Modelamiento y simulación dinámica

del proceso de digestión anaerobia

Oscar Dario Yepez Ceron

Universidad Nacional de Colombia

Facultad de Ingenierı́a y Arquitectura, Departamento de Ingenierı́a Quı́mica
Manizales, Colombia
April, 2021
Dynamic modeling and simulation of
the anaerobic digestion process

Oscar Dario Yepez Ceron

Thesis or degree project presented as a partial requirement to qualify for the title of:
Maestrı́a en ingenierı́a - Ingenierı́a quı́mica

Director:
Ph.D. Oscar Andrés Prado-Rubio

Co-director:
Ph.D. Beatriz Helena Aristizábal Zuluaga

Research area:
Bioprocess design and control
Grupo de Investigación en Aplicación de Nuevas Tecnologı́as, GIANT
Grupo de Trabajo Académico en Ingenierı́a Hidráulica y Ambiental, GTAIHA

Universidad Nacional de Colombia

Facultad de Ingenierı́a y Arquitectura, Departamento de Ingenierı́a Quı́mica
Manizales, Colombia
April, 2021
4

“Focus your wit and will”

“Hope is the greatest ally”
 5

Agradecimientos
Quiero agradecer a mi padre Hector y a mi madre Marleny por apoyarme, ser mi ejemplo y
enseñarme que la nobleza, la honestidad y la humildad son los valores que hacen grande a
una persona. A mis hermanos Mario y Laura, gracias por el cariño, los ánimos y por con-
siderarme su modelo a seguir. Todo esto es por y para ustedes. Sepan que cada cosa que
hago es para hacerlos sentir orgullosos. Igualmente quiero agradecer a mi Director de tesis,
el profesor Oscar Andrés Prado, una persona a la que admiro y respeto demasiado. Gracias
por su exigencia, su paciencia, confianza, y a las enseñanza académicas y humanas que me
brindo en este proceso. A mi Co-directora de tesis, la profesora Beatriz Aristizabal. Una per-
sona admirable que agradezco por el constante seguimiento, apoyo, motivación, confianza,
comprensión en mi trabajo y por todas las enseñanzas inculcadas. A los dos, muchas gracias
por hacerme un mejor profesional, y aún más, una mejor persona.

A mis dos amigos de toda la Universidad, Carlitos y Luna. Gracias por ser indispensables,
por el apoyo académico y emocional, por ser de verdad mis parceros, por los consejos, re-
gaños, paciencia y disposición. Me alegro saber que todos lo hemos conseguido. A Lau, una
persona que se convirtió en mi parcera. Gracias por los consejos, ayuda académica y sobre
todo por las risas y su comprensión. A Sophi, por enseñarme que la sinceridad siempre es la
mejor elección, por los buenos momentos y en especial, por recordarme que todo lo puedo y
nunca dudar de mı́. A Xime, una de las personas más nobles, sensatas, cariñosas e incondi-
cionales que he conocido. Gracias por todo el apoyo académico y moral en este proceso. A
mis compañeros y amigos: Ebti, Jei, Luisito, viejo Julio, Miguelin, Jhonny, Laura M, Ana,
Camilo Z, Santiago B. Gracias por el apoyo y los buenos momentos compartidos. A mis
amigos S.P. por el apoyo, por ser indispensables y por todos estos años de amistad.

Agradezco a la Universidad Nacional de Colombia-sede Manizales. A la Fundación Ceiba con

el programa Bécate Nariño por el apoyo financiero. A la colaboración de los laboratorios de
Procesos Productivos a cargo de la profesora Maria Fanny, Laboratorio de Aprovechamiento
de Residuos a cargo del profesor Andrés Felipe Rojas, al Laboratorio de Calidad del Aire, al
grupo de Investigación en Aplicación de Nuevas Tecnologı́as, GIANT y Grupo de Trabajo
Académico en Ingenierı́a Hidráulica y Ambiental, GTAIHA, ası́ como también al apoyo de
Wilmar, Juan, Jhon y Offir.
 6

Resumen
Modelamiento y simulación dinámica del proceso de digestión anaerobia

El principal modelo productivo que tiene actualmente la sociedad se basa en producir-

consumir-descartar. Este consiste en la extracción de materias primas que se suelen utilizar
una sola vez, y cuando se desechan, no se vuelven a usar. Este sistema se conoce como
economı́a lineal y funciona bajo el principio de que todo lo fabricado tiene un final y acaba
saliendo del ciclo de producción. Sin embargo, existe otro modelo, la economı́a circular, y
se basa en un proceso sostenible, donde los materiales y recursos se mantienen en el ciclo
productivo y en la economı́a durante el mayor tiempo posible sin perder su valor y con-
servando su vida útil. Una de las tecnologı́as que implementa la filosofı́a de la economı́a
circular es el proceso de digestión anaerobia. Es un proceso en el que microorganismos des-
componen materia orgánica para producir energı́a renovable en forma de metano y productos
plataforma de alto valor agregado. El desarrollo y estudio constante de esta tecnologı́a une
aspectos ambientales y energéticos, promoviendo la implementación de una economı́a circu-
lar, mitigando el impacto negativo en el medio ambiente, la preservación de los recursos y el
desarrollo económico y social.

Esta tesis, a través de ensayos experimentales y modelado y simulación computacional, inves-

tiga escenarios de producción de metano y compuestos de valor agregado utilizando digestión
anaerbia de residuos orgánicos locales. Desde una perspectiva experimental, se evaluó el Po-
tencial Bioquı́mico de Metano (PBM) de residuos de alimentos del Departamento de Nariño,
Colombia. El rendimiento de metano se determinó en condiciones mesófilas, monitoreando
las variables del proceso como el pH, demanda quı́mica de oxigeno soluble (DQOs), la pro-
ducción de biogás y la calidad del metano a través de un ensayo de medición de metano
en lı́nea y un ensayo con botellas destructivas. Los resultados de PBM de los residuos de
alimentos con una relación C/N de 12.81, mostraron una producción de biogás de 306 mL
gSV−1 , una concentración de metano del 73 % y un rendimiento especı́fico de metano de 251
mL CH4 gSV−1 en 31 dı́as. Asi mismo, la remoción de DQOs fue del 62 % y el pH se mantuvo
dentro del rango operativo óptimo para la digestión anaerobia (6.5-7.5).

Después, se implementó un estudio tecnoeconómico para comparar la producción de biome-

tano y la producción de Ácidos Grasos Volátiles (AGV) a partir de la digestión de residuos
de alimentos en un contexto Colombiano. Para evaluar la viabilidad económica se utilizó
el ı́ndice de Valor Presente Neto (VPN) con un modelo económico riguroso, utilizando va-
riables de entrada como: costos de construcción, costos de logı́stica y servicios, cantidad de
sustrato disponible, tasa de inflación, precios de venta del producto, calidad de metano en-
contrada en ensayos los de PBM, entre otros. Esta evaluación económica permitió encontrar
que el proceso de digestión anaerobia orientado a la producción de AGV puede ser 11 veces
más lucrativo que la generación de biometano. A su vez, el biometano podrı́a reemplazar y
 7

satisfacer hasta el 15 % (545455 m3 biometano año−1 ) del consumo de gas natural en el sec-
tor de transporte y el 5 % (378 ton de AGV año−1 ) de las exportaciones de AGV de Colombia.

Finalmente, se propuso un caso de estudio para investigar la bioproducción óptima de AGV

bajo perturbaciones de entrada. Anaerobic Digestion Model 1 (ADM1) se utilizó para di-
señar e implementar la optimización de la estructura de control basada en una estrategia
“override”, donde el pH fue la variable controlada. Un algoritmo de optimización en tiempo
real encontró los mejores escenarios de producción bajo perturbaciones de entrada. La im-
plementación de la optimización en tiempo real en un control de lazo cerrado de digestión
anaerobia permitió mejorar la producción y los ingresos por ventas de AGV, lo que se evi-
denció en un aumento de 8771 USD dı́a−1 respecto al proceso nominal, y un aumento del
7 % cuando el sistema fue perturbado en condiciones desfavorables, con respecto al punto de
operación.

Los resultados obtenidos de los ensayos experimentales y de la simulación y control del mo-
delo matemático propuesto, mostraron el potencial de producción de metano a partir de
residuos de alimentos en Colombia, ası́ como la rentabilidad económica de la producción
de AGV como una nueva perspectiva de la digestión anaerobia. Este estudio contribuye a
la transferencia de conocimiento del proceso, se promueven prácticas para la economı́a sus-
tentable y se presentan oportunidades de obtener productos de valor agregado y a su vez,
demostrar el potencial económico de productos no convencionales en digestión anaerobia.

Palabras clave: Digestión anaerobia; Residuos de alimentos; ADM1; Modela-

miento y simulación; Análisis tecnoeconómico.
 8

Dynamic modeling and simulation of the anaerobic digestion process

Abstract
The main productive system that society has is to produce-consume-discard. It consists of
extracting raw materials, which are usually used only once and when discarded, they are not
used again. This approach is known as linear economy and works under the principle that
everything manufactured has an end and ends up leaving the production cycle. However,
there is another approach, the circular economy it is based on a sustainable process, where
materials and resources are to be maintained in the production cycle and in the economy,
for as long as possible, without losing its value and conserving its lifespan. One of the tech-
nologies that implements the philosophy of the circular economy is the anaerobic digestion
process. A process by which microorganisms break down organic material to produce renewa-
ble energy in the form of methane and high value-added platform products. The development
and constant study of this technology unites environmental and energy aspects, promoting
the implementation of a circular economy, mitigating negative impact on the environment,
resource preservation and economic and social development.

This thesis, through experimental work and computational modeling and simulation, inves-
tigates methane and value-added compounds production scenarios using anaerobic digestion
from local organic waste. From experimental perspective, Biochemical Methane Potential
(BMP) of domestic food waste, Colombia was evaluated. Methane yield was determined un-
der mesophilic conditions, monitoring process variables such as pH, soluble chemical oxygen
demand (CODs), biogas production, and methane quality through a on-line methane mea-
surement assay and destructive bottle assay. The BMP results of the anaerobic digestion
of food waste with a C/N ratio of 12.81, showed a biogas production of 306 mL gVS−1 , a
methane concentration of 73 %, and a specific methane yield of 251 mL CH4 gVS−1 in 31
days. Likewise, CODs removal was 62 % and the pH remained within the optimal operating
range for the anaerobic digestion (6.5-7.5).

Next, a techno-economic study was implemented to compare the biomethane production and
Volatile Fatty Acids (VFAs) production from the digestion of food waste in a Colombian
context. To evaluate the economic viability, the Net Present Value (NPV) index with a ri-
gorous economic model was used, using as inputs: construction costs, logistics and services
costs, the amount of substrate available, inflation rate and product selling prices, methane
quality found in BMP assays, among others. This economic evaluation allowed to find that
the anaerobic digestion process oriented towards VFAs production can be 11 times more lu-
crative than biomethane generation. In turn, biomethane could replace and satisfy as much
as 15 % (545455 m3 biomethane year−1 ) of natural gas consumption in the transport sector
and 5 % (378 Ton VFAs year−1 ) of VFAs exports in Colombia.
 9

Finally, case study was proposed to investigate optimal bio-production of VFAs under input
disturbances. The Anaerobic Digestion Model 1 (ADM1) was used to design and implement
a optimizing control structure based on an override strategy where pH is the controlled va-
riable. A real-time optimization algorithm found the best production scenarios under inlet
disturbances. The implementation of a real-time optimization layer in a closed-loop control
of anaerobic digestion allowed to enhance the production and incomes from sales of VFAs.
This was evidenced in an increase of 8771 USD day−1 concerning to the nominal process,
an 7 % increase when the system was disturbed in unfavorable conditions, concerning to the
point of operation.

The results obtained from the experimental assays and from the simulation and control of
the proposed mathematical model, showed the potential of methane production from food
waste in Colombia, as well as the economic profitability of the VFAs production as a new
perspective of anaerobic digestion. This study contributes to the transfer of knowledge of
the process, practices for sustainable economy are promoted, opportunities are presented to
obtain scenarios to obtain other value-added and in turn demonstrate the economic potential
of non-conventional products in anaerobic digestion.

Keywords: Anaerobic digestion; Food waste; ADM1; Techno-economic analysis;

Modeling and simulation.
 10

Products
Journal papers
Oscar Dario Yepez-Ceron, Beatriz Aristizabal, Oscar Andrés Prado-Rubio, 2021. Eva-
luation of anaerobic digestion beyond biogas. Journal of Environmental Chemical En-
gineering. Status: submitted.

Oscar Dario Yepez-Ceron, Beatriz Aristizabal, Oscar Andrés Prado-Rubio, 2021. Real-
time optimization for enhanced closed-loop control of anaerobic digestion for VFAs
production under disturbances. Status: paper under development 80 %.

Academic awards
Oscar Dario Yepez-Ceron, Beatriz Aristizabal, Oscar Andrés Prado-Rubio, 2020. Po-
tential Volatile Fatty Acids production through controlled Anaerobic Digestion. Euro-
pean Biosolids Bioresources VIRTUAL Conference, 24th – 25th November. AquaEnvi-
ro, Bristol-England. WINNER OF THE POSTER AWARD.

Project funding
This work was carried out with the scholarship project: “Formación del talento humano
de alto nivel para el fortalecimiento de necesidades estratégicas en ciencia, tecnologı́a
e innovación en el departamento Nariño” by Fundación Ceiba and Bécate Nariño pro-
gram. 2019-2020.

Peer reviewed conference papers

Oscar Dario Yepez-Ceron, Beatriz Aristizabal, Oscar Andrés Prado-Rubio, (2020).
Valoración de la Digestión Anaerobia más Allá del Biogás. In Proceedings of “XLI
Encuentro Nacional de la AMIDIQ” (ISBN: en trámite). Marı́a del Rosario Enrı́quez
Rosado (Editor). Pages: BIO-327-332. Academia Mexicana de Investigación y Docencia
en Ingenierı́a Quı́mica (AMIDIQ).

Conference Presentations
Conference oral presentation

Oscar Dario Yepez-Ceron, Beatriz Aristizabal, Oscar Andrés Prado-Rubio, 2019. Eva-
luación económica del proceso de digestión anaerobia para la producción de biogás y
productos plataforma. VI Feria Ambiental, 29th and 30th October, Manizales, Colombia.
 11

Conference poster presentation

Oscar Dario Yepez-Ceron, Beatriz Aristizabal, Oscar Andrés Prado-Rubio, 2020. Po-
tential Volatile Fatty Acids production through controlled Anaerobic Digestion. Euro-
pean Biosolids Bioresources VIRTUAL Conference, 24th – 25th November. AquaEnvi-
ro, Bristol-England.

Oscar Dario Yepez-Ceron, Beatriz Aristizabal, Oscar Andrés Prado-Rubio, 2020. VA-
LORACION DE LA DIGESTIÓN ANAEROBIA MÁS ALLÁ DEL BIOGÁS. XLI
Encuentro Nacional de la Academia Mexicana de Investigación y Docencia en Inge-
nierı́a Quı́mica A.C. (AMIDIQ), 22th − 24th October. Ixtapa Zihuatanejo, México.

Oscar Dario Yepez-Ceron, Beatriz Aristizabal, Oscar Andrés Prado-Rubio, 2019. Eva-
luación económica del proceso de digestión anaerobia para la producción de biogás y
productos plataforma. 8 Congreso Internacional de Desarrollo Sostenible y Medio Am-
biente. Cambio climático en las perspectivas de los Objetivos de Desarrollo Sostenible
- ODS, 22th ,23th and 24th Octubre, Manizales-Colombia.

Oscar Dario Yepez-Ceron, Andrea Jaramillo-Carmona, Rosa Camila-Parra, 2019. Pro-

ducción de biogás como aprovechamiento de residuos de alimentos en el departamento
de Caldas. 8 Congreso Internacional de Desarrollo Sostenible y Medio Ambiente. Cam-
bio climático en las perspectivas de los Objetivos de Desarrollo Sostenible - ODS,
22th ,23th and 24th Octubre, Manizales-Colombia.
Content
Abstract 11

Products 11

1 Introduction 14
1.1 Hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.2 Thesis objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.2.1 General objective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.2.2 Specific objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.3 Thesis content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

2 Anaerobic Digestion Process 20

2.1 Fundamentals of the anaerobic digestion process . . . . . . . . . . . . . . . . 21
2.2 Microorganisms involved in the anaerobic digestion process . . . . . . . . . . 24
2.3 Types of anaerobic biodigesters . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.4 Influence of operation conditions . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.4.1 Anaerobic process temperature . . . . . . . . . . . . . . . . . . . . . 28
2.4.2 pH and alkalinity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.4.3 Total and volatile solids . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.4.4 Carbon/Nitrogen ratio . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.4.5 Food waste as substrate . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.4.6 Retention Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.5 Inhibition and toxicity effects . . . . . . . . . . . . . . . . . . . . . . . . . . 34
2.6 Alternative products from the anaerobic digestion process . . . . . . . . . . . 37
2.6.1 Volatile Fatty Acids . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.6.2 Digestate or biol as organic fertilizer . . . . . . . . . . . . . . . . . . 42
2.7 Benefits of the anaerobic digestion process . . . . . . . . . . . . . . . . . . . 43
2.7.1 Health benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
2.7.2 Environmental benefits . . . . . . . . . . . . . . . . . . . . . . . . . . 43
2.7.3 Social Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
2.8 Case study: Food waste from the Nariño region as a substrate for anaerobic
digestion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
2.9 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
2.9.1 Substrate sampling: Food Waste . . . . . . . . . . . . . . . . . . . . . 48
Content 13

2.9.2 Food Waste characterization . . . . . . . . . . . . . . . . . . . . . . . 49

2.9.3 Inoculum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
2.9.4 Biochemical Methane Potential (BMP) . . . . . . . . . . . . . . . . . 51
2.10 Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
2.10.1 Characterization of food waste and inoculum . . . . . . . . . . . . . . 56
2.10.2 Biochemical methane potential of food waste . . . . . . . . . . . . . . 59
2.10.3 Theoretical Methane Yield . . . . . . . . . . . . . . . . . . . . . . . . 64
2.11 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

3 Techno-economic evaluation of anaerobic digestion beyond biogas 75

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
3.2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
3.2.1 Economic potential evaluation . . . . . . . . . . . . . . . . . . . . . . 80
3.3 Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
3.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
3.5 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

4 Modelling and simulation of the anaerobic digestion process 102

4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
4.2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
4.2.1 Mathematical implementation of ADM1 and verification . . . . . . . 107
4.2.2 Optimal operating point determination . . . . . . . . . . . . . . . . . 108
4.2.3 Control structure design . . . . . . . . . . . . . . . . . . . . . . . . . 109
4.2.4 Hierarchical control structure: real-time optimization and closed-loop
control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
4.3 Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
4.3.1 Implementation and validation of ADM1 model . . . . . . . . . . . . 111
4.3.2 Operating point: VFAs production enhance . . . . . . . . . . . . . . . 112
4.3.3 Control structure design . . . . . . . . . . . . . . . . . . . . . . . . . 115
4.3.4 Hierarchical control structure: real-time optimization layer for closed-
loop control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
4.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

5 Conclusions and Recommendations 128

5.1 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
5.2 Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
1 Introduction
About 80 % of global energy consumption comes from fossil fuels, which are considered the
main source of acidifying pollutants and greenhouse gases that accelerate global warming
and climate change (Jimenez et al., 2015). Hence, the importance of studying and evaluating
new unconventional technologies to produce sustainable and self-sustaining energy without
causing damage to the environment, following the philosophy of a circular economy. In a
circular economy, unlike the linear economy, resources are regenerated within the biological
cycle, recovered, and restored thanks to the technical cycle (Macarthur, 2006). Renewable
energy that follows this philosophy, with great potential and that currently has great rele-
vance worldwide, is the production of biogas through a biological process called anaerobic
digestion.

Anaerobic digestion is a technology developed for the degradation of organic waste within
biodigesters or anaerobic digesters. Four main steps for degradation occur in these: hydroly-
sis, acidogenesis, acetogenesis and methanogenesis, which, due to the action of different
microorganisms, organic substrates such as food residues, pruning residues, domestic garba-
ge, solid manure, sewage, agro-industrial residues, etc., are converted into energy in the form
of biogas containing mainly methane (main component for energy production) and carbon
dioxide (Guo et al., 2017).

Food waste is one of the most promising organic waste suitable substrate for anaerobic di-
gestion due to its high energy content and its availability is relatively vast. The anaerobic
digestion process is a proven and effective solution for the treatment and valorization of
food waste (Zhang et al., 2014). In Colombian context, the characterization of food waste is
limited, as well as its study in the anaerobic digestion process. Cadavid-Rodrı́guez L.S. and
Bolaños-Valencia (2015), Guerrero Pinzón and Delgadillo Mirquez (2016), Solarte Toro et al.
(2017), present some of the few studies with these local wastes for the production of biogas.
The physico-chemical characterization of these wastes is arduous, however, it allows to know
the potential for the generation of renewable energy, along with technology development for
waste management.

Additionally, an alternative approach to the anaerobic digestion process is to target the pro-
duction of Volatile Fatty Acids (VFAs): acetic acid, propionic acid, butyric acid and valeric
acid. The direct recovery of these digestion products (which are produced in the acidogenesis
1.1 Hypothesis 15

process) or further processing to obtain other molecules (for example, polyhydroxyalkanoates

(PHAs) or medium chain length fatty acids), can result in the production of end products
with more added value than the conventional product, biogas (Kleerebezem et al., 2015).

Although the application of anaerobic digestion to degrade organic compounds and produ-
ce biogas has been successful (in addition to being a well-established process), the optimal
design of anaerobic digesters for maximum methane production, and even for production of
VFAs, remains a challenge. The above due to the variability of the substrate and its weak de-
composition, the complexity of microbial consortium, destabilization, low biogas production,
complicated biochemical, physical and chemical interactions involved in the processes and
inhibitions that can destabilize the process (Donoso-Bravo et al., 2011; Yu et al., 2018). To
address these drawbacks, increased efforts to optimize the process performance are crucial.
One way to do this issue can be through the simulation and mathematical modeling of the
anaerobic process (Manjusha and Beevi, 2016).

Recently, several mathematical models of AD have been proposed and in turn they have been
implemented in different computer platforms for their simulations. The review document by
Batstone et al. (2015) covers the latest studies and developments of mathematical models
of the anaerobic digestion process that describe this process, being the ADM1 model (An-
aerobic Digestion Model No 1) (Batstone et al., 2002) more complete and widely used. The
use of this model allows a better understanding of the process, optimizing the production of
biogas and alternative products and as a platform for the implementation of a control system.

In order to study the anaerobic digestion process for the production of biogas and VFAs, this
study evaluates the potential of the production of biogas (biomethane) from local organic
residues using Biochemical Methane Potential (BMP) assays. Compare through a rigorous
economic model and the Net Present Value (NPV) index, the economic viability of the pro-
duction of biomethane and VFAs in a local context. Finally, implement a closed loop control
system, using ADM1 as a platform, to identify variables and conditions processes that can
improve VFAs production, using a real-time optimization layer to ensure the highest profit
from VFA sales.

1.1. Hypothesis
Upon completion of the proposed investigation, the following hypotheses are expected to be
verified:

It is possible to determine design and operation limitations of the anaerobic digestion

process through rigorous dynamic modeling and simulation of the system.
 16 1 Introduction

It is possible that the anaerobic digestion process together with the production of other
compounds is a techno-economically viable option in a Colombian context.

Design limitations are related to operating conditions (i.e. temperature, pH, input substrate
flow, soluble compounds and substrate particulates concentration).

1.2. Thesis objectives

1.2.1. General objective
Investigate biogas production scenarios and value-added compounds using anaerobic
digestion from local organic waste.

1.2.2. Specific objectives

Propose and implement a mathematical model of anaerobic digestion using the MATLAB-
SIMULINK platform.

Determine the optimal biogas (biomethane) yield by simulating the mathematical mo-
del proposed using food waste as a substrate.

Evaluate modifications to the mathematical model focused on the generation of alter-

native products.

Note regarding specific objective 3: during the research, it was noticed that VFAs accumula-
tion was achieved using the original proposed model, when exploring the operational window
in a broader spectrum. This makes it unnecessary to include structural changes in original
model ADM1 for preliminary research of VFAs production by anaerobic digestion.

Although the thesis was planned to basically perform simulation, the scope of the project was
increased due to the formulation and funding achieved through the project: “Formación del
talento humano de alto nivel para el fortalecimiento de necesidades estratégicas en ciencia,
tecnologı́a e innovación en el departamento Nariño” by Fundación Ceiba and Bécate Nariño
program, allowed to carry out part of the experimental part and complement the research
focused on the production of methane from local organic waste.

1.3. Thesis content

This thesis is organized into chapters written in the form of a scientific article (except for the
second chapter). Due to this distribution, some information is repeated so that each chapter
is understandable to the reader. The thesis consists of the following chapters:
1.3 Thesis content 17

Chapter 2 entitled: “Anaerobic Digestion Process” describes the fundamentals of the

anaerobic digestion process and the potential for methane production from food waste
form the Nariño region, Colombia, is evaluated through Biochemical Methane Potential
(BMP) assays.

Chapter 3 entitled: “Economic evaluation of the anaerobic digestion process” presents

a techno-economic comparison of the products of the anaerobic digestion process: bio-
methane and volatile fatty acids (VFAs), from waste in Colombian context, through a
rigorous economic model, applying the Net Present Value (NPV) index.

Chapter 4 entitled: “Modeling and simulation of the anaerobic digestion process” in-
cludes simulation of Anaerobic Digestion Model No 1 (ADM1) as a platform for the
implementation of a control structure with real-time optimization (RTO) algorithm for
enhance the VFAs bio-production under inlet disturbances.

Chapter 5 entitled: “Conclusions and recommendations” shows an overview of the con-

tributions to knowledge made in this thesis together with proposed recommendations
for future research.
References
Batstone, D. J., Keller, J., Angelidaki, I., Kalyuzhnyi, S., Pavlostathis, S., Rozzi, A., Sanders,
W., Siegrist, H., and Vavilin, V. (2002). Anaerobic digestion model no. 1, volume 1.

Batstone, D. J., Puyol, D., Flores-Alsina, X., and Rodrı́guez, J. (2015). Mathematical mo-
delling of anaerobic digestion processes: applications and future needs. Rev Environ Sci
Biotechnol, 14:595–613.

Cadavid-Rodrı́guez L.S. and Bolaños-Valencia, I. (2015). Aprovechamiento de residuos

orgánicos para la producción de energı́a renovable en una ciudad colombiana Use of organic
waste for renewable energy production in a Colombian city. Energética, 46:23–28.

Donoso-Bravo, A., Mailier, J., Martin, C., Rodrı́guez, J., Aceves-Lara, A., and Wouwer, A. V.
(2011). Model selection, identification and validation in anaerobic digestion: A review.

Guerrero Pinzón, K. and Delgadillo Mirquez, L. R. (2016). Caracterización de residuos

de alimento como materia prima para digestión anaerobia. PhD thesis, Universidad de
Ibagué, Ibagué, Colombia.

Guo, Z., Liu, W., Yang, C., Gao, L., Thangavel, S., Wang, L., He, Z., Cai, W., and Wang,
A. (2017). Computational and experimental analysis of organic degradation positively
regulated by bioelectrochemistry in an anaerobic bioreactor system.

Jimenez, J., Latrille, E., Harmand, J., Robles, A., Ferrer, J., Gaida, D., Wolf, C., Mairet,
F., Bernard, O., Alcaraz-Gonzalez, V., Mendez-Acosta, H., Zitomer, D., Totzke, D., Span-
jers, H., Jacobi, F., Guwy, A., Dinsdale, R., Premier, G., Mazhegrane, S., Ruiz-Filippi,
G., Seco, A., Ribeiro, T., Pauss, A., and Steyer, J.-P. (2015). Instrumentation and con-
trol of anaerobic digestion processes: a review and some research challenges. Reviews in
Environmental Science and Bio/Technology, 14(4):615–648.

Kleerebezem, R., Joosse, B., Rozendal, R., and Van Loosdrecht, M. C. (2015). Anaerobic di-
gestion without biogas? Reviews in Environmental Science and Biotechnology, 14(4):787–
801.

Macarthur, E. (2006). Towards a Circular Economy. Journal of Industrial Ecology, 10(1-

2):4–8.
References 19

Manjusha, C. and Beevi, S. (2016). Mathematical Modeling and Simulation of Anaerobic

Digestion of Solid Waste. Procedia Technology, 24:654–660.

Solarte Toro, J. C., Mariscal Moreno, J. P., and Aristizábal Zuluaga, B. H. (2017). Evaluación
de la digestión y co-digestión anaerobia de residuos de comida y de poda en bioreactores
a escala laboratorio. Revista Ion, 30:105–116.

Yu, Z., Leng, X., Zhao, S., Ji, J., Zhou, T., Khan, A., Kakde, A., Liu, P., and Li, X. (2018). A
review on the applications of microbial electrolysis cells in anaerobic digestion. Bioresource
Technology, 255:340–348.

Zhang, C., Su, H., Baeyens, J., and Tan, T. (2014). Reviewing the anaerobic digestion of
food waste for biogas production. Renewable and Sustainable Energy Reviews, 38:383–392.
2 Anaerobic Digestion Process
In this chapter, as a preamble, a literature review of the fundamentals of anaerobic digestion
process was conducted, where information is collected on factors that affect the process,
microorganisms that are involved, types of bioreactors, alternative products of anaerobic di-
gestion and its benefits. Then, the Biochemical Methane Potential (BMP) of food waste sam-
ples collected from a restaurant (Nariño region, Colombia) was evaluated. Physico-chemical
characterization of the food waste was performed, including proximate analysis, ultimate
analysis, calorific value, among others. The methane yield was determined under mesophilic
conditions, using two complementary assays. The first assay determined the methane yield
using a German online biogas measurement equipment, Yieldmaster-Bluesense, and the se-
cond assay monitored process variables such as pH and soluble CODs through the sacrifice
bottle method. The results of this chapter serve to understand the anaerobic digestion pro-
cess, to identify variables that affect the process and to quantify the methane yield that food
waste can generate. In addition, this chapter shows the methane concentration obtained
through the anaerobic digestion of food waste (BMP assays), a necessary parameter for a
techno-economic analysis that evaluates the economic viability when producing biomethane
and Volttie fatty acids (VFAs).
2.1 Fundamentals of the anaerobic digestion process 21

2.1. Fundamentals of the anaerobic digestion process

Anaerobic digestion is a biological process where organic matter, in the absence of oxygen, is
degraded and converted by microorganism families to macromolecules and then in methane
and carbon dioxide, also called biogas (Toerien and Hattingh, 1969). This process occurs
in biodigesters, which are hermetic and waterproof constructions that prevent the entry of
air and supply the conditions for microorganisms to degrade organic matter. At the same
time, this process is also known as an unconventional, environmentally sustainable techno-
logy, able to convert a variety of waste: municipal solid waste (organic part), animal manure,
wastewater, and agricultural waste in the form of biogas (Horan et al., 2018).

Biogas, the main product of the anaerobic digestion process, is composed of 50 to 75 %

methane, 25 to 50 % carbon dioxide and 2 to 8 % of other gases such as nitrogen, oxygen,
and traces of gases (hydrogen sulfide (H2 S ), ammonia (NH3 ) and water vapor) (House and
Surratt, 2013). A summary of the typical composition and properties of biogas is presented
in Table 2-1. The degradation of organic matter to methane is complex and is achieved
through a sequence of reactions in four steps: hydrolysis, acidogenesis, acetogenesis, and
methanogenesis (Gavala et al., 2003).

Table. 2-1: Chemical composition of biogas, substances and component properties. (Braun,
2007)
Component Concentration Properties
CH4 50-70 % (v/v) Energy Carrier
CO2 25-50 % (v/v) Corrosive, especially in the presence of water
H2 S 0-5000 ppm Corrosive, SO2 emissions during combustion
NH3 0-500 ppm NOx emissions during combustion
N2 0-5 % (v/v) Decrease heating value
Water steam 1-5 % (v/v) Corrosive, decrease heating value

In the first stage, enzymatic hydrolysis, complex materials degrade in their monomers. In
the acidogenesis or fermentation phase, volatile fatty acids (VFAs) are generated along with
alcohols, lactic acid, CO2 , H2 , NH3 , H2 S, and new cellular material. Acetogenesis is the third
step in which acetate and molecular hydrogen are produced through anaerobic oxidation of
higher fatty acids and the conversion of propionate, butyrate, and valerate into acetate and
hydrogen. Methanogenesis, the final stage, involves the production of methane from carbon
substrates (Nguyen, 2014). An outline of the anaerobic digestion process is presented in Fi-
gure 2-1.
 22 2 Anaerobic Digestion Process

Figure. 2-1: Sketch of the reaction pathways in the anaerobic digestion process. The brown
dotted line represents the degradation of lipids in monosaccharides; the violet
dotted line represents the breakdown of monosaccharides in acetate, propionate,
butyrate, and valerate; the dashed green line represents the breakdown of amino
acids into acetate, propionate, valerate, and butyrate, and hydrogen; the black
dotted line represents the breakdown of long chain fatty acids in acetate and
hydrogen. Adapted from Manchala et al. (2017).

Biogas can be used as a fuel for cooking, as heating and, as a cogeneration fuel, for the pro-
duction of electrical energy (Guo et al., 2017). Besides the energy generated, the digestate
2.1 Fundamentals of the anaerobic digestion process 23

or biol is the main side product of the anaerobic digestion process. This has been used as
biofertilizer to provide nutrients to the plants and increase the organic composition of the
soil (Kumar and Tuohy, 2018). Also, the use of biol as a raw material for the construction
of ceramic materials has been studied, giving a new alternative for its use (Salazar, 2019).
More recently, the anaerobic digestion process has gained attention as a cost-effective and
ecological alternative for the production of VFAs, which include substances such as acetic
acid, propionic acid, butyric acid, valeric acid, lactic acid, and caproic acid (Wang et al.,
2014). As can be seen in Figure 2-1, these compounds are intermediaries for the production
of hydrogen thus for the production of biogas, being, second most limiting stage of the pro-
cess, since the accumulation of these compounds would inhibit process. Their bioproduction
from waste could become a sustainable alternative to petro and chemical synthesis of these
platform compounds.

In developed countries of Europe, North America and Asia, biogas production is highly
industrialised and large-scaled, where biogas is mainly used for energy generation in engines
(combined heat and power plants) or burners. For a technical use of biogas, calorific value,
and ignition temperature should be taken into account as some of the important physical
properties. Table 2-2 shows the expected physical properties of biogas.

Table. 2-2: Physical properties of biogas with an average composition of 60 % (v/v) CH4 ,
38 % (v/v) CO2 and 2 % (v/v) trace gas components (Braun, 2007)
Parameter Unit
Net calorific value 21.48 MJ m−3
Density 1.21 kg m−3
Ignition temperature 700 °C
Explosive mixture lı́mits of CH4 with air 4.4-16.5 % (v/v)
Rate of flame propagation 0.25 m s−1
Air requirement for combustion 5.71 m3 m−3 biogas
Odor 500000 Odor units m−3

In this way, organic waste can be transformed into valuable products following the circular
economy philosophy. The circular economy aims to ensure that the value of products and
materials is maintained for as long as possible in the production cycle. The ability and fle-
xibility of the anaerobic digestion process to digest a large amount of organic waste, and in
turn produce a variety of products, consolidate the role of the anaerobic digestion process
in the circular economy.
 24 2 Anaerobic Digestion Process

2.2. Microorganisms involved in the anaerobic digestion

process
The anaerobic digestion process is a complex bioprocess performed mainly by bacteria, but
higher trophic groups such as protozoa and anaerobic fungi may also be present. The micro-
bial population contains many genera (types) of strictly anaerobic bacteria and facultative
anaerobic bacteria (Murphy and Thamsiriroj, 2013).

Currently, four different trophic groups are common in anaerobic processes. Figure 2-2 shows
these groups. The coordinated activity of these trophic groups as a whole ensures the stability
of the process.

Figure. 2-2: Four trophic groups involved in the anaerobic digestion process. 1. Acidogenic
bacteria. 1.1. Hydrolytic bacteria 1.2. Fermentative bacteria 2. Acetogenic bac-
teria. 3. Homoacetogenic bacteria. 4. Methanogenic bacteria. 4.1. Methanogenic
hydrogenotrophic bacteria. 4.2. Acetoclastic methanogenic bacteria. Adapted
from Murphy and Thamsiriroj (2013).
2.2 Microorganisms involved in the anaerobic digestion process 25

Acidogenic bacteria: This group consists of fermentative and hydrolytic bacteria. Hydroly-
tic bacteria break down the carbohydrate into monomers and detects soluble material
in particulate matter. Fermentative bacteria transform the resulting monomers in a
wide range of products. The final products of the acidogenic stage include acetic acid,
hydrogen and carbon dioxide. However, most of the products are volatile fatty acids
with a greater number of carbons, such as propionate, butyrate, valerate, and also
complementing substance as alcohols.

Acetogenic bacteria: This is a group of bacteria known as forced proton-reducing acetoge-

nic microorganisms. This a syntropic group, which implies that they must act together
with bacteria in a different trophic group to digest a substrate. A product of their
own metabolism is hydrogen, which is toxic to them, thus, these bacteria require to
interact with other species that can use hydrogen. These bacteria are fundamental in
the anaerobic digestion process since they convert fermentative intermediates (VFAs)
into methanogenic substrates: hydrogen, carbon dioxide, acetic acid, and uncarbonated
compounds.

Methanogenic bacteria: This group consists of hydrogenotrophic methanogenic bacteria

and acetoclastic methanogenic bacteria. The hydrogenotrophic bacteria utilise hydro-
gen which the acetogens produce. The relationship between acetogens and hydroge-
notrophic methanogenic bacteria is a good example of syntropic mutualism: bacteria
in different trophic groups converting propionate, butyrate, and long-chain acids into
methane and water.

Species of only two genera Methanosarcina and Methanothrix can produce methane
from acetic acid and are referenced to as aceticlastic. Hydrogenotrophic methanogens
have relatively fast growth rates, which make them very efficient. The final product is
methane and the contribution of the hydrogenotrophic pathway is from 27 to 30 % and
by the acetoclastic route, it is 70 %.

Homoacetogenic Bacteria: They are strict bacteria that catalyze the formation of acetate
from carbon dioxide and hydrogen. Hydrogen-consuming acetogens appear to be out-
competed by methanogens for hydrogen. The net result, however, is the maintenance
of low hydrogen partial pressures and increased significance of acetate as an immediate
methane precursor.
 26 2 Anaerobic Digestion Process

2.3. Types of anaerobic biodigesters

There are several types of biodigesters that are currently in use for biogas production. Table
2-3 shows the characteristics and description of the main biodigesters in which the anaerobic
digestion process is carried out. Depending on the required application, the type of biodi-
gester is chosen and used. For laboratory scale studies, Anaerobic Sequential Batch Reactor
(ASBR) biodigesters are the most appropriate. The main considerations for the biodigesters
design are substrate type, phase, product inhibition, bioenergy recovery, and mass transfer
limitations (Uçkun Kiran et al., 2016).

There are three main biodigesters groups. The first is the discontinuous digester. These
biodigesters are fed with organic raw materials and inoculum and then, the anaerobic degra-
dation occurs. Once all the organic material has been degraded, the content of the reactor
is discharged, cleaned and a new batch is added for digestion (Kumaran et al., 2016).

The second type is the continuous stirred tank digester, where all biochemical reactions take
place in the same reactor. The substrate is fed constantly(continuously or in small batches
with a defined interval of time), and as a direct consequence, the biogas production is almost
stable. The main advantage of this biodigester is its simplicity in construction and operation
(Gonzalez-Fernandez et al., 2015). Finally, there is the two-stage (or even multi-stage) con-
tinuous feeding system, where the hydrolysis/acidogenesis and acetogenesis/methanogenesis
stage are separated in different reactors (Moo-Young, 2011).

Anaerobic biodigesters can also be classified as low rate or high-rate biodigesters. Table 2-4
shows a comparison of these types of reactors. The main difference between them is the
application. High-rate biodigesters are used for bioenergy production, while low-rate bio-
digesters are used to treat and dispose of organic waste (sewage, feces, etc.). Some examples
of low-rate biodigesters are anaerobic ponds, septic tanks and Imhoff tanks (Varnero Mo-
reno, 2011). High-rate biodigesters have been used to formulate the mathematical models
for biogas production.

On the other hand, the selection of the biodigester, the biomass retention capacity should
be considered since anaerobic microorganisms slowly grow during the metabolic generation
of methane and hydrogen. It is often essential to select a biodigester configuration that
decouples the hydraulic retention time (HRT) from the solid retention time (SRT). Such
decoupling can maintain a significantly high SRT/HRT ratio which favors a faster growth of
anaerobic microorganisms (Khanal, 2008).
2.3 Types of anaerobic biodigesters 27

Table. 2-3: Biodigesters used in the anaerobic digestion process

Biodigester Description and characteristics References

ACP is composed of a conventional anaerobic reactor

with stirring
The influent reactor is directly in contact with the an- Varnero
Anaerobic
aerobic biomass Moreno
Contact
HRT is 12 to 24 hours (2011);
Process
Khanal
(ACP) The SRT in the system is of the order of 25 to 40 days,
(2008)
producing the hydrolysis of solids and their subsequent
mechanization

It is a suspended growth system in which adequate hy-

draulic and organic loading conditions are maintained Cruz-
to facilitate the formation of dense lumps of biomass Salomón
Upflow known as granules et al.
anaerobic The diameter of the granules varies from 1 to 3 mm (2017);
sludge An extremely long 200-day SRT can be achieved with a Hulshoff
blanket- low HRT, approximately 6 h Pol et al.
(UASB) (2004);
It is ideal for the production of biogas with a high con-
Khanal
centration of methane from a feed stream with high so-
(2008)
lubility

It is a batch variation of the UASB, where a single reac-

tor is used to fill, react, sediment and decant Borja
Anaerobic
The ASBR process is recommended for bioenergy pro- (2011);
Sequential
duction from animal manure and another biowaste (was- Mao et al.
Batch
tewater, food waste) (2015);
Reactor
Khanal
(ASBR) High levels of biomass can be achieved in the reactor
(2008)
regardless of HRT

Upflow Anaerobic Filter (UAF): corresponds to a type

of tubular anaerobic reactor that operates in a conti- Meng
Anaerobic nuous and upward flow regime. Typically, HRT varies et al.
Filter from 0.5 to 4 days and the loading rate varies from 5 to (2016);
(AF) 15 kg COD m−3 Khanal
Downflow Anaerobic Filter (DAF): it is similar to the (2008)
UAF, only that it operates at a downstream flow
 28 2 Anaerobic Digestion Process

Table. 2-4: Comparison of high and low-rate biodigesters (Varnero Moreno, 2011; Fujihira
et al., 2018; Rico et al., 2017)
Low-rate biodigesters High-rate biodigesters
They are not mixed They are mixed
Conditions such as Temperature, SRT and They maintain a high level of biomass in the
others are not controlled bioreactor
Organic load rate is a low range of 1-2 kg Organic load rates vary from 5 to 30 kg
COD m−3 day−1 COD m−3 day−1 or even higher
They are not suitable for bioenergy produc- They are more appropriate for bioenergy
tion production

2.4. Influence of operation conditions

In the anaerobic digestion process, microorganisms must remain in optimal-operating condi-
tions to achieve maximum process performance. Depending on the component to be produ-
ced (biogas or VFAs), the factors that affect the process will determine the stability and the
products quality. The following is a description of the most relevant factors that affect the
process for the production of biogas, while Section 2.6.1 describes the factors that govern
the process for the production of VFAs.

2.4.1. Anaerobic process temperature

Temperature is the most important physical factor affecting the kinetics of the biochemi-
cal reactions, especially methanogenic bacteria growth (Lier et al., 1997). The temperature
ranges are classified into:

Psychrophilic: less than 20 °C

Mesophilic: between 20 °C and 40 °C

Thermophilic: greater than 40 °C

Methanogenic bacteria reach a max growth rate of 60 °C for thermophilic conditions. For
mesophilic conditions, optimal growth is between 30 and 36 °C. For the psychrophilic con-
ditions, the worst growth conditions are found, reaching only 23 % at temperatures between
13 and 18 °C. Figure 2-3 shows how growth trend of methanogenic bacteria decreases expo-
nentially when they exceed the specific temperature where the growth rate is maximum.
2.4 Influence of operation conditions 29

Figure. 2-3: The relative growth rate of psychrophilic, mesophilic and thermophilic metha-
nogens. Taken from Lier et al. (1997)

The thermophilic conditions, although the elimination of pathogenic microorganisms is hig-

her, it is usually more sensible at any change in operating conditions. In addition, it exhibits
inhibition problems due to the higher toxicity of certain compounds at high temperatures
(i.e. ammoniacal nitrogen or long-chain fatty acids). Under thermophilic conditions, biogas
production up to 30 % higher can be achieved, the biogas has more carbon dioxide concen-
tration. For this reason, mesophilic conditions are considered as the optimal condition for
the anaerobic digestion process (35-37 °C). In addition, small changes in temperature in this
regime are less sensitive and do not influence the efficiency of the process.

2.4.2. pH and alkalinity

Small variations in pH cause the anaerobic digestion process to be negatively affected, espe-
cially for methanogenic microorganisms. To maintain the activity of methanogenic archaea,
the optimum pH range should be between 6.8 and 7.5. For fermentative bacteria, the pH
range is between 5.0 and 6.0, with tolerance for pH values of up to 4.5 (López-Hernández
et al., 2017). A decrease in the pH value increases the activity of fermentative microorga-
nisms, while consumers of their (slower) products are inhibited by the increase in acidity.
The differences are one of the main operational problems in the anaerobic digestion process.

The pH value affects both biogas production and methane composition. When the pH is
below 6, the biogas has low percentage of methane, negatively affecting its energy capacity.
 30 2 Anaerobic Digestion Process

Studies suggest that for anaerobic digestion at mesophilic conditions, the optimal pH range
should be between 6.5-7.5 (Ishida et al., 1982; Van Ginkel et al., 2001; Khalid et al., 2011).
The pH is controlled primarily by the natural alkalinity as a buffer system. Each mole of or-
ganic nitrogen theoretically generates an alkalinity equivalent. Ammonia reacts with carbon
dioxide to produce ammonium bicarbonate. Bicarbonate ions (HCO− 3 ) are the main con-
tributors to the alkalinity of the system. On the other hand, under unfavorable conditions
(accumulation of VFAs), the pH also decreases if the system has a small buffer capacity
which can cause process failures due to inhibitions (Nguyen, 2014).

To maintain the pH within the optimal ranges in the biodigester, it is often necessary to
add substances that provide alkalinity to the system, such as sodium bicarbonate (preferred
due to its high solubility and low toxicity), sodium carbonate, ammonium hydroxide, gas
ammonia, lime, sodium, and potassium hydroxide.

2.4.3. Total and volatile solids

For the process to develop normally, the percentage of total solids (TS) contained in the
substrate is an important factor to account for. Mobility of microorganisms is very limited
as the amount of total solids increases, affecting efficiency and biogas production.

Experimentally, it has been shown that a load in semicontinuous digesters should not have
more than 8 to 12 % of total solids to ensure the proper functioning, unlike discontinuous
digesters, which have between 40 to 60 % of TS (Varnero Moreno, 2011). Volatile solids (VS)
indicates the amount of material that can be used as food by microorganisms and theoreti-
cally be converted to methane.

2.4.4. Carbon/Nitrogen ratio

The substrate is the main source of carbon and nitrogen to microorganisms. However, the
microbiological process also requires mineral salts (nutrients) such as sulfur, phosphorus, po-
tassium, calcium, magnesium, iron, manganese, molybdenum, zinc, cobalt, selenium, tungs-
ten, nickel and another minor. Residues such as manures and sewage sludge contribute to
these elements in an adequate proportion, however, with the use of some industrial organic
waste, the addition of these nutrients may be necessary (Varnero Moreno, 2011).

The Carbon:Nitrogen (C/N) ratio is a factor that relates the proportion of these two com-
pounds present in the organic material. Carbon is the energy source (catabolism) and nitro-
gen is used for the formation of new cells (anabolism). Normally, bacteria consume 30 times
2.4 Influence of operation conditions 31

more carbon than nitrogen, so the optimal ratio of these two elements in the raw material
is considered in a range of 30:1 to 20:1 (Varnero Moreno, 2011).

If the C/N ratio is too high, nitrogen is rapidly consumed by methanogenic microorganisms
to supply their protein requirements and is no longer available to react to the excess carbon
content in the material. As a result, biogas production decreases. On the other hand, if the
C/N ratio is very low, the nitrogen is released and accumulates in the form of ammonia,
causing the pH to increase creating a toxic effect on methanogenic bacteria (Abbasi et al.,
2012).

2.4.5. Food waste as substrate

Conventionally, the anaerobic digestion process can use organic waste of animal, vegetable,
agroindustrial, domestic origin among others. Table 2-5 shows a classification of the different
organic substrates used for the process.

Table. 2-5: Organic waste from different origins Varnero and Arellano (1990).
Residue Example
Animal origin Manure, urine, guano, slaughterhouse waste
Vegetable origin Weeds, crop stubble, straws, spoiled fodder
Human origin Stool, trash, urine
Agribusiness Crops, molasses, seed residues
Forestry Leaves, stems, branches, and barks
Aquatic crops Seaweed and aquatic weeds

Food waste is interesting as substrate for the anaerobic digestion process since its energy
content is high, large quantity and the availability of this is relatively high. The anaerobic
digestion process is an effective solution for the treatment and recovery of food waste (Zhang
et al., 2014). Compared to other organic residues, food waste could be a good substrate for
anaerobic digestion process due to their high degradability and physicochemical characte-
ristics, where a high methane yield can be obtained. Table 2-6 shows a comparison of food
waste with other organic substrates, where a favorable methane yield for food waste is esti-
mated.

According to the different eating habits, the food waste composition depends on the region
from which it is obtained and the presence of: rice, vegetables, meat, eggs, and other main
components. Table 2-7 shows the composition and characteristics of food waste from diffe-
rent regions of the world. As shown in the table, the contents of total solids (TS) and volatile
solids (VS) of food waste are in the ranges of 18.1-30.9 % and 17.1-26.35 %, respectively. This
indicates that water represents 70- 80 % of waste.
 32 2 Anaerobic Digestion Process

Table. 2-6: Substrate characteristics for the anaerobic digestion (House and Surratt, 2013)
Substrate Organic content C/N Volatile Solids Methane
( %) yield
(m3 CH4
kgVS−1 )
Animal waste
Pig manure Carbohydrates, 7 4 0.3
proteins, lipids
Beef manure Carbohydrates, 13 6.4 0.2
proteins, lipids
Bird droppings Carbohydrates, 7 4 0.3
proteins, lipids
Vegetable waste
Straw Carbohydrates, 90 56-81 0.15-0.35
lipids
Pruning waste Carbohydrates, 125 54-63 0.2-0.5
lipids
Pasture Carbohydrates, 18 18-23 0.3-0.55
lipids
Agroindustrial Waste
Serum 75-80 % lactose, - 4.5 0.33
20-25 % protein
Soy oil 90 % vegetable - 85.5 0.8
oil
Olive pulp - 18 0.33
Food waste Carbohydrates, 24 20 0.5-0.6
proteins, lipids
Sewage sludge
Sewage sludge - 3.7 0.4
Sewage sludge - 7.5 0.4
concentrate

The highest methane potential of food waste is in the range of 0.5-1.1 m3 of CH4 kgVS−1 ,
generally higher than other food waste substrates such as lignocellulosic biomass, animal
manure and sewage sludge (Mao et al., 2015). The highest methane yield is obtained from
household and restaurant food waste since it has a relatively high lipid content and also a
2.4 Influence of operation conditions 33

balanced nutrient composition. Carbohydrates and proteins are generally considered rapidly
degradable, thus residues of lipid-rich foods (i.e, used oil, ice cream) and easily degradable
carbohydrates can achieve high methane yields. In contrast, food waste with a high ligno-
cellulosic fraction and low lipid content, such as fruit and vegetable residues and brewery
residues, have methane potentials of less than approximately 0.16-0.35 m3 kgVS−1 (Grim-
berg et al., 2015).

Table. 2-7: Composition and characteristics of food waste reported in the literature for dif-
ferent regions
Parameter Zhang Zhang Zhang Solarte To-
et al. et al. et al. ro et al.
(2011) (2013) (2007) (2017)
Region Korea China USA Colombia
TS % 18.1 23.1 30.9 29
VS % 17.1 21 26.35 25.3
VS/TS ( %) 0.94 90.9 85.3 87.2
pH 6.5 4.2 NR NR
C (ppm) 46.67 56.3 46.78 48.3
N (ppm) 3.54 2.3 3.16 2.1
C/N 13.2 24.5 14.8 27
S (ppm) 0.33 NR 2508 0.2
P (ppm) 1.49 NR NR NR
−1
Methane yield (mLCH4 gVS ) 479 540 440 240
TS are the total solids ( %), VS are the Volatile Solids ( %), C is Carbon, N nitrogen,
S Sulfur, P Phosphorus. NR is not reporting.

2.4.6. Retention Time

As previously mentioned, retention time is the period for which organic material (substrate)
or microorganisms (solids) remain together in a digester to degrade the substrate to get a
high consumption of carbon, nitrogen, and nutrients. A biodigester is more efficient when the
retention time is shorter (Abbasi et al., 2012). However, to achieve low “substrate retention
times” it is necessary to simultaneously achieve high micro-organism (“solids”) retention
times as explained in the following sub-sections.
 34 2 Anaerobic Digestion Process

Hydraulic retention time: The hydraulic retention time (HRT) is the time that an orga-
nic material spends in a digester from the instant of its entry into the digester to its exit to
be degraded (Abbasi et al., 2012). Aiming for a complete degradation under mesophilic con-
ditions, the optimal hydraulic retention time is 15-30 days and for thermophilic conditions
is necessary 12-14 days (Mir et al., 2016).

Solids Retention Time: “Solids” is the term used to denote microorganisms in the anaero-
bic process. So, solids retention time (SRT) is microorganisms retention time. In conventional
low-rate digesters and Continuously Stirred Tank Reactors (CSTR), there is no disposal to
retain ”solids”. In those systems, HRT is the same SRT due to the solids pass out of the
digesters at the same rate as the substrate to be degraded. On the other hand, for high-rate
digesters, SRT > HRT since suspended growth systems allow microorganisms retention. In
a typical high-rate anaerobic digester, SRT=3HRT (Mir et al., 2016).

2.5. Inhibition and toxicity effects

The inhibition is conceived as a reversible reduction of the activity of the microorganisms,

while, the toxicity imposes an irreversible effect on population death. Compounds present in
the organic material (feedstock) and subproducts of the metabolic activity, can affect nega-
tively the population in the digester, causing partial or total cessation of their activity. Some
effects on the anaerobic digestion process of specific feedstock compounds are presented in
Table 2-8.

Some minor compounds are present such as: sodium, calcium, potassium, magnesium, iron,
chromium, manganese, selenium, iodine, lead, molybdenum, nickel, cobalt, arsenic. Their
salts and heavy metals serve as micronutrients to the bacteria’s activity. A high concentra-
tion of these substances, or interaction with other species, could cause toxic effect. Table 2-9
resumes salts and heavy metals that present toxicity to anaerobic metabolism.

Concerning anaerobic treatment, the most critical substances are oxygen, ammoniacal nitro-
gen, sulfurized compounds, and organic acids. Table 2-10 shows the inhibitory concentrations
of these compounds and some descriptions.
2.5 Inhibition and toxicity effects 35

Table. 2-8: Feedstock compounds and their effect on the digestion process (Steffen et al.,
1998; House and Surratt, 2013)
Compound Digestibility Feedstock source Process distur- Process inhibi-
bing effects tion
Foaming Layering
Fats Very good Slaughterhouses High VFAs le-
Poor water
Rendering vels Low pH
solubility
plants
Dairy processes pH decrease
Proteins Very good Foaming
Pharmaceutical High ammonia
industry concentrations
Carbohydrates
Very good
Sugars

Starch Very good Agro-industries Foaming pH decrease

Crop residues
Cellulose Poor Animal manures
Rendering plants
Poor water
Volatile fatty Very good High VFAs le-
solubility
acids (VFAs) Oil mills vels
Animal manures

Crop and crops

Organic Poor Foaming Antibiotic
residues
pollutants Pesti- effects
cides Antibiotics Organic wastes
Detergents
 36 2 Anaerobic Digestion Process

Table. 2-9: The concentration of inhibitory salts and heavy metals of the anaerobic digestion
process (Appels et al., 2018)
Compound Stimulating concen- Moderately inhibitory Strongly inhibitory
−1
tration (mg L ) concentration (mg concentration (mg
−1 −1
L ) L )
+
Na 3500-5500 8000
+
K 200-400 2500-4500 12000
2+
Ca 100-200 2500-4000 8000
2+
Mg 75-150 1000-1500 3000
2−
S 200 200
0.5 (soluble)
2+
Cu
50-70 (total)
3(soluble)
6+
Cr 10
200-250(soluble)
2 (soluble)
3+
Cr
180-240 (total)
2+
Ni 30 (total)
2+
Zn 1 (soluble)
Chlorides 6000
Lead 5
compounds
Cyanide 1–2 (acclimatization
possible up to 50)
Copper 1
compounds
Potassium >10000 (acclimatiza-
chloride tion possible up to
40000)
2.6 Alternative products from the anaerobic digestion process 37

Table. 2-10: Inhibitory substances for the anaerobic digestion (Appels et al., 2018).
Inhibitor Inhibitor concentra- Description
tion
Oxygen >0.1 mg L−1 Inhibition of anaerobic methanogenic
bacteria
−1
Hydrogen sulfide >50 mg L The inhibitory effect increases with the
decrease of the pH value
The effect of inhibitors increases with
−1
Volatile Fatty >2000 mg L the decrease in the pH value
Acids (pH=7) High adaptability of bacteria
The inhibitory effect increases with the
increase in the pH value (due to the
−1
Ammonium ni- >35000 mg L formation of ammonia), as well as the
trate (pH=7) temperature
High adaptability of bacteria
Antibiotics, Does not apply The inhibitory effect is product specific
disinfectants

2.6. Alternative products from the anaerobic digestion

process
2.6.1. Volatile Fatty Acids
Conventionally, the anaerobic digestion process has been a study point and interest mainly
in biogas production. However, other products can be obtained. An alternative approach to
conventional anaerobic digestion process is to target the production of Volatile Fatty Acids
(VFAs). These are building block chemicals with high demand in the market. Traditionally,
their production is based on non-renewable petrochemical sources that cause serious negative
effects on health and the environment.

VFAs are linear molecules of short-chain aliphatic monocarboxylate, having two (acetic acid)
to six (caproic acid) carbon atoms. In general, they have a wide range of applications, such
as a carbon source for: biogas, polyhydroxyalkanoate biopolymers (PHAs), biofuel precur-
sors, chemical components for industries or used for the biological removal of phosphorus or
nitrogen (Jankowska et al., 2017).

The most common VFAs produced from the anaerobic digestion process are acetic, propionic,
and butyric acid. These are important intermediates produced in acidogenesis and acetoge-
 38 2 Anaerobic Digestion Process

nesis steps of the digestion process. Properties, applications, market size, and production
methods are summarized in Table 2-11. VFAs are mostly used in fields of food and beve-
rages as acidifiers, but also in the cosmetic industry, tanning industry, and pharmaceutical
industry.

Table. 2-11: Volatile Fatty acids general properties. Atasoy et al. (2018); Zacharof and
Lovitt (2013).
Market Market
Chemical for- size Price
VFAs Application Production methods
mula (kton (USD
year−1 ) ton−1 )
Vinyl ace- Chemical synthe-
tate mono- sis: Carboxylation
mer(polymers, of methanol, Ace-
Acetic adhesives, dyes), taldehyde oxidation,
3500000 400-800
acid food additive, Ethylene oxidation
solvent, vinegar,
Bioprocess: oxidati-
ester produc-
ve and anaerobic
tion, chemicals
Chemical synthe-
Animal and hu- sis: Hydrocarboxy-
man food additi- lation of ethylene,
Propionic 1500- ve, Chemical in- Aerobic oxidation of
180000
acid 1650 termediate, Sol- propionaldehyde
vent, Flavouring
Bioprocess: Anaero-
agent
bic process

Ester used food

industry as
aroma additive, Chemical synt-
Food additive, hesis: oxidation of
Butyric 2000- flavoring, Phar- Butyraldehyde
30000
acid 2500 maceuticals,
Bioprocess: Fungal
Animal feed
fermentation of gluco-
supplement,
se
Fishing bait
additive
2.6 Alternative products from the anaerobic digestion process 39

Usually, VFAs are obtained from petrochemical derivates synthetically and biologically pro-
cess using the fermentation process. Although bio-based production methods are environment-
friendly, currently these can not compete economically with the petro-based production
methods (Liu et al., 2018). The lower manufacturing cost of oil-based production methods
and the lower efficiency of bio-based production methods are the main reasons why petro-
based production methods are dominant (Atasoy et al., 2018). However, adverse effects on
the environment such as Greenhouse Gas emissions, high energy requirements, chemical re-
quirements, and large waste and wastewater production are the driving forces shift from
petrol-based production to bio-based methods (Besselink et al., 2017). For the study of bio-
based VFAs production methods have particular importance nowadays. Research has been
focused on increasing efficiency, enhancement the operation conditions, providing renewable
sources as substrate, characterizing and assessing the microbial communities with their in-
teractions, and new separation processes.

The optimization of the operating conditions is the most significant topic of bio-based VFAs
production methods. Table 2-12 shows operation conditions in terms of pH, temperature,
type of substrate, retention time, and reactor type effects on VFAs production through
bio-based process. pH has the biggest impact on concentration and composition of VFAs.
pH values between 5 and 11 favor the production of VFAs, having a wide range compared
to the optimum pH for the production of biogas that is between 6.5 and 7.5. Regarding
temperature, as well as for the production of VFAs and biogas, mesophilic conditions promote
the production of these compounds. The type of substrate influences the production of
VFAs, the degree of acidification being an important characteristic to be considered. On the
other hand, the C/N ratio influences directly the biogas production. Alternatively, VFAs
production requires a lower SRT than biogas oriented process, because low SRT can prevent
methanogen dominance.
40 2 Anaerobic Digestion Process

Table. 2-12: Parameters that affect VFAs production.

Influence of operational pa-

Operation
rameter on VFAs produc- Study cases References
parameter
tion
The optimal pH values for
the production of VFAs
are mainly in the range of
5.25–11, but the specific
ranges are dependent on
pH is one of the most
the type of waste usedb .
critical parameters that
have a very strong effect
The stepwise pH fermenta-
on the VFAs concentration
tion strategy envelopment
and composition. Also, pH
by Zhao et al. (2018)shows
assumes an important role
that the activity of acid
in increasing the production
producing bacteria (pH 9)
rate and yield of VFAs in
was improved, as well as
the anaerobic digestion
inhibition of the activities
process. a. Liu et al.
of methanogens (pH 11),
(2012)
pH which resulted in an increa-
pH affects the microor- [Link] et al.
sed production of VFA.
ganism activity because (2014)
of the majority of enzy-
Huang et al. (2018) studied
mes do not tolerate acidic
the effect of pH (3, 5, 7,
(pH<3) or alkaline (pH>12)
9, 10 and 12) on VFAs
environmentsa .
production. Their results
concluded that under pH
Any change in pH va-
10, the optimal VFAs pro-
lue can control the type
duction occurred.
of VFAs produced from
acidogenesis fermentationb .
Optimal pH for a specific
VFAs production is highly
dependent on the type of
substrate used.
2.6 Alternative products from the anaerobic digestion process 41

Temperature is a key para-

VFAs yields are pretty si-
meter that improves VFAs
milar at both thermophilic
production due to that it
and mesophilic conditions.
affects the enzyme activity,
However, the mesophilic
the growth of microorga- c. Zhou
temperature (35 °C) is the
nisms, and hydrolysis ratec . et al.
most optimum and econo-
(2018)
mical favorable condition
A change of operation d. Straz-
for VFAs productione .
Temperature temperature can alter zera et al.
the microbial structure of (2018)
Jiang et al. (2013) stated
the microbial consortium e. Gruhn
that butyric acid was the
involved in acidogenic et al.
d
main product at a wor-
fermentation . Also, tem- (2016)
king temperature of 55 °C,
perature affects the type
meanwhile acetic and pro-
of main VFAs product in
pionic acids were the main
fermentation.
product at 35 °C.
The production of VFAs is
significantly influenced by Silva et al. (2013) studied
the type of substrate used. the fermentation of eight
organic wastes streams.
The degree of acidifica- These presented significant
tion is an important factor variance in the degree of
in the fermentation process acidification, resulting in
for VFAs production and it different VFAs productions.
is defined as the percentage
of initial Chemical Oxygen Cheese whey, sugarcane mo- f. Jin et al.
Demand (COD) converted lasses and organic fraction (1999)
Substrate into organic acidsf . This of municipal solid waste [Link]
factor indicates the amount streams presented the hig- et al.
of VFAs produced in the hest degree of acidification (2018)
anaerobic digestion process. (up to 40 %) with total
VFAs production of 2700-
The VFAs composition 3300 mg L−1 as COD. On
produced from waste the other hand, landfill lea-
streams is due to the cha- chate produced 634 and 240
racteristics (carbohydrate, mg L−1 as COD with the lo-
lipid and protein content) of west degrees of acidification
the organic matter content of 2 %.
of the waste streamg .
 42 2 Anaerobic Digestion Process

Retention time includes

hydraulic retention time
(HRT) and solid retention
time (SRT).
Feng et al. (2009) stated
VFAs production dependsthat increasing the STR
more on the hydraulic from 4 d to 12 d produced
retention time compareda 44 % higher VFAs concen-
to the temperature of atration because of the high
reactorh . soluble substrates amount.
However, further, an in-
h. Kim
SRT can prevent the crease in SRT to 16 d led in
et al.
dominance of methanogens lower VFAs concentration,
(2013)
in the anaerobic process although there were even
Retention i. Ferrer
as the growth rate of met- more soluble substrates.
time et al.
hanogens is lower than
(2010)
that of acidogensi. This
j. Lee et al.
makes that a lower STR, Lim et al. (2008) demons-
(2014)
the production of VFAs is trated that the production
beneficial. of VFAs increased as the
HRT increased from 96 h to
A higher HRT could 192 h, but there was no furt-
be advantageous to VFAs her increase in VFAs pro-
production due to the duction once the HRT ex-
microorganisms have more ceeded to 288 h.
time to react with the
waste. However, prolonged
HRT could lead to stagnant
VFAs productionj .

2.6.2. Digestate or biol as organic fertilizer

In addition to generating biogas as fuel, anaerobic digestion of organic matter produces an
organic residue with excellent properties as fertilizer. The solid or liquid digester or biol has
no bad smell, does not attract flies, and can be applied directly to the field in liquid form.
This applications are limited to the presence of pollutants like pathogens, heavy metals, etc.
Otherwise, the solid biol can be dehydrated and stored for later use. In general, biol contains
N2 (1.8 %), P2 O5 (1.0 %), K2 O (0.9 %), Mn (188 ppm), Fe (3550 ppm), Zn (144 ppm) and
Cu (28 ppm) (Surendra et al., 2014).
2.7 Benefits of the anaerobic digestion process 43

Biol does not leave toxic residues in the soil, raises the quality of the soil, and can be con-
sidered as a good fertilizer that can compete or be complemented with chemical fertilizers
(Fregoso Soria et al., 2001) This results in money and time savings on fertilizers and also
helps regenerate the land.

2.7. Benefits of the anaerobic digestion process

2.7.1. Health benefits
In many rural areas, direct biomass burning (wood, crop residues) is used as fuel in tradi-
tional cookstoves. This activity results in higher emissions of carbon monoxide and carbon
dioxide, hydrocarbons, and particulate matter. These emissions result in severe health issues
due to indoor pollution (IAP), mainly indoors without proper ventilation. Additionally, IAP
has been linked to other health problems, such as child pneumonia, chronic obstructive pul-
monary diseases, lung cancer, asthma and cataracts, tuberculosis among others (Surendra
et al., 2014). Women and children are more susceptible to IAP due to prolonged exposure
to smoke when they spend hours cooking and doing other household chores.

Through the traditional and economical construction of digesters, organic waste from farms
(which are more accessible than firewood) is used to produce biogas. Unlike burning biomass,
biogas (H2 S from biogas should be removed previously) can provide a clean and smoke-free
environment and could significantly reduce IAP.

2.7.2. Environmental benefits

The use of firewood as fuel has an important negative impact on local forests. Fuel wood con-
tribute to deforestation and have negative impacts on forest health and biodiversity. World-
wide deforestation causes 17-25 % of all anthropogenic Greenhouse gas emissions (Strassburg
et al., 2009). Also, unlike the carbon dioxide emission of firewood, the carbon dioxide emis-
sions that originate from the use of biogas are equal to the amount of carbon dioxide that
affects plants to grow and produce renewable resources. Therefore, no additional carbon dio-
xide is produced, which is considered harmful to the climate.

Thus, when replacing firewood with biogas can reduce deforestation, mainly in developing
countries. In the same way, Greenhouse gas emissions can be mitigated.
 44 2 Anaerobic Digestion Process

2.7.3. Social Benefits

In rural areas, collecting firewood can be a task that requires time and a lot of human effort.
Commonly, women and children can get to travel more than 5 km and spend nearly 6 hours
a day gathering firewood (Topa et al., 2004). This labor is difficult and may cause neck-pain,
back-pain, and other physical issues. Likewise, the time spent by women and children in this
labor, deprive them of opportunities for education and social activities(Surendra et al., 2014).

With the implementation of biodigesters to produce biogas, greater and better opportunities
can be provided for the beneficiaries. In addition, the quality of life of people can improve,
since, in areas where there is no electricity, biogas (through a cogenerator) can meet this need.

2.8. Case study: Food waste from the Nariño region as a

substrate for anaerobic digestion
The type, composition, and amount of food waste is defined by economic development and
population growth, where the main sources are collection centers, agribusinesses, hotels, res-
taurants, family homes, dining rooms and companies (Zhang et al., 2014). Currently, the
problem of food waste is rising. Worldwide, it is estimated that about 1.3 billion tons year−1
of solid waste are generated and this amount is expected to increase to 2.2 billion tons by
2025 (One third corresponds to food waste) (Han et al., 2016). Loss refers to food that occurs
in the stages of agricultural production, post-harvest and industrial storage, and processing.
On the other hand, waste is food that occurs in the stages of distribution and retail and
consumption.

In Colombia, this problem is not foreign. According to a study by Departamento Nacional

de Planeación (2016), Colombia lost 9.96 million tons of food per year, and in the Pacific
region (i.e. Chocó, Nariño, Cauca, Valle del Cauca) appears in fourth place with 1’063.159
(17.1 %) tons year−1 and in third place with 488.539 (13.8 %) tons year−1 of wasted food.

The disposal and elimination of this waste is a problem that affects environmental, econo-
mic and social aspects. Traditional approaches for food waste disposal and elimination are
mainly in landfills, by incineration, and to a limited extent by aerobic decomposition (com-
post). However, it is relevant to mitigate this type of practice due to adverse consequences.
Incineration is high energy demanding since these wastes have high water content and often
pollute the air (Zhang et al., 2014). On the other hand, sanitary landfills disposal have been
banned in many countries due to serious pollution problems because of high chemical and
biological oxygen demand (COD and BOD) (Lin et al., 2013). At the same time, a series
of problems arise that include the increase in the cost of waste disposal, the lack of space
2.8 Case study: Food waste from the Nariño region as a substrate for anaerobic digestion
45

on the ground, the contamination of groundwater by leachate and the emission of toxic and
greenhouse gases (Uçkun Kiran and Liu, 2015).

According to the study carried out by Superintendencia de Servicios Públicos Domiciliarios

(2016), in Colombia the number of final waste disposal sites has been reduced in the last 3
years of evaluation with respect to 2016, in 2014 there are 400 sites, in 2015 a total of 227 sites
and in 2016 there were 269 sites. To dispose of waste, there are adequate and inadequate sites.
Figure 2-4 shows the number of adequate sites (sanitary landfill and contingency cell) and
inadequate sites (transitional cell, open-air dump, burial, dumping of water bodies, burning
of solid waste). In Colombia, landfills were and still a priority, however, the lifespan of many
of these are ending. In 2016, the country had 275 sites between adequate and inadequate
(related to operation and management), mainly are landfills (158 sites), open-air dumps (54
sites) and transitional cells (34 sites). The Nariño department has 17 final disposal sites, of
which 11 are suitable sites in the form of landfills, of which it is estimated that the useful life
of these is between 3 to 15 years (Superintendencia de Servicios Públicos Domiciliarios, 2016).

Figure. 2-4: The number of final disposal sites in Colombia for 2014, 2015 and 2016 validities.
Adapted from: Final Disposal of Solid Waste-National Report-2016 (Superin-
tendencia de Servicios Públicos Domiciliarios, 2016)

The Nariño region generates 235.322 ton year−1 of solid waste (Superintendencia de Servi-
cios Públicos Domiciliarios, 2016), of which, it is generally estimated that about one third
is food waste (Han et al., 2016). Then, it could be said that per year, approximately 84.500
 46 2 Anaerobic Digestion Process

ton year−1 of food is wasted in Nariño. The generation of food waste is expected to increase
every year since this is proportional to the population increase, where Nariño since 1985,
each year has had a change in the average population of +1.3 % (DANE, 2017).

Considering the negative environmental impact, the potential decrease and high costs of
adequate waste disposal sites, as well as the unfeasibility of incinerating them, the anaerobic
digestion process has been proposed as a technology relatively profitable to dispose and take
advantage of this organic waste aimed primarily at the production of renewable energy. Ta-
bles 2-13 and 2-14 present situations in which the biogas produced by food waste generated
in the Nariño department can be used. It should be noted that these calculations are very
ideal, but it helps to have an overview of the maximum extent of biogas.

Table. 2-13: Ideal generation of electricity from food waste in the Nariño department (this
study).
Food waste Biogas produced Electricity genera- Satisfied Colombian
−1 3
(Ton year ) (m ) ted (kWh-year) houses
235 322 47 064 400 282 386 400 47 064
3
Equivalences: Biogas contains 60 % methane, 1 m biogas equals 6 kWh
Colombian house consumes approximately 6000 kWh year-house .

Table. 2-14: Ideal equivalence of natural gas from food waste in the Nariño department (this
study).
Food waste Biogas produced Equivalence to na- Satisfied Colombian
−1 3
(Ton year ) (m ) tural gas produced houses
(m3 )
235 322 47 064 400 28 238 640 67 234
Equivalences: Biogas contains 60 % Methane, 1 m Biogas equals 0.6 m3 of natural gas
3

(Varnero Moreno, 2011), Colombian house consumes approximately 420 m3 year−1

of natural gas.

The development of organic waste management through anaerobic digestion technology

would articulate the environmental and energy aspects, promoting the implementation of
a circular economy, aiming that the value of products and materials is maintained during
the productive cycle (Consejo Nacional de Polı́tica Económica y Social, 2016). The circu-
lar economy provides multiple value creation mechanisms not linked to the consumption of
finite resources. As shown in Figure 2-5, in a circular economy, unlike the linear economy,
resources are regenerated within the biological cycle, recovered, and restored thanks to the
2.8 Case study: Food waste from the Nariño region as a substrate for anaerobic digestion
47

technical cycle. Within the biological cycle, the anaerobic digestion process allows the dis-
carded materials to be regenerated. In the technical cycle, with sufficient available energy,
human intervention recovers the different resources (reuse, use, treatment, among others)
and recreates the order within the time scale that arises (Macarthur, 2006).

Figure. 2-5: The linear economy(left) ignores the environment impacts hat come with re-
source consumption and waste disposal, and results in too much virgin resource
extraction, pollution, and waste. In contrast, the circular economy (right) takes
into account the impact of resource consumption and waste on the environment.
The objective of the circular economy is to optimize the use of virgin resources
and reduce pollution and waste at each step, inasmuch as possible and desirable
(Sauvé et al., 2016)

For a specific study, it was decided to analyze the panorama of food waste generation in
the city of Ipiales-Nariño. The “Instituto de Servicios Varios de Ipiales” (ISERVI) provided
information on the current state of solid waste generation and the current state of the land-
fill “La Victoria” in the city of Ipiales. Table 2-15 shows these characteristics for the year
2019. The data reveal that 60 % of solid waste is organic waste, making the city have a great
chance of taking advantage of this waste.

As mentioned, anaerobic digestion is a process that produces biogas as renewable energy and
VFAs as economically attractive products. Factors such as pH, temperature, type of substra-
te, and type of digester can define the desired product. To evaluate the anaerobic digestion of
an organic mixture, parameters to determine the quantity and quality of the products obtai-
ned are necessary: characterization of the substrate, evaluation of the Biochemical Methane
Potential, monitoring of variables such as pH, COD, temperature, and concentration of VFAs
throughout the process. Food waste is a promising substrate for the production of biogas and
 48 2 Anaerobic Digestion Process

VFAs, an evaluation of this waste with the anaerobic digestion process at the local context,
will give a vision of the potential uses, producing compounds with energy value and added
value. Thus, in this study, the potential for methane and VFAs production from food waste
from the Department of Nariño, Colombia was evaluated. The substrate was characterized
and the evaluation of methane production was carried out, using the assay of the biochemical
potential of methane in batch digesters. To monitor the abovementioned process variables,
and this test was complemented with a sacrifice bottle assay.

Table. 2-15: The current state of generation and management of solid residues in Ipiales,
Nariño. Data provided by ISERVI
Item Description
Solid residues amount 1450 Ton month−1
Organic residues amount 870 Ton month−1
Landfill use time 8 years
Landfill lifetime 15 years
Municipalities served 10
Landfill area 26.34 hectares
Type of solid residues treatment Combined landfill: area and trench type
2 UASB reactors
2 Mechanical oxidation lagoons
Leachate Treatment
2 sedimentation tank
2 maturation lagoons

2.9. Methodology
2.9.1. Substrate sampling: Food Waste
Food waste (FW) was obtained at the Casa Colombia restaurant, located in the city of
Ipiales-Nariño Department. They were collected considering sampling protocol ASTM D5231-
92. Six samples (FW1, FW2, FW3, FW4, FW5 and FW6) were collected daily for six con-
secutive days in the afternoon. Additionally a composite sample (CFW) of the six previous
samples was prepared. They were stored in Ziploc plastic bags and no physical or chemical
treatment was performed. They were weighed using a weighing hook and preserved in a
freezer at a temperature of -15 ° C until use. The samples taken were transported to the city
of Manizales using portable expanded polystyrene refrigerators with dry ice to keep them at
a low temperature.

In the Laboratory of Productive Processes of Universidad Nacional de Colombia-Manizales,

the low biodegradability compounds separation of each sample was carried out, characte-
2.9 Methodology 49

rizing and identifying qualitatively the type of organic matter. Next, a JAVAR industrial
blender was used to crush each sample until a particle size of 0.2 to 1 cm was achieved, as
recommended by Sharma SK, Mishra IM, Sharma MP (1988). The average grinding time
was 20 seconds. Each sample was weighed with a FEMTO ABT series analytical balance.
Finally, the samples were stored at a temperature of -15 ° C until use.

2.9.2. Food Waste characterization

The analyzes that were carried out were: total solids (TS), volatile solids (VS), proximate
analysis: total humidity, dry matter, ash, volatile matter and fixed carbon, ultimate analy-
sis: the amount of carbon, hydrogen, oxygen, and nitrogen and calorific Value. The analyses
were carried out in the Waste Exploitation laboratory and Water Laboratory at Universidad
Nacional de Colombia-Manizales. All analyses were performed with a replica.

Total solids and volatile solids: TS are used to describe the dry matter of a subs-
trate, and VS represent the amount of material that can be used as food by microorga-
nisms, and which theoretically must be converted to methane. For the quantification
of TS and VS of food waste, the protocol of Standard Methods for the examination of
water and wastewater (APHA, 2008) was followed. Table 2-16 shows the methods used.

Table. 2-16: Methods used to determine the parameters of TS and VS

Parameter Unit Method Technique
−1
TS mg TS mg sample S.M.2540 B Gravimetry
−1
VS mg VS mg sample S.M.2540 E Gravimetry

Proximate analysis: The proximate analysis of CFW separates the products into
four groups: (1) moisture (M), (2) volatile Matter (VM) consisting of gases and vapors
released during pyrolysis, (3) fixed Carbon (FC), non-volatile fraction and (4) ash con-
tent (ASH), inorganic residue after combustion. A summary of the standard methods
used to perform the proximate analysis is presented in Table 2-17.

Ultimate analysis: The ultimate analysis provides information about the content
of elements such as carbon (C), hydrogen (H), oxygen (O) and nitrogen (N). From
the proximate pnalysis and through correlations developed by Shen et al. (2010) it is
possible to give an approximation of the content of these elements.
50 2 Anaerobic Digestion Process

Table. 2-17: Standard methods used to carry out the Proximate analysis of CFW
Analysis Method
Moisture ASTM E871-82 (2013)
Volatile Matter ASTM E872-82 (2013)
Ashe ASTM E1755 - 01(2015)
Fixed Carbon ASTM E870 – 82 (2013)

%C = 0.635( %F C) + 0.460( %V M ) − 0.095( %ASH) (2-1)

%H = 0.059( %F C) + 0.060( %V M ) + 0.01( %ASH) (2-2)
%O = 0.340( %F C) + 0.469( %V M ) − 0.023( %ASH) (2-3)

According to Shen et al. (2010), equations 2-1, 2-2 and 2-3 are valid for raw materials
containing:
9.2 % ≤ F C ≤ 32.79 %
57.2 % ≤ V M ≤ 90.6 %
0.1 % ≤ ASH ≤ 24.6 %
36.2 % ≤ C ≤ 53.1 %
4.7 % ≤ H ≤ 6.61 %
31.37 % ≤ O ≤ 48.0 %

These values are expressed in mass fraction ( % wt.) on dry basis.

For the purposes of this study, it will be assumed that the rest of the composition is
nitrogen. This is because nitrogen is a compound that is present in this class of waste
in a larger quantity in relation to other components that this residue contains (Na, S,
K, P, etc) (Esteves and Devlin, 2010). The nitrogen content is calculated with equation
2-4.

%N = 100 − %C − %H − %O (2-4)

Calorific value: It is the amount of energy emitted by a raw material when it is sub-
mitted a combustion process. To determine the calorific value, a calorimetric technique
was used by means of an immersion bath in the SDACM3100 Bomb Calorimeter unit
with pressurization of the sample with oxygen at 3 MPa and operating temperatures
of 24 to 42 ° C.
2.9 Methodology 51

2.9.3. Inoculum
The inoculum is an important part of the start-up of the anaerobic digestion process since
it is the one that contributes most of the microbial consortium in the process (Demirel and
Scherer, 2011). The inoculum should preferably be fresh and extracted from an anaerobic
reactor in operation (stirred tank, UASB, etc.) in the proportion (or volume) desired for the
tests (Angelidaki et al., 2009).

For this study, an inoculum from the UASB of a coffee industry was used. The company
Buencafé (Chinchiná, Caldas) supplies this inoculum. It was required to homogenize and
acclimation for 10 days at a temperature of 36 °C in a thermostatic bath until the remaining
organic matter was consumed (Angelidaki et al., 2009). During the acclimation process, 0.5
g L−1 day−1 of food waste was added to adapt the inoculum to the substrate used (Solarte
Toro et al., 2017). Additionally, the inoculum was characterized in terms of volatile solids,
total solids, and pH following the methods presented in Table 2-17.

2.9.4. Biochemical Methane Potential (BMP)

To evaluate the application of anaerobic treatment systems in the degradation of an organic
substrate, it is necessary to determine its methanogenic potential (Angelidaki et al., 2009).
Through a discontinue essay known as Biochemical Methane Potential (BMP) is possible to
evaluate the methanogenic potential on a laboratory scale. BMP is widely used to determine
the concentration of organic matter present in a residue that can be anaerobically converted
to methane (CH4 ). As well as BMP can evaluate the efficiency of an anaerobic process in
the degradation of specific wastes. This test has an approximate duration of 30 days where
methane production is followed and registered, which is related to the reduction of organic
matter expressed as COD (Chemical Oxygen Demand) (Esposito et al., 2012).

Several norms aimed at standardization of BMP tests such as DIN 38414 TL8 (1985), ASTM
D 5210 (1992), ASTM D 5511 (1994), ISO 11734 (1995), ISO 14853 (1998), and ISO 15985
(2004). However, to perform the BMP test with food waste as substrate, the German stan-
dard VDI 4630 was used (VDI-Handbuch Technik Biomasse, 2016). In order to analyze the
stages and development of the anaerobic digestion of food waste, two BMP assays were ca-
rried out. In the first experiment, only methane production was monitored on-line and in the
second, methane production, COD, VFAs, and pH were monitored by sacrifice bottle every
3 days on average. The following is the process used for these experiments.

Macronutrients and micronutrients solution: To guarantee the most favorable condi-

tions that enhance the anaerobic degradation of the substrate, it is necessary to use a solution
of macronutrients and micronutrients (Owen et al., 1979). As suggested by Angelidaki et al.
 52 2 Anaerobic Digestion Process

(2009), a ratio of 1:100 v/v of macronutrients, 0.1:100 v/v of micronutrients, and 0.2:100 v/v
of yeast extract was added in the liquid phase volume for the BMP assays. NaOH and HCl
solutions were used to adjust the pH. The solutions used in the BMP assays are presented
in Table 2-18.

Table. 2-18: Composition of the nutrient solution for the BMP assay (Angelidaki et al.,
2009)
Solution Solution:liquid Chemical compound Unit Concentration
phase ratio
(v/v)
NH4 Cl 100
NaCl 10
Macronutrients 1:100 MgCl2 -6H2 O g L−1 10
CaCl2 -2H2 O 5
K2 HPO4 - 3H2 O 200
ZnCl2 0.05
AlCl3 0.05
CoCl2 -6H2 O 0.05
Micronutrients 0.1:100 H3 BO3 g L−1 0,05
(NH4 )6Mo7 O24 -4H2 O 0.05
C10 H16 N2 O8 0.5
FeCl2 - 4H2 O 2
0.2:100 Yeast extract (source of vi- g L−1 0.2
Other tamins)
- HCl 2
N
- NaOH 2

Methane production by on-line measurement assay: The on-line methane quanti-

fication was carried out using the German Yieldmaster by BlueSense equipment which per-
forms real-time measurements of methane concentration and volume produced. It measures
up to 80 mL min−1 with a resolution of 1 mL and consists of a volume measurement cell
as well as temperature and pressure sensor. The sensor contains the infrared light source
and detector. The infrared light beam is reflected by the gas-filled measuring adapter and
the light weakened by the analyte gas is measured by the detector. Yieldmaster equipment
sensors report methane concentration data at 0.8 bar gauge pressure and at an average tem-
perature of 20 °C. Figure 2-6 shows the equipment used. Two bottles were used under the
same experimental conditions (inoculum, macronutrient, micronutrient, and yeast extract
solution). One corresponds to the blank (without substrate) only with water to determine
the endogenous activity of the inoculum. The other using food waste as a substrate. The
2.9 Methodology 53

measurements were taken every 30 minutes and the data is saved in a computer, using the
equipment software BACVis®.

Figure. 2-6: On-line methane measurement assay setup.

The bottles have a volume of 300 mL (working volume of 200 mL). The bottles consist of
an upper nozzle that attaches to the methane sensor of the Yieldmaster equipment. They
also have an upper side nozzle through which the biogas leaves the bottle to be quantified.
The assay was performed under mesophilic conditions of 37 °C in a Memert WNB 22 ther-
mostatic bath with daily manual stirring for 31 days. The pH was adjusted to 7.0 before
the operation. According to the VDI 4630 standard, the substrate-inoculum ratio was set on
0.4 gVSsubstrate :gVSinoculum and for every 500 mL of digestion volume, the inoculum should
provide 7.5 gVS (VDI-Handbuch Technik Biomasse, 2016). The amounts of inoculum, subs-
trate, and nutrients can be seen in Table 2-19. The bottles were made up to working volume
with tap water. To ensure removal of the oxygen content within the bottle, nitrogen was
bubbled into each bottle for 3 minutes. Then the bottles were sealed and capped ensuring
that there were no leaks. The assay had no duplicate. The pH, CODs, VFAs were evaluated
at the beginning and end of the assay.

Table. 2-19: Conditions for the BMP assay with on-line methane production measurement
Item Units Value
Inoculum g 24.42
Substrate g 3.4
Macronutrient solution mL 2
Micronutrient solution mL 0.2
Yeast extract solution mL 0.4
 54 2 Anaerobic Digestion Process

Methane production by sacrifice bottle assay: To monitor several operational para-

meters throughout the anaerobic digestion process such as CODs and pH, a sacrifice bottle
assay was performed. 30 bottles of 120 mL (working volume of 80 mL) was used, of which 20
contain substrate, inoculum and nutrients and the rest only inoculum and nutrients (blank).
The bottles consist of a small upper nozzle that is hermetically closed with a rubber lid
and aluminum cap. The assay was carried out under mesophilic conditions of 37 ° C in a
BINDERED 260 incubator. Each day manual stirring was performed on each bottle. The
assay lasted 31 days.

The substrate-inoculum ratio was set on 0.4 gVSsubstrate :gVSinoculum and for every 500 mL of
digestion volume, the inoculum should provide 7.5 gVS (VDI-Handbuch Technik Biomasse,
2016). The amount of inoculum, substrate, and nutrients can be seen in Table 2-20. The pH
was adjusted to 7.0 before operation and the bottles were made up to working volume with
tap water. Nitrogen was bubbled into each bottle for 3 minutes. The volume of methane was
measured every 2 days by the alkaline displacement method(Cárdenas et al., 2016). Figure
2-7 shows the setup of the assay and the methane quantification from each bottle. Every
3 ± 1 days, 3 bottles were discarded (2 bottles contained substrate and 1 bottle was the
blank) and subsequently, pH and CODs were measured. For the initial value of CODs, a 5
mL sample was taken from a bottle with substrate before starting the assay.

Figure. 2-7: Sacrifice bottle assay setup

Soluble chemical oxygen demand (CODs): Chemical oxygen demand (COD) is a mea-
sure of the quantity of carbon present in the sample. Specifically, soluble chemical oxygen
demand (CODs) allows to periodically monitor the development of the biodegradation pro-
cess of organic matter (Owen et al., 1979). For the CODs measurement, the Standard Method
“5220 CHEMICAL OXYGEN DEMAND (COD)” was followed (APHA, 2008). Previously,
each sample was subjected to centrifugation at 5500 rpm for 10 min in an INDULAB Ref
004 dynamic centrifuge.
2.9 Methodology 55

Table. 2-20: Conditions for the BMP assay with sacrifice bottles
Item Units Value
Inoculum g 9.77
Substrate g 1.36
Macronutrient solution mL 0.8
Micronutrient solution mL 0.08
Yeast extract solution mL 0.16

Theoretical biochemical methane potential: The theoretical BMP of an anaerobic di-

gestion process is limited by stoichiometry and can be calculated if the elemental composition
of the substrate is known. For this study, a stoichiometric model proposed by Buswell and
Mueller (1952) was used. Likewise, knowing the theoretical BMP and experimental BMP
(obtained from the previous tests) it is possible to estimate the biodegradability index ( %B)
as described by Sosnowski et al. (2003), Equations 2-5-2-7 were used to calculate the theo-
retical BPM and the index of biodegradability.

Noticed that the analyzes performed corresponds to preliminary exploration since at this
stage it was not possible to perform triplicates.

   
a b 3c n a b 3c
Cn Ha Ob Nc + n − − + H2 O → − − + CO2
4 2 8 2 8 4 8
 
n a b 3c
+ + − − CH4 + cN H3 (2-5)
2 8 4 8

n a b 3c


22, 4 + − − 1000
BM Ptheoretical (mL CH4 gV S −1 ) = 2 8 4 8
(2-6)
12n + a + 16b + 14c
BM Pexperimental
%B = ∗ 100 (2-7)
BM Ptheoretical

In these equations, n,a,b and c indicate the number of moles of carbon, hydrogen, oxy-
gen, and nitrogen respectively, 22.4 corresponds to the volume (L) occupied by an ideal gas
under standard conditions, 1000 is the volume conversion factor for conversion from L to
mL, and 12, 1, 16, and 14 are the molecular weights (g mol−1 ) of C, H, O and N, respectively.
 56 2 Anaerobic Digestion Process

2.10. Results and discussion

2.10.1. Characterization of food waste and inoculum
Table 2-21 shows the characteristics of the samples collected. Potato, rice, and chicken were
the common foods in each food waste sample. On the other hand, fruit peel, chicken skin,
and bones were the predominant low biodegradability waste (compounds that cannot be or
are hardly degraded in the anaerobic digestion process) in the samples, which were remo-
ved. The physical pretreatment of food waste by reducing the particle size was carried out
to homogenize the substrate and improve the performance of the process in an increase in
biogas. According to the study by Hajji and Rhachi (2013), this pretreatment can improve
biogas production by 20 % when the particle size is reduced to 10 mm. Figure 2-8 shows
food waste before and after size reduction.

Figure. 2-8: (a): Separation of organic matter and low biodegradability waste from the subs-
trate. (b): Food waste after size reduction.
2.10 Results and discussion 57

Table. 2-21: Food waste amples

Weight or-
ganic matter
Organic Low biode- Weight for
without low
Sample Organic Matter matter gradability composite
biodegrada-
weight (g) waste sample * (g)
bility waste
(g)
Pea, potato,
beef, chicken, Lulo peel,
FW2 carrot, banana, 1500 paprika peel, 983 191.2
fish, beans, let- thorns, seeds
tuce, pumpkin
Onion, to-
Banana, potato, mato peel,
rice, peas, cassa- paprika,
FW3 va, beans, nood- 1500 cucumber, 1120 217.9
les, fish, chicken, thorns, chic-
carrots ken skin

Paprika,
Cassava, potato,
corncob, fish
FW4 rice, banana, 1250 1143 222.3
skin, chicken
corn, beef
skin
Potato, pork,
beef, banana,
beans, arepa, Tomato,
FW5 rice, tomatoes, 1500 lettuce, fish 1166 226.8
carrots, onions, skin
cassava
Banana, rice, Tomato,
peas, fish, meat, lettuce, fish
FW6 1500 729 141.8
potatoes, corn, skin, thorns,
carrots, beans orange peel
FW1 was discarded due to the decomposition of the waste.
*A composite sample (CFW) with a total weight of 1000 g was made

Table 2-22 shows the results of the physicochemical characterization of food waste and
inoculum. To determine the moisture content, the food waste was initially subject to a pre-
drying at 60 °C for 24 h. The decrease in moisture by this process is not depicted in Table
 58 2 Anaerobic Digestion Process

2-23. The moisture of 10 % should be interpreted as the remaining moisture of partially dry
food waste. The VS/TS ratio (94 %) shows similarity with the values reported by various
authors, which are between 87 % - 95 % ( Zhang et al. (2013),Solarte Toro et al. (2017),Pra-
manik et al. (2019)). This relationship suggests that the wastes have a high potential for
biodegradability and are suitable for use in the anaerobic digestion process. Regarding the
inoculum, it is a granular inoculum and the results show a VS/TS ratio > 80 %, which indi-
cates a high content of active biomass present, making it a reliable inoculum for use in the
anaerobic digestion process.

Table. 2-22: Physicochemical characterization of food waste and inoculum

Parameter Unit Food Waste Inoculum
Total solid 37.50 ± 1.63 13.82 ± 1.23
% wt
Volatile Solid 35.33 ± 1.62 12.28 ± 1.22
VS/TS % 94.23 88.861

Concerning the proximate analysis of food waste, the results are in agreement with the re-
sults reported by the authors of Table 2-23. Ash content represents the minerals remaining
when moisture and organic matter are driven off from a sample. The study by Lo et al.
(2012) showed that the addition of ash can improve the anaerobic digestion process. Nevert-
heless, ash also can increase the concentration of metals resulting in detrimental effects on
the process (Lo et al., 2010). The value of 3 % of ashes for this waste does not represent re-
levance in the improvement or decline in the production of biogas. On the other hand, when
calculating the combustibility index (VM/FC) of the food waste, a value of 5 was obtained.
A low value like this suggests that the waste has a higher percentage of fixed carbon, which
favors longer combustion and higher caloric power (Rojas-González et al., 2019). The above
suggests that food waste can be incinerated and used as fuel, however, the calorific value
of methane is 53 MJ kg−1 (Scott, 2000), 2.7 times higher than that found in this study (19
MJ kg−1 ), therefore, the anaerobic digestion process for the production of methane suggests
being the most indicated to take advantage of these wastes.

The C/N ratio of food waste is 13. Haug (1993) proposed an optimal ratio of 15-30. The
differences between the elemental composition of food waste are determined by the region
of origin. As mentioned before, this food waste was composed mainly of rice, potatoes, and
chicken, indicating a lower content of animal protein and in turn, high content of carbohy-
drates. As can be seen in Table 2-7, similar methane yields were obtained when the food
waste presents different C/N ratios. Therefore, the substrate for this study indicates the
potential to be used in the anaerobic digestion process in the production of biogas.
2.10 Results and discussion 59

Table. 2-23: Elemental and proximate analysis of food waste

Parameter Unit This study Chae et al. Singh and Yadav
(2020) (2021)
C 47.14 ±0.16 51.74 45.71
H 5.83 ± 0.009 6.79 6.72
% wt.
O 43.34 ± 0.065 36.53 44.66
N 3.68 ± 0.47 3.36 2.91
C/N - 12.81 15.4 15.70
Moisture* % wt. 9.69 ± 0.28 NR NR
Ash 3.01 ± 0.37 9.9 3.62
Volatile matter (VM) % wt. 80.88 ± 0.064 68.9 73.78
Fixed Carbon (FC) 16.11 ± 0.13 12.1 13
−1
Calorific power MJ kg 19.02 ± 0.003 16.87 16.07
NR: Not reporting. wt: weight fraction on dry base. *Pre-dried at 60 °C for 24 h.

2.10.2. Biochemical methane potential of food waste

The behavior of the evolution of biogas production and its quality ( %v/v methane) for the
anaerobic digestion of food waste is shown in Figure 2-9(a). Similarly, the accumulated and
daily methane production is seen in Figure 2-9(b). These data were obtained from the online
measurement of the YieldMaster equipment every 30 minutes for 31 days.

The total volume of biogas generated in 31 days of the process was 306 mL gVS−1 for the
degradation of food waste, with a methane content of 73 % v/v. 80 % of the total biogas
generated is produced after 9 days. Solarte Toro et al. (2017) report that this percentage is
reached after 22 days of digestion, with a total biogas production of 450 mL gVS−1 and with
a theoretical methane content of 52.5 % v/v at 37 °C and 40 days. Similary, Zhang et al.
(2007) obtained 600 mL gVS−1 of biogas and 73 % v/v methane at 50°C after 30 days. For
this study, the high methane content in the biogas produced may be because the food waste
was relatively easy to degrade (low amount of lignocellulosic components) and therefore a
large amount of soluble organic compounds is available for rapid conversion into CH4 , ge-
nerating the high content of CH4 . In turn, this is also related to the low moisture content
and particle size of the substrate. Likewise, according to the data in Table 2-2, an energy
content of 26,13 MJ m−3 could be estimated for the biogas produced with 73 % methane
concentration.
 60 2 Anaerobic Digestion Process

Figure. 2-9: BMP assay by on-line measurement of methane during anaerobic digestion of
food waste. (a) Accumulated biogas production and methane concentration. (b)
Accumulated BMP and daily methane production.

From Figure 2-9(b) it can be seen that a yield of 251 mL gVS−1 of methane is produced.
The highest daily methane production (33 mL gVS−1 ) occurs on day 6, suggesting that the
adaptation time of the microbial biomass present in the inoculum was short, rapidly de-
grading the substrate. This is because the inoculum had an adequate conditioning process.
Methane production increased until day 10 and then remained almost constant at a low level
until the end of the assay (day 31). This pattern can be explained because the substrate,
due to its high biodegradability, begins to quickly deplete. The methane yield generated in
this study (251 mL gVS−1 ) was lower than that reported by Heo, Park, Kang, (2004) who
obtained 489 mL gVS−1 in 40 days under mesophilic conditions using food waste as substrate.

For the BMP test of food waste by on-line measurement of methane, pH and CODs were
monitored at both the beginning and end of the assay. Table 2-24 presents these parameters.
2.10 Results and discussion 61

Regarding pH, the process remained in the optimal range of 6.5-7.5 (Khalid et al., 2011).
This indicates that there was no inhibition by pH, which was to be expected due to the
amount of biogas generated. At the beginning of the test, a few drops of 2N NaOH had
to be added, since the pH of the food waste with inoculum was 6.8. The pH of the blank
was not adjusted. The CODs at the end of the assay shows a decrease concerning the initial
one, giving an indication of degradation of organic matter in the digestion process. Usually,
the percentage of CODs removal efficiency is reported as an indicator of the effectiveness of
the anaerobic treatment on the stabilization of the biodegradable organic waste (Cendales
Ladino, 2011). For this study, at the end of the assay, a CODs removal efficiency of 62 % was
obtained.

Table. 2-24: Control parameters for BMP assay of food waste by on-line measurement
Food Waste Blank (inoculum)
Parameter
Initial Final Initial Final
pH 7.02* 7.257 7.15 7.25
−1
CODs (mg L ) 13354.2 5054.2 2062.5 1829.2
*Value adjusted with NaOH (2N).

Obtaining the initial and final control parameters of the anaerobic digestion process is useful
to give a general evaluation of the process. However, the process can be periodically moni-
tored using sacrifice bottle. A sacrifice bottle assay was performed simultaneously with the
BMP assay by online measurement of methane. Table 2-25 shows the day on which sacrifice
bottle are discarded for their corresponding characterization. This assay aims to evaluate the
behavior of the control parameters (pH, CODs, VFAs) along the anaerobic digestion process.

Table. 2-25: Sample collection during BMP assay using the scarified bottle method
Samples A B C D E F G H I J
Sacrifice day 5 10 12 14 17 20 24 26 28 31

Figure 2-10(a) shows accumulate BMP production for each sample as a function of the sa-
crifice day. As expected, the highest methane yield produced was that of sample J (31 days)
with 129 mL gVS−1 of methane, because it was the last to be discarded (day 31). Sample A (5
sacrifice day) shows the lowest methane yield with 16 mL gVS−1 and after discarding sample
F (20 sacrifice day) the accumulated methane yield stabilizes at an average value of 123.5 mL
gVS−1 . Compared with the yield found in the on-line methane measurement assay (Figure 2-
9 (a)), the yield obtained from sample J is 1.9 times lower. The methane concentration in the
 62 2 Anaerobic Digestion Process

biogas generated by each sample had a stable value throughout the assay, which was 61.9 %
(v/v), 1.18 times lower than that reported in the assay by on-line measurement of methane.
Therefore, the two assays (methane online measurement and sacrifice sample) should not be
interpreted as comparable, but as complementary, being the sacrifice bottle assay that pro-
vided information on the behavior of the relevant variables throughout the process and the
on-line methane measurement assay who provides the methane yield produced by food waste.

Figure. 2-10: BMP assay by sacrifice bottle during anaerobic digestion of food waste. (a)
Accumulate BMP and methane concentration in biogas of each sample on its
sacrifice day. (b) CODs concentration and % CODs removal of each sample on
its sacrifice day.

Organic substances in food waste are degraded and transformed into biogas during the an-
aerobic digestion process, resulting in fluctuations in the concentration of CODs. Figure 8
(b) shows the CODs values obtained for each sacrifice sample, with the percentage of CODs
removal. At the beginning of the trial, the CDOs reported a value of 12.8 g L−1 , while, from
2.10 Results and discussion 63

sample F (day 20) to the end of the assay, a stable average value of 4.6 g L−1 was reported.
A maximum COD removal of 65.6 % was obtained for sample G, which was discarded on day
24. However, after discarding sample C (day 14), it can be seen that the % CODs removal
tends to stabilize and an average value of 61.7 % of CODs removed was obtained. % CODs
removal obtained in this study was lower than the values reported by Ma et al. (2018) who
obtained 89 % CODs removal at 35 °C and 12 days. The high percentage of CODs removal
is mainly since the authors carry out a pretreatment of food waste with fungal mash rich in
various enzymes. Nevertheless, similar values were obtained to those reported by Wijayanti
et al. (2018), who obtained 59,45 % CODs removal at 28°C in 60 days.

Figure. 2-11: pH variation in BMP assay by sacrifice bottle during anaerobic digestion of
food waste. The initial pH values reported were adjusted to pH 7 with 2N HCl.

In Figure 2-11 it can be seen that until sample E (day 17) was discarded, there is a slight
decrease in pH. This decreasing at the beginning of the digestion process is due to volatile
fatty acids production from the substrate. This shows the hydrolysis and acidogenesis stages
 64 2 Anaerobic Digestion Process

in the process. Subsequently, the stability of the pH at an average value of 7.12 suggests that
the acids are decomposed until the end of the assay. The pH range throughout the assay
was 6.9-7.2, which is within the optimal operating range for the anaerobic digestion process
between 6.5-7.5 (Khalid et al., 2011). The behavior of the pH of this study reports similar
results to those reported by Solarte Toro et al. (2017), who reported pH ranges of 6.8-7.5
where the minimum value recorded occurs in the first 5 days of the assay. The fact that the
behavior of the pH did not show large fluctuations, indicates that food waste provides the
necessary nitrogen to control the pH by natural alkalinity as a buffer system since ammonia
reacts with carbon dioxide to produce ammonium bicarbonate. Bicarbonate ions (HCO− 3)
expected to be the main contributors to the alkalinity of the system.

2.10.3. Theoretical Methane Yield

According to the element contents of food waste shown in Table 2-26, the organic matters
in this waste could be expressed as formulations of C15 H22 O10 N, similar to that reported by
Browne and Murphy (2013) for food waste of C16.4 H29 O9.8 N, but different from that repor-
ted by Solarte Toro et al. (2017) C27 H45 O16 N. The similarity or difference in the empirical
formula for food waste is because food preferences and cuisine may vary from one region to
another. Table 2-26 presents the theoretical and experimental BMP, as well as the biodegra-
dability index obtained for the food waste used in the anaerobic digestion process reported
in this study and by various authors.

Table. 2-26: Theoretical and experimental BMP for food waste during anaerobic digestion
Item Unit This study Solarte Toro Browne and
et al. (2017) Murphy
(2013)
Theoretical BMP 439.49 517.06 550
mL CH4 gVS−1
Experimental BMP 251 240 529
Biodegradability % 58.4 46.4 96.1
index (B)

For this study, a biodegradability index of 58.4 % was obtained, 1.64 less than that reported
by Browne and Murphy (2013), who mentions that the high reported value was associated
with acclimatized inoculum and wet samples of food waste. However, it was favorable com-
pared to those reported by Solarte Toro et al. (2017) 46.4 %). This result can be attributed
to the content of elements of the waste, where the authors report almost double the carbon,
2.11 Conclusions 65

obtaining a theoretically optimal C:N ratio (30:1) for biogas production, which generates
that the theoretical BMP is a high value, decreasing the value of Biodegradability index.

2.11. Conclusions
Food waste from the Nariño region presented mainly the presence of rice, potatoes, chicken,
and meat. The characterization of these waste showed great similarity to those reported
in the bibliography, with a VS/TS ratio of 0.94 and a C/N ratio of 12.81. The proximate
analysis found a calorific value of food waste of 19 MJ kg−1 and a low value of combustibility
index (VM / FC) of 5. The experimental results of the anaerobic digestion of food waste
showed a biogas production of 306 mL gVS−1 , and a methane concentration of 73 % at 37
° C in 31 days. The hydrolysis phase occurs in the first 8 days, where the highest daily
methane production was obtained on day 6 (33 mL gVS−1 ), obtaining a methane yield of
251 mL CH4 gVS−1 . A biodegradability index of 58.4 % was obtained, considering that the
theoretical BMP calculated for these waste was 429.49 mL CH4 gVS−1 . Likewise, the con-
trol parameters to monitor digestion show that the percentage of CODs removal was 62 %
and the pH remained within the optimal operating range for the anaerobic digestion (6.5-7.5).

The sacrifice bottle assay was convenient and complementary to the behavior analysis during
the anaerobic digestion process. The pH and COD control helped to identify the hydrolysis
and fermentation phase and rule out inhibitions throughout the process. A similar percen-
tage of COD elimination (62 %) was presented in the two proposed tests, which allowed
complemented the analysis of the biogas production from food waste.

Finally, these results show that this food waste can be used for the production of renewable
energy through the anaerobic digestion process, which, according to this study, the biogas
produced (73 % methane) could theoretically contribute 26.13 MJ m−3 , being able to supply
energy to interconnected areas and, in turn, reducing environmental problems by replacing
firewood with biogas.

The following experimental information served to implement the economic model that com-
pares the economic viability of the production of biomethane and VFAs from food waste by
the anaerobic digestion process: methane concentration of 73 % (v/v) and process pH of 7.2.
References
Abbasi, T., Tauseef, S., and Abbasi, S. (2012). Biogas Energy. Springer, India.

Angelidaki, I., Alves, M., Bolzonella, D., Borzacconi, L., Campos, J. L., Guwy, A. J., Kal-
yuzhnyi, S., Jenicek, P., Van Lier Abstract, J. B., and Van Lier, J. B. (2009). Defining
the biomethane potential (BMP) of solid organic wastes and energy crops: a proposed
protocol for batch assays. Water Science & Technology—WST, 595.

APHA (2008). STANDARD METHODS FOR THE EXAMINATION OF WATER AND

WASTEWATER (11th ed.). American Journal of Public Health and the Nations Health,
51(6):940–940.

Appels, L., Baeyens, J., Degreve, J., and Dewil, R. (2018). Principles and potential of
the anaerobic digestion of waste-activated sludge. Progress in Energy and Combustion
Science., 34.

Atasoy, M., Owusu-Agyeman, I., Plaza, E., and Cetecioglu, Z. (2018). Bio-based volatile fatty
acid production and recovery from waste streams: Current status and future challenges.
Bioresource Technology, 268(July):773–786.

Besselink, H., Brouwer, B., and van der Burg, B. (2017). Validation and regulatory accep-
tance of bio-based approaches to assure feedstock, water & product quality in a bio-based
economy. Ind. Crops Prod, 106:138–145.

Borja, R. (2011). Biogas Production. In Comprehensive Biotechnology, pages 785–798.

Elsevier.

Braun, R. (2007). Anaerobic digestion: A multi-faceted process for energy, environmental

management and rural development. Improvement of Crop Plants for Industrial End Uses,
pages 335–416.

Browne, J. D. and Murphy, J. D. (2013). Assessment of the resource associated with bio-
methane from food waste. Applied Energy, 104:170–177.

Buswell, A. M. and Mueller, H. F. (1952). Mechanism of Methane Fermentation. Industrial

& Engineering Chemistry, 44(3):550–552.
References 67

Cárdenas, L. M., Parra, B. A., Torres, P., and Vásquez, C. H. (2016). Perspectivas del ensayo
de Potencial Bioquı́mico de Metano - PBM para el control del proceso de digestión anae-
robia de residuos Perspectives of Biochemical Methane Potential - BMP test for control
the anaerobic digestion process of wastes Perspectivas do t. Revista ION, 29(1):95–108.

Cendales Ladino (2011). Producción de biogás mediante la co-digestión anaeróbica de la

mezcla de residuos cı́tricos y estiércol bovino para su utilización como fuente de energı́a
renovable. Chemical Engineering Transactions, 49:385–390.

Chae, J. S., Kim, S. W., Lee, J. H., Joo, J. C., and Ohm, T. I. (2020). Combustion charac-
teristics of solid refuse fuel derived from mixture of food and plastic wastes. Journal of
Material Cycles and Waste Management, 22(4):1047–1055.

Consejo Nacional de Polı́tica Económica y Social (2016). Polı́tica Nacional Para La Gestión
De Residuos Sólidos. Documento CONPES 3874, I:73.

Cruz-Salomón, A., Meza-Gordillo, R., Rosales-Quintero, A., Ventura-Canseco, C., Lagunas-

Rivera, S., and Carrasco-Cervantes, J. (2017). Biogas production from a native beverage
vinasse using a modified UASB bioreactor. Fuel, 198:170–174.

Demirel, B. and Scherer, P. (2011). Trace element requirements of agricultural biogas diges-
ters during biological conversion of renewable biomass to methane. Biomass and Bioenergy,
35(3):992–998.

Departamento Nacional de Planeación (2016). PÉRDIDA Y DESPERDICIO DE ALIMEN-

TOS EN COLOMBIA. Estudio de la Dirección de Seguimiento y Evaluación de Polı́ticas
Públicas.

Esposito, G., Frunzo, L., Panico, A., and Pirozzi, F. (2012). Bio-Methane Potential Tests
To Measure The Biogas Production From The Digestion and Co-Digestion of Complex
Organic Substrates. Environmental Technology (United Kingdom), 33(24):2733–2740.

Esteves, S. and Devlin, D. (2010). Food Waste Chemical Analysis. Material Change for a
better enviroment.

Feng, L., Wang, H., Chen, Y., and Wang, Q. (2009). Effect of solids retention time and
temperature on waste activated sludge hydrolysis and short-chain fatty acids accumulation
under alkaline conditions in continuous-flow reactors. Bioresour. Technol., 100:44–49.

Ferrer, I., Vázquez, F., and Font, X. (2010). Long term operation of a thermophilic anaerobic
reactor: Process stability and efficiency at decreasing sludge retention time. Bioresour.
Technol, 101:2972–2980.
 68 References

Fregoso Soria, M. D. J., Ferrera-cerrato, R., Barra, J. E., González, A., Santos, J. T., Borges,
L., and Pereyda, G. (2001). Producción de biofertilizantes mediante biodigestión de excreta
liquida de cerdo. Terra Latinoamericana, 19(4):353–362.

Fujihira, T., Seo, S., Yamaguchi, T., Hatamoto, M., and Tanikawa, D. (2018). High-rate
anaerobic treatment system for solid/lipid-rich wastewater using anaerobic baffled reactor
with scum recovery. Bioresource Technology, 263:145–152.

Gavala, H., Angelidaki, I., and Ahring, B. (2003). Kinetics and modeling of anaerobic
digestion process. Advances in biochemical engineering/biotechnology, 81:58–93.

Gonzalez-Fernandez, C., Sialve, B., and Molinuevo-Salces, B. (2015). Anaerobic digestion of

microalgal biomass: Challenges, opportunities and research needs. Bioresource Technology,
198:896–906.

Grimberg, S., Hilderbrandt, D., Kinnunen, M., and Rogers, S. (2015). Anaerobic digestion of
food waste through the operation of a mesophilic two-phase pilot scale digester – Assess-
ment of variable loadings on system performance. Bioresource Technology, 178:226–229.

Gruhn, M., Frigon, J., and Guiot, S. (2016). Acidogenic fermentation of Scenedesmus sp.-
AMDD: comparison of volatile fatty acids yields between mesophilic and thermophilic
conditions. Bioresour. Technol, 200:624–630.

Guo, Z., Liu, W., Yang, C., Gao, L., Thangavel, S., Wang, L., He, Z., Cai, W., and Wang,
A. (2017). Computational and experimental analysis of organic degradation positively
regulated by bioelectrochemistry in an anaerobic bioreactor system.

Hajji, A. and Rhachi, M. (2013). The influence of particle size on the performance of
anaerobic digestion of municipal solid waste. Energy Procedia, 36:515–520.

Han, W., Fang, J., Liu, Z., and Tang, J. (2016). Techno-economic evaluation of a combined
bioprocess for fermentative hydrogen production from food waste. Bioresource Technology,
202:107–112.

Haug, R. (1993). The Practical Handbook of Compost Engineering. USA, lewis publ edition.

Horan, N., Zahrim, A., and Newati, Y. (2018). Anaerobic Digestion Processes Applications
and Effluent Treatment. Cham, Switzerland, springer edition.

House, D. and Surratt, V. (2013). The biogas handbook: Science, production and applications.
Woodhead Publishing Limited, Cambridge, UK.

Huang, X., Shen, Y., Luo, H., Liu, J., and Liu, J. (2018). Enhancement of extracellular lipid
production by oleaginous yeast through preculture and sequencing batch culture strategy
with acetic acid. Bioresour. Technol., 247(395-401).
References 69

Hulshoff Pol, L., de Castro Lopes, S., Lettinga, G., and Lens, P. (2004). Anaerobic sludge
granulation. Water Research, 38(6):1376–1389.

Ishida, M., Haga, R., and Odawara, Y. (1982). Anaerobic digestion process. Google Patents.

Jankowska, E., Chwialkowska, J., Stodolny, M., and Oleskowicz-Popiel, P. (2017). Volati-
le fatty acids production during mixed culture fermentation – The impact of substrate
complexity and pH. Chemical Engineering Journal, 326:901–910.

Jiang, J., Zhang, Y., Li, K., Wang, Q., Gong, C., and Li, M. (2013). Volatile fatty acids pro-
duction from food waste: Effects of pH, temperature, and organic loading rate. Bioresour.
Technol, 143:525–530.

Jin, D., Chen, J., and Lun, S. (1999). Production of poly(hydroxyalkanoate) by a composite
anaerobic acidification–fermentation system. Process Biochem, 34:829–833.

Khalid, A., Arshad, M., Anjum, M., Mahmood, T., and Dawson, L. (2011). The anaerobic
digestion of solid organic waste. Waste management, 31:1737–1744.

Khanal, S. K. (2008). Anaerobic Reactor Configurations for Bioenergy Production. In Anae-

robic Biotechnology for Bioenergy Production, chapter 5, pages 93–114. Wiley-Blackwell,
Oxford, UK.

Kim, W., Shin, S., Lim, J., and Hwang, S. (2013). Effect of temperature and hydraulic
retention time on volatile fatty acid production based on bacterial community structure
in anaerobic acidogenesis using swine wastewater. Bioprocess Biosyst, 36(6):791–798.

Kumar, V. and Tuohy, M. G. (2018). Biogas Fundamentals, Process, and Operation. Cham,
Switzerland, springer edition.

Kumaran, P., Hephzibah, D., Sivasankari, R., Saifuddin, N., and Shamsuddin, A. H. (2016).
A review on industrial scale anaerobic digestion systems deployment in Malaysia: Oppor-
tunities and challenges. Renewable and Sustainable Energy Reviews, 56:929–940.

Lee, W. S., Chua, A. S. M., Yeoh, H. K., and Ngoh, G. C. (2014). A review of the production
and applications of waste-derived volatile fatty acids. Chemical Engineering Journal,
235:83–99.

Lier, J. B., Rebac, S., and Lettinga, G. (1997). High-rate anaerobic wastewater treatment
under psychrophilic and thermophilic conditions. Water Science and Technology, pages
199–206.

Lim, S., Kim, B., Jeong, C., Choi, J., Ahn, Y., and Chang, H. (2008). Anaerobic organic
acid production of food waste in once-a-day feeding and drawing-off bioreactor. Bioresour
Technol, 99 (16):7866–7874.
 70 References

Lin, C. S. K., Pfaltzgraff, L. A., Herrero-Davila, L., Mubofu, E. B., Abderrahim, S., Clark,
J. H., Koutinas, A. A., Kopsahelis, N., Stamatelatou, K., Dickson, F., Thankappan, S.,
Mohamed, Z., Brocklesby, R., and Luque, R. (2013). Food waste as a valuable resource for
the production of chemicals, materials and fuels. Current situation and global perspective.
Energy and Environmental Science, 6(2):426–464.

Liu, H., Han, P., Liu, H., Zhou, G., Fu, B., and Zheng, Z. (2018). Full-scale production
of VFAs from sewage sludge by anaerobic alkaline fermentation to improve biological
nutrients removal in domestic wastewater. Bioresource Technology, 260(February):105–
114.

Liu, H., Wang, J., Liu, X., Fu, B., Chen, J., and Yu, H. (2012). Acidogenic fermentation of
proteinaceous sewage sludge: effect of pH. Water Res., 46 (3):799–807.

Lo, H. M., Chiu, H. Y., Lo, S. W., and Lo, F. C. (2012). Effects of different SRT on anaerobic
digestion of MSW dosed with various MSWI ashes. Bioresource Technology, 125:233–238.

Lo, H. M., Kurniawan, T. A., Sillanpää, M. E., Pai, T. Y., Chiang, C. F., Chao, K. P., Liu,
M. H., Chuang, S. H., Banks, C. J., Wang, S. C., Lin, K. C., Lin, C. Y., Liu, W. F., Cheng,
P. H., Chen, C. K., Chiu, H. Y., and Wu, H. Y. (2010). Modeling biogas production
from organic fraction of MSW co-digested with MSWI ashes in anaerobic bioreactors.
Bioresource Technology, 101(16):6329–6335.

López-Hernández, J. E., Ramı́rez-Higareda, B. L., Gomes-Cabral, C. B., and Morgan-

Sagastume, J. M. (2017). Guı́a técnica para el manejo y aprovechamiento de biogás en
plantas de tratamiento de aguas residuales.

Ma, Y., Gu, J., and Liu, Y. (2018). Evaluation of anaerobic digestion of food waste and
waste activated sludge: Soluble COD versus its chemical composition. Science of the Total
Environment, 643:21–27.

Macarthur, E. (2006). Towards a Circular Economy. Journal of Industrial Ecology, 10(1-

2):4–8.

Manchala, K. R., Sun, Y., Zhang, D., and Wang, Z.-W. (2017). Anaerobic Digestion Mode-
lling, volume 2. Elsevier Ltd.

Mao, C., Feng, Y., Wang, X., and Ren, G. (2015). Review on research achievements of biogas
from anaerobic digestion. Renewable and Sustainable Energy Reviews, 45:540–555.

Meng, Y., Jost, C., Mumme, J., Wang, K., and Linke, B. (2016). Oxygen tolerance capacity
of upflow anaerobic solid-state (UASS) with anaerobic filter (AF) system. Journal of
Environmental Sciences, 45:200–206.
References 71

Mir, M. A., Hussain, A., and Verma, C. (2016). Design considerations and operational
performance of anaerobic digester: A review. Cogent Eng, 3:1–20.

Moo-Young, M. (2011). Comprehensive biotechnology. Elsevier.

Murphy, J. D. and Thamsiriroj, T. (2013). Fundamental science and engineering of the anae-
robic digestion process for biogas production. The Biogas Handbook: Science, Production
and Applications, pages 104–130.

Nguyen, H. H. (2014). Modelling of food waste digestion using ADM1 integrated with Aspen
Plus.

Owen, W. F., Stuckey, D. C., Healy, J. B., Young, L. Y., and McCarty, P. L. (1979). Bioassay
for monitoring biochemical methane potential and anaerobic toxicity. Water Research,
13(6):485–492.

Pramanik, S. K., Suja, F. B., Porhemmat, M., and Pramanik, B. K. (2019). Performance and
kinetic model of a single-stage anaerobic digestion system operated at different successive
operating stages for the treatment of food waste. Processes, 7(9).

Rico, C., Montes, J. A., and Rico, J. L. (2017). Evaluation of different types of anaerobic seed
sludge for the high rate anaerobic digestion of pig slurry in UASB reactors. Bioresource
Technology, 238:147–156.

Rojas-González, A. F., Flórez-Montes, C., and López-Rodrı́guez, D. F. (2019). Prospectivas

de aprovechamiento de algunos residuos agroindustriales. Revista Cubana de Quı́mica,
31(1):31–52.

Salazar, J. D. (2019). Guı́a metodológica para el manejo y aprovechamiento de biosólidos en

Colombia. Colombia, universida edition.

Sauvé, S., Bernard, S., and Sloan, P. (2016). Environmental sciences, sustainable develop-
ment and circular economy: Alternative concepts for trans-disciplinary research. Environ-
mental Development, 17:48–56.

Scott, G. (2000). Green polymers (Review). Polymer Degradation and Stability, 68:1–7.

Sharma SK, Mishra IM, Sharma MP, S. J. (1988). Effect of particle size on biogas generation
from biomass residues. Biomass, 17(4):251–63.

Shen, J., Zhu, S., Liu, X., Zhang, H., and Tan, J. (2010). The prediction of elemental
composition of biomass based on proximate analysis. Energy Conversion and Management,
51:983p–987p.
 72 References

Silva, F., Serafim, L., Nadais, H., Arroja, L., and Capela, I. (2013). Acidogenic fermentation
towards valorisation of organic waste streams into volatile fatty acids. Chem. Biochem,
Q.27:467–476.

Singh, D. and Yadav, S. (2021). Steam gasification with torrefaction as pretreatment to

enhance syngas production from mixed food waste. Journal of Environmental Chemical
Engineering, 9(1):104722.

Solarte Toro, J. C., Mariscal Moreno, J. P., and Aristizábal Zuluaga, B. H. (2017). Evaluación
de la digestión y co-digestión anaerobia de residuos de comida y de poda en bioreactores
a escala laboratorio. Revista Ion, 30:105–116.

Sosnowski, P., Wieczorek, A., and Ledakowicz, S. (2003). Anaerobic co-digestion of sewage
sludge and organic fraction of municipal solid wastes. Advances in Environmental Research,
7(3):609–616.

Steffen, R., Szolar, O., and Braun, R. (1998). Feedstocks for Anaerobic Digestion.

Strassburg, B., Turner, R., Fisher, B., Schaeffer, R., and Lovett, A. (2009). Reducing emis-
sions from deforestation – the combined incentives mechanism and empirical simulations.
Global Environ Change, 19:265(78).

Strazzera, G., Battista, F., Garcia, N. H., Frison, N., and Bolzonella, D. (2018). Volatile
fatty acids production from food wastes for biorefinery platforms: A review. Journal of
Environmental Management, 226(August):278–288.

Superintendencia de Servicios Públicos Domiciliarios (2016). Disposición Final de Residuos

Sólidos - Informe Nacional. I Simposio Iberoamericano de I Simposio Iberoamericano de
Ingenierı́a de Residuos, 2015(2012-2015):181.

Surendra, K. C., Takara, D., Hashimoto, A. G., and Khanal, S. K. (2014). Biogas as a sus-
tainable energy source for developing countries: Opportunities and challenges. Renewable
and Sustainable Energy Reviews, 31:846–859.

Toerien, D. F. and Hattingh, W. H. (1969). Anaerobic digestion I. The microbiology of

anaerobic digestion. Water Research, 3(6):385–416.

Topa, N., Mizoue, N., Kaib, S., and Nakao, T. (2004). Variation in woodfuel consumption
patterns in response to forest availability in Kampong Thom Province, Cambodia. Biomass
Bioenergy, 27(57-78).

Uçkun Kiran, E. and Liu, Y. (2015). Bioethanol production from mixed food waste by an
effective enzymatic pretreatment. Fuel, 159:463–469.
References 73

Uçkun Kiran, E., Stamatelatou, K., Antonopoulou, G., and Lyberatos, G. (2016). Production
of biogas via anaerobic digestion.

Van Ginkel, S., Sung, S., and Lay, J. J. (2001). Biohydrogen production as a function of pH
and substrate concentration. Environmental science & technology, 35:4726–4730.

Varnero, M. and Arellano, J. (1990). Aprovechamiento racional de desechos orgánicos. Tech-

nical report, Santiago, Chile.

Varnero Moreno, M. T. (2011). Manual de Biogás. Santiago de Chile.

VDI-Handbuch Technik Biomasse (2016). VDI 4630 - Fermentation of organic materials –

Characterization of the substrate, sampling, collection of material data and fermentation
test.

Wang, K., Yin, J., Shen, D., and Li, N. (2014). Anaerobic digestion of food waste for volatile
fatty acids (VFAs) production with different types of inoculum: Effect of pH. Bioresource
Technology, 161:395–401.

Wijayanti, D. F., Suwartha, N., and Priadi, C. R. (2018). Effect of the addition of fat oil
and grease (FOG) on the performance of a dry anaerobic digestion food waste reactor.
International Journal of Technology, 9(2):267–274.

Zacharof, M. P. and Lovitt, R. W. (2013). Complex effluent streams as a potential source of

volatile fatty acids. Waste and Biomass Valorization, 4(3):557–581.

Zhang, C., Su, H., Baeyens, J., and Tan, T. (2014). Reviewing the anaerobic digestion of
food waste for biogas production. Renewable and Sustainable Energy Reviews, 38:383–392.

Zhang, C., Su, H., and Tan, T. (2013). Batch and semi-continuous anaerobic digestion of
food waste in a dual solid–liquid system. Bioresource Technology, 145:10–16.

Zhang, L., Lee, Y.-W., and Jahng, D. (2011). Anaerobic co-digestion of food waste and
piggery wastewater: Focusing on the role of trace elements. Bioresource Technology,
102(8):5048–5059.

Zhang, R., El-Mashad, H. M., Hartman, K., Wang, F., Liu, G., Choate, C., and Gamble, P.
(2007). Characterization of food waste as feedstock for anaerobic digestion. Bioresource
Technology, 98(4):929–935.

Zhao, J., Wang, D., Liu, Y., Ngo, H., Guo, W., Yang, Q., and Li, X. (2018). Novel stepwise
pH control strategy to improve short chain fatty acid production from sludge anaerobic
fermentation. Bioresour. Technol., 249:431–438.
 74 References

Zhou, M., Yan, B., Wong, J. W., and Zhang, Y. (2018). Enhanced volatile fatty acids pro-
duction from anaerobic fermentation of food waste: A mini-review focusing on acidogenic
metabolic pathways. Bioresource Technology, 248:68–78.
3 Techno-economic evaluation of
anaerobic digestion beyond biogas
In this chapter, a technical comparison and an economic evaluation of the products that
can be generated in the anaerobic digestion process was carried out. First, biomethane is
presented, and on the other hand platform products such as volatile fatty acids (VFAs) are
obtained. For this analysis, the Net Present Value method was used under different condi-
tions of supply (available food waste) and demand (natural gas consumption in the transport
sector and VFAs exports) in a Colombian context. Besides, the results of this chapter serve
as a perspective towards the enhancement of the VFAs production, justifying the study of
new methodologies to guide the anaerobic digestion process towards the production of plat-
form compounds. The content of this chapter is presented in article format.
 76 3 Techno-economic evaluation of anaerobic digestion beyond biogas

TECHNO-ECONOMIC EVALUATION OF
ANAEROBIC DIGESTION BEYOND BIOGAS
Oscar Dario Yepez−Cerona , Beatriz Aristizabalb ,Oscar Andrés Prado−Rubioc
Departamento de Ingenierı́a Quı́mica, Universidad Nacional de Colombia, Manizales,
Caldas,170001, Colombia.
Grupo de Investigación en Aplicación de Nuevas Tecnologı́as, GIANTa,c
Grupo de Trabajo Académico en Ingenierı́a Hidráulica y Ambiental, GTAIHAb
a
odyepezc@[Link], b bharistizabalz@[Link], c oaprador@[Link]

ABSTRACT
The anaerobic digestion process has been an object of study and interest mainly for the pro-
duction of biogas, referred to as biomethane. However, the production of platform products,
such as volatile fatty acids (VFAs), has a growing interest due to its high economic potential
and positive environmental impact by harnessing organic waste within the circular economy
philosophy. In the present work, a rigorous techno-economic analysis is performed comparing
production scenarios for anaerobic digestion using food waste as substrate in the Colombian
context. The Net Present Value (NPV) index was used to evaluate the economic viability
of the process when producing biomethane and VFAs, using as inputs construction costs,
logistics and services costs, amount of substrate available, inflation rate and product selling
prices, among others. Eight production scenarios were investigated at different scales from
the available food waste and ve production scenarios for the demand of potential products.
It was found that the anaerobic digestion process oriented towards VFAs production can
be 11 times more lucrative than biomethane generation. In turn, biomethane could replace
and satisfy as much as 15 % (545455 m3 biomethane year−1 ) of natural gas consumption in
the transport sector and 5 % (378 Ton VFAs year−1 ) of VFAs exports in Colombia, being
also an interesting production scenario. These results indicate that the anaerobic digestion
process could contribute to harnessing food waste for the production of renewable energy
and high-added platform components and these scenarios can be extrapolated to other de-
veloping countries.

Keywords: Anaerobic Digestion; Biomethane; Volatile Fatty Acids (VFAs); Techno-economic

analysis; Viability.

3.1. Introduction
Currently, the Latin American economy can be considered basically based on the linear con-
cept, where goods from raw materials are bought by consumers. Once the products have
3.1 Introduction 77

been used or consumed, consumers discard the remaining materials, without being reused.
As the world population grows and new industrial and developed areas expand, it is evident
that linear economy is already producing scarcity of commodities and energy, including food.
This is leading to financial hardship, human suffering, and conflict (Sariatli, 2017).

In accordance with the 2030 Agenda for Sustainable Development and the Sustainable De-
velopment Goals (SDGs), Goal 7 includes: “Ensure access to affordable, reliable, sustainable
and modern energy for all”. Therefore, the sustainable management of the environment and
natural resources is a commitment assumed for economic growth and human well-being.
Only through proper management of natural resources, a society can develop in a sustai-
nable way, with a healthy ecosystem where water, air, and land are not contaminated as
a result of residual discharges generated in the linear economy. In this sense, the circular
economy is proposed as an alternative to avoid the drastic consequences that may arise from
continuing to apply the linear economy (Macarthur, 2006).

The circular economy is defined as an economic-environmental strategy that aims to reduce

both the entry of materials and the production of waste, creating a closed-loop or circle whe-
re the materials that have been discarded are treated to reduce re-entering the production
system (Sauvé et al., 2016). The circular economy is an economic concept directly related
to sustainability. Its main objective is that the value of products, materials, and natural
resources remain in the economy for as long as possible, considerably reducing the genera-
tion of waste (Rizos et al., 2017). Anaerobic digestion process matches within this concept
since it transforms bio-based waste such as food waste, pruning waste, household waste, solid
manure, wastewater, agro-industrial waste, among others, into valuable products. Anaero-
bic digestion is carried out in hermetic containers called biodigesters. The bioprocess takes
advantage of the degrading action of the biological material that certain microorganisms
carry it out in the absence of oxygen, conventionally aimed to obtain biogas, which is mainly
composed of carbon dioxide (40 %) and methane (60 %) (House and Surratt, 2013). Biogas
is a renewable energy resource that does not add dioxide carbon load in the atmosphere and
it has many uses in a sustainable society. Biogas can be used as a fuel for cooking, for heat
and electricity production through co-generation (Hengeveld et al., 2016; Guo et al., 2017),
upgraded biogas (biomethane) is used as vehicle fuel (Cucchiella et al., 2017; Balkenhoff
et al., 2010) and also in injection into the gas grid (Wall et al., 2018; Aryal and Kvist, 2018).

During the anaerobic digestion process, four stages are carried out: the hydrolysis of car-
bohydrates in simpler soluble compounds that, later through acidogenesis, are metabolized
producing volatile acids. Subsequently, in the acetogenesis stage, these compounds are trans-
formed into acetic acid, hydrogen and carbon dioxide, to finally obtain the main compound of
interest in methanogenesis, biogas (Diaz-Baez et al., 2002). Once the organic matter complies
with the anaerobic digestion process and the biogas is extracted, a wet organic residue called
 78 3 Techno-economic evaluation of anaerobic digestion beyond biogas

digestate or biol is produced, which is pumped out of the biodigester. Digestate consists
of slow degradable, stable organic components such as lignin, nitrogen, and phosphorous
in various forms, inorganic salts containing phosphate, ammonium, potassium, and other
minerals (Fagerström et al., 2018). Digestate has no bad smell, does not attract flies, and
can be applied directly to the field in liquid form as fertilizer, as long as it complies with the
soil application regulations.

An alternative approach to increase anaerobic digestion profitability is to target the produc-

tion of Volatile Fatty Acids (VFAs): acetic acid, propionic acid, butyric acid, valeric acid
and caproic acid (Wang et al., 2014). Previous research indicate that pH, temperature, type
of substrate and retention time are parameters that favor the production of VFAs instead of
biogas (Feng et al., 2009; Lee et al., 2014; Silva et al., 2013; Jiang et al., 2013). The direct
recovery of these products from fermentation broth or the subsequent processing to produce
other molecules (for example, polyhydroxyalkanoates (PHAs) or fatty acids of medium chain
length), can result in valorization of the process beyond biogas and digestate (Kleerebezem
et al., 2015). As advantages, the storage and transport of VFAs are easier and safer than
biogas, and the added value of VFAs is 1.650 USD ton−1 (Calt, 2015), greater than the
value of purified biogas (95 % methane), which is 996 USD ton−1 (Zhou et al., 2018). VFAs
have a wide range of applications, valuable industrial chemicals for cosmetics production,
the pharmaceutical industry, petrochemical synthesis, and the food and beverage industries
(Zacharof and Lovitt, 2013). Additionally, VFAs production has created an opportunity for
novel applications, such as polyhydroxyalkanoate biopolymers (PHA) production (Bluemink
et al., 2016) nutrient removal in wastewater treatment plants (Lim et al., 2008), chain elon-
gation (Cabrera Rodrı́guez et al., 2017), or bulk fuel and solvent production (Agler et al.,
2011). Therefore, VFAs have the potential to be one of the main platforms in the new bio-
refineries, which can be used to produce biofuels and valuable chemicals by transforming
biomass-based raw materials (Satinder et al., 2017).

In this sense, focusing on the production of VFAs through the anaerobic digestion process
could turn the process into a system with a higher net income compared to the process des-
tined to generate biogas. The process focused on biogas production has a limited financial
profitability because biogas is not as efficient as fossil fuels, the biogas purification process
is costly and the biogas is cheaper than VFAs. Despite VFAs potential, there is few techno-
economical assessment studies focused on the production of VFAs through the anaerobic
digestion process, and even less on the economic comparison of the production of biogas and
VFAs by this process (Kleerebezem et al., 2015).

Therefore, this contribution aims to perform a rigorous techno-economic analysis of the an-
aerobic digestion comparing the net economic income depending on the product of interest
to provide a comprehensive view of the conventional anaerobic digestion process (biogas pro-
3.2 Methodology 79

duction), assessing the economic advantages and disadvantages of VFAs recovery compared
to biogas production. The investigated production scenarios are based on substrate offer and
products demand. This assessment shows the anaerobic digestion economic viability in short
and long term as a function of the production scale. The simulation results are particularly
relevant since by the year 2025, 2.2 billion tons per year of urban municipal waste will be
generated in the world, 46 % of which represents organic waste, which could be used by
implementing technologies that promote the circular economy (Hoornweg and Bhada Ta-
ta, 2012). Moving towards a more circular economy could deliver benefits such as reducing
environmental issues, preserve the supply of raw materials, increasing competitiveness, sti-
mulating innovation, boosting economic growth, and creating jobs.

3.2. Methodology
In the present work, an a model-based techno-economic assessment in Colombian context is
performed using technologies validated for the production of biomethane and VFAs through
anaerobic digestion, using food waste as substrate. Food waste was chosen as substrates
because of their high energy content (oriented to biogas production), low buffer capacity
(oriented to VFAs production), high degradability, and large quantity and abundant availa-
bility (Zhang et al., 2014). In addition, food waste has better yields for biogas production
and VFAs production compared to other types of waste (House and Surratt, 2013; Mao et al.,
2015; Zhou et al., 2018).

From a general perspective, the same base process was considered but with a different ob-
jective product. The first was focused on biomethane production and the second on VFAs
production (acetic acid, propionic acid, butyric acid and valeric acid). Depending on the
focus, the process varies in design parameters such as: solids retention time (determines the
size of the biodigester), operating pH, operating temperature, and volatile solids content of
the substrate.

For the economic analysis and scaling up of anaerobic digestion, two approaches are con-
sidered. First, the amount of food waste available in Colombia (supply) to be used as a
substrate in the anaerobic digestion process. Second, the demand for biomethane and VFAs
in Colombia. When biomethane is produced, the income was calculated by comparing the
consumption of natural gas in the transportation sector in Colombia, and when VFAs is
generated, the income was determined by comparing the export of industrial monocarboxy-
lic fatty acids. Figure 3-1 shows the flow diagram that summarizes the methodology of the
study carried out. It consists of 4 main steps: 1) Systematic literature review, 2) Design of the
process, 3) Economic model development and 4) Scenario based techno-economic analysis.
Each main step composed of complementary items that are developed for the implementa-
 80 3 Techno-economic evaluation of anaerobic digestion beyond biogas

tion and analysis.

Figure. 3-1: Methodology for the economic study of the anaerobic digestion process.

3.2.1. Economic potential evaluation

A systematic literature review was carried out to identify the optimal conditions of the an-
aerobic digestion process, using food waste as substrate, in order to obtain the maximum
yields for biogas and VFAs reported. For this purpose, 3 scientific search engines were used to
find information: Science Direct, Scopus, and Web of Science. The application of the scheme
to explore databases was carried out using a search equation that limits the results to the
words: “anaerobic digestion”, ”biogas”, “volatile fatty acids”, “VFAs”, “methane, “biomet-
hane”, “yield”, ”production”, “efficiency”.

The economic feasibility is evaluated through the Net Present Value (NPV) index, which
makes possible to analyze the profitability of a long-term investment. NPV (measured in
USD) is defined as the sum of the present values of the individual cash flows, and it consi-
ders only cash inflows and outflows. In addition, the determination of these flows is based
on the incremental approach. This method is described below (Gebrezgabher et al., 2010):

n
X It − Ot
N P V = −Io + (3-1)
t=1
(1 + IRR)t
3.2 Methodology 81

Where Io : Initial investment, It : Discounted cash inflows, Ot : Discounted cash outflows, n:

Lifetime of investment, t: Time of the cash flow, and IRR: Internal Rate of Return.

a. Cash flows
The costs, revenues, and investment calculations are based on the methodology used
by Cucchiella and D’Adamo (2016) and Ferella et al. (2019). For the revenues side
(discounted cash inflows), purified biogas (biomethane) is considered as vehicle fuel.
While for the VFAs production process, the main revenue was from the sale of the sum
of individual VFAs. Additionally, another part of the revenues is represented by the
disposal of organic waste. In Colombia, the final disposal of urban waste in landfills
is a paid service and for this study, it was taken into account as income for the two
processes.

For the costs side (discounted cash outflows), three phases were considered: (i) subs-
trate acquisition (transport), (ii) production (maintenance, operation, and services),
and (iii): purification (biogas) or separation (VFAs). Figure 3-2 depicts these 3 pha-
ses of biomethane process, where the acquisition of substrate, the production stage
in a Chinese-type biodigester and the purification stage are identified. Likewise, for
the process to obtain VFAs, the acquisition of substrate, the production (Chinese-type
biodigester) and the separation of the product stage are observed. For both processes,
the initial investment costs are the construction of a Chinese type concrete biodigester
with integrated polyester membrane gas-holder, with two concrete tanks for the stora-
ge of the input and output of the substrate.

For the first process, the biogas purification system was carried out using a non-rough
polymer membrane, which uses the driving force generated by pressure difference for
separation (Cucchiella and D’Adamo, 2016). While for the second process, VFAs se-
paration was performed by an electrodialysis process (Moresi and Sappino, 2000). For
the substrate acquisition, only the cost of transporting the waste to the biodigester
was considered. For both biomethane production and VFAs production, the estimated
costs were: maintenance and operation of the biodigester, cost of electricity service and
the biodigester depreciation. For the purification of biogas into biomethane, the cost
of maintenance and operation of the purifier, the cost of electricity service of the puri-
fier, and depreciation of the purifier were considered. For the separation of the VFAs,
the costs of maintenance and operation, the cost of electricity, and the depreciation of
the electrodialysis system were included. Finally, labor costs were considered in both
processes. For one worker, it was estimated 1 legal minimum wage in force, along with
a transport subsidy, as stipulated by the Colombian standards.
 82 3 Techno-economic evaluation of anaerobic digestion beyond biogas

Figure. 3-2: Phases of the anaerobic digestion process from food waste for the production
of: (a) biomethane (b) VFAs.

b. Economic model description

In order to use equation 3-1, the economic model was implemented in Microsoft Excel
and corresponds to an adjusted version of the model to the economic model proposed
by Cucchiella et al. (2018), where the main differences lies on the scale of the process,
local economic indices, and the design of separation and purification processes.

Process oriented to the production of biomethane

inv inv
Io = CB+ST + CBP (3-2)
It = RF W D + RBiomethane (3-3)
Ot = CBOM + CBd,t + CBe,t + CL,t + CF W T,t + CP OM,t + CP d,t + CP e,t (3-4)

Process oriented to the production of VFAs

3.2 Methodology 83

inv
Io = CB+ST + CEinv (3-5)
It = RF W D + RV F A (3-6)
Ot = CBOM + CBd,t + CBe,t + CL,t + CF W T,t + CEOM,t + CEd,t + CEe,t (3-7)

Initial investment: Biodigester and storage tanks

SRB = (ηV SR ) (HB ) (N WSR ) + 2π (ηhSR ) (rB ) (N WSR ) (3-8)

4πrB
ηV SR = (3-9)
dSR
2HB
ηhSR = (3-10)
dSR
SRST = (ηV SRST ) (eST ) (N WSR ) + 4 (ηHSRST ) (eST ) (N WSR ) (3-11)
16 (eST )
ηV SRST = (3-12)
dSR
4 (eST )
ηHSRST = (3-13)
dSR
VST C = 10 (eST )2 (tB ) (3-14)
(F W ) (fo )
VST = (3-15)
DF W
aM = 2π (rB )2 (3-16)

Initial investment: Biogas Purification

inv
CBP = (CU BP ) (Sbiomethane ) (3-17)
(F W ) (V SF W ) (Ymethane )
Sbiomethane = (3-18)
24000DF W

Initial investment: VFAs separation - electrodialysis equipment

CEinv = CEU + CEM + CEC (3-19)

CEU = α + βAM (3-20)
(ef ) (SV F A )
AM = (3-21)
2 (tw ) (Js )
ηL (F W ) (YV F A ) (SRT )
SV F A = (3-22)
DF W
CEM = γAM (3-23)
CEC = δAM (3-24)
84 3 Techno-economic evaluation of anaerobic digestion beyond biogas

Revenue: Final disposal of organic waste

RF W D = (F W ) (RF W DU ) (tW ) (3-25)

Revenue: Biomethane sale

RBiomethane = 24 (SBiomethane ) (PBiomethane ) (tw ) (3-26)

PBiomethane = (PP T V )(Cf gas )(1 − %rpP T V ) (3-27)

Revenue: VFAs sale

RV F A = 12 (SV F A ) (PV F A ) (3-28)

Costs: Production costs

inv
(OM F ) (1 + RP I OM )t−1


CBOM = CB+ST (3-29)
inv


CB+ST (Pdf )
CBd,t = (3-30)
100ηdebt
 
inf
CBd,t+1 = CBd,t 1 + (3-31)
100
CBe,t+1 = (CU eB ) (Qbiogas ) (Pe ) (3-32)
100
QBiogas = (SBiomethane ) (tw ) (3-33)
60  
inf
CBe,t+1 = CBe,t 1 + (3-34)
100
CL,t = 12 (CU L ) (ηop ) (3-35)
 
inf
CL,t+1 = CL,t 1 + (3-36)
100
CF W T,t = (CU F W T ) (F W ) (tw ) (3-37)
 
inf
CF W T,t+1 = CF W T,t 1 + (3-38)
100

Costs: Biogas Purification

inv


CP OM,t = CBP (Pmo ) (3-39)
 
inf
CP OM,t+1 = CP OM,t 1 + (3-40)
100
(3-41)
3.2 Methodology 85

inv
(CBP ) (Pdf )
CP d,t = (3-42)
100ηdebt
 
inf
CP d,t+1 = CP d,t 1 + (3-43)
100
CP e,t = (CU eP ) (QBiogas ) (Pe ) (3-44)
 
inf
CP e,t+1 = CP e,t 1 + (3-45)
100
(3-46)

Costs: VFAs separation - electrodialysis equipment

CEM
CEOM,t = 0.03 CEinv + (3-47)
ηm
 
inf
CEOM,t+1 = CEOM,t 1 + (3-48)
100
inv
C
CEd,t = E (3-49)
ηdebt
 
inf
CEd,t+1 = CEd,t 1 + (3-50)
100
CEe,t = 12 (EW h ) (PE ) (Pe ) (3-51)
SV F A
EW h = (3-52)
(Js ) (AM )
 
inf
CEe,t+1 = CEe,t+1 1 + (3-53)
100

c. Model assumptions and limitations

For both processes, the initial investment, revenues, and costs were estimated at the
Colombian setting. The investment cost was covered with bank financing. An Internal
Rate of Return (IRR) of 10 % was chosen and the lifetime of investment was 15 years.
Eight cases were investigated for the substrate supply, i.e. the production of these
compounds (biomethane and VFAs) depends on the available substrate. The substra-
te was estimated from the food waste that can produce a small city to a large city in
Colombia. Inflow varied between 5 Ton day −1 (small city) to 200 Ton day −1 (large city).

On the other hand, for the production of biomethane and VFAs, five cases were in-
vestigated based on the demand for these products. For the process oriented to the
production of biomethane, studies were carried out to satisfy between 0.1 % and 15 %
of the demand for natural gas in the transportation sector for Colombia in 2019. On
the other hand, for the process oriented to the production of VFAs, between 5 % and
 86 3 Techno-economic evaluation of anaerobic digestion beyond biogas

50 % of the export of industrial monocarboxylic fatty acids from Colombia in 2017 was
taken into account. In addition, it was assumed that there are no production interrup-
tions and that both biomethane (purification) and VFAs (electrodialysis separation)
were obtained at 100 %. 330 days of annual production were used for all processes.

In terms of costs, the price of pre-treatment and conditioning of the food waste was not
taken into account. Tools and equipment such as solid/liquid separator pipes, ducts,
pumps, mixers, among others, were considered negligible expenses that, out of the costs
included, make up the plant of the anaerobic digestion process. Likewise, the purchase
and conditioning of the land for the plant was not contemplated in the economic study.
On the revenue side, it was estimated that the purchase of biomethane and VFAs is
total for the project duration. The sales price of VFAs was estimated as an average of
the sales price of acetic acid, propionic acid, and butyric acid. The final specifications
of the biomethane (like composition and pressure) are adjusted to be in line with their
final use, therefore these costs were not taken into account. Economic and technical
inputs are proposed in Table 3-1.

This study can be considered as a conceptual analysis to evaluate different production

scenarios, where it is assumed that the production of biomethane and VFAs through
the process of anaerobic digestion is ideal according to the yields found in literature
(Jiang et al., 2018; Wang et al., 2016).

3.3. Results and discussion

Through the systematic literature review, 36 articles were selected with relevant informa-
tion on the production of biogas and VFAs through anaerobic digestion, using food waste
as a substrate. For the production of biogas, yields between 142 to 926 mL biomethane
gV S −1 were reported, while for the production of VFAs, yields between 21.54 to 47.31 gV-
FAs L−1 digestate were found. For this study, the best yields reported was selected, along
with the ideal processing conditions to achieve that yield. A summary of the operating con-
ditions of the both process are shown in Table 3-2 and Table 3-3. These conditions remain
constant for each of the cases to be economically evaluated.
3.3 Results and discussion 87

Table. 3-1: Input data for economic model

Variable Value Unit Reference
% rPP T V 5 − Ferella et al. (2019)
−3
CC 143.85 USD m Gobernación del Cauca (2016)
−3
Cf gas 0.0105 MWh m Ferella et al. (2019)
−2
CM 8.63 USD kg [Link]
−1
CSR 1.18 USD kg Gobernación del Cauca (2016)
−3
CU Be 0.13 KWh m Cucchiella and D’Adamo (2016)
−1
CU F W T 5.17 USD Ton Mintransporte (2018)
−1
CU L 293.70 USD month This study
−3
CU P 0.29 MWh m Cucchiella et al. (2018)
−3 −1
CU P B 5428.57 USD m h Cucchiella et al. (2018)
dSR 0.3 m This study
−1
EW h 600 h month This study
ef 1.1 − Moresi and Sappino (2000)
eST 0.2 m This study
fo 1.1 − This study
fH/r 2 − This study
inf 3.86 % Portafolio (2019)
−2 −1
JS 0.529 kg m h This study
−1
N WSR 0.56 kg m Gobernación del Cauca (2016)
OM F 0.04 − Lauer et al. (2018)
Pdf 20 % Cucchiella and D’Adamo (2016)
PE 1.7 KW This study
−3
Pe 0.16 USD m Enel Codensa (2019)
Pmo 5 % Ferella et al. (2019)
−1
RF W DU 9.12 USD Ton Grupo EPM (2019)
RP I OM 0.065 − Lauer et al. (2018)
−3
PP T V 28.17 USD MWh Ferella et al. (2019)
−1
PV F A 1.83 USD kg Zhou et al. (2018)
t 15 year This study
tB 0.2 m This study
tW 330 day This study
α 185587.3 USD Moresi and Sappino (2000)
−2
β 843.4 USD m Moresi and Sappino (2000)
δ 742.3 USD m−2 Moresi and Sappino (2000)
γ 327.7 USD m−2 Moresi and Sappino (2000)
ηdebt 15 year Cucchiella and D’Adamo (2016)
−1
ηL 3 Batch month Moresi and Sappino (2000)
ηm 3 year Moresi and Sappino (2000)
ηop 4 − This study
 88 3 Techno-economic evaluation of anaerobic digestion beyond biogas

Table. 3-2: Operating conditions of the anaerobic digestion of food waste for the production
of biomethane.
Item Units Value Reference
Substrate - Food waste -
−1
Density substrate kg L 0.514 Assumed
Operation temperature °C 55
pH - 7.2
Solids Retention Time day 29
Jiang et al. (2018)
Volatile Solids substrate gVS L−1 20.12
Product - Biomethane
−1
Biomethane Yield mL Biomethane gVS 556
Concentration product % v/v 100 Assumed
−1
Sale Price product USD kg 0.28 Ferella et al.
(2019)

Table. 3-3: Operating conditions of the anaerobic digestion of food waste for the production
of VFAs.
Item Units Value Reference
Substrate - Food waste -
−1
Density substrate kg L 0.514 Assumed
Operation temperature °C 35
pH - 7
Solids Retention Time day 8.75
−1 Wang et al. (2016)
Volatile Solids substrate gVS L 20.12
Product - VFAs
VFAs Yield gVFAs 47.31
−1
L digestate
Concentration product % v/v 100 Assumed
−1
Sale Price product USD kg 1.833 Zhou et al. (2018)

Regarding the scale of the process, in Table 3-4 can be seen the amount of substrate (supply),
which defines the capacity of the plant, and which was taken into account for the 8 cases
of the process focused on the production of biomethane and VFAs. The total amount of
food waste in Colombia per year was used as a reference. On the other hand, Table 3-5
3.3 Results and discussion 89

summarizes the 5 cases with the estimated amounts of biomethane and VFAs production, in
order to satisfy a percentage of the natural gas for the transportation sector and industrial
monocarboxylic fatty acids, respectively demand in Colombia.

Table. 3-4: Amount of substrate for the case studies based on the supply of food waste
in Colombia for the production of biomethane and VFAs through anaerobic
digestion.
Food waste in Colombia (Ton day −1 ) Reference
26575 Departamento Nacional de Planeación (2016)
Case Substrate inflow (Ton day −1 )
1a 5
2a 10
3a 20
4a 30
5a 50
6a 100
7a 150
8a 200

Applying the economic model proposed together with the NPV method for each situation
and case study of the process proposed, the surface curve for supply and demand of biomet-
hane production (Figure 3-3) and VFAs production (Figure 3-4) were obtained. Taking into
account the supply (amount of substrate), the process oriented to VFAs production presents
a positive accumulated NPV when 10 Ton day −1 of substrate are used onwards. With a
smaller amount of substrate, the accumulated NPV is unfavorable until the final year of the
project. For the biomethane process, favorable NPVs are seen from 20 Ton day −1 of subs-
trate used. Substrate values lower than this will generate losses profits until the final year of
the project. The time since it generates economic profit varies with respect to the amount
of substrate to use. For the biomethane process, a maximum of 7 years (from 20 Ton day −1 )
and at least 4 years (from 100 Ton day −1 ) are required to start generating economic profit.
However, in the case of VFAs production, profit is made from the first year when an amount
of substrate greater than or equal to 20 Ton day −1 is used. Thus, the project to produce
VFAs from the anaerobic digestion process is more economically attractive. This is mainly
because the time in which it begins to generate profits is shorter compared to the biomethane
process. The above is concluded taking into account the same amount of substrate to be used.
 90 3 Techno-economic evaluation of anaerobic digestion beyond biogas

Table. 3-5: Amount of products for the case studies based on the demand for natural gas
and industrial monocarboxylic fatty acids in Colombia.
Export of industrial mono-
Gas natural demand in the Reference carboxylic fatty acids from Reference
transport sector for Colombia Colombia (Ton year −1 )
3 −1
(m year )
Unidad de
Planeación
Datawheel
3 636 364 Minero 7 556
(2017)
Energética
(2017)
Natural gas de-
Biomethane VFAs demand percentage VFAs (Ton
Case mand percenta-
(m3 year −1 ) ( %) year −1 )
ge ( %)
1b 0,1 3636 5 378
2b 1 36364 10 756
3b 5 181818 25 1889
4b 10 363636 40 3022
5b 15 545455 50 3778

Figure. 3-3: Accumulated NPV of biomethane production through the anaerobic digestion
process. (a): Supply. (b): Demand. White surface is zero.

Comparing results individually, the process aimed at VFAs production has an overall in-
vestment (equipment, maintenance, depreciation, labor, services) 3.5 times higher than the
biomethane process. This is mainly due to the cost of the electrodialysis equipment, which
represents 89 % of the total investment. Compared to biogas purification, the costs of elec-
trodialysis equipment are on average 6 times higher (depending on the scale of the process
3.3 Results and discussion 91

the investment costs are inversely proportional to the waste in flow). From the bioreactor
point of view, due to the low solids retention time of the VFAs process, the smaller size of
the biodigester is needed compared to the biomethane process. This results in lower costs
in the construction of the biodigester (2 times smaller than the biodigester needed for the
production of biomethane), mitigating the overall investment costs of the process aimed at
the production of VFAs. The costs of labor, maintenance, and services make a similar value
between the two processes in all evaluated scales.

Figure. 3-4: Accumulated NPV of VFAs production through the anaerobic digestion process.
(a): Supply. (b): Demand. White surface is zero.

From a product point of view, revenue from final VFAs products sales is on average 11 times
higher than sales per biomethane. As a result, the initial investment cost can be recovered
in a few years (less than 3) as long as the substrate flow is greater than 10 Ton day −1 . In
contrast, for biomethane production, a positive NPV was reflected when the substrate flow
is greater than 18 Ton day −1 . As a consequence, profits were noticed after the maximum
evaluated period (15 years).

Cucchiella et al. (2018) conducted an economic study using the NPV method on small-scale
plants (50, 100, and 150 m3 h−1 of substrate) for the production of biomethane through the
anaerobic digestion process. As substrate, corn waste and municipal organic waste generated
in Italy were employed. The results showed that a positive NPV is generated for biogas plants
that can treat at least 50 m3 h−1 of substrate equivalent to 600 Ton day −1 . In comparison
to this study, positive results were obtained when treating an organic waste rate from 100
Ton day −1 . This difference could be partially explained since the availability of the substrate
and process factors such as storage, compression, and distribution of the methane which are
taken into account by the authors and which were not considered here since that information
was not available in our context. This makes the process economically feasible when treating
larger substrate flows, as the scale is inversely proportional to the process costs. Besides,
another difference can be seen in the cost of methane production. The authors report pro-
duction cost between 0.74 and 0.97 e m−3 of methane produced (Cucchiella et al., 2018).
 92 3 Techno-economic evaluation of anaerobic digestion beyond biogas

For this study, the methane production price ranges from 0.32 to 0.63 e m−3 . Production
costs for this study are derived from annual costs without taking into account the cost of
fixed capital investment, these being inversely proportional to the scale of the process, as
mentioned above. According to Arteconi et al. (2017), the cost of biomethane production
depends by the feed-stock composition, which is related to the biomethane yield that can
be obtained from the substrate, suggesting a higher production cost for low-yield substrates.
In this study, the highest biomethane yields for food waste reported in the literature were
used, thus generating low biomethane production costs.

On the other hand, Kleerebezem et al. (2015) have performed also a study s comparing the
economic feasibility of the products of the anaerobic digestion. Although the authors did
not make an in-depth assessment of the revenues and costs of the entire process, as done in
this contribution, they report that the VFAs oriented anaerobic digestion process can be 5.6
times more profitable than when oriented to biogas production (Kleerebezem et al., 2015).
For this study, it can be seen from the accumulated NPV as a function of the supply scale
(see Figure 3-5 (a)) that VFAs production is more economically attractive. Specifically for
the last year of the economic study (15 day), the scale of the process is directly proportional
to the income generated. As a result, net profits are up to 28 times higher when the process
is oriented to VFAs production rather than to the production of biomethane (when 200 Ton
day −1 of substrate are treated). On average, the percentage increase in net revenues from
the process aimed at VFAs production with respect to the process for generating biometha-
ne is 97 %, as shown in Figure 3-5 (b). This is mainly due to the difference in sales price
of the Figure 3-5 (b) products, which approximately VFAs have 13 times the commercial
value of biomethane. This means that the investment costs are recouped in less time, quickly
generating a positive NPV and therefore higher net profits.

Figure. 3-5: Cumulative NPV of biomethane and VFAs production for the year of project
completion (15 years) taking into account the supply of substrate. (a): Compa-
rison of net revenues. (b): Percentage increase of net revenues from the VFAs
production process with respect to the biomethane production process
3.4 Conclusions 93

3.4. Conclusions
The results of this study showed that the economic investment feasibility of the project ai-
med at producing VFAs from the anaerobic digestion process is more attractive than the
investment project focused on biomethane production. Despite the fact that the former re-
quires a higher initial investment. Technically, it is possible to valorizate the supply of food
waste utilization of at least 20 Ton day −1 and a percentage from 5 % of the national demand
for natural gas for the transport sector, with a favorable economic balance in a period of
15 years. While for the production of VFAs by means of the process of anaerobic digestion,
the supply of the reuse of food waste of minimum 9 Ton day −1 can be supplied and could
satisfy 5 % of the annual VFAs exported by Colombia, generating profitable income for the
implementation of the process within a period of 15 years.

The study of these scenarios is important because it can encourage and promote the analysis
and improvement of the anaerobic digestion process for VFAs production. The methodology
used in this study can be implemented in regions that have information on at least economic
indicators (inflation rate), amount of substrate, prices of construction materials, services and
logistics, as well as sale prices of products. At the same time, it theoretically demonstrates
the economic sustainability of the anaerobic digestion process at a national level, opening
the doors to a circular economy with a great capacity for profit, providing a response to
the current problems derived from non-renewable energies, such as environmental pollution,
limited availability of resources and the supply of energy to interconnected areas.

3.5. Acknowledgments
We thank the Fundación Ceiba and Becate Nariño, with the project: ”Formación del talento
humano de alto nivel para el fortalecimiento de necesidades estratégicas en ciencia, tecnologı́a
e innovación en el departamento Nariño”, for funding this work. We also express our gratitude
to the Universidad Nacional de Colombia-sede Manizales for providing the tools and resources
for this research.
94 3 Techno-economic evaluation of anaerobic digestion beyond biogas

Nomenclature
AM Membrane area for electrodialysis, m2
aM Polyester membrane area, m2
inv
CB+ST Cost biodigester + 2 storage tanks, USD
CBe Biodigester electricity cost, USD year −1
CBd Biodigester depreciation fund, USD year −1
CBOM Biodigester operation and maintenance cost, USD year −1
inv
CBP Biomethane purification investment cost, USD
CC Concrete cost 3000 PSI (includes installation), USD m−3
CEinv Electrodialysis equipment investment cost, USD
CEC Electrodialysis cell cost, USD
CEd Electrodialysis depreciation, USD year −1
CEe Electrodialysis electricity cost, USD year −1
CEM Electrodialysis membrane cost, USD
CEOM Electrodialysis operation and maintenance cost, USD year −1
CEU Electrodialysis equipment unit cost, USD
Cf gas Conversion factor biogas, MWh m−3
CF W T Substrate transportation cost, USD year −1
CL Labor cost, USD year −1
CM Polyester membrane cost, USD m−2
CP d Purification biomethane depreciation fund, USD year −1
CP e Purification biomethane electricity cost, USD year −1
CP OM Purification operation and maintenance cost, USD year −1
CSR Steel rod 3/80 cost (includes installation), USD kg −1
CU Be Biodigester electricity consumption unit, KWh m−3
CU P Unit purification biomethane electricity consumption, MWh m−3
CU P B Biomethane purification unit cost, USD m−3 h−1
CU L Unit labor cost USD, month−1
CU F W T Unit substrate transportation cost, USD Ton−1
dSR Distance between steel rods, m
DF W Food Waste density, kg L−1
ef Correction factor for safety over design of the required membrane surface area
eST Edge of storage tank, m2
EW h Electrodialysis equipment working hour, h month−1
fH/r Height-radius ratio
fo Oversizing factor
FW Food waste flow, Ton day −1
HB Biodigester height, m
inf Rate of inflation, %
Io Initial investment, USD year −1
It Discounted cash inflows, USD year −1
3.5 Acknowledgments 95

Nomenclature
JS VFA flux, kg m−2 h−1
n Lifetime of investment, year
NhSR Number of horizontal steel rods
N WSR Nominal weight steel rods 3/80 , kg m−1
Ot Discounted cash outflows, USD year −1
OM F Operation and maintenance cost annual factor
PBiomethane Selling price of biomethane, USD m−3
PP T V Price of biomethane in the virtual trading point, USD MWh−3
Pdf % of depreciation found
Pe Unit price of electricity, USD m−3
PE Electrodialysis power, KW
Pmo % of maintenance and overhead cost
PV F A Selling price VFAs, USD kg −1
Qbogas Biogas flow, m3 h−1
% rPP T V % reduction of the price in the PTV
rB Biodigester radius, m
RBiomethane Biomethane sale revenue, USD year −1
RF DF W Food waste disposal revenue, USD year −1
RF W DU Food waste disposal price unitary, USD Ton−1
RP I OM Rate price increase operation and maintenance costs
RV F A VFAs sale revenue, USD year
Sbiomethane Biomethane production, m3 h−1
SRT Solid retention time, day
SRB Amount of steel rod 3/80 for the biodigester, kg
SRST Amount of steel rod 3/80 for 2 storage tanks, kg
SV F A VFAs production, kg month−1
t Time of the cash flow, year
tB Wall thickness, m
tW Working time, day
VB Biodigester volume, m3
VBC Biodigester concrete volume, m3
VBF C Biodigester floor concrete volume, m3
VBW C Biodigester wall concrete volume, m3
VST Storage tank volume, m3
VST C Storage tanks concrete volume (2), m3
VT G Biodigester volume-gas phase , m3
VT L Biodigester volume-liquid phase , m3
V SF W Volatile solid food waste, gV S L−1
Ybiomethane Biomethane yield, mL methane gV S −1
YV F A VFAs yield, gVFAs L−1
96 3 Techno-economic evaluation of anaerobic digestion beyond biogas

Greek symbols
α Constant for electrodialysis equipment unit cost, USD
β Constant for electrodialysis equipment unit cost, USD m−2
δ Constant for electrodialysis cell cost, USD m−2
γ Constant for electrodialysis membrane cost, USD m−2
ηdebt Period of loan, year
ηhSR Number of horizontal steel rods for biodigester
ηL Number of monthly loads, Batch month−1
ηhSRST Number of horizontal steel rods for 2 storage tanks
ηm Membrane shelf life, year
ηop Number of operators
ηV SR Number of vertical steel rods for biodigester
ηV SRST Number of vertical steel rods for 2 storage tanks
References
Agler, M. T., Wrenn, B. A., Zinder, S. H., and Angenent, L. T. (2011). Waste to bio-
product conversion with undefined mixed cultures: The carboxylate platform. Trends in
Biotechnology, 29(2):70–78. doi:[Link]

Arteconi, A., Spitoni, M., Polonara, F., and Spigarelli, F. (2017). The feasibility of lique-
fied biomethane as alternative fuel: a comparison between european and chinese markets.
International Journal of Ambient Energy, 38(5):481–488. doi:10.1080/01430750.2016.
1191040.

Aryal, N. and Kvist, T. (2018). Alternative of Biogas Injection into the Danish Gas Grid
System—A Study from Demand Perspective. ChemEngineering, 2(3):43. doi:https://
[Link]/10.3390/chemengineering2030043.

Balkenhoff, B., Jamieson, D., and Consulting, S. L. R. (2010). Upgraded Biogas as Renewable
Energy. pages 1–10.

Bluemink, E. D., Van Nieuwenhuijzen, A. F., Wypkema, E., and Uijterlinde, C. A. (2016).
Bio-plastic (poly-hydroxy-alkanoate) production from municipal sewage sludge in the Net-
herlands: A technology push or a demand driven process? Water Science and Technology,
74(2):353–358. doi:10.2166/wst.2016.191.

Cabrera Rodrı́guez, C., Moreno González, M., de Weerd, F., Viswanathan, V., V., and der
Wielen, L.A.M., Straathof, A. (2017). Esters production via carboxylates from anaerobic
paper mill wastewater treatment. Bioresource Technology. doi:[Link]
1016/[Link].2017.02.030.

Calt, E. A. (2015). Products Produced from Organic Waste Using Managed Ecosystem Fer-
mentation. Journal of Sustainable Development, 8(3):43–51. doi:10.5539/jsd.v8n3p43.

Cucchiella, F. and D’Adamo, I. (2016). Technical and economic analysis of biomethane:

A focus on the role of subsidies. Energy Conversion and Management, 119:338–351.
doi:[Link]

Cucchiella, F., D’Adamo, I., and Gastaldi, M. (2017). Biomethane: A renewable resource as
vehicle fuel. Resources, 6(4). doi:10.3390/resources6040058.
 98 References

Cucchiella, F., D’Adamo, I., Gastaldi, M., and Miliacca, M. (2018). A profitability analy-
sis of small-scale plants for biomethane injection into the gas grid. Journal of Cleaner
Production, 184:179–187. doi:[Link]

Datawheel (2017). Observatory of Economic Complexity (OEC). url:[Link]

en/visualize/stacked/hs92/export/col/show/63823/1995.2017/.

Departamento Nacional de Planeación (2016). PÉRDIDA Y DESPERDICIO DE ALIMEN-

TOS EN COLOMBIA. Estudio de la Dirección de Seguimiento y Evaluación de Polı́ticas
Públicas.

Diaz-Baez, C., Espitia Vargas, S., and Molina Perez, F. (2002). Digestión anaerobia, una
aproximación a la tecnologı́a. Bogotá D.C., unibibios edition. url:[Link]
[Link]/handle/unal/49691.

Enel Codensa (2019). Tarifas Energı́a Eléctrica. url:[Link]

dam/enel-co/espa~nol/personas/1-17-1/2019/[Link].

Fagerström, A., Al Seadi, T., Rasi, S., and Briseid, T. (2018). The Role of Anaerobic
Digestion and Biogas in the Circular Economy. IEA Bioenergy Task 37, 37(8):1–24.
ISBN:978-1-910154-45-8.

Feng, L., Wang, H., Chen, Y., and Wang, Q. (2009). Effect of solids retention time and
temperature on waste activated sludge hydrolysis and short-chain fatty acids accumulation
under alkaline conditions in continuous-flow reactors. Bioresour. Technol., 100:44–49.
doi:[Link]

Ferella, F., Cucchiella, F., D’Adamo, I., and Gallucci, K. (2019). A techno-economic as-
sessment of biogas upgrading in a developed market. Journal of Cleaner Production,
210:945–957. doi:[Link]

Gebrezgabher, S. A., Meuwissen, M. P., Prins, B. A., and Lansink, A. G. (2010). Economic
analysis of anaerobic digestion-A case of Green power biogas plant in the Netherlands.
NJAS - Wageningen Journal of Life Sciences, 57(2):109–115. doi:ttp://[Link]/10.
1016/[Link].2009.07.006.

Gobernación del Cauca (2016). Data Cauca. url:[Link]

Grupo EPM (2019). Tarifas aseo segundo semestre 2019. url:[Link]

co/.

Guo, Z., Liu, W., Yang, C., Gao, L., Thangavel, S., Wang, L., He, Z., Cai, W., and Wang, A.
(2017). Computational and experimental analysis of organic degradation positively regu-
lated by bioelectrochemistry in an anaerobic bioreactor system. doi:10.1016/[Link].
2017.08.039.
References 99

Hengeveld, E. J., Bekkering, J., van Gemert, W. J., and Broekhuis, A. A. (2016). Biogas
infrastructures from farm to regional scale, prospects of biogas transport grids. Biomass
and Bioenergy, 86:43–52. doi:[Link]

Hoornweg, D. and Bhada Tata, P. (2012). What a Waste : A Global Review of Solid Waste
Management. Urban development series. World Bank, Washington, DC, knowledge,:29–
43. doi:10.1201/9781315593173-4.

House, D. and Surratt, V. (2013). The biogas handbook: Science, production and applications.
Woodhead Publishing Limited, Cambridge, UK. ISBN:978085709011.

Jiang, J., Li, L., Cui, M., Zhang, F., Liu, Y., Liu, Y., Long, J., and Guo, Y. (2018). Anaerobic
digestion of kitchen waste: The effects of source, concentration, and temperature. Bioche-
mical Engineering Journal, 135:91–97. doi:[Link]
004.

Jiang, J., Zhang, Y., Li, K., Wang, Q., Gong, C., and Li, M. (2013). Volatile fatty acids pro-
duction from food waste: Effects of pH, temperature, and organic loading rate. Bioresour.
Technol, 143:525–530. doi:[Link]

Kleerebezem, R., Joosse, B., Rozendal, R., and Van Loosdrecht, M. C. (2015). Anaerobic di-
gestion without biogas? Reviews in Environmental Science and Biotechnology, 14(4):787–
801. doi:10.1007/s11157-015-9374-6.

Lauer, M., Hansen, J. K., Lamers, P., and Thrän, D. (2018). Making money from waste: The
economic viability of producing biogas and biomethane in the Idaho dairy industry. Ap-
plied Energy, 222(February):621–636. doi:[Link]
04.026.

Lee, W. S., Chua, A. S. M., Yeoh, H. K., and Ngoh, G. C. (2014). A review of the production
and applications of waste-derived volatile fatty acids. Chemical Engineering Journal,
235:83–99. doi:[Link]

Lim, S., Kim, B., Jeong, C., Choi, J., Ahn, Y., and Chang, H. (2008). Anaerobic organic
acid production of food waste in once-a-day feeding and drawing-off bioreactor. Bioresour
Technol, 99 (16):7866–7874. doi:[Link]

Macarthur, E. (2006). Towards a Circular Economy. Journal of Industrial Ecology, 10(1-

2):4–8. ISBN:1530-9290.

Mao, C., Feng, Y., Wang, X., and Ren, G. (2015). Review on research achievements of
biogas from anaerobic digestion. Renewable and Sustainable Energy Reviews, 45:540–555.
doi:10.1016/[Link].2015.02.032.
 100 References

Mintransporte (2018). Costos por movilización y por tiempos logı́sticos.

[Link]

Moresi, M. and Sappino, F. (2000). Electrodialytic recovery of some fermentation products

from model solutions: Techno-economic feasibility study. Journal of Membrane Science,
164(1-2):129–140. doi:[Link]

Portafolio (2019). Inflación no cede en Colombia. url:[Link]

economia/inflacion-en-colombia-durante-octubre-de-2019-535288.

Rizos, V., Rizos, V., Tuokko, K., Tuokko, K., Behrens, A., Behrens, A., Vasileios Rizos,
V. R., Katja Tuokko, K. T., and Arno Behrens, A. B. (2017). The Circular Economy:
A review of definitions, processes and impacts. CEPS Research Report No 2017/8, April
2017. ISBN:9461385978.

Sariatli, F. (2017). Linear Economy versus Circular Economy: A comparative and analy-
zer study for Optimization of Economy for Sustainability. Bioeconomy and Sustainable
Development, 31. doi:10.1515/vjbsd-2017-0005.

Satinder, K. B., Saurabh, J. S., and Kannan, P. (2017). Platform Chemical Biorefinery.
Elsevier, future gre edition. doi:9780128030042.

Sauvé, S., Bernard, S., and Sloan, P. (2016). Environmental sciences, sustainable deve-
lopment and circular economy: Alternative concepts for trans-disciplinary research. En-
vironmental Development, 17:48–56. doi:[Link]
09.002.

Silva, F., Serafim, L., Nadais, H., Arroja, L., and Capela, I. (2013). Acidogenic fermentation
towards valorisation of organic waste streams into volatile fatty acids. Chem. Biochem,
Q.27:467–476.

Unidad de Planeación Minero Energética (2017). Balance de gas natural 2017 - 2026. Upme,
27(4):1–53. ISSN:07180764.

Wall, D. M., Dumont, M., and Murphy, J. D. E. (2018). Green gas - Facilitating a future
green gas grid through the production of renewable gas, volume 2. ISBN:9781910154373.

Wang, K., Yin, J., Shen, D., and Li, N. (2014). Anaerobic digestion of food waste for volatile
fatty acids (VFAs) production with different types of inoculum: effect of pH. Bioresource
technology, 161:395–401. doi:10.1016/[Link].2014.03.088.

Wang, Y., Zang, B., Li, G., and Liu, Y. (2016). Evaluation the anaerobic hydrolysis
acidification stage of kitchen waste by pH regulation. Waste Management, 53:62–67.
doi:[Link]
References 101

Zacharof, M. P. and Lovitt, R. W. (2013). Complex effluent streams as a potential source

of volatile fatty acids. Waste and Biomass Valorization, 4(3):557–581. doi:10.1007/
s12649-013-9202-6.

Zhang, C., Su, H., Baeyens, J., and Tan, T. (2014). Reviewing the anaerobic digestion of
food waste for biogas production. Renewable and Sustainable Energy Reviews, 38:383–392.
doi:10.1016/[Link].2014.05.038.

Zhou, M., Yan, B., Wong, J. W., and Zhang, Y. (2018). Enhanced volatile fatty acids pro-
duction from anaerobic fermentation of food waste: A mini-review focusing on acidogenic
metabolic pathways. Bioresource Technology, 248:68–78. doi:[Link]
[Link].2017.06.121.
4 Modelling and simulation of the
anaerobic digestion process
The linear system of production and consumption worldwide has not undergone changes and
is still based on the extraction of raw materials, the production of goods, consumption and
the generation of waste. The circular economy starts from the change in this system towards
a more sustainable approach avoiding waste and mitigating negative impacts for the envi-
ronment, the climate and human health. Anaerobic digestion is a technology governed by the
circular economy, which in addition to producing biogas as renewable energy, can produce
added value products such as Volatile Fatty Acids (VFAs).

In this chapter, a model-based techno-economic investigation was proposed to enhance the

VFAs bio-production through anaerobic digestion process. The Anaerobic Digestion Model
No 1 (AMD1) was used to design and implement a control system based on an override
strategy where pH is the controlled variable, and a real-time optimization algorithm finds
the best production scenarios under inlet disturbances such as protein particle components.
With this simulation approach, the understanding of the system is improved, and an integral
vision is given to the conventional anaerobic digestion process, promoting and contributing
to the study of the generation of non-conventional products in the anaerobic digestion. The
implementation of the model was carried out from the research presented by Rosen and
Jeppsson (2006). The content of this chapter is presented in article format.
 103

Real-time optimization for enhanced closed-loop

control of anaerobic digestion for VFAs production
under disturbances
Oscar Dario Yepez−Cerona , Beatriz Aristizabalb ,Oscar Andrés Prado−Rubioc
Departamento de Ingenierı́a Quı́mica, Universidad Nacional de Colombia, Manizales,
Caldas,170001, Colombia.
Grupo de Investigación en Aplicación de Nuevas Tecnologı́as, GIANTa,c
Grupo de Trabajo Académico en Ingenierı́a Hidráulica y Ambiental, GTAIHAb
a
odyepezc@[Link], b bharistizabalz@[Link], c oaprador@[Link]

ABSTRACT
Anaerobic Digestion have shown to play an important role in circular economy considering
its ability to transform waste in biogas. Beyond that, the production of platform substances
through the anaerobic digestion process, such as volatile fatty acids (VFAs) is of growing
interest due to the expected higher process profitability. However, the production and ex-
traction of these compounds are still challenging. Computational simulation is considered a
relevant tool to provide process understanding and improve the design and operation of the
anaerobic digestion process. This work proposes a model-based techno-economic investiga-
tion to enhance the VFAs bio-production form waste under disturbance. Anaerobic Digestion
Model No 1 (AMD1) is implemented for designing and evaluating a control structure based
on an override strategy. pH is the controlled variable, and a real-time optimization (RTO)
algorithm finds the best production scenarios under inlet disturbances. Through simulation
it was possible to obtain concentration of 34 times larger of total sum of VFAs, and increase
of 8771 USD day−1 (concerning to the nominal value) for sales of VFAs was obtained (by
implementing the VFAs production improvement), and through the RTO layer implemen-
tation, 7 % increase on process profit could be obtained when the system was disturbed in
unfavorable conditions (concerning to the point of operation). The results show that it is
possible to improve VFAs production compared to an uncontrolled reactor and demonstrate
that the control structure with real-time optimization guarantees the maximum possible in-
come when the system is subject to disturbances, finding the optimal operation point for this
purpose. Modeling efforts showed the production potential of VFAs through the anaerobic
digestion process, opening the doors to new perspectives of the process, being an important
step to implement profitable renewable technologies.
 104 4 Modelling and simulation of the anaerobic digestion process

4.1. Introduction
The rapid development of modern society and the constantly increasing population are fac-
tors that cause the generation of organic waste to increase significantly annually. Worldwide,
2.01 billion tons per year−1 of solid waste is generated, and this amount is expected to in-
crease to 3.4 million tons by 2025 (Paes et al., 2019). Improper waste management is not
only harmful to humans and the environment, but also increases climate change and makes
the sanitation system compel nations and governments to invest more financial and material
resources for its remediation (Wainaina et al., 2020). Consequently, renewable technologies
for waste management are necessary to promote environmental protection, sustainable de-
velopment and are also essential to implement and project a circular economy. The main
objective of the circular economy is that the value of products, materials and natural resour-
ces remain in the economy for as long as possible, considerably reducing the generation of
waste (Rizos et al., 2017).

One technology that implements the philosophy of the circular economy is the anaerobic
digestion process. Anaerobic digestion is a biological process where organic matter, in the
absence of oxygen, is degraded by microorganisms families through a sequence of reactions
in four stages: hydrolysis, acidogenesis (or fermentation), acetogenesis, and methanogenesis
and is mainly converted into methane and carbon dioxide, also called biogas (Toerien and
Hattingh, 1969). The anaerobic digestion process for biogas production has been highly suc-
cessful and well-established. Nevertheless, biogas is not one of the most valuable products
that can be obtained form anaerobic digestion. Then, this process has recently gained at-
tention as a cost-effective and environmentally friendly alternative for volatile fatty acids
(VFAs) production (Wang et al., 2014). VFAs including acetic acid, propionic acid, buty-
ric acid, valeric acid are potentially renewable carbon sources that have applications in the
pharmaceutical industry, precursors for polyhydroxyalkanoates (PHAs), cosmetic produc-
tion, petrochemical synthesis, and food and beverage industries (Zacharof and Lovitt, 2013).
Besides, production and extraction of VFAs during the anaerobic digestion process, though
is a technical challenge, could offer an economic advantage compared to biogas due to market
value for VFAs is 1.650 USD ton−1 (Calt, 2015), greater than the market value for purified
biogas (95 % methane), which is 996 USD ton−1 (Zhou et al., 2018a).

Several strategies to enhance VFAs production through the anaerobic digestion process have
been investigated. According to Zhou et al. (2018b), the production of VFAs can be impro-
ved based on the stages of anaerobic digestion, and these can be generally classified into:
I) Improving hydrolysis rate to produce more soluble substrates for further fermentation.
Optimization of key operational factors such as pH, temperature, and substrate pretreat-
ment before fermentation are methods that enhance hydrolysis (Kim et al., 2005; Zhang
et al., 2005; Wang et al., 2014). II) Promoting the acidogenic process: experimentally, it has
4.1 Introduction 105

been shown that substrate, inoculum, hydraulic retention time (HRT), organic loading rate
(OLR), pH, temperature, and headspace gas are critical factors that determine the efficiency
of VFAs production and quality of VFAs (Lim et al., 2008; Chen et al., 2013; Jiang et al.,
2013). III) Removing the inhibiting factors: the accumulation of VFAs may cause the acido-
genic reactions thermodynamically unfavorable, resulting in the shift of metabolic pathway
to produce other products (Pind et al., 2003). This limitation can be overcome if the VFAs
were removed from the system continuously through precipitation, extraction, crystalliza-
tion, distillation, and, primarily, electrodialysis technique (Redwood et al., 2012; Jones et al.,
2015; Arslan et al., 2016) .

Despite previous research efforts, the optimal design, control, and operation of the anaerobic
digestion process VFAs production remain as challenge due to the variability of the substrate
and its weak decomposition, the complexity of the microbial consortium, the destabilization
by inhibition, pH variability, inhibition by ammonia, complicated biochemical, physical, and
chemical interactions involved in the process (Donoso-Bravo et al., 2011; Yu et al., 2018).
To address these drawbacks, the computational simulation of mathematical models is and
interesting approach to improve the design and efficiency of the anaerobic digestion process
(Manjusha and Beevi, 2016). The computational simulation allows a better understanding
of the system, formulation, and validation of some hypotheses, prediction of the behavior
of the system under different conditions, thereby reducing experimental information requi-
rements, costs, risk, and time (Donoso-Bravo et al., 2011). On the other hand, according to
Nguyen et al. (2015), implementing an anaerobic digestion process control system enables
quick process stabilization with less operation and maintenance inconveniences, maximizing
the efficiency of the process.

Research in the last two decades has focused on implementing sophisticated models to si-
mulate and understand the complex degradation of organic matter that occurs in anaerobic
processes. These models are developed to discover specific aspects of the composite substrate
degradation process, taking into account the identification of key factors that affect system
performance (Batstone et al., 2002). Possibly the most widely applied model in the area of
anaerobic digestion process research is the Anaerobic Digestion Model No. 1 (ADM1) deve-
loped by the modelling task group of the International Water Association IWA (Batstone
et al., 2002). ADM1 takes into account the main steps of disintegration, hydrolysis, acido-
genesis, acetogenesis, and methanogenesis. It consists of descriptions of 7 groups of bacteria
and archaea, 19 biochemical process rates, 24 components (12 soluble and 12 particulate
elements), 3 gas-liquid transfer kinetic processes (CH4 , CO2 , H2 ), and 6 acid-base balances
in association with the pH calculation together with a set of 105 kinetic and stoichiometric
parameters/variables (Batstone et al., 2002; Kleerebezem and van Loosdrecht, 2006).

ADM1 has been applied and validated by simulating the digestion of various organic waste
 106 4 Modelling and simulation of the anaerobic digestion process

(Batstone and Keller, 2003; Lohani et al., 2016; Jurado et al., 2016; Nordlander et al., 2017).
Likewise, ADM1 modifications have been made to improve and optimize the biogas produc-
tion by anaerobic digestion process (Pessoa et al., 2019; Li et al., 2020; Sun et al., 2021), as
well as the implementation of control systems with ADM1 as a test platform, where these
studies focused on the design of control strategies that avoid the accumulation of VFAs to
guarantee the process stability and optimal production of biogas (Garcı́a-Sandoval, 2009;
Ramirez et al., 2019; Zhou et al., 2020). In contrast, research on the implementation and
control of ADM1 to enhanced VFAs production is limited. Bai et al. (2015) developed a
modified ADM1 to simulate anaerobic fermentation of waste activated sludge degradation
and evaluate the influence of pH on VFAs production. The authors used particle swarm
optimization (PSO) to improve acidogenic reactor design and optimize the operation pa-
rameters. They found that alkaline conditions are beneficial for VFAs production through
anaerobic fermentation, being one of the few studies conducted that focuses on improving
VFAs production through modeling and simulation of the anaerobic digestion process.

Due to the lack of understanding of the factors that enhance VFAs production in anaerobic
digestion systems, a study to predict and improve VFAs production by anaerobic digestion
using a robust mathematical model such as ADM1 is required. Additionally, mentioned abo-
ve, orienting the anaerobic digestion process toward the production of VFAs can be a more
economically attractive process than when producing biogas. A strategy that allows imple-
menting these two purposes is to include a real-time optimization (RTO) layer to a control
structure to estimate the optimal anaerobic digestion conditions under input disturbances.
A RTO layer is a framework that continuously evaluates the process operating conditions
to maximize the economic productivity of a process (Mirlekar et al., 2018), thus it could
be used to find the best production scenarios when the substrate change properties and the
control structure guarantee stable operation. In this study, ADM1 model is used to simu-
late and evaluate possible VFAs production scenarios using the RTO strategy coupled with
a control structure. The implementation and manipulation of this model proposes strate-
gies to obtain VFAs instead of biogas. Through the Matlab/simulink® tool, a sensitivity
analysis is performed to identify the input variables of the model that promote the increase
in VFAs concentration and decrease methane production. Likewise, the implementation of
closed-Loop control with an RTO layer (economic optimizer) is designed and investigated to
guarantee, not only the optimal operation of VFAs production under potential disturbances
but also to maximizes the economic income from the sale of VFAs produced.

4.2. Methodology
The methodology that was proposed to implement a closed-loop real-time optimization for
ADM1 to increase the production and incomes from sales of the VFAs consists of 4 primary
4.2 Methodology 107

steps: 1. Mathematical implementation of ADM1, 2. Operation point estimation that en-

hances VFA production under inlet standard conditions, 3. Control structure design where
manipulated and controlled variables are selected and 4. Hierarchical control structure: real-
time optimization layer implementation for closed-loop control. Figure 4-1 shows the flow
diagram that summarizes the methodology of the study carried out.

Figure. 4-1: Methodology for the implementation of closed-loop real-time optimization that
enhances VFAs production in anaerobic digestion.

4.2.1. Mathematical implementation of ADM1 and verification

ADM1 model was programmed in Matlab/simulink® R2018a through an Ordinary Diffe-
rential Equations (ODEs) system as well as a Differential-algebraic equation (DAE) system.
The equations covering the biological kinetic rate expressions and coefficients were obtai-
ned from the Peterson matrix (Batstone et al., 2002). All of the parameters and initial
conditions necessary for the implementation of the model come from Rosen and Jeppsson
(2006). Because ADM1 is a very stiff system with time constants ranging from fractions of
a second to months (Rosen and Jeppsson, 2006), ODEs were solved with the Euler method
solver (ode15s) available in Matlab® . The in-house implementation was done since the
open-source implementation is available in C++ connected with Matlab® , which limits
the potential of model adaptations
 108 4 Modelling and simulation of the anaerobic digestion process

In order to verify the model implementation, simulation outputs were compared to (Rosen
and Jeppsson, 2006)) simulations. The comparison between the benchmark data and the
results derived from the simulation of the model of this study was carried out by calculating
the absolute error of each output variable of the model. The simulation time (tspan ) for the
model verification was 2000 days, it was also used to obtain final values at steady-state con-
ditions.

4.2.2. Optimal operating point determination

The optimal operating point corresponds to the sum of the highest concentration of each
VFA that can be produced. To determine this, a sensitivity analysis was first performed
to determine the input variables: soluble components, particulate components, soluble ionic
components and soluble gaseous components (presented in the nomenclature) that promote
the generation of VFAs and inhibit the generation of methane. Each of the input variable
was individually modified in a range of 5 times less and 5 times more than the initial value.
This sensitivity analysis allowed to define the operative window and potential numerical
problems for the subsequent optimization.

Due to the process unstability, it was necessary to implement an open-loop control that
allows stabilizing the production of VFAs, where the manipulated variables were the input
variables found from the sensitivity analysis carried out before. To optimize the reactor with
the open-loop control, equation 4-1 presents the economic objective function that allow fin-
ding the highest economic income from selling VFAs. The optimization problem depicted in
Equation 4-1 was solved using the Matlab® function “Fmincon” with the “interior-point”
algorithm. This solver is a powerful tool able to minimize non linear objective functions,
considering inequality and equality constraints (Pataro et al., 2020). The prices of the VFAs:
valeric acid, butyric acid, propionic acid, and acetic acid are presented in Table 4-1. The
extraction and purity of the VFAs obtained from the anaerobic digestion process were con-
sidered ideal.

Table. 4-1: VFAs sale prices. (Zhou et al., 2018b)

VFA unit value
−1
Valeric acid USD kg 2.5
Butyric acid USD kg−1 2
−1
Propionic acid USD kg 1.65
−1
Acetic acid USD kg 0.8
4.2 Methodology 109

( 4
)
X
(P riceV F Ai )([V F A]i ) + V F A− i
 
min −qin (4-1)
i=1

Where: qin : Flow rate m3 day−1 , Price VFAi : VFAs sale prices USD kg−1 , [V F A]i : VFAs
concentration kgCOD m−3 , [V F A− ]i : ions VFAs concentration kgCOD m−3

4.2.3. Control structure design

Once the operating point of the system is found, a control structure must be designed, i.e.,
identify potential controlled and manipulated variables to achieve the control objective. For
this, the system controllability and stability is determined through the evaluation of the
Kaman’s controllability matrix (Seborg et al., 2010). For this purpose, it was necessary to
calculate the linear representation of the system through numerical approach (MathWorks,
2021). For determining the best pairing of controlled and manipulated variables, Bristol’s
Relative Gain Array (RGA) method was used (Bristol, 1966; Skogestad and Postlethwaite,
2005).

Due to the non linear nature of the model, the potential manipulated variables: soluble com-
ponents, particulate components, soluble ionic components and soluble gaseous components
(presented in the nomenclature), were identified through a sensitivity analysis instead of
using a linear approach. The sensitivity analysis was carried out using step disturbances on
each input variable of (-50 %: 5: 50 %) at time 0 and tspan of 500 days. The interest output
variable is the pH, because pH is one of the factors that most affects the concentration and
composition of VFAs (Liu et al., 2012; Lee et al., 2014). Through the sensitivity analysis,
this statement was corroborated. To quantify output sensitivity, the accumulated root-mean-
square error (RMSE) was calculated, which was calculated according to equations 4-2 and
4-3.

r Pn
i=1 (Pi − Oi )2
RM SEk = (4-2)
n
21
X
RM SEac = RM SEk (4-3)
k=1

Where RM SEk : Root-mean-squeare error for a k % disturbance, Pi : Operating point value,

Oi : Observable point value, n: Data number, RM SEac : Root-mean-square error accumulated
for a disturbed input variable.
 110 4 Modelling and simulation of the anaerobic digestion process

4.2.4. Hierarchical control structure: real-time optimization and

closed-loop control
In practice, the control system is usually divided into several layers. Typically, layers in-
clude scheduling (weeks), site-wide optimization (day), local optimization (hour), supervi-
sory/predictive control (minutes) and regulatory control (seconds) (Larsson and Skogestad,
2000). Referring to local optimization, the purpose of this is to identify the active constraints
and recompute optimal set points for the controlled variables (Skogestad, 2004). This opti-
mization is the task of the real-time optimization (RTO) layer.

Initially, the linear relation between controlled and manipulated variable is required. A li-
near representation of first order plus time-delay (FOPTD) model is obtained by optimal
parameter estimation using the step response method with 10 % and -10 % input disturbance
(Seborg et al., 2010). This is approach is preferred to exploit knowledge of input disturbances
and to have a better representation of the non linear response.

 t

y(t) = KM 1 − e− τ (t − td ) (4-4)

Where K: Gain, M: Amplitude of the input step disturbance, τ : Time constant, t: Time, td :
Time delay

The PID controller implemented was a parallel type with a derivative filter. Derivative filter
reduces the sensitivity of the control calculations to noisy measurement (Seborg et al., 2010).
The PID controller parameters were estimated with the Internal Model Control (IMC) (Gar-
cia and Morari, 1982). Because the pH (controlled variable), depends on the manipulation
of the inlet cations and anions concentration (Scat in and San in i.e,. base or acid addition),
which are complementary, it is proposed to implement a override controller to exploit this
particular system characteristic. Manipulated variables can be switched according to a spe-
cified situation.

On the top of the control structure, a real-time optimization (RTO) layer is implemented.
Considering the economic profit as target, the implementation of the RTO closed-loop aims
to increase the daily revenue according to a positive or negative disturbance. The economic
objective function used for process design is re-used (equation 4-1) which takes into account
the sale price of the VFAs (Table 4-1). Once again, Matlab® solver ”fmincon” was used
to solve the optimization problem. A step disturbance of the input variable Xpr in of +10 %
and -10 % was carried out in a time of 200 days.
4.3 Results and discussion 111

4.3. Results and discussion

4.3.1. Implementation and validation of ADM1 model
The implemented model was used as an independent model to test a specific system to en-
hance VFAs production and inhibit biogas production. ADM1 consists of 35 components
that are presented in the nomenclature along with their units (Batstone et al., 2002). The
process operating conditions are also presented.

Model verification is important due to that it verifies that model results are reliable. The
verifying was performed by comparing results from the model with results from Rosen and
Jeppsson (2006) ADM1 implementation. Error absolute between the benchmark outputs and
outputs derived from ADM1 simulation in this study for the same input values are presen-
ted in Figure 4-2. As can be seen from the comparison between the benchmark data and
steady-state results derived from the model simulation, there was a very close agreement for
all values. The prediction errors were encountered in the range of 10−3 , where the variable
Sgas,ch4 had the highest absolute error with 4.3x10−3 . Due to the small values encountered,
the implementation is considered appropriate and the reasons for the differences could be
attributed to potential updated in Matlab® numerical methods.

Figure. 4-2: Steady-state output comparison between Rosen and Jeppsson (2006) and actual
model implementation
 112 4 Modelling and simulation of the anaerobic digestion process

4.3.2. Operating point: VFAs production enhance

Through a sensitivity analysis, ADM1 input variables were individually modified, without
altering parameters, coefficients, and operational conditions stipulated by Rosen and Jepps-
son (2006)). It was found that the input variable with the highest effect promoting the VFAs
production was San in (soluble components anion input) which defines the concentration ad-
dition of acidic anions. Figure 3 shows the behavior of the VFAs concentration, together
with other key components, when San in was modified.

As previously mentioned, Scat in and San in are complementary variables. However, Scat in
still does not modify the system significantly. A disturbance affects the results in Figure
4-3, however, this variable remained constant and San in varied. Lower San in values than
0.15 kmol m−3 did not affect the system behavior. On the other hand, for higher values,
the VFAs concentration increased and methane production decreased. The San in value that
maximizes the VFAs output concentration was 0.16 kmol m−3 , but notice that the system
does not stabilize. The fact of increasing the San in concentration implies that the H+ ions
concentration increase, causing a decrease in the pH. Besides, as can be seen in Figure 4-3,
and according to the chemical reaction: H+ + N H3 ↔ N H + +
4 , the increase in H ions causes
the charge balance to destabilize, generating ammonium production (SN H4+ ), resulting in
a decrease in pH. According to the literature, pH has been identified as a the one of the
main variables that can modify the behavior, being a critical factor that controls the VFAs
production during acidogenic fermentation of the process (Zhou et al., 2018a). Wang et al.
(2014) found that butyric acid was the main product under pH 5.0 with the percentage of
butyrate above 80 % under anaerobic conditions. Jiang et al. (2013) found that when the pH
decreases to 5 in an anaerobic digestion process, acetic acid was the dominant product, follo-
wed by butyrate, propionate, and valerate. Likewise, Min et al. (2005) observed in anaerobic
digestion a maximum fraction of propionic acid (80 %) at a pH of 6.5.
According to Fukuzaki et al. (1990), methane production from propionic acid is an important
intermediate step in the bioconversion of organic matter to biogas. This step is inhibited by
the accumulation of H2 . The partial pressure of H2 must be kept below 10−6 to 10−4 bar
for methane production. If this step is inhibited, it would be expected that propionate and
perhaps acetate would accumulate. Figure 4-3 shows an increase in the partial pressure of H2
(Pgash2 ) above the range stipulated by Fukuzaki et al. (1990) (maximum value of 0.43 bar for
San in =0.16 kmol m−3 ), thus being an important factor for the accumulation of VFAs. Also,
due to this accumulation of propionic acid, a significant inhibition is generated, causing the
concentration of methanogenic bacteria to decrease, and therefore the decrease in methane
production (Wang et al., 2009). The accumulation of VFAs is also due to the decrease of
the growth of the degrading microorganisms of the VFAs (Xc4 , Xpro , Xac ). This is due to an
inhibition by pH since micro-organisms are intolerable to acidic conditions.
4.3 Results and discussion 113

Figure. 4-3: Sensitivity analysis of the input variable San in . Where San in =0.02 kmol m−3
(nominal value)

As previously mentioned, although an improvement in the VFAs production is observed,

the system does not stabilize since it is completely inhibited, mainly due to the dropping
pH with ranges from 0.96 to 0.34. That is the reason to implement an open-loop control an
stabilize the VFAs production, during the economic optimization, where the manipulated
variables were San in and Scat in , are critical variables for the production of VFAs. For this,
ADM1 was simulated by changing San in from 0.02 to 0.16 kmol m−3 . Then, the economic
optimizer manipulates San in and Scat in to avoid a drastic pH drop, guarantee the stability
of the system and generate the highest economic income from sales of VFAs.

Solving equation 4-1 (economic objective function), the optimal inlet concentration of so-
luble component cations (Sscat in op ) was 0.04 kmol m−3 (value equal to nominal) while for
soluble component anions (San in op ) was 0.0495 kmol m−3 . The simulation time to solve the
economic optimizer was 28.19 seconds. Finally, ADM1 was simulated with tspan =300 days
again, by performing a step disturbance from San in to San in op in a time of 80 days. The
disturbance time corresponds to a range of 55 to 108 days (time where the VFAs accumu-
late), and the choice of the disturbance time within this range did not influence the results.
Figure 4-4 shows a comparison between the results of the nominal ADM1 simulation (so
called nominal process (San in =0.02 kmol m−3 )), ADM1 simulation when VFAs production
is maximum (so called unstable process (San in =0.16 kmol m−3 )), and ADM1 simulation
when open-loop control with economic optimizer was implemented (so called optimal pro-
 114 4 Modelling and simulation of the anaerobic digestion process

cess (San in op = 0.0495 kmol m−3 )). The implementation of the open-loop control coupled
with the economic optimization stabilized the system due to the pH control, avoiding the
process inhibition. In both the unstable process and in the optimal process, degraders of
VFAs showed the same behavior, suggesting that the inhibition of the system was due to
the pH dropping and the accumulation of VFA was caused by the low growth rate of VFAs
degraders. The ammonia concentration also stabilized because the generation of H+ ions is
smaller, stabilizing the charges of the aforementioned reaction.

Figure. 4-4: ADM1 simulations: Nominal (San in =0.02 kmol m−3 , maximum VFAs produc-
tion (San in =0.16 kmol m−3 ) and open-loop control with economic optimizer
(San in op = 0.0495 kmol m−3 ).

In comparison with the simulation with the unstable process, the optimal process obtained a
concentration of 34 times larger of total sum of VFAs, which generated a profit of 8800 USD
day−1 per sale of VFAs. This results was expected since the unstable process is completely
inhibited at the steady-state. Regarding the nominal process, the methane production of
the nominal process is 16 times greater than that of the optimal process, logical result since
the nominal ADM1 model was developed to reproduce biomethane production. Table 4-2
presents the relevant comparisons of results obtained for the nominal process, the unstable
process, and the nominal process for the steady-state at a tspan of 300 days. Finally, the op-
timal process was simulated with a tspan of 2000 days (steady-state) to obtain the operating
point to implement closed-loop control.
4.3 Results and discussion 115

Table. 4-2: Steady-state results comparisons from ADM1 simulations

Parameter Unit Nominal Unstable Optimal % Enhancement
process process process
−3
San in kmol m 0.02 0.16 0.0495 -
−3
Scat in kmol m 0.04 0.04 0.04 -
−3
Total VFAs kgCOD m 0.478 0.0051 34.217 7058.37
−1
Total VFAs sale USD day 82.317 1.4834 8853.72 10655.64
3 −1
qgasCh4 m d 1709.23 0.000618 107.144 -
pH - 7.466 0.958 4.929 -

4.3.3. Control structure design

To calculate the state-space model, 35 states (35 ADM1 components from table 3), 36 inputs
(35 ADM1 components + qin ), and 36 outputs (35 components + pH) were considered. On
the other hand, flow rate acts as a disturbance to the system and, as mentioned earlier, the
pH as a variable to control due to this is a critical factor that controls the VFAs production
during acidogenic fermentation of the process. Through the sensitivity analysis, the pH was
corroborated as the critical parameter for the production of VFAs. Based on that, the RGA
was limited to 5 input variables that can control pH. From the linear approximation, it is
validated that the system is stable and controllable. For this case the RGA is interpreted as
the gains of each manipulated input towards the pH, and using the linear system, the RGA
matrix obtained is depicted in Table 4-3.

Table. 4-3: The relative gain array for ADM1

Manipulated variables
Controlled variable
SIC in SIN in Scat in San in qin
−6
pH 1.625x10 0.0548 0.481 0.463 1.518x10−11

From Table 4-3 only Scat in and San in present values close to 0.5, while the other variables
present a value close to zero, which indicates that the later manipulated variables do not
affect the variable to be controlled. Thus, Scat in and San in present a high degree of inter-
action, having the same effect on the variable to control. This was to be expected since the
addition of soluble component anions and cations directly influence the pH in opposite direc-
tions. With this, the selected variable to control to maintain the production of VFAs was pH
and the manipulated variables were Scat in and San in , two complementary variables that be
controlled according to the override control strategy. This strategy involving one constraint
variable implemented using a control selector, two PID and one associated input and output.
 116 4 Modelling and simulation of the anaerobic digestion process

Once the controlled variable and manipulated variables were identified, disturbances that
modify the operating point were determined using a sensitivity analysis. Figure 4-5 shows
the accumulated RMSE results for each of the 36 input variables. The variables that pre-
sented the highest accumulated RMSEs were Xpr in and Xch in , indicating disturbances in
those variables, have the highest impact on the system pH, thus represent the most difficult
control scenario to be further investigated.

Figure. 4-5: Accumulated RMSE of disturbance of input variables concerning the pH of the
operating point

4.3.4. Hierarchical control structure: real-time optimization layer for

closed-loop control
The FOPTD model obtained from the sensitivity analysis is shown in Table 4-4. The system
requires controlling the pH using Scat in and San in , These variables are complementary and to
stabilize the system, it is only necessary to add them. Thus, to increase the pH, adding Scat in
is necessary, and to decrease the pH, adding San in is necessary. The time constant for the
two variables is similar, therefore, the response speed of the process is also similar. With K
and t defined, the PID controller settings were estimated. The results are shown in Table 4-5.

Through the override control strategy, also called selective control, it was possible to imple-
ment two controllers (PID1 and PID2) that manipulated the same process. The selection of
4.3 Results and discussion 117

the controller was determined by the difference between the signal of the variable to control
(pH) and the value of the set-point, for which the following constraint was established:

if(set-point pH - signal pH)>1x10−10 ⇒ PID1 on ∧ PID2 off, else PID1 off ∧ PID2 on

Table. 4-4: Parameters for FOPTD function

Manipulated variable K (m3 kmol−1 ) τ (day)
Scat in 6.447 15.99
San in -6.591 16.61

Table. 4-5: PID controllers settings

Controller Manipulated P I D α N
variable
PID1 Scat in 0.4653 0.0291 0.4653 0.05 20
PID2 San in -0.4552 -0.0274 -0.4552 0.05 20

Figure. 4-6: Simulink® implementation scheme of closed-loop for real-time optimization

implementation that enhances VFAs production in anaerobic digestion

The implementation of RTO in the process should ensure maximum profit from the sale
of VFAs produced when the system under disturbances. RTO collects the pH measurement
from the process, updates the model data, and performs economic optimization using Equa-
tion 4-1. As a result, the RTO layer sends new set-points to the control layer to achieve
the optimal operating point. A control scheme implemented in Matlab/simulink® of the
 118 4 Modelling and simulation of the anaerobic digestion process

closed-loop control with the RTO layer can be seen in Figure 4-6.

Figure. 4-7: ADM1 simulation results with and without closed-loop real-time optimization
when a step disturbance of 10 % Xpr in was performed

Figures 4-7 and 4-8 show the system response when the optimal control (RTO) was imple-
mented and when it was not. For the positive disturbance, it can be seen that the system
benefits from the increase in the concentration of Xpr in since more VFAs are produced and
therefore, greater profits concerning the initial operating point. Due to the control structure,
process inhibition and unstability are avoided. The results indicate that when the distur-
bance was performed and the optimal control is not implemented, more total VFAs were
produced (37.02 kgCOD m−3 ) but fewer gains (9524.5 USD day−1 ) concerning the optimal
process (36.92 kgCOD m−3 total VFAs and 9569.9 USD day−1 ). This is because, in the un-
controlled process, more acetic acid is produced, which has the lowest price of the 4 potential
VFAs. Acetic acid is the main intermediate of VFAs for the conversion into methane, which
also explains the increase in methane flow by 7 m3 day−1 when the control is not applied
in the process. The system presented acidification in response to the positive disturbance
of Xpr in , with a pH value of 4.941, which decreases 34 % with respect to the nominal value
(7,466). To control this, the optimizer determined an optimal operating point where the pH
should be kept at 4.928. Thus, according to the override control constraint, the variable to
manipulate was San in (PID2 on), and an increase in San in of 0.0029 kmol m−3 was needed
to guarantee the highest profit from the sale of VFAs. On the other hand, when the process
was disturbed with Xpr in of -10 %, PID1 (manipulated variable Scat in ) was the controller
that acted. Although the disturbance did not destabilize or inhibit the process, the response
4.4 Conclusions 119

variables decreased. For this case, the optimizer determined the best possible operating point
under these conditions. Regarding the uncontrolled process, losses of 783.98 USD day−1 were
generated, while for the optimal process, losses of 733.14 USD day−1 were generated after
the system was disturbed.

Figure. 4-8: ADM1 simulation results with and without closed-loop real-time optimization
when a step disturbance of -10 % Xpr in was performed

The implementation of the RTO allowed to obatin 0.48 % increase in process profit when
the system was disturbed with Xpr in +10 %. On the other hand, 7 % improvement was
obtained when Xpr in is disturbed by -10 %. % enhancement refers to the situation with
control plus RTO, compared to the situation only with control. For the second disturbance,
RTO increased profit by 0.51 % and 0.54 % for Xch in disturbances of +10 % and -10 %,
respectively. From both scenarios, it can be seen that it is worth implementing RTO when
the disturbance is negative, since the percentage of improvement is greater when it is positive,
and generating the least possible loss of income. Significant results are observed when Xpr in
is disturbed, the disturbance of the other variables is negligible.

4.4. Conclusions
A real-time optimization closed-loop with ADM1 as a test platform to improve VFAs produc-
tion was implemented. The ADM1 implementation was carried out in Matlab/Simulink®.
ADM1 model results were verified with the results of Rosen and Jeppsson (2006). Through
 120 4 Modelling and simulation of the anaerobic digestion process

sensitivity analysis, San in and Scat in were the most important variables that enhance VFAs
production and inhibited methane production, obtaining a value of San in =0.16 kmol m−3
(Scat in remained constant). To stabilize the system, an open-loop control was implemented
with an economic optimizer. It was determined that a disturbance of San in op = 0.0495 kmol
m−3 in 80 days, allowed VFAS production that generated maximum profit corresponding
to de optimal operation condition. The results showed that the production of total VFAs
about the nominal ADM1 model increased from 0.478 kgCOD m−3 to 34.217 kgCOD m−3 ,
generating economic incomes of 8853.72 USD day−1 from sales of VFAs.

To implement the closed-loop control at the operating point, the state-space model was
calculated. Using the RGA, the variable to control was the pH and the manipulated variables
were San in and Scat in . Through a disturbance analysis, Xpr in was the variable with the
highest effect on the process behavior with an accumulated RMSE of 0.65. An override control
strategy was implemented with two PID1 and PID2 controllers and a selector, where the first
manipulates Scat in and the second manipulates San in , to control the pH. Finally, RTO layer
was implemented in the control process to find the optimum operating point (pH set-point)
that maximizes the economic profit from the sale of VFAs, in response to disturbances step
by ±10 % of Xpr in . The RTO layer allowed up to 7 % increase on process profit. This low
value could decrease the interest to have a complex control structure, however, for a larger
scale this value is significant. Other input disturbances should be investigated that could
generate the need of an RTO on the top of the control structure. Finally, the implementation
of this control system could provide theoretical guidance for practical applications and the
development of control strategies that allow promoting the production of VFAs through the
anaerobic digestion process.
4.4 Conclusions 121

Nomenclature
No Components Descriptions Unit
1 Ssu Monochaccharides (sugars) kgCOD m−3
2 Saa Amino acids kgCOD m−3
3 Sf a Fatty acids kgCOD m−3
Soluble components

4 Sva Total valerates kgCOD m−3

5 Sbu Total butyrates kgCOD m−3
6 Spro Total propionates kgCOD m−3
7 Sac Total acetate kgCOD m−3
8 Sh2 Hydrogen gas kgCOD m−3
9 Sch4 Methane gas kgCOD m−3
10 SIC Inorganic carbon KmolC m−3
11 SIN Inorganic nitrogen kmolN m−3
12 SI Soluble inerts kgCOD m−3
13 Xc Composites kgCOD m−3
14 Xch Carbohydrates kgCOD m−3
Particulate Components

15 Xpr Protein kgCOD m−3

16 Xli Lipid kgCOD m−3
17 Xsu Sugar degraders kgCOD m−3
18 Xaa Amino acids degraders kgCOD m−3
19 Xf a Fatty acids degraders kgCOD m−3
20 Xc4 Valerate and Butyrate degraders kgCOD m−3
21 Xpro Propionate degraders kgCOD m−3
22 Xac Acetate degraders kgCOD m−3
23 Xh2 Hydrogen degraders kgCOD m−3
24 XI Particulate inerts kgCOD m−3
25 cat Soluble component cations kmol m−3
25 San Soluble component anions kmol m−3
components

27 Sva− Soluble component valerate ions kgCOD m−3

28 Sbu− Soluble component butyrate ions kgCOD m−3
Soluble

kgCOD m−3
ionic

29 Spro− Soluble component propionate ions

30 Sac− Soluble component acetate ions kgCOD m−3
31 Shco3− Soluble component hydrogen carbonate KmolC m−3
32 Snh3 Soluble component ammonia ions kmolN m−3
33 Sgas,h2 Soluble component hydrogen gas kgCOD m−3
Soluble gaseous
components

34 Sgas,h2 Soluble component methane gas kgCOD m−3

35 Sgas,co2 Soluble component carbon dioxide gas KmolC m−3
122 4 Modelling and simulation of the anaerobic digestion process

Nomenclature
No Components Descriptions Unit
36 Top Operation temperature °C
conditions
operating
Process

37 qin Flow rate m3 day−1

38 Patm Atmospheric pressure bar

References
Arslan, D., Steinbusch, K. J. J., Diels, L., Hamelers, H. V. M., Strik, D. P. B. T. B., Buisman,
C. J. N., and Wever, H. D. (2016). Selective short-chain carboxylates production: A
review of control mechanisms to direct mixed culture fermentations. Critical Reviews in
Environmental Science and Technology, 46(6):592–634.

Bai, J., Liu, H., Yin, B., and Ma, H. (2015). Modeling of enhanced VFAs production from
waste activated sludge by modified ADM1 with improved particle swarm optimization for
parameters estimation. Biochemical Engineering Journal, 103:22–31.

Batstone, D., Keller, J., Angelidaki, I., Kalyuzhnyi, S., Pavlostathis, S., Rozzi, A., Sanders,
W., Siegrist, H., and Vavilin, V. (2002). Anaerobic digestion model No 1 (ADM1). Water
science and technology : a journal of the International Association on Water Pollution
Research, 45:65–73.

Batstone, D. J. and Keller, J. (2003). Industrial applications of the IWA anaerobic digestion
model No. 1 (ADM1). Water science and technology : a journal of the International
Association on Water Pollution Research, 47(12):199–206.

Bristol, E. (1966). On a new measure of interaction for multivariable process control. IEEE
Transactions on Automatic Control, 11(1):133–134.

Calt, E. A. (2015). Products Produced from Organic Waste Using Managed Ecosystem
Fermentation. Journal of Sustainable Development, 8(3):43–51.

Chen, Y., Luo, J., Yan, Y., and Feng, L. (2013). Enhanced production of short-chain fatty
acid by co-fermentation of waste activated sludge and kitchen waste under alkaline condi-
tions and its application to microbial fuel cells. Applied Energy, 102:1197–1204.

Donoso-Bravo, A., Mailier, J., Martin, C., Rodrı́guez, J., Aceves-Lara, A., and Wouwer, A. V.
(2011). Model selection, identification and validation in anaerobic digestion: A review.

Fukuzaki, S., Nishio, N., Shobayashi, M., and Nagai, S. (1990). Inhibition of the fermentation
of propionate to methane by hydrogen, acetate, and propionate. Applied and Environmen-
tal Microbiology, 56(3):719–723.

Garcia, C. E. and Morari, M. (1982). Internal model control. A unifying review and some new
results. Industrial Engineering Chemistry Process Design and Development, 21:308–323.
 124 References

Garcı́a-Sandoval, J. P. (2009). Cascade hybrid control for anaerobic digestion systems. IFAC
Proceedings Volumes (IFAC-PapersOnline), 7(PART 1):183–188.

Jiang, J., Zhang, Y., Li, K., Wang, Q., Gong, C., and Li, M. (2013). Volatile fatty acids pro-
duction from food waste: Effects of pH, temperature, and organic loading rate. Bioresour.
Technol, 143:525–530.

Jones, R. J., Massanet-Nicolau, J., Guwy, A., Premier, G. C., Dinsdale, R. M., and Reilly,
M. (2015). Removal and recovery of inhibitory volatile fatty acids from mixed acid fer-
mentations by conventional electrodialysis. Bioresource Technology, 189:279–284.

Jurado, E., Antonopoulou, G., Lyberatos, G., Gavala, H., and Skiadas, I. (2016). Continuous
anaerobic digestion of swine manure: ADM1-based modelling and effect of addition of swine
manure fibers pretreated with aqueous ammonia soaking. Applied Energy, 172:190–198.

Kim, H. J., Choi, Y. G., Kim, D. Y., Kim, D. H., and Chung, T. H. (2005). Effect of
pretreatment on acid fermentation of organic solid waste. Water Science and Technology,
52(1-2):153–160.

Kleerebezem, R. and van Loosdrecht, M. C. M. (2006). Critical analysis of some concepts

proposed in ADM1. Water science and technology : a journal of the International Asso-
ciation on Water Pollution Research, 54(4):51–57.

Larsson, T. and Skogestad, S. (2000). Plantwide control - a review and a new design proce-
dure. Modeling, Identification and Control, 21(4):209–240.

Lee, W. S., Chua, A. S. M., Yeoh, H. K., and Ngoh, G. C. (2014). A review of the production
and applications of waste-derived volatile fatty acids. Chemical Engineering Journal,
235:83–99.

Li, H., Chen, Z., Fu, D., Wang, Y., Zheng, Y., and Li, Q. (2020). Improved ADM1 for
modelling C, N, P fates in anaerobic digestion process of pig manure and optimization
approaches to biogas production. Renewable Energy, 146:2330–2336.

Lim, S., Kim, B., Jeong, C., Choi, J., Ahn, Y., and Chang, H. (2008). Anaerobic organic
acid production of food waste in once-a-day feeding and drawing-off bioreactor. Bioresour
Technol, 99 (16):7866–7874.

Liu, H., Wang, J., Liu, X., Fu, B., Chen, J., and Yu, H. (2012). Acidogenic fermentation of
proteinaceous sewage sludge: effect of pH. Water Res., 46 (3):799–807.

Lohani, S. P., Wang, S., Lackner, S., Horn, H., Khanal, S. N., and Bakke, R. (2016). ADM1
modeling of UASB treating domestic wastewater in Nepal. Renewable Energy, 95:263–268.
References 125

Manjusha, C. and Beevi, S. (2016). Mathematical Modeling and Simulation of Anaerobic

Digestion of Solid Waste. Procedia Technology, 24:654–660.

MathWorks (2021). linmod. Extract continuous-time linear state-space model around ope-
rating point.

Min, K. S., Khan, A. R., Kwon, M. K., Jung, Y. J., Yun, Z., and Kiso, Y. (2005). Acidogenic
fermentation of blended food-waste in combination with primary sludge for the production
of volatile fatty acids. Journal of Chemical Technology & Biotechnology, 80(8):909–915.

Mirlekar, G., Gebreslassie, B., Diwekar, U., and Lima, F. V. (2018). Biomimetic model-based
advanced control strategy integrated with multi-agent optimization for nonlinear chemical
processes. Chemical Engineering Research and Design, 140:229–240.

Nguyen, D., Gadhamshetty, V., Nitayavardhana, S., and Khanal, S. K. (2015). Automatic
process control in anaerobic digestion technology: A critical review. Bioresource Techno-
logy, 193:513–522.

Nordlander, E., Thorin, E., and Yan, J. (2017). Investigating the possibility of applying an
ADM1 based model to a full-scale co-digestion plant. Biochemical Engineering Journal,
120:73–83.

Paes, L. A. B., Bezerra, B. S., Deus, R. M., Jugend, D., and Battistelle, R. A. G. (2019).
Organic solid waste management in a circular economy perspective – A systematic review
and SWOT analysis. Journal of Cleaner Production, 239:118086.

Pataro, I. M., da Costa, M. V., and Joseph, B. (2020). Closed-loop dynamic real-time
optimization (CL-DRTO) of a bioethanol distillation process using an advanced multilayer
control architecture. Computers and Chemical Engineering, 143:107075.

Pessoa, R. W., Mendes, F., Oliveira, T. R., Oliveira-Esquerre, K., and Krstic, M. (2019). Nu-
merical optimization based on generalized extremum seeking for fast methane production
by a modified ADM1. Journal of Process Control, 84:56–69.

Pind, P., Angelidaki, I., and Ahring, B. (2003). Dynamics of the anaerobic process: Effects
of volatile fatty acids. Biotechnology and Bioengineering, 82(7):791–801.

Ramirez, I. D., Ramı́rez Villareal, A., and Ospina Nieto, J. D. (2019). Modelado y control no
lineal de ácidos grasos volátiles en un biorreactor anaerobio de lecho fijo y flujo ascendente
(UAFBR). Revista Ingenierı́as Universidad de Medellı́n, 19(36):53–69.

Redwood, M. D., Orozco, R. L., Majewski, A. J., and Macaskie, L. E. (2012). Electro-
extractive fermentation for efficient biohydrogen production. Bioresource Technology,
107:166–174.
 126 References

Rizos, V., Rizos, V., Tuokko, K., Tuokko, K., Behrens, A., Behrens, A., Vasileios Rizos,
V. R., Katja Tuokko, K. T., and Arno Behrens, A. B. (2017). The Circular Economy:
A review of definitions, processes and impacts. CEPS Research Report No 2017/8, April
2017.

Rosen, C. and Jeppsson, U. (2006). Aspects on ADM1 Implementation within the BSM2
Framework. Technical report, pages 1–37.

Seborg, D. E., Edgar, T. F., Mellichamp, D. A., and Doyle III, F. J. (2010). Process Dynamics
and Control. United States of America, 3rd editio edition.

Skogestad, S. (2004). Control structure design for complete chemical plants. Computers and
Chemical Engineering, 28(1-2):219–234.

Skogestad, S. and Postlethwaite, I. (2005). Multivariable Feedback Control: Analysis and

Design, volume 2.

Sun, H., Yang, Z., Zhao, Q., Kurbonova, M., Zhang, R., Liu, G., and Wang, W. (2021).
Modification and extension of anaerobic digestion model No.1 (ADM1) for syngas bio-
methanation simulation: From lab-scale to pilot-scale. Chemical Engineering Journal,
403:126177.

Toerien, D. F. and Hattingh, W. H. (1969). Anaerobic digestion I. The microbiology of

anaerobic digestion. Water Research, 3(6):385–416.

Wainaina, S., Awasthi, M. K., Sarsaiya, S., Chen, H., Singh, E., Kumar, A., Ravindran,
B., Awasthi, S. K., Liu, T., Duan, Y., Kumar, S., Zhang, Z., and Taherzadeh, M. J.
(2020). Resource recovery and circular economy from organic solid waste using aerobic
and anaerobic digestion technologies.

Wang, K., Yin, J., Shen, D., and Li, N. (2014). Anaerobic digestion of food waste for volatile
fatty acids (VFAs) production with different types of inoculum: effect of pH. Bioresource
technology, 161:395–401.

Wang, Y., Zhang, Y., Wang, J., and Meng, L. (2009). Effects of volatile fatty acid concentra-
tions on methane yield and methanogenic bacteria. Biomass and Bioenergy, 33(5):848–853.

Yu, Z., Leng, X., Zhao, S., Ji, J., Zhou, T., Khan, A., Kakde, A., Liu, P., and Li, X. (2018). A
review on the applications of microbial electrolysis cells in anaerobic digestion. Bioresource
Technology, 255:340–348.

Zacharof, M. P. and Lovitt, R. W. (2013). Complex effluent streams as a potential source of

volatile fatty acids. Waste and Biomass Valorization, 4(3):557–581.
References 127

Zhang, B., Zhang, L.-L., Zhang, S.-C., Shi, H.-Z., and Cai, W.-M. (2005). The Influence of
pH on Hydrolysis and Acidogenesis of Kitchen Wastes in Two-phase Anaerobic Digestion.
Environmental Technology, 26(3):329–340.

Zhou, H., Ying, Z., Cao, Z., Liu, Z., Zhang, Z., and Liu, W. (2020). Feeding control of
anaerobic co-digestion of waste activated sludge and corn silage performed by rule-based
PID control with ADM1. Waste Management, 103:22–31.

Zhou, M., Yan, B., Wong, J. W., and Zhang, Y. (2018a). Enhanced volatile fatty acids pro-
duction from anaerobic fermentation of food waste: A mini-review focusing on acidogenic
metabolic pathways. Bioresource Technology, 248:68–78.

Zhou, M., Yan, B., Wong, J. W., and Zhang, Y. (2018b). Enhanced volatile fatty acids pro-
duction from anaerobic fermentation of food waste: A mini-review focusing on acidogenic
metabolic pathways. Bioresource Technology, 248:68–78.
5 Conclusions and Recommendations

5.1. Conclusions
This thesis evaluated and analyzed the anaerobic digestion process from a techno-economic
point of view under the circular economy philosophy. For this, experimental assays focused
on the methane production of food waste from Colombia were carried out. A comparison of
the economic viability of the process for the production of methane and platform products
such as VFAs was performed. Finally, by using Anaerobic digestion model No 1 (ADM1), a
closed-loop control system was implemented with a real-time optimization layer that enhan-
ces the production and the maximum possible economic profits from the sale of VFAs under
input disturbances.

The experimental part of this thesis focused on characterization food waste from the de-
partment of Nariño-Colombia and the evaluation of the methane production through the
anaerobic digestion process. The purpose of this study was to evaluate the potential for
organic waste reuse for the renewable energy production, identifying properties of the subs-
trate such as Carbon/Nitrogen ratio, Volatile Solids (VS) and Total Solids (TS), proximate
analysis and ultimate analysis, which provide relevant information for waste management.
Likewise, the experimental assays contributed to the understanding of the performance of
the digestion process of these wastes by monitoring process variables such as VS/TS ratio,
pH, temperature, removal of soluble COD, biogas production and methane concentration.
The results obtained show that local food waste has great potential for the renewable energy
production, suggesting that the application of this technology can contribute and encourage
sustainable development, providing social, economic and environmental solutions.

The economic comparison of the production of biomethane and VFAs from the anaerobic
digestion of food waste, found a different perspective of the digestion process, since it was
proven that orienting the process towards the production of VFAs may be more economically
viable than the process oriented to the generation of biogas. From this, the digestion process
could be classified as a profitable and lucrative process. The results of this study identified
that the main reason why VFAs production is economically viable was mainly due to the low
hydraulic retention time (HRT), lower reactor volume and high sales price of VFAs. Addi-
tionally, by implementing this economic study in the Colombian context, a circular economy
in underdeveloped countries is promoted, generating new business perspectives that in turn
5.2 Recommendations 129

provide a response to current problems derived from non-renewable energies, such as envi-
ronmental pollution and limited availability of resources.

The special focus of this thesis was the simulation and modeling of the anaerobic digestion
process, towards on improving the production of VFAs. The implementation of ADM1 was a
platform that allowed to understand the dynamic behavior of the anaerobic digestion process
and, through its manipulation, scenarios to obtain products of interest could be estimated an
evaluation of techno-economic viability incorporating a real-time optimization layer closed-
loop control. The results of this investigation showed that it is possible to maximize the
production of VFAs by controlling the pH of the process, being a variable governed by the
concentrations of cations and anions in the system. The developed control computational
platform brings good perspectives to simulate various scenarios, such as different control
strategies, with and without the real-optimization layer as well, allowing the study of va-
rious scenarios of the anaerobic digestion process. The proposed control structure can help
VFAs production investigations through anaerobic digestion, to operate optimally and to
achieve the highest possible economic benefit.

5.2. Recommendations
Through the cases studied in this thesis, it was possible to identify problems and situations
that can promote further research:

In the Biochemical Methane Potential (BMP) assays, it was not possible to quantify
the VFAs concentration. The monitoring of these properties can identify the phases
of the digestion process, as well as the performance of these products can serve as an
input variable for the implemented economic model.

In the sacrifice bottle assay, methane production was not measured continuously. Fac-
tors such as access to the setup of the experiment prevented this activity from being
carried out. Measurement of this variable can make the results from this assay compa-
rable with the results of the on-line methane measurement assay.

According to the experimental assays, it is recommend to degas the inoculum for 20

days and measure the gas production of the bottles every 5 hours for the first two days.

The recommended method to determine the amount of nitrogen is the TKN method,
because it measures biodegradable nitrogen, however, this measurement can be faster
and more reliable using the methodology described in this research.

It is suggested to validate the methane concentration measurement reported by the

Yieldmaster equipment with a gas chromatography measurement of the same sample,
130 5 Conclusions and Recommendations

since values greater than 70 % v/v of methane are very unlikely.

For the economic evaluation, methane production and VFAs production parameters
from the literature were used. The economic study with local values could not be
carried out, because the VFAs yield was not quantified (only the quality of the biogas
was used) in the experimental assays. It is recommended to implement the economic
model with these results to evaluate of the feasibility of these processes in Colombia.

Substrate pretreatment costs, tools and equipment such as solid/liquid separator pipes,
conduits, pumps, mixers, among others, were not taken into account in the economic
model, because the study was a first approximation to give an overview of the process.
The reason for this was due to the lack of information in the literature. It is suggested
to carry out a process scheme with more equipment and tools to obtain more accurate
results.

In the techno-economic analysis, it is recommended to carry out a sensitivity analysis

in terms of variability once the performance data of the anaerobic digestion products
are available in the local context.

In a more rigorous analysis of the comparison and economic evaluation of the products
generated in the anaerobic digestion process, it is recommended to include unit pro-
cesses and operations that contemplate the pretreatment of waste, such as: pumps,
valves, compounds separation stage with low biodegradability, crushing stage, and sta-
ge of disposition of low biodegradability compounds and biol.

An ideal separation of the interest products (biomethane and VFAs) was determined, in
order to find the maximum economic profitability of the processes. It is recommended
to use real values that determine the purification or separation of the products.