapp) (Kapp) (Kapp) (Kapp) (Kapp)
86 3 6.29 61 884 0.07 55 880 0.06 45 869 0.05
88 3 6.16 58 851 0.07 32 867 0.04 30 828 0.04
90 4 6.23 123 797 0.15 121 811 0.15 78 803 0.10
92 4 6.24 124 845 0.15 158 771 0.21 106 761 0.14
94 5 6.35 124 845 0.15 89 835 0.11 84 813 0.10
96 5 6.15 193 804 0.24 195 855 0.23 164 793 0.21
98 0 6.43 52 981 0.05 46 1002 0.05 35 961 0.04
100 0 6.77 75 1121 0.07 83 1140 0.07 64 1136 0.06
103 2 6.39 132 1136 0.12 86 1168 0.07 63 1141 0.06
105 2 6.45 102 1129 0.09 67 1151 0.06 75 1161 0.06

To compare the accuracy of the method, the Nordmann titration was further applied
over the same period of time (Table 19). The pre-treating method consisted of filtering
the effluent samples first through a kitchen sieve and then through a textile mash. The
presented results are again mean values of two repetitions.

Table 19: VFA, alkalinity and A/TIC results (titration according to Nordmann)

sieve & textile filtered

Feedstock VFA Alkalinity A/TIC

FW pH [mg/l] [mg/l] (Nord- It should be noted that
Day [kg/d] (effluent) (Nordmann) (Nordmann) mann)
86 3 6.29 43 847 0.05
previous to the analyses
88 3 6.16 31 817 0.04 on day 94 (12.11.08), a
90 4 6.23 91 741 0.12 heavy rain occurred. It
92 4 6.24 118 703 0.17
94 5 6.35 81 789 0.10
eventually diluted the ef-
96 5 6.15 147 734 0.20 fluent sample to some ex-
98 0 6.43 40 943 0.04 tent.
100 0 6.77 63 1028 0.06
103 2 6.39 121 1073 0.11
105 2 6.45 121 1069 0.11

Table 18 and Table 19 reveal that the influences of different titration and pre-treatment
methods on alkalinity and A/TIC results are negligible. However, for measurements of
the VFA concentration, the Kapp titration is strongly favoured over the Nordmann
method because of its better accuracy.

35
ZHAW Bachelor Thesis 09 Christian Lohri UI05

All results of the various A/TIC analyses are presented in Figure 24, showing its devel-
opments and deviations. As far as statements in terms of process stability are con-
cerned, both titrations (according to Kapp and Nordmann) and all pre-treatment meth-
ods are applicable, the simple ‘sieve&textile’ treatment being preferable due to its sim-
plicity.

Figure 24: Results of A/TIC titrations (Kapp and Nordmann) with different pre-treatment

36
ZHAW Bachelor Thesis 09 Christian Lohri UI05

4.1.4 Operational parameters

Below are the calculated operational parameters of the research plant at ARDHI Uni-
versity, while constantly feeding 2kg of substrate over a defined period of time (Food
waste: 18.08 – 08.09.2008, 7 samples, 2 repetitions), market waste: 25.09 –
16.10.2008, 5 samples, 2 repetitions).

y Hydraulic Retention Time (HRT) of ARDHI plant

3
Food waste and market waste: 0.85m / 0.02m3 = 42.5days

y Solid Retention Time (SRT) of ARDHI plant:

3 3
Food waste and market waste: 0.85m * 10.6kg VS/m / 0.03kg VS/day = 300days

y Organic Loading Rate (OLR) of ARDHI plant:

3 3 3 3
Food waste (TS) 0.02m /d * 26.7kg/m / 0.85m = 0.628kg/m reactor d
3 3 3 3
(VS) 0.02m /d * 22.6kg/m / 0.85m = 0.532kg/m reactor d
3 3 3 3
(COD) 0.02m /d * 28.3kg/m / 0.85m = 0.666kg/m reactor d

3 3 3 3
Market waste (TS) 0.02m /d * 8.2kg/m / 0.85m = 0.193kg/m reactor d
3 3 3 3
(VS) 0.02m /d * 7.0kg/m / 0.85m = 0.165kg/m reactor d
3 3 3 3
(COD) 0.02m /d * 8.6kg/m / 0.85m = 0.202kg/m reactor d

y Gas Production Rate (GPR) of ARDHI plant:

3 3 3 3
Food waste 0.234Nm /d / 0.85m = 0.275Nm gas/m reactor day

3 3 3 3
Market waste 0.122Nm /d / 0.85m = 0.144Nm gas/m reactor day

y Specific Gas Production (SGP) of ARDHI plant:

3 3 3
Food waste 0.234Nm /d / 0.02m /d * 22.6kg VS/m =
3
264.4Nm biogas/kg feed

3 3 3
Market waste 0.122Nm /d / 0.02m /d * 7.0kg VS/m =
42.7Nm3 biogas/kg feed

37
ZHAW Bachelor Thesis 09 Christian Lohri UI05

y Substrate removal effectiveness (η%) of ARDHI plant:

Food waste: (TS) (0.02m3/d * 26.7kg/m3 – 0.02m3/d * 3.7kg/m3) /

3 3
(0.02m /d * 26.7kg/m ) * 100 = 86.1%
3
(VS) (0.02m /d * 22.6kg/m3 – 0.02m3/d * 1.7kg/m3) /
3 3
(0.02m /d * 22.6kg/m ) * 100 = 92.5%
3
(COD) (0.02m /d * 28.3kg/m3 – 0.02m3/d * 4.8kg/m3) /
3 3
(0.02m /d * 28.3kg/m ) * 100 = 83.0%

3
Market waste (TS) (0.02m /d * 8.2kg/m3 – 0.02m3/d * 2.7kg/m3) /
3 3
(0.02m /d * 8.2kg/m ) * 100 = 67.1%
3
(VS) (0.02m /d * 7.0kg/m3 – 0.02m3/d * 1.3kg/m3) /
3 3
(0.02m /d * 7.0kg/m ) * 100 = 81.4%
(COD) (0.02m3/d * 8.6kg/m3 – 0.02m3/d * 1.2kg/m3) /
3 3
(0.02m /d * 8.6kg/m ) * 100 = 86.0%

38
ZHAW Bachelor Thesis 09 Christian Lohri UI05

4.1.5 Reduction of w aste volume and organic load

Figure 25 shows the degree of daily waste reduction (orange) by comparing TS concen-
tration of the influent with TS of the effluent.

-84.9%

-72.8%

Figure 25: Reduction of waste volume (TS) when feeding FW (day 8-32) & MW (day 43-67)

One option to describe the reduction of organic load is given by measuring the VS of
influent and effluent (Figure 26). The average VS of food waste influent is 451g/d and
of its effluent 35g/d, which results in a reduction rate of 92.2%. The mean VS of market
waste influent is 177g/d and of its effluent 26g/d, which leads to a reduction of 85.3%.

-92.2%

-85.3%

Figure 26: Reduction of VS content when feeding FW (day 8-32) and MW (day 43-67)

Another option to describe the reduction of organic load is through COD. As Figure 27
illustrates, the daily COD content of influent and effluent between FW (average influent
567g/d, average effluent 96g/d) and MW (average influent 152g/d, average effluent
24g/d) differ considerably. However, the reduction in term of COD for both substrates
was found to be comparable at approximately 83%.

39
ZHAW Bachelor Thesis 09 Christian Lohri UI05

-83.1%

-83.9%

Figure 27: Reduction of COD content when feeding FW (day 8 -32) & MW (day 43 - 67)

Figure 28 reveals the per-

centage of COD dissolved
28.3% compared to COD total
from influent (2kg food
waste & market waste) and
effluent.

56.5% Figure 29 & 30 present an

43.3%
58.6% overview of TS, VS and
COD reduction, both when
feeding 2kg FW and MW
Figure 28: Comparison of COD solid and COD dissolved from
FW and MW per day.

Figure 29: Input/Output with daily load of 2kg/d Food waste

Figure 30: Input/Output with daily load of 2kg/d Marke twaste

40
ZHAW Bachelor Thesis 09 Christian Lohri UI05

4.1.6 Gas resulting from anaerobic digestion of food waste and market waste

After reaching a steady-state condition, the two feeding substrates were characterized
and recorded in terms of wet weight, TS, VS and CODtotal over a period of 10 days:
Food waste: 29.08.- 07.09.2008 (day 19 – 28); Market waste: 25.09 - 04.10.2008 (day
46 – 55). Table 20 presents the average values of these parameters and the quantity of
the gas produced (measured on site and adjusted to norm conditions according to the
general gas equation, see 3.1.6)

Table 20: Mean values during 10-day steady state conditions for both substrates
Gas Gas
Feedstock Wet weight TS VS CODtotal
production production
(dilution 1:20) [kg/d] [kg/d] [kg/d] [kg/d]
DSM [l/d] Norm [Nl/d]
Food waste 2 0.482 0.451 0.567 289 259
Market waste 2 0.202 0.180 0.172 126 113

Table 21 shows the average biogas production for the two different feed substrates per
kg wet weight, TS, VS and COD. In brackets are the results transformed to norm condi-
tions (0°C, 1013mbar).

Table 21: Gas production in dependence to daily wet weight, TS, VS and COD for both sub-
strates (transformed to norm conditions)

Average gas Average gas Average gas Average gas

Feedstock
per kg wet weight per kg TS per kg VS per kg COD
(dilution 1:20)
[Nl/kg WW] [Nl/kg TS] [Nl/kg VS] [Nl/kg COD]
Foodwaste 144 (129) 599 (537) 640 (574) 509 (457)
Marketwaste 63 (57) 624 (559) 700 (628) 733 (657)

Figure 31 shows the daily gas production over 10 days of steady-state conditions re-
sulting from 2kg FW (wet weight) and as comparison 2kg MW (ww). The daily VS aver-
age of FW was 0.451kg/d and 0.177kg/d for MW respectively and is marked as yellow
spots. The average gas production during this period was 289l/d (259Nl/d) from FW and
126l/d (113Nl/d) from MW.

Average: 289l/d

Average: 259Nl/d

Average: 126l/d

Average: 113Nl/d

Figure 31: Comparison of daily gas production from FW and MW during 10 days of steady-state
conditions

41
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Figure 32 presents the development of daily gas production during phase 3, when daily
feeding load was increased to 5kg/d wet weight MW (0.5kg/d TS) and subsequently
5kg/d FW (1.2kg/d TS). Between feeding MW and FW, a 3-day feeding break was inter-
mitted in which the gas production dropped to 70Nl/d. For detailed gas productions see
Appendix A3.

Figure 32: Gas production while increasing daily feed:MW (day 64-76), FW (day 80-100)

Figure 33 reveals the correlation between the daily feeding load and the consequential
gas production. The gas quantity resulting from food waste (FW) is approximately dou-
ble as high as from market waste (MW). While FW shows an almost linear increase in
gas quantity, the relatively high gas amount resulting from 1kg MW can possibly be
explained with the heterogeneity of the input material.

Figure 33: Gas production in dependence with daily feeding load

Approximate maximum biogas yield

As presented in Figure 29 and Figure 30, the average COD reduction was 470.6g/d
(food waste) and 147.3g/d (market waste). This corresponds to the decomposition of
176.5g C per day and 55.2g C per day respectively. The calculated approximate maxi-
mum biogas potential is therefore 330Nl (FW) and 103Nl (MW) per day (Appendix A3),
which is 78.6% and 109.6% of the measured gas production. The deviation between
results from measurement and calculation can to some extent be explained by the inac-
curacy of the conversion (COD Æ C) and the assumption that biogas consist of 50%
CO2 and 50% CH4 (see 3.1.6).

42
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Gas composition
The results of the daily gas composition measurements are shown in Figure 34 as aver-
age values recorded during the 10-days steady-state conditions (ww 2kg/d) each for FW
and MW. The gas resulting from MW with 66.4% show a higher methane content com-
pared to the gas from FW (56.8%).

Figure 34: Average gas composition from 2kg/d FW (left) and MW (right), H2S >100ppm and NH3
>200ppm are not shown

Biogas utilization: Daily running time of burner

Numerous tests have been done to measure the burning period
of 100l biogas produced at the research plant at ARDHI, using
the biogas stove distributed by ARTI-TZ (Figure 35).

Table 22 presents the contents of methane and results of all

tests when no weight was applied on the gasholder.

Figure 35: Gas stove

attached on research
CBS for burning tests

Table 22: Tests of burning duration for 100l biogas without weight on gasholder
Pressure in
Methane Burning duration
Date reactor [mbar]
content of 100l biogas
no weight
[Vol-%] [min]
on gasholder
28.08.2008 60 2 25
60 2 27
06.09.2008 62 2 24
10.09.2008 62 2 20.5
62 2 22.5
62 2 23
25.09.2008 70 2 19.5
15.11.2008 59 2 23
25.10.2008 56 2 19
03.11.2008 61 2 19.5
Average 61.4 2 22.3

43
ZHAW Bachelor Thesis 09 Christian Lohri UI05

ARTI-TZ occasionally advises their customers to increase

the gas pressure by permanently applying weight on the
gasholder. Table 23 shows the test results, when ½ cement
brick (approximately 12kg) was applied. The burning dura-
tion of biogas consequently decreases, yet the required
cooking time for the same food also reduces.

Figure 36: Weight applied on

gasholder

Table 23: Tests of burning duration for 100l biogas with weight on gasholder
Date Methane Pressure in Burning duration
content reactor [mbar] of 100l biogas
[Vol-%] 1/2brick (12kg) [min]
on gasholder
06.09.2008 62 4 13.5
25.10.2008 56 4 9
03.11.2008 61 4 12
21.11.2008 68 4 12.5
68 4 13.5
Average 62.8 4 12.1

4.1.7 Effluent quality

The average concentration of total phosphorus in the effluent samples while feeding
food waste was 248mg/l (6.8% of TS), of which 171mg/l (69%) occurred as phosphate.
After changing substrate to market waste, 147mg/l of phosphate were found, contribut-
ing 66% to the total phosphorus concentration of 225mg/l (8.2% of TS). As comparison,
162mg/l of total P were found in the influent consisting of diluted market waste.

Figure 37: Concentration of Ptotal and PO4 in effluent while feeding FW

(day 8-32) and MW (day 43-67)

44
ZHAW Bachelor Thesis 09 Christian Lohri UI05

While average ammonium-nitrogen concentration in the influent was found to be

31.6mg/l in food waste and 27.9mg/l in market waste, the NH4-N concentration discov-
ered in the effluent was 74.1mg/ (FW) and 85.5mg/l (MW). In other words, the bacterial
activity resulted in an NH4-N increase of 134.5% for FW, and 206.5% for MW respec-
tively (Figure 38).

Figure 38: Ammonium-Nitrogen difference between influent and effluent

Heavy metals
The results of the Atomic Absorption Spectrophotometer (AAS) analyses for Lead (Pb),
Copper (Cu) and Cadmium (Cd) are presented in Table 24.

Table 24: Heavy metal concentrations in effluents

from research plant at ARDHI

Date Pb [mg/l] Cu [mg/l] Cd [mg/l]

28.08.08 0.054 0.000 0.000 It should be considered, that the range
28.08.08 0.072 0.000 0.000 for inhibiting effects on anaerobic proc-
08.09.08 0.093 0.000 0.000
ess (Cu: 40-250mg/l, Cd: 150-600mg/l)
11.09.08 0.061 0.000 0.000
were by far not reached. On the other
22.09.08 0.036 0.000 0.018
side, the requirement for Pb concentra-
03.10.08 0.095 0.000 0.013
tion (0.020-200mg/l) is fulfilled in all
16.10.08 0.057 0.006 0.021
samples (Eder & Schulz, 2006).
Average 0.067 0.001 0.007

The quality of the effluent as organic fertilizer can only to some limited extent be
evaluated, as the suitability depends on the kind of plant it is applied. Furthermore, the
content of Potassium (K), but also Fe, Ca, Mg and Zn are of vital interest because
phosphate builds chelates with these essential trace elements which potentially prevent
the reception of PO4.

Comparison with Quality Standards of Compost is also of limited validity. Nevertheless,

the measured heavy metal concentrations are far below the tolerated values for com-
post (see Appendix C5).

45
ZHAW Bachelor Thesis 09 Christian Lohri UI05

4.2 Inspection of installed ARTI-plants in DSM

4.2.1 Result ove r view of first inspection tour

Table 25 presents an overview of all visited ARTI plants during the first inspection tour.
On 18/19 October 2008 out of 12 ARTI Compact biogas systems (CBS) at household
level, 4 were in operation, whereas 8 were not in use due to various reasons, which are
explained in 4.2.2. 5 additional CBS were furthermore visited on the first inspection
tour to serve as comparison (marked grey in Table 25). These plants include the CBS at
ARTI-office, which is used as demonstration plant and fed irregularly, the research
plant at ARDHI University and the three recently installed 4000l-digesters of Azania
Secondary School at institutional level (Appendix B7).

Table 26 reveals the gas composition of the inspected systems, which in general corre-
spond well with the findings of the research plant at ARDHI. The high methane content
of CBS #14 can be explained with the break-down and thus congested feeding. Because
the CO2-releasing process of hydrolysis became deactivated, the content of methane
increased disproportionately. Contrarily, the low methane concentration of CBS #30 can
be explained by its recent set-up (September 2008) and hence shows the predominant
product of the hydrolysis (CO2).

46
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Table 25: Overview of effluent results from first ARTI inspection tour (18/19 October 2008)

E f f l u e n t
Si ze Alka-
ARTI VFA
(dig./ COD NH4- P linity A/TIC
CSB-# Temp. Redox TS VS Pb Cu Cd (Nord-
gas- Status pH total N total (Nord- (Nord- Remarks
(installa- [°C] [mV] [%] [%] [mg/l] [mg/l] [mg/l] mann)
holder) [mg/l] [mg/l] [mg/l] mann) mann)
tion date) [mg/l]
[l] [mg/l]
2 1000/ In Many inver-
(02/2007) 750 use 6.78 30.8 -270 0.37 61.3 3340 93.0 166 0.04 0.00 0.00 57.8 1200 0.05 tebrate larvae
3 1000/ In For dog food
(04/2007) 750 use 6.47 28.6 -420 0.22 56.1 790 117.0 244 0.08 0.02 0.01 124.2 1180 0.11 H2O heater
4 1000/ Not Inlet ran down
(04/2007) 750 in use No digester liquid in CBS (by dogs)
5 1000/ Not Cause of break-
(04/2007) 750 in use No digester liquid in CBS down unknown
6 1000/ Not Inlet pipe bro-
(05/2007) 750 in use No digester liquid in CBS ken
7 1000/ Not
(06/2007) 750 in use No digester liquid in CBS 4th breakdown
14 1000/ Not Inlet blocked,
(03/2008) 750 in use 7.60 36.0 +37 364.0 157.4 1780 0.09 not enough feed
15 1000/ Not Gas tap
(03/2008) 750 in use 6.48 31.3 -312 0.24 54.6 2640 119.0 94 0.14 0.02 0.00 91 820 0.11 broken
17 2000/ Not
(05/2008) 1500 in use 4.15 31.8 -182 63.5 Overfed
18 2000/ In Inlet slightely
(05/2008) 1500 use 6.88 33.5 -437 1.75 63.7 5640 312.0 304 290.2 2410 0.12 blocked
25 1000/ Not
(08/2008) 750 in use 6.61 31.3 -174 0.18 54.8 2020 67.5 368 124.2 810 0.15 Inlet blocked
30 1000/ In
(09/2008) 750 use 6.08 34.8 -60 0.23 62.7 3480 31.5 256 0.00 0.03 0.02 456.2 590 0.77 New (22.9.08)
1 1000/ In Demo-plant
(11/2006) 750 use 6.26 33.0 -503 0.16 59.3 3270 95.5 78 57.8 630 0.09 (ARTI-office)
24 1000/ In Research plant
(07/2008) 750 use 6.31 33.2 -455 0.31 58.6 1130 73.0 272 91.0 850 0.11 at ARDHI
26 4000/ In Institutional
(09/2008) 3000 use 6.46 31.2 -336 0.88 83.3 6100 45.0 314 423.0 650 0.65 New (1.9.08)
27 4000/ In Institutional
(09/2008) 3000 use 6.42 32.0 -347 0.47 75.8 45.0 328 223.8 750 0.30 New (1.9.08)
28 4000/ In Institutional-
(09/2008) 3000 use 6.40 31.1 -366 1.37 83.8 48.0 320 655.4 800 0.82 New, larvae

47
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Table 26: Gas composition of inspected ARTI plants in DSM (18/19 October 2008)
G a s c o m p o s I t I o n
ARTI
CH4 CO2 O2 H2S NH3
CBS- Status
[Vol-%] [Vol-%] [Vol-%] [ppm] [ppm]
#
2 In use 68 32 0.2 72 167

3 In use 54 46 0 >100 >200

4 Not in use no gas in gasholder

5 Not in use no gas in gasholder

6 Not in use no gas in gasholder

7 Not in use no gas in gasholder

14 Not in use 80 11 0.3 >100 >200

15 Not in use 59 41 0.5 >100 >200

17 Not in use 52 48 0.2 >100 >200

18 In use 57 43 0.1 >100 >200

25 Not in use 53 33 0.5 >100 >200

30 In use 48 52 0.3 >100 >200

1 In use 59 41 0.2 >100 >200

24 In use 57 43 0.3 >100 >200

26 In use 55 45 0 >100 >200

27 In use 54 46 0 >100 >200

28 In use 56 44 0 >100 >200

4.2.2 Problems and consequences

The poor results of the first inspection tour in terms of functioning are mainly a conse-
quence of

- Lack of proper instruction by ARTI-TZ (leads to low understanding of the operator)

- Poor maintenance by the operator (leads to inappropriate operation and damage of CBS)
- Lack of follow-up service by ARTI-TZ (leads to unidentified failures of CBS)

The concrete causes for the break-down of the 8 ARTI systems include breaking of inlet
pipe, overfeeding, blocking of inlet pipe and broken gas tab. In some cases, the digester
was not in use for several months and the real cause of the failure could not be deter-
mined. In general, the operators of the CBS did not seem well informed in terms of poten-
tial feedstock, quantity of water to dilute, feedback of effluent, correction of defects, etc.
Considering the fact that even minor problems can lead to a complete standstill of the
system (e.g condense water in the gas hose), the necessity of an accurate training and
follow-up service is inevitable.

As a result of the above mentioned findings, the following improvements were strived:

- Employing an additional ARTI technician for fixing all damaged systems. He is further-
more responsible for the follow-up services, especially during the first months after the
purchase of a new CBS until the daily operation becomes habitual.
- Conducting a workshop for the ARTI technicians to teach them how to inspect a CBS
systematically by means of a check list (see Appendix B4).

48
ZHAW Bachelor Thesis 09 Christian Lohri UI05

- Creating a simple customer manual in English and Kiswahili which can be distributed as
part of the instruction training. The illustrated paper can be consulted in case of questions
about the feeding procedure and when minor problems occur (see Appendix B3).
- Encouraging the customers to call ARTI-TZ immediately in case of any difficulties so that
a complete restart of the CBS can be avoided.

4.2.3 Results of second inspection tour

The second inspection tour on 27 November disclosed that 10 CBS were functioning
whereas 2 were still not in use.

One of the CBS still not functioning is located in a remote rural area outside Dar es Sa-
laam and its digester size is 2000l. According to the operator, he prefers to sell the
kitchen waste to nearby pig farmers and therefore does not have sufficient feedstock to
operate the plant. The second CBS not in use was heavily overfed already during the first
inspection tour. According to the advice of ARTI-TZ, the feeding was interrupted in order
for the pH to rise up to at least 6, before starting regular feeding again. However, the di-
gester still showed high acidity on the second inspection tour (pH<5) and its gas was not
combustible. Limestone (previously diluted with water) could be added to increase the pH.

All other CBS were functioning and proved that the services done by ARTI-TZ following
the first inspection were successful. Points of concern on all plants remained the frequent
blocking of the gas hose by condense water and in two cases the long distance (up to
15m) between digester and burner.

49
ZHAW Bachelor Thesis 09 Christian Lohri UI05

While in 3 household digesters, a handful insect larvae was discovered, one household
CBS and one digester at Azania Secondary School displayed a large population of insect
larvae (Figure 39).

Figure 39: Insect larvae found in digester of AZANIA Sec. Scool

A closer look at the Zoology Department (DSM University) revealed that

- 80-90% of the larvae (and pupae) found in the digester were species of the Psychodidae
family (engl. sewage flies or moth flies), which do not bite and are no serious disease
transmitter.
- 10-20% of the larvae (and pupae) were Culex, which are the most widespread mosquito
species in Dar es Salaam. They act as human nuisance and as vector of filarial parasites.
- Anopheles mosquitoes (transmitter of Malaria parasites) were not found due to the high
organic pollution of the digester liquid. According to Marquardt et al. (2000), Anopheles
larvae occur in a wide range of habitats but most species prefer clean, unpolluted water.

50
ZHAW Bachelor Thesis 09 Christian Lohri UI05

5 Discussion

5.1 Assessment
To guarantee a transparent evaluation regarding the suitability of ARTI Compact biogas
systems as a treatment option for organic solid waste on household level in Dar es Sa-
laam, the assessment has been divided into the following five criteria groups.

5.1.1 Technical aspects

If operated properly, the ARTI Compact biogas system is robust in terms of structural and
biological stability.
All material necessary for the installation of a CBS is locally available. Various recom-
mendations regarding the use of material and enhanced feeding procedure can be found in
5.2.1. Longevity is not considered to impose a problem, as there is a 30-year guarantee
on the water tanks given by the manufacturer.
Blocking of the inlet pipe seems to occur fre-
quently as a result of insufficient cutting or diluting
of the feedstock. In addition, it was observed that
the moistness of the system attracts African Giant
Land Snails (Achatina fulica) which can also lead
to clogging of the inlet (Figure 40).
The gas loss through the rim between digester and Figure 40: African Giant Snail

gasholder (approximately 22%) leaves room for improvement as the emission of biogas
reduces the efficiency of the system and spreads undesired greenhouse gases.
Although the research CBS at ARDHI did not cause any problems and was regarded as
simple to operate, the inspection tours clearly proved that a proper follow-up service is
not optional but absolutely essential. This is predominantly important during the first 3
months after the purchase of a CBS, as long as the daily routine of the operator has not
been consolidated. Climatic conditions in DSM are favourable for the CBS, as average air
temperature is optimal for anaerobic process (30-35°C) and the negative influence on
process stability and gas production caused by rainfall is minimal.

Biological performance
Average gas production of the research plant when feeding FW and MW are positive and
rather high compared to literature findings (Table 27).

Table 27: Comparison of gas production from research results with findings from literature

Source Feedstock TS [%] VS [%] CH4 Gas

content [%] [Nl/kg VS]
Lohri, 2009 Food waste 24 91 57 572
Eder&Schulz, 2003 Kitchen waste 9-37 80-98 45-61 200-500
Deublein, 2008 Leftovers (Canteen) 9-37 75-98 - 400-1000

Lohri, 2009 Market waste 10 88 66 628

Eder&Schulz, 2003 Market waste 28-45 50-80 62 450
Deublein, 2008 Market waste 8-20 75-90 - 400-600

51
ZHAW Bachelor Thesis 09 Christian Lohri UI05

The average pH value of all visited CBS (in operation) was 6.5, which is at the low end of
the recommended pH range (6.5-7.5) for one-stage systems (Welliger et al., 1991, Mata-
Alvarez, 2003, Eder & Schulz, 2006).

5.1.2 Economical aspects

According to Costech (2006), the major barrier for dissemination of the biogas technology
in Tanzania is the high investment cost. This statement is also applicable for the ARTI
CBS. The monthly income of the majority of population in the city of Dar es Salaam varies
between 20,000 TZS (20 US$) and 200,000 TZS (200 US$) and the Tanzania Department
of Statistics estimates the city’s monthly per capita income at around 40 US$ (Sanga,
2003). Considering these facts, the 850’000 TZS investment cost for a household plant
limits the range of potential customers to members of the high-income class only.
It is further questionable, if the gap between material cost of a CBS (360’000 TZS) and
price of the system (850’000TZS) needs to be as high as it is or if another strategy would
be more helpful in accomplishing the goal of wide disseminating of the technology.

The biomethanization of 2kg kitchen waste (consisting of 1kg food waste and 1kg market
waste Æ 17%TS, 90%VS) in the ARTI research CBS resulted in an average of 200l biogas
(61% CH4), which in turn resulted in a burning duration of 45 min. This is approximately
1/3 of the average cooking time of 2.5h per day and family (household with 5 members).
An average household using 84kg of charcoal per month (Sanga, 2003) can therefore save
one third of charcoal (28kg), which is an equivalent of 23’000 TZS (current monthly ex-
penditures per household and month: 70’000 TZS). The amortisation period hence adds up
to approximately 3 years.

Charcoal industry is one of the largest informal sectors in Tanzania, providing employment
especially to the very poor part of society. As these people may probably not benefit di-
rectely from the ARTI technology because of its high investment costs, they might even
see a decrease in possibilities for income generating activities. The poverty reduction po-
tential is thus assessed to be very low.

As there are currently no competing products on the Tanzanian market for anaerobic di-
gestion of organic solid waste on household level, the potential market viability is re-
garded to be good. The most critical market factor is quality and reliability of installed
digesters, as mouth to mouth propaganda can also work counterproductive in case of bad
performances. Consequently, an accurate quality control service is imminent.

5.1.3 Environmental aspects

Viewed from an environmental perspective, the anaerobic digestion of organic solid waste
in an ARTI Compact biogas system bears several positive consequences:

- 83% - 92% reduction of organic load and hence less pollution through open dumps
- 30% diminution of charcoal use and therefore less deforestation
- Reduction of methane emission, if produced biogas is properly burned
- An odourless high-quality fertilizer
- A high potential for awareness-rising as it perfectly shows that waste can be looked at
and used as a valuable resource (‘from waste to value’).

52
ZHAW Bachelor Thesis 09 Christian Lohri UI05

5.1.4 Socio-cultural and political aspects

The acceptance of biogas in Tanzania is high and it even carries the perception of being
modern (Schmitz, 2007). Yet there is low awareness about waste separation, which can
result in non-biodegradable material landing inside the digester (bones, wood, plastic…)
or biodegradable matter not being fed into the digester. The day-to-day operation of a
CBS requires a high level of discipline and routine to maintain a high gas production. It is
therefore important to lay emphasis on proper instructions and facilitate the daily handling
as much as possible.
There is no explicit political support, but also no specific obstacles have been observed in
regard to biogas dissemination.

5.1.5 Safety issues

Although a theoretical risk of explosion with biogas exists, such catastrophe is not likely
to occur. Danger arises mainly in closed chambers, where a mixture of air and biogas (6-
12%) evolves. However, no explosion related to biogas utilisation has been reported in
Tanzania (Schmitz, 2003), which is probably a result of the commonly open and therefore
well ventilated rooms where digesters and burners are located. Both methane (lighter than
air) and carbon dioxide (heavier than air) are toxic gases but are not considered to im-
pose a serious threat to human health because of the airy architecture of Tanzanian build-
ings.

The hazard of attracting serious disease transmitters by the open area of digester liquid
(rim) is relatively low, but can not be completely neglected. While Anopheles mosquitoes
are unlikely to appear in digesters because of the high organic pollution, Culex mosqui-
toes will not despise this kind of breeding ground. Due to the physical construction of
mosquito larvae (Appendix C4), they need to live underneath the liquid surface to suck
oxygen through their breathing tube. Any turbulence disturbs this vitally important activity.
As flooding or flushing of breeding place prevent the breeding of mosquito larvae (Chees-
brough, 1987), stirring the rim surface sporadically with a stick can therefore diminish the
larval mosquito population.

If biogas is continuously produced but not constantly used, the capacity of the gasholder
will be exceeded. There is hardly any danger for the dome to topple over, as the gas-
holder will be slightely lifted by the pressure of the gas. The abundant biogas can thus
bubble out through the rim which is comparable to a safety pressure valve.

Although hydrogen sulphide is a toxic gas (Table 10), desulphurisation on the biogas re-
sulting from an ARTI CBS at household level is not necessary. Since CH4 and CO2 are
both odourless (but nevertheless dangerous), the appearance of H2S can even be helpful
in detecting gas leakages. Simple experiments have been done by adding a low-tech sul-
phur trap to the research plant. Yet the efficiency of the trap could not be fully measured
due to the low H2S detection range of the Dräger X-am 7000 sensor.

It is worth mentioning that a well functioning CBS does not release any bad odour and is
therefore user-friendly in this regard.

53
ZHAW Bachelor Thesis 09 Christian Lohri UI05

5.1.6 Overview of suitability assessment

Table 28 presents an overview of the previously discussed criteria and their evaluation in
terms of suitability. The criteria listed do not claim completeness or equality of impor-
tance, but demonstrate a possibility to assess a question incorporating different aspects.

Table 28: Overview of suitability assessment

Very Very
Criteria of suitability low Low .OK. Good good
-- - o + ++
y Technical aspects X
Local availability of material X
Suitability of parts used X
Biological performance X
Simplicity of operation X
Optionality for maintenance (follow-up service) X
Climatical suitability X
y Economical aspects X
Widespread affordability (investment cost) X
Savings through energy substitution X
Potential market viability X
Poverty reduction potential X
y Environmental aspects X
Reduction of organic waste X
Reduction of deforestation (charcoal use) X
Benefits from organic fertilizer X
Awareness rising potential (nutrient circle) X
Reduction of greenhouse gas emissions X
y Socio-cultural and political aspects X
Acceptance and reputation of biogas X
Awareness of waste separation X
Change of daily routine X
Political support X
y Safety issues (nuisances) X
From explosion X
From disease spreading (mosquitoes) X
From toxic effect of CH4 & CO2 X
From H2S X
Odour X
y OVERALL SUITABILITY X

54
ZHAW Bachelor Thesis 09 Christian Lohri UI05

5.2 Recommendations

5.2.1 General

The ARTI Compact biogas system (CBS) seems more feasible for households in urban ar-
eas, as organic solid waste generated in rural region are preferably used as animal feed.
Discussions with customers previous to their purchase of a CBS should emphasize on

- The realistic amount of organic waste possibly generated in their household (and the
possibility to organize additional waste in their neighbourhood).
- The approximate operational effort per day and the responsibility of the person operating
the ARTI Compact biogas system.
- The presumable output of biogas and its energy equivalent (hours of cooking).
(- The possibility of getting money refunded through well maintenance of the CBS Æ re-
bate system, see 5.2.3).

As a result of the inspection tours, the CBS seems suitable for households aiming to save
money in the long term (by replacing conventional cooking fuels). These households
should be targeted and could be reached by a more attracting (reduced) price per unit.
Most household of the high-income class however did not seem to pay much attention to
the functioning of the system as the financial factor is not essential to them.
On an institutional level, the CBS seem to be more successful, as the schools can pur-
posely appoint an employee responsible for the daily feeding of the plant. This employee
is thus interested in good maintenance and performance of the system, as his salary di-
rectly depends on it. In addition, CBS can help eco-lodges with the aim of offering an en-
vironmental friendly alternative to mass tourism to live up to their philosophy.

5.2.2 Technical

Installation
The most essential technical recommendation for improvement concerns the size of di-
gester and gasholder. A better fit between these two elements would considerably reduce
the loss of biogas to the atmosphere. While this loss is about 22% of total digester area
using SIMTANKs (see 2.3.2), AFRITANKs (1000l tank: Ø1.07m, 750l tank: Ø0.98m) would
reduce this waste area to approximately 14%. According to ARTI-TZ, negotiations with
AFRITANK are on its way. Instead of clueing parts like nipple, elbow and gas tap with Ar-
aldite, it is advisable to use solid tape. While one broken part can thereby easily be un-
screwed and replaced, broken objects which are clued together eventually require the re-
placement of the whole gasholder.
To minimize the risk of a blocked inlet-pipe, ARTI-TZ started to install a swift-elbow in-
stead of a T-connector at the lower end of the inlet-pipe. However, it is more recommend-
able to attach a 2”-ball valve to the T-connector (as it has been done on the research
plant), through which the blocking material can easily be removed.
Even properly operated CBS permanently create condense water which lead to clogging of
the gas hose on a weekly basis. Although such minor problems can easily be corrected by
removing the water, it can also lead to a complete standstill of the system if no actions
are undertaken. In Appendix C6, three suggestions for condense water drains can be
found.

55
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Daily operation
It is highly recommended to adopt the enhanced feeding procedure as described in 3.1.2
(Feeding plan, phase 2) which includes

- previous dilution of cut-up feedstock to prevent blocking

- feedback of effluent (20l) to mix bacteria with fresh influent and prevent distinctive pH
gradients inside the digester

The inlet pipe should permanently be covered to avoid entering of giant snails. A 5l-
bucket can facilitate the operators’ understanding of the required daily feeding load. Addi-
tionally, a 20l bucket with a clearly visible 10l-mark can be recommended to measure the
amount of water to dilute the feedstock with. Complaints of customers about the weak
flame of the burner can be prevented by advising them to lay half a cement brick on the
gasholder. A phone number of an ‘ARTI biogas doctor’ on the digester can help in case of
any questions.

Follow -up service

A clear after-sales service strategy should be developed. After the installation, three ser-
vices should be conducted which are included in the price of a Compact biogas system.

st
1 service (2 weeks after installation before the first feeding): Demonstration and expla-
nation of the feeding procedure (potential feedstocks, pre-treatment, dilution, feedback of
effluent, condense water removal, indication of possible problems, utilisation of effluent,
distribution of manual, encouragement to call ‘biogas doctor’ in case of questions)

nd
2 service: (3-4 weeks after 1st service): Check-up of CBS (see check-list, Appendix B4),
short interview with operator about experiences and problems

rd
3 service: (6 months after installation): Check-up of CBS, short interview with operator
about experiences and problems

All services should be documented and the date and status marked on a sticker on the
nd
digester. The 2 and 3rd follow-up services should possibly be arranged to take place
during feeding time. This allows observation and gives room for the ARTI technician to
rd
comment the procedure. After the 3 service, a sporadic inquiry by phone can give valu-
able information about the status of functioning and the satisfaction of the customer. An-
other service package over a certain period can be offered against payment.

Each time an ARTI technician visits an operating CBS, 5-10l of effluent should be col-
lected and brought back to a large demonstrational, regularly fed ‘mother plant’ at the
ARTI office. The aggregate of bacterial samples from all existing plants which are culti-
vated there act as a reservoir of diverse inoculum and can be used when setting-up a new
CBS to guarantee a potent, odourless start-up mix.

56
ZHAW Bachelor Thesis 09 Christian Lohri UI05

5.2.3 Economical

Previous to the sale of an ARTI Compact biogas system (CBS), the realistic amount of
kitchen waste generated in the household should be estimated and the customer given the
option to choose between two alternatives:

- A regular CBS (1000l) which is only fed with the daily waste generated in the household.
The output can substitute up to 1/3 of the expenses for cooking fuel.
- A larger CBS (1500-2000l), which requires the organisation of additional waste from res-
taurants, markets and/or chips sellers nearby. On one hand the investment cost is higher,
on the other hand also the biogas output and therefore the substitution of cooking fuel
rises. It should be clearly communicated that this alternative requires more logistical ef-
fort to collect the waste and more operational effort to feed the plant.

Since the price of 850’000 TZS for a household-CBS constitutes the main barrier for a
wide dissemination of the ARTI technology in Tanzania, the price strategy should be re-
considered. A rewarding rebate system could eventually motivate the users to take better
nd
care of their CBS. On the 2 and 3rd follow-up service, the customers could be awarded
with 50’000 TZS each, if their CBS is properly maintained and functioning well. One year
after the purchase of the system, another 50’000 TZS could be given back to the owner if
the plant is working without problems. If the CBS needs to be fixed on one of these three
occasions, the price of 50’000 TZS can be considered as repairing fee. Hence the actual
cost of a well maintained CBS can be reduced from 850’000 to 700’000 TZS.
The aim of this rebate system is a high percentage of smoothly operating CBS which di-
rectly helps the ARTI technology by evoking positive mouth-to-mouth propaganda.

5.2.4 Environmental

A better fit of digester and gasholder will also benefit the environment as it reduces the
emission of greenhouse gases to the atmosphere. In cases of water shortage, the feed-
stock can as well be diluted with effluent (yet inhibiting components like heavy metals will
thereby not be diluted).
One advantages, which can as well be communicated is the saving of wood and thus the
reduction of deforestation (0.9kg charcoal per day and household = 5.6kg
wood/day/household).

5.2.5 Socio-cultural and political

Since waste separation is not common in many developing countries, emphasis should be
laid on patient explanation. Moreover, concrete examples should be given to illustrate the
different kind of wastes suitable as feedstock. It is recommended to give the operational
instructions not only to the owner of the biogas plant, but particularly to the person re-
sponsible for the daily feeding.

It should be considered that every new technology which requires a change in daily habits
needs to convince its customers in the beginning, regardless to society and culture.

57
ZHAW Bachelor Thesis 09 Christian Lohri UI05

5.2.6 Safety

Attention should be paid to the location of the digester and the gas burner. The plant
should stand in direct sunlight and as close to the gas burner as possible, whereas the
burner itself should be placed in a well ventilated room. The customers should be in-
formed about the dangers of gas leakages (intoxication, explosion) and advised to check
vulnerable spots regularly by smelling. To avoid breeding of mosquitoes, it is further ad-
visable to stir the rim surface with a stick on a weekly basis.

5.2.7 Simple monitoring of ARTI-plants

A check list as suggested in Appendix B4 should guide the ARTI technicians during their
monitoring. In terms of chemical-physical parameters, pH (optimum 6.5–7.5) and digester
temperature (optimum 25-35°C) should be recorded. An interview with the operator should
give indications regarding the comprehension of the system, the satisfaction of the cus-
tomer and eventual problems that have occurred.

Table 29: Recommended equipment for monitoring of ARTI CBS

Equipment Purpose
Check-list and pen Guidance, keeping record of services
pH-Litmus stripes or pH meter (+ calibration standards) pH measurement
Lighter or matches Checking combustibility of biogas
Soap Checking air tightness (with soapwater)
Tools (screwdriver, knife, glue, seal tape, hose clips…) Repairing
Condense water traps Installation
Empty water bottles (5-10l) Collection of digester liquid
Digital Camera Picture of problems
Customers manual and contact number To be contacted in case of questions
Sticker: “checked by ARTI on (date) by (name)” Proof of service, attaching to digester

For more detailed information about the process stability and efficiency of the CBS, the
A/TIC-ratio (Appendix A2) and composition of biogas could be further analyzed. However,
these measurements require either deeper understanding of the chemical background of
anaerobic digestion (A/TIC) or expensive instruments (Dräger).

58
 ZHAW Bachelor Thesis 09 Christian Lohri UI05

References
y Books and scientific reports

Chaggu E.J., Kaseva M.E., Kassenga G.R., Mbuligwe S.E. (1998): Research and pilot
scale demonstration project on composting of domestic solid waste. Case study Sinza, Dar
es Salaam. University College of Lands and Architectural Studies (UCLAS)

nd
Cheesbrough M. (1987): Medical Laboratory Manual for tropical Countries, Vol.1, 2 Edi-
tion, ELBS (Educational Low-Priced Biooks Scheme), Cambridge

Clesceri, L.S., Greensberg A.E., Trussell R.R. (1998): Standard methods (for examination
th
of water and wastewater), 17 edition; American Public Health Association, Washington

Costech (2006): Energy resources in Tanzania, Volume 1; Tanzania Commission for Sci-
ence and Technology

Day J.A., Harrison A.D., de Moor I.J. (2003): Guides to the freshwater invertebrates of
southern Africa, Volume9: Diptera; Water Research Commission, South Africa

Deublein, D., Steinhauser A. (2008): Biogas from waste and renewable resources: An in-
troduction; Wiley-VCH-Verlag, Weinheim

Edelmann W., Engeli H. (1996): Biogas aus festen Abfällen und Industrieabwässern; Bun-
desamt für Konjunkturfragen, Pacer

Eder B., Schulz H. (2006): Biogas Praxis; Oekobuch Magnum, Staufen

Gunseelan V.N (1997): Anaerobic Digestion of biomass for methane production: a review;
Biomass and Bioenergy 13 (1997) 83-114

Gunaseelan V.N. (2004): Biochemical methane potential of fruits and vegetable solid
waste feedstocks; Biomass and Bioenergy 26 (2004), 389-399

th
Hach Company (2002): Water analysis handbook, 4 edition, Loveland

Henry R.K., Yongsheng Z., Jun D. (2006): Municipal solid waste management challenges
in developing countries-Kenyan case study, Waste Management26 (2006) 92-100

Kaseva M.E., Gupta S.K. (1996): Recycling – an environmentally friendly and income gen-
erating activity towards sustainable solid waste management. Case study- Dar es Salaam,
Tanzania; Resources, Conservation and Recycling 17 (1996) 299-309

Kaseva M.E., Mbuligwe S.E (2005): Appraisal of solid waste collection following private
sector involvement in Dar es Salaam city, Tanzania; Habitat International (2005) 353-366

59
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Kassenga G.R. (1997): Promotion of renewable energy technologies in Tanzania; Re-

sources, Conservation and Recycling 19 (1997) 257-263

Kassim S.M., Ali M. (2006): Solid waste collection by the private sector: Households’ per-
spective-Findings from a study in Dar es Salaam city, Tanzania; Habitat International 30
(2006) 769-780

Kossmann W., Pönitz U., Habermehl S. (): Biogas Digest Volume 1-4-Biogas basics [publi-
cation]; GTZ and ISAT

Koottatep S., Ompont M., Hwa T.J (): Biogas: Green Productivity Option for Community
Development

Lansing S., Botero R.B., Martin J.F. (2008): Waste treatment and biogas quality in small-
scale agricultural digesters; Bioresource Technology 99 (2008) 5881-5890

Lee D.H., Behera S.K., Kim J.W., Park H.S. (2008): Methane production potential of
leachate generated from Korean food waste recycling facilities: A lab-scale study; Waste
Management (2008), doi:10.1016/j.wasman.2008.06.033

Marquardt, W.C., Demaree R.S., Grieve R.B. (2000): Parasitology & Vector Biology, Har-
court/Academic Press, Burlington

Mata-Alvarez J. (2003): Biomethanization of the organic fraction of municipal solid

wastes; IWA Publishing, Cornwall

Mbuligwe, S.E. (2002): Institutional solid waste management practices in developing coun-
tries: a case study of three academic institutions in Tanzania; Resources, Conservation
and Recycling 35 (2002) 131-146

Mbuligwe S.E., Kassenga G. (2004): Feasibility and strategies for anaerobic digestion of
solid waste for energy production in Dar es Salaam city, Tanzania; Resources, Conserva-
tion and Recycling 42 (2006), 183-203

Mwakaje A.G. (2007): Dairy farming and biogas use in Rungwe district, South-west Tanza-
nia: A study of opportunities and constraints; Renewable and Sustainable Energy Reviews

Müller, Ch. (2007): Anaerobic digestion of biodegradable solid waste in low- and middle-
income countries, Sandec-report

Parrot, L., Sotamenou J., Dia, B.K. (2008): Municipal solid waste management in Africa:
Strategies and livelihoods in Yaoundé, Cameroon; Waste Management (2008)
doi:10.1016/j.wasman.2008.05.005

Sanga G.,A. (2003): Evaluation of impact of clean technologies and substitution of cooking
fuels in urban residences in Tanzania; Academic Master’s degree dissertation, Faculty of
Mechanical Engineering, Campinas State University, Brasil

60
 ZHAW Bachelor Thesis 09 Christian Lohri UI05

Schillig F., Tijmensen M. (2004) Biogasanlagen. in: Bioenergieanlagen - Planung & Instal-
lation. Hrsg: DGS Deutsche Gesellschaft für Sonnenenergie e.V.

Schmitz T.D., (2007): Feasibility study for a national domestic biogas programme in Tan-
zania; published by GTZ, Eschborn

Sheya M.S., Mushi S.J.S. (2000): The state of renewable energy harnessing in Tanzania;
Applied Energy 65 (2000) 257-271

SNV (2008): Biogas for better life- a Tanzanian Initiative, Draft Programme Implementa-
tion Document fo a national programme on Domestic Biogas Dissemination in Tanzania;
August 2008

Troschinetz A.M., Mihelcic J.R. (2008): Sustainable recycling of municipal solid waste in
developing countries; Waste Management (2008), doi:10.1016/j.wasman.2008.04.016

Wellinger A., Baserga U. Edelmann W. Egger K., Seiler B. (1991): Biogas Handbuch; Ver-
lag Wirz, Aarau

Yhdego M. (1995): Urban solid waste management in Tanzania: Issues, concepts and chal-
lenges; Resources, Conservation and Recycling 14 (1995) 1-10

Zurbrügg C. (2002): Solid Waste Management in developing countries, Sandec, Dübendorf

y Internet
[1] www.wordtravels.com/images/map/Tanzania_map.jpg (Map of Tanzania), November 08
[2] www.fastonline.org/CD3WD_40/BIOGSHTM/EN/IMAGES/CAMARTEC.GIF (CAMARTEC)
September 08
[3] www.ashdenawards.org/winners/arti06 (ARTI Compact digester), July 08
[4] www.howtopedia.org/en/How_to_Build_the_ARTI_Compact_Biogas_Digestor%3F (ARTI
Compact digester), June 08
[5] www.apo-tokyo.org/gp/e_publi/biogas/BiogasGP3.pdf (Asian Productivity Organization
(APO): Biogas - A GP Option For Community Development), August 08
[6] www.arti-india.org (ARTI Compact digester), July 08
[7] www.gtz.de/de/dokumente/en-biogas-volume1.pdf (GTZ-GATE 1999: Biogas Basics)
August 08
[8] www.gtz.de/de/dokumente/en-biogas-volume2.pdf (Application & Product Develop-
ment), August 08
[9] www.gtz.de/de/dokumente/en-biogas-volume3.pdf (Costs & Benefits/Biogas Pro-
gramme), August 08
[10] www.gtz.de/de/dokumente/en-biogas-volume4.pdf (Biogas Country Reports), August
08
[11] www.hdrstats.undp.org/countries/data_sheets/cty_ds_TZA.html (Human Development
Report 2007), October 08
[12] www.tanzania.go.tz/census (Statistical figures); October 08

61
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Table of Figures
Figure 1: Uncontrolled dumping in Dar es Salaam, Tanzania ......................................... 1
Figure 2: Map of Tanzania [1] .................................................................................. 3
Figure 3: Illegally dumped waste in Dar es Salaam ...................................................... 5
Figure 4: Waste pushcart, waste truck and waste collection point in Dar es Salaam ......... 6
Figure 5: CAMARTEC fixed-dome plant [2] ................................................................. 9
Figure 6: Scheme of the biodegradation steps of complex matter (Mata-Alvarez, 2003) .. 12
Figure 7: ARTI Compact biogas plant scheme (gasholder empty and gas-filled) ............. 17
Figure 8: Experimental site .................................................................................... 19
Figure 9: Preparing of cow dung ............................................................................. 20
Figure 10: Food waste (FW, left) and market waste (MW, right)................................... 20
Figure 11: Food waste raw, mashed, minced (left to right) .......................................... 21
Figure 12: Market waste raw, manually cut, blended and diluted (left to right) ............... 21
Figure 13: Sampling points .................................................................................... 24
Figure 14: Scaling on gasholder ............................................................................. 25
Figure 15: Continuous rising of gasholder proportional to gas production ..................... 25
Figure 16: pH development inside digester during phase 1: start-up ............................ 31
Figure 17: pH development inside digester during phase 2: feeding strategy ................. 31
Figure 18: pH development inside digester during phase 3: maximum load ................... 32
Figure 19: Development of A/TIC-ratio (Nordmann) while feeding FW (day 8-32) and MW
(day 43-67)..................................................................................................... 32
Figure 20: Comparison of TS concentration in influent, digester liquid and effluent ........ 33
Figure 21: Comparison of VS concentration in influent, digester liquid and effluent ....... 33
Figure 22: Comparison of COD concentration in influent, digester liquid and effluent ..... 33
Figure 23: Stratification of digester liquid in ARDHI research plant ............................. 34
Figure 24: Results of A/TIC titrations (Kapp and Nordmann) with different pre-treatment 36
Figure 25: Reduction of waste volume (TS) when feeding FW (day 8-32) & MW (day 43-67)
.................................................................................................................... 39
Figure 26: Reduction of VS content when feeding FW (day 8-32) and MW (day 43-67) .... 39
Figure 27: Reduction of COD content when feeding FW (day 8 -32) & MW (day 43 - 67).. 40
Figure 28: Comparison of COD solid and COD dissolved from FW and MW.................... 40
Figure 29: Input/Output with daily load of 2kg/d Food waste ....................................... 40
Figure 30: Input/Output with daily load of 2kg/d Marke twaste .................................... 40
Figure 31: Comparison of daily gas production from FW and MW during 10 days of steady-
state conditions .............................................................................................. 41
Figure 32: Gas production while increasing daily feed:MW (day 64-76), FW (day 80-100) 42
Figure 33: Gas production in dependence with daily feeding load ................................ 42
Figure 34: Average gas composition from 2kg/d FW (left) and MW (right), H2S >100ppm
and NH3 >200ppm are not shown ....................................................................... 43
Figure 35: Gas stove attached on research CBS for burning tests ............................... 43
Figure 36: Weight applied on gasholder ................................................................... 44
Figure 37: Concentration of Ptotal and PO4 in effluent while feeding FW (day 8-32) and
MW (day 43-67)............................................................................................... 44
Figure 38: Ammonium-Nitrogen difference between influent and effluent ...................... 45
Figure 39: Insect larvae found in digester of AZANIA Sec. Scool ................................. 50
Figure 40: African Giant Snail ................................................................................ 51
62
ZHAW Bachelor Thesis 09 Christian Lohri UI05

List of Tables
Table 1: Solid waste generation in DSM city ............................................................... 4
Table 2: Composition of household waste in DSM ....................................................... 5
Table 3: Quantities of waste disposed of [t/d] by different methods in TZ, 1997............... 6
Table 4: Primary energy consumption in TZ 1999 by source (Schmitz, 2007): .................. 7
Table 5: Household (HH) energy sources and utilization (Mwakaje, 2007) ....................... 7
Table 6: Preferences in the use of cooking fuels in DSM in 2001 (Sanga, 2003) .............. 8
Table 7: The maximum gas yield per kg VS of different organic waste substrates (+cow
manure) found in literature. .............................................................................. 13
Table 8: Typical biogas components from organic fraction of MSW (Mata-Alvarez, 2003) 13
Table 9: General features of biogas (Deublein, 2008) ................................................ 13
Table 10: Effects of H2S (adopted from Eder & Schulz, 2006) ..................................... 14
Table 11: ARTI-CBS installed by 30 November 2008 .................................................. 17
Table 12: Costs of the ARTI-biogas plants (November 2008) ...................................... 18
Table 13: List of materials and prices used for the installation of ARTI plant ................ 19
Table 14: Feeding plan of the ARTI research biogas plant at ARDHI University ............. 23
Table 15: Analysed feedstock parameters (TS, VS) and diluted influent parameters (all) 23
Table 16: Analyzed effluent parameters ................................................................... 26
Table 17: TS and VS of layers inside the digester (26.11.2008) .................................. 34
Table 18: VFA, alkalinity and A/TIC results applying different pre-treatment methods
(titration according to Kapp) ............................................................................. 35
Table 19: VFA, alkalinity and A/TIC results (titration according to Nordmann) ............... 35
Table 20: Mean values during 10-day steady state conditions for both substrates .......... 41
Table 21: Gas production in dependence to daily wet weight, TS, VS and COD for both
substrates (transformed to norm conditions) ........................................................ 41
Table 22: Tests of burning duration for 100l biogas without weight on gasholder ........... 43
Table 23: Tests of burning duration for 100l biogas with weight on gasholder ............... 44
Table 24: Heavy metal concentrations in effluents .................................................... 45
Table 25: Overview of effluent results from first ARTI inspection tour (18/19 October 2008)
...................................................................................................................... 1
Table 26: Gas composition of inspected ARTI plants in DSM (18/19 October 2008) ........ 48
Table 27: Comparison of gas production from research results with findings from literature
.................................................................................................................... 51
Table 28: Overview of suitability assessment ........................................................... 54
Table 29: Recommended equipment for monitoring of ARTI CBS ................................. 58

63
ZHAW Bachelor Thesis 09 Christian Lohri UI05

Appendix
CONTENT OF APPENDIX

A Laboratory
A1 Detailed methodologies (TS, VS, COD, Ptot, PO4, NH4-N, Ntot, A/TIC (Nordmann))
A2 Titration methodology according to Kapp for monitoring of anaerobic digestion:
VFA, alkalinity and A/TIC-ratio
A3 Results: Data sheets

B ARTI
B1 List of installed ARTI-plants
B2 Documentation of Installation
B3 ARTI customer manual
B4 ARTI Service checklist
B5 ARTI results of inspection
B6 ARTI Questionnaire
B7 ARTI CBS at Azania Sec. School, DSM

C Various
C1 Rocking displacement gasmeter (scheme)
C2 Positive displacement (scheme)
C3 Biogas-Versuchsanlage in Trubschachen
C4 Invertebrates Identification key / Mosquito information
C5 Compost Quality Standards
C6 Condense water drains

D Presentation