1 Introduction

World production will require at least 70% increase of food by the year 2050 [32, 34]. However, chances of negative impact are there with the increasing global food productivity on the environment [34]. Responsible agricultural production based on the recycling of organic waste is adequate for increasing global agricultural production without major negative impacts on the environment [29, 44]. Agricultural production generates a large amount of waste and agricultural byproducts. When these waste and agricultural byproducts are not well managed, they constitute a real environmental problem [13, 22, 39, 40]. Good management of agricultural byproducts can not only reduce their negative effects on the environment but also allow transformations that promote their valorization as an input in agriculture. Several studies have focused on the recycling of organic waste and agricultural byproducts to produce reusable products in agriculture. The use of vermicomposting to improve the quality of compost and to ensure that earthworms are usable in animal production is some example [12, 25, 44, 47]. Organic waste is also composted by various methods with the aim of improving the quality of compost for better yield in crop production [1, 4, 36, 42]. Biogas production use in cooking is also a way for decomposing of organic wastes and agricultural byproducts [17]. In the way of decomposing of organic waste and agricultural byproducts, the production of fly larvae is becoming a topic of interest for many researchers [2, 24]. In this context, several studies have been carried out on the production of black soldier fly larvae for the recycling of organic waste [16, 35]. Hermetia illucens are insects with a low impact on the environment. They can grow on waste and byproducts; additionally, they have a high feed conversion efficiency and a low risk of transmission of zoonotic infections [27, 45, 46]. Among the insect species capable of rapidly producing significant biomass under controlled rearing conditions, the black soldier fly is currently the main species widely studied for its high capacity for waste degradation and its use as a feed ingredient [26, 28]. It is a protein source with a well-balanced essential amino acid (EAA) profile that is almost comparable to that of fish meal [3, 11, 15, 30]. The protein content of H. illucens varies from 37 to 63%, while the lipid content also varies from 7 to 39%, depending mainly on the rearing substrate [10, 21]. In addition, BSFL are saprophagous and produced on an industrial scale worldwide because of their suitability and nutritional value [46]. Given its role as a tool for recycling organic animal and plant waste, several studies have used kitchen waste, animal droppings, cakes and bran for its production. Tuber peels also constitute agricultural byproducts that can be valorized because of their nutritional composition, especially starch. Cassava, yam and sweet potato peels can therefore be used as substrates for the production of black soldier fly larvae. Cassava is a widely consumed tuber in Africa [6, 9]. Before processing, cassava tuber is often peeled off. Its peels constitute waste for cassava processing companies. Yam is a tuber of great importance that is produced mainly in Africa and accounts for approximately 96% of its global production. In 2014, its global production was estimated at approximately 50 million tons. It constitutes the staple food of more than 155 million people worldwide [7]. In Benin, production is much more concentrated in the central and northern regions of the country [8]. It is used in the composition of several dishes, and its peels are sometimes a source of pollution for large processors. Sweet potatoes are also part of several diets and generate waste through their peels. It is a tuber from Central America and has very interesting nutritional value [38]. In a context of recycling and valorization of agricultural by-products, these tubers peelings constitute potential production substrates for BSFL. The present study aimed to enhance the value and evaluate the effects of these tuber peeling on the production of black soldier fly larvae.

2 Materials and methods

2.1 Experimental location

This study was carried out in Benin in the district of Sèmè-Kpodji on the AWARA ET FILS farm (6°23′11.9ʺN 2°36′16.2ʺE). Sèmè-Kpodji is located in south of Benin in Ouémé department.

2.2 Experimental design

Before the rearing of BSFL, 30 g of prepupae were used in the love cage with an egg-laying attractant (pineapple fruit) to attract females to the egg-laying site for its reproduction. A wooden nest for females to lay eggs was placed in the egg-laying structure to shelter the eggs [20]. A cloth soaked in water was placed in the love cage to allow the flies to hydrate. Egg collection was discontinued when all the flies (spawners) died. After brood stock spawning, the eggs were harvested from the spawning structure, collected and incubated in trays. The egg hatching substrate was chick feed to allow optimal development of the larvae during the first 5 days before the growth experiment [20]. The chick feed used is for “Poussin 1” for “GVS Company” Five days after incubation, the larvae were sieved through 2 mm and 1 mm mesh sieves to obtain a homogeneous sample. A uniform ratio of 100 mg/larva/day was used. Twelve tanks (25 cm × 14 cm × 12 cm) were arranged in triplicate and constituted the 4 experimental treatments. A density of 1 larva/g of substrate was used for seeding [20].

2.3 Rearing substrates

Four rearing substrates used were soy bran (SB) materialized by treatment (TSB), cassava peel (CP) materialized by treatment (TCP), yam peel (YP) materialized by treatment (TYP) and potato peel (PP) materialized by treatment (TPP). Soy bran (SB) was obtained by processing soy cheese. CP was obtained after peeling cassava from the “Gari” transformation factory. Gari is a staple food produced in abundance in southern Benin. YP was obtained after peeling yam from a pounded yam restaurant. Pounded yam is a staple food produced in northern Benin. PP was obtained after potato peeling from potato chip sellers in southern Benin.

2.4 Substrates treatment

After collecting the substrates, they were dried in the sun for 3 days. The substrates dried were finely ground with the mill and sieved through a mesh of 0.1 mm (Fig. 1). All rearing substrates were dosed in dry matter, organic matter, ash, total nitrogen, and carbon, and the carbon/nitrogen ratio was calculated.

Fig. 1
figure 1

a cassava peel; b yam peel; c sweet potato peel before treatment. d cassava peel; e yam peel; f sweet potato peel after treatment

Full size image

2.5 Harvest of larvae and experimental follow-up

The substrates were moistened to 70% moisture [10]. The larval growth duration was ten days up to the end of larvae stage. Growth monitoring was carried out every two days with 50 individuals weighing 0.001 g on a 500 g scale. The physicochemical parameters, such as the pH and temperature of the substrates, were measured with a pH meter and a thermometer “ELMETRON CP-411” brand throughout the experimental period. These parameters were taken twice daily (at 7:00 a.m. and 5:00 p.m.). At the end of the experiment, the larvae were harvested and weighed. These larvae were dried in an oven at 50 °C for 6 h [20]. The harvested and dried larvae were assayed in the laboratory for proteins, lipids and dry matter according to the treatments according to the [5] standard methods. The rest of the substrates were dried according to the treatments at 105 °C in an oven and weighed to assess the loss of substrate.

2.6 Chemical analysis of the samples

Total carbon was measured using the method of [31]. Organic matter content was determined by the following equation: \(OM\%= \text{C\% }\times 1.724\). Nitrogen was determined according to Bremner & Mulvaney (1982) method [14]. Total phosphorus was analyzed using the colorimetric method with molybdenum in sulfuric acid. The dry matter content was determined by dehydrating the samples in a drying oven at 105 °C overnight. Ash contents were determined by incinerating samples in a muffle furnace heated to 550 °C at an increasing rate of 50 °C every 30 min for 4 h and then cooling in a desiccator. The proteins and lipids in the larvae at the end of the experiment were measured in triplicate according to the AOAC [5] standard methods to evaluate the quality of the larvae.

2.7 Growth parameters

Ten days after growth experiment, zootechnical parameters such as the survival rate (SR), daily weight gain (DWG), larvae weight (LW), production (P), and degradation rate (DR) were calculated to assess substrate performance.

$${\varvec{S}}{\varvec{R}}\boldsymbol{ }(\boldsymbol{\%})=\frac{FN}{IN}\times 100$$
(1)

where: IN: initial number of larvae and FN: final number of larvae.

The initial number of larvae was measured on the first day of waste feeding. The final number of larvae was measured after 10 days of feeding when the first pupae were observed.

$$\mathbf{D}\mathbf{W}\mathbf{G}(\mathbf{g}/\mathbf{d})=\frac{FB-IB}{\Delta t}$$
(2)

IB: initial biomass (g), FB: final biomass (g) and Δt: duration of the experiment in number of days

$$\mathbf{L}\mathbf{W}(\mathbf{g})=\frac{FB}{FN}$$
(3)

LW (g): Larvae weight

$$\mathbf{P} (\text{g of larvae}/\text{kg of substrate})=\frac{FB-IB}{Q}$$
(4)

where Q is the quantity of substrate (kg)

$$\mathbf{D}\mathbf{R} (\text{\%}) =\frac{W-R}{W}\times 100$$
(5)

W: the initial quantity of substrate; R: residue of substrate at the end of breeding.

2.8 Data processing

The mean and standard deviation were calculated for each parameter. Statistical analysis was performed using STATVIEW software version (5.01) by analysis of variance with one classification criterion (ANOVA 1). The zootechnical data obtained were subjected to Fisher's LSD tests to compare the different means calculated at the 5% threshold. A Z test was performed to assess the correlation between growth parameters and the chemical composition of the substrates.

3 Results

3.1 Physicochemical and biochemical parameters during BSFL rearing

The physicochemical parameters measured during the experiment (pH and temperature) are summarized in Table 1. The biochemical parameters of the different substrates used before starting the experiment are summarized in Table 2. The average pH of the substrates during the experiment varied from 6.70 ± 0.31 to 7.70 ± 0.15. The temperature of the substrates during the experiment varied from 27.26 ± 0.20 °C to 27.59 ± 0.28 °C. For the biochemical parameters of the different substrates used before starting the experiment, significant differences were observed (p < 0.05). The dry matter content ranged from 88.44 ± 0.13% to 91.44 ± 0.07%. The organic matter content ranged from 23.68 ± 0.77% to 67.45 ± 0.06%. The ash content ranged from 32.54 ± 0.06% to 76.32 ± 0.80%. The carbon content varied from 10.87 ± 0.10% to 34.54 ± 0.18%. The nitrogen content ranged from 2.13 ± 0.14% to 4.40 ± 0.25%. Phosphorus varied from 0.43 ± 0.02% to 0.72 ± 0.12%. The C/N ratio varied from 5.15 ± 0.36% to 8.51 ± 0.46%.

Table 1 Physicochemical parameters during BSFL rearing
Full size table
Table 2 Biochemical parameters of the different substrates used
Full size table

3.2 Growth and substrate parameters

During the experiment, the growth and substrate parameters were monitored and are summarized in Table 3. The final biomass ranged from 26.10 ± 0.77 g to 49.40 ± 0.26 g. The DWG ranged from 2.19 ± 0.09 g/d to 4.53 ± 0.02 g/d. The production ranged from 43.90 ± 1.87 g/kg of substrate to 90.56 ± 0.54 g/kg of substrate. The degradation rate ranged from 36.00 ± 2.30% to 65.00 ± 2.88%. The survival rate ranged from 77.66 ± 1.45% to 99.00 ± 0.28%. For all the parameters, the Yam peel treatment presented the lowest value among the treatments. The highest value was observed in the reference treatment (TSB).

Table 3 Growth performance and feed utilization
Full size table

The growth curves of the BSF larvae varied depending on the rearing substrate. We noticed more interesting growth on the substrates composed of TSB and poorer growth on the substrates composed of TYP (Fig. 2). The same observation was made for the larvae weight (PW). A significant difference was observed between the different treatments (p < 0.05) depending on the treatment (Fig. 3). Larvae weight ranged from 0.16 ± 0.01 g (TYP) to 0.24 ± 0.01 g (TSB). Among the treatments, TYP yield was the lowest, and TCP yield was the highest (Fig. 3).

Fig. 2
figure 2

Growth curves of BSF larvae (Hermetia illucens) on various substrates

Full size image
Fig. 3
figure 3

Final larvae weights of different substrates used for production

Full size image

3.3 Correlations between the physicochemical parameters of the substrates and DWG

The correlation between the biochemical parameters of the substrates and the DWG was established by the “correlation Z test”, which allowed us to establish the relationship between the biochemical composition of the substrate and the growth parameters and substrate degradation (Table 4). There was a strong correlation between the organic matter content of the substrate and that of the DWG (99.1%). It was also observed that phosphorus was strongly correlated with DWG (92.9%). The carbon/nitrogen ratio also presented a high correlation with DWG (78.6%).

Table 4 Correlations between several physicochemical parameters of the substrates and DWG
Full size table

3.4 Effects of substrates on the nutritional quality of BSFL

After the larvae were harvested, it was dried and assayed for dry matter, protein and lipids (Table 5). The use of cassava, yam and sweet potato peels in the production of BSFL affects the nutritional composition of the larvae. No significant difference (p > 0.05) was observed in the dry matter content of the BSFL according to the treatment. It varied from 38.89 ± 0.58% (TCB) to 39.42 ± 0.36% (TSB). A significant difference (p < 0.05) was observed between the different treatments with regard to the lipid and protein content. The protein content varied from 33.51 ± 0.94% (TYP) to 40.64 ± 032% (TSB). The lipid content varied from 26.24 ± 0.35% (TYP) to 30.44 ± 0.18% (TSB). Notably, the substrate composed of yam peels induced a low lipid content in BSFL among the treatments composed of peel substrates. There was no difference in lipids between the BSFL product and the cassava peel or sweet potato peel substrates.

Table 5 Nutritional values of BSF larvae produced on different substrates
Full size table

4 Discussion

4.1 Physicochemical parameters during the experiment and biochemical composition of the rearing substrates

The physicochemical parameters during the experiment and the nutritional composition of the rearing substrates are therefore necessary not only for the nutritional quality of the larvae but also for the growth of BSFL and for the reproduction of adults [33]. The average temperature of the substrates during the experiment varied from 27.26 ± 0.20 °C (TSB) to 27.59 ± 0.28 °C (TPP). These values are within the temperature range accepted by BSFL for development, which is between 24 and 36 °C according to studies of Makkar et al. [28]. The pH of the substrates during the experiment varied from 6.70 ± 0.31 (TYP) to 7.70 ± 0.15 (TSB). This value is within the range accepted by the species according to studies of Makkar et al. [28], which showed that the optimum pH for the development of BSFL is between 6 and 8. With regard to the other parameters, dry matter, organic matter, ash, carbon, nitrogen phosphorus, and carbon/nitrogen, significant differences were observed (p < 0.05). These different parameters that define the nature of the substrates have impacted the growth and nutritional value of BSFL. The nitrogen application rate (2.13 ± 0.14%) in this study was lower than the nitrogen application rate associated with the use of manure (3.60%) and vegetables and fruits (3.21%) for BSFL rearing [33]. Growth and substrate parameters; relationship between biochemical parameters of the substrates and zootechnical performance; and biochemical quality of the substrates; stocking density and humidity are the parameters that influence the growth and survival of BSFL during rearing [20].

4.2 Growth parameters and bioconversion

The differences in growth, survival and substrate utilization observed are due to the biochemical quality of the rearing substrates used. The stocking density (1 larva/g of substrate) applied in the present study is in line with the studies of Gougbedji et al. [20]. Additionally, the humidity (70%) applied in the present study is in line with the studies of Gougbedji et al. [20]. For the larvae weight, a significant difference (p < 0.05) was observed between the treatments. A high value (0.25 g) for final larvae weights was observed with the reference treatment composed of soy bran, which was approximately the same value (0.24 g) as that reported in the studies of [20] and (0.25) as that reported in the studies of vodounnou et al. [43]. All the other values obtained with the cassava, yam and potato peel substrates were lower than those obtained with the reference substrate. In the study by Gougbedji et al. [20], a purely vegetal substrate composed of soybean meal, rapeseed oil and Euphorbia heterophilla leaves was used. In the studies of [37] and [23], the larvae weight varied between 0.22 g and 0.25 g, corroborating the results obtained for soy bran and cassava peel substrates. We noted that a low degradation rate was observed for the yam peel substrate. This result can be explained by the low organic matter content of this substrate. This value is also lower than that obtained by [20] (82.05%) using a purely vegetal substrate composed of soybean meal, rapeseed oil and Euphorbia heterophilla leaves. Those results show that biochemical quality of the rearing substrates affect the growth of BSFL. It is corroborating the results Fitriana et al. [48] which shows that a low nutrient content of the substrate could reduce growth performance. The survival rate of BSFL varied from 77.66% (TYP) to 99.00% (TSB). The lowest value was also observed in the yam peel treatment. Despite this low survival rate with the substrate composed of yam peels (77.66%), which is higher than the survival rate obtained by [20] when using chick feed to produce BSFL (50%).

The bioconversion capacity of BSFL was evaluated in this study by the degradation rate of different substrates by BSFL. The degradation capacity of the substrates varied depending on the substrates. Among the peels used, cassava (48.13 ± 1.04) and sweet potato (45.33 ± 1.45) peels have a higher bioconversion rate than yam peels (36.00 ± 2.30). Several studies have shown the ability of BSFL to decompose organic substrates. These studies have also focused on the variability of bioconversion depending on the substrates. The different degradation rates obtained in this study corroborate the studies of [48] which evaluated the bioconversion of several substrates such as animal waste, agricultural by-products and animal feed ingredients by BSFL and obtained different conversion rates.

4.3 Correlations between the physicochemical parameters of the substrates and DWG

The DWG of the BSFL was monitored during the study. The organic matter, carbon/nitrogen ratio and phosphorus content of the substrate allow good zootechnical performance of BSFL. A high correlation (99.1%) was observed between the organic matter content of the substrate and the DWG of the BSFL. Similarly, the phosphorus content of the substrate was strongly correlated with the DWG (92.9%). The C/N ratio also promoted the growth of BSFL, with a correlation of 78.6%. It is also noted that the growth of BSFL strongly degrades the substrate. An increase in the abundance of organic matter, nitrogen and phosphorus in the substrate would better promote the growth of BSFL and enhance its degradation. The abundance of substrates in organic matter and phosphorus increases BSFL production [41]. The present study also shows that the abundance of Nitrogen in the substrate favors the growth of BSFL.

4.4 Relationship between rearing substrates and nutritional quality of BSFL

There was no difference in protein content between the BSFL product in the yam peel or sweet potato substrates. These values are lower than those contained in BSFL produced in cassava peels. The protein content of the substrates can therefore act not only on production of BSFL but also on nutritional quality. The protein content in BSFL produced by yam and sweet potato substrates was lower than that reported by Henry et al. [21] and Barragan et al. [10]. The lipid content varied from 26.24% (TYP) to 30.44% (TSB). These values corroborate the findings of Henry et al. [21] and Barragan et al. [10] who reported that the lipid content of BSFL remained in the range of 7 to 39%. Notably, rearing substrates affect the nutritional quality of BSFL [18, 19]. This result is in opposite with the result of study [48] where the nutrient content of the substrates did not significantly affect the nutritional quality of BSFL. According this study, A low nutrient content of the substrate could still produce BSF larvae with a high crude protein and ether extract contents.

5 Conclusion

The current study highlights the importance role of BSFL in waste management converting organic waste into food resource. It can be noted that cassava, yam and sweet potato peels can be used in the rearing of BSFL. The best growth, degradation of the substrate and survival of BSFL were obtained with the cassava peel substrate. We also note that the productivity and nutritional quality of BSFL are strongly correlated with the abundance of the substrate in organic matter, Nitrogen and Phosphorus. The protein in substrates affected the growth performances and the nutritional quality of BSFL. In the future, the cassava, yam and sweet potato peels flour can be manufactured for BSFL production.