Accepted Manuscript

Short Communication

Cocoa residues as viable biomass for renewable energy production through

anaerobic digestion.

Nayaret Acosta, Jo De Vrieze, Verónica Sandoval, Danny Sinche, Isabella

Wierinck, Korneel Rabaey

PII: S0960-8524(18)30768-5
DOI: https://doi.org/10.1016/j.biortech.2018.05.100
Reference: BITE 20006

To appear in: Bioresource Technology

Received Date: 21 March 2018

Revised Date: 29 May 2018
Accepted Date: 30 May 2018

Please cite this article as: Acosta, N., De Vrieze, J., Sandoval, V., Sinche, D., Wierinck, I., Rabaey, K., Cocoa
residues as viable biomass for renewable energy production through anaerobic digestion., Bioresource
Technology (2018), doi: https://doi.org/10.1016/j.biortech.2018.05.100

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Title: Cocoa residues as viable biomass for renewable energy production through anaerobic

digestion.

Nayaret Acosta1, Jo De Vrieze1, Verónica Sandoval2, Danny Sinche3, Isabella Wierinck4,

Korneel Rabaey1*
1
Center for Microbial Ecology and Technology (CMET), Ghent University, Coupure Links

653, B-9000 Gent, Belgium

2
Universidad de las Fuerzas Armadas ESPE, Av. Gral. Rumiñahui s/n, Sangolquí, Ecuador
3
Instituto de Eficiencia Energética y Energías Renovables, Iñaquito N35-37, Quito, Ecuador
4
Organic Waste Systems (OWS), Dok Noord 5, B-9000 Gent, Belgium

Manuscript submitted to Bioresource Technology

Abstract

The aim of this work was to evaluate the bioenergy potential of cocoa residue via anaerobic

digestion. Batch and fed-batch lab-scale reactors were operated under low and high solids

conditions. In the batch tests, 59 ± 4% of Chemical Oxygen Demand (COD) was recovered as

methane. This corresponded with an average methane yield of 174 (wet) and 193 (dry) L kg-1

volatile solids fed, whereas a series of fed-batch reactors produced 70 ± 24 (wet) and 107 ± 39

(dry) L CH4 kg-1 volatile solids fed during stable conditions. A case study was developed for

canton Balao (Ecuador) based on our experimental data, operational estimates and available

cocoa waste in the area. Annually, 8341 MWh could be produced, meeting 88% of the current

electricity demand in Balao. This case study proves the potential for cocoa waste as a source

of renewable energy in rural areas.

Keywords: biogas, cocoa waste, bio-energy, methane, agricultural residues

*
Correspondence to: Korneel Rabaey, Ghent University; Faculty of Bioscience Engineering; Center for
Microbial Ecology and Technology (CMET); Coupure Links 653; B-9000 Gent, Belgium; phone: +32 (0)9 264
59 76; fax: +32 (0)9 264 62 48; E-mail: Korneel.Rabaey@UGent.be; Webpage: www.cmet.Ugent.be.

1
1. Introduction

The use of agricultural biomass for anaerobic digestion (AD) can be considered a suitable

approach with environmental and economic benefits (Paudel et al., 2017). The main product

of AD is biogas, which contains methane, a versatile carrier of renewable energy.

The AD is a well-studied and mature technology, but there remain several ways in which

energy recovery can be improved. One strategy is the introduction of new feedstocks for

biomass conversion. To our knowledge, the potential for energy production from cocoa

residue using AD has not been described. Many of the challenges faced in AD nowadays are

linked to the nature of the feedstock. Thus, it is highly relevant to assess potential biogas

feedstocks to evaluate their suitability for long-term AD (Ammenberg & Feiz, 2017).

Cocoa is one of the most important crops produced in tropical regions, such as Ecuador,

which has a production of 234,000 metric tons of cocoa beans per year (ICCO, 2016). Each

ton of dry cocoa beans generates 10 to 15 tons of wet cocoa waste, consisting mainly of the

pods (Mansur et al., 2014; Preinversión, 2014). Pods are typically left to decompose on the

cocoa plantation, producing bad odors and exacerbating the propagation of plant diseases,

including black pod rot (Mansur et al., 2014).

The large volumes generated, combined with their composition, make cocoa waste an

attractive feedstock for AD. Currently, cocoa waste is discarded directly onto the land The

AD can stabilize the organic matter, and help retain the nutrients, thereby increasing the value

towards land application and reducing potential negative impacts, e.g., plant diseases.

The main objective of this research was to determine biogas and methane yields from cocoa

residues. The results were compared with agricultural residues to evaluate the quality of this

biomass as a potential new feedstock for energy production. The biogas potential was initially

assessed through a biochemical methane potential (BMP) test, followed by fed-batch reactors

(FBR). The data were used for a case study design of a full-scale plant in Balao (Ecuador).

2
2. Methods

2.1. Inoculum and feedstock characterization

Two inocula were used for the experiments. Inoculum 1 was obtained from a full-scale biogas

plant treating fruit residues and cow manure (Sangolquí, Ecuador), and used for the first BMP

test (BMP1) and for the FBR. Inoculum 2 was used for BMP2 and was obtained from a lab-

scale mesophilic digester using waste activated sludge as feedstock (Gent, Belgium). Cocoa

pods were collected from the National Institute of Agricultural Research (Portoviejo,

Ecuador). Cocoa waste includes the pericarp, exocarp, mesocarp and a thin layer of endocarp

of the fruit. The particle size of the feedstock was reduced with a kitchen blender into pieces

of less than 2 mm. and stored at 4°C until use.

2.2. Experimental design

2.2.1. Biochemical Methane Potential tests

The BMPs were run in Schott bottles with a liquid volume of 400 mL (500 mL total volume).

The temperature of the reactors was kept at 29 ± 1 °C, using an electronic aquarium heater

(Atman, China) submerged in water in plastic containers as water baths. The treatments were:

BMP1 (cocoa + inoculum 1, wet), BMP2 (cocoa + inoculum 2, wet) and Dry AD (Dranco).

The concentration of feedstock was 4 g COD L-1. A set of blank assays was performed for

endogenous methane production determination. Triplicate assays for all the conditions were

performed. Experiments were finished when the biogas production reached a plateau. Biogas

production was measured daily using liquid displacement in separated columns connected to

the digesters. Biogas composition was measured at the end of the experiment. The TS,

volatile solids (VS), total Kjeldahl nitrogen (TKN), chemical oxygen demand (COD) and

volatile fatty acids (VFA) were determined at the beginning and at the end of the experiment.

The results are expressed at standard temperature (273 K) and pressure (101,325 Pa)

conditions (STP). The Dranco test followed the method ISO15985(2014).

3
2.2.2. Fed-batch Reactors

Experiments were carried out in the same type of reactors and in similar operational

conditions as in BMP tests. Wet AD reactors were fed with cocoa residues and distilled water,

while dry AD reactors were fed with cocoa residues only. The amount of cocoa waste added

was based on the organic loading rate (OLR) and hydraulic retention time (HRT). Feeding

and mixing was performed manually three times per week. Biogas production and

composition were evaluated in the same way as for the BMP. Gas composition was monitored

weekly. Samples of the effluent were collected and stored at 4 °C until use to evaluate TS,

VS, TKN, COD and VFA. The pH was controlled by the addition of NaOH (0.1 M) when

needed to maintain neutral pH of ~ 7.0. Results are expressed as L per kg of VS fed.

2.2.3. Case study- Theoretical Energy Potential

With the biogas and methane values obtained from experimental results and biomass

information from Preinversión (2014), a theoretical energy potential was calculated for a

specific region in Ecuador, the canton Balao. This canton was selected as this area gathers the

highest residue density generation, more than 90 000 tons of fresh cocoa waste per year in a

radius of 11.4 km. The region, with a surface area of 409.10 km2, hosts 5238 families

(houses) (INEC, 2012). The equations and the standard values used for the calculation of the

theoretical potential energy production were from Bundhoo et al. (2016) and Ruffino et al.

(2015). Within the scope of our study, only a theoretical energy potential was calculated to

simplify the evaluation of cocoa waste as feedstock for energy production in the Balao region,

the technical and economic potential were not considered.

2.3. Analytical techniques

The TS, VS, TKN and COD were measured using Standard Methods (Greenberg AE, 1992).

The VFA content was analyzed according to Standard methods for the examination of water

and wastewater (APHA, 2012). The pH was monitored with a pH meter C532 (Consort,

4
Turnhout, Belgium). A compact gas chromatograph (Global Analyser Solutions, Breda, The

Netherlands) was used to measure the biogas composition. In the first channel, CH4, O2, N2

and H2 (Porabond Q pre-column and Molsieve 5A column) were measured, while in the

second channel (Rt-QS-bond column and pre-column) CO2 was measured. The detection limit

was 100 ppmv for each gas, using a thermal conductivity detector.

2.4 Calculation and analysis of data

The R Studio software version 3.3.1, was used for statistical analysis (http://www.r-

project.org). The Kolmogorov-Smirnov and Shapiro-Wilk test were used to check normality,

and homoscedasticity was analyzed with the Levene test. When normality and

homoscedasticity were confirmed, the Student t-test was used to evaluate significant

differences. If normality and homoscedasticity were not confirmed, the non-parametric

Wilcoxon Signed Rank test was used.

3. Results and Discussion

3.1. Biochemical Methane Potential tests

The BMP test evaluates the quality of the feedstock for biogas production, and it is typically

used to estimate the potential of a full-scale biogas plant. The biogas yields obtained in this

experiment were 307 ± 15 L kg-1 VS and 271 ± 70 L kg-1 VS, which corresponds to methane

yields of 175 ± 9 L kg-1 VS and 173 ± 45 L kg-1 VS for BMP1 and BMP2, respectively, under

mesophilic conditions, after 30 days of operation (Figure 1). In thermophilic conditions, the

Dranco process resulted in 345 L kg-1 VS, with a methane yield of 193 L kg-1 VS. All

experiments reached a plateau after approximately 10 days. After 21 days, no additional gas

production was observed. The methane content in the biogas was 57% for BMP1, 64% for

BMP2 and 56% for Dranco. No H2S was observed. The values obtained from BMP tests are

comparable to yields of other agricultural feedstocks rich in carbohydrates, such as residues

from palm oil, banana and rice plantations (Preinversión, 2014). Biogas yields from these

5
crops range from 220 to 630 L kg-1 VS and methane yields from 81 to 400 L kg-1 VS (Fiore et

al., 2016; Raposo et al., 2012). AD of agricultural residues targets both stabilization of

organic waste and in-situ energy recovery. Small and medium scale anaerobic digesters, using

agricultural residues, are becoming a common approach in Latin America.

In terms of conversion, 51 ± 2 %, 50 ± 13 % and 51% of the initial COD was converted into

methane in BMP1, 2 and the Dranco test, respectively. These values are similar to conversion

efficiencies of cow manure, food grease, food waste and sugar beets (Moody et al., 2011).

Most of the biomass conversion to biogas occurred within the first 5 days of experiment for

all BMP tests (Figure 1), indicating that cocoa waste was easily biodegraded within the first

week in all three treatments. The operational conditions, such as TS content and type of

inoculum, did not affect the rate of transformation into biogas. The fractions were rapidly

degraded to biogas might had corresponded to the sugars present in the outer layer of the

endocarp (rich in pectin),as well as the lighter fractions of hemicellulose (11%); cellulose

(35%) and pectin (6%) in the mescocarp (Sobamiwa & Longe, 1994).

These results indicate that cocoa waste could indeed be an attractive feedstock for AD due to

(i) a similar biogas and methane production rate compared with other above-mentioned

feedstocks (ii) availability of large quantities relative to crop and (iii) the potential benefits of

stabilized waste back to agricultural land over the direct application of untreated waste.

3.2. Semi-continuous reactors

The results obtained from BMP were used as a basis for long-term tests in which the

degradation potential can be assessed under more realistic conditions. Two types of

treatments were performed for AD of cocoa waste in semi-continuous mode: wet AD and dry

AD, which differ in the TS content of the reactor. Dry AD performed significantly better than

wet AD in terms of biogas yield (P=0.0001) and methane yield (P<0.0001), as well as in

terms of pH, VFA accumulation and VS conversion. In wet AD reactors during the first 15

6
days, the biogas yield was 246 ± 96 L kg-1 VS with a methane yield of 75 ± 30 L kg-1 VS

(Figure 2). The highest biogas yield was obtained on day 7, with 418 ± 3 L kg-1 VS. The

biogas contained more CO2 than CH4, which is common during the start-up process. On day

18, there was an operational disruption on both treatments, which is reflected in the biogas

and methane production. From day 20 to 39, wet AD reached stable conditions, with a biogas

yield of 192 ± 33 L kg-1 VS and a methane content of 39 ± 4 %. From day 40 on, a permanent

decrease in the biogas and methane yield occurred, with even lower production when the

OLR was further increased up to 4 g VS L−1 d−1 on day 49. The biogas and methane yield

during period III were 46 ± 13 L kg VS−1 and 9 ± 4 L kg-1 VS, respectively.

In dry AD reactors, the biogas and methane production were similar to wet AD during the

first period, being 205 ± 52 and 72 ± 18 L kg-1 VS, respectively (Figure 2). From period II on,

the performance of dry AD was better than wet AD. At an OLR of 2 g VS L−1 d−1, the biogas

yield was 288 ± 87 L kg-1 VS and the methane yield was twice as high as in the wet AD,

reaching 122 ± 34 L kg-1 VS even though the methane content of the biogas was similar in

both cases (43 ± 7 % dry AD). However, the dry AD also failed 7 days after the operational

failure of the wet AD. From day 46 on, there was a decrease in the biogas and methane yield.

The failure was more evident when the OLR was increased to 4 g VS L−1 d−1 on day 49.

During the III period, the average biogas and methane yield were 109 ± 31 and 37 ± 12 L kg-1

VS, respectively. Due to a low biogas yield, the experiments were terminated on day 74.

Dry AD results were better in comparison to wet AD in terms of biogas and methane yields

and operational parameters. Nowadays, dry AD is being studied widely, because of its

multiple benefits, as it requires lower input energy, less wastewater is generated, and a low

moisture digestate is remaining (Arelli et al., 2018). Dry AD could work at higher OLR and

TS values than wet AD (Arelli et al., 2018). In the present study, the performance of dry AD

was better than wet AD, but both failed when the OLR was set at 4 g VS L-1d-1.

7
The VFAs and pH were monitored to assess incomplete degradation of organic matter from

biomass to methane and the buffering capacity of the media, respectively. An accumulation of

VFAs was observed in both treatments, which resulted in a decrease in pH and inhibition of

methanogenesis, as well as an increase of CO2 in the headspace. Control of pH with 0.1 M

NaOH was needed from day 32 and 60 onwards in the wet and dry AD reactors, respectively.

The VFA concentration, expressed as the acetic acid equivalents, increased in both systems.

For wet AD, the VFA concentration increased from 297 ± 14 mg L-1 (day 32) to 1066 ± 59

mg L-1 at the end of the experiment (day 74). In case of dry AD, the VFA accumulation

started from 109 ± 51 mg L-1 on day 37 until a final value of 807 ± 136 mg L-1 on day 74.

The fact that both treatments struggled to maintain methane production and a balanced pH,

indicates that the nature of the substrate, rich in lignin and with a lack of essential nutrients, is

the main point to consider on the long term. Low biogas production from AD fed with a

single feedstock is quite common, and is usually a result of nutrient deficiency and low buffer

capacity(Romero-Güiza et al., 2016). The limitation might be caused by the nature of the

agricultural residues, which involves recalcitrant structures, such as lignin (Sobamiwa &

Longe, 1994). The AD of this type of feedstock can become sustainable if the process can

overcome this issue (Mancini et al., 2018). There are several alternatives to increase biogas

and methane yields from agricultural residues, including co-digestion with nutrient-rich co-

substrates, pretreatment of the feedstock to increase the biodegradability, improved reactor

design and operational conditions and/or the supply of macro and micro-nutrients to stimulate

microbial activity and/or reduce the concentration of inhibitory agents (Mancini et al., 2018;

Romero-Güiza et al., 2016). The application of these methods to enhance cocoa waste

digestion can be an important subject of future studies.

For wet AD, the VS removal efficiency was 96 ± 1%, 82 ± 8% and 38 ± 13 % for period I, II

and III, respectively. Similarly, dry AD presented a VS removal of 64 ± 29% during period I,

8
73 ± 20% during period II and 49 ± 14 % in the last period. The majority of the feedstock was

converted to CO2. The low pH, increasing VFA concentrations and high CO2 in the resulting

biogas indicate the development of an acidogenic culture instead of the desired mixed

community with methanogenic archaea (Mancini et al., 2018).

The accumulation of VFA in AD can result from feedstock overload and is one of the most

common reasons for process deterioration (Franke-Whittle et al., 2014). Even though the

organic loading rate used here was relatively low, the BMP tests indicate that the cocoa waste

was rapidly degraded, allowing for subsequent VFAs accumulation. The higher pH buffering

capacity in dry AD is likely the reason for better performance compared to wet AD.

The FBR had a lower methane yield than the indicated by the BMP tests, which is expected

due to the optimized conditions in BMP. Overall, the yields of both BMP and FBR

configurations were comparable to those obtained for other agricultural residues, including

feedstocks, such as rice husks, rice straw, grape stalk, pomace, olive mill, and winery waste

(Fiore et al., 2016). The values of methane production are lower compared to a variety of

agricultural feedstocks with lower TS concentrations (Liotta et al., 2016), indicating the

importance of loading rate, buffering capacity and the nutrients inherent to each feedstock.

3.3 Case study- Preliminary electricity potential analysis from cocoa waste through

Anaerobic Digestion

The results obtained from the dry AD BMP tests were used as inputs in the design of a full-

scale digester using cocoa waste as feedstock. The dry AD was chosen for its biogas and

methane yields, as well as, higher operational stability. Moreover, considering the consistency

of the feedstock, dry AD appears a suitable approach.

As mentioned in section 2.2.3, the canton Balao was selected, as this area has the highest

density in cocoa waste generated per year (Preinversión, 2014). The parameters for the design

of the full-scale plant are shown in Table 1. It was assumed that cocoa waste is generated

9
continuously during the year at a rate of 93,377 tons per year (Preinversión, 2014). According

to Batista (2009), the harvest of cocoa can be considered mostly constant, with two or three

peaks during the year when the weather conditions (mainly rainfall) allow so. The full-scale

plant is designed to operate in mesophilic conditions, at a HRT of 30 days. The total volume

of the digester is 10,442 m3, implying that three or more reactors would need to be built. The

energy is transformed through a combined heat and power (CHP) unit into electricity and

heat. The power range of the CHP unit was estimated at 1MW and total energy production

was calculated assuming CHP efficiencies of 34.6% for electricity and 42.0% for thermal

energy (Bundhoo et al., 2016). The internal energy demand from the plant was assumed to be

equal to 3.4% for electrical consumption and 11% for thermal consumption (Bundhoo et al.,

2016), and these values were subtracted from the total energy produced by the plant. In this

scenario, the net available electrical energy production of the biogas plant reached 8341 MWh

per year. The theoretical energy generated from the plant would be sufficient to cover up to

88% of the estimated 9428.4 MWh of electricity consumed in Balao each year (INEC, 2012).

This model scenario, based on actual laboratory results, shows the potential of cocoa waste in

biogas-based power generation. However, several issues that need to be addressed. First, the

biogas and methane potential must be optimized, particularly in terms of overall operational

stability of the process. Among the different alternatives for optimization, co-digestion seems

the most suitable for this case. Second, the reuse of the digestate will need to be investigated,

including practices such as composting and direct application.

The advantage of AD is the creation of bio-energy while retaining the fertilizer value of the

organic waste. Decentralized energy production in rural areas has several advantages,

including lower costs in distribution and transmission of electricity (Nguyen, 2007) and local

employment (Nguyen, 2007). AD fits within a broader bio-refinery concept, allowing

maximal synergy between waste treatment and the production of value-added compounds.

10
The application of AD feedstocks back to land–either directly, as digestate, upon composting

or after pyrolysis to biochar demands further investigation as well (Mansur et al., 2014).

4. Conclusions

Up to 50% of the organic matter from cocoa waste was transformed into biogas with 60%

methane content. These values are equivalent to similar agricultural feedstocks, such as rice

husks, and winery waste. Dry AD process is preferred, due to higher methane yields and more

stable conditions. There is an opportunity to further optimize the process at lab- and pilot-

scale conditions. In this sense, co-digestion could be the key to optimize energy production

and operational stabilization. Due to its direct availability and abundance, cocoa waste will be

a valuable feedstock for renewable energy production in full-scale plants in rural areas.

E-supplementary data of this work can be found in online version of the paper.

Acknowledgments

This work was supported by the grant SENESCYT Convocatoria Abierta 2014.and INER. Jo

De Vrieze is supported as postdoctoral research fellow by the Research Foundation Flanders

(FWO-Vlaanderen). The company OWS contributed with dry batch test. We thank Pablo

Carvajal, Kristof Verbeeck and Stephen Andersen for critically reading the manuscript.

References

1. Ammenberg, J., Feiz, R. 2017. Assessment of feedstocks for biogas production, part
II—Results strategic decision making. Resources, Conservation and Recycling, 122, 388-404.
2. Angelonidi, E., Smith, S.R. 2015. A comparison of wet and dry anaerobic digestion
processes for the treatment of municipal solid waste and food waste. Water Environ. J., 29(4),
549-557.
3. APHA. 2012. Standard Methods for the Examination of Water and Wastewater.
4. Arelli, V., Begum, S., Anupoju, G.R., Kuruti, K., S, S. 2018. Dry anaerobic co-
digestion of food waste and cattle manure: Impact of total solids, substrate ratio and thermal
pre treatment on methane yield and quality of biomanure. Bioresource Technology, 253, 273-
280.
5. Batista, L. 2009. El cultivo de cacao. Santo Domingo: Centro para el Desarrollo

11
Agropecuario y Forestal.
6. Bundhoo, Z.M.A., Mauthoor, S., Mohee, R. 2016. Potential of biogas production from
biomass and waste materials in the Small Island Developing State of Mauritius. Renewable
and Sustainable Energy Reviews, 56, 1087-1100.
7. Fiore, S., Ruffino, B., Campo, G., Roati, C., Zanetti, M.C. 2016. Scale-up evaluation
of the anaerobic digestion of food-processing industrial wastes. Renewable Energy, 96, 949-
959.
8. Franke-Whittle, I.H., Walter, A., Ebner, C., Insam, H. 2014. Investigation into the
effect of high concentrations of volatile fatty acids in anaerobic digestion on methanogenic
communities. Waste Management, 34(11), 2080-2089.
9. Greenberg AE, C.L., Eaton AD. 1992. Standard methods for the examination of water
and wastewater. 18th edition 1992 ed.
10. ICCO. 2016. ICCO Quarterly Bulletin of Cocoa Statistics. International Cocoa
Organization.
11. INEC. 2012. Ecuador en cifras, Vol. 2015.
12. ISO15985. 2014. Plastics – Determination of the ultimate anaerobic biodegradation
and disintegration under high-solids anaerobic-digestion conditions. in: Method by analysis of
released biogas.
13. Liotta, F., Esposito, G., Fabbricino, M., van Hullebusch, E.D., Lens, P.N., Pirozzi, F.,
Pontoni, L. 2016. Methane and VFA production in anaerobic digestion of rice straw under
dry, semi-dry and wet conditions during start-up phase. Environ Technol, 37(5), 505-12.
14. Mancini, G., Papirio, S., Riccardelli, G., Lens, P.N.L., Esposito, G. 2018. Trace
elements dosing and alkaline pretreatment in the anaerobic digestion of rice straw.
Bioresource Technology, 247, 897-903.
15. Mansur, D., Tago, T., Masuda, T., Abimanyu, H. 2014. Conversion of cacao pod
husks by pyrolysis and catalytic reaction to produce useful chemicals. Biomass and
Bioenergy, 66, 275-285.
16. Moody, L.B., Burns, R.T., Bishop, G., Sell, S.T., Spajic, R. 2011. Using Biochemical
Methane Potential Assays to Aid in Co-substrate Selection for Co-digestion. 27(3).
17. Nguyen, K.Q. 2007. Alternatives to grid extension for rural electrification:
Decentralized renewable energy technologies in Vietnam. Energy Policy, 35(4), 2579-2589.
18. Paudel, S.R., Banjara, S.P., Choi, O.K., Park, K.Y., Kim, Y.M., Lee, J.W. 2017.
Pretreatment of agricultural biomass for anaerobic digestion: Current state and challenges.
Bioresource Technology, 245, 1194-1205.
19. Preinversión, I.N.d. 2014. Atlas Bioenergetico del Ecuador. First Edition ed.
20. Raposo, F., De la Rubia, M.A., Fernández-Cegrí, V., Borja, R. 2012. Anaerobic
digestion of solid organic substrates in batch mode: An overview relating to methane yields
and experimental procedures. Renewable and Sustainable Energy Reviews, 16(1), 861-877.
21. Romero-Güiza, M.S., Vila, J., Mata-Alvarez, J., Chimenos, J.M., Astals, S. 2016. The
role of additives on anaerobic digestion: A review. Renewable and Sustainable Energy
Reviews, 58, 1486-1499.
22. Ruffino, B., Fiore, S., Roati, C., Campo, G., Novarino, D., Zanetti, M. 2015. Scale
effect of anaerobic digestion tests in fed-batch and semi-continuous mode for the technical
and economic feasibility of a full scale digester. Bioresour Technol, 182, 302-13.
23. Scano, E.A., Asquer, C., Pistis, A., Ortu, L., Demontis, V., Cocco, D. 2014. Biogas
from anaerobic digestion of fruit and vegetable wastes: Experimental results on pilot-scale
and preliminary performance evaluation of a full-scale power plant. Energy Conversion and
Management, 77, 22-30.
24. Sobamiwa, O., Longe, O.G. 1994. Utilization of cocoa-pod pericarp fractions in
broiler chick diets. Animal Feed Science and Technology, 47(3), 237-244.

12
25. Weiland, P. 2010. Biogas production: current state and perspectives. Applied
Microbiology and Biotechnology, 85(4), 849-860.
Tables:

Table 1 Design parameters of a full-scale anaerobic digester using cocoa waste as single

feedstock

Design parameters Unit Value

Waste generation ton FW y-1 93,377
Volatile Solids VS content ton VS Tn-1 FW 0.13
Specific biogas production m3 ton-1 VS 345
Specific methane production m3 ton-1 VS 193
Hydraulic Retention Time HRT Days 30
Digester filling coefficient – 0.75
Electrical engine efficiency Ηelectrical % 34.6
Thermal engine efficiency Ηthermal % 42
Biogas Low Heating Value LHV MJ m-3 22
Substrate density ρm ton m-3 0.98
Working year Days 365
Engine work hours Hours 24
Total volume of the digester m3 10,442
Functional volume of the digester m3 7,831
Biogas generation m3 y-1 4,083,761
Methane generation m3 y-1 2,281,567
Total energy production MWh y-1 24,956
Total thermal energy generation MWh y-1 10,482
Total electrical energy generation MWh y-1 8,635
On site thermal energy consumption MWh y-1 1,153
On site electrical energy consumption MWh y-1 294
Net thermal energy generation MWh y-1 9,329
Net electricity production MWh y-1 8,341
Combined Heat and Power CHP output MW 1

13
Figures:

200
Cumulative methane production ( L kg-1 VS)

150

100

Dranco
BMP 1
50
BMP 2

0
0 5 10 15 20 25 30
Time (days)

Figure 1 Cumulative methane production in BMP 1, BMP2 and Dranco tests using cocoa
residues as feedstock minus the methane value obtained from the control (only with inoculum).
The difference between BMP1 and BMP2 is the type of the inoculum used for each experiment.
For BMP1 the inoculum used was anaerobic sludge from a pilot anaerobic digester treating fruit
waste in Sangolquí, Ecuador For BMP 2 the inoculum was obtained from a mesophilic lab scale
digester using activated sludge as feedstock (Gent, Belgium). Dranco corresponds with dry
anaerobic digestion.

14
Figure 2 Specific biogas and methane yield in wet (a) and dry (b) anaerobic-digestion of cocoa
residue at different OLR values, divided into the three phases. These correspond to 1 (wet), 1.6
(dry), 2 and 4 g VS L-1 d-1, respectively, for periods I, II and III.

15
Highlights

 Cocoa waste is a feasible biomass resource for energy production through anaerobic

digestion.

 Dry anaerobic digestion of cocoa waste has more stable performance

 Up to 88% of the electricity demand of a small region in Ecuador could be potentially

covered.

16