Black Soldier Fly (Hermetia Illucens)

foods

Article Black Soldier Fly (Hermetia illucens) Protein Concentrates as a Sustainable Source to Stabilize O/W Emulsions Produced by a Low-Energy High-Throughput Emulsiﬁcation Technology

Junjing Wang 1 , Morane Jousse 2, Jitesh Jayakumar 1, Alejandro Fernández-Arteaga 3 , Silvia de Lamo-Castellví 1 , Montserrat Ferrando 1 and Carme Güell 1,*

1 Departament d’Enginyeria Química, Universitat Rovira i Virgili, 43007 Tarragona, Spain; [email protected] (J.W.); [email protected] (J.J.); [email protected] (S.d.L.-C.); [email protected] (M.F.) 2 AgroParisTech, 75005 Paris, France; [email protected] 3 Departamento Ingeniería Química, Facultad Ciencias, Campus Universitario Fuentenueva, Universidad de Granada, 18071 Granada, Spain; [email protected] * Correspondence: [email protected]

Abstract: There is a pressing need to extend the knowledge on the properties of insect protein fractions to boost their use in the food industry. In this study several techno-functional properties of a black soldier fly (Hermetia illucens) protein concentrate (BSFPC) obtained by solubilization and precipitation at pH 4.0–4.3 were investigated and compared with whey protein isolate (WPI), a conventional dairy Citation: Wang, J.; Jousse, M.; protein used to stabilize food emulsions. The extraction method applied resulted in a BSFPC with a Jayakumar, J.; Fernández-Arteaga, A.; protein content of 62.44% (Kp factor 5.36) that exhibited comparable or higher values of emulsifying de Lamo-Castellví, S.; Ferrando, M.; activity and foamability than WPI for the same concentrations, hence, showing the potential for Güell, C. Black Soldier Fly emulsion and foam stabilization. As for the emulsifying properties, the BSFPC (1% and 2%) showed (Hermetia illucens) Protein the capacity to stabilize sunflower and lemon oil-in-water emulsions (20%, 30%, and 40% oil fraction) Concentrates as a Sustainable Source produced by dynamic membranes of tunable pore size (DMTS). It was proved that BSFPC stabilizes to Stabilize O/W Emulsions Produced by a Low-Energy sunflower oil-in-water emulsions similarly to WPI, but with a slightly wider droplet size distribution. ◦ High-Throughput Emulsification As for time stability of the sunflower oil emulsions at 25 C, it was seen that droplet size distribution Technology. Foods 2021, 10, 1048. was maintained for 1% WPI and 2% BSFPC, while for 1% BSFPC there was a slight increase. For https://doi.org/10.3390/ lemon oil emulsions, BSFPC showed better emulsifying performance than WPI, which required to foods10051048 be prepared with a pH 7 buffer for lemon oil fractions of 40%, to balance the decrease in the pH caused by the lemon oil water soluble components. The stability of the emulsions was improved when Academic Editor: Massimo Castellari maintained under refrigeration (4 ◦C) for both BSFPC and WPI. The results of this work point out the feasibility of using BSFPC to stabilize O/W emulsions using a low energy system. Received: 8 April 2021 Accepted: 9 May 2021 Keywords: black soldier fly; insect protein; techno-functional properties; membrane emulsification; Published: 11 May 2021 dynamic membrane

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affil- 1. Introduction iations. The Food and Agriculture Organization (FAO) of the United Nations reports that the global population is likely to grow up to nine billion by 2050 [1,2] inducing a dramatic increase of food demand. There is a general agreement that the conventional protein sources will not be able to provide the approximately 260 million tons of proteins required by 2050, Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. hence, there is a need to explore alternative protein sources that are both sustainable This article is an open access article and possess high nutritional value. Beans and legumes containing high protein content distributed under the terms and (e.g., beans 23.5%, lentils 36.7%, and soybean 41.1% in dry matter) involved in the human conditions of the Creative Commons diet since ancient times can be a good alternative, and soy is nowadays the most valuable Attribution (CC BY) license (https:// protein source for feed; however, to rely solely on plant proteins to fill the expected protein creativecommons.org/licenses/by/ gap puts huge pressure on the already pressured agricultural fields and raises serious 4.0/). environmental concerns [3–8].

Foods 2021, 10, 1048. https://doi.org/10.3390/foods10051048 https://www.mdpi.com/journal/foods Foods 2021, 10, 1048 2 of 21

Among the alternative protein sources, edible insects have drawn great attention in recent decades and FAO has claimed that edible insects are a good source of protein for dietary purposes [6]. Even though entomophagy, eating insects as foods, is being practiced by more than two billion people worldwide, most western developed countries have serious objections to consuming edible insects as a result of distaste and disgust about their nature and appearance [6,9–13]. Some studies showed that insect consumption was more acceptable in a masked way or when it was invisible in the food [11,14,15], which encouraged the integration of edible insects as milled powders and/or pastes in food products, as well as using their different functional fractions such as proteins, fatty acids, and chitin [16–19]. Moreover, in January 2021, the EFSA Panel on nutrition has adopted a scientific opinion on the safety of yellow mealworm as a novel food pursuant to Regulation (EU) 2015/2283. There are several applications of insect functional fractions, from a recent study on the production of nano-emulsions from insect oils with potential application as drug/bioactive delivery systems [20] to other studies that explore the antimicrobial activity of peptides purified from insect proteins [21]. There are several studies testing protocols for extract- ing/purifying the protein fraction from crude insect meal for black soldier fly (Herme- tia illucens), mealworm (Tenebrio militor), grasshopper (Schistocerca gregaria), honey bee (Apis mellifera), locust (Locusta migratoria), and cricket (Gryllodes sigillatus)[16,18,22–25]. Their procedures were based on the solubilization of protein in aqueous solution by adjust- ing pH values. Most of the studies on protein extraction/purification explore the potential functionality of the protein fractions. Gould and Wolf [18] used T. molitor protein as an emulsifier to stabilize sunflower oil-in-water emulsions produced by mechanical stirring; they obtained stable emulsions with droplet size smaller than emulsions stabilized with whey protein. Mishyna et al. [22] concluded that both S. gregaria and A. mellifera protein extracts have an emulsifying capacity comparable to whey protein; however, the insect source leads to some differences in the stability of the emulsions. Insect protein fractions have also been investigated with promising results as an ingredient in food formulations to replace meat in sausages [26] and meat batters [27]. Therefore, insect proteins, besides being used as a protein source, have the potential to be used in food and feed formulations to replace conventional proteins (such as dairy proteins) as gelling and emulsifying agents. This potential and the variations in the properties depending on the insect species and the development stage deserve more research studies to better understand their techno- functional properties, which will be highly relevant for food/feed industry regarding the preparation, processing, and storage of their edible insect food products. Black soldier fly (Hermetia illucens) is a true fly belonging to the family Stratiomyidae, whose larvae contain 42% crude protein and 29% fat on average in the dry matter [28]. Although the protein content in black soldier fly (BSF) larvae is lower than in insects from the orthoptera species such as adult locusts, grasshoppers, and crickets which were reported to have up to 77% protein content in the dry matter [29], the advantage of BSF is the survival rate and the efficiency of converting organic materials into their own biomass [4,5]. Promising data on techno-functional properties of BSF protein fractions has been reported by Bußler et al. [16], while Caligliani et al. [17] presented a systematic approach for BSF meal fractionation, and suggested the use of enzymatic treatment to tailor the properties of the protein fractions. Even though the data regarding the emulsifying activity of BSF protein extracts is scarce, its potential to replace conventional dairy proteins, such as whey protein, in food/feed formulations is worthy to study. As for the emulsification process, membrane emulsification is a widely used technique with advantages such as reduced mechanical stress, low energy consumption, and uniform droplet formation [30]. One of the reported drawbacks of membrane emulsification is the low productivity, which can prevent widespread application in the industry. Recently, dynamic membranes of tunable pore size (DMTS) have been applied to produce lemon oil emulsions [31] and double emulsions encapsulating carob polyphenols [32] in premix mode, that is, producing a coarse emulsion which is then refined by passing through the microstructured system. The DMTS system consists of a layer of glass microbeads Foods 2021, 10, 1048 3 of 21

supported by a nickel microsieve placed on the bottom of the membrane module. Because of this configuration, the DMTS system is easy to clean and re-use, and most importantly, requires low-energy input and yields high fluxes, showing its potential as a sustainable emulsification technology for industrial applications. This emulsification system has already been proved successful to produce sunflower oil-in-water emulsions stabilized with legume proteins [33], but it has never been tested for insect proteins. This research aimed to assess the potential of black soldier protein extracts to stabilize O/W emulsions produced by a low-energy high-throughput emulsification technology. To achieve this goal, first, a protocol for protein extraction was implemented to obtain BSF protein concentrate (BSFPC), second, relevant techno-functional properties (solubility, foaming capacity and foam stability, emulsifying activity, water/oil binding capacity, and interfacial tension) of BSFPC were obtained and compared to whey protein isolate, and third, sunflower and lemon oil emulsions were produced by the DMTS system and compared with whey protein. This study provides relevant data towards the use of more sustainable protein sources and technologies in food production.

2. Materials and Methods 2.1. Materials Partially defatted black soldier fly powder (BSF powder) was kindly provided by Hexafly (County Meath, Ireland). Sodium hydroxide pellets (NaOH, Chem-Lab NV, Zedel- gem, Belgium) and hydrochloric acid (37–38% HCl, J.T. Baker, Griesheim, Germany) were used for BSF protein extraction. Hexane (≥99%, Sigma-Aldrich, Poznań,Poland) was used for removing lipid fraction. The BCA (bicinchoninic acid) assay kit (Pierce Biotechnology, Thermo Scientific, Rockford, IL, USA) was used for protein quantification giving the results in bovine serum albumin (BSA) equivalent value. Note that BSF concentrations that are BSA eq % are given hereinafter as % for simplicity. Phosphate buffer (pH 7) was prepared using sodium phosphate dibasic heptahydrate (HNa2O4P7H2O, ACROS, Barcelona, Spain) and sodium phosphate monobasic monohydrate (H2NaO4PH2O, ACROS, Barcelona, Spain). Acetic buffers (pH 3 and 5) were prepared with sodium acetate (Sigma-Aldrich, USA) and acetic acid (96%, Panreac, Barcelona, Spain). Buffers for pH 9 and 11 were prepared by sodium hydroxide (Chem-lab, Zedelgem, Belgium) and sodium tetraborate (ACROS, Spain). Whey protein isolate (WPI) was purchased from Davisco Foods International, Inc. (97.6%, Lot.JE151-4-420, Eden Prairie, MN, USA). Sunflower oil used as the oil phase for the O/W emulsions was purchased from a local supermarket (Borges S.A., Tarragona, Spain) and lemon oil (24L120 Limón Aroma) was ordered from Dallant S.A., Barcelona, Spain.

2.2. Protein Extraction The protein extraction process was based on Zhao et al. [34] with some modifications. A total of 30 g of BSF powder was mixed with 150 mL of 0.25 M NaOH solution and the mixture was heated to 40 ◦C for one hour with constant agitation at 400 rpm on a magnetic stirrer (RCT ST, IKA, Staufen, Germany). The mixture was centrifuged (Meditronic 7000599, J.P. SELECTA, Barcelona, Spain) at 4490 rpm for 15 min and the lipid fraction on the top was carefully separated by pipets. The pellet was reserved for further extraction (twice more), while the pH of the supernatant was adjusted to 4.0–4.3 by adding 37% HCl and 1N HCl successively, followed by centrifugation (3750 rpm, 15 min) to obtain the precipitated proteins. After the centrifugation, the precipitated proteins were collected in plastic petri dishes and were kept at −60 ◦C until freeze-drying. The precipitated proteins were freeze- dried (LYOQUEST-85 PLUS, Telstar, Barcelona, Spain) for 24 h at 0.2 mbar with the plates heated to 20 ◦C. Freeze-dried samples were combined, ground, and defatted. Defatting was carried out by stirring 50 g of freeze-dried powder and 250 mL of hexane for one hour, after which powders were settled down and the hexane layer was decanted. This procedure was repeated until no color was observed in the hexane layer. The remaining hexane in the powders was evaporated in the fume hood for 2 days. Next, samples were collected and were kept in a desiccator at 4 ◦C until being used. Foods 2021, 10, 1048 4 of 21

The yield of the extraction process and the yield of the protein extraction in this study were deﬁned as Equations (1) and (2), respectively.

gram of BSFPC output %Extraction yield = × 100 (1) gram of BSF powder input

gram of protein in BSFPC output %Protein yield = × 100 (2) gram of protein in BSF powder input

2.3. Characterization of the BSF Powder and the BSFPC 2.3.1. Amino Acid Composition, Total Nitrogen, and Protein Content Total nitrogen and amino acid contents were analyzed by AGROLAB Ibérica S.L.U (Tarragona, Spain) based on Kjeldahl and the European Union Commission Regulation REG(UE) 152/2009, III, F: 2009-02 methods, respectively. Nitrogen to protein conversion factor, Kp, was calculated from the ratio of the sum of amino acid residue weights to nitrogen content [35]. The calculated Kp factor was used to estimate the protein content in the samples.

2.3.2. Fourier Transform Mid Infrared Spectroscopy (FTIR) The BSF powder and the BSFPC obtained after freeze-drying and defatting were analyzed by following the protocol of Mellado-Carretero et al. [36] to qualitatively determine the efﬁciency of the defatting process. For the acquisition of the spectral proﬁles, 4 mg of each sample was taken randomly and placed onto the sample stage of a portable spectrometer Cary 630 (Agilent Technologies Spain SL, Madrid, Spain), equipped with a single bounce ATR diamond crystal accessory and a deuterated triglycine sulfate (DTGS) detector. A pressure clamp was used to ensure optimal contact between samples and the diamond crystal. A background scan was extracted from every sample scan to prevent the effect of environmental changes. Spectra were acquired from 4000 to 800 cm−1 with 8 cm−1 of resolution using MicroLab PC software (Agilent Technologies SL, Madrid, Spain).

2.4. Techno-Functional Properties of the BSFPC 2.4.1. Protein Solubility Protein solubility was examined with the similar method as described in the literature [22]. As it is pH, temperature, and ionic strength dependent, the measurements were performed at room temperature and different pH buffers with the same molarity. A sample of BSFPC (0.2 g of powder) was dispersed into 10 mL of 0.2 M buffer (pH 3, 5, 7, 9, and 11) and stirred for 2 h. The mixtures were centrifuged (Biocen 22R, Orto Alresa, Madrid, Spain) at 3250 g for 20 min. The protein content in the supernatants was quantiﬁed using BCA assay kit. All the experiments were performed at least in duplicate. The protein solubility was calculated as Equation (3).

Soluble protein in the supernatant %PS = × 100 (3) Total protein content in the powder

2.4.2. Isoelectric Point (pI) Determination pI of soluble protein fraction from BSFPC was determined by zeta potential using Zetasizer Nano-ZS (Malvern Instruments, Worcestershire, UK). A total of 0.1% of BSFPC soluble protein solution was prepared at pH 7, and the pH was subsequently adjusted to 5.5, 5.0, 4.5, 4.0, and 3.5 by adding 1 N HCl and 0.01 N HCl gradually. Zeta potential of supernatant after centrifugation at 3250 g for 20 min was measured and plotted against pH. The isoelectric point set to the pH region where the Zeta potential was close to 0.

2.4.3. Water and Oil Binding Capacity Water binding capacity (WBC) and oil binding capacity (OBC) were analyzed by the method as reported in the literature [25]. Brieﬂy, for WBC, 0.5 g of powders were mixed Foods 2021, 10, 1048 5 of 21

with 2.5 mL of 0.2 M phosphate buffer (pH 7) and vortexed in a centrifugation tube for 60 s followed by centrifugation at 3250 g for 20 min at room temperature. The supernatant was decanted and the tube with the residual pellet was placed upside down on a ﬁlter paper for 60 min, to drain the residual non-bound water, before recording the weight. The WBC is calculated as shown in Equation (4),

g m − m WBC water = 1 0 × 100 (4) gDM mDM

where m0 is the initial weight of the sample, m1 is the weight of residual after 60 min, and mDM is the dry matter of the initial sample which can be calculated by measuring the weight changes of the fresh sample being kept in the oven (UN55, Memmert, Büchenbach, Germany) at 105 ◦C until no difference in weight is observed. As for OBC, similarly, 0.5 g powders were mixed with 2.5 mL of commercial sunﬂower oil and vortexed for 60 s twice with 5 min of pause in between. The rest of the procedure is identical to WBC analysis using Equation (5) for the calculation. Both WBC and OBC analyses were done in triplicate.

g m − m OBC oil = 1 0 × 100 (5) gDM mDM

2.4.4. Foaming Capacity (FC) and Foam Stability (FS) The foaming properties were analyzed following the literature [22] with minor mod- iﬁcations. A sample of 20 mL of 0.1% BSFPC solution prepared with 0.2 M pH 7 buffer was placed in a 50 mL plastic tube and subjected to vigorous rotor-stator homogenization (Ultra Turrax T18 digital, IKA, Staufen, Germany) at 1200 rpm for 2 min. The height of the foam layer after 10 s and 120 min was recorded. FC and FS (from experiments run in duplicate) were calculated using Equations (6) and (7), respectively [22],

H %FC = t × 100 (6) H0 FC %FS = 120 × 100 (7) FC0

where H0 is the initial height of protein solution in the tube, Ht is the height of generated foam after agitation, FC0 is the initial FC (after 10 s), and FC120 is the one obtained after 120 min.

2.4.5. Emulsifying Activity (EA) EA was evaluated at different protein concentrations (0.1%, 0.5%, and 1.0%) of both BSFPC and WPI using the method described by Purschke et al. [25]. Briefly, in a beaker 10 mL of protein solution and 10 mL of sunflower oil were homogenized using ULTRA TURRAX at 11,000 rpm for 30 s. An aliquot of 10 mL of the emulsion was transferred into a 15 mL scaled tube and centrifuged at 3250 g for 20 min at room temperature. Triplicates were performed for each sample. The height of the emulsified layer was noted, and the EA was calculated using Equation (8) [25],

H %EA = EL × 100 (8) HS

where HEL is the height of emulsified layer and HS is the total height of solution in the tube.

2.4.6. Interfacial Tension Analysis The pendant drop method was applied to monitor the interfacial tension changes of the interface of commercial sunﬂower oil and water phase by the tensiometer (CAM 200, KSV instrument, Espoo, Finland) coupled with the software CAM 2008. A capillary syringe Foods 2021, 10, 1048 6 of 21

filled up with water or protein solution (0.1% w/w) with a plastic cylindric tip (diameter 0.71 mm) was fitted in the tensiometer. The tip was immersed in the sunflower oil in a quartz cuvette. A pendant drop was generated by a capillary syringe with a controlled volume of 15 µL(±2 µL), and the interfacial tension was recorded immediately and every 5 min for 90 min or until it reached equilibrium. Reflective indices of 1.480 and 1.475 for sunflower oil and lemon oil were used for the calculation. Duplicates were performed for each water solution.

2.5. Premix Membrane Emulsiﬁcation 2.5.1. Preparation of O/W coarse emulsion A total of 150 g of O/W emulsion was prepared based on the formulation in Table1 . The whey protein solution was prepared by dissolving the desired amount of WPI in deionized water (or in 0.2 M phosphate buffer pH 7) and stirring for 2 h. The solution was kept in the fridge overnight for complete hydration. BSFPC solution was prepared by dissolving BSFPC powder and stirring for 1 h followed by the pH adjustment to 7 by 1 M NaOH. After further stirring for 2 h, the solution was kept overnight in the fridge. Protein concentration was quantiﬁed using the BCA assay kit after centrifugation twice at 3750 rpm for 15 min, and the required concentration of BSFPC solution was obtained by dilution with deionized water (or 0.2 M phosphate buffer pH7). The coarse emulsion was prepared by homogenizing the oil phase and water phase in a beaker using a rotor-stator homogenizer (Ultra Turrax T18 digital, IKA, Staufen, Germany) at 15,800 rpm for 2 min and at 15,400 rpm for 1 min with 30 s of pause every 1 min to avoid temperature rising.

Table 1. Composition of O/W emulsions.

Oil Phase Oil Fraction Emulsiﬁer in Water Phase Water Phase Medium 1% WPI Deionized water or 0.2 M 20 wt.%, 30 wt.%, or 40 wt.% 1% BSFPC Sunﬂower oil or lemon oil phosphate buffer pH7 1 2% BSFPC 1 0.2M phosphate buffer pH 7 was tested only for 20 and 40 wt.% lemon oil emulsions stabilized by 1% WPI, 1% BSFPC and 2%BSFPC.

2.5.2. Emulsiﬁcation with Dynamic Membranes of Tunable Pore Size (DMTS) In a cylindric module, 2 g of 94.2 µm glass microbeads (resulting in a height of 8.3 mm layer) were placed on top of a nickel sieve having a pore size of 289 × 13 µm (length × width) with a thickness of 120 µm as illustrated on Figure1. A pressure vessel was connected to the DMTS module where the coarse emulsions were pressurized by nitrogen to pass through the micron-sized interstitial voids formed by the glass microbeads layer to achieve droplet breakup. The emulsions were collected in an Erlenmeyer placed above an electronic balance to record the mass gain over time. The same procedure was repeated four times (five emulsification cycles in total) to further refine the emulsions. The interstitial void diameter (dv) of the channels formed by glass microbeads was 78.4 µm, calculated as Equation (9),

4ε dv = (9) (1 − ε)6/db

where db is the microbeads diameter (94.2 µm) and ε is the porosity (0.55), which can be calculated using the particle (ρp) and bulk (ρb) densities of glass microbeads using Equation (10). ρ ε = 1 − b (10) ρp

Transmembrane ﬂux, JDMTS, was calculated using Equation (11), φ JDMTS = (11) ρe A Foods 2021, 10, x FOR PEER REVIEW 7 of 23

where d is the microbeads diameter (94.2μm) and ε is the porosity (0.55), which can be Foods 2021, 10, 1048 calculated using the particle (𝜌) and bulk (𝜌) densities of glass microbeads using 7Equa- of 21 tion (10). ε=1 . (10) where φ is the mass ﬂow rate acquired from the mass/time data recorded with the electronic Transmembrane flux, 𝐽, was calculated using Equation (11), balance, ρe is the emulsion density, A is the effective surface area of the DMTS. The dynamic membrane was disassembled once the emulsiﬁcation was completed 𝐽 = , (11) (5 cycles) and the nickel sieve and glass microbeads were reused after cleaning based on thewhere protocol 𝜙 is the applied mass by flow Kaade rate et acquired al. [37]. Dishwashingfrom the mass/time detergent data and recorded ethanol with were the used elec- to cleantronic glass balance, microbeads. 𝜌 is the emulsion density, 𝐴 is the effective surface area of the DMTS.

FigureFigure 1.1.Premix Premix emulsiﬁcationemulsification andand dynamicdynamic membrane membrane of of tunable tunable pore pore size size emulsiﬁcation emulsification setup. setup.

2.5.3.The Particle dynamic Size andmembrane Distribution was disassembled once the emulsification was completed (5 cycles)The and particle the nickel size distribution sieve and glass of O/W microbea emulsionsds were was reused measured after after cleaning every based emulsifica- on the tionprotocol cycle applied by laser diffractionby Kaade et using al. [37]. Mastersizer Dishwashing 2000 (Malvern detergent Instruments, and ethanol Worcestershire, were used to UK).clean Particle glass microbeads. reflective indices were set to 1.480 and 1.475 for sunflower oil and lemon oil, respectively, and the dispersant reflective index was set to 1.330. Mean droplet size and2.5.3. droplet Particle size Size dispersion and Distribution can be calculated, and expressed as Sauter mean diameter d3,2 (Equation (12)) and the span factor (Equation (13)), respectively, The particle size distribution of O/W emulsions was measured after every emulsification cycle by laser diffraction using Mastersizer 2000− 1(Malvern Instruments, Worcester- ns ! shire, UK). Particle reflective indices were6 set tov 1.480i and 1.475 for sunflower oil and d3,2 = = ∑ (12) lemon oil, respectively, and the dispersantSv reflectivei=1 di index was set to 1.330. Mean droplet size and droplet size dispersion can be calculated, and expressed as Sauter mean diameter where S is the droplet surface area per unit volume, v is the volume fraction of droplets d3,2 (Equationv 12) and the span factor (Equation (13)), respectively,i in the ith size class of the discretized distribution, di is the mean droplet diameter in that class and ns is the number of size classes, 𝑑, = =∑ , (12) d90 − d10 where 𝑆 is the droplet surface area perδ = unit volume, 𝑣 is the volume fraction of droplets(13) d50 in the 𝑖 size class of the discretized distribution, 𝑑 is the mean droplet diameter in that class and 𝑛 is the number of size classes, where dx is the droplet diameter corresponding to x% volume on a cumulative droplet size distribution curve. 𝛿= , (13) 2.5.4. Zeta Potential where 𝑑 is the droplet diameter corresponding to x% volume on a cumulative droplet size Zetadistribution potential curve. of emulsions was measured using Zetasizer Nano-ZS (Malvern Instru- ments, Worcestershire, UK). Samples were diluted 100 times by deionized water. The same values2.5.4. Zeta of reflective Potential indices for the particle (sunflower or lemon oil droplets) and dispersant as those used in Section 2.5.3 were also applied here.

2.5.5. Analysis of Emulsion Stability Several 10 mL aliquots were collected in tubes for every freshly produced emulsion.

Emulsions were kept at room temperature (25 ◦C) and in refrigerator (4 ◦C) for 7 days. Droplet size distribution, as well as zeta potential were measured after 1, 3, and 7 days. Foods 2021, 10, 1048 8 of 21

2.6. Statistical Analysis The data described are mean ± standard deviation. Signiﬁcant differences between the groups were determined using ANOVA and Tukey test (p < 0.05).

3. Results and Discussion 3.1. Chemical Composition of BSF Powder and BSFPC 3.1.1. Amino Acids, Total Nitrogen, and Protein Content The protein content in the BSFPC was determined based on the amino acids and total nitrogen contents and compared with the BSF powder. Nitrogen-to-protein conversion factor (Kp) for insects is lower than the conventional value (Kp = 6.25) due to the existence of non-protein nitrogen compounds such as chitin [38]. Therefore, it is necessary to recalculate the Kp from amino acids contents as described in Section 2.3.1. The Kp values calculated shown in Table2 are slightly lower than the true values as they were calculated based on only 17 amino acids, which led to slightly lower amount of protein contents estimated. Most of the amino acid contents measured in BSFPC were higher than those in the BSF powder resulting in an increase of the Kp value from 4.71 to 5.36. It indicates that the extraction process effectively reduced non-protein nitrogen compounds [17]. Based on the calculated Kp values and the total nitrogen contents of the BSF powder and BSFPC, the protein contents were 39.00% and 62.44%, respectively, which are comparable to those (36.7% and 67.6%) reported by Janssen et al. [38] using pH 6 phosphate buffer in protein extraction. Results showed the essential amino acid contents in BSFPC meet the FAO recommendation of protein quality. Some amino acids were two (histidine and threonine) and three times (phenylalanine and tyrosine) higher than the recommended values, except for undetermined content of tryptophan (Table2). It also showed comparable protein quality to soybean, a plant sourced agriproduct. However, BSFPC contained slightly lower amount of leucine and lysine compared to casein protein. The protein extraction process resulted in a loss of total amount of sulphur amino acids (methionine and cysteine), which decreased from 35.4 mg/g protein in BSF powder to 28.2 mg/g protein in BSFPC, which is mainly due to the reduction of cysteine (from 0.43% DM in BSF powder to 0.36% DM in BSFPC). Even though cysteine is not classiﬁed as an essential amino acid, it is the key amino acid conforming disulphide bonds in high order protein structure [39]. The reduction of cysteine can be attributed to the strengthened protein’s tertiary and quaternary structure with multiple disulphide bridges of cysteines, which makes it difﬁcult to extract out [40]. In this study, a 35.8% of extraction yield and a 59.3% of protein yield were obtained which was higher than the protein yield reported by Janssen et al. [38] (17.1%) and in the range (47% to 91%) of which obtained by Caligiani et al. [17] using three different extraction methods. As for the studies on other insect species, 17% to 30.8% of extraction yields and 26.4% to 59.9% of protein yields were reported from T. molitor [16,18,41,42], Z. morio, A. diaperinus [43], A. domesticus, B. dubia [42], L. migratoria [23], S. gregaria, and A. mellifera [22]. Foods 2021, 10, 1048 9 of 21

Table 2. Amino acid composition, total nitrogen, and calculated Kp (nitrogen to protein conversion factor) value and protein content for BSF (black soldier ﬂy) powder and BSFPC (black soldier ﬂy protein concentrate), compared to soybean, casein, and FAO (Food and Agriculture Organization) recommendation for adults.

BSF Powder BSFPC BSF Powder BSFPC 2013 FAO Soybean Casein (% DM) (% DM) (mg/g Protein) (mg/g Protein) (mg/g Protein) (mg/g Protein) 5 (mg/g Protein) 5 Essential amino acid His 1.71 2.60 43.8 41.6 16 25 32 Ile 2.12 3.68 54.4 58.9 30 47 54 Leu 3.36 5.80 86.9 92.9 61 85 95 Lys 3.06 5.12 78.5 82.0 48 63 85 Met+Cys 3 1.38 1.76 35.4 28.2 23 24 35 Phe+Tyr 4 5.23 9.43 134.1 151.1 41 97 111 Thr 1.96 3.21 50.3 51.4 25 38 42 Val 2.71 4.79 69.5 76.7 40 49 63 Trp 1.87 1 nd 2 nd 2 Nd 2 6.6 11 14 Non-essential amino acid Asp 4.79 8.07 122.8 129.2 Glu 6.17 8.53 158.2 136.6 Ala 3.08 4.84 80.0 77.5 Arg 2.52 3.92 64.6 62.8 Gly 2.66 3.72 68.2 59.6 Pro 2.74 4.10 70.3 65.7 Ser 1.99 3.16 51.0 50.6 Kp 4.71 5.36 Total Nitrogen (%) 3 8.28 11.65 Protein content (%) 39.00 62.44 Extraction yield (%) 35.8 Protein yield (%) 59.3 1 value is referenced from Janssen et al. [38]; 2 not determined; 3 sum of sulphur amino acids (Cys and Met); 4 sum of aromatic amino acids (Phe and Try); 5 values from Young and Pellet [44]. Foods 2021, 10, 1048 10 of 21

3.1.2. Attenuated total Reﬂectance Fourier Transform Mid-Infrared Spectroscopy (ATR-FT-MIR) Figure2 shows the attenuated total reﬂectance FTIR raw spectra and Savitzky–Golay’s second derivatives (11 points) of BSF powder and BSFPC before and after defatting. Raw spectra and second derivatives of BSF powder and BSFPC before and after defatting show four important spectral regions: 2930–2850 cm−1, 1753 cm−1, 1630–1510 cm−1, and 1150– 1020 cm−1. These spectral regions have also been reported by other authors that have analyzed different insect species, P. succincta, C. roseapbrunner [45], T. molitor, A. diaperinus, G. sigillatus, A. domesticus, and L. migratoria [36]. Strong IR bands at 2924 cm−1, 2853 cm−1, −1 and 1753 cm linked to CH2 asymmetric stretching, CH2 symmetric stretching, and C=O stretching of lipids, respectively [46], are present in BSF powder and BSFPC before defatting − Foods 2021, 10, x FOR PEER REVIEW spectra. In the spectrum of BSFPC after defatting, the IR band at 1753 cm 110disappeared of 23

and the absorbance of the IR bands at 2924 cm−1, 2853 cm−1 signiﬁcantly decreased.

FigureFigure 2. 2. AttenuatedAttenuated total total reflectance reflectance Fourier Fourier transf transformorm mid midinfrared infrared spectroscopy spectroscopy (FTIR) (FTIR) raw raw spectra spectra(black line)(black of line) BSF of (black BSF (black soldier soldier fly) powder fly) powder and and BSFPC BSFPC (black (black soldier soldier fly fly protein protein concentrate) concen- before trate) before and after defatting and the corresponding Savitzky–Golay’s second derivatives (grey and after defatting and the corresponding Savitzky–Golay’s second derivatives (grey line). line). 3.2. Techno-Functional Properties 3.2. Techno-Functional Properties To determine the techno-functional properties of BSFPC and assess its potential appli- To determine the techno-functional properties of BSFPC and assess its potential ap- Foods 2021, 10, x FOR PEER REVIEW cations, solubility, WBC and OBC, FC and FS, EA and interfacial tension11 were of 23 analyzed plications, solubility, WBC and OBC, FC and FS, EA and interfacial tension were analyzed and summarized in Figures3 and4. and summarized in Figures 3 and 4.

3.2.1. Protein Solubility and Isoelectric Point (pI) As for the protein application in aqueous media, solubility is of importance. Figure 3a shows that solubility of BSF powder and BSFPC depends largely on pH value, with the highest solubilities (19.1% and 38.0%) at pH 11, and lowest solubility (10.5 % and 12.4 %) at pH 5. Isoelectric point (pI) of BSFPC was determined to be in the range of pH 4.0 to 4.5 (Figure 3a), which explained the lowest solubility observed at pH 5. This is in agreement with the range of protein pI (4–5) found in the literature of T. molitor [16,24,47], S. gregaria [22,24], A. domesticus [24], A. mellifera [22], and H. illucens [16]. Overall, the solubility values for BSF powder and BSFPC are comparatively lower than the ones reported for T. molitor [16,47], and S. gregaria and A. mellifera [22]. Nevertheless, the BSFPC presented an improved protein solubility compared to the BSF powder shown by the two-fold increase of solubility at pH 7–11. Purschke et al. [23] also observed the synergic effect of ionic FigurestrengthFigure 3. ( 3.a (1) (Proteina %) ProteinNaCl solubility to solubility 3 % NaCl) at pH at rangingbetween pH ranging from pH from 34 to and 11 3 pHand to 11 9zeta andon potentialprotein zeta potential solubility at pH ranging at pH for rangingT. from molitor from 3.5 to 3.5protein7; to (b 7;) WBC(b concentrate.) WBC (water (water binding Moreover,binding capacity) capacity) protein and and solubi OBC OBClity (oil(oil can bindingbinding be improved capacity) ofafter of BSF BSF enzymatic (black (black soldier soldier hydrol- fly) fly) powder powderysisand of BSFPC and A. domesticusBSFPC (black (black soldier and soldier L. flymigratoria fly protein protein powders concentrate). concentrate). [19,25]. Different None lowercaseof lowercase these strategies letters letters indicate were indicate sig- used significant nificant differences at the p < 0.05 level. indifferences this study at to the increasep < 0.05 the level. solubility of the BSFPC but they could be applied when solubility must be enhanced. Since some techno-functional properties such as foaming and 3.2.2. WBC and OBC emulsifying depend on the soluble protein fraction, based on the solubility results obtained,WBC protein and OBC solutions are critical at pH features 7 were preparedof food ingredients to conduct in foaming food processing and emulsifying and appli- tests cations.in the laterThey sections. are related to the ability of taking up and retaining water and oil, respectively, which directly affect the texture and the flavor of the products, especially in meat and bakery [47]. As for foods with high protein content, protein molecule structure, amino acid composition, pH, hydrophilicity, and hydrophobicity on protein surface are determining factors of WBC and OBC [24,47]. BSFPC had WBC of 2.2 g water/g DM that is higher than the reported 0.4 g water/g DM and 1.5 g water/g DM for T. molitor protein extract and enzymatic hydrolysate of L. migratoria protein [16,23], respectively, but lower than hexane defatted T. molitor powders (2.7 g water/g DM) [48] (Figure 3b). As for the OBC, the value obtained for BSFPC (1.1 g oil/g DM) was higher than the reported one for T. molitor protein extract [16], in the range of the ones for defatted T. molitor [48], and lower

than the one reported from enzymatic hydrolysate of L. migratoria protein [23]. The relatively higher WBC might be due to the higher protein content in the BSFPC which contains more hydrophilic groups to bind with water molecules. On the contrary, OBC can relate to the hydrophobic groups on the surface of protein molecules to bind with oil. Zielińska et al. [24] reported higher values of WBCs (2.18–3.95 gwater/gDM) and OBCs (1.98–3.33 goil/gDM) of protein extracts from T. molitor, L. migratoria, and A. domesticus than the ones found for BSFPC. The WBC and OBC of the BFS powder and BSFPC are comparable to plant-based flours such as wheat and rice, which were reported to have WBC from 1.4 to 1.9 gwater/gDM, and OBC from 1.5 to 1.9 goil/gDM, respectively [49]. Therefore, the BSFPC obtained in this study seems to be suitable in broad food applications entailing high protein content.

Foods 2021, 10, x FOR PEER REVIEW 12 of 23

the protein structure, the positive effect of carbohydrates on foaming was explained by Foods 2021, 10, 1048 11 of 21 Zielińska et al. [24] which might be also the case in this study due to the BSFPC contained 62.44% of protein and the remaining parts are possibly carbohydrates and soluble fibers.

FigureFigure 4. 4.(a)( aFC) FC (foaming (foaming capacity) capacity) and FS and (foam FS (foamstability) stability) among different among differentconcentration concentration of WPI of WPI (whey(whey protein protein isolate) isolate) and and BSFPC BSFPC (black (black soldier soldier fly protein ﬂy protein concentrate); concentrate); (b) EA (emulsifying (b) EA (emulsifying ac- activ- tivity)ity) compared compared WPI with with BSFPC. BSFPC. Different Different lowerc lowercasease letters letters indicate indicate significant signiﬁcant differences differences at at the thep < p 0.05< 0.05 level. level.

3.2.1.Regardless Protein Solubilityof the FC, 2% and WPI Isoelectric solution Pointshowed (pI) the highest FS at 120 min after foam generation (31.1%), followed by 0.1% BSFPC (25.2%) and 0.1% WPI (17.3%). Due to the As for the protein application in aqueous media, solubility is of importance. Figure3a higher concentration of protein in the 2% WPI solution, the viscoelastic film formed at the air-watershows that interface solubility [22,25] of was BSF more powder durable. and Ne BSFPCvertheless, depends BSFPC largelyis considered on pH to value,be suit- with the ablehighest for preparing solubilities food (19.1% products and based 38.0%) on foams. at pH 11, and lowest solubility (10.5% and 12.4%) at pH 5. Isoelectric point (pI) of BSFPC was determined to be in the range of pH 4.0 to 4.5 3.2.4.(Figure EA3 a), which explained the lowest solubility observed at pH 5. This is in agreement with the rangeEAs of ofWPI protein and BSFPC pI (4–5) at foundconcentrations in the literature of 0.1%, of0.5%,T. molitor and 1.0%[16 ,were24,47 assessed], S. gregaria as [22,24], describedA. domesticus at Subsection[24], A. mellifera 2.4.5. BSFPC[22], andat 0.1%H. illucenspresented[16 higher]. Overall, EA than the solubility0.1% WPI; valueshow- for BSF ever,powder both andwere BSFPC not able are to comparativelyemulsify all oil. A lower visible than top theoil layer ones was reported observed for afterT. molitor cen- [16,47], trifugation,and S. gregaria whichand wasA. more mellifera significant[22]. Nevertheless, for 0.1% WPI. the When BSFPC the presented protein concentration an improved protein wassolubility raised to compared 0.5%, the gap to the in EA BSF between powder WP shownI and BSFPC by the was two-fold reduced increase (51.7% and of solubility 54.2%, at pH respectively)7–11. Purschke and the et al.EA[ values23] also were observed nearly equal the synergic (59.2% and effect 60.0%) of ionicat 1.0% strength protein con- (1% NaCl to centration.3 % NaCl) It between is worth pHmentioning 4 and pH that 9 on0.5% protein BSFPC solubility was able to for emulsifyT. molitor all proteinoil fraction, concentrate. whereasMoreover, 0.5% protein WPI was solubility not competent can be to improved do so until after further enzymatic increased hydrolysis to 1%. of A. domesticus andTheL. migratoria results of powdersEAs showed [19 ,the25]. same None trend of these as the strategies FC of WPI were and used BSFPC in thisdiscussed study in to increase thethe previous solubility section. of the As BSFPC both properties but they ar coulde dependent be applied on the when functionality solubility of musthydropho- be enhanced. bic groups on the protein surface and its molecular flexibility, it can be assumed that Since some techno-functional properties such as foaming and emulsifying depend on the BSFPC can be used in emulsification as other conventional emulsifiers, such as WPI. This wassoluble also proteinexplained fraction, by Zieli basedńska et on al. the[24] solubility who reported results improved obtained, EAs protein in insect solutions protein at pH 7 extractswere prepared compared to to conduct their whole foaming fraction and powders, emulsifying which tests was in due the to later the increased sections. protein contents in the protein extracts, especially with the amount of the hydrophobic amino 3.2.2. WBC and OBC acids. Purscke et al. [25] studied the effect of pH and ionic strength on the EA of L. migratoria proteinWBC andconcentrates OBC are concluding critical features EA increas of fooded with ingredients the increase in foodof ionic processing strength, and applica- thetions. highest They EA are was related reached to near the ability the isoelectric of taking point up of and the retaining protein. However, water and regarding oil, respectively, thewhich functionality directly of affect a protein the textureto be used and as thean emulsifier, ﬂavor of there the products, are more factors especially to be con- in meat and sidered,bakery such [47]. as As the for stability foods with of emulsions, high protein droplet content, size distributions, protein molecule and zeta-potential, structure, amino acid whichcomposition, are explained pH, hydrophilicity,in the later sections. and hydrophobicity on protein surface are determining factors of WBC and OBC [24,47]. BSFPC had WBC of 2.2 g water/g DM that is higher than 3.2.5.the reportedInterfacial 0.4 Tension g water/g DM and 1.5 g water/g DM for T. molitor protein extract and enzymatic hydrolysate of L. migratoria protein [16,23], respectively, but lower than hexane defatted T. molitor powders (2.7 g water/g DM) [48] (Figure3b). As for the OBC, the value obtained for BSFPC (1.1 g oil/g DM) was higher than the reported one for T. molitor protein extract [16], in the range of the ones for defatted T. molitor [48], and lower than the one reported from enzymatic hydrolysate of L. migratoria protein [23]. The relatively higher WBC might be due to the higher protein content in the BSFPC which contains more hydrophilic groups to bind with water molecules. On the contrary, OBC can relate to the hydrophobic groups on the surface of protein molecules to bind with oil. Zieli´nskaet al. [24] reported higher values of WBCs (2.18–3.95 gwater/gDM) and OBCs (1.98–3.33 goil/gDM) of protein Foods 2021, 10, 1048 12 of 21

extracts from T. molitor, L. migratoria, and A. domesticus than the ones found for BSFPC. The WBC and OBC of the BFS powder and BSFPC are comparable to plant-based ﬂours such as wheat and rice, which were reported to have WBC from 1.4 to 1.9 gwater/gDM, and OBC from 1.5 to 1.9 goil/gDM, respectively [49]. Therefore, the BSFPC obtained in this study seems to be suitable in broad food applications entailing high protein content.

3.2.3. FC and FS A total of 0.1 wt.% and 2 wt.% whey protein isolate (WPI) solutions and 0.1 wt.% soluble BSFPC solutions were compared in terms of FC and FS. As it is shown in Figure4a , 0.1% BSFPC displayed higher FC (51%) than 0.1% WPI (21.4 %) and even than 2% WPI (41.7%). The mechanical agitation by Ultra Turrax introduced not only air bubbles but also energy into the whole system which can unfold the protein structure and change the protein conformation favoring the stabilization of air bubbles at the air-water interface. It is reported that protein extraction and enzymatic hydrolysis resulted in the generation of small peptides with surface-stabilizing residues which can rapidly diffuse onto the interface and rearrange the structure to improve FC [19,22–24]. Yi et al. [42] and Purschke et al. [25] discussed the impact of pH and ionic strength on foaming and found improved foaming behavior both at near protein pI and at increased NaCl concentration. Apart from the protein structure, the positive effect of carbohydrates on foaming was explained by Zielińskaet al. [24] which might be also the case in this study due to the BSFPC contained 62.44% of protein and the remaining parts are possibly carbohydrates and soluble fibers. Regardless of the FC, 2% WPI solution showed the highest FS at 120 min after foam generation (31.1%), followed by 0.1% BSFPC (25.2%) and 0.1% WPI (17.3%). Due to the higher concentration of protein in the 2% WPI solution, the viscoelastic film formed at the air-water interface [22,25] was more durable. Nevertheless, BSFPC is considered to be suitable for preparing food products based on foams.

3.2.4. EA EAs of WPI and BSFPC at concentrations of 0.1%, 0.5%, and 1.0% were assessed as described at Section 2.4.5. BSFPC at 0.1% presented higher EA than 0.1% WPI; however, both were not able to emulsify all oil. A visible top oil layer was observed after centrifugation, which was more significant for 0.1% WPI. When the protein concentration was raised to 0.5%, the gap in EA between WPI and BSFPC was reduced (51.7% and 54.2%, respectively) and the EA values were nearly equal (59.2% and 60.0%) at 1.0% protein concentration. It is worth mentioning that 0.5% BSFPC was able to emulsify all oil fraction, whereas 0.5% WPI was not competent to do so until further increased to 1%. The results of EAs showed the same trend as the FC of WPI and BSFPC discussed in the previous section. As both properties are dependent on the functionality of hydrophobic groups on the protein surface and its molecular flexibility, it can be assumed that BSFPC can be used in emulsification as other conventional emulsifiers, such as WPI. This was also explained by Zielińskaet al. [24] who reported improved EAs in insect protein extracts compared to their whole fraction powders, which was due to the increased protein contents in the protein extracts, especially with the amount of the hydrophobic amino acids. Purscke et al. [25] studied the effect of pH and ionic strength on the EA of L. migratoria protein concentrates concluding EA increased with the increase of ionic strength, and the highest EA was reached near the isoelectric point of the protein. However, regarding the functionality of a protein to be used as an emulsifier, there are more factors to be considered, such as the stability of emulsions, droplet size distributions, and zeta-potential, which are explained in the later sections.

3.2.5. Interfacial Tension The effect of WPI and BSFPC on sunflower oil-water interfacial tension was analyzed as described in Section 2.4.6. As indicated in Figure5, the interfacial tension of sunflower oil-water was around 25 mN/m and no significant reduction was observed due to the Foods 2021, 10, x FOR PEER REVIEW 13 of 23

Foods 2021, 10, 1048 The effect of WPI and BSFPC on sunflower oil-water interfacial tension was analyzed13 of 21 as described in Subsection 2.4.6. As indicated in Figure 5, the interfacial tension of sunflower oil-water was around 25 mN/m and no significant reduction was observed due to the absence of surface-active compounds. The addition of WPI or BSFPC lowered the in- absence of surface-active compounds. The addition of WPI or BSFPC lowered the interfacial terfacial tension almost instantaneously to 13.7 and 8.4 mN/m, respectively. During the tension almost instantaneously to 13.7 and 8.4 mN/m, respectively. During the next 10 min, next 10 min, the interfacial tension decreased quickly, and from that point on the decrease the interfacial tension decreased quickly, and from that point on the decrease was slower, was slower, reaching a plateau value of 10.3 mN/m and 3.4 mN/m for WPI and BSFPC, reaching a plateau value of 10.3 mN/m and 3.4 mN/m for WPI and BSFPC, respectively. respectively. The dynamics of the interfacial tension is limited by the diffusion rate of The dynamics of the interfacial tension is limited by the diffusion rate of emulsifiers and, emulsifiers and, subsequently, the time required to reach and adsorb at the interface. subsequently, the time required to reach and adsorb at the interface. When analyzing When analyzing the progress of interfacial tension in similar oil-water systems stabilized the progress of interfacial tension in similar oil-water systems stabilized with food grade with food grade proteins [33] three different phases have been identified: (i) a lag-time proteins [33] three different phases have been identified: (i) a lag-time controlled by the controlled by the initial migration of protein molecules to the interface, (ii) a sharp de- initial migration of protein molecules to the interface, (ii) a sharp decrease by absorption creaseof protein by absorption molecules of on protein interface, molecules and (iii) on a interface, slow decrease and (iii) to reacha slow pseudo-equilibrium decrease to reach pseudo-equilibriuminterfacial tension byinterfacial rearrangement tension ofby protein rearrangement molecules of andprotein multilayer-film molecules and formation. multilayer-filmResults in formation. Figure5 show Results a two-phasein Figure 5 processshow a two-phase with no significant process with lag time,no significant and a faster lag time,diffusion and a andfaster adsorption diffusion and of BSFPC adsorption resulted of BSFPC in lower resulted interfacial in lower values interfacial at time values 0 than atWPI. time Differences0 than WPI. in Differences the molecular in the weight, molecular protein weight, composition, protein andcomposition, structure mayand struc- lead to turedifferences may lead in to the differences performance in the of performanc BSFPC to reducee of BSFPC the interfacial to reduce tension the interfacial in the sunflower tension inoil-water the sunflower [18,50 ].oil-water [18,50].

Figure 5. Interfacial tension between aqueous solutions of water, 0.1% WPI (whey protein isolate) Figure 5. Interfacial tension between aqueous solutions of water, 0.1% WPI (whey protein isolate) and 0.1% BSFPC (black soldier fly protein concentrate), respectively, and sunflower oil (25 ◦C). and 0.1% BSFPC (black soldier fly protein concentrate), respectively, and sunflower oil (25 °C). This finding agrees with the results reported by Gould and Wolf [18] who compared theThis effects finding of whey agrees protein with the and resultsT. molitor reportedprotein by Gould on the and purified Wolf [18] sunflower who compared oil-water theinterface, effects of and whey obtained protein equilibrium and T. molitor interfacial protein tensions on the purified of 13 mN/m sunflower and 12oil-water mN/m in- for terface,whey proteinand obtained and T. equilibrium molitor protein interfacial extract, tensions respectively. of 13 mN/m and 12 mN/m for whey protein and T. molitor protein extract, respectively. 3.3. Dynamic Membrane of Tunable Pore Size (DMTS) Emulsification 3.3. DynamicAfter proving Membrane the of emulsifying Tunable Pore capacity Size (DMTS) of BSFPC Emulsification at certain conditions of rotor-stator homogenizationAfter proving (see the Sectionemulsifying 3.2.4), capacity we assessed of BS theirFPC ability at certain to stabilize conditions emulsions of rotor-stator produced homogenizationwith a low-energy (see membrane Section 3.2.4), emulsification we assessed technique. their ability Premix to stabilize membrane emulsions emulsification pro- ducedwith DMTSwith a low-energy was used to membrane obtain BSFPC emulsifica and WPI-stabilizedtion technique. O/W Premix emulsions membrane formulated emulsi- ficationwith sunflower with DMTS (SO) was or used lemon to oilobtain (LO). BSFPC and WPI-stabilized O/W emulsions formulated with sunflower (SO) or lemon oil (LO). 3.3.1. Droplet Size Distribution 3.3.1. DropletBoth oils Size are Distribution widely used in the food industry in the form of an emulsion, although theyBoth show oils important are widely differences used in inthe composition food indust andry in water the form solubility. of an SO,emulsion, broadly although used as a theymedium show forimportant delivering differences lipophilic in micronutrients composition and such water as carotenoids solubility. SO, and broadly vitamin used E [51 ],as is a amedium complex for mixture delivering of fatty lipophilic acids, totally micronutrien immisciblets such in water,as carotenoids while LO, and used vitamin as a flavoring E [51], and antimicrobial agent, is rich in terpenes and partially water soluble [52,53]. As can be seen in Figure6, BSFPC was able to stabilize O/W emulsions to a similar or even to a higher extent than WPI all along the emulsification process. Notice that all FoodsFoodsFoods 2021 202120212021,, , ,10 10 1010,, , ,x x xx FOR FOR FORFOR PEER PEER PEERPEER REVIEW REVIEW REVIEWREVIEW 14141414 of of ofof 23 23 2323

Foods 2021, 10, x FOR PEER REVIEWisisisis a aaa complex complexcomplexcomplex mixture mixturemixturemixture of ofofof fatty fattyfattyfatty acids, acids,acids,acids, totally totallytotallytotally immiscible immiscibleimmiscibleimmiscible in ininin water, water,water,water, while whilewhilewhile LO, LO,LO,LO, used usedusedused14 of as 23asasas a aaa fla- fla-fla-fla-

voringvoringvoring and andand antimicrobial antimicrobialantimicrobial agent, agent,agent, is isis rich richrich in inin te teterpenesrpenesrpenesrpenes and andandand partially partiallypartiallypartially wa wawawaterterterter soluble solublesolublesoluble [52,53]. [52,53].[52,53].[52,53]. AsAs cancan bebe seenseen inin FigureFigure 6,6, BSFPCBSFPC waswas ableable toto stabilizestabilize O/WO/W emulsionsemulsions toto aa similarsimilar Foods 2021, 10, 1048 AsAs cancan bebe seenseen inin FigureFigure 6,6, BSFPCBSFPC waswas ableable toto stabilizestabilize O/WO/W emulsionsemulsions toto aa similarsimilar14 of 21 orororis eveneven evena complex tototo aaa higherhigher highermixture extentextentextent of fatty thanthanthan acids, WPIWPIWPI totally allallall alongalong alongimmiscible thethethe emulsificationemulsificationemulsification in water, while process.process. process.LO, used NoticeNotice Noticeas a fla- thatthatthat allallall proteinsproteinsproteinsvoring and werewerewere antimicrobial preparedpreparedprepared withwithagent,with nono nois rich bufferingbufferingbuffering in terpenes sososolution.lution.lution.lution. and partially InInInIn addition,addition,addition,addition, water soluble thethethethe differentdifferentdifferentdifferent [52,53]. naturenature naturenature ofofofof eacheacheach oil oiloilAs strongly strongly stronglycan be seen impacted impactedimpacted in Figure the thethe 6, progress progress progressBSFPC was of ofof drop abledropdrop toletletletlet stabilize size sizesizesize and andandand O/W span, span,span,span, emulsions especially especiallyespeciallyespecially to a at atatatsimilar the thethethe highest highesthighesthighest oilproteinsoiloilor fraction. fraction.fraction. even wereto Duringa DuringDuring higher prepared DMTS DMTSextentDMTS with thanemulsification, emulsification,emulsification, no WPI buffering all along solution. drop dropdrop theletslets letsletsemulsification In break breakbreakbreak addition, up upupup as asas asprocess. the they theytheythey different go gogogo Notice through throughthroughthrough nature that the thethetheall of inter- inter-inter-inter- each stitialoilstitialstitialstitialproteins strongly voids voidsvoidsvoids were impactedbetween betweenbetweenbetween prepared the thethethe with microbeads microbeadsmicrobeads progressmicrobeads no buffering of that that dropletthatthat so ma mamamalution.kekeke size upupup In and the thetheaddition, span, dynamicdynamicdynamic especiallythe membrane,differentmembrane,membrane, at nature the ininin highest thisthisthisof casecasecase oil each oil strongly impacted the progress of droplet size and span, especially at the highest withfraction.withwith a aa mean meanmean During diameter diameterdiameter DMTS of ofof emulsification, 78.4 78.478.4 μ μμm.m.m. In InIn most mostmost droplets cases, cases,cases, break droplet dropletdroplet up assize sizesize they reduction reductionreductionreduction go through followed followedfollowedfollowed the interstitial an ananan anal- anal-anal-anal- oil fraction. During DMTS emulsification, droplets break up as they go through the inter- ogousvoidsogousogous between pattern:pattern:pattern: a theaa largelargelarge microbeads dropletdropletdroplet sizesize thatsize reductionreduction makereduction up the tttookookook dynamic placeplaceplace atatat membrane, thethethe firstfirstfirst emulsificationemulsificationemulsification in this case with cycle,cycle,cycle, a stitial voids between the microbeads that make up the dynamic membrane, in this case aftermeanafter whichwhich diameter therethere of werewere 78.4 onlyonlyµm. minor Inminor most reductionsreductions cases, droplet withwith sizeaa slightslight reduction increaseincrease followed inin span,span, an whichwhich analogous agreesagrees afterafterwith which which a mean there there diameter were were ofonly only 78.4 minor minorμm. In reductions reductionsmost cases, with dropletwith a a slight slightsize reduction increase increase followedin in span, span, anwhich which anal- agrees agrees pattern: a large droplet size reduction took place at the first emulsification cycle, after withwithwithogous several severalseveral pattern: studies studiesstudies a large using usingusing droplet the thethe sizesame samesame reduction set-ups set-upsset-ups t[32,54,55]. [32,54,55].[32,54,55].ook place at the first emulsification cycle, whichafter which there werethere were only only minor minor reductions reductions with with a slight a slight increase increase in inspan, span, which agrees agrees with severalwith several studies studies using using the same the same set-ups set-ups [32 ,[32,54,55].54,55].

Figure 6. Sauter mean diameter (d3,2, full symbols) and span (empty symbols) versus the number of emulsification cycles Figure 6. Sauter mean diameter (d , full symbols) and span (empty symbols) versus the number of emulsification cycles FigureFigureFigurefor 6. 6.emulsions6. Sauter SauterSauter mean mean meanat different diameter diameterdiameter oils and (d (d(d3,23,23,2 3,2oil,, ,, full full fullfullfractions: symbols) symbols)symbols)symbols) (a) 20% and andandand SO span spanspanspan (sunflower (empty (empty(empty(empty oil); symbols) symbols)symbols)symbols) (b) 30% versus versus versusversusSO; (c) the the40%thethe number number numbernumberSO; (d) 20% of ofofof emulsification emulsificationemulsificationLOemulsification (lemon oil); cycles cyclescyclescycles for emulsions at different oils and oil fractions: (a) 20% SO (sunflower oil); (b) 30% SO; (c) 40% SO; (d) 20% LO (lemon oil); forforforfor emulsions emulsions emulsionsemulsions(e) 30% LO; at at atat and different different differentdifferent (f) 40% oils oilsLO.oilsoils and and( andand d oil 3,2oil oiloil-1% fractions: fractions: fractions:fractions: WPI (whey ( ( (a(aa))) ) 20% protein20% 20%20% SO SO SOSO isolate); (sunflower (sunflower (sunflower(sunflower d3,2 oil); oil); -1%oil);oil); BSFPC( ( (b(bb))) ) 30% 30% 30%30% (black SO; SO; SO;SO; ( (soldier (c(ccc))) ) 40% 40% 40%40% fly SO; SO; SO;SO; protein ( ( (d(dd))) ) 20% 20% 20%20% concentrate); LO LO LOLO (lemon (lemon (lemon(lemon oil); oil); oil);oil); (((e(ee))) ) 30% 30% 30%30% d LO; LO;3,2 LO;LO;-2% and and andand BSFPC; ( ( (f(ff)f)) ) 40% 40% 40%40% span-1%WPI; LO. LO. LO.LO. ( ( (( d d dd3,23,23,23,2-1%-1%-1%-1% span-1% WPI WPI WPIWPI (whey (whey (whey(whey BSFPC; protein protein proteinprotein span-2% isolate); isolate); isolate);isolate); BSFPC). d d dd3,23,23,23,2 -1%-1%-1%-1% BSFPC BSFPC BSFPCBSFPC (black (black (black(black soldier soldier soldier soldier fly fly fly flyfly protein protein protein protein concentrate); concentrate); concentrate);concentrate); d ddd3,23,23,23,2-2%-2%-2%-2% BSFPC; BSFPC; BSFPC; BSFPC; span-1%WPI; span-1%WPI; span-1%WPI;span-1%WPI; span-1% span-1% span-1%span-1% BSFPC; BSFPC; BSFPC; BSFPC; span-2% span-2% span-2%span-2% BSFPC). BSFPC). BSFPC).BSFPC). As for SO emulsions, d3,2 and span reached values of 12.5 ± 1.8 μm and 1.1 ± 0.1, respectively,AsAsAs for forfor SOSOSO after emulsions,emulsions,emulsions, five cycles ddd of3,23,23,23,2 DMTS andandandand span spanspanspan emulsi reachedreachedreachedreachedfication values valuesregardlessvaluesvalues ofofofof 12.5 12.512.5of12.512.5 the ± ±±±± emulsifier 1.81.81.81.8 μµμμmmm andandand and 1.11.1 1.11.1oil ± ±±± 0.1,0.1,0.1, respectively,respectively,respectively,respectively,fraction, even after afterafterafter though fivefive fivefivefive d cycles cycles3,2cyclescycles of the of ofofof coarse DMTS DMTSDMTSDMTS emulsion emulsificationemulsi emulsiemulsiemulsi fication(cycleficationficationfication 0) ranged regardless regardlessregardlessregardless from ofof ofof28 the thetheμthem emulsifieremulsifier emulsifieremulsifiertoemulsifier 40 μm for and andandand oil oiloiloil emulsions with 20% and 40% oil fraction, respectively. Therefore, the surface-active prop- fraction,fraction,fraction,fraction, even even eveneven though thoughthough though d ddd 3,23,23,23,2 of ofofofof the thethethe the coarse coarsecoarsecoarse coarse emulsion emulsionemulsionemulsion emulsion (cycle (cycle(cycle(cycle (cycle 0) 0)0)0) ranged0) rangedrangedranged ranged from fromfromfrom from 28 282828 μ 28 μμmmmµ to tomto 40 4040 to μ μ 40μmmmµ for forform erties of BSFPC and WPI were able to stabilize even the highest oil-in-water interface area emulsionsforemulsionsemulsions emulsions with withwith with 20% 20%20% 20% and andand 40% and40%40% 40%oil oiloil fraction, fraction,fraction, oil fraction, respec respecrespec respectively.tively.tively.tively.tively. Therefore, Therefore, Therefore,Therefore, Therefore, the the thethe surface-active surface-active surface-activesurface-active the surface-active prop- prop- prop-prop- created during emulsification of 40% oil fraction emulsions. ertiespropertiesertieserties of ofof BSFPC BSFPCBSFPC of BSFPC and andand WPI WPIWPI and were WPIwere were were able ableable to toableto stabilize stabilizestabilize to stabilize even eveneveneven even the thethethe highest highesthighesthighest the highest oil-in-water oil-in-wateroil-in-wateroil-in-water oil-in-water interface interfaceinterfaceinterface interface area areaareaarea Compared to SO, LO emulsions showed smaller droplet size of the coarse emulsions createdareacreated created duringduring during emulsificationemulsification emulsification ofof 40%40% of oil 40%oil fractionfraction oil fraction emulsions.emulsions. emulsions. createdcreatedwith values duringduring ranging emulsificationemulsification from 18 μ m ofof to 40% 40%40% 22 μ oil oiloilm fraction fractionbecausefraction emulsions.of emulsions.emulsions. the lower interfacial tension of the Compared to SO, LO emulsions showed smaller droplet size of the coarse emulsions LO-waterComparedComparedCompared system to toto (12.9 SO, SO,SO, LO LO±LO 0.2 emulsions emulsionsemulsions mN/m [56] showed showedshowed and 11.82 smalle smallesmalle mN/mrrrr droplet dropletdroplet[57])droplet compared size sizesizesize of ofofof the thetothethe the coarse coarsecoarsecoarse SO-water emulsions emulsionsemulsionsemulsions withwithwith valuesvaluesvalues rangingrangingranging from fromfrom 181818 μµμμmmm tototo 222222 µμμμmmm because becausebecause of ofof the thethe lower lowerlower interfacial interfacialinterfacial tensiontensiontension ofofof thethethe LO-waterLO-waterLO-water systemsystemsystem (12.9 (12.9(12.9(12.9 ± ±±± 0.20.20.2 mN/m mN/mmN/m [56][56][56] [56] andandand 11.8211.82 11.8211.82 mN/mmN/m mN/m [57])[57])[57]) [57]) comparedcomparedcompared to toto thethe the SO-water SO-waterSO-water system (25 mN/m, Figure5). After the refining process with DMTS, d 3,2 decreased to values between 4 µm and 7 µm and span ranged from 1 to 2 when oil fraction was kept below 30% (Figure6d–f). However, when the oil fraction increased to 40%, WPI stabilized emulsions could not further get refined in DMTS after three cycles what resulted in higher values of d3,2 and span than those obtained with BSFPC stabilized emulsions, which could Foods 2021, 10, x FOR PEER REVIEWFoodsFoods 2021 2021 , ,10 10, ,x x FOR FOR PEER PEER REVIEW REVIEW 14 of 23 1414 of of 23 23

Foods 2021, 10, x FOR PEERFoods REVIEW 2021 , 10, x FOR PEER REVIEW 14 of 23 14 of 23

Foods 2021, 10, x FOR PEER REVIEW 15 of 23

is a complex mixture ofisis fattya a complex complex acids, mixture totallymixture immiscible of of fatty fatty acids, acids, in water, totally totally while immiscible immiscible LO, used in in water, aswater, a fla- while while LO, LO, used used as as a a fla- fla- is a complex mixtureis of a fatty complex acids, mixture totally ofimmiscible fatty acids, in totallywater, immisciblewhile LO, used in water, as a fla-whilevoring LO, and used antimicrobial as a fla-voringvoring agent, and and is antimicrobialrich antimicrobial in terpenes agent, agent, and ispartially is rich rich in in wate terpenesrpenester soluble and and partially [52,53].partially wa waterter soluble soluble [52,53]. [52,53]. voring and antimicrobialvoring agent,system and is antimicrobial rich(25 mN/m, in terpenes Figure agent, and 5). is Afterpartially rich thein te refiningwarpenester soluble processand partially [52,53]. with DMTS, wa ter d soluble3,2As decreased can [52,53].be seen to val- in FigureAsAs 6,can can BSFPC be be seen seen was in in able Figure Figure to stabilize 6, 6, BSFPC BSFPC O/W was was emulsions able able to to stabilize stabilize to a similar O/W O/W emulsions emulsions to to a a similar similar Foods 2021As, 10 can, 1048 be seen in FigureAsues can 6, between BSFPC be seen 4was μ inm Figureandable 7 to μ m 6,stabilize andBSFPC span O/Wwas ranged ableemulsions from to stabilize 1 to to 2 whena similarO/W oil orfractionemulsions even wasto ato kept higher a similar below extent or15 or ofeven eventhan 21 to toWPI a a higher higher all along extent extent the than thanemulsification WPI WPI all all along along process. the the emulsification emulsificationNotice that all process. process. Notice Notice that that all all or even to a higher extentor even than30% to (FigureWPIa higher all 6d–f). along extent However, the than emulsification WPI when all the along oil process. fraction the emulsification increasedNotice that to 40%, allprocess.proteins WPI stabilized Noticewere prepared that emul- allproteins proteins with no were were buffering prepared prepared solution. with with no Inno addition,buffering buffering theso solution. lution.different In In addition,natureaddition, of the the different different nature nature of of proteins were preparedproteins withsions nowere couldbuffering prepared not furtherso lution.with get no Inrefined buffering addition, in DMTS so thelution. differentafter In three addition, nature cycles ofwhateachthe different oilresulted strongly naturein higher impacted of each each the oil oil progressstrongly strongly impactedof impacted droplet the sizethe progress progress and span, of of especiallydrop dropletlet size size at and theand highestspan, span, especially especially at at the the highest highest values of d3,2 and span than those obtained with BSFPC stabilized emulsions, which could each oil strongly impactedeach oil the strongly progress impacted of droplet the size progress and span, of drop especiallylet size at and the span, highest especiallyoil fraction. at theDuring highest DMTSoil oil fraction. fraction.emulsification, During During dropDMTS DMTSlets emulsification, emulsification, break up as they drop drop goletslets through break break theup up as inter-as they they go go through through the the inter- inter- bebe successfullysuccessfully refined refined over over five five emulsification emulsification cycles cycles (Figure (Figure 6f).6 f).The The poor poor performance performance of oil fraction. During DMTSoil fraction. emulsification, During DMTS droplets emulsification, break up as theydrop golets through break up the as inter- theystitial go through voids between the inter- thestitialstitial microbeads voids voids between between that ma the theke microbeads upmicrobeads the dynamic that that ma membrane,makeke up up the the in dynamic dynamic this case membrane, membrane, in in this this case case WPIof WPI as as an an emulsifier emulsifier in in the the LO/W LO/W system could could be be linked linked to tothe the chemical chemical composition composition of stitial voids between the microbeads that make up the dynamic membrane, in this casewith a mean diameter of 78.4 μm. In most cases, droplet size reduction followed an anal- stitial LO,voidsof LO, containing betweencontaining the anhydrous anhydrous microbeads acids that and and phenolicma phenolicke up acids the acids fromdynamic from the peel the membrane, peel[58,59], [58 able,59 in], tothis able diffuse case to diffuse with with a a mean mean diameter diameter of of 78.4 78.4 μ μm.m. In In most most cases, cases, droplet droplet size size reduction reduction followed followed an an anal- anal- with a mean diameter of 78.4 μm. In most cases, droplet size reduction followed an anal-ogous pattern: a large droplet size reduction took place at the first emulsification cycle, with ain inmean thethe waterdiameter phase phase of and 78.4 and reduce μ reducem. In pH most pH below below cases, the theWPIdroplet WPI isoelectric size isoelectric reduction point, point, in followeda range in a rangeof an 4.8–5.1. anal- of 4.8–5.1. ogousogous pattern: pattern: a a large large droplet droplet size size reduction reduction t ooktook place place at at the the first first emulsification emulsification cycle, cycle, ogous pattern: a largeogous droplet TheThepattern: size water reduction a phase phaselarge of ofdroplet LO/Wt LO/Wook place emulsionssize emulsions reductionat the containing first containing t ookemulsification 40%place 40% oil at fraction oilthe cycle, fraction first showedafter emulsification showed which a pH thereof a pH5.03 cycle,were of that 5.03 after onlyafter that whichminor which therereductions there were were onlywith only minor aminor slight reductions reductions increase in with with span, a a slight whichslight increase increaseagrees in in span, span, which which agrees agrees after which there wereafter only whichmay mayminor cause causethere reductions aggregationwere only with minor and anda slight precipitation precipitation reductions increase ofwithin of wheyspan, whey a slight proteinswhich proteins increase agrees and, and,with in in turn,span, turn,several a which areduction reduction studies agrees ofusing of with with their the several several same studies set-upsstudies using [32,54,55].using the the same same set-ups set-ups [32,54,55]. [32,54,55]. with several studies usingwith severalthesurface-activetheir same surface-active studies set-ups using capacity. [32,54,55]. capacity. the Atsame theseAt set-ups these conditions, conditions, [32,54,55]. the the slightly slightly lower lower isoelectric isoelectric point point of of BSFPC (pHBSFPC 4.0–4.5) (pH 4.0–4.5) was more was favorablemore favorable to stabilize to stabilize LO/W LO/W emulsions emulsions with with a 40%a 40% oil oil fraction. fraction. To confirm the impact of pH on protein performance to stabilize LO/W emulsions, a set of experimentsTo confirm wasthe impact carried of out pH usingon protein 0.2 M pe phosphaterformance to buffer stabilize pH 7LO/W as water emulsions, phase. a LO/W set of experiments was carried out using 0.2 M phosphate buffer pH 7 as water phase. emulsions having 20% and 40% oil fraction were formulated with 1% WPI, 1% BSFPC, or LO/W emulsions having 20% and 40% oil fraction were formulated with 1% WPI, 1% 2% BSFPC. Figure7 shows how, under these pH conditions, whey proteins were able to BSFPC, or 2% BSFPC. Figure 7 shows how, under these pH conditions, whey proteins successfullywere able to stabilizesuccessfully LO/W stabilize emulsions LO/W duringemulsions emulsification. during emulsification. After five After emulsification five cycles, emulsions with 20% oil fraction showed d of 3.1 µm and span of 0.98, while those emulsification cycles, emulsions with 20% oil fraction3,2 showed d3,2 of 3.1 μm and span of µ emulsions0.98, while withthose 40%emulsions oil fraction with 40% had oil a dfraction3,2 of 3.6 hadm a andd3,2 of span 3.6 μ ofm 1.15. and span Regarding of 1.15. BSFPC, althoughRegarding it BSFPC, also stabilized although LO/Wit also stabilized emulsions LO/W in the emulsions five cycles in the of five DMTS cycles emulsification, of DMTS it wasemulsification, not able to it maintain was not theable span to maintain that increased the span over that the increased process. over The the results process. obtained The from WPIresults emulsified obtained LOfrom emulsions WPI emulsified with 20% LO emul and 40%sions oil with fractions 20% and were 40% similar oil fractions to what were reported bysimilar Kaade to etwhat al. [reported31] which by had Kaade slightly et al. smaller [31] which d3,2 duehad slightly to the smaller-size smaller d3,2 emulsifierdue to the Tween 20smaller-size was used. emulsifier Tween 20 was used.

Figure 7. Sauter mean diameter3,2 (d , full symbols) and span (empty symbols) versus the number Figure 7. Sauter mean diameter (d , full3,2 symbols)Figure 6. and Sauter span mean(empty diameter symbols)FigureFigure (d 6. versus6. 3,2Sauter Sauter, full the symbols)mean meannumber diameter diameter and span (d (d3,2 3,2(empty, ,full full symbols) symbols) symbols) and and versus span span the(empty (empty number symbols) symbols) of emulsification versus versus the the number cyclesnumber of of emulsification emulsification cycles cycles ofof emulsificationemulsification cycles cycles for for produced produced LO LO (lemon (lemon oil) emulsions oil) emulsions with 0.2 with M 0.2phosphate M phosphate buffer in buffer the in the Figure 6. Sauter mean diameterFigure 6. (d Sauter3,2, full mean symbols) diameter and span(d3,2, (emptyfull symbols) symbols) and versus span (empty the number symbols) forof emulsificationemulsions versus the at number different cycles offor oilsfor emulsification emulsions emulsionsand oil fractions: at at cyclesdifferent different ( a) 20% oils oils SOand and (sunflower oil oil fractions: fractions: oil); ( a( a)( b)20% )20% 30% SO SO SO; (sunflower (sunflower (c) 40% SO; oil); oil); (d ( b)( b20%) )30% 30% LO SO; SO; (lemon ( c(c) )40% 40% oil); SO; SO; ( d(d) )20% 20% LO LO (lemon (lemon oil); oil); water phase at different oil fractions: (a) 20% LO and (b) 40% LO. ( d3,2-1 % WPI (whey protein for emulsions at differentfor oils emulsions and oil fractions: at different (a) oils 20% and SO oilwater(sunflower fractions: phase oil); ( ata) different 20%(b) 30% SO (sunflowerSO; oil fractions:(c) 40% oil);SO; (a ( (()ebd) 20%) )30% 30%20% LO LO; SO;LO and and (lemon(c) (40% b(f)) 40%40% oil);SO; LO.LO. (d(e()e) 20% ()30% 30% d LO 3,2LO; LO;-1%-1 (lemon and % andWPI WPI ( f( )f (wheyoil); )40% (whey40% LO. LO.protein protein ( ( d d 3,2isolate);3,2-1%-1% WPI WPI (wheyd (whey3,2-1% protein BSFPCprotein isolate);(black isolate); soldier d d3,23,2 -1%fly-1% protein BSFPC BSFPC concentrate);(black (black soldier soldier fly fly protein protein concentrate); concentrate); isolate); d3,2-1 % BSFPC (black soldier fly protein concentrate); d3,2-2 % BSFPC; span- (e) 30% LO; and (f) 40% LO. ( d3,2-1% WPI (whey protein3,2 isolate); d3,2-1%-1 % BSFPC BSFPC (black (black soldier soldier3,2 fly fly d protein3,2 protein-2% BSFPC; concentrate); concentrate); span-1%WPI; d d3,23,2-2%-2-2% % BSFPC; BSFPC; BSFPC; span-1% span-1%WPI; BSFPC;span-1%WPI; span-2% span-1% span-1% BSFPC). BSFPC; BSFPC; span-2% span-2% BSFPC). BSFPC). (e) 30% LO; and (f) 40% LO. ( d -1%1%WPI; WPI (whey span-1%3,2 protein BSFPC; isolate); span-2% d BSFPC).-1% BSFPC (black soldier fly protein3,2 concentrate); d3,2-2% BSFPC; span-1%WPI; d3,2-2% BSFPC; span-1% span-1%WPI; BSFPC; span-2% span-1% BSFPC). BSFPC; span-2% BSFPC). As for SO emulsions, d3,2 and span reached values of 12.5 ± 1.8 μm and 1.1 ± 0.1, 3.3.2. Productivity AsAs for for SO SO emulsions, emulsions, d d3,23,2 and and span span reached reached values values of of 12.5 12.5 ± ± 1.8 1.8 μ μmm and and 1.1 1.1 ± ± 0.1, 0.1, 3.3.2. Productivity As for SO emulsions,As d 3,2for and SO span emulsions, reached d 3,2values and spanof 12.5 reached ± 1.8 μvaluesm and of 1.1 12.5 ± 0.1, ±respectively, 1.8 μm and after1.1 ± five 0.1, respectively, cyclesrespectively, of DMTS after after emulsi five five cycles ficationcycles of of regardless DMTS DMTS emulsi emulsi of theficationfication emulsifier regardless regardless and oil of of the the emulsifier emulsifier and and oil oil The productivity of the emulsification process, measured as flux during emulsifica- respectively, after five cycles of DMTSThe productivity emulsification of the regardless emulsification of the process, emulsifier measured and oilfraction, as flux during even though emulsification, d3,2 of the coarse emulsion (cycle 0) ranged from 28 μm to 40 μm for respectively,tion, is aftera key fiveparameter cycles forof DMTSprocess emulsiscale-up.fication Fluxes regardless obtained during of the the emulsifier fifth emulsifica- and oilfraction, fraction, even even though though d d3,23,2 of of the the coarse coarse emulsion emulsion (cycle (cycle 0) 0) ranged ranged from from 28 28 μ μmm to to 40 40 μ μmm for for fraction, even thoughfraction, d3,2 ofis tionthe aeven keycoarsecycle though parameter of emulsion SO emulsid3,2 of for (cycle onsthe process coarseranged 0) ranged scale-up. emulsionfrom from206 Fluxesm (cycle283m μ−2mh − obtained0)1 to to ranged 40481 μ m 3duringfrommforemulsions−2 h− 128 and the μ mthe fifth towith ones 40 emulsification 20%μ ofm LO for and emulsions emulsions 40% oil fraction, with with 20% 20% respec and and 40%tively. 40% oil oil Therefore, fraction, fraction, respec respecthe surface-activetively.tively. Therefore, Therefore, prop- the the surface-active surface-active prop- prop- 3 −2 −1 3 −2 −1 emulsions with 20% andemulsions 40%cycleemulsions oil with fraction, of SO 20% were emulsions respecand between 40%tively. ranged oil 231 fraction, Therefore, m3 fromm−2h −respec 2061 and the m 617surface-activetively.m m3hm Therefore,−2hto−1, 481depending prop- m theertiesm surface-active onh of theBSFPCand oil the fraction and prop- ones WPI erties oferties were LO of of able BSFPC BSFPC to stabilize and and WPI WPI even were were the able able highest to to stabilize stabilize oil-in-water even even the theinterface highest highest area oil-in-water oil-in-water interface interface area area 3 −2 −1 3 −2 −1 erties of BSFPC and WPIerties were ofemulsions(Figure BSFPC able to8). and stabilize wereThe WPI higher between wereeven values theable 231 highestobtained mto stabilizem oil-in-waterh for andevenLO emulsions 617 the interface m highestm canh oil-in-waterarea be,created dependingattributed during interface to on the the emulsification lower oilarea fraction created created ofduring during 40% oilemulsification emulsification fraction emulsions. of of 40% 40% oil oil fraction fraction emulsions. emulsions. created during emulsificationcreated(Figure during of 40%8). emulsification oil The fraction higher emulsions. values of 40% obtained oil fraction for emulsions. LO emulsions can beCompared attributed toto theSO, lowerLO emulsionsComparedCompared showed to to SO, SO, LOsmalle LO emulsions emulsionsr droplet showed showed size of smallethe smalle coarserr droplet droplet emulsions size size of of the the coarse coarse emulsions emulsions ◦ Compared to SO, LO Comparedemulsionsviscosity ofshowedto thisSO, LO oil smalle (1.41emulsionsr mPadroplet· sshowed at size 25 ofC) smalle the [56 coarse]r compared droplet emulsions size to of thewith the one valuescoarse of SO rangingemulsions (48.8 mPa fromwith with·s 18 at values valuesμm to ranging ranging22 μm becausefrom from 18 18 ofμ μm mthe to to lower22 22 μ μm minterfacial because because of oftension the the lower lower of the interfacial interfacial tension tension of of the the ◦ with values ranging from 1826 μmC) to [60 22] sinceμm because as in any of membrane the lower system,interfacial flux tension is inversely of theLO-water proportional system to the (12.9 viscosity. ± 0.2 mN/m [56] and 11.82 mN/m [57]) compared to the SO-water with values ranging from 18 μm to 22 μm because of the lower interfacial tension of theLO-water LO-water system system (12.9 (12.9 ± ± 0.2 0.2 mN/m mN/m [56] [56] and and 11.82 11.82 mN/m mN/m [57]) [57]) compared compared to to the the SO-water SO-water LO-water system (12.9LO-water ± 0.2Consequently, mN/m system [56] (12.9 and the 11.82± lowest0.2 mN/mmN/m flux [56] values[57]) and compared always 11.82 mN/m correspond to the [57]) SO-water tocompared the emulsions to the withSO-water the highest oil fraction. As for the effect of the protein type and concentration, Figure8 shows that regardless of the emulsion formulation, refining emulsions stabilized with WPI resulted in higher fluxes than the ones stabilized with BSFPC. Given that BSFPC had a protein content of 62.4% and the impurities can be carbohydrates and soluble fibers, it is thought that they contributed to increase the viscosity of the continuous phase. The effect of the viscosity Foods 2021, 10, x FOR PEER REVIEW 16 of 23

viscosity of this oil (1.41 mPa·s at 25 °C) [56] compared to the one of SO (48.8 mPa·s at 26 °C) [60] since as in any membrane system, flux is inversely proportional to the viscosity. Consequently, the lowest flux values always correspond to the emulsions with the highest oil fraction. As for the effect of the protein type and concentration, Figure 8 shows that Foods 2021, 10, 1048 16 of 21 regardless of the emulsion formulation, refining emulsions stabilized with WPI resulted in higher fluxes than the ones stabilized with BSFPC. Given that BSFPC had a protein content of 62.4% and the impurities can be carbohydrates and soluble fibers, it is thought ofthat the they continuous contributed phase to increase on the the flux viscosit for they of DMTS the continuous system hasphase. been The previously effect of the seen by viscosity of the continuous phase on the flux for the DMTS system has been previously Kaade et al. [31], who reported higher fluxes during the emulsification of LO emulsions seen by Kaade et al. [31], who reported higher fluxes during the emulsification of LO stabilized with 2 wt.% Tween 20 than the ones obtained using 1% WPI. Moreover, proteins emulsions stabilized with 2 wt.% Tween 20 than the ones obtained using 1% WPI. More- andover, other proteins compounds and other in compounds BSFPC stabilized in BSFPC emulsions stabilized canemulsions also result can also in fouling result in of foul- the DMTS system,ing of the compared DMTS system, to emulsions compared stabilized to emulsions with stabilized WPI as with already WPI as observed already observed duringpremix emulsificationduring premix withemulsification several microstructured with several microstructured systems [31 systems,61–63]. [31,61–63].

FigureFigure 8. Transmembrane 8. Transmembrane fluxes fluxes obtained obtained atat cyclecycle 1, 1, 3, 3, and and 5 5during during DMTS DMTS (dynam (dynamicic membrane membrane of tunable of tunable pore size) pore size) emulsificationsemulsifications of ( aof) ( SOa) SO (sunflower (sunflower oil) oil) emulsions emulsions and ( (bb)) LO LO (lemon (lemon oil) oil) emulsions emulsions with with different different formulations. formulations.

3.3.3.3.3.3. Stability Stability of the Emulsions Emulsions SOSO emulsions emulsions stabilized stabilized with with WPI WPI (1%) (1%) and and BSFPC BSFPC (1 and (1 2%) and with 2%) oil with fractions oil fractions of of 20%,20%, 30%, 30%, andand 40%40% were kept kept at at room room temperature temperature (25 (25°C) for◦C) seven for seven days. days.Samples Samples were were measuredmeasured onon daysdays 1, 3, 3, and and 7 7of of storage storage to fo tollow follow changes changes in the in droplet the droplet size distribution. size distribution. AlthoughAlthough allall the emulsions emulsions had had creaming creaming after afterone day one of day storage, of storage, it can be itseen can from be seenthe from thedroplet droplet size size distribution distribution measurements measurements (Figure (Figure9) that SO9) emulsions, that SO emulsions, in general, can in general,main- can tain its d3,2 and span during seven days of storage at 25 °C with a minor◦ increase in d3,2 (<2 maintain its d3,2 and span during seven days of storage at 25 C with a minor increase in d μm)(<2 andµ m)span and (<0.5). span There (<0.5). was Therea moderate was incr a moderateease in span increase for the inemulsions span for stabilized the emulsions 3,2with 1% BSFPC with 20% SO at day 7 which can be the aggregation of oil droplets due to stabilized with 1% BSFPC with 20% SO at day 7 which can be the aggregation of oil droplets the protein-protein interaction [64]. In agreement with droplet size distribution evolution due to the protein-protein interaction [64]. In agreement with droplet size distribution in time, zeta potential of 1% BSFPC SO emulsions kept at 25 °C was below −30 mV◦ (Figure evolution10), the reference in time, value zeta potentialfor sufficient of 1%electrostatic BSFPC SOrepulsion emulsions to prevent kept droplet at 25 Ccoales- was below −cence.30 mV SO (Figure emulsions 10), stabilized the reference by WPI value showed for sufﬁcienta stronger electrostaticsurface repulsive repulsion effect (zeta to prevent dropletpotentials coalescence. between −45 SO mV emulsions and −40 stabilizedmV) than the by ones WPI stabilized showed aby stronger BSFPC ( surface−40 mV repulsiveto effect−35 mV). (zeta potentials between −45 mV and −40 mV) than the ones stabilized by BSFPC (−40 mV to −35 mV). As for LO emulsions, the ones prepared with unbuffered proteins were kept at room temperature (Figure9), while the emulsions prepared with buffered proteins were kept at both room temperature and refrigeration (Figure 11). The emulsions without buffer showed ◦ a signiﬁcant increase in d3,2 and span after seven days of storage at 25 C, regardless of the protein used. These changes agree with the zeta potential values (−28.4 mV to −17.3 mV, Figure 10d–f) of WPI stabilized emulsions, which indicate a tendency to droplet coalescence. In LO emulsions the pH decreases to values close to the Ip of WPI, affecting the protein conformation and hence emulsion stability. Since the Ip of BSFPC used in this study is slightly lower than the one of WPI, the decrease of pH has a lower impact on the protein conformation. The emulsions showed a lower increase of d3,2 and span after seven days of storage at 25 ◦C than emulsions stabilized with WPI. For BSFPC the zeta potential values were maintained in the range of −35.9 mV to −30.3 mV for all the emulsions. As for the LO emulsions prepared with phosphate buffer, the ones with WPI showed a slight increase ◦ of d3,2 and span after seven days of storage at 25 C and almost no changes when stored at 4 ◦C. The stability of these emulsions correlates well with their zeta potential values, −70 mV to −50 mV (Figure S1). However, when using a phosphate buffer to produce emulsions with BSFPC, d3,2 and span increased more than for the unbuffered emulsions at 25 ◦C. Decreasing the storage temperature had a positive effect on the emulsion stability, Foods 2021, 10, 1048 Foods 2021, 10, x FOR PEER REVIEW 17 of 21 17 of 23

mainly for the lowest oil fraction. From these results, it seems that the BSFPC obtained in Foods 2021, 10, x FOR PEER REVIEW 17 of 23 this work could be a better option for encapsulating lemon oil without the need of buffering Foods 2021, 10, x FOR PEER REVIEW 17 of 23 the pH of the continuous phase.

Foods 2021, 10, x FOR PEER REVIEW 14 of 23

is a complex mixture of fatty acids, totally immiscible in water, while LO, used as a flavoring and antimicrobial agent, is rich in terpenes and partially water soluble [52,53]. As can be seen in Figure 6, BSFPC was able to stabilize O/W emulsions to a similar or even to a higher extent than WPI all along the emulsification process. Notice that all proteins were prepared with no buffering solution. In addition, the different nature of each oil strongly impacted the progress of droplet size and span, especially at the highest oil fraction. During DMTS emulsification, droplets break up as they go through the interstitial voids between the microbeads that make up the dynamic membrane, in this case with a mean diameter of 78.4 μm. In most cases, droplet size reduction followed an analogous pattern: a large droplet size reduction took place at the first emulsification cycle, after which there were only minor reductions with a slight increase in span, which agrees with several studies using the same set-ups [32,54,55].

Figure 9. Particle size distribution of SO (sunflower oil) and LO (lemon oil) emulsions (no buffer added) stabilized by 1%WPI (whey protein isolate), 1% BSFPC (black soldier fly protein concentrate), and 2% BSFPC at day 0, day 1, day 3, and day 7 of storage at room temperature (25 °C). The color transparency scale indicates the length of storage time: from FigureFigureFigure 9. Particle9. 9.Particle Particle size size size distribution distribution distribution of of of SO SOSO (sunﬂower darkest(sunflower(sunflower to lightest oil)oil) andand refer LO LO (lemon(lemon fresh emulsionoil) oil) oil) emulsions emulsions emulsions (day (no 0),(no (no bufferday buffer buffer 1, added)day added) added) 3, andstabilized stabilized stabilizedday 7. by by by 1%WPI1%WPI1%WPI (whey (whey (whey protein protein protein isolate), isolate), isolate), 1% 1% 1% BSFPC BSFPC BSFPC (black (black (black soldier soldiersoldier ﬂyfly proteinproteinein concentrate), concentrate), and and and 2% 2% 2% BSFPC BSFPC BSFPC at at day at day day 0, 0,day 0,day day1, day1, 1,day 3, day and3, and 3, and dayday 7day of 7 storage of7 ofstorage storage at room at at room room temperature temperature temperature (25 ◦ (25C).(25 °C). The°C). colorTheThe color transparency transparencysparency scale scalescale indicates indicates indicates the the length the length length of storageof ofstorage storage time: time: time: from from from darkest to lightestdarkestdarkest referto tolightest lightest fresh refer emulsion refer fresh fresh (dayemulsion emulsion 0), day (day (day 1, 0), day0), dayday 3, and1,1, day day 3,3, andand 7. dayday 7. 7.

Figure 6. Sauter mean diameter (d3,2, fullFigure symbols) 10. Zeta and potential span (empty versus symbols) storage timeversus at theroom number temperature of emulsification (25 °C) for cyclesproduced emulsions ◦ FigureFigurefor 10. emulsions 10.Zeta Zeta potential potential at different versusversus oils storage storage and oil time time fractions:(no at at roombuffer room temperature(a added)) temperature 20% SO at (sunflower (25different °C) (25 forC) oilsproduced oil); for andproduced (b )oil 30%emulsions fractions: SO; emulsions (c ) 40%(a) 20% SO; SO(d) (sunflower20% LO (lemon oil); (oil);b) 30% SO; (c) 40% (no buffer added) at different oils and oil fractions: (a) 20% SO (sunflower oil); (b) 30% SO; (c) 40% (noFigure buffer(e) 30% 10. added) Zeta LO; potentialand at different(f) 40% versus LO. oils storage( and d3,2 oil-1% time fractions: SO;WPI at (room d(whey) 20% ( atemperature) 20%proteinLO (lemon SO isolate); (sunﬂower (25 oil); °C) ( efor) d 30 oil);3,2produced -1%% (LO;b BSFPC) 30% and emulsions SO; (black(f) 40 (c) % soldier 40% LO. ( fly 1% protein WPI (wheyconcentrate); protein isolate); 1% (noSO; buffer (d) 20% added) LO (lemon at different oil); (e )oils 30 %and LO; oil and fractions: (f) 40 % ( aLO.) 20% ( SO1% (sunflowerWPI (whey oil);protein (b) isolate);30% SO; (c 1%) 40% SO; (d) d 20%3,2-2% LO BSFPC; (lemon oil); span-1%WPI; (e) 30 % LO; and span-1%BSFPC (f) 40 (black BSFPC; % LO. soldier ( span-2%1% fly WPI protein (wheyBSFPC). concentrate); protein isolate); and 2%1% BSFPC). SO;BSFPC (d) 20% (black LO soldier (lemon fly oil); protein (e) 30 concentrate); % LO; and and(f) 40 % 2% LO. BSFPC). ( 1% WPI (whey protein isolate); 1% BSFPC (black soldier ﬂy protein concentrate); and 2% BSFPC). BSFPC (black soldier fly protein concentrate); and 2% BSFPC). As for SO emulsions, d3,2 and span reached values of 12.5 ± 1.8 μm and 1.1 ± 0.1, respectively, after five cycles of DMTS emulsification regardless of the emulsifier and oil fraction, even though d3,2 of the coarse emulsion (cycle 0) ranged from 28 μm to 40 μm for emulsions with 20% and 40% oil fraction, respectively. Therefore, the surface-active properties of BSFPC and WPI were able to stabilize even the highest oil-in-water interface area created during emulsification of 40% oil fraction emulsions. Compared to SO, LO emulsions showed smaller droplet size of the coarse emulsions with values ranging from 18 μm to 22 μm because of the lower interfacial tension of the LO-water system (12.9 ± 0.2 mN/m [56] and 11.82 mN/m [57]) compared to the SO-water

Foods 2021, 10, x FOR PEER REVIEW 18 of 23

As for LO emulsions, the ones prepared with unbuffered proteins were kept at room temperature (Figure 9), while the emulsions prepared with buffered proteins were kept at both room temperature and refrigeration (Figure 11). The emulsions without buffer showed a significant increase in d3,2 and span after seven days of storage at 25 °C, regardless of the protein used. These changes agree with the zeta potential values (−28.4 mV to −17.3 mV, Figure 10d–f) of WPI stabilized emulsions, which indicate a tendency to droplet coalescence. In LO emulsions the pH decreases to values close to the Ip of WPI, affecting the protein conformation and hence emulsion stability. Since the Ip of BSFPC used in this study is slightly lower than the one of WPI, the decrease of pH has a lower impact on the protein conformation. The emulsions showed a lower increase of d3,2 and span after seven days of storage at 25 °C than emulsions stabilized with WPI. For BSFPC the zeta potential values were maintained in the range of −35.9 mV to −30.3 mV for all the emulsions. As for the LO emulsions prepared with phosphate buffer, the ones with WPI showed a slight increase of d3,2 and span after seven days of storage at 25 °C and almost no changes when stored at 4 °C. The stability of these emulsions correlates well with their zeta potential values, −70 mV to −50 mV (Figure S1). However, when using a phosphate buffer to produce emulsions with BSFPC, d3,2 and span increased more than for the unbuffered emulsions at 25 °C. Decreasing the storage temperature had a positive effect on the emulsion stability, mainly for the lowest oil fraction. From these results, it seems that the BSFPC Foods 2021, 10, 1048 18 of 21 obtained in this work could be a better option for encapsulating lemon oil without the need of buffering the pH of the continuous phase.

FigureFigure 11. 11. ParticleParticle size size distribution distribution of ofLO LO (lemon (lemon oil)oil) emulsions emulsions stabilized stabilized by buffered by buffered 1% WPI 1% WPI (whey(whey protein protein isolate), isolate), buffered buffered 1% 1% BSFPC BSFPC (black (black soldier soldier fly ﬂy protein protein concentrate), concentrate), and and buffered buffered 2% 2% BSFPCBSFPC after after day day 0 0fresh fresh emulsion, emulsion, day day 1, 1,day day 3, 3,an andd day day 7 of 7 ofstorage storage at atroom room temperature temperature (25 (25 °C)◦ C) andand in in fridge fridge (4 (4 °C).◦C). The The color color tr transparencyansparency scale scale indicates indicates the the leng lengthth of of storage storage time: time: from from darkest darkest toto lightest lightest refer refer fresh fresh emulsion emulsion (d (dayay 0), 0), day day 1, 1, day day 3, 3, and and day day 7. 7.

4. Conclusions This study presents a holistic approach to the use of black soldier fly protein to stabilize food emulsions. The extraction process enabled enrichment of the protein content from 39% (original powder) to almost 63% thereby resulting in a protein concentrate (BSFPC). The essential amino acid profile of BSFPC is similar to soybean and casein and meets the FAO recommendation. As for the techno-functional properties of the BSFPC compared to WPI, it has been proven that BSFPC has higher FC and FS than WPI at low protein concentration (0.1%). Moreover, it was found that BSFPC has higher values of EA for low protein concentration (0.1%) than WPI and comparable values to WPI for 2% concentration. These results show the potential of BSFPC to be used in food formulations to replace totally or partially WPI. Moreover, the ability of BSFCP to lower the interfacial tension in the sunflower oil/water system as well as the values for WBC and OBC, point out the high potential of this protein to stabilize food emulsions. This has been proved using BSFCPC to stabilize emulsions with two oils frequently used in the food industry, such as sunflower oil and lemon oil. The emulsions have been produced using a low-energy high-throughput system previously tested with conventional food emulsifiers. Droplet size distribution and fluxes obtained for sunflower oil emulsions stabilized with BSFPC are comparable to the ones obtained with emulsions stabilized with WPI. For lemon oil emulsions, however, BSFPC successfully reduced droplet size distribution of emulsions with 20% to 40% oil fraction, while the ones produced with WPI for 30% and 40% oil fraction clearly showed an increase in the droplet size distribution after each emulsification cycle. It has been seen that since lemon oil is partially soluble in water when the emulsions have more than 30% oil Foods 2021, 10, 1048 19 of 21

fraction, there is a decrease in the pH of the emulsion. For WPI this pH decrease leads to a value close to pI of the protein, and therefore lowering its ability to stabilize the oil-water interface. Since the BSFPC has a lower pI, this phenomenon is not that important. As for the storage stability of the emulsions, the results point out that BSFPC has comparable results to WPI for sunﬂower oil at 4 ◦C. For lemon oil, or at higher temperatures (25 ◦C) WPI can better maintain the droplet distribution if the pH can be controlled. Even though further research is required to improve the protein extraction process to improve its solubility, the results of this study show the potential of BSFPC to become a sustainable protein for the food industry.

Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/foods10051048/s1, Figure S1: Zeta potential of LO emulsions (0.2 M phosphate buffer pH 7) versus storage time. Author Contributions: Conceptualization, C.G. and M.F.; methodology, J.W., C.G., M.F., S.d.L.-C., and A.F.-A.; validation, C.G., M.F., S.d.L.-C., and A.F.-A.; formal analysis, J.W.; investigation, J.W., M.J., and J.J.; resources, C.G., M.F., S.d.L.-C., and A.F.-A.; data curation, J.W.; writing—original draft preparation, J.W.; writing—review and editing, C.G., M.F., S.d.L.-C., and A.F.-A.; visualization, J.W.; supervision, C.G. and M.F.; project administration, C.G. and M.F.; funding acquisition, C.G. and M.F. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement (No. 713679) and the Universi- tat Rovira i Virgili (URV), and the Ministerio de Economía i Competitividad (CTQ 2014-54520-P) Ministerio de Ciencia e Innovación (PGC2018-097095-B-I00). Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Data is contained within the article or supplementary material. Acknowledgments: The authors are thankful to Hexafly (Ireland) for kindly providing black soldier fly powders. Conflicts of Interest: The authors declare no conflict of interest.

References 1. United Nations. World Population Prospects 2019; United Nations: New York, NY, USA, 2019. 2. Alexandratos, N.; Bruinsma, J. World Agriculture towards 2030/2050 the 2012 Revision; FAO: Rome, Italy, 2012. 3. Charlton, A.J.; Dickinson, M.; Wakefield, M.E.; Fitches, E.; Kenis, M.; Han, R.; Zhu, F.; Kone, N.; Grant, M.; Devic, E.; et al. Exploring the chemical safety of fly larvae as a source of protein for animal feed. J. Insects Food Feed 2015, 1, 7–16. [CrossRef] 4. Oonincx, D.G.A.B.; Van Broekhoven, S.; Van Huis, A.; Van Loon, J.J.A. Feed conversion, survival and development, and composition of four insect species on diets composed of food by-products. PLoS ONE 2015, 10, e0144601. [CrossRef] 5. Spranghers, T.; Ottoboni, M.; Klootwijk, C.; Ovyn, A.; Deboosere, S.; De Meulenaer, B.; Michiels, J.; Eeckhout, M.; De Clercq, P.; De Smet, S. Nutritional composition of black soldier fly (Hermetia illucens) prepupae reared on different organic waste substrates. J. Sci. Food Agric. 2017, 97, 2594–2600. [CrossRef] 6. Van Huis, A.; Van Itterbeeck, J.; Klunder, H.; Mertens, E.; Halloran, A.; Muir, G.; Vantomme, P. Edible Insects: Future Prospects for Food and Feed Security; FAO: Rome, Italt, 2013; Volume 171, ISBN 9789251075951. 7. Van Krimpen, M.; Bikker, P.; Van der Meer, I.; Van der Peet-Schwering, C.; Vereijken, J. Cultivation, Processing and Nutritional Aspects for Pigs and Poultry of European Protein Sources as Alternatives for Imported Soybean Products; Wageningen UR Livestock Research: Lelystat, The Netherlands, 2013. 8. Zielińska,E.; Baraniak, B.; Kara´s,M.; Rybczyńska,K.; Jakubczyk, A. Selected species of edible insects as a source of nutrient composition. Food Res. Int. 2015, 77, 460–466. [CrossRef] 9. Amato, M. Insects as Food: A Cross-Cultural Comparison of Consumers’ Intention and Behaviour. Ph.D. Thesis, Università degli Studi di Napoli Federico II, Napoli, Italy, 2017; p. 121. 10. DeFoliart, G.R. Insects as food: Why the Western Attitude Is Important. Annu. Rev. Entomol. 1999, 44, 21–50. [CrossRef] 11. Hartmann, C.; Shi, J.; Giusto, A.; Siegrist, M. The psychology of eating insects: A cross-cultural comparison between Germany and China. Food Qual. Prefer. 2015, 44, 148–156. [CrossRef] 12. Tan, H.S.G.; Fischer, A.R.H.; Tinchan, P.; Stieger, M.; Steenbekkers, L.P.A.; van Trijp, H.C.M. Insects as food: Exploring cultural exposure and individual experience as determinants of acceptance. Food Qual. Prefer. 2015, 42, 78–89. [CrossRef] Foods 2021, 10, 1048 20 of 21

13. Yen, A.L. Edible insects: Traditional knowledge or western phobia? Entomol. Res. 2009, 39, 289–298. [CrossRef] 14. Balzan, S.; Fasolato, L.; Maniero, S.; Novelli, E. Edible insects and young adults in a north-east Italian city an exploratory study. Br. Food J. 2016, 118, 318–326. [CrossRef] 15. Schösler, H.; de Boer, J.; Boersema, J.J. Can we cut out the meat of the dish? Constructing consumer-oriented pathways towards meat substitution. Appetite 2012, 58, 39–47. [CrossRef] 16. Bußler, S.; Rumpold, B.A.; Jander, E.; Rawel, H.M.; Schlüter, O.K. Recovery and techno-functionality of flours and proteins from two edible insect species: Meal worm (Tenebrio molitor) and black soldier fly (Hermetia illucens) larvae. Heliyon 2016, 2. [CrossRef] 17. Caligiani, A.; Marseglia, A.; Leni, G.; Baldassarre, S.; Maistrello, L.; Dossena, A.; Sforza, S. Composition of black soldier fly prepupae and systematic approaches for extraction and fractionation of proteins, lipids and chitin. Food Res. Int. 2018, 105, 812–820. [CrossRef] 18. Gould, J.; Wolf, B. Interfacial and emulsifying properties of mealworm protein at the oil/water interface. Food Hydrocoll. 2018, 77, 57–65. [CrossRef] 19. Hall, F.G.; Jones, O.G.; O’Haire, M.E.; Liceaga, A.M. Functional properties of tropical banded cricket (Gryllodes sigillatus) protein hydrolysates. Food Chem. 2017, 224, 414–422. [CrossRef][PubMed] 20. Chou, T.H.; Nugroho, D.S.; Cheng, Y.S.; Chang, J.Y. Development and Characterization of Nano-emulsions Based on Oil Extracted from Black Soldier Fly Larvae. Appl. Biochem. Biotechnol. 2020, 191, 331–345. [CrossRef] 21. Park, S.I.; Kim, J.W.; Yoe, S.M. Purification and characterization of a novel antibacterial peptide from black soldier fly (Hermetia illucens) larvae. Dev. Comp. Immunol. 2015, 52, 98–106. [CrossRef][PubMed] 22. Mishyna, M.; Martinez, J.-J.J.I.; Chen, J.; Benjamin, O. Extraction, characterization and functional properties of soluble proteins from edible grasshopper (Schistocerca gregaria) and honey bee (Apis mellifera). Food Res. Int. 2019, 116, 697–706. [CrossRef] [PubMed] 23. Purschke, B.; Tanzmeister, H.; Meinlschmidt, P.; Baumgartner, S.; Lauter, K.; Jäger, H. Recovery of soluble proteins from migratory locust (Locusta migratoria) and characterisation of their compositional and techno-functional properties. Food Res. Int. 2018, 106, 271–279. [CrossRef] 24. Zielińska,E.; Kara´s,M.; Baraniak, B. Comparison of functional properties of edible insects and protein preparations thereof. LWT 2018, 91, 168–174. [CrossRef] 25. Purschke, B.; Meinlschmidt, P.; Horn, C.; Rieder, O.; Jäger, H. Improvement of techno-functional properties of edible insect protein from migratory locust by enzymatic hydrolysis. Eur. Food Res. Technol. 2018, 244, 999–1013. [CrossRef] 26. Kim, H.; Setyabrata, D.; Lee, Y.J.; Jones, O.G.; Kim, Y.H.B. Pre-treated mealworm larvae and silkworm pupae as a novel protein ingredient in emulsion sausages. Innov. Food Sci. Emerg. Technol. 2016, 38, 116–123. [CrossRef] 27. Scholliers, J.; Steen, L.; Glorieux, S.; Van de Walle, D.; Dewettinck, K.; Fraeye, I. The effect of temperature on structure formation in three insect batters. Food Res. Int. 2019, 122, 411–418. [CrossRef] 28. Wang, Y.-S.; Shelomi, M. Review of Black Soldier Fly (Hermetia illucens) as Animal Feed and Human Food. Foods 2017, 6, 91. [CrossRef] 29. Ramos-Elorduy, J.; Manuel, J.; Moreno, P.; Prado, E.E.; Perez, M.A.; Otero, J.L.; Guevara, O.L. De Nutritional Value of Edible Insects from the State of Oaxaca, Mexico. J. Food Compos. Anal. 1997, 157, 142–157. [CrossRef] 30. Nazir, A.; Schroën, K.; Boom, R. Premix emulsification: A review. J. Memb. Sci. 2010, 362, 1–11. [CrossRef] 31. Kaade, W.; Güell, C.; Ballon, A.; Mellado-Carretero, J.; De Lamo-Castellví, S.; Ferrando, M. Dynamic membranes of tunable pore size for lemon oil encapsulation. LWT 2020, 123, 109090. [CrossRef] 32. Wang, J.; Martínez-Hernández, A.; de Lamo-Castellví, S.; Romero, M.-P.; Kaade, W.; Ferrando, M.; Güell, C. Low-energy membrane- based processes to concentrate and encapsulate polyphenols from carob pulp. J. Food Eng. 2020, 281, 109996. [CrossRef] 33. Ladjal-Ettoumi, Y.; Berton-Carabin, C.; Chibane, M.; Schroën, K.; Ladjal Ettoumi, Y.; Berton-Carabin, C.; Chibane, M.; Schroën, K. Legume Protein Isolates for Stable Acidic Emulsions Prepared by Premix Membrane Emulsification. Food Biophys. 2017, 12, 119–128. [CrossRef] 34. Zhao, X.; Vázquez-Gutiérrez, J.L.; Johansson, D.P.; Landberg, R.; Langton, M. Yellow mealworm protein for food purposes— Extraction and functional properties. PLoS ONE 2016, 11, e0147791. [CrossRef] 35. Lourenço, S.O.; Barbarino, E.; De-Paula, J.C.; Pereira, L.O.D.S.; Lanfer Marquez, U.M.; Lourenco, S.O.; Barbarino, E.; De-Paula, J.C.; da S. Pereira, L.O.; Marquez, U.M.L. Amino acid composition, protein content and calculation of nitrogen-to-protein conversion factors for 19 tropical seaweeds. Phycol. Res. 2002, 50, 233–241. [CrossRef] 36. Mellado-Carretero, J.; García-Gutiérrez, N.; Ferrando, M.; Güell, C.; García-Gonzalo, D.; De Lamo-Castellví, S. Rapid discrimina- tion and classification of edible insect powders using ATR-FTIR spectroscopy combined with multivariate analysis. J. Insects Food Feed 2020, 6, 141–148. [CrossRef] 37. Kaade, W.; Ferrando, M.; Khanmohammed, A.; Torras, C.; De Lamo-Castellví, S.; Güell, C. Low-energy high-throughput emulsification with nickel micro-sieves for essential oils encapsulation. J. Food Eng. 2019, 263, 326–336. [CrossRef] 38. Janssen, R.H.; Vincken, J.-P.P.; Van Den Broek, L.A.M.M.; Fogliano, V.; Lakemond, C.M.M.M. Nitrogen-to-Protein Conversion Factors for Three Edible Insects: Tenebrio molitor, Alphitobius diaperinus, and Hermetia illucens. J. Agric. Food Chem. 2017, 65, 2275–2278. [CrossRef][PubMed] 39. Oshimura, E.; Sakamoto, K. Amino acids, peptides, and proteins. Cosmet. Sci. Technol. Theor. Princ. Appl. 2017, 285–303. [CrossRef] 40. Burns, A.; Olszowy, P.; Ciborowski, P. Biomolecules. In Proteomic Profiling and Analytical Chemistry: The Crossroads, 2nd ed.; Elsevier Inc.: Amsterdam, The Netherlands, 2016; pp. 7–24. ISBN 9780444636881. Foods 2021, 10, 1048 21 of 21

41. Azagoh, C.; Ducept, F.; Garcia, R.; Rakotozafy, L.; Cuvelier, M.-E.; Keller, S.; Lewandowski, R.; Mezdour, S. Extraction and physicochemical characterization of Tenebrio molitor proteins. Food Res. Int. 2016, 88, 24–31. [CrossRef][PubMed] 42. Yi, L.; Lakemond, C.M.M.; Sagis, L.M.C.; Eisner-Schadler, V.; Van Huis, A.; Boekel, M.A.J.S.V. Extraction and characterisation of protein fractions from five insect species. Food Chem. 2013, 141, 3341–3348. [CrossRef] 43. Laroche, M.; Perreault, V.; Marciniak, A.; Gravel, A.; Chamberland, J.; Doyen, A. Comparison of conventional and sustainable lipid extraction methods for the production of oil and protein isolate from edible insect meal. Foods 2019, 8, 572. [CrossRef] [PubMed] 44. Young, V.R.; Pellett, P.L. Protein evaluation, amino acid scoring and the Food and Drug Administration’s proposed food labeling regulations. J. Nutr. 1991, 121, 145–150. [CrossRef] 45. Chatsuwan, N.; Nalinanon, S.; Puechkamut, Y.; Lamsal, B.P.; Pinsirodom, P. Characteristics, Functional Properties, and Antioxi- dant Activities of Water-Soluble Proteins Extracted from Grasshoppers, Patanga succincta and Chondracris roseapbrunner. J. Chem. 2018, 2018.[CrossRef] 46. Socrates, G. Infrared and Raman Characteristic Group Frequencies Contents, 3rd ed.; Wiley: Hoboken, NJ, USA, 2004; ISBN 978-0-470- 09307-8. 47. Haque, M.A.; Timilsena, Y.P.; Adhikari, B. Food Proteins, Structure, and Function. In Reference Module in Food Science; Elsevier: Amsterdam, The Netherlands, 2016; pp. 1–8. ISBN 9780081005965. 48. Son, Y.J.; Lee, J.C.; Hwang, I.K.; Nho, C.W.; Kim, S.H. Physicochemical properties of mealworm (Tenebrio molitor) powders manufactured by different industrial processes. LWT 2019, 116, 108514. [CrossRef] 49. Chandra, S. Assessment of functional properties of different flours. Afr. J. Agric. Res. 2013, 8, 4849–4852. [CrossRef] 50. McClements, D.J.; Gumus, C.E. Natural emulsifiers—Biosurfactants, phospholipids, biopolymers, and colloidal particles: Molecular and physicochemical basis of functional performance. Adv. Colloid Interface Sci. 2016, 234, 3–26. [CrossRef] 51. Onwulata, C.I. Microencapsulation and functional bioactive foods. J. Food Process. Preserv. 2013, 37, 510–532. [CrossRef] 52. Rao, J.; McClements, D.J. Lemon oil solubilization in mixed surfactant solutions: Rationalizing microemulsion & nanoemulsion formation. Food Hydrocoll. 2012, 26, 268–276. [CrossRef] 53. Burt, S. Essential oils: Their antibacterial properties and potential applications in foods—A review. Int. J. Food Microbiol. 2004, 94, 223–253. [CrossRef][PubMed] 54. Nazir, A.; Boom, R.M.; Schroën, K. Droplet break-up mechanism in premix emulsification using packed beds. Chem. Eng. Sci. 2013, 92, 190–197. [CrossRef] 55. Sahin, S.; Sawalha, H.; Schroën, K. High throughput production of double emulsions using packed bed premix emulsification. Food Res. Int. 2014, 66, 78–85. [CrossRef] 56. Rao, J.; McClements, D.J. Impact of lemon oil composition on formation and stability of model food and beverage emulsions. Food Chem. 2012, 134, 749–757. [CrossRef][PubMed] 57. Treesuwan, W.; Neves, M.A.; Uemura, K.; Nakajima, M.; Kobayashi, I. Preparation characteristics of monodisperse oil-in-water emulsions by microchannel emulsification using different essential oils. LWT Food Sci. Technol. 2017, 84, 617–625. [CrossRef] 58. González-Mas, M.C.; Rambla, J.L.; López-Gresa, M.P.; Amparo Blázquez, M.; Granell, A. Volatile compounds in citrus essential oils: A comprehensive review. Front. Plant. Sci. 2019, 10, 1–18. [CrossRef] 59. Cheong, M.W.; Chong, Z.S.; Liu, S.Q.; Zhou, W.; Curran, P.; Yu, B. Characterisation of calamansi (Citrus microcarpa). Part I: Volatiles, aromatic profiles and phenolic acids in the peel. Food Chem. 2012, 134, 686–695. [CrossRef][PubMed] 60. Diamante, L.M.; Lan, T. Absolute Viscosities of Vegetable Oils at Different Temperatures and Shear Rate Range of 64.5 to 4835 s-1. J. Food Process. 2014, 2014, 1–6. [CrossRef] 61. Trentin, A.; Ferrando, M.; López, F.; Güell, C. Premix membrane O/W emulsification: Effect of fouling when using BSA as emulsifier. Desalination 2009, 245, 388–395. [CrossRef] 62. Surh, J.; Jeong, Y.G.; Vladisavljević,G.T. On the preparation of lecithin-stabilized oil-in-water emulsions by multi-stage premix membrane emulsification. J. Food Eng. 2008, 89, 164–170. [CrossRef] 63. Sawalha, H.; Sahin, S.; Schroën, K. Preparation of polylactide microcapsules at a high throughput with a packed-bed premix emulsification system. J. Appl. Polym. Sci. 2016, 133, 3–9. [CrossRef] 64. Zecca, E. Investigating the Role of Surface Hydrophobicity in Protein Aggregation. Ph.D. Thesis, University of Connecticut Graduate School, Storrs, CT, USA, 2017.