Waste Management 100 (2019) 151–160

Contents lists available at ScienceDirect

Waste Management
journal homepage: www.elsevier.com/locate/wasman

Pre-treatment of banana peel to improve composting by black soldier fly

(Hermetia illucens (L.), Diptera: Stratiomyidae) larvae
A. Isibika a,b,⇑, B. Vinnerås a, O. Kibazohi b, C. Zurbrügg c, C. Lalander a
a
Department of Energy and Technology, Swedish University of Agricultural Sciences, Box 7032, 75007 Uppsala, Sweden
b
Department of Chemical and Mining Engineering, College of Engineering and Technology, University of Dar es Salaam, P.O. Box 35131, Dar es Salaam, Tanzania
c
Eawag: Swiss Federal Institute of Aquatic Science and Technology, Überlandstrasse 133, 8600 Dübendorf, Switzerland

a r t i c l e i n f o a b s t r a c t

Article history: Use of black soldier fly (Hermetia illucens (L.), Diptera: Stratiomyidae) larvae (BSFL) is among the solutions
Received 22 July 2019 being explored to shift the value chain in organic waste management by producing valuable products.
Revised 10 September 2019 Although BSFL consume a range of substrates, nutrient-imbalanced materials with high hemicellulose
Accepted 12 September 2019
and lignin content, e.g. manure and banana peel, yield low conversion into larval biomass. This study
Available online 17 September 2019
explored pre-treatment methods to improve the nutrient composition and digestibility of banana peel
to achieve higher substrate conversion into BSFL biomass. The pre-treatment methods evaluated were
Keywords:
microbial, chemical (non-protein nitrogen), heat-based, and combinations of these. All pre-treatments
Food industry waste
Fruit peel
tested except heating resulted in more efficient BSFL conversion in terms of final larvae weight. The
Eco-technologies low BSFL responses in pre-treatments were caused by the observed high amounts of tannins and phenolic
Trichoderma reesei compounds mainly from the heating pre-treatment. Waste to biomass conversion ratio correlated nega-
Rhizopus oligosporus tively with substrate volatile solids (VS) and positively with the decrease in VS in pre-treatment.
Organic waste management Microbial
14 days pre-treatments provided the optimum pre-treatment time for the microorganisms
to achieve maximum degradation of the substrates, facilitating larval assimilation of the released nutri-
ents. Rhizopus oligosporus-14 days and ammonia + Rhizopus resulted in the most efficient BSFL treatment,
measured as protein produced per kg incoming material.
Ó 2019 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction waste allows biodegradable materials to decompose under unhy-

gienic conditions leading to accumulation of greenhouse gases
Population growth and urbanisation is expected to increase causing climate change (Cogut, 2016). The dumped waste attracts
municipal solid waste generation to 2.2 billion tonnes by 2025 insects and rodent vectors that spread diseases such as cholera and
and predicted to affect low- and middle-income countries (LMIC) malaria (Chowdhury et al., 2017). Untreated leachate from the
most, due to unsustainable waste management practices decomposing dumped waste contaminate surface and groundwa-
(WorldBank, 2017). Solid waste management in most LMIC ter supplies (Nagarajan et al., 2012).
involves open dumping and burning (WorldBank, 2018). Municipal Besides mixed municipal solid waste, homogeneous organic
solid waste in those countries contains a mixture of plastics (<13%), solid waste is generated by agricultural and industrial activities
metals (<4%), paper (<9%), glass (<4%) and more than 50% organic in LMIC during agricultural harvests and agro-industrial processes
waste (Hoornweg and Perinaz, 2012). Solid waste dumped at land- comprising stems, stalks, peel, seeds and pulp (Krishna and
fill undergoes sorting by waste pickers, who remove a large pro- Chandrasekaran, 1996). For instance banana, belonging to the fam-
portion of the non-organic waste leaving high proportions of ily Musaceae and genus Musa, is a crop grown in 120 countries
biodegradable organic waste (Komakech et al., 2014). Open burn- throughout the world (Byarugaba-Bazirake, 2008). Tanzania con-
ing is a source of dangerous carcinogenic gases and of black carbon, tribute large quantities of bananas grown in Africa while generat-
a short-lived climate pollutant (CCAC, 2015). Open dumping of ing around 60 tonnes ha
1 of banana-related organic solid waste
per harvest (Tock et al., 2010; Emmanuel et al., 2014). Banana is
a source of staple food and cash crop, and its juice is used commer-
⇑ Corresponding author at: Department of Energy and Technology, Swedish
cially to produce alcoholic drinks (Carter et al., 2010). Banana
University of Agricultural Sciences, Box 7032, 75007 Uppsala, Sweden.
E-mail address: [email protected] (A. Isibika).
waste from harvesting consists of stems and leaves, while banana

https://doi.org/10.1016/j.wasman.2019.09.017
0956-053X/Ó 2019 The Author(s). Published by Elsevier Ltd.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
152 A. Isibika et al. / Waste Management 100 (2019) 151–160

waste from industrial production of e.g. banana crisps, banana 2. Materials and methods
juice and wine consists of peel and pulp, and both types are nor-
mally dumped in landfill, rivers or unregulated dumping grounds 2.1. Material and preparation
(Osma et al., 2007). There is also a growing industry of mass pro-
duction of clear banana juice (Kibazohi et al., 2017). This creates 2.1.1. BSFL
an extra need for sustainable management systems to utilise the The BSFL used in this study were obtained from a colony that
nutrients from the banana waste. has been in operation since 2015 at the Department of Energy
Use of insects for treatment of organic wastes is gaining increas- and Technology, SLU. The BSFL were reared on chicken feed (Gran-
ing interest, as it uses organic solid wastes as a resource to produce ngården Hönsfoder Start, metabolisable energy content of
valuable products (Čičková et al., 2015). Use of BSFL for organic 11.2 MJ kg
1, 80% moisture) for around 5 d. They were then trans-
waste treatment has the potential to add value to non-utilised ferred to the banana peel substrates and pre-treatments described
organic wastes and also to act as an additional income generator below.
for waste managers (Lohri et al., 2017). This technology converts
organic waste efficiently and rapidly into protein-rich (40% dry
2.1.2. Banana peel
matter (DM) and fat-rich (30% DM) larvae suitable for use in ani-
Banana peel used in this study were of two types: 1) Peel from
mal feed (Newton et al., 1977; St-Hilaire et al., 2007; Stamer,
Musa acuminata, Cavendish bananas, referred to in this paper as
2015) and biodiesel production (Li et al., 2011; Zheng et al.,
dessert peel; and 2) peel from ripe Pisang Awak bananas used
2011), while the treatment residue is valuable fertiliser
specifically for juice production, referred to as juice peel. The des-
(Sheppard et al., 1994; Setti et al., 2019).
sert peel was collected every day in plastic bags at the Swedish
BSFL can be reared on different substrates, including animal
University of Agricultural Sciences (SLU) campus in Uppsala, Swe-
manures (Sheppard et al., 1994; Myers et al., 2008; Zhou et al.,
den. The juice bananas were bought in Mabibo market, Dar es Sal-
2013) pig liver, fish rendering waste and fruit waste (Nguyen
aam, Tanzania, and transported to SLU, where it was separated
et al., 2013), human excreta (Lalander et al., 2013; Banks et al.,
from edible flesh. The banana peel wastes collected were kept in
2014) and food waste (Diener et al., 2011; Nguyen et al., 2015).
the fridge and homogenised using a blender (Robot Coupe Blixer
Previous studies revealed that BSFL growth and waste conversion
4 V). They were then portioned and frozen at
20 °C until use.
into biomass depend on the type, quantity and quality/nutrient
composition of the substrate (Myers et al., 2008; Nguyen et al.,
2013; Nyakeri et al., 2017; Gold et al., 2018; Meneguz et al., 2.1.3. Pre-treated banana peel
2018). On substrates with imbalanced amounts of nutritional sub- Only the dessert peel was used in the pre-treatments of banana
stances such as protein, fats and fibre, BSFL grow at a slower rate peel as this was the only banana peel we found in Sweden where
and achieve lower body weight (Nyakeri et al., 2017). Food sub- the study took place. The pre-treatment methods tested are shown
strates with a low protein to carbon ratio, high amounts of fat in Table 1. In all experiments, after the defined pre-treatment per-
and fibre have also been shown to inhibit BSFL growth rate and iod the pre-treated peel was mixed well, weighed, portioned into
biomass yield because of poor digestibility and low nutrient utili- small daily feeding portions and frozen at
20 °C until use.
sation (Tomberlin et al., 2009; Tschirner and Simon, 2015;
Lalander et al., 2019). Growth is particularly poor on homogeneous 2.1.4. Chemicals
wastes with a high hemicellulose and lignin content, such as dairy The chemical tested as pre-treatment was 24% technical ammo-
manure and banana peel (Nyakeri et al., 2017; Rehman et al., nia (aq) (Nitor, Sweden) and analytical grade concentrated sul-
2017a; Kumar et al., 2018). phuric acid (Fisher Chemicals, Uppsala, Sweden) was used for
A few studies have been conducted on methods to improve the neutralising the pH after the pre-treatment period.
biodegradability/digestibility of substrate in order to increase con- All chemicals and reagents used in the nutrient analyses were of
version by BSFL (Yu et al., 2011; Li et al., 2015; Rehman et al., analytical grade and were purchased from Sigma-Aldrich (Stein-
2017b). One suggested method is to improve the nutrient balance heim, Germany). These were: 80% ethanol, 1% methanol, 7.5%
by mixing waste with other substrates, e.g. rice straw with glucose sodium hydrogen carbonate, chloroform, D-glucose, anthrone
and restaurant waste (Zheng et al., 2012; Li et al., 2015) or dairy reagent, Folin-Denis reagent, tannic acid, diethyl ether, 1.25% sul-
manure with soybean curd residue (Rehman et al., 2017b). Another phuric acid, 1.25% sodium hydroxide, Folin-Ciocalteau reagent, gal-
suggested method involves addition of microorganisms to the sub- lic acid, sodium potassium tartrate (Rochelle salt), Kjeldahl tablets,
strate, to improve the conversion efficiency and degradation of cel- concentrated sulphuric acid (NH3-free grade), sodium salicylate,
lulose and hemicellulose, especially in homogeneous wastes such as sodium nitroprusside and 0.15% sodium hypochlorite solution.
poultry manure (Yu et al., 2011). A combination of these two meth-
ods can also be applied, e.g. microbe treatment together with mixing 2.1.5. Microorganisms
restaurant waste and rice straw (Zheng et al., 2012). However, due to Isolation of BSFL gut bacteria: Instar 5 BSFL (reared on food
the high lignin-cellulose content in rice straw, using this as a co- waste) were collected and washed with deionised water, followed
substrate has been found to produce small final BSFL weight com- by rinsing with ethanol (70% EtOH). The disinfected larvae were
pared with using other organic wastes rich in protein and lipids manually chopped into pieces and immersed (4 larvae) in 10 mL
(Zheng et al., 2012; Li et al., 2015; Manurung et al., 2016). There is unselective microbial nutrient broth (National Veterinary Institute,
thus a need to identify inexpensive, simple and effective techniques Uppsala, Sweden) at 70 °C for 20 min. Upon cooling to room tem-
to enable treatment of homogeneous and fibrous wastes with BSFL. perature (RT), 1 mL broth was spread on taurocholate gelatine agar
The main aim of this study was to explore the possibility of (TGA) (Miclev, Sweden) plates (£ 14 cm) and incubated at 37 °C
improving BSFL conversion of banana peel by using a pre- for 24 h. Ten visually different colonies were clean-spread on indi-
treatment method. Specific objectives were to: vidual TGA plates and incubated at 37 °C for 24 h. Single individual
colonies were then taken and heat-treated (70 °C, 20 min) once
a. Evaluate the efficiency of BSFL in degrading banana peel more in unselective nutrient broth (as described above). Upon
waste pre-treated with heat, ammonia or microorganisms. reaching RT, 10 mL of broth were spread onto a TGA plate and incu-
b. Identify the nutritional parameters that affect BSFL treat- bated at 37 °C for 24 h. Once again, single colonies were collected
ment efficiency. from each plate, clean-spread on TGA plates and incubated at
 A. Isibika et al. / Waste Management 100 (2019) 151–160 153

Table 1
Summary of the banana peel wastes used and the pre-treatments tested.

Experiment Banana peel type Code Pre-treatment 1 Pre-treatment 2

Treatment Conditions Treatment Conditions
1. Non-pre-treated
Juice peel Juice peel – – – –
Dessert peel Dessert peel – – – –
2. Microbial pre-treatment Dessert peel Trich7d Trichoderma reesei 7d – –
Trich14d 14d – –
Trich21d 21d – –
Rhiz7d Rhizopus 7d – –
Rhiz14d oligosporus 14d – –
Rhiz21d 21d – –
Bac7d Bacteria 7d – –
Bac14d 14d – –
Bac21d 21d – –
3. Non-protein nitrogen pre-treatment Dessert peel NH30.8 %N Ammonia 0.8%N, 7d – –
Dessert peel NH31 %N 1%N, 7d – –
Dessert peel NH30.8%N_14d Ammonia 0.8%N, 14d
4. Heating pre-treatment Dessert peel Heat Heat, 120 °C 1hr – –
5. Non-protein nitrogen and microbial pre-treatment Dessert peel NH30.8 % Ammonia 0.8%N, 7d Trichoderma reesei 7d
N + Trich7d
Dessert peel NH30.8 %N + Rhiz7d 0.8%N, 7d Rhizopus 7d
oligosporus
Dessert peel NH30.8 %N + Bac7d 0.8%N, 7d Bacteria 7d
6. Heating and microbial pre-treatment Dessert peel Heat + Trich7d Heat, 120 °C 1hr Trichoderma reesei 7d
Dessert peel Heat + Rhiz7d 1hr Rhizopus 7d
oligosporus
Dessert peel Heat + Bac7d 1hr Bacteria 7d

37 °C for 24 h. The clean-spread colonies were kept at 4 °C until microorganisms (Haddadin et al., 2009). It was also demonstrated
use. In total, 10 visually different bacterial colonies were observed. that solid-state fermentation with Trichoderma reesei increases the
No bacterial typing was performed for the isolated bacteria. Before digestibility of banana peel to monogastric animals (Katongole
use, one colony from each clean-spread plate was added to 10 mL et al., 2017). In this study, fungal species (Trichoderma reesei or Rhi-
unselective bacterial nutrient broth and incubated at 37 °C for zopus oligosporus) or BSFL gut bacteria were seeded amounting to
24 h. A mixture of bacteria isolated from the BSFL gut was used 1% (w/w) of dessert peel waste. The inoculated wastes were left
in this study (Lundgren, 2019). for 7 d, 14 d and 21 d at 28 °C. The concentration of the each inoc-
The fungi used in pre-treatment were Trichoderma reesei and ulation added was not measured but it approximated to 1.45  109
Rhizopus oligosporus, obtained from the Department of Molecular cells/mL.
Sciences, SLU. Both were pre-cultured on malt extract agar (MEA)
at 28 °C for 7 d. The spores were harvested by pouring sterile 2.2.3. Experiment 3 – Chemical pre-treatment
0.9% NaCl onto the MEA plates to wet the mycelium, inoculation This experiment evaluated the effect of chemical treatment by
loop removed mycelium from the agar surface. The spore suspen- addition of non-protein nitrogen to the peel. This was done to bal-
sion was transferred to a 50 mL centrifuge tube and diluted with ance the carbon-nitrogen ratio, because BSFL gut microorganisms
0.9% NaCl to 50 mL. The suspension was kept at 3–6 °C until use. can probably utilise and convert non-protein nitrogen into pro-
teins, as can other animals like ruminants (Amha, 2015). Ammonia
2.2. Experimental set-up solution with a concentration of 24.5% was added to dessert peel
waste. Two different amounts of ammonia solution were tested,
The study comprised six experiments (Table 1). The first exper- 0.8% and 1%, the former based on estimated amount of nitrogen
iment involved feeding the larvae with non-pre-treated dessert and fibre in the dessert peel and the latter calculated as 4% (w/
and juice peel, while the remaining five experiments involved w) ammonia solution per wet weight of dessert peel. The peel
feeding the larvae with dessert peel that had undergone one pre- pre-treated with ammonia solution was left to stand for 7 d at
treatment (Experiments 2–4) or two pre-treatments (Experiments 28 °C. The pH of the substrate was then neutralised to pH 7 by
5–6) prior to BSFL composting. addition of 95% concentrated sulphuric acid.

2.2.1. Experiment 1 – No pre-treatment 2.2.4. Experiment 4 – Heating

The first experiment involved feeding the BSFL with dessert The fourth experiment evaluated the effect of heat pre-
peel and juice peel to evaluate the effect of different banana peel treatment of dessert peel prior to BSFL composting. The peel was
types on BSFL growth. heated at 120 °C under 2 bar for 1 h. The hypothesis was that heat-
ing would degrade macromolecules in the peel, such as fibre-
2.2.2. Experiment 2 – Microbial pre-treatment bound polyphenols, to enhance availability of micronutrients for
The second experiment aiming to make the nutrients in waste easy extraction and utilisation for growth by the BSFL.
more available to the BSFL, evaluated the effect of microbial pre-
treatment for different periods. Fungi, e.g. Trichoderma spp., have 2.2.5. Experiment 5 – Combined chemical and microbial pre-treatment
been demonstrated to facilitate composting, as they break down This set of experiments assessed the effect of addition of non-
complex compounds like fibre into forms more available to other protein nitrogen followed by microbial pre-treatment prior to BSFL
154 A. Isibika et al. / Waste Management 100 (2019) 151–160

composting. Ammonia (0.8%) solution was added to dessert peel as 1989). Total phenolic content was analysed according to the
described in Experiment 3. The treated waste was left to stand at Folin-Ciocalteu method (Singleton et al., 1999), while soluble car-
28 °C for 7 d, and then the pH was neutralised using concentrated bohydrates were determined using the anthrone method (Allen
sulphuric acid as described for Experiment 3, but with further et al., 1989). Crude fibre content was analysed according to AOAC
addition of 1% (w/w) of the selected inoculate and left at 28 °C methods (AOAC, 1984), while total lipid content was determined
for 7 d. Hence the combined pre-treatment lasted a total of 14 d. using a chloroform-methanol mixture (Bligh and Dyer, 1959).

2.2.6. Experiment 6 – Heating and microbial pre-treatment 2.5. Calculations

The sixth experiment examined the effect of combined heating
and microbial pre-treatment of dessert peel. The heated peel To evaluate the efficiency of BSFL composting, material reduc-
(Experiment 4) was seeded with 1% (w/w) of the selected inoculate tion and waste to biomass conversion ratio were calculated.
(Trichoderma reesei / Rhizopus oligosporus / BSFL gut bacteria) and The percentage material reduction on a VS basis
left at 28 °C for 7 d before BSFL composting. (MR%VS Þ during BSFL composting was calculated as:
 
mres  TS%res  VS%res
2.3. BSFL composting MR%VS ¼ 1
 100 ð1Þ
minflow  TS%inflow  VS%inflow
Composting in this study is defined as the degradation of where mres and minflow is the mass, TS%res and TS%inflow is the total
organic material by BSFL under aerobic conditions for a specific solids content, and VS%res and VS%inflow is the total volatile solids
period of time resulting in a larval biomass and compost, nutrient
content in treatment residues and inflow substrate, respectively.
rich-product that can be used as fertilizer (Čičková et al., 2015;
The percentage biomass conversion ratio on a VS basis (BCR%VS Þ
Lalander et al., 2015). All feeding experiments were conducted in
was calculated as:
triplicate plastic containers (Smartstore classic 2, 21x17x11 cm3)
with plastic lids with a hole covered with netting for air circulation, mpp  TS%pp  VS%pp
BCR%VS ¼  100 ð2Þ
under the same controlled environment of 28 °C for 1 month. For minflow  TS%inflow  VS%inflow
feeding experiment, 200 larvae < 2 mm (around 10 d old) were
where mpp and minflow is the mass, TS%pp and TS%inflow is the total
added to each box, giving a density of 0.6 larvae cm
2. The larvae
were fed with the substrates every second or third day, at a feeding solids content, and VS%pp and VS%inflow is the total volatile solids
rate of 40 mg dry matter (DM) larva
1 d
1. The substrate portions content in pre-pupae and inflow substrate, respectively.
were thawed and brought to room temperature before feeding. The percentage survival rate (SR% Þ of the BSFL was calculated
Growth of the larvae was monitored three days per week by as:
 
weighing 10 randomly selected larvae. The larvae were then rinsed ppout
with water, dried on paper and placed back into the appropriate SR% ¼  100 ð3Þ
BSFLin
treatment container.
The BSFL composting units were run for one month to study the where ppout is the number of larvae that survived to the end of the
growth of BSFL on pre-treated and non-pre-treated banana peel. A experiment and BSFLin is the initial number of larvae used in the
sample of the treatment residues was collected once a week for experiment.
total volatile solids (VS) and pH determination. After sampling, a Volatile solids consumption (VScon: Þ during pre-treatments was
fresh thawed ration of substrate was added to the containers. After calculated as:
30 d of feeding, BSFL/Prepupae were manually picked from the VScon: ¼ VS%banana
VS%pre
treated ð4Þ
peel peel
residues. The DM and VS content of all harvested biomass of
BSFL/prepupae and of treatment residues were determined. For C/N ratio
the treatment residues, the pH was also measured. The carbon to nitrogen ration was calculated by dividing the
percentage organic carbon by the percentage of total nitrogen on
2.4. Physico-chemical and nutrient analyses dry matter basis. The the percentage of organic carbon was esti-
mated by dividing the percentage VS by 1.8 (Haug, 1980).
2.4.1. Dry matter and total volatile solids
The DM content of weighed samples during the experiments, 2.6. Statistical analysis
residues at the end of the experiments and pre-pupae was deter-
mined by drying the measured biomass at 85 °C for 48 h in a lab- Shapiro test of normality was conducted to verify normality of
oratory termaks oven. For VS measurements, the dried samples data. The survival rate was found to be non-normal and was con-
were heated to 200 °C for 2 h, to prevent sample losses due to heat- verted into log10 death rate that was found to be normally dis-
ing directly at very high temperatures, and then heated to 550 °C tributed (p > 0.05). Analysis of variance (ANOVA) with 95%
for 4 h. confidence interval was performed to identify statistically signifi-
cant differences in treatment response variables between different
2.4.2. pH treatments. When a significant difference was found, Tukey’s HSD
The pH of the feed substrates was determined using an InoLab (honestly significant difference) post-hoc test with 95% confidence
Laboratory pH meter. For this, 10 g sample were placed in a interval was performed. Principal component analysis (PCA) was
50 mL centrifuge tube, diluted to 50 mL with deionised water performed to find the variables that mostly contributed to the data
and left to stand for 1 h at room temperature before the pH variance, while multi-linear regression was used to evaluate the
readings. variables selected in the PCA. R statistical software (R Core Team,
2016) was used for statistical analysis and for graphical represen-
2.4.3. Nutrient analysis tation of the data.
Nutrient analysis was performed in order to evaluate the nutri- All experiments were conducted in triplicate with the exception
tional parameters influencing the efficiency of BSFL composting of of the microbial 14- and 21-day pre-treatments, which were
banana peels. Total nitrogen content was determined using a semi- planned and performed on single samples. The NH30.8 %N + Trich7d
micro Kjeldahl method and indophenol-blue method (Allen et al., and Heat + Trich7d were also performed on single samples, because
 A. Isibika et al. / Waste Management 100 (2019) 151–160 155

when performed on triplicates had very low survival rate and the between pre-treatments with heat pre-treatments having 80 as
experiment were repeated in singlets for verification of previous the highest ratio while the non-protein nitrogen pre-treatments
results. Singlet values were not included in ANOVA analysis, but having the lowest ratio (13–16). The pH varied greatly among
were included in multi-linear regression analysis. Nutrition analy- the different pre-treatments. Juice peel had the lowest pH values
sis were not conducted on all replicates due to lack of material. The (around 4), while Trich21d pre-treatment of dessert peel gave the
nutritional analysis was conducted in order to find the substrate highest pH, ranging from 8 to 9.
properties that contributed the most to the variance in treatment
response values. The values of triplicates and doublets from the
3.2. Effect of pre-treatment on the nutrient content of banana peel
nutrition analysis were averaged prior to the multi-regression
analysis. The failed NH30.8 %N + Trich7d and Heat + Trich7d triplicate
The heating pre-treatment and Heat + Bac7d gave the highest
experiments were not included in the statistical analysis.
amounts of total phenolic compounds (3 mg gallic acid equiva-
lents (GA) 100 g
1 extract), while the other pre-treatments
gave  1 mg GA 100 g
1 extract (Table 3). Juice peel had different
3. Results
amounts of phenolic compounds and tannins (>0.1 mg GA
100 g
1 extract, 2% tannins) than dessert peel (1 mg GA 100 g
1
3.1. Banana peels and properties of pre-treated peel
extract, 0.5% tannins). Trich14d gave the lowest amount of total
phenolic compounds (0.05 mg GA 100 g
1 extract). The heating,
Dry matter content and VS content differed significantly
Heat + Rhiz7d and NH30.8 %N pre-treatments gave the largest
between juice peel and dessert peel (Table 2). Juice peel had a
amounts of tannin, >2%.
30% DM content and 97% VS content on DM basis, while dessert
All the pre-treatments with addition of ammonia solution gave
peel had a DM content of 13% and a VS content of 85% on a
total nitrogen content of >2%, with the exception of NH30.8 %N -
DM basis.
+ Trich7d, which had 1% total nitrogenNH30.8 %N gave the highest
Pre-treatments of dessert peel altered DM content and total VS
amount (4%), while Rhiz21d gave the lowest (0.2%). Soluble car-
content. The pre-treatments that resulted in the highest DM con-
bohydrates, crude fibre and total lipids were not significantly dif-
tent was Heat + Bac7d (22%). The lowest DM content (8%) was
ferent between the pre-treated and non-pre-treated dessert peel.
observed for Heat + Trich7d and all longer pre-treatments of 14 d
and 21 d. Bac21d resulted in the highest amount of VS (88%).
Pre-treatments resulting in lower VS content (73–78%) were 3.3. BSFL composting efficiency
Heat + Bac7d, Heat + Trich7d, NH30.8 %N + Rhiz7d and 14 d and 21 d
microbial pre-treatments except for Bac21d which had 88% VS, Without any pre-treatment, BCR%VS (7% VS) and the final larvae
the highest content. The carbon to nitrogen ratio was different weight (134 mg larva
1) were higher for BSFL composting of des-
between the two banana peels, juice peel having a higher carbon sert peel than for BSFL composting of juice peel (0.9% VS and
to nitrogen ratio of 62.6 while dessert peel had 53.8. The carbon 33 mg larva
1, respectively). The latter were the lowest values
to nitrogen ration was altered by the pre-treatments and varied observed in all experiments (Table 4).

Table 2
Physico-chemical characteristics of peels before (experiment 1) and after pre-treatments (experiments 2–6). In cases when triplicates were conducted, values are presented as
mean (n = 3) ± standard deviation, in other cases the value of the singlet sample is presented. Same letter column wise represent no significant differences (p < 0.05). Only sample
conducted in triplicate are included in the statistical comparison. The juice peel were not included in the comparison as no pre-treatment was conducted on these peels.

Dry matter (%) Total volatile solids (% DM) C/N ratio pH Lowest - Highest
Experiment 1
Juice peel 29.9 ± 0.40 96.9 ± 0.22 62.6 3.1 ± 0.03–4.4 ± 0.06
Dessert peel 12.5 ± 0.34a,d 85.2 ± 0.08a,b,f 53.8 5.3 ± 0.11–6.6 ± 0.37
Experiment 2
Trich7d 9.8 ± 0.12b,c 81.6 ± 0.93b,e,d 75.6 4.2 ± 0.08 –5.8 ± 0.13
Trich14d 8.3 76.4 83.2 8.3–8.8
Trich21d 7.6 73.5 – 8.1–9.3
Rhiz7d 10.2 ± 0.06b,d,c 80.8 ± 0.56e – 6.6 ± 0.19–7.4 ± 0.19
Rhiz14d 8.3 76.4 49.3 8.2–8.8
Rhiz21d 8.0 77.1 78.5 6.8–8.5
Bac7d 11.8 ± 0.23a,b,c 78.6 ± 2.3c,e 76.6 7.0 ± 0.42–8.2 ± 0.15
Bac14d 6.8 71.4 57.5 6.5–8.7
Bac21d 7.9 87.8 – 7.3–8.8
Experiment 3
NH30.8 %N 12.6 ± 0.65a,e 84.4 ± 0.32a,d,f 12.67 6.3 ± 0.45–7.7 ± 0.05
NH31 %N 10.1 ± 2.4d,c 83.1 ± 2.2b,d,f 13.19 5.4 ± 0.25–7.7 ± 0.09
NH30.8%N_14d 12.2 85.3 16.34 5.7–7.0
Experiment 4
Heat 9.4 ± 0.43b 86.7 ± 0.82a 80.28 4.4 ± 0.04–5.0 ± 0.18
Experiment 5
NH30.8 %N + Trich7d 10.5 82.3 45.72 8.0–8.6
NH30.8 %N + Rhiz7d 10.0 ± 0.21b,c 78.2 ± 0.53e 17.38 8.5 ± 0.11–8.9 ± 0.03
NH30.8 %N + Bac7d 8.5 ± 0.25b 80.3 ± 0.34e 19.40 4.8 ± 0.08–7.8 ± 0.09
Experiment 6
Heat + Trich7d 8.2 76.5 – 7–8.41
Heat + Rhiz7d 12.2 ± 0.66c,d,e 86.0 ± 0.70a,f 47.78 4.8 ± 0.21–6 ± 0.30
Heat + Bac7d 21.8 ± 1.0 78.2 ± 1.0c,e 52.34 6.1 ± 0.21–7.5 ± 0.15

Different letters within columns indicate significant differences (p < 0.05). Only sample conducted in triplicate are included in the statistical comparison.
156 A. Isibika et al. / Waste Management 100 (2019) 151–160

Table 3
Nutritional characteristics of juice peel and of dessert peel before (Experiment 1) and after pre-treatments (experiments 2–6). In cases when samples from more than one
replicate were conducted, values are presented as mean ± standard deviation, in other cases the value of the singlet sample is presented.

Soluble Total Soluble Crude fibre Total Total phenolic compounds (mg Volatile solids consumption
tannins (% nitrogen (% carbohydrates (% (% DM) lipids (%) GA 100 g
1 extract) (percentage points)
DM) DM) DM)
Experiment 1
Juice peel 2.1 0.86 12.2 68 1.3 0.089 0
Dessert peel 0.53 ± 0.01 0.88 ± 0.26 13.6 ± 0.08 68.6 ± 2.8 1.4 ± 0.04 0.98 ± 0.05 0
(n = 2)
Experiment 2
Trich7d (n = 2) 0.60 ± 0.005 0.60 ± 0.12 13.7 ± 3.5 32.4 ± 2.9 1.4 ± 0.04 0.14 ± 0.12 4.0 ± 0.46
Trich14d 1.1 0.51 15.4 40.8 1.4 0.05 8.8
Rhiz7d (n = 3) 0.79 ± 0.43 0.91 ± 0.25 7.1 ± 3.0 61.5 ± 1.4 1.4 ± 0.08 0.76 ± 1.0 4.4 ± 0.56
Rhiz21d 1.2 0.24 44.9 61.9 1.3 0.07 8.1
Bac7d (n = 3) 1.3 ± 0.33 0.57 ± 0.18 6.0 ± 1.9 67.9 ± 15.1 1.3 ± 0.07 1.0 ± 0.10 6.6 ± 2.3
Bac14d 1.3 0.69 3.7 59.5 1.4 1.1 13.7
Experiment 3
NH30.8 %N 2.6 3.7 12.5 60.9 1.4 0.09 1.2
NH31 %N(n = 2) 0.80 ± 0.29 3.5 ± 0.09 10.1 ± 2.4 72.1 ± 2.9 1.4 ± 0.07 0.47 ± 0.30 1.2 ± 2.2
NH30.8%N_14d 1.3 2.9 5.8 61.7 1.4 1.1 0
Experiment 4
Heat(n = 3) 2.7 ± 0.59 0.60 ± 0.22 16.1 ± 5.4 63.0 ± 3.0 4.1 ± 0.27 3.5 ± 2.8 0
Experiment 5
NH30.8 % 1.6 1.0 11 54.3 1.3 0.50 2.9
N + Trich7d
NH30.8 % 0.85 ± 0.54 2.5 ± 0.09 7.9 ± 3.6 63.1 ± 1.5 1.4 ± 0.04 0.42 ± 0.16 7.0 ± 0.53
N + Rhiz7d
(n = 3)
NH30.8 % N + Bac7d(n = 2) 1.1 ± 0.02 2.3 ± 1.1 12.0 ± 0.06 63.7 ± 1.5
1.4 ± 0.05 0.74 ± 0.98 5.1 ± 0.16
Experiment 6
Heat + Rhiz7d 3.1 1.0 12.6 62.4 3.6 0.20 0
Heat + Bac7d 1.9 ± 0.48 0.83 ± 0.02 11.3 ± 1.1 55.6 ± 1.9 2.8 ± 1.9 3.1 ± 1.1 6.5 ± 1.0
(n = 2)

Table 4
Process parameters in BSFL composting of peels before (experiment 1) and after pre-treatments (experiments 2–6). In cases when triplicates were conducted, values are
presented as mean (n = 3) ± standard deviation, in other cases the value of the singlet sample is presented. Same letter column wise represent no significant differences (p < 0.05).
Only sample conducted in triplicate are included in the statistical comparison.

Final larval weight (mg larva
1) Survival rate (%) Death rate (log10 %)i Biomass conversion ratio (% VS) Material reduction (% VS)
Experiment 1
Juice peel 33 ± 2a 97.7 ± 1.9 0.65 ± 0.28a 0.87 ± 0.09a 70.2 ± 3.2a
Dessert peel 134 ± 3b,c 91.5 ± 6.8 0.80 ± 0.45a,b 7.2 ± 1.2b,c,e 48.9 ± 17.4a,c
Experiment 2
Trich7d 169 ± 60b,d,c 90.8 ± 4.5 0.91 + 0.27a,b 8.2 ± 1.4c,d 36.3 ± 2.9b,c
Trich14d 199 93.5 0.81 11.6 35.2
Trich21d 226 60.5 1.60 9.5 35.1
Rhiz7d 213 ± 11d,e,f 59.3 ± 18.1 1.6 ± 0.20b 6.7 ± 1.9b,c,d,e 32.3 ± 3.7b,c
Rhiz14d 220 99.5
0.3 15.0 38.3
Rhiz21d 119 98.0 0.3 8.0 38.5
Bac7d 194 ± 15b,d 89.2 ± 7.7 0.89 ± 0.51a,b 8.1 ± 1.0c,d,e 51.0 ± 1.2a,c
Bac14d 152 97.0 0.48 14.5 26.5
Bac21d 151 94.0 0.79 9.9 36.0
Experiment 3
NH30.8 %N 177 ± 23b,d 97.7 ± 2.4 0.19 ± 0.5a 7.1 ± 0.5b,c,e 40.3 ± 7.9c
NH31 %N 176 ± 44b,e 89.3 ± 9.4 0.69 ± 0.86a,b 9.6 ± 3.9c,d,e 25.6 ± 17.5b,c
NH30.8%N_14d 140 97.0 0.19 ± 0.50a,b 6.0
Experiment 4
Heat 113 ± 21b 66.3 ± 7.8 1.5 ± 0.10c,b 3.5 ± 0.8a,b 8.9 ± 0.6b
Experiment 5
NH30.8%N + Trich7d 231 51.0 1.7 6.1 44.8
NH30.8 %N + Rhiz7d 229 ± 22d,e 94.0 ± 2.3 0.76 ± 0.16a,b 14.8 ± 1.2d 47.8 ± 4.3a,c
NH30.8 %N + Bac7d 187 ± 32b 90.5 ± 10.0 0.81 ± 0.47a,b 11.0 ± 1.7d,e 29.1 ± 3.7b,c
Experiment 6
Heat + Trich7d 136 94.5 0.74 7.4 24.0
Heat + Rhiz7d 105 ± 11a,c,f 74.0 ± 16.2 1.3 ± 0.36a,b 2.6 ± 0.10a,b 25.7 ± 3.3b,c
Heat + Bac7d 143 + 12b,f 88.0 ± 8.0 1.0 ± 0.29a,b 3.3 + 0.43b 66 + 1.0a
 A. Isibika et al. / Waste Management 100 (2019) 151–160 157

Most of the pre-treatments improved BCR%VS of dessert peel ticides used, degree of ripeness and post-harvest handling. Larvae
compared with the non-pre-treated peel, except for Rhiz7d, growth and biomass conversion ratio were much higher on the
NH30.8%N_14d, NH30.8%N + Trich7d, heat, Heat + Bac7d and Heat non-pre-treated dessert peel than on the non-pre-treated juice
+ Rhiz7d. The Bac7d, Heat + Rhiz7d and heat pre-treatments resulted peel. Previous studies on BSFL composting of banana peel have
in lower larval weights (<120 mg larva
1) and BCR%VS (2–4% VS), reported a final larvae weight of 55 mg larva
1 (Nyakeri et al.,
with exception of Bac7d with 8% VS and addition of Heat + Bac7d 2017). Final weight of the larvae in the present study was lower
with 3% VS. BSFL reared on dessert peel subjected to a combination on juice peel (30 mg larva
1) than on dessert peel (134 mg larva
1).
of microbial and ammonia pre-treatments developed rapidly and However, the BCR%VS values obtained for non-pre-treated juice and
weighed more than those reared on non-pre-treated dessert peel, dessert peel in this study (0.9 and 7.2% VS) were lower than that
and had a final larval weight >190 mg larva
1 and BCR%VS ranging reported in the previous study (11% VS) (Nyakeri et al., 2017).
from 6 to 14% VS. Higher BCR%VS values were achieved with the The reason for this could be the differences in the banana peel used
14 d long pre-treatments: 12% VS for Trich14d, 15% VS for Rhiz14d and in the BSFL composting set-up, as it has been demonstrated
and 15% VS for Bac14d, while that for the 7 d and 21 d pre- that factors such as larval density and larval feeding rates affect
treatments was less than 10% VS. None of the substrate properties process efficiency (Parra Paz et al., 2015).
measured was found to contribute to the material reduction. Sim-
ilarly, the survival rate did not change considerably when rearing
the larvae on differently pre-treated peel except for NH30.8%N -
4.2. Efficiency of BSFL composting with microbial pre-treatments
+ Trich7d where SR% was considerably lower than in the other pre-
treatments (Table 4).
Microbial pre-treatment of dessert peel yielded larger larvae
(>150 mg larva
1) than produced on the non-pre-treated dessert
i) The ANOVA analysis was conducted on the log10 death rate
peel (134 mg larva
1), with the exception of larvae in the pre-
in order to comply with the test requirements of normality.
treatment with Rhiz21d (190 mg larva
1). Trichoderma reesei treat-
ment also improved peel digestibility in monogastric animals
The PCA results demonstrated that BCR%VS was the only BSFL
(Katongole et al., 2017). Many types of yeast, bacteria and fungi,
composting parameter affected by the pre-treatments and that it
when applied to organic waste, produce enzymes that degrade
correlated with the nutritional parameters (Table 5, Fig. 1). Final
the cellulose (the enzyme cellulase) and tannins (tannase) in the
larval weight, material reduction and survival rate were not
waste (Katongole et al., 2017). Furthermore, microbial pre-
affected significantly by the pre-treatments. A correlation between
treatment is reported to increase the lactic acid bacteria content
substrate VS and BCR%VS was observed in this study, with the larger
in banana peel, which when consumed by animals serves as a feed
the substrate VS, the lower the ratio (Table 5, Model 1.0). It was
supplement and improves the intestinal microbial balance. Accord-
also observed that the higher the substrate VS consumption during
ing to those authors, the increase in final larvae weight due to
pre-treatments, the higher the BCR%VS value achieved (model 2.0).
microbial pre-treatment of banana peel might be explained by
The nitrogen-enriched treatments had higher BCR%VS , while the
increased digestibility and utilisation of the nutrients in the peel
heated samples had lower values. On including both VS consump-
caused by either improved availability or the presence of lactic acid
tion during pre-treatments and the nitrogen content (Table 5,
bacteria. Similar microbial effects could explain the improved
model 2.1), the adjusted R2 value increased from around 0.3 to over
growth and development of BSFL fed on poultry manure inoculated
0.6. The adjusted R2 value increased to almost 0.8 when the sub-
with bacteria from the gut of BSFL (Yu et al., 2011). Comparing the
strate content of tannins was also included (Table 5, model 2.2).
7 d, 14 d and 21 d microbial pre-treatments in the present study,
the impact on biomass conversion ratio suggests that 14 d might
be closest to the optimum pre-treatment time for the microorgan-
4. Discussion
isms to achieve maximum degradation of the substrates, facilitat-
ing larval assimilation of the released nutrients (Table 3, Fig. 1).
4.1. BSFL composting efficiency with non-pre-treated banana peel
Increasing the pre-treatment time to 21 d resulted in lower
BCR%VS , possibly because the microorganisms themselves con-
Although both juice peel and dessert peel were from ripe bana-
sumed the nutrient required by the BSFL.
nas, they differed in their composition (Tables 2 and 3). The differ-
ences might have been due to the different varieties/cultivars of
bananas, geographical locations and soils at the growing sites, pes-
4.3. Efficiency of BSFL composting with chemically pre-treated
substrates
Table 5
Model strength (adjusted R2) and F-test significance value (p) for the evaluated
models. The BSFL grown on nitrogen-enriched dessert peel were larger
and BCR%VS was higher than found for non-pre-treated dessert peel
Dependent Predictors Adjusted
variable R2
(Table 4). This could be due to the overall higher nitrogen level and
lower carbon to nitrogen (C/N) ratio, which may improve BSFL con-
Model Biomass Substrate VS 0.438***
version efficiency (Rehman et al., 2017b). However, it has been
1.0 conversion ratio
Model Biomass VS consumption 0.331** demonstrated that low C/N ratio is not enough for good conversion
2.0 conversion ratio of substrate into larval biomass, and that the larval feeding dose of
Model Biomass VS consumption, nitrogen 0.558*** protein and VS is important (Lalander et al., 2019). The high final
2.1 conversion ratio
larvae weights and BCR%VS observed in the ammonia pre-
Model Biomass VS consumption, nitrogen, tannin 0.788***
2.2 conversion ratio
treatments suggest that the BSFL process may be able to convert
Model Biomass VS consumption, nitrogen, tannin, 0.795*** non-protein nitrogen to protein (Tables 2 and 3). Ruminant ani-
2.3 conversion ratio phenolic compounds mals are also known to convert non-protein nitrogen into protein,
Significance level:
since fermentation bacteria in the rumen can synthesis amino
**
p < 0.01. acids from non-protein nitrogen (Andrade-Montemayor et al.,
***
p < 0.001. 2009; Tadele and Negassie, 2015).
158 A. Isibika et al. / Waste Management 100 (2019) 151–160

Fig. 1. Correlation between waste to biomass conversion ratio (BCR%VS ) and (a) initial substrate VS (% of DM); (b) VS consumption during the pre-treatment (percent points);
(c) predicted BCR values for Model 2.1, taking into account VS consumption during pre-treatment and substrate nitrogen content; and (d) predicted BCR values of Model 2.2,
taking into account VS consumption during pre-treatment and the substrate content of nitrogen and soluble tannins. The points circled in red represent 14-day pre-
treatments, while those circled in blue represent 21-day pre-treatments.

4.4. Efficiency of BSFL composting with heat pre-treatments forms more available to the BSFL (demonstrated by the VS con-
sumption during the pre-treatments).
Heating the banana peel seemed to increase the amounts of tan-
nins, which could explain the lower BCR%VS (4% VS) and larvae
with low final weight (Fig. 1). Interestingly, similar findings in a 4.6. Efficiency of BSFL composting with heating pre-treatment and
herbivore study concluded that tannins, among other substances, chemical pre-treatment combined
are anti-nutritional parameters in food substrates that result in
lower weight gain and lower efficiency of utilisation of dietary The non-pre-treated dessert peel yielded BSFL with smaller
dry matter in herbivores (Butler, 1992). final larval weight compared to the combined heat + chemical
pre-treatments with exception of Heat + Rhiz7d. All the combined
heat + chemical pre-treatments gave lower waste to biomass con-
version ratio, except for Heat + Trich7d (7.4% VS). The low BSFL
4.5. Efficiency of BSFL composting with microbial pre-treatment and responses in these treatments are likely to have been caused by
chemical pre-treatment combined the observed high amounts of tannins from the heating pre-
treatment. The higher BCR%VS for Heat + Trich7d probably means
The BSFL grew larger (231, 229, 187 mg larva
1) on dessert peel that Trichoderma reesei was able to break the compounds present
pre-treated with ammonia, followed by microbial pre-treatment. in the peel following heat pre-treatment into more readily avail-
These combined pre-treatments also yielded a BCR%VS of 11% able nutrients that aided its conversion into BSFL biomass. This
(NH30.8 %N + Bac7d) to 15% (NH30.8 %N + Rhiz7d), although NH30.8%N - suggests that effect of the substrate VS on BSFL efficiency depends
+ Trich7d gave a value of 6.1%. The good BSFL responses to this com- on the composition of the VS under different pre-treatments,
bined method can be attributed to a more adequate protein dose rather than the amount of VS in the substrate. Pre-treatments
and to the pre-treatments degrading complex carbohydrates into may then have the ability to transform the substrate VS into either
 A. Isibika et al. / Waste Management 100 (2019) 151–160 159

more digestible products or products that still do not enhance sub- Bligh, E.G., Dyer, W.J., 1959. A rapid method of total lipid extraction and
purification. Can. J. Biochem. Physiol. 37, 911–917.
strate nutrient utilisation by the BSFL.
Butler, L.G., 1992. Antinutritional effects of condensed and hydrolyzable tannins. In:
Hemingway, R.W., Laks, P.E. (Eds.), Plant Polyphenols: Synthesis, Properties,
4.7. Feasibility of using these pre-treatments on BSF treatment of Significance. Springer, US, Boston, MA, pp. 693–698.
Byarugaba-Bazirake, G.W., 2008. The Effect of Enzymatic Processing on Banana Juice
banana peel waste
and Wine. Stellenbosch University, Stellenbosch.
Carter, B.A., Reeder, R., Mgenzi, S.R., Kinyua, Z.M., Mbaka, J.N., Doyle, K., Nakato, V.,
Pre-treatment increases waste to biomass conversion ratio, but Mwangi, M., Beed, F., Aritua, V., Lewis Ivey, M.L., Miller, S.A., Smith, J.J., 2010.
Identification of Xanthomonas vasicola (formerly X. campestris pv.
also requires additional infrastructure and a longer retention time,
musacearum), causative organism of banana xanthomonas wilt, in Tanzania,
of more than 7 d. The value of the increased productivity needs to Kenya and Burundi. Embo J. 59, 403.
be weighed against the additional operating and investment costs, CCAC, 2015. Municipal Solid Waste http://ccacoalition.org/en/resources/municipal-
especially for an industrial set-up. Operationally, to optimise pre- solid-waste-knowledge-platform, Climate and Clean Coalition Knowledge
Platforms. Retrieved on 10th April 2019.
treatment solutions larger batches should be treated for cost effi- Chowdhury, F.R., Nur, Z., Hassan, N., von Seidlein, L., Dunachie, S., 2017. Pandemics,
ciency, but BSFL treatment requires thin layers of substrate. pathogenicity and changing molecular epidemiology of cholera in the era of
Shorter pre-treatment may be the most economically viable global warming. Ann. Clin. Microbiol. Antimicrob. 16, 10.
Čičková, H., Newton, G.L., Lacy, R.C., Kozánek, M., 2015. The use of fly larvae for
option, even though it does not yield the highest output. Further organic waste treatment. Waste Manage. 35, 68–80.
process optimisation studies are required to find the best combina- Cogut, A., 2016. R20 regions of climate action:OPEN-BURNING-OF-WASTE-A-
tion of pre-treatment and full process design. GLOBAL-HEALTH-DISASTER. https://regions20.org/wp-content/uploads/2016/
08/OPEN-BURNING-OF-WASTE-A-GLOBAL-HEALTH-DISASTER_R20-Research-
Paper_Final_29.05.2017.pdf. Retrieved on 10th August 2018.
5. Conclusions Diener, S., May Studt Solano, N., Roa-Gutiérrez, F., Zurbrügg, C., Tockner, K., 2011.
Biological treatment of municipal organic waste using black soldier fly larvae.
Waste Biomass Valorizat.
Most pre-treatments of banana peel improved the waste to bio- Emmanuel, E.S., Anthony, M.M., Amelia, K.K., 2014. Biovalorization of Banana
mass conversion ratio compared with peel that was not pre- waste: Auditing and improvement of bio-methane production by fungal pre-
treatment. Journal of Chemical. Biol. Phys. Sci. Environ. Sci. 4, 818–833.
treated. The exception was heat pre-treatment and combined heat
Gold, M., Tomberlin, J.K., Diener, S., Zurbrügg, C., Mathys, A., 2018. Decomposition of
and microbial pre-treatment. Factors affecting BSFL composting biowaste macronutrients, microbes, and chemicals in black soldier fly larval
were the efficiency of fungi/bacteria pre-treatment and the com- treatment: A review. Waste Manage. 82, 302–318.
pounds present in the substrate after pre-treatment, such as total Haddadin, M.S.Y., Haddadin, J., Arabiyat, O.I., Hattar, B., 2009. Biological conversion
of olive pomace into compost by using Trichoderma harzianum and
nitrogen, tannins and total phenolic compounds. Pre-treated sub- Phanerochaete chrysosporium. Bioresour. Technol. 100, 4773–4782.
strates with high amounts of tannins and total phenolic com- Haug, R.T., 1980. Compost Engineering; Principles and Practice. Technomic
pounds produced small, low- weight larvae and resulted in low Publishing.
Hoornweg D.B.-T., Perinaz, 2012 What a Waste: A Global Review of Solid Waste
biomass conversion ratio. Chemical pre-treatment with non- Management. Urban development series; knowledge papers no. 15 World Bank,
protein nitrogen produced largest final larvae weight and high bio- Washington, DC.Ó.
mass conversion ratios. Applying a pre-treatment such as those Katongole, C.B., Bakeeva, A., Passoth, V., Lindberg, J.E., 2017. Effect of solid-state
fermentation with Arxula adeninivorans or Hypocrea jecorina (anamorph
evaluated in this study could thus enable BSFL composting of cur- Trichoderma reesei) on hygienic quality and in-vitro digestibility of banana
rently unsuitable nutrient-imbalanced and fibrous wastes. peels by mono-gastric animals. Livestock Sci. 199, 14–21.
Kibazohi, O., Kyamuhangire, W., Leonard Kaunga, D., Rokoni, C., 2017. Process
improvement for mechanical extraction of low-viscosity clear banana juice. Afr.
Acknowledgements J. Food Sci. 11, 291–295.
Komakech, A.J., Banadda, N.E., Kinobe, J.R., Kasisira, L., Sundberg, C., Gebresenbet, G.,
Trichoderma reesei and Rhizopus oligosporus were kindly provided Vinnerås, B., 2014. Characterization of municipal waste in Kampala, Uganda. J.
Air Waste Manag. Assoc. 64, 340–348.
by Volkmar Passot, Department of Molecular Sciences, SLU, Upp- Krishna, C., Chandrasekaran, M., 1996. Banana waste as substrate for a-amylase
sala. This study was funded by the Swedish International Develop- production by Bacillus subtilis (CBTK 106) under solid-state fermentation. Appl.
ment Cooperation Agency (Sida - Styrelsen för Internationellt Microbiol. Biotechnol. 46, 106–111.
Kumar, S., Negi, S., Mandpe, A., Singh, R.V., Hussain, A., 2018. Rapid composting
Utvecklingssamarbete) as part of the research Sustainable agricul-
techniques in Indian context and utilization of black soldier fly for enhanced
tural productivity, processing, and value chain for enhancing food decomposition of biodegradable wastes - A comprehensive review. J. Environ.
security in Tanzania. The authors wish to thank Prof. Amelia Kivaisi Manage. 227, 189–199.
of the University of Dar es Salaam and Moritz Gold of Eawag - Lalander, C., Diener, S., Magri, M.E., Zurbrugg, C., Lindstrom, A., Vinneras, B., 2013.
Faecal sludge management with the larvae of the black soldier fly (Hermetia
Swiss Federal Institute of Aquatic Science and Technology, for their illucens)–from a hygiene aspect. Sci Total Environ 458–460, 312–318.
valuable comments and inputs to this article. Lalander, C., Diener, S., Zurbrügg, C., Vinnerås, B., 2019. Effects of feedstock on larval
development and process efficiency in waste treatment with black soldier fly
(Hermetia illucens). J. Cleaner Prod. 208, 211–219.
Declaration of Competing Interest Lalander, C.H., Fidjeland, J., Diener, S., Eriksson, S., Vinnerås, B., 2015. High waste-to-
biomass conversion and efficient Salmonella spp. reduction using black soldier
fly for waste recycling. Agron. Sustainable Dev. 35, 261–271.
None. Li, Q., Zheng, L., Qiu, N., Cai, H., Tomberlin, J.K., Yu, Z., 2011. Bioconversion of dairy
manure by black soldier fly (Diptera: Stratiomyidae) for biodiesel and sugar
References production. Waste Manage. 31, 1316–1320.
Li, W., Li, M., Zheng, L., Liu, Y., Zhang, Y., Yu, Z., Ma, Z., Li, Q., 2015. Simultaneous
utilization of glucose and xylose for lipid accumulation in black soldier fly.
Allen, S.E., Grimshaw, H.M., Parkinson, J.A., Quarmby, C., 1989. Chemical Analysis of
Biotechnol. Biofuels 8, 117.
Ecological Materials. Blackwell Scientific Publications.
Lohri, C.R., Diener, S., Zabaleta, I., Mertenat, A., Zurbrugg, C., 2017. Treatment
Amha, Y.T.N., 2015. Use of different non protein nitrogen sources in ruminant
technologies for urban solid biowaste to create value products: a review with
nutrition: A review. International knowledge sharing plattform IISTE. Adv. Life
focus on low- and middle-income settings. Rev. Environ. Sci. Bio-Technol. 16,
Sci. Technol. 29.
81–130.
Andrade-Montemayor, H., García Gasca, T., Kawas, J., 2009. Ruminal fermentation
Lundgren, K., 2019. Impact of fly specific bacteria on fly larvae composting. UPTEC
modification of protein and carbohydrate by means of roasted and estimation of
W, 41.
microbial protein synthesis. Revista Brasileira de Zootecnia 38, 277–291.
Manurung, R., Supriatna, A., Esyanthi, R.R., Putra, R.E., 2016. Bioconversion of rice
AOAC, 1984. Official Methods of Analysis. Association of Official Analytical
straw waste by black soldier fly larvae Hermetia illucens L.): Optimal feed rate
Chemists. AOAC, Arlington.
for biomass production. J. Entomol. Zool. Stud. 4, 1036–1041.
Banks, I.J., Gibson, W.T., Cameron, M.M., 2014. Growth rates of black soldier fly
Meneguz, M., Schiavone, A., Gai, F., Dama, A., Lussiana, C., Renna, M., Gasco, L., 2018.
larvae fed on fresh human faeces and their implication for improving sanitation.
Effect of rearing substrate on growth performance, waste reduction efficiency
Trop. Med. Int. Health 19, 14–22.
160 A. Isibika et al. / Waste Management 100 (2019) 151–160

and chemical composition of black soldier fly (Hermetia illucens) larvae. J. Sci. Sheppard, D.C., Newton, G.L., Thompson, S.A., Savage, S., 1994. A value added
Food Agric. 98, 5776–5784. manure management system using the black soldier fly. Bioresour. Technol.,
Myers, H.M., Tomberlin, J.K., Lambert, B.D., Kattes, D., 2008. Development of black 275–279
soldier fly (Diptera: Stratiomyidae) larvae fed dairy manure. Environ. Entomol. Singleton, V.L., Orthofer, R., Lamuela-Raventós, R.M., 1999. [14] Analysis of total
37, 11–15. phenols and other oxidation substrates and antioxidants by means of folin-
Nagarajan, R., Thirumalaisamy, S., Lakshumanan, E., 2012. Impact of leachate on ciocalteu reagent. In: Methods in Enzymology. Academic Press, pp. 152–178.
groundwater pollution due to non-engineered municipal solid waste landfill St-Hilaire, S., Sheppard, C., Tomberlin, J.K., Irving, S., Newton, L., McGuire, M.A.,
sites of erode city, Tamil Nadu, India. Iranian J. Environ. Health Sci. Eng. 9, 35. Mosley, E.E., Hardy, R., Sealey, W., 2007. Fly prepupae as a feedstuff for rainbow
Newton, G.L., Booram, C.V., Barker, R.W., Hale, O.M., 1977. Dried hermetia illucens trout (Oncorhynchus mykiss). J. World Aquacult. Soc., 59–67
larvae meal as a supplement for swine. J. Anim. Sci. 44, 395–400. Stamer, A., 2015. Insect proteins—a new source for animal feed. The use of insect
Nguyen, T.T., Tomberlin, J.K., Vanlaerhoven, S., 2013. Influence of resources on larvae to recycle food waste in high-quality protein for livestock and
Hermetia illucens (Diptera: Stratiomyidae) larval development. J. Med. aquaculture feeds is held back largely owing to regulatory hurdles 16, 676–80.
Entomol. 50, 898–906. Tadele, Y., Negassie, A., 2015. Use of Dif-ferent non-protein nitrogen sources in
Nguyen, T.T.X., Tomberlin, J.K., Vanlaerhoven, S., 2015. Ability of black soldier fly ruminant nutrition: A review. Scholarly J. Agric. Sci. 5, 100–105.
(diptera: stratiomyidae) larvae to recycle food waste. Environ. Entomol. 44, Tock, C.L.L., Lee, K.T., Tan, K.T., Bhatia, S., 2010. Renew. Sustain. Energ. Rev 14, 789–
406–410. 805.
Nyakeri, E.M., Ogola, H.J.O., Ayieko, M.A., Amimo, F.A., 2017. Valorisation oforganic Tomberlin, J.K., Adler, P.H., Myers, H.M., 2009. Development of the black soldier fly
waste material: growth performance of wild black soldier fly larvae (Hermetia (Diptera: Stratiomyidae) in relation to temperature. Environ. Entomol. 38, 930–
illucens) reared on different organic wastes. J. Insects Food Feed 3, 193–202. 934.
Osma, J.F., Herrera, J.L.T., Couto, S.R., 2007. Banana skin: A novel waste for laccase Tschirner, M., Simon, A., 2015. Influence of different growing substrates and
production by Trametes pubescens under solid-state conditions. Application to processing on the nutrient composition of black soldier fly larvae destined for
synthetic dye decolouration. Dyes Pigm., 32–37 animal feed. J. Insects Food Feed 1, 249–259.
Parra Paz, A.S., Carrejo, N.S., Gómez Rodríguez, C.H., 2015. Effects of larval density WorldBank, 2017. Solid Waste Management. http://www.worldbank.org/en/topic/
and feeding rates on the bioconversion of vegetable waste using black soldier urbandevelopment/brief/solid-waste-management. Retrieved on 17 October
fly larvae Hermetia illucens (L.), (Diptera: Stratiomyidae). Waste Biomass 2017.
Valorizat. 6, 1059–1065. WorldBank, 2018 .Understanding poverty, Waste Management. http://
R Core Team (2016) R: A Language and Environment for Statistical Computing. In: R www.worldbank.org/en/topic/urbandevelopment/brief/solid-waste-
Foundation for Statistical Computing, Vienna, Austria. management. Retrieved on 17 October 2017.
Rehman, K.U., Cai, M., Xiao, X., Zheng, L., Wang, H., Soomro, A.A., Zhou, Y., Li, W., Yu, Yu, G., Cheng, P., Chen, Y., Li, Y., Yang, Z., Chen, Y., Tomberlin, J.K., 2011. Inoculating
Z., Zhang, J., hman et al. 2017a. Cellulose decomposition and larval biomass poultry manure with companion bacteria influences growth and development
production from the co-digestion of dairy manure and chicken manure by mini- of black soldier fly (Diptera: Stratiomyidae) larvae. Environ. Entomol. 40, 30–35.
livestock (Hermetia illucens L.). J. Environ. Manage. 196, 458–465. Zheng, L., Hou, Y., Li, W., Yang, S., Li, Q., Yu, Z., 2012. Biodiesel production from rice
Rehman, K.U., Rehman, A., Cai, M., Zheng, L., Xiao, X., Somroo, A.A., Wang, H., Li, W., straw and restaurant waste employing black soldier fly assisted by microbes.
Yu, Z., Zhang, J., hman et al. 2017b. Conversion of mixtures of dairy manure and Energy 47.
soybean curd residue by black soldier fly larvae (Hermetia illucens L.). J. Cleaner Zheng, L., Li, Q., Zhang, J., Yu, Z., 2011. Double the biodiesel yield: Rearing black
Prod. 154, 366–373. soldier fly larvae, Hermetia illucens, on solid residual fraction of restaurant
Setti, L., Francia, E., Pulvirenti, A., Gigliano, S., Zaccardelli, M., Pane, C., Caradonia, F., waste after grease extraction for biodiesel production. Renew. Energy, 75–79.
Bortolini, S., Maistrello, L., Ronga, D., 2019. Use of black soldier fly (Hermetia Zhou, F., Tomberlin, J.K., Zheng, L., Yu, Z., Zhang, J., 2013. Developmental and waste
illucens (L.), Diptera: Stratiomyidae) larvae processing residue in peat-based reduction plasticity of three black soldier fly strains (Diptera: Stratiomyidae)
growing media. Waste Manage. 95, 278–288. raised on different livestock manures. J. Med. Entomol 50, 1224–1230.