RESOURCE RECOVERY & REUSE SERIES 16 Global Experiences on Waste Processing with Black Soldier Fly (Hermetia illucens): From Technology to Business Gabrielle Joly and Josiane Nikiema 16 ISSN 2478-0529 About the Resource Recovery & Reuse Series Resource Recovery and Reuse (RRR) is a subprogram of the CGIAR Research Program on Water, Land and Ecosystems (WLE) dedicated to applied research on the safe recovery of water, nutrients and energy from domestic and agro-industrial waste streams. This subprogram aims to create impact through different lines of action research, including (i) developing and testing scalable RRR business models, (ii) assessing and mitigating risks from RRR for public health and the environment, (iii) supporting public and private entities with innovative approaches for the safe reuse of wastewater and organic waste, and (iv) improving rural-urban linkages and resource allocations while minimizing the negative urban footprint on the peri-urban environment. This subprogram works closely with the World Health Organization (WHO), Food and Agriculture Organization of the United Nations (FAO), United Nations Environment Programme (UNEP), United Nations University (UNU), and many national and international partners across the globe. The RRR series of documents present summaries and reviews of the subprogram’s research and resulting application guidelines, targeting development experts and others in the research for development continuum. IN PARTNERSHIP WITH: Global Experiences on Waste Processing with Black Soldier Fly (Hermetia illucens): From Technology to Business Gabrielle Joly and Josiane Nikiema RESOURCE RECOVERY & REUSE SERIES 16 ii The authors Ms. Gabrielle Joly is an environmental engineer. She has an MSc in Environmental Engineering from KTH Royal Institute of Technology, Sweden, and an MSc in General Engineering from the Ecole Centrale de Lyon, France. Her fields of expertise include organic solid waste valorization, process engineering for waste management, and sustainable social and economic development. Dr. Josiane Nikiema is a Senior Researcher – Environmental Science at the International Water Management Institute (IWMI) and is based in Accra, Ghana. She has a PhD in Chemical Engineering from the Université de Sherbrooke, Canada. Her fields of expertise include wastewater treatment and reuse, recovery of nutrients and organic matter from fecal sludge and organic solid waste, and testing business models for safe resource recovery and reuse. Joly, G.; Nikiema, J. 2019. Global experiences on waste processing with black soldier fly (Hermetia illucens): from technology to business. Colombo, Sri Lanka: International Water Management Institute (IWMI). CGIAR Research Program on Water, Land and Ecosystems (WLE). 62p. (Resource Recovery and Reuse Series 16). doi: 10.5337/2019.214 / resource recovery / resource management / reuse / waste management / waste treatment / waste reduction / technology / black soldier fly / Hermetia illucens / life cycle / bioconversion / insect farming / breeding / larvae / pupae / yields / harvesting / lipid content / animal feeding / feedstocks / biomass / biofuels / biodiesel / chitin / residues / organic wastes / fertilizers / nutrients / energy conservation / infrastructure / monitoring / economic impact / economic value / costs / environmental effects / legal aspects / social benefits / public health / business models / markets / case studies / Indonesia / South Africa / Ghana / Canada / ISSN 2478-0510 (Print) ISSN 2478-0529 (Online) ISBN 978-92-9090-893-7 Copyright © 2019, CGIAR Research Program on Water, Land and Ecosystems, International Water Management Institute (IWMI). Fair use: Unless otherwise noted, you are free to copy, duplicate or reproduce, and distribute, display, or transmit any part of this paper or portions thereof without permission, and to make translations, adaptations or other derivative works under the following conditions: ATTRIBUTION. The work must be referenced according to international citation standards, while attribution should in no way suggest endorsement by WLE, IWMI or the author(s). NONCOMMERCIAL. This work may not be used for commercial purposes. SHARE ALIKE. If this work is altered, transformed or built upon, the resulting work must be distributed only under the same or similar license to this one. Front cover photograph: Informal fecal sludge management in Guntur, India (photo: C.S. Sharada Prasad). Series editor (science): Pay Drechsel, IWMI English editor: Robin Leslie Designer: W. D. A. S. Manike Disclaimer The opinions expressed in this paper and any possible errors are the responsibility of the authors. They do not reflect the position of the CGIAR Research Program on Water, Land and Ecosystems or of the institutions and individuals who were involved in the preparation of the report. iii Acknowledgments The authors would like to thank Prof. Ofusu-Budu (University of Ghana), Emmanuel K. Boadu (Animal Research Institute, Accra) and Cecilia Lalander (Swedish University of Agricultural Sciences) for kindly letting them visit their black soldier fly (BSF) waste valorization systems and for sharing useful information on the BSF technology. Bram Dortmans (Swiss Federal Institute of Aquatic Science and Technology [Eawag]) and Pierre-Olivier Maquart (PhD student, University of Stirling, Scotland) are thanked for the information provided on case studies from the following projects: From Organic Waste to Recycling for Development (FORWARD) and Ento- Prise. Logistical support provided by Mr. Randy Adjei (intern, International Water Management Institute [IWMI], Ghana) and Mr. Eric Nartey (Research Officer, Recycling and Reuse, IWMI, Ghana) is gratefully acknowledged. The authors are also grateful for the administrative support provided by KTH Royal Institute of Technology which enabled Gabrielle Joly to conduct her master’s degree project at IWMI, Ghana. Project This report was prepared in the context of the project named “Scaling out the Recovery of Nutrients and Organic Matter from Faecal Sludge for Food Production in Ghana: From Waste to Food (WaFo)”. Collaborators Jekora Ventures Ltd. (JVL) Tema Metropolitan Assembly (TMA) Training Research and Networking for Development (TREND) Donors This research was carried out as part of the CGIAR Research Program on Water, Land and Ecosystems (WLE) and supported by Funders contributing to the CGIAR Trust Fund (https://www.cgiar.org/ funders/). This report is based on research funded in part by: • Bill & Melinda Gates Foundation • CGIAR Research Program on Water, Land and Ecosystems (WLE) iv v CONTENTS List of Figures ..............................................................................................................................................vii List of Tables ..............................................................................................................................................vii Acronyms and Abbreviations ........................................................................................................................viii SUMMARY ........................................................................................................................................................... ix 1 INTRODUCTION .............................................................................................................................................. 1 2 THE BLACK SOLDIER FLY (BSF) ................................................................................................................... 2 3 WASTE PROCESSING BY THE BSF............................................................................................................... 2 3.1 Feedstock Selection .................................................................................................................................. 2 3.1.1 Sourcing ............................................................................................................................................. 2 3.1.2 Waste Preprocessing .......................................................................................................................... 6 3.2 Breeding Conditions .................................................................................................................................. 7 3.2.1 Mating and Oviposition ....................................................................................................................... 9 3.2.2 Egg Harvesting and Hatching ........................................................................................................... 10 3.2.3 Larvae Breeding ................................................................................................................................ 11 3.2.4 Collection of Migrating Prepupae ...................................................................................................... 11 3.2.5 Pupation ........................................................................................................................................... 12 3.2.6 Monitoring of Breeding Performance ................................................................................................. 12 3.3 Waste Treatment ...................................................................................................................................... 13 3.3.1 Operating Conditions ........................................................................................................................ 13 3.3.2 Operational Designs .......................................................................................................................... 14 3.3.3 Monitoring of the Waste Treatment Unit’s Performance ..................................................................... 15 3.4 Product Harvesting and Post-treatment ................................................................................................ 15 3.4.1 Product Yields .................................................................................................................................. 15 3.4.2 Harvesting Techniques ...................................................................................................................... 15 3.4.3 Post-treatments ................................................................................................................................ 17 4 PRODUCTS: PROPERTIES AND APPLICATIONS ....................................................................................... 18 4.1 BSF Larvae ............................................................................................................................................... 18 4.1.1 Properties of BSF Larvae .................................................................................................................. 18 4.1.2 Use of BSF Larvae as Animal Feed ................................................................................................... 20 4.1.3 Production of Biodiesel ..................................................................................................................... 21 4.1.4 Production of Chitin .......................................................................................................................... 22 4.2 Waste Residue ......................................................................................................................................... 22 4.2.1 Properties ......................................................................................................................................... 22 4.2.2 Use as Fertilizer ................................................................................................................................ 22 4.2.3 Safety ............................................................................................................................................... 23 5 ECONOMIC, ENVIRONMENTAL, LEGAL AND SOCIAL DIMENSIONS OF THE BSF TECHNOLOGY ..... 23 5.1 Economic Impact ..................................................................................................................................... 23 5.1.1 Economic Benefits ............................................................................................................................ 23 5.1.2 Costs Associated with the Process ................................................................................................... 27 5.1.3 Overall Economic Performance ......................................................................................................... 29 5.2 Environmental Impact ............................................................................................................................. 30 5.2.1 Environmental Benefits...................................................................................................................... 30 5.2.2 Adverse Environmental Impacts ........................................................................................................ 32 5.2.3 Overall Environmental Performance ................................................................................................... 32 vi 5.3 Legal Aspects .......................................................................................................................................... 33 5.4 Social Aspects ......................................................................................................................................... 34 5.4.1 Public Health .................................................................................................................................... 34 5.4.2 Social Benefits ................................................................................................................................. 34 5.4.3 Social Acceptance ............................................................................................................................ 34 6 IMPLEMENTATION OF BSF TECHNOLOGIES: CASE STUDIES ................................................................ 36 6.1 Overview .................................................................................................................................................. 36 6.2 Case Study 1: FORWARD ....................................................................................................................... 36 6.2.1 Context ............................................................................................................................................. 36 6.2.2 Technology and Process ................................................................................................................... 36 6.2.3 Economic Viability and Impacts ......................................................................................................... 38 6.3 Case Study 2: AgriProtein ....................................................................................................................... 38 6.3.1 Context ............................................................................................................................................. 38 6.3.2 Technology and Process ................................................................................................................... 39 6.3.3 Economic Viability and Impacts ......................................................................................................... 39 6.4 Case Study 3: Ento-Prise ........................................................................................................................ 40 6.4.1 Context ............................................................................................................................................. 40 6.4.2 Technology and Process ................................................................................................................... 40 6.4.3 Economic Viability and Impacts ......................................................................................................... 41 6.5 Case Study 4: Enterra Feed .................................................................................................................... 41 6.5.1 Context ............................................................................................................................................. 41 6.5.2 Technology and Process ................................................................................................................... 41 6.5.3 Economic Viability and Impacts ......................................................................................................... 42 6.6 Lessons Learned from the Case Studies ............................................................................................... 42 7 STATE OF THE RESEARCH AND THE NEED FOR FURTHER STUDIES ................................................... 43 7.1 Overview of the Literature Published on BSF Technology .................................................................... 43 7.2 Research Gaps ........................................................................................................................................ 46 REFERENCES .................................................................................................................................................... 48 vii LIST OF FIGURES FIGURE 1. Lifecycle and characteristics of the BSF 3 FIGURE 2. The conventional waste treatment process using the BSF 4 FIGURE 3. Average composition of a mature BSF larva (% DM) (based on data provided in Table 9) 18 FIGURE 4. Main focus of the studies reviewed (n=90) 43 FIGURE 5. Main aspects examined by the studies reviewed (n=90) 44 FIGURE 6. Types of waste investigated in the studies reviewed (n=47) 44 FIGURE 7. A) Contexts, and B) climate zones examined by the studies reviewed (n=90) 45 FIGURE 8. Publication dates of the studies reviewed (n=89) 45 LIST OF TABLES TABLE 1. The optimal parameter values for feedstock 5 TABLE 2. Optimal breeding conditions and operational designs suggested in the literature 8 TABLE 3. Comparison of breeding performance in two Indonesian facilities 12 TABLE 4. Optimal operating conditions for BSF waste treatment 13 TABLE 5. Operational designs proposed in the literature for the BSF rearing containers 13 TABLE 6. Optimal feeding rate values in terms of biomass production and/or waste reduction for different feedstocks 13 TABLE 7. Bioconversion performance for different feedstocks 16 TABLE 8. BSF products’ properties and applications 17 TABLE 9. General composition of mature BSF larvae 19 TABLE 10. Lipid content of BSF larvae obtained from different feedstocks 19 TABLE 11. Proportions of selected fatty acids in BSF larvae for different feedstocks 19 TABLE 12 Recommended standards for selected chemical contaminants 21 TABLE 13. Biodiesel yields obtained for different feedstocks (1,000 BSF larvae per kg waste) 21 TABLE 14. Fuel properties of the biodiesel produced from BSF larvae’s lipids 22 TABLE 15. Economic benefits and costs associated with a BSF facility 24 TABLE 16. Comparison of infrastructure costs, space requirement and capacity of several BSF waste treatment plants 28 TABLE 17. Labor requirements and costs depending on the waste treatment capacity 28 TABLE 18. Labor costs according to biomass production and total running costs 29 TABLE 19. Environmental benefits and negative impacts associated with a BSF facility 30 TABLE 20. Comparison of biodiesel yields for different feedstocks 32 TABLE 21. Comparison of dried larvae with alternative feedstocks for feed and biodiesel production in terms of environmental impacts 32 TABLE 22. Legislation regarding the use of BSF larvae as animal feed in different parts of the world 33 TABLE 23. Social issues and benefits associated with the BSF technology 35 TABLE 24. Comparison between the BSF technology and other organic waste treatment options 35 TABLE 25. Overview of and comparison between the case studies documented 37 TABLE 26. The FORWARD BSF facility in East Java: Production characteristics 38 TABLE 27. The AgriProtein facility in Cape Town: Production characteristics 39 TABLE 28. The Ento-Prise project in Greater Accra: Production characteristics 40 TABLE 29. The Enterra Feed facility: Production characteristics 41 TABLE 30. Research gaps pertaining to the BSF technology 46 viii ACRONYMS AND ABBREVIATIONS BSF Black Soldier Fly C Carbon CAD Canadian Dollar CH4 Methane CO2 Carbon Dioxide DM Dry Matter DW Dry Weight GHG Greenhouse Gas GWP Global Warming Potential IDR Indonesian Rupiah MSW Municipal Solid Waste N Nitrogen R&D Research and Development SDG Sustainable Development Goal USD United States Dollar WW Wet Weight ix SUMMARY The black soldier fly (BSF) can cope with a wide range of environmental conditions and the adult fly is not a vector of disease. BSF larvae can consume different organic materials, including various organic wastes generated in large volumes within urban areas. By doing so, they reduce waste volume, grow into a protein-rich biomass and leave behind a nutrient- rich residue. The harvested larvae can then be used in formulating feed for monogastric animals such as poultry, fish and pigs. Given their high fat content, they may also be processed into high quality biodiesel. The waste residue could constitute a valuable soil conditioner. Therefore, the BSF- based technology is viewed as one of the most promising technologies for organic waste processing. This report gives an extensive overview of the different aspects of BSF-based technology when used for organic waste processing. It describes the different process components, i.e. (1) waste preprocessing, (2) BSF breeding, (3) waste treatment, (4) product harvesting, and (5) post- treatment of the final products. For each of these key steps, the report describes recommended operating conditions and possible designs. It also reviews the economic, environmental, legal and social aspects of the BSF-based treatment method and presents four business examples on the implementation of the BSF technology in different parts of the world and at different scales. The analysis reveals that the BSF technology could be a promising business option for organic waste valorization. However, it highlights that most research has so far focused on the technical aspects of the technology, resulting in limited data on its economic and environmental performance in support of business start-ups and development. 1 GLOBAL EXPERIENCES ON WASTE PROCESSING WITH BLACK SOLDIER FLY (HERMETIA ILLUCENS): FROM TECHNOLOGY TO BUSINESS 1. INTRODUCTION Waste management constitutes one of the most pressing challenges of the twenty-first century and plays a key role in sustainable development (Scheinberg et al. 2010; Wilson et al. 2015). As waste management is a cross-cutting issue, which impacts many aspects of societies, economies and the environment, addressing this challenge also contributes towards the achievement of more than half of the United Nations Sustainable Development Goals (SDGs) for 2030, including those related to health, climate change, food security, poverty alleviation, responsible consumption and production (Wilson et al. 2015). The global amount of waste generated is increasing rapidly due to population growth, rapid urbanization and economic growth associated with changes in consumption patterns (Karak et al. 2012; Wilson et al. 2015). Most of this growth is occurring in low- and middle-income countries. At the same time, natural resources are being depleted. This calls for a paradigm shift toward a circular economy focusing on ‘closing the loop’, which can be achieved through waste valorization (Lohri et al. 2017). Organic waste recycling is often overlooked because the value of its products is perceived to be lower than that of other waste materials such as plastics, glass or metal (Scheinberg et al. 2010). However, treating organic waste in low-income countries, where it often accounts for the greatest fraction of the municipal solid waste (MSW) generated (typically 50 to 80%), would significantly improve the whole waste management system (Wilson et al. 2015; Lohri et al. 2017; Zurbrügg et al. 2018); in addition, health and environmental hazards related to inappropriate disposal practices would be reduced and nutrient loss would be avoided. Different technologies for the valorization of organic waste have been extensively studied and implemented successfully, the most widespread being composting and anaerobic digestion (Zurbrügg et al. 2018). However, in low- and middle-income countries, the implementation of these solutions has been hindered due to limited financial resources, lack of political support, poor legislative frameworks and legal barriers hampering the marketing of products from the valorization process, as well as the lack of viable business models (Ali 2004; Zurbrügg 2013). Therefore, promoting the value-adding opportunity of organic waste valorization and designing viable business models are crucial to make organic waste recycling more attractive (Rao et al. 2017). The valorization of organic waste using the black soldier fly (BSF) has been promoted as a promising technology, especially in developing countries, as it combines waste reduction and value creation through the bioconversion of low-value organic waste into high-value energy-rich larvae that can be sold. This innovative technology consists of feeding organic waste to BSF larvae to reduce its volume and to recover nutrients. As the larvae feed, they grow into a highly nutritional biomass that can be harvested and further processed into animal feed ingredients or biodiesel, while the waste residue can also be postprocessed into biofertilizer. In addition to improving waste management, BSF technology could contribute to food security. Indeed, BSF larvae-based ingredients constitute a potential alternative to increasingly costly and highly unsustainable feed products (e.g. fishmeal and soybean meal), currently used in the animal production industry (St-Hilaire et al. 2007b; Newton et al. 2008; Salomone et al. 2017; Quilliam et al. 2017). By addressing two major global challenges, BSF waste treatment may constitute “the missing link in designing a circular economy” (van Huis et al. 2013). As research on BSF technology for organic waste valorization is relatively recent, few comprehensive review of this technology is currently available. Therefore, this study provides an extensive overview of the BSF technology; it describes the status of the research, different aspects of this treatment method (technical, economic, environmental, social and so forth), compares it to other options for organic waste valorization, presents case studies on technology implementation and highlights the need for further research. A thorough literature search was carried out in 2017 using the Web of Science and Science Direct databases, Google Scholars, as well as specific libraries, such as the Wiley Online Library, Sage Journals and Springer Link. The search strings used for the literature review included ‘black soldier fly’, ‘Hermetia illucens’ and ‘organic waste’. Additional publications were then identified based on the references used in the articles found through the database search. In total, more than 90 studies were selected and reviewed. In addition, BSF systems in Ghana and Sweden were visited and actors working with BSF technology were interviewed in order to provide concrete case studies of the implementation of a BSF system. This analysis was guided by the following research questions: (1) How does the waste treatment by BSF work? (2) How is it implemented? (3) How does such a system perform technically, economically and environmentally? (4) What are the prospects and constraints associated with the implementation of BSF technology as a business? 2 RESOURCE RECOVERY & REUSE SERIES 16 2.THE BLACK SOLDIER FLY (BSF) The BSF (Hermetia illucens), also known as latrine larva, is a dipterian from the Straiomyidae family (Diener 2010; Caruso et al. 2013; Dortmans et al. 2017; Lohri et al. 2017). It was originally native to the tropical region of Central and South America but has spread to other parts of the world through the transport of goods and human migrations (James 1935; Callan 1974; Leclercq 1997). Today, it is commonly found in tropical and warm temperate regions between latitudes 45°N and 40°S (Diener 2010; Caruso et al. 2013; Dortmans et al. 2017; Lohri et al. 2017). The BSF has a short lifecycle of about six to seven weeks (Tomberlin et al. 2002; Alvarez 2012; Caruso et al. 2013; Dortmans 2015), which according to some authors can be extended by up to four months when unfavorable conditions (food shortage, low temperature, oxygen depletion, drought, etc.) decelerate BSF activity (Furman et al. 1959; Sheppard et al. 1994; Diener 2010; Banks 2014; Tran et al. 2015; Zurbrügg et al. 2018). Five main stages can be distinguished in the BSF’s lifecycle: egg, larval, prepupal, pupal and adult (Banks 2014; Oliveira et al. 2015). The larval and pupal stages constitute most of the lifecycle’s duration, the egg hatching and adult stages being relatively short in comparison. The larval stage is particularly important as it is the only step of the lifecycle in which the BSF feeds. Therefore, larvae need to store enough fat and protein to sustain their biological activities in the latter stages (Diclaro and Kaufman 2009; Caruso et al. 2013; Dortmans et al. 2017). Figure 1 illustrates the different lifecycle stages of the BSF and its main characteristics. 3. WASTE PROCESSING BY THE BSF Essentially, waste treatment by the BSF consists of feeding organic waste to BSF larvae to produce energy-rich larvae and organic fertilizer. Several BSF characteristics make this insect particularly attractive for valorizing organic waste: � The voracious appetite of the BSF larvae for decaying organic matter enables efficient conversion of a wide range of organic waste materials; � The shortness of the BSF’s lifecycle allows its frequent reproduction, therefore ensuring a steady source of larvae to convert the organic waste, as well as a reliable supply of energy-rich larvae that can be used as animal feed; � The resilience of the BSF facilitates its rearing and makes its use in waste treatment less constraining; and � Finally, by crawling naturally out of the waste, the prepupae can be easily harvested. To take advantage of the natural features of the BSF in waste management, its natural lifecycle must be engineered to optimize waste reduction and biomass production. In addition, waste should be treated in a reliable and consistant manner, to stabilize the treatment and production processes and facilitate operations (Zurbrügg et al. 2018). Therefore, this section addresses the technical aspects of the BSF technology, describing how it works and how it can be optimized, based on pilot and/or experimental research literature. Several aspects should be taken into consideration when siting a BSF processing facility. They include (Zurbrügg et al. 2018): � Access to utilities (water, electricity); � Options for wastewater management; � Existence of an environmental and physical barrier to minimize nuisances (visual or olfactory) to the surrounding environment or intrusion into the premises; � Secured supplies of quality raw materials; and � Processing facilities offering suitable growth conditions for the BSF. The BSF treatment process can be typically disaggregated into: (1) waste preprocessing, (2) BSF breeding, (3) waste treatment, (4) product harvesting, and (5) post-treatment of the products (Dortmans et al. 2017; Zurbrügg et al. 2018). The next section describes the different components of a BSF system and discusses the optimal operating conditions and designs proposed in the literature for each of them (Figure 2). 3.1 Feedstock Selection 3.1.1 Sourcing The waste received at the treatment facility should be controlled at all times; inorganic and hazardous waste must be removed from the waste stream. Waste nutritive composition is known to play a critical role in BSF activity and growth performance. But many uncertainties remain concerning how the feedstock type and quality could affect the outcome of BSF conversion of waste. Hence, selecting a suitable BSF feedstock has become a complex process, which in many practical cases, is based on waste availability and cost. 3 GLOBAL EXPERIENCES ON WASTE PROCESSING WITH BLACK SOLDIER FLY (HERMETIA ILLUCENS): FROM TECHNOLOGY TO BUSINESS F IG U R E 1 . L IF E C Y C LE A N D C H A R A C T E R IS T IC S O F T H E B S F. S ou rc es : B as ed o n da ta p ro vi de d by B oo th a nd S he pp ar d 19 84 ; S he pp ar d et a l. 19 94 ; T om be rli n an d S he pp ar d 20 02 ; D ic la ro a nd K au fm an 2 00 9; D ie ne r 20 10 ; C ar us o et a l. 20 13 ; B an ks 2 01 4; C ic ko vá e t al . 2 01 5; D or tm an s et a l. 20 17 ; L oh ri et a l. 20 17 ; Z ur br üg g et a l. 20 18 . M at in g ~ 2 d ay s af te r em er ge nc e O vi po si tio n ~ 4 d ay s af te r em er ge nc e Li fe tim e ~ 1 w ee k Le ng th ~ 1 m ilim et er (m m ) W ei gh t ~ 2 8 m ic ro gr am s (u g) C ol or : p al e ye llo w o r cr ea m y w hi te 30 0- 1, 00 0 eg gs pe r fe m al e fly A bo ut 2 -3 w ee ks to em er ge a s a fly La rv ae g ro w fr om 1 m m to ab ou t 2 50 m m in le ng th , 5 0 m m in w id th a nd 2 00 m g in w ei gh t, th ro ug h 6 in st ar s (i. e. la rv al s ta ge s) Lo ok s lik e a w as p; o ne p ai r of w in gs , no s tin ge r, 6 le gs a nd 2 lo ng an te nn ae Le ng th : 1 0- 15 m m W ei gh t: 10 ti m es h ea vi er th an th e ho us e fly A s pu pa , t he B S F st op s m ov in g an d st ar ts tr an sf or m in g in to fl y M ou th pa rt tr an sf or m ed in to a ho ok -s ha pe d st ru ct ur e C ol or : d ar k br ow n to c ha rc oa l g re y P re pu pa e cr aw l n at ur al ly o ut o f th ei r fe ed s ou rc e in s ea rc h of a pu pa tio n si te 6 - 7 w ee ks E G G LA R VA A D U LT P U P A P R E P U P A E E gg s ha tc h af te r~ 3 -4 d ay s La rv ae re ac h th e pr ep up al s ta ge in ~ 2 -3 w ee ks u nd er o pt im al c on di tio ns 4 RESOURCE RECOVERY & REUSE SERIES 16 FIGURE 2. THE CONVENTIONAL WASTE TREATMENT PROCESS USING THE BSF. BSF breeding Biogas Animal feed Larvae oil Young larvae Biodiesel Composting Waste treatment Waste preprocessing (sorting, shredding, drying/watering, blending, etc.) Larvae refinement (boiling, drying, freezing, oil extraction, etc.) Product harvesting Biogas production Biodiesel production Waste residue postprocessing Fertilizer and soil conditioner Digestate Photo: Gabrielle Joly 5 GLOBAL EXPERIENCES ON WASTE PROCESSING WITH BLACK SOLDIER FLY (HERMETIA ILLUCENS): FROM TECHNOLOGY TO BUSINESS In theory, BSF larvae can process a wide range of organic materials due to their powerful mouthparts, the unique composition of their gut microbiota, including bacterial species not found in the microbiota of other insects, as well as the high activity of their digestive enzymes, such as amylase, lipase and protease in their salivary glands and gut (Jeon et al. 2011; Kim et al. 2011; Caruso et al. 2013; Banks 2014). According to the literature, the feedstocks used in the BSF treatment include: � Mixed municipal organic waste (Diener et al. 2011); � Food, restaurant and market waste, such as fruit and vegetable waste (Nguyen et al. 2015; Parra Paz et al. 2015; Saragi and Bagastyo 2015; Cheng and Lo 2016; Leong et al. 2016; Lalander et al. 2019); � Animal manure, such as poultry, cow and pig manure (Sheppard et al. 1994; Yu et al. 2011; Myers et al. 2008; Li et al. 2011b; Newton et al. 2005; Nguyen et al. 2015; Lalander et al. 2019); � Human feces and fecal sludge (Lalander et al. 2013; Banks et al. 2014; Joly 2018; Lalander et al. 2019); and � Agroindustrial waste, such as: ° food processing waste (Lardé 1989; Caruso et al. 2013; Dortmans et al. 2017; Mohd-Noor et al. 2017), ° spent grains (Dortmans et al. 2017), ° slaughterhouse waste (Dortmans et al. 2017), and ° fish waste (Nguyen et al. 2015; Saragi and Bagastyo 2015; St-Hilaire et al. 2007b). Despite the flexibility of BSF larvae and feedstock, some authors have highlighted key parameters influencing the ability of BSF larvae to process a material (see Table 1). According to Lalander et al. (2019), feedstock could affect three main BSF- related parameters, such as larval development time, the final prepupal weight and the waste-to-biomass conversion rate. TABLE 1. THE OPTIMAL PARAMETER VALUES FOR FEEDSTOCK. Parameters Optimal values Suggested References pertaining to the preprocessing feedstock methods for optimization Nutrient content Feedstock rich in protein and carbohydrates Mixing different St-Hilaire et al. 2007a; (e.g. 21% protein and 21% carbohydrate); waste types Gobbi et al. 2013; Suitable C/N ratio: 10-40 (optimal nutrient Lalander et al. 2015; balance not established). High contents of Cammack and Tomberlin volatile solids are preferable 2017; Dortmans et al. 2017; Lohri et al. 2017; Rehman et al. 2017a, 2017b; Lalander et al. 2019 Fiber content Not too high (no optimal value established) Prefermentation Zheng et al. 2012a; Caruso et al. 2013; Lohri et al. 2017; Mohd-Noor et al. 2017; Rehman et al. 2017a Moisture content 60-90% (wet weight (WT)) Dewatering, water addition Cammack and Tomberlin and/or mixing different 2017; Cheng et al. 2017; waste types Dortmans et al. 2017; Lohri et al. 2017 Particle size 1-2 cm Shredding Dortmans et al. 2017; Lohri et al. 2017 pH 5-8 (suitable values) Mixing different Caruso et al. 2013; waste types Dortmans 2015; Lalander et al. 2015; Rehman et al. 2017a, 2017b Structure Sufficient structure to allow the larvae to Addition of matrix material, Barry 2004; move through the feedstock, consume it such as pine shavings or Perednia 2016 and breathe crushed charcoal 6 RESOURCE RECOVERY & REUSE SERIES 16 Overall, according to Lalander et al. (2019), protein and total volatile solid contents remain the most critical waste-related factors. Therefore, feedstocks with higher concentrations of these two components should be preferred, such as abattoir waste, food waste and human feces. But pure fruit and vegetable wastes and sewage sludges may not be suitable, unless they are mixed with other acceptable materials. Typically, larval development is favored if the feedstock is rich in protein and easily available carbohydrates (Dortmans et al. 2017; Lalander et al. 2019). When Cammack and Tomberlin (2017) used a balanced diet, i.e. containing 21% protein and 21% carbohydrate, larval development was optimal. In addition, a suitable C/N ratio is also critical for the biological activity of BSF larvae. Feedstocks with a C/N ratio ranging from 10 to 40 have been reported to be efficiently converted by BSF larvae (Saragi and Bagastyo 2015; Lalander et al. 2015; Rehman et al. 2017a, 2017b). But Rehman et al. (2017b) observed, when comparing different mixtures of dairy manure and soybean curd residue, with C/N ratios ranging from 16.2 to 18.4, that BSF treatment performed best, in terms of fiber reduction and biomass production, for the substrate with a C/N ratio of 16.2. Similarly, Rehman et al. (2017a) recommended a C/N ratio of 14.2 for co- digestion of dairy manure and chicken manure by the BSF. On the other hand, high fat content could be detrimental to BSF growth (Lalander et al. 2019). 3.1.2 Waste Preprocessing Mixing different types of waste is a preprocessing method that helps to optimize the nutrient balance of the feedstock and could enhance waste reduction, larvae growth and the nutritional content of the larvae (St-Hilaire et al. 2007a; Gobbi et al. 2013; Cammack and Tomberlin 2017; Rehman et al. 2017a, 2017b; Lalander et al. 2019). However, the effects of co-digesting different feedstocks have not yet been fully documented. Other process parameters highlighted in the literature (see Table 1) were moisture content, particle size and nutrient content (Cheng et al. 2017; Dortmans et al. 2017; Lohri et al. 2017). Lignocellulosic waste, characterized by a high fiber content, such as vegetable waste or dairy manure, has been reported to be harder to convert by the BSF (Zheng et al. 2012a; Lohri et al. 2017; Rehman et al. 2017a, 2017b). Allowing such materials to ferment, so that complex organic molecules are broken down into simpler elements more easily assimilated by BSF larvae, could be a beneficial pretreatment. In the case of waste from oil palm and coconut milk extraction industries, fermentation for typically up to a few weeks is sufficient (Caruso et al. 2013; Mohd-Noor et al. 2017). During a longer fermentation period, too many microorganisms could grow in the substrate and compete with BSF larvae for common nutrients (Mohd-Noor et al. 2017; OVRSol 2010). The BSF feedstock should be sufficiently moist to allow the larvae to ingest the material. Materials that are too dry cannot be processed by BSF larvae. But if the food source is too wet, the larvae will crawl out of the waste to search for a drier location or their separation from the residue at the end of the treatment will be more difficult (Alvarez 2012; Caruso et al. 2013; Cheng et al. 2017). Alvarez (2012) pointed out that the larvae’s development rate can be controlled via the feedstock’s moisture content. Latest studies suggested a range of suitable moisture content of approximately 60 to 90% of WW (Cammack and Tomberlin 2017; Cheng et al. 2017; Dortmans et al. 2017; Lohri et al. 2017), while Fatchurochim et al. (1989) reported that moisture contents ranging from 40 to 70% of WW were optimal for BSF larvae development. Therefore, wastes that are too moist, such as fecal sludge or fruit and vegetable waste, may require dewatering before being processed, while water should be added to drier materials such as chicken manure. The ideal solution may be to mix materials of different water content to easily achieve a suitable moisture level (Furman et al. 1959; Dortmans et al. 2017; Lohri et al. 2017). Regarding particle size, the literature suggests its reduction before processing, for example by using a shredder or a hammer mill (Dortmans et al. 2017; Lohri et al. 2017). Feedstocks with particle size smaller than 1 to 2 centimeters (cm) in diameter allow the larvae, which have no chewing mouthparts, to access food more easily (Dortmans et al. 2017). Another important parameter, though rarely addressed in the literature, is the structure of the feedstock. Barry (2004) and Perednia (2016) highlighted the importance of ensuring that the feedstock has enough structure to allow the larvae to move through the material, consume it and obtain an adequate supply of oxygen. Perednia (2016) recommended adding matrix materials, such as pine shavings and crushed charcoal, to improve the ability of BSF larvae to burrow and move through the waste pile. The literature does not identify pH as a key factor that influences the ability of BSF larvae to process a feedstock, and none of the studies reviewed proposes an optimal range for pH value. Feedstocks with pH ranging from 5 to 8 have been processed successfully in experiments involving BSF larvae (Caruso et al. 2013; Dortmans 2015; Lalander et al. 2015; Rehman et al. 2017a). However, Rehman et al. (2017b) reported that buffer capacity is crucial for the biological activity of BSF larvae. Comparing BSF treatment performance for different mixtures of dairy manure and soybean curd residue, whose pH ranged from 5.1 to 7.9, Rehman et al. (2017b) observed the greatest fiber reduction and biomass production for the substrate exhibiting a pH of 6.7, while a pH of 7.8 was recommended by Rehman et al. (2017a) for co-digesting dairy and chicken manures with the BSF. 7 GLOBAL EXPERIENCES ON WASTE PROCESSING WITH BLACK SOLDIER FLY (HERMETIA ILLUCENS): FROM TECHNOLOGY TO BUSINESS 3.2 Breeding Conditions Two main types of BSF waste treatment systems can be distinguished, namely systems relying on natural colonization by the BSF and artificial breeding systems (Cicková et al. 2015; Lohri et al. 2017). Systems relying on natural colonization are mainly used at the household level, typically for backyard applications (Lohri et al. 2017). They are unsuitable in the context of a controlled waste treatment facility (Cicková et al. 2015; Lohri et al. 2017). Therefore, recent literature mostly focuses on artificial breeding systems, which include a breeding unit where the BSF are bred in captivity to produce young larvae (Diener et al. 2015a; Lohri et al. 2017; Dortmans et al. 2017). Such systems are more expensive and complex than those depending on natural BSF populations but allow for a controlled operation and stable production (Cicková et al. 2015; Lohri et al. 2017). Therefore, the present review focuses on the latter system. Diener et al. (2015a) and Lohri et al. (2017) highlighted the key role played by the breeding unit in a BSF waste treatment facility as the production of enough young larvae is crucial to ensure the running of the waste treatment process. They also pointed out that maintaining a large enough and healthy BSF colony is the most delicate step of the process (Diener et al. 2015a; Lohri et al. 2017). To maintain the colony, a fraction of the young larvae is typically kept in the breeding unit or, alternatively, prepupae harvested from the waste treatment unit are reintroduced in the breeding unit to pupate into flies (Nature 2016; Dortmans et al. 2017). Flies are then used to produce eggs, which are incubated until they hatch into larvae. The optimal operating conditions and designs to rear the BSF at each stage of its lifecycle are discussed in the following sections and summarized in Table 2. Photo: Gabrielle Joly 8 RESOURCE RECOVERY & REUSE SERIES 16 TA B LE 2 . O P T IM A L B R E E D IN G C O N D IT IO N S A N D O P E R AT IO N A L D E S IG N S S U G G E S T E D IN T H E L IT E R AT U R E . O p tim al o p er at in g c o nd iti o ns S ug g es te d R ef er en ce s Li fe cy cl e st ag e Te m p er at ur e H um id ity Li g ht D ie t O th er o p er at io na l d es ig ns E g g s C on st an t > 6 0% D ar k N on e - E gg s in cu ba te d in a S he pp ar d et a l. 20 02 ; te m pe ra tu re en vi ro nm en t, co ve re d co nt ai ne r an d Zh an g et a l. 20 10 ; (e .g . ~ 2 7 °C ) w ith 0 -5 0% pl ac ed a bo ve a fe ed D ie ne r et a l. 20 11 ; da ily li gh t so ur ce fo r ne on at e la rv ae A lv ar ez 2 01 2; H ol m es ex po su re et a l. 20 12 , 2 01 7; M ut af el a 20 15 Ju ve ni le la rv ae C on st an t R el at iv el y co ns ta nt D ar k S pe ci al d ie t ( e. g. - Ju ve ni le la rv ae k ep t S he pp ar d et a l. 20 02 ; ( 4- 6 d ay s o ld ) te m pe ra tu re in th e hu m id ity le ve l en vi ro nm en t w he at b ra n, r ab bi t fo r 4- 6 da ys a fte r D ie ne r et a l. 20 11 ; 24 -3 3 °C r an ge or c hi ck en fe ed ) ha tc hi ng in th e C ar us o et a l. 20 13 ; w ith e no ug h in cu ba tio n co nt ai ne r D or tm an s et a l. 20 17 ; st ru ct ur e Ya ng 2 01 7 La rv ae 24 -3 3 °C Th e lit er at ur e fo cu se s D ar k W el l-d efi ne d di et o r - La rv ae fe d w ith a S he pp ar d et a l. 20 02 ; on th e m oi st ur e en vi ro nm en t or ga ni c w as te to b e w el l-d efi ne d fe ed u nt il To m be rli n et a l. 20 02 ; co nt en t o f t he tr ea te d th ey re ac h th e pr ep up al A lv ar ez 2 01 2; fe ed st oc k st ag e or a re u se d fo r C ar us o et a l. 20 13 ; w as te tr ea tm en t H ar nd en a nd T om be rli n 20 16 ; D or tm an s et a l. 20 17 P re p up ae /p up ae In th e sa m e ra ng e 60 -7 0% D ar k en vi ro nm en t, N on e P up at io n m ed iu m (e .g . P re pu pa e co lle ct ed in a N ew to n et a l. 20 05 ; as th e la rv al s ta ge w ith 0 -5 0% d ai ly w oo d ch ip s, c oc o pe at , co nt ai ne r fil le d w ith D ie ne r et a l. 20 11 ; (2 4- 33 ° C ) lig ht e xp os ur e co m po st ) e xh ib iti ng a a dr y an d w at er - A lv ar ez 2 01 2; m oi st ur e le ve l o f ab so rb in g m at er ia l, C ar us o et a l. 20 13 ; 50 -8 5% a nd a d ep th co nn ec te d to th e fe ed in g B an ks 2 01 4; of 1 5- 20 c m co nt ai ne r th ro ug h a pi pe M ut af el a 20 15 ; L in 2 01 6; (in cl in at io n: 2 8- 45 °) o r N ak am ur a et a l. 20 16 ; fe ed in g co nt ai ne r pl ac ed D or tm an s et a l. 20 17 ; di re ct ly in th e co lle ct io n H ol m es e t a l. 20 17 co nt ai ne r A d ul ts 25 -3 2 °C > 6 0% M or ni ng s un lig ht N on e, b ut p ro vi di ng S uf fic ie nt s pa ce to G re en ho us e or n et te d ca ge B oo th a nd S he pp ar d w at er w ith s ug ar is m at e in fl ig ht . H ig h fly (s iz e ra ng in g fro m 0 .2 7 x 19 84 ; H ol m es e t a l. 20 12 ; re co m m en de d de ns ity (5 ,0 00 fl ie s m -3 ). 0. 27 x 0 .2 7 m et er s (m ) t o 3 S he pp ar d et a l. 20 02 ; P la nt to fa vo r le kk in g. x 3 x 6 m ). O vi po si tio n To m be rli n an d S he pp ar d m ed ia w ith c av iti es m ad e of 20 02 ; Z ha ng e t a l. 20 10 ; ca rd bo ar d or w oo d an d D ie ne r et a l. 20 11 ; pl ac ed o n or c lo se to A lv ar ez 2 01 2; C ar us o or ga ni c m at te r w ith a et a l. 20 13 ; M ut af el a su ffi ci en tly s tr on g sm el l 20 15 ; N ak am ur a et a l. 20 16 ; D or tm an s et a l. 20 17 9 GLOBAL EXPERIENCES ON WASTE PROCESSING WITH BLACK SOLDIER FLY (HERMETIA ILLUCENS): FROM TECHNOLOGY TO BUSINESS 3.2.1 Mating and Oviposition About two days after emerging, BSF mate whilst in fly through lekking, a mating behavior characterized by the clustering of males in a given location and attraction of females through competitive display (Tomberlin and Sheppard 2001; Diclaro and Kaufman 2009; Furman et al. 1959; Diener 2010; Caruso et al. 2013). Then, about two days after mating, females extend their ovipositor to lay their eggs in the form of a single clutch (Tomberlin and Sheppard 2002; Cicková et al. 2015). In general, the number of eggs laid by each BSF female ranges from 320 to 1,000 (Tomberlin et al. 2002; Diclaro and Kaufman 2009; Caruso et al. 2013; Banks 2014; Dortmans et al. 2017). The adult fly dies once its fat reserve is depleted (Alvarez 2012; Myers et al. 2008), i.e., typically a few hours after oviposition for females (Tomberlin et al. 2002). Oviposition usually takes place close to decaying organic matter so that, immediately after hatching, the larvae have access to a feed source. In addition, for oviposition, female flies seem to choose media that have small cavities into which they can lay their egg packages to ensure their protection from predators and prevent their dehydration by direct sunlight (Caruso et al. 2013; Dortmans et al. 2017). The eggs hatch after about three to four days (Sheppard et al. 2002; Diclaro and Kaufman 2009; Dortmans et al. 2017). Operating Conditions Three main environmental parameters influence mating and oviposition of the BSF, namely temperature, light and humidity. Temperature plays an important role in ensuring mating and oviposition (Tomberlin and Sheppard 2002; Alvarez 2012). BSF females require temperatures greater than 26 °C to lay eggs (Tomberlin and Sheppard 2002). Booth and Sheppard (1984) in particular observed that 99.6% of oviposition took place when temperatures were between 27.5 and 37.5 °C. Dortmans et al. (2017) recommended an optimal range of 25 to 32 °C to rear adult BSF. Tomberlin and Sheppard (2002) established that high light intensity promotes mating. Specifically, they observed that, under sunlight, most mating (75%) occurred when light intensity was greater than 200 µmol m-2 s-1 and a minimum light intensity of 63 µmol m-2 s-1 was required for mating to take place. But Zhang et al. (2010) recorded that over 110 µmol m-2 s-1, the mating activity of flies exposed to sunlight decreased. This difference could be because, in addition to light intensity, the time of day has been shown to influence the mating rate. Mating occurs generally early in the day (with a peak at 10:00), while oviposition generally takes place later in the day (Tomberlin and Sheppard 2002; Zhang et al. 2010). Some authors report that light source and wavelength range also influence mating activity. In particular, mating is stimulated by sunlight, as well as artificial light with wavelengths from 450 to 700 nanometers (nm). However, greater fertility and hatchability may be achieved with sunlight than with artificial light (Tomberlin and Sheppard 2002; Zhang et al. 2010; Nakamura et al. 2016). On the other hand, light does not seem to influence oviposition. Under unsuitable light conditions, BSF females could lay eggs without having mated. However, these eggs are infertile (Tomberlin and Sheppard 2002). In addition, Zhang et al. (2010) recorded similar numbers of eggs laid under sunlight and light from a quartz-iodine lamp with a 350 to 2,500 nm spectrum. Moreover, the light source used to stimulate mating does not seem to affect larval development and pupation later because Zhang et al. (2010) observed similar larval and pupal development times with the quartz-iodine lamp and natural sunlight. Furthermore, it appears that humid conditions could prolong the lifespan of BSF adults and thus promote their reproduction (Caruso et al. 2013). Typically, adults reared under a 70% relative humidity level live two to three day longer than those in drier environments (Holmes et al. 2012). Similarly, providing water for the flies to drink is also reported to be beneficial because flies provided with water live one to two days longer than those without water (Tomberlin et al. 2002). Adding sugar to the water is also reported to promote reproduction (Caruso et al. 2013; Nakamura et al. 2016). Humidity may also have an influence on oviposition as Tomberlin and Sheppard (2002) reported that 80% of eggs are laid when humidity exceeds 60%. However, Sheppard et al. (2002) observed mating and oviposition in a range of humidity conditions as wide as 30 to 90%. To choose an oviposition site, females use the tip of their abdomen, which contains the ovipositor, i.e. the body part used to lay eggs, which is covered by sensors. Therefore, when searching for an oviposition site, BSF females drag the tip of their abdomens along the surface of a substrate to probe its characteristics. This process gives females information about the presence of BSF eggs, larvae, competitors or pathogens in the substrate, as well as the availability of nutrients (Tomberlin 2017). Zheng et al. (2013a) observed that ovipositing females were attracted by substrates containing bacteria isolated from BSF eggs, while they were generally repelled by the presence of bacteria isolated from competing insects, such as blow flies or beetle larvae. Females, in search of an oviposition site, may also be attracted by the effluent from decomposing waste and leave chemical markers to attract other females to the laying site (Sheppard et al. 2002; Alvarez 2012). Operational Designs For the mating unit, different designs can be found in the literature. The most widespread are greenhouses (Sheppard et al. 2002; Diener et al. 2011; Alvarez 2012; Caruso et al. 2013) and netted cages (Zhang et al. 2010; Mutafela 2015; Popoff and Maquart 2016a; Dortmans 10 RESOURCE RECOVERY & REUSE SERIES 16 et al. 2017; C. Lalander, pers. comm., June 16, 2017). Depending on the availability of sunlight, mating units are either equipped with lamps (Mutafela 2015) or exposed to sunlight (Diener et al. 2009b; Alvarez 2012; Popoff and Maquart 2016a). However, in tropical regions, it is recommended not to place the cage in direct sunlight to avoid the rapid dehydration of the flies (B. Dortmans, pers. comm., September 28, 2017). The mating unit has to be big enough to allow the flies to mate in flight (Barry 2004; Alvarez 2012; Caruso et al. 2013; Banks 2014). Sizes reported in the literature for the mating cage typically range from 0.7 x 0.7 x 1.4 m to 3 x 3 x 6 m (Sheppard et al. 2002; Tomberlin and Sheppard 2002; Zhang et al. 2010; Diener et al. 2011; Charlton et al. 2015; Mutafela 2015; Dortmans et al. 2017), with fly density ranging from about 100 up to 5,200 flies m-3 (Tomberlin and Sheppard 2002; Zhang et al. 2010; Charlton et al. 2015). Caruso et al. (2013) recommended that the ceiling of the mating cage should be higher than 1.5 m, as Tomberlin and Sheppard (2001) observed that in nature BSF couples can fly up to 1.5 m above the ground while mating. However, Nakamura et al. (2016) showed that fertilized eggs could be obtained in a cage as small as 27 x 27 x 27 cm with a high fly density (5,000 flies m-3). In addition, several authors, based on the observations of Tomberlin and Sheppard (2001), suggested placing plants, either natural or artificial, in the mating unit to favor lekking and thus mating (Caruso et al. 2013; Mutafela 2015). In addition, plants provide sites for flies to rest on (Cicková et al. 2015). In several set-ups, the flies are also kept hydrated by regularly spraying them with water and/or by placing a wet cloth on a container filled with water or wet cotton so that the flies do not drown in it (Alvarez 2012; Caruso et al. 2013; Popoff and Maquart 2016a; Dortmans et al. 2017). Alvarez (2012) and Dortmans et al. (2017) highlighted the importance of providing a suitable medium for oviposition so that all females lay their eggs in the same location, thus facilitating egg harvesting. The oviposition medium needs to fulfil several conditions. Booth and Sheppard (1984) observed that BSF females prefer to lay their eggs on dry media. In the literature, several designs with different materials and shapes are proposed for the oviposition medium. Cardboard and wood are the most widely used materials (Booth and Sheppard 1984; Diener et al. 2011; Tomberlin et al. 2002; Mutafela 2015). Other materials include paper towels (Nakamura et al. 2016) or dry banana tree leaves (Caruso et al. 2013; Popoff and Maquart 2016a). Examples of designs include strips of cardboard or wood tied together so that they are separated by small gaps (Mutafela 2015; Dortmans et al. 2017), cardboard honeycomb (Dortmans et al. 2017; C. Lalander, pers. comm., June 16, 2017), cardboard rolls (Zhang et al. 2010), blocks made of three layers of corrugated cardboard glued together (Sheppard et al. 2002; Tomberlin et al. 2002) and strips of corrugated cardboard wrapped around skewers and tucked into rings of bamboo (Diener et al. 2011). Dortmans et al. (2017) also suggested using ‘bioballs’, normally designed as biofilters for aquariums of fish ponds. In addition, the oviposition medium should preferably be placed on or close to organic matter with a sufficiently strong smell to attract females to lay their eggs into the oviposition medium. In most experimental settings, decomposing organic waste is used (Dierner et al. 2011; Mutafela 2015). Dortmans et al. (2017) suggested that dead flies and eggs themselves could be used as an attractant and thus recommended mixing the fresh attractant substrate with the residue from an old container used to collect eggs, which is also suggested by Tomberlin (2017). On the other hand, Furman et al. (1959) and Tomberlin (2017) suggested that females are more attracted to substrates already containing BSF larvae and thus recommended placing larvae in the attractant container, but other authors did not back this hypothesis (Kemppineen 1998; Tomberlin and Sheppard 2002). Mutafela (2015) also pointed out that the attractant should not be too wet to prevent females from drowning in it. Finally, Dortmans (2015) reported that female flies prefer shaded sites, perceived as safer, to lay their eggs. Therefore, Dortmans et al. (2017) proposed placing a shading basket above the oviposition medium. 3.2.2 Egg Harvesting and Hatching Operating Conditions According to Alvarez (2012), eggs are particularly vulnerable to changes in environmental parameters. Therefore, eggs should preferably be held at a constant temperature until hatching. For example, Sheppard et al. (2002) reported that keeping eggs at 27°C yields satisfactory results as they observed egg-hatching rates exceeding 80% under sufficient humidity. Regarding humidity, Holmes et al. (2012) reported that egg-hatching success increases as the relative humidity level increases. More precisely, they reported that humidity levels of more than 60% result in optimal egg- hatching rates and prevent desiccation of the eggs (Holmes et al. 2012). Operational Designs In all the designs proposed in the literature reviewed, oviposition media containing eggs are harvested and transferred to another location for hatching. The oviposition media are usually placed above (Dortmans et al. 2017) or directly on a high-quality feed source adapted to the newly hatched larvae (Zhang et al. 2010; Diener et al. 2011; Mutafela 2015). In a few cases, eggs clusters are removed manually from the oviposition medium and placed directly in a hatching container. However, this method is labor- intensive and therefore, not recommended for large-scale operations (Caruso et al. 2013; Popoff and Maquart 2016a). The hatching container is usually covered, for example, with a fine mesh to protect the eggs and the juvenile larvae from predators (Sheppard et al. 2002; Zhang et al. 2010; Diener et al. 2011; Mutafela 2015; Nakamura et al. 2016; Popoff and Maquart 2016a). In the study led by Diener et al. 11 GLOBAL EXPERIENCES ON WASTE PROCESSING WITH BLACK SOLDIER FLY (HERMETIA ILLUCENS): FROM TECHNOLOGY TO BUSINESS (2011), the hatching containers were placed in a dark and warm environment; however this is not in line with findings by Holmes et al. (2017) who found that eggs hatched faster when they are exposed to 12 hours of light day-1 than if they are exposed to 0 or 8 hours of light day-1 (2012). 3.2.3 Larvae Breeding Newly hatched larvae are particularly sensitive to changes in environmental conditions and food competition. Therefore, feeding them with a special diet and keeping them in a controlled and protected environment for a few days, typically four to six days, increases their survival rate (Diener et al. 2011; Popoff and Maquart 2016a; Dortmans et al. 2017; C. Lalander, pers. comm., June 16, 2017). Various diets for the young larvae are suggested in the literature. They include a mixture of corn meal, wheat bran and water (Mutafela 2015 adapted from Sheppard et al. 2002), rabbit feed mixed with water (Diener et al. 2011) or chicken feed for starter chicks mixed with water (Dortmans et al. 2017). Yang (2017) indicated that the lack of structure of the feed source is particularly problematic for juvenile larvae which are not strong enough to create pore space to breathe. They advised against using diets characterized by too-fine particle size, such as alfalfa and corn meal, or that are too sticky, like cooked sorghum. Instead, it is recommended to add matrix materials that have low density but high rigidity, such as wood branches, wheat bran, rice bran or wood- shaving dust to the diet in order to create a loose texture that will allow the juveniles to breathe (C. Lalander, pers. comm., June 16, 2017; Yang 2017). In the literature, two main scenarios are encountered regarding the fate of four-to-six day-old larvae. In the first, a fraction of the young larvae is kept in the breeding unit to reach the adult stage and hence produce new larvae (Popoff and Maquart 2016a; Dortmans et al. 2017). Another option is using all the larvae in the treatment process but later on reintroducing a fraction of the prepupae into the breeding unit so that they pupate and emerge as adult flies (Newton et al. 2005; Caruso et al. 2013; Nature 2016). In the first scenario, the larvae kept in the breeding unit are placed into a container filled with a well-defined feed until they reach the prepupal stage (Dortmans et al. 2017; Zurbrügg et al. 2018). An example of artificial diet, recommended by Sheppard et al. (2002) and Tomberlin et al. (2002) for breeding BSF larvae, is the Gainseville diet, which consists of 50% wheat bran, 30% alfalfa meal and 20% corn meal. In the second scenario, there is a risk that the colony could collapse in the case of system failure, for instance if the waste is contaminated. On the other hand, feeding the larvae with a controlled diet, despite being more expensive, reduces the risk of failure. In addition, as diet has been shown to influence both the physiological and morphological characteristics of the adult fly, and especially female fertility, controlling the larval diet maintains a healthy and productive colony (Gobbi et al. 2013). Some authors reported that larvae are photophobic and should therefore be held in a dark environment (Caruso et al. 2013; B. Dortmans, pers. comm., September 28, 2017). However, a recent study by Holmes et al. (2017) established that if the larvae are kept in the dark, they require about one- third more time to develop into adults. Ideal temperature for larval development ranges between 24 and 33 °C (Alvarez 2012; Dortmans et al. 2017). If the temperature is too high in the waste, the larvae will crawl out of the food source to find a cooler location. On the other hand, larvae subjected to low temperatures will slow down their metabolisms to survive, which means that they will feed less and thus grow at a slower pace (Dortmans et al. 2017). Because larvae generate heat by moving into the food source as they feed, Alvarez (2012) suggested that larvae better withstand temperatures which are lower compared to the optimal range than higher temperatures. The temperature at which the larvae are reared, besides influencing the larval growth rate, also affects the size and lifespan of the future adult fly. Tomberlin et al. (2009) reported that, above 27 °C, as the temperature increases smaller adults with shorter lifespan are observed. In addition, between 27 °C and 30 °C, they observed trade-offs between the larval development time (the larvae take less time to reach maturity as the temperature increases), the adult lifespan (adults live for a shorter time when the development temperature is higher) and the prepupal weight (prepupae are heavier at lower temperatures). On the other hand, Harnden and Tomberlin (2016) observed that larvae reared at 24.9 °C reached maturity faster but their final weight was on average 30% lower than larvae reared at 27.6 °C and 32.2 °C, which suggests that below ~ 27 °C, the trend is reversed. Furthermore, Tomberlin et al. (2009) observed that only 0.1% of larvae reared at 36°C reached the adult stage, which suggests that sustained high temperatures are not suitable for breeding larvae. Based on these studies, the upper temperature limit seems to lie between 33 and 36 °C (Tomberlin et al. 2009; Harnden and Tomberlin 2016). 3.2.4 Collection of Migrating Prepupae When the larvae attain the prepupal stage, they have reached their maximum size. They stop feeding and empty their digestive tracts (Diener 2010; Banks 2014; Dortmans et al. 2017). Using their hook-shaped mouthparts, they emerge from the food source and reach a dry, dark and protected location to pupate into adult flies (Diener 2010; B. Dortmans, pers. comm., September 28, 2017). The average time of migration is not reported in the literature. According to Alvarez (2012), this depends on the larvae’s ability to find a suitable pupation site. Alvarez (2012) also suggested that larvae may leave a chemical trail during their search for a pupation site for other larvae to find, resulting in a migration path. To contain the prepupae that crawl out of the feed source, the feeding container must be connected to another 12 RESOURCE RECOVERY & REUSE SERIES 16 container filled with a dry and water-absorbing material (Dortmans et al. 2017). This latter container can either be used directly as a pupation container (Diener et al. 2011; Mutafela 2015) or as a transfer container (Dortmans et al. 2017). Regarding the connection between the two containers, ramps connected to a pipe leading to the pupation container can be used (Diener et al. 2011; Mutafela 2015). For the exit ramp, inclinations ranging from 28° to 45° have been successfully tested (Newton et al. 2005; Diener et al. 2011; Banks 2014; Mutafela 2015). However, Banks (2014) reported that BSF prepupae can climb up vertical surfaces if the moisture is sufficient to maintain surface tension. Therefore, instead of using ramps to connect the feeding container to the collection container, the feeding container can be placed directly into the collection container in which the prepupae will fall once they have climbed up the internal wall of the feeding container (Nakamura et al. 2016; Dortmans et al. 2017). 3.2.5 Pupation Pupation is reportedly favored by stable temperature conditions, in the same range as the larval stage (Dortmans et al. 2017). However, the impact of light is somewhat unclear. While some authors assert that pupae are photophobic and require a dark environment (B. Dortmans, pers. comm., September 28, 2017; Caruso et al. 2013), a recent study by Holmes et al. (2017) established that pupae exposed to light 12 hours day-1 emerged faster as adults compared to pupae held in the dark. Adult emergence success seems to increase with rising relative humidity levels. Typically, a humidity level of 60 to 70% is recommended as it prevents the desiccation of pupae (Alvarez 2012; Holmes et al. 2012). Various materials have been proposed in the literature as pupation medium. Some authors suggested dry materials, such as wood chips (Diener et al. 2009b, 2011; Alvarez 2012), hay (Diener et al. 2011), peat (Mutafela 2015), dried coffee grounds (Nakamura et al. 2016) and even pieces of empty arboreal termites’ nests (Diener et al. 2011). On the other hand, Dortmans et al. (2017) recommended using a moist material such as compost, moist coco peat or pot soil, into which the pupae can bury. This is supported by Lin (2016), who found that optimum adult emergence rates are obtained by maintaining a moisture level of 50 to 85% in the pupation medium. Finally, Alvarez (2012) recommended providing a pupation medium at a depth of between 15 and 20 cm. To prevent the emerging flies from escaping, Diener et al. (2011) reported using a nylon net, while Dortmans et al. (2017) recommended placing the pupation containers inside a dark cage, which, in addition to containing the newly emerged flies, provides stable environmental conditions, favoring the pupation process. Regarding the location of the pupation container, two main options are described in the literature. The first consists of placing the pupation container directly in the mating cage (Mutafela 2015) so that when the flies emerge from the pupation shell, they can directly mate. However, in most set-ups, pupation occurs in sealed containers and once the flies have emerged, they are released into the mating cage (Diener et al. 2011; Popoff and Maquart 2016a; Dortmans et al. 2017). 3.2.6 Monitoring of Breeding Performance Dortmans et al. (2017) recommended monitoring survival rates at every stage of the BSF lifecycle, as well as the oviposition rate in order to assess the performance of the breeding process and identify potential problems. Table 3 compares values, for commonly used breeding performance indicators, recorded in two Indonesian BSF waste treatment facilities. TABLE 3. COMPARISON OF BREEDING PERFORMANCE IN TWO INDONESIAN FACILITIES. Performance indicators Values (%) reported by Values (%) reported Dortmans et al. (2017) by Caruso et al. (2013) Hatching rate 70 80 Larval survival rate 70 60 Adult emergence rate 80 90 Oviposition rate 350 eggs female-1 18 eggs female-1 a a Calculated based on the value provided by Dortmans et al. (2017) for the average weight of an egg (25 µg). As pointed out by Caruso et al. (2013), this value is very low compared to values reported in the literature, which could be explained by a range of physical, behavioral, abiotic or technical factors. 13 GLOBAL EXPERIENCES ON WASTE PROCESSING WITH BLACK SOLDIER FLY (HERMETIA ILLUCENS): FROM TECHNOLOGY TO BUSINESS TABLE 4. OPTIMAL OPERATING CONDITIONS FOR BSF WASTE TREATMENT. Operating parameter Optimal value References Feeding rate 20-130 for high biomass production and 4-60 for Myers et al. 2008; Diener et al. 2009b; high reduction rate (mg [milligram] larva-1 day-1, Caruso et al. 2013; Banks 2014; dry weight [DW]), depending on the waste type Parra Paz et al. 2015 Larval density 1.2-5 larvae cm-² Parra Paz et al. 2015 Waste layer thickness < 7.5 cm or < 15 cm if matrix materials are added to Perednia 2016; Yang 2017 the waste 3.3.1 Operating Conditions Optimal environmental conditions and diet for BSF larvae have been discussed in Sections 3.1 and 3.2.3. Additional key operating conditions for waste treatment are larval density, feeding rate and the feeding regime. When choosing an appropriate feeding rate and larval density, there is a trade-off between waste reduction efficiency (waste management perspective), promoted by a low feeding rate and high larval density, and biomass production (economic perspective), favored by a high feeding rate and low larval density (Diener et al. 2009b; Parra Paz et al. TABLE 6. OPTIMAL FEEDING RATE VALUES IN TERMS OF BIOMASS PRODUCTION AND/OR WASTE REDUCTION FOR DIFFERENT FEEDSTOCKS. Feedstock Optimal feeding rate (mg larva-1 day-1) in terms of… References …biomass …waste …both biomass production production reduction and waste reduction Chicken feed (UFA 625) ≥200a 100c 100 Diener et al. 2009b (60% moisture content) Vegetable and fruit waste ≥130a ≤20b 163d Parra Paz et al. 2015; (DW) Saragi and Bagastyo 2015 Dairy manure ≥70a ≤27b - Myers et al. 2008 (~ 70% moisture content) Human feces (65-85% ≥200a ≤50b - Banks 2014 moisture content) Palm kernel meal (DW) ≥64a ≤4b - Caruso et al. 2013 Note: a ≥ indicates that it was the maximal value tested in the experiment; b ≤ indicates that it was the minimal value tested in the experiment; c this value was established using the waste reduction index, which in addition to taking into account the waste reduction, considered the larval development time; d this value was established using additional parameters, besides biomass production and waste reduction, i.e. the temperature change, the final pH and the leachate production rate. 2015; Manurung et al. 2016). In addition, Dortmans et al. (2017) pointed out that if the feeding rate is too high, BSF larvae are not able to process all the waste, resulting in an unprocessed waste layer, where heat can build up due to bacterial activity, creating an unfavorable environment for the larvae. On the other hand, a feeding rate that is too low results in food shortage, which hinders larval development and waste treatment efficiency (Dortmans et al. 2017). Table 6 presents optimal feeding rate values for different feedstocks in terms of larval growth, waste reduction and both parameters. 3.3 Waste Treatment The waste treatment itself consists of feeding the young larvae produced in the breeding unit with the organic waste to be processed. The larvae fed on the waste grow into energy-rich prepupae while reducing the waste (Dortmans et al. 2017). Compared to breeding, the waste treatment step is relatively simple (Lohri et al. 2017). Optimal operating conditions for BSF waste treatment are summarized in Table 4, while the main operational designs proposed in the literature for BSF rearing containers are described in Table 5. TABLE 5. OPERATIONAL DESIGNS PROPOSED IN THE LITERATURE FOR THE BSF REARING CONTAINERS. Characteristics References Type Individual containers or larger basins Tomberlin et al. 2002; Newton et al. 2005; Volume 40-400 liters (L) Diener et al. 2011; Caruso et al. 2013; Material Plastic, metal or concrete Devic 2014; Charlton et al. 2015; Special features Drainage system, system to prevent disturbance Lalander et al. 2015; Mutafela 2015; from other insects or predators Popoff and Maquart 2016a, 2016b; Dortmans et al. 2017 14 RESOURCE RECOVERY & REUSE SERIES 16 Besides the influence of feeding rate on bioconversion performance, Parra Paz et al. (2015) also showed that larval density was a key parameter that had an even greater impact than the feeding rate. Studying vegetable and fruit waste, they suggested an optimal larval density of 1.2 larvae cm-². They established however that high biomass production could be achieved with a larval density as high as 5 larvae cm-² as long as the feeding rate was below 95 mg larva-1 day-1 (DW). Indeed, they observed that using both high larval density (over 5 larvae cm-²) and a high feeding rate (over 95 mg larva-1 day-1, DW) reduces the performance of the system in terms of both waste reduction and larval growth. In addition, it results in increased acidity, temperatures and leachate production (Parra Paz et al. 2015). Another important aspect to consider is the feeding regime. First, a choice has to be made between continuous and batch regimes. In a continuous system, larvae and waste are added continuously to a container, which is emptied only when it is full. On the other hand, in a batch operation, a defined amount of waste and number of larvae are added to a container, which is harvested when the larvae have reached maturity (Dortmans et al. 2017). By comparing batch and continuous operations, Mutafela (2015) observed better performances, in terms of waste reduction, prepupal weight and larval development time, for the batch mode. Alvarez (2012) and Dortmans et al. (2017) also recommended operating a BSF system in the batch mode to simplify maintenance of the system and isolate risks such as diseases to avoid whole system failure. Furthermore, within the batch mode, several feeding regimes can be distinguished. Banks et al. (2014) compared incremental feeding, where larvae were fed every two days with fresh feces, and lump amount feeding, consisting of providing larvae with one sample of human feces at the beginning of the feeding period. Lump amount feeding yielded better results in terms of larval growth but resulted in longer development time compared to incremental feeding, while waste reduction efficiency was similar between the two feeding regimes (Banks et al. 2014). The thickness of the waste layer is another important operational parameter, as it affects the ability of larvae to obtain a sufficient supply of oxygen. If the waste layer is too thick, larvae that tend to dig down without stopping will die of lack of oxygen as they get too deep into the waste pile. In addition, the bottom layer of the waste pile, where the larvae cannot live because of anaerobic conditions, will remain unprocessed by the larvae. In this regard, Dortmans et al. (2017) suggested that the waste layer should not exceed 5 cm thickness, while Perednia (2016) and Yang (2017) recommended a maximal depth of about 7.5 cm for the feedstock. As pointed out by Cicková et al. (2015), the waste layer thickness significantly limits the volume of waste that can be processed per square meter in a BSF facility, and implies greater space requirement, or using many shallow trays or basins, resulting potentially in a more labor-intensive process. However, Perednia (2016) reported that using matrix materials, such as pine shavings or crushed charcoal, improves the ability of BSF larvae to move through the waste pile and thus aerate it, which allows, at air pressure of about 100 hPa, to at least double the maximum depth at which oxygen supply is sufficient for the larvae to live, from about 7.5 cm to 15 cm. This allows, in turn, to at least double the waste-processing capacity per square meter. In addition, Perednia (2016) suggested mixing, turning or pumping air through the waste pile in order to ensure optimal oxygen supply for the larvae. Furthermore, some authors have explored the use of microorganisms to optimize bioconversion by the BSF. Dortmans et al. (2017) suggested the potential role of symbiotic microorganisms, which make nutrients available, through the degradation of cell structures, for the larvae to assimilate them. Yu et al. (2011) investigated the effect of inoculating poultry manure with companion bacteria (B. subtilis strains S15, S16, S19 and B. natto strain D1) on larval development. They reported that adding these four strains of Bacillus subtilis to the substrate enhanced larval development, as larger larvae were produced in a shorter time (Yu et al. 2011). Findings reported by Yu et al. (2011) are supported by a study conducted by Zheng et al. (2012a), who tested the co-conversion of rice straw and restaurant waste by the BSF and microbes (Rid-X). They reported that associating BSF with Rid-X microbes enhanced conversion of cellulose and hemicellulose into sugar, which was used by the BSF for development, as well as lignin degradation. In addition, by making more nutrients available, Rid-X microbes promoted nutrient utilization by BSF larvae and the incorporation of these nutrients into their biomass, resulting in greater biomass production. Protein utilization by the BSF increased from 74 to 92% by adding Rid-X microbes to the substrate. Zheng et al. (2012a) particularly recommended the use of microorganisms to assist the BSF in the conversion of lignocellulosic materials, which the BSF have more difficulty in processing. 3.3.2 Operational Designs In the treatment unit, the waste to be processed is typically placed in containers where the small larvae from the breeding unit are added to quickstart the waste decomposition process. As reported in the literature, these are usually made of plastic (Tomberlin et al. 2002; Lalander et al. 2015; Mutafela 2015; Dortmans et al. 2017), metal (Diener et al. 2011; Devic 2014) or concrete (Newton et al. 2005; Caruso et al. 2013; Popoff and Maquart 2016a, 2016b). Containers with a wide range of volumes, i.e. from 40 to 400 L, are reported in the literature (Diener et al. 2011; Caruso et al. 2013; Lalander et al. 2015; Charlton et al. 2015; Mutafela 2015; Popoff and Maquart 2016b; Dortmans et al. 2017). They consist typically of either individual containers that can be handled by operators (Diener et al. 2011; Lalander et al. 2015; Dortmans et al. 2017) or larger basins (Newton et al. 2005; Caruso et al. 2013; Popoff and Maquart 2016a, 2016b). However, Dortmans et al. (2017) recommended avoiding very large containers so that risk is divided in the 15 GLOBAL EXPERIENCES ON WASTE PROCESSING WITH BLACK SOLDIER FLY (HERMETIA ILLUCENS): FROM TECHNOLOGY TO BUSINESS event of a problem. In addition, to save space, several authors suggest taking advantage of vertical space by stacking individual containers upon each other with ventilation frames in-between levels to allow air to flow or placing them on vertical shelves (Popoff and Maquart 2016a, 2016b; Dortmans et al. 2017; Zurbrügg et al. 2018). Most containers are rectangular in shape, however Caruso et al. (2013) described an Indonesian BSF facility that uses circular basins. Additionally, some authors reported fitting the containers with a drainage system, usually consisting of a plastic pipe leading to a tap, in order to prevent liquid from stagnating and creating anaerobic conditions (Diener et al. 2011; Mutafela 2015). The literature also describes various systems to prevent the invasion of insects, like wasps or flies, or predators such as lizards, which can disturb the process. To trap other insects, authors have suggested using buckets (Diener et al. 2009b; Dortmans et al. 2017) or building a concrete channel that surrounds the facility (Popoff and Maquart 2016b). Filling them with water and a few drops of liquid detergent or oil enables the reduction of water surface tension and thus drowning of insects (Diener et al. 2009b; Popoff and Maquart 2016b; Dortmans et al. 2017). Popoff and Maquart (2016b) also suggested using double-door systems to isolate the waste treatment unit. 3.3.3 Monitoring of the Waste Treatment Unit’s Performance Common parameters used in the literature to assess the system’s performance are waste reduction rate (Diener et al. 2009b, 2011; Banks et al. 2014; Lalander et al. 2015; Dortmans et al. 2017; Lohri et al. 2017; Lalander et al. 2019), bioconversion rate (Banks et al. 2014; Lalander et al. 2015; Lohri et al. 2017; Dortmans et al. 2017; Lalander et al. 2019), mean larval/prepupal weight (Cicková et al. 2015; Lalander et al. 2019), larval development time (Diener et al. 2009b; Cicková et al. 2015; Lohri et al. 2017) and feed/food conversion rate (FCR) (Diener et al. 2011; Caruso et al. 2013; Banks et al. 2014). As the performance of a BSF system depends on the type of waste that is being processed, values for the main performance indicators used in the literature are presented for different feedstocks in Table 7. 3.4 Product Harvesting and Post- treatment The BSF process yields two main products, namely mature BSF larvae, and the waste residue, whose properties, applications and post-treatment are described in the following sections and summarized in Table 8. 3.4.1 Product Yields Product yields vary significantly depending on the waste type being processed. Overall, yield values reported in the literature for mature larvae and waste residue range respectively from 40 to 118 kg of larvae tonne of waste-1 and 210 to 810 kg of waste residue tonne of waste-1, on a dry basis (Newton et al. 2005; Myers et al. 2008; Diener et al. 2011; Nguyen et al. 2013; Banks et al. 2014; Saragi and Bagastyo 2015; Rehman et al. 2017a). Yields of mature larvae and waste residue for different feedstocks can be deduced from, respectively, the bioconversion and the waste reduction rates presented in Table 7. 3.4.2 Harvesting Techniques The technique used to separate the BSF from the waste residue depends on the stage at which it is being harvested, i.e. larval or prepupal stage. When harvesting is carried out at the prepupal stage, the most common method reported in the literature is self-harvesting, i.e. prepupae, naturally migrating from the waste to find a pupation site, are guided to a given location, typically via a ramp, to be harvested (Diener et al. 2011; Mutafela 2015; Popoff and Maquart 2016a). The advantage of self-harvesting is that it is a simple and non-labor-intensive method. On the other hand, to harvest the larvae before they turn into prepupae, a manual sieve or an automated shaking sieve are used (Popoff and Maquart 2016a; Cheng et al. 2017; Dortmans et al. 2017). Dortmans et al. (2017) recommended a sieve mesh size of 3 millimeters (mm) for manual sieving and 5 mm for automated sieving. Cheng et al. (2017) demonstrated that larvae could be harvested using a manual 2.36-mm sieve from the residue of food waste, whose initial moisture content was below 80% (wet basis). On the other hand, if the initial moisture content of the waste is above 80% (wet basis), instead of obtaining a crumbly waste residue, the waste residue will be in the form of a slurry with unprocessed chunks (Cheng et al. 2017; Dortmans et al. 2017). In that case, Dortmans et al. (2017) suggested using non-shaking flat screens with a 5-mm mesh, through which both the liquid and larvae that want to avoid sunlight will flow and fall into a container placed below, while unprocessed chunks will remain on top of the screen. Larvae that have fallen through the mesh into the collection container can then be harvested from the liquid using a strainer spoon (Dortmans et al. 2017). 16 RESOURCE RECOVERY & REUSE SERIES 16 TA B LE 7 . B IO C O N V E R S IO N P E R F O R M A N C E F O R D IF F E R E N T F E E D S T O C K S . F ee d st o ck W as te M ea n fin al La rv al d ev el o p m en t B io co nv er si o n F o o d R ef er en ce s re d uc tio na (% ) la rv al w ei g ht b (m g ) tim ec (d ay s) ra te d (% ) co nv er si o n ra tio e P ig m an ur e 56 (D W ) 11 3 (W W ) 25 -4 6 4 (D W ) 10 (D W ) N ew to n et a l. 20 05 ; N gu ye n et a l. 20 13 ; B an ks e t a l. 20 14 D ai ry m an ur e 33 -5 8 (D W ) 13 7- 17 9 (W W ) 26 -3 0 2- 4* (D W ) - M ye rs e t a l. 20 08 C hi ck en m an ur e 50 (W W ) 22 0 (W W ) - 4 (W W ) 13 (W W ) S he pp ar d et a l. 19 94 ; B an ks e t a l. 20 14 D ai ry m an ur e an d 43 -5 5 (D W ) 60 -1 00 (W W ) 18 -2 2 4- 10 (D W ) 6- 10 (D W ) R eh m an e t a l. 20 17 a ch ic ke n m an ur e H um an f ec es 25 -5 5 (W W ) 19 4- 31 5 (W W ) - 2- 22 (W W ) 2- 16 (W W ) B an ks e t a l. 20 14 M S W 66 -7 9 (D W ) 13 8- 22 0 (W W ) - 12 (D W ) 15 (D W ) D ie ne r et a l. 20 11 ; B an ks e t a l. 20 14 K itc he n w as te - 17 3 (W W ) 20 -3 3 - - N gu ye n et a l. 20 13 R es ta ur an t w as te - 15 4 (W W ) 19 - - S pr an gh er s et a l. 20 17 F ru it an d v eg et ab le s 43 -6 4 (D W ) 12 3 (W W ) 22 -4 0 - - N gu ye n et a l. 20 13 ; S ar ag i a nd B ag as ty o 20 15 Ve g et ab le w as te - 14 0 (W W ) 16 - - S pr an gh er s et a l. 20 17 F is h w as te 19 -5 4 (D W ) 14 3 (W W ) 20 -3 6 - - N gu ye n et a l. 20 13 ; S ar ag i a nd B ag as ty o 20 15 O ve ra ll ra ng e 19 -7 9 60 -3 15 16 -4 6 2- 22 2- 16 a W ei gh t p er ce nt ag e of th e in iti al w as te a dd ed th at is re du ce d ov er th e fe ed in g pe rio d. b M ea n w ei gh t o f o ne la rv a at th e en d of th e fe ed in g pe rio d. c T im e re qu ire d fo r th e ju ve ni le la rv ae a dd ed to th e w as te to re ac h th e pr ep up al s ta ge . d W ei gh t p er ce nt ag e of w as te a dd ed th at is c on ve rt ed in to la rv al b io m as s. It in di ca te s ho w m an y ki lo gr am s (k g) o f m at ur e la rv ae c an b e ob ta in ed fr om 1 00 k g of w as te . e R at io o f t he w ei gh t o f f ee d in ge st ed a nd w ei gh t g ai ne d by th e la rv ae o ve r th e fe ed in g pe rio d. It m ea su re s th e ef fic ie nc y of th e la rv ae to c on ve rt th e fe ed in ge st ed in to b od y m as s. * Va lu e ob ta in ed b y us in g an e qu at io n : D ry fe ed in ta ke (g ) d iv id ed b y W et w ei gh t g ai n (g ) N ot e: D W : d ry w ei gh t; W W : w et w ei gh t. 17 GLOBAL EXPERIENCES ON WASTE PROCESSING WITH BLACK SOLDIER FLY (HERMETIA ILLUCENS): FROM TECHNOLOGY TO BUSINESS TABLE 8. BSF PRODUCTS’ PROPERTIES AND APPLICATIONS. Mature BSF larvae Waste residue Yield 40-118 kg of larvae tonne of waste-1 (DW basis) 210-810 kg of waste residue tonne of Typically, 200 kg (WW) of larvae tonne of waste-1 waste-1 (DW) Properties High protein (40% DW) and lipid content (35% DW). The waste residue contains nutrients, Relatively rich in Ca, P and K. Main fatty acids: lauric including increased concentrations of acid, palmitic acid and oleic acid. Main essential amino ammonium nitrogen. The residual C/N ratio acids: lysine, valine and leucine. depends on the initial C/N ratio of the input waste. pH between 7 and 8. Compost obtained is immature. Safety The levels of most chemical contaminants are lower BSF waste treatment removes, in animal than those recommended. The only chemical risk and human waste, bacteria from the identified pertains to the bioaccumulation of cadmium Enterobacteriaceae family (Salmonella in larvae. There is also a risk of presence of pathogens spp. and E. coli) under sufficient in larvae reared on animal or human waste despite the temperature (27-32 °C) and alkaline antibacterial properties of the larvae. conditions but has no effect on the destruction of other pathogens such as Enterococcus spp., bacteriophage or Ascaris suum ova. BSF treatment also accelerates the degradation of different types of pharmaceuticals and pesticides in the waste. Applications The main application for BSF larvae is their use as Fertilizer feed ingredients for monogastric animals. The oil extracted from the larvae can also be used to produce biodiesel and the chitin contained in the exoskeleton of the larvae can be sold as a chelating agent. Post-treatment Sanitization (e.g. boiling), drying, lipid extraction, etc. Thermophilic composting or vermicomposting or anaerobic digestion References Hale 1973; Newton et al. 1977, 2005; Bondari and Erickson et al. 2004; Newton et al. 2005; Sheppard 1981, 1987; Erickson et al. 2004; St-Hilaire Liu et al. 2008; Choi et al. 2009; et al. 2007a, 2007b; Diener 2010; Diener et al. 2011, Diener et al. 2011; Green and Popa 2012; 2015b; Li et al. 2011b; Sealey et al. 2011; Zheng et al. Lalander et al. 2013, 2015, 2016; Banks 2012a, 2012b; Caruso et al. 2013; Finke 2013; et al. 2014; Adeku 2015; Dortmans 2015; Lalander et al. 2013, 2016; Banks et al. 2014; Lock Saragi and Bagastyo 2015; Murray 2016; et al. 2014; Makkar et al. 2014; Charlton et al. 2015; Dortmans et al. 2017; Lohri et al. 2017; Leong et al. 2015, 2016; Park et al. 2015; Tran et al. Quilliam et al. 2017; Rehman et al. 2017a 2015; Cummins Jr. et al. 2017; Devic et al. 2017; Dortmans et al. 2017; Gao et al. 2017; Liu et al. 2017; Rehman et al. 2017a; Liland et al. 2017; Schiavone et al. 2017; Spranghers et al. 2017; Zurbrügg et al. 2018 Note: Ca = calcium; P = phosphorus; K = potassium. 3.4.3 Post-treatments BSF Larvae Further processing of the harvested larvae is usually required for sanitization, storage and transport purposes (Zurbrügg et al. 2018). Sanitization can be achieved by placing the larvae in boiling water for about two minutes, which kills off the bacteria on the larvae and allows them to empty their guts (Dortmans et al. 2017). Alternatively, Charlton et al. (2015) suggested washing the larvae with water and placing them in sawdust overnight to allow them to empty their guts. Then, depending on the market demand, the larvae can also be frozen or dried (Dortmans et al. 2017). Drying is particularly interesting as it is less energy-intensive and further sanitizes the product (Lalander et al. 2013). In addition, as the dry matter content of fresh prepupae is quite high (30 to 45%), dehydrating them could be easier and less costly compared to other fresh by-products (Newton et al. 2008; Makkar et al. 2014; Tran et al. 2015). Dortmans et al. (2017) recommended drying the larvae until their moisture content drops below 10% so they can be stored efficiently. The fat content should also be kept low for storage (Zurbrügg et al. 2018). Different 18 RESOURCE RECOVERY & REUSE SERIES 16 drying methods are suggested in the literature. Charlton et al. (2015) suggested placing the larvae in a gas oven at 60 to 80 °C for two hours. Alternatively, solar drying is a low-cost and energy-saving solution particularly adapted to low-income and tropical countries. Caruso et al. (2013), who used bamboo baskets to sun dry BSF larvae in Indonesia, established that under a light intensity greater than 2,000 lux, a temperature of 38 °C and air humidity of about 50%, 17 hours of sunshine were required to dry 95% of the larvae