Isophorone Diisocyanate (IPDI) Microencapsulation for Mono-Component Adhesives: Effect of the Active H and NCO Sources

polymers

Article Isophorone Diisocyanate (IPDI) Microencapsulation for Mono-Component Adhesives: Effect of the Active H and NCO Sources

Mahboobeh Attaei 1 ID ,Mónica V. Loureiro 1,Mário do Vale 1, José A. D. Condeço 1, Isabel Pinho 2, João C. Bordado 1 and Ana C. Marques 1,* ID 1 CERENA, Departamento de Engenharia Química, Instituto Superior Técnico, Universidade de Lisboa, Avenida Rovisco Pais, 1049-001 Lisboa, Portugal; [email protected] (M.A.); [email protected] (M.V.L.); [email protected] (M.d.V.); [email protected] (J.A.D.C.); [email protected] (J.C.B.) 2 CIPADE—Indústria e Investigação de Produtos Adesivos, SA. Av. 1◦ de Maio, 518 Z. Industrial 1, Apt. 82, 3701-909 São João da Madeira, Portugal; [email protected] * Correspondence: [email protected]; Tel.: +351-21-8417-335

Received: 15 June 2018; Accepted: 22 July 2018; Published: 26 July 2018

Abstract: Polyurea/polyurethane (PUa/PU) shell microcapsules (MCs), containing high loadings of isophorone diisocyanate (IPDI) in the core, were developed to enable the production of mono-component, eco-friendly and safer adhesive formulations for the footwear industry. IPDI microencapsulation was obtained via oil–in–water (O/W) microemulsion combined with interfacial polymerization. A methylene diphenyl diisocyanate (MDI) compound (a commercial blend of monomeric and polymeric species), with higher reactivity than IPDI and low viscosity, was added to the O phase to competitively contribute to the shell formation, improving its quality. Four different active H sources were tested, aimed at achieving a high encapsulation yield. The successful encapsulation of IPDI was conﬁrmed by Fourier transformed infrared spectroscopy (FTIR) and thermogravimetric analysis (TGA), while the MCs’ morphology and size distribution were assessed by scanning electron microscopy (SEM). The incorporation of a multifunctional isocyanate silane in the O phase, as “latent” active H source, led to the formation of impermeable PUa/PU-silica hybrid shell MCs with more than 60 wt.% of pure encapsulated IPDI. A proof-of-concept study shows high peeling strength and a structural type of failure of the adhesive joint, revealing an effective IPDI release. These new engineered MCs are found to be promising crosslinkers for mono-component adhesives for high demanding applications.

Keywords: microcapsules; isocyanate; IPDI; 3-isocyanatopropyltriethoxysilane; microemulsion; interfacial polymerization; controlled release; adhesives; mono-component; eco-innovative; footwear

1. Introduction Isocyanates have been used for over 60 years in polyurethane (PU) and polychloroprene adhesives formulations, mainly because of their high reactivity and capability to promote a good adhesion. The formation of PUs occurs by the reaction between the extremely reactive polyisocyanates’ NCO groups and any chemical group that contains an active hydrogen (H) atom [1]. Current commercial polyurethane-based adhesives are bicomponent, i.e., they are composed by two components, a polyol base (OH prepolymer) and a crosslinker. High quality, strong and long-lasting adhesives used in the footwear industries are typically isocyanate-based, which provides appropriate high strength quality adhesive joints when gluing the different parts of the footwear. However, the high toxicity of isocyanate compounds is restricting their use in the industry, based on current legislation. Companies

Polymers 2018, 10, 825; doi:10.3390/polym10080825 www.mdpi.com/journal/polymers PolymersPolymers 20182018,, 1010,, 825x FOR PEER REVIEW 22 ofof 1616

However, the high toxicity of isocyanate compounds is restricting their use in the industry, based on incurrent the adhesive legislation. sector Companies are therefore in the willingadhesive to sector offer aare safe therefore adhesive willing solution to offer to these a safe industries adhesive butsolution still basedto these on industries isocyanate but chemistrystill based dueon isocya to itsnate high chemistry adhesive due efficiency. to its high This adhesive work efficiency. refers to theThis synthesis work refers of microcapsulesto the synthesis (MCs) of microcapsules with a polyurea/polyurethane (MCs) with a polyurea/polyurethane (PUa/PU) shell containing (PUa/PU) isophoroneshell containing diisocyanate isophorone (IPDI), diisocyanate as the encapsulated (IPDI), as isocyanate,the encapsulated to be added isocyanate, to footwear to be adhesivesadded to formulations.footwear adhesives The encapsulation formulations. ofThe isocyanate, encapsulation the crosslinker, of isocyanate, enables the thecrosslinker, production enables of a new the generationproduction ofof onea new component generation adhesives, of one component contributing adhesives, to an advance contributing of the to statean advance of the artof the in what state regardsof the art adhesive in what technology regards adhesive and an technology improvement and of an the improvement productivity of in the the productivity footwear industry. in the Thesefootwear novel industry. one-component These novel adhesives one-component will contain ad encapsulatedhesives will isocyanates contain encapsulated which decreases isocyanates the risk towhich health decreases by avoiding the risk the to user’s health contact by avoiding with this the toxic user’s chemical. contact Inwith addition, this toxic these chemical. adhesives In addition, will not requirethese adhesives the weighing will and not mixing require of components,the weighing which and preventsmixing manufacturingof components, errors which and prevents reduces packagingmanufacturing by 50%, errors decreasing and reduces its ecological packaging footprint. by 50%, decreasing its ecological footprint. InIn thisthis work,work, thethe IPDIIPDI waswas encapsulatedencapsulated usingusing anan interfacialinterfacial polymerizationpolymerization methodmethod inin combinationcombination withwith anan oil–in–wateroil–in–water (O/W)(O/W) microemulsionmicroemulsion technique.technique. Microencapsulation can bebe defineddefined asas aa wayway ofof coatingcoating smallsmall particles, liquid droplets or gas bubbles within a thin filmfilm or shell, whichwhich protectsprotects oror isolatesisolates thethe corecore materialmaterial fromfrom thethe surrounding environment.environment. Many different activeactive materialsmaterials suchsuch asas drugs,drugs, enzymes,enzymes, vitamins,vitamins, pesticides,pesticides, flavors,flavors, andand catalystscatalysts havehave beenbeen successfullysuccessfully encapsulatedencapsulated insideinside shellsshells mademade fromfrom aa variety of polymeric and non-polymeric materials,materials, depending onon thethe endend useuse ofof thethe encapsulatedencapsulated products.products. TheThe employedemployed techniquetechnique involvedinvolved anan emulsificationemulsification processprocess inin combinationcombination withwith interfacialinterfacial polymerization.polymerization. TheThe emulsion emulsion is obtainedis obtained through through vigorous vigorous stirring stirring of two of or two more or immisciblemore immiscible phases composedphases composed by at least by twoat least liquids, two usuallyliquids, in usually the presence in the ofpresence an emulsifier. of an emulsifier. Regarding Regarding the interfacial the polymerizationinterfacial polymerization process, it process, is composed it is composed by two steps. by two During steps. During the first the step, first thestep, diffusion the diffusion of two of differenttwo different reactive reactive monomers monomers occurs, occurs, and and they they come come in contact in contact at the at the interface, interface, i.e., i.e., atthe at the surface surface of theof the oil oil droplets droplets of of the the emulsion emulsion system, system, leading leading to to the the formation formation of of an an initial initial thinthin polymericpolymeric layerlayer shellshell [[2].2]. DuringDuring thethe secondsecond step,step, anan increaseincrease ofof thethe shellshell thicknessthickness occurs,occurs, byby thethe continuationcontinuation ofof thethe polymerizationpolymerization reactionsreactions towardstowards thethe organicorganic (dispersed)(dispersed) phase.phase. TheThe polymerization is controlled by diffusion,diffusion, soso thethe growth’sgrowth’s speed of the MCs’ shell will decrease as the shell thickness increases [3]. [3]. TheThe shellshell formationformation isis duedue toto thethe reactionreaction betweenbetween thethe –NCO–NCO groupsgroups fromfrom isocyanateisocyanate andand activeactive HH sourcessources present in in the the aqueous aqueous phase phase of of the the emulsi emulsionon system, system, which which may may be provided be provided by amines by amines (NH (NHgroups) groups) or polyols or polyols (OH (OH groups). groups). Different Different active active H Hsources sources may may have have different different chain chain length andand chemicalchemical structurestructure and,and, therefore,therefore, differentdifferent stericsteric hindrancehindrance and reactivity, whichwhich stronglystrongly influencesinfluences thethe MCs’MCs’ shellshell thickness,thickness, asas wellwell asas theirtheir morphologymorphology [[4].4]. As represented in FigureFigure1 1,, the the reaction reaction betweenbetween –NCO–NCO groupsgroups andand differentdifferent activeactive HH sourcessources cancan leadlead toto differentdifferent finalfinal compounds:compounds: whenwhen reactingreacting withwith –OH–OH groups,groups, itit leadsleads toto thethe formationformation ofof aa urethaneurethane linkage;linkage; withwith waterwater itit leadsleads toto thethe formationformation ofof anan amine;amine; andand withwith aminoamino groupsgroups itit leadsleads toto thethe formationformation ofof aa ureaurea linkage.linkage.

Figure 1. The reaction of isocyanate with hydroxylhydroxyl group, water,water, andand aminoamino group.group.

The major efforts already reported for microencapsulation of isocyanate compounds belong to The major efforts already reported for microencapsulation of isocyanate compounds belong the encapsulation of monomeric isocyanate IPDI, as a healing agent. Yang et al. reported the to the encapsulation of monomeric isocyanate IPDI, as a healing agent. Yang et al. reported the successful microencapsulation of IPDI, as a healing agent, via interfacial polymerization, with a successful microencapsulation of IPDI, as a healing agent, via interfacial polymerization, with a microencapsulation yield of 70 wt.%, taking into account the amount of IPDI and solvent [5]. In their

Polymers 2018, 10, 825 3 of 16 microencapsulation yield of 70 wt.%, taking into account the amount of IPDI and solvent [5]. In their work, a prepolymer was previously prepared using a solution of toluene diisocyanate (TDI) in cyclohexanone, followed by further addition of 1,4-butanediol as an active H source. Carefully analyzing the paper, the encapsulated isocyanate yields at ca. 35 wt.% of the MCs’ weight. The encapsulation of IPDI within MCs’ shells composed of poly(urea-formaldehyde), polyurethane, polyurea, bi-layer polyurethane/poly (urea-formaldehyde) and a double-layered shell composed of polyurea/melamine formaldehyde has already been reported [6–8]. Different compounds have also been tested as active H sources, for example poly vinyl alcohol, glycerol, 1,4-butanediol and 1,6-hexanediol [9–11]. Some of these authors reported on the impossibility of achieving MCs using commercial pre-polymers [8] and employed pre-polymers which had to be synthesized on purpose, for promoting the shell formation. In addition, other authors have reported on shelf-life issues, solvent resistance issues, or weak mechanical performance of the shells. Moreover, the yield of isocyanate encapsulation is relatively poor in most of the cases found in the literature. Indeed, the encapsulation efficiency is sometimes referred to the encapsulated isocyanate and solvent, instead of the pure isocyanate only. In the present work, IPDI was efficiently microencapsulated using a readily commercially available MDI source, Ongronat® 2500, which consists of a mixture of monomeric and polymeric MDI species, responsible for an acceptable low viscosity of the dispersed phase of the emulsion and, therefore, no solvent was required. It displays significantly higher reactivity than IPDI, favorably competes in the formation of the polymeric shell, enhancing the encapsulated IPDI content, and has not yet been described in the literature for this application. In addition, four different active H sources were tested, namely 3-(2-aminoethylamino) propyltrimethoxysilane (aminosilane) in combination with tetraethyl orthosilicate (TEOS), diethylenetriamine (DETA) and 3-(triethoxysilyl) propyl isocyanate or 3-isocyanatopropyltriethoxysilane (IPES), to achieve a faster polymerization reaction with an increased encapsulation yield, as well as durable (long shelf-life) MCs. The novelty of this work relies on: (i) optimization studies involving multifunctional active H sources and a commercially available high reactivity and low viscosity isocyanate compound (Ongronat® 2500), which eliminates the need for a solvent and increases the encapsulated pure isocyanate content; (ii) the use of aminosilane and TEOS in the synthesis, which not only contributed as an active H source, but also for a PUa/PU-silica hybrid shell; (iii) the addition of a “latent” active H source, isocyanato silane (IPES), which has both a Si-ethoxy (Si–OC2H5) and isocyanato functionality and, which can be placed in the dispersed, organic phase, contrary to common active H sources; and (iv) the application of MCs obtained through interfacial polymerization, containing IPDI, as a crosslinking agent to be applied in the adhesive industry for high demanding applications. The work reported in this paper was carried out within the framework of a Technology Platform on Microencapsulation and Immobilization led by A.C. Marques, which deals with the synthesis of MCs, microscaffolds and microspheres with tailored morphologies, organic functionalities and a wide porosity range [12–16].

2. Materials and Methods

2.1. Materials and Methods

2.1.1. Materials The MDI compound and shell precursor Ongronat® 2500 was purchased from BorsodChem. The isocyanate to be encapsulated, Isophorone diisocyanate (IPDI, with 98% purity), was obtained from Bayer. Regarding the active H source compounds, the 3-(2-aminoethylamino) propyltrimethoxysilane (with 90% purity) and tetraethyl orthosilicate (TEOS, with 99% purity) were purchased from VWR Chemicals; Diethylenetriamine (DETA) (with 99% purity) was obtained from Alfa Aesar; and the 3-isocyanatopropyltriethoxysilane (IPES, with 95% purity) was purchased from Aldrich Chemistry. Polymers 2018, 10, 825 4 of 16

The surfactant Gum Arabic (GA) was obtained from Labchem. All chemicals were used as received, without further puriﬁcation.

2.1.2. Preparation of Microcapsules The oil–in–water (O/W) emulsion system was obtained through vigorous stirring of the isocyanates, as the oil phase, consisting of 6.7 wt.% of the total emulsion system, with the water phase consisting of 95.3 wt.% Milli-Q water and 4.7 wt.% GA, at a stirring speed of 3200 rpm with an Ultra-Turrax (IKA T25 digital ULTRA TURRAX, Germany) during 10 min at room temperature (RT). The oil phase is composed by two isocyanates, being IPDI, present at 70 wt.%, the one to be encapsulated and Ongronat® 2500, present at 30 wt.%, the one to competitively contribute to the MCs’ shell formation. The IPDI is less reactive than Ongronat® 2500, which allows its encapsulation due to the faster reaction of Ongronat® 2500 with the OH groups of the aqueous phase. Once the emulsion was ready, the active H sources (aminosilane, TEOS and DETA) were added, dropwise, to the emulsion system (W phase) while under mechanical stirring at 400 rpm, using a mechanical stirrer Heidolph RZR 2051 control. IPES, on the other hand, was added to the O phase used to prepare the emulsion. The emulsification and stirring rates had been previously optimized for these microencapsulation syntheses. During this procedure, the temperature was increased gradually from RT to the maximum synthesis temperature, listed in Table1. The reactional medium was maintained at the maximum temperature, under mechanical stirring for several hours, depending on the synthesis, until the MCs shell attains enough maturity to tolerate the pressure applied during the filtration procedure. The MCs were then filtrated using a vacuum filtration system, while washed with Milli-Q water. The final product was dried at atmospheric pressure and room temperature for 24 h. It should be noted that these syntheses require a very frequent monitoring of the emulsion quality by means of optical microscopy, in order to control the stability of the emulsion, as well as the maturity degree of the shell.

Table 1. MCs’ synthesis parameters: active H source employed, maximum reactional temperature and synthesis.

MCs Water (W) Active H Source Maximum Duration Oil (O) Phase Acronym Phase (wt.% of Total Emulsion) Temp. (◦C) (Hours) Aminosilane (5.7 wt.%) + TEOS (1.5 wt.%) AT 65 4 (added to the W phase) Ongronat® DETA (5.0 wt.%) D Water + GA 70 2 2500 + IPDI (added to the W phase) IPES (1.7 wt.%) I 60 3 (added to the O phase)

2.2. Characterization The adopted techniques and methods for the MCs’ characterization include optical microscopy, scanning electron microscopy (SEM), Fourier transformed infrared spectroscopy (FTIR) and thermogravimetric analysis (TGA). The photomicrographs obtained from SEM were also used to evaluate the MCs’ size distribution using the software ImageJ. Proof of concept studies were carried out to evaluate the MCs’ performance as encapsulating agents to enable mono-component and self-reactive adhesives.

2.2.1. Optical Microscopy A Kruss, MSZ 5600 optical microscope was used to evaluate the MC’ size, shape and maturity (stiffness, qualitatively assessed by punching, or tearing with the point of a needle), as well as the stability of the emulsion and droplets’ size during the synthesis procedure. Polymers 2018, 10, 825 5 of 16

2.2.2. Scanning Electron Microscope (SEM) SEM analysis was carried out to evaluate the morphology, size distribution, roughness and porosity of the obtained MCs, using a JEOL JSM7001F (JEOL, Tokyo, Japan) with a FEG-SEM (Field Emission Gun) system. Previously to the analysis, the MCs were placed in a sample holder using conductive adhesive tape with a double face. Afterwards, the samples were coated with a conductive Au/Pd thin ﬁlm, through sputtering, using a Quorum Technologies sputter coater model Q150T ES. To evaluate the MCs’ size distribution, the images obtained through SEM were imported to the software ImageJ (v 1.51p, National Institutes of Health, Bethesda, MD, USA). After calibrating the scale, the contrast of the SEM images was enhanced, and the MCs were highlighted by a thresholding function. After treating the images to remove residual debris and agglomerates, the data for the size distribution of MCs was obtained using the Analyze Particles functionality of the software.

2.2.3. Fourier Transformed Infrared Spectroscopy (FTIR) FTIR characterization was used to study the chemical structure of the obtained MCs by identification of specific characteristic groups, in order to assess the presence of certain compounds in the MCs’ shell and core. This technique enabled to confirm the presence of unreacted NCO groups in the MCs’ core, as well as to make a comparative evaluation of the encapsulated NCO content among different syntheses. A PerkinElmer, Spectrum Two, FTIR spectrometer was used, equipped with a Pike Technologies MIRacle® Attenuated Total Reflectance (ATR) accessory. The spectra were achieved with a resolution of 4 cm−1 and data collection of 8 scans.

2.2.4. Thermogravimetric Analysis (TGA) Through TGA analysis, it was possible to quantify the amount of encapsulated compound in the obtained MCs. The analyses were performed under a controlled nitrogen atmosphere, at a temperature increase rate of 10 ◦C·min−1 using a TGA HITACHI STA 7200.

2.2.5. Peeling Strength Test The peeling strength test was essential to evaluate the adhesive effectiveness of the adhesive joint, in the bonding process. Peeling strength is the average load per unit width of the bond, required to separate bonded materials. A separation angle of 180 degrees was applied. A high peeling strength value, i.e., similar to that of the reference sample (adhesive joint with non-encapsulated IPDI), confirms the release of the microencapsulated IPDI content and its effective reaction with the surrounding environment (OH pre-polymer). For this test, MCs with encapsulated IPDI (I MCs) were added to an adhesive formulation base component, OH pre-polymer (having OH groups available to react), supplied by CIPADE S.A., to act as a crosslinker. The quantity of MCs added to the solution was such that the quantity of the encapsulated isocyanate corresponds to 5 wt.% of the OH pre-polymer (base component). The resulting mono-component adhesive material was applied as a homogeneous thin film (0.2 mm of thickness) onto previously carded leather substrates. The film was allowed to dry at RT for 15 min. The substrates were then placed on top of each other, with the adhesive faces in contact and were subjected to the conditions of pressure and temperature used in the footwear industry in the adhesive application process. For this test, infrared radiation and a manual press with electrically heated platens, Carver, were used. The samples were left to stand for 48 h at RT before carrying out the peeling strength test. A universal testing machine Instron 5566 was used at a constant rate of speed of 100 mm/min. Tests using a reference sample, namely the same OH pre-polymer but with non-encapsulated IPDI, were also performed for results comparison. Polymers 2018, 10, 825 6 of 16

3. Results and Discussion The morphology and size distribution of the MCs were assessed by SEM. Figures2–4 show the SEM photomicrographs of the AT MCs, D MCs and I MCs, respectively. Some of the MCs were crushed on purpose to make possible identifying their characteristic morphology, including their shell thickness.PolymersPolymers 2018 All 2018, 10 the, x10 FOR, x MCs FOR PEER PEER present REVIEW REVIEW a core–shell morphology, have spherical shape and do not6 of6 16 evidenceof 16 significant aggregation. An average shell thickness of 8 ± 1, 2 ± 0.5 and 3 ± 0.5 µm was found for AT shell thickness. All the MCs present a core–shell morphology, have spherical shape and do not MCs,shell D thickness. MCs and All I MCs, the MCs respectively. present a For core–shell all cases, morphology, the outer surfacehave spherical of the MCsshape is and relatively do notsmooth. evidenceevidence significant significant aggregation. aggregation. An An average average shell shell thickness thickness of 8of ± 81, ± 2 1, ± 2 0.5 ± 0.5 and and 3 ± 30.5 ± 0.5 μm μ wasm was found found Some non-uniformity and wrinkles that might be observed are due to the fast formation of a mature, forfor AT AT MCs, MCs, D MCsD MCs and and I MCs, I MCs, respectively. respectively. For For all allcases, cases, the the outer outer surface surface of theof the MCs MCs is relativelyis relatively stiff shell through the reaction of the active H sites with the highly reactive isocyanate compounds, smooth.smooth. Some Some non-uniformity non-uniformity and and wrinkles wrinkles that that might might be beobserved observed are are due due to theto the fast fast formation formation of of followeda mature,a mature, by stiff shrinking stiff shell shell through of through the corethe the reaction [17 reaction]. Another of ofthe the ac reason tiveactive H might sitesH sites with be with the the inhomogeneousthe highly highly reactive reactive isocyanate reaction isocyanate kinetics andcompounds, fluid-inducedcompounds, followed followed shear by forces byshrinking shrinking [5,11 of]. ofthe The the core inner core [17]. surface[17]. Another Another is generallyreason reason migh migh smoothert bet bethe the inhomogeneous since inhomogeneous little shear flow occursreactionreaction inside kinetics kinetics the MCsand and fluid-induce during fluid-induce thed formation sheard shear forces forces of [5,11]. the [5,11]. shell The The wall.inner inner su rfacesurface is generallyis generally smoother smoother since since littleBylittle shear optical shear flow flow microscopy, occurs occurs inside inside it the was the MCs possible MCs during during to the follow the formation formation a MCs’ of theof shell the shell formationshell wall. wall. behavior and to conclude about theBy Byoptical shortest optical microscopy, reactionmicroscopy, time, it was it was for possible eachpossible synthesis, to followto follow a whichMCs’ a MCs’ shell allows shell formation formation enough behavior maturity behavior and ofand to the concludeto shellconclude material toabout withstandabout the the shortest the shortest filtration reaction reaction step time, time, without for for each damageeach synthe synthe ofsis, thesis, which MCs.which allows Inallows Figure enough enough5, it maturity is maturity possible of ofthe to the observe shell shell that material to withstand the filtration step without damage of the MCs. In Figure 5, it is possible to thematerial MCs’ shell to withstand broke due the to filtration the pressure step without applied dama andge a significantof the MCs. amount In Figure of 5, liquid it is possible isocyanate to (IPDI) observeobserve that that the the MCs’ MCs’ shell shell broke broke due due to tothe the pres pressuresure applied applied and and a significanta significant amount amount of ofliquid liquid was released from the MCs, confirming the success of the encapsulation process. It should be stressed isocyanateisocyanate (IPDI) (IPDI) was was released released from from the the MCs, MCs, confir confirmingming the the success success of theof the encapsulation encapsulation process. process. It It thatshould thereshould be is bestressed no stressed solvent that that insidethere there is the no is no capsules.solvent solvent inside inside the the capsules. capsules.

FigureFigure 2. SEM2. SEM image image of ATof AT MCs MCs (Active (Active H source:H source: aminosilane aminosilane and and TEOS). TEOS). The The image image on onthe the right right Figure 2. SEM image of AT MCs (Active H source: aminosilane and TEOS). The image on the right displays displaysdisplays a broken a broken MC MC which which demonstrates demonstrates the the core core–shell–shell morphology morphology typical typical of these of these MCs. MCs. Scale Scale bar bar a broken MC which demonstrates the core–shell morphology typical of these MCs. Scale bar = 10 µm. = 10= μ10m. μ m.

FigureFigure 3. SEM 3. SEM image image of Dof MCsD MCs (Active (Active H source:H source: DETA). DETA). The The image image on onthe the right right displays displays a broken a broken Figure 3. SEM image of D MCs (Active H source: DETA). The image on the right displays a broken MCMC which which demonstrates demonstrates the the core–shell core–shell morpho morphologylogy typical typical of theseof these MCs. MCs. Scale Scale bar bar = 10 = μ10m. μ m. MC which demonstrates the core–shell morphology typical of these MCs. Scale bar = 10 µm.

Polymers 2018, 10, 825 7 of 16 Polymers 2018, 10, x FOR PEER REVIEW 7 of 16 Polymers 2018, 10, x FOR PEER REVIEW 7 of 16

Figure 4. SEM image of I MCs (Active H and NCO source: isocyanato silane, IPES). The image on the Figure 4. SEM image of I MCs (Active H and NCO source: isocyanato silane, IPES). The image on right displays a broken MC which demonstrates the core–shell morphology typical of these MCs. Figurethe right 4. displaysSEM image a broken of I MCs MC (Active which H demonstrates and NCO source: the core–shell isocyanato morphology silane, IPES). typical The ofimage these on MCs. the rightScale bardisplays = 10 μ µam. broken MC which demonstrates the core–shell morphology typical of these MCs. Scale bar = 10 μm.

Figure 5. Optical microscopy images: (a) I MCs before tearing of the shell by the point of a needle; and (b) the same image after breaking the shell, exhibiting release of the core content (IPDI). Figure 5. Optical microscopy images: images: ( (aa)) I I MCs MCs before before tearing tearing of of the the shell shell by by the the point point of of a a needle; needle; and ( b) the same image after breaking the shell, exhibitingexhibiting releaserelease ofof thethe corecore contentcontent (IPDI).(IPDI). Through the analysis of the size distribution of the MCs, using the ImageJ software, it was possibleThrough to conclude the analysis that the of particles the size have distribution an average of diameterthe MCs, of using 95 ± 7,the 81 ImageJ± 4 and 82software, ± 7 μm forit was the Through the analysis of the size distribution of the MCs, using the ImageJ software, it was possibleAT MCs, to D conclude MCs and that I MCs the samples,particles respectively.have an average For diameterthe stirring of 95rate ± 7,employed 81 ± 4 and (400 82 ±rpm), 7 μm the for size the possible to conclude that the particles have an average diameter of 95 ± 7, 81 ± 4 and 82 ± 7 µm ATof the MCs, MCs D isMCs herein and found I MCs to samples, be smaller respectively. than that Foreportedr the stirring in the rateliterature, employed for instance (400 rpm), by Yangthe size et for the AT MCs, D MCs and I MCs samples, respectively. For the stirring rate employed (400 rpm), ofal. the[5]. MCsIndeed, is herein besides found shear/stirring to be smaller rate, itthan is well that known reported that in the the size literature, of MCs foris influenced instance by by Yang several et the size of the MCs is herein found to be smaller than that reported in the literature, for instance by al.factors [5]. Indeed, including besides the viscosity shear/stirring of the rate,dispersed it is well and known dispersant that thephase, size interfacial of MCs is influencedtension of theby severalmedia, Yang et al. [5]. Indeed, besides shear/stirring rate, it is well known that the size of MCs is influenced factorsthe geometry including of thethe viscosity mixing ofdevice, the dispersed blade hydrod and dispersantynamics, phase, temperature, interfacial and tension surfactant of the media,effects by several factors including the viscosity of the dispersed and dispersant phase, interfacial tension the[5,11,18,19]. geometry In allof cases,the mixing it is possible device, to stateblade that hydrod the particleynamics, size temperature, distribution closely and surfactant follows a normaleffects of the media, the geometry of the mixing device, blade hydrodynamics, temperature, and surfactant [5,11,18,19].distribution, In because all cases, all it theis possible distributions to state had that thethe particlecharacteristic size distribution bell-shaped closely curve, follows as observed a normal in effects [5,11,18,19]. In all cases, it is possible to state that the particle size distribution closely follows a distribution,Figures 6–8, andbecause the pall-values the distributions obtained through had the the ch Shapiro–Wilkaracteristic bell-shaped normality curve,test, listed as observed in Table in2, normal distribution, because all the distributions had the characteristic bell-shaped curve, as observed Figureswere all 6–8,higher and than the thep-values defined obtained alpha value through of 0.05 the, notShapiro–Wilk rejecting the normality null hypothesis test, listed that inthe Table values 2, in Figures6–8, and the p-values obtained through the Shapiro–Wilk normality test, listed in Table2, wereare normally all higher distributed. than the defined alpha value of 0.05, not rejecting the null hypothesis that the values were all higher than the defined alpha value of 0.05, not rejecting the null hypothesis that the values are normallyThe FTIR distributed. spectra of IPDI and the IPDI loaded MCs were obtained to further reveal both the are normally distributed. compositionThe FTIR and spectra the IPDI of IPDI encapsulat and theion IPDI effectiveness. loaded MCs From were the obtained FTIR spectra, to further shown reveal in Figure both 9,the it The FTIR spectra of IPDI and the IPDI loaded MCs were obtained to further reveal both the compositionis possible to andobserve the IPDIan intense encapsulat bandion peaked effectiveness. at ca. 2260 From cm− 1the, related FTIR to spectra, the N=C=O shown bond in Figure stretching 9, it composition and the IPDI encapsulation effectiveness. From the FTIR spectra, shown in Figure9, isvibration, possible which to observe indicates an intense the presence band peaked of unreacted at ca. 2260NCO cm groups−1, related in the to MCs, the N=C=O confirming bond a successfulstretching it is possible to observe an intense band peaked at ca. 2260 cm−1, related to the N=C=O bond vibration,encapsulation which of indicatesIPDI. The the peaks presence ascribed of unreacted to the presence NCO groups of amine in thegroups MCs, can confirming be observed a successful at 3200– stretching vibration, which indicates the presence of unreacted NCO groups in the MCs, confirming a encapsulation3400 cm−1 from of N–H IPDI. stretching The peaks of theascribed amine to bonds the pr andesence at 1510 of amine cm−1 fromgroups N–H can bending. be observed The presenceat 3200– 3400of amine cm−1 groups from N–H in the stretching MCs can of be the related amine with bonds the and presence at 1510 of cm PUa−1 from in the N–H shell bending. composition, The presence derived ofmainly amine from groups the inreaction the MCs of canNCO be groups related with with the the active presence H sources of PUa inused the in shell the composition,synthesis of AT derived MCs mainly from the reaction of NCO groups with the active H sources used in the synthesis of AT MCs

Polymers 2018, 10, 825 8 of 16 successful encapsulation of IPDI. The peaks ascribed to the presence of amine groups can be observed at 3200–3400 cm−1 from N–H stretching of the amine bonds and at 1510 cm−1 from N–H bending. Polymers 2018, 10, x FOR PEER REVIEW 8 of 16 ThePolymers presence 2018, 10, ofx FOR amine PEER groups REVIEW in the MCs can be related with the presence of PUa in the8 shell of 16 composition, derived mainly from the reaction of NCO groups with the active H sources used in the synthesisandand DD MCs,MCs, of AT namelynamely MCs aminosilaneaminosilane and D MCs, and namelyand DETA,DETA, aminosilane respectively.respectively. and TheThe DETA, carbonylcarbonyl respectively. groupsgroups Theobservedobserved carbonyl atat ca.ca. groups 1715–1715– 1730 cm−1 (C=O from urethane)−1 and at 1680–1700 cm−1 (C=O from urea),− the1 C=C groups detected at observed1730 cm−1 at (C=O ca. 1715–1730 from urethane) cm (C=O and at from 1680–1700 urethane) cm and−1 (C=O at 1680–1700 from urea), cm the(C=O C=C from groups urea), detected the C=C at 1522 cm−1, the C–O–C group− 1at 1214 cm−1 and C-O stretching− at1 ca. 1300 cm−1 confirm the presence− of1 groups1522 cm detected−1, the C–O–C at 1522 group cm at, the1214 C–O–C cm−1 and group C-O at stretching 1214 cm at andca. 1300 C-O cm stretching−1 confirm at the ca. presence 1300 cm of conﬁrmurethaneurethane the andand presence ureaurea linkages,linkages, of urethane evidencingevidencing and urea thatthat linkages, thethe MCs’MCs’ evidencing shellshell isis composed thatcomposed the MCs’ byby shellPUPU andand is composed PUa.PUa. Moreover,Moreover, by PU andthethe peak PUa.peak Moreover,assignedassigned toto the thethe peak carbonylcarbonyl assigned groupgroup to the ofof carbonylureaurea isis moremore group intenseintense of urea thanthan is more thethe intensesamesame peakpeak than relatedrelated the same toto urethane bonds, which means that the majority of the shell material is indeed polyurea. peakurethane related bonds, to urethane which means bonds, that which the means majority that of the the majority shell material of the shellis indeed material polyurea. is indeed polyurea.

Table 2. p-values for AT MCs, D MCs and I MCs analysis, obtained through the Shapiro–Wilk TableTable 2.2. p-values for AT MCs,MCs, DD MCsMCs andand II MCsMCs analysis,analysis, obtainedobtained throughthrough thethe Shapiro–WilkShapiro–Wilk normality test. normalitynormality test.test. MCs Shapiro–Wilk p-Value MCs Shapiro– Shapiro–WilkWilk p-Valuep-Value ATAT 0.3020.302 ATD 0.336 0.302 DD 0.3360.336 III 0.5780.5780.578

AT MCs AT MCs

Frequency (a.u.) Frequency Frequency (a.u.) Frequency

0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 Diameter (μm) Diameter (μm)

Figure 6. Size distribution of AT MCs. FigureFigure 6.6. Size distribution ofof ATAT MCs.MCs.

D MCs D MCs

Frequency (a.u.) Frequency Frequency (a.u.) Frequency

0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 Diameter (μm) Diameter (μm)

FigureFigure 7. 7.Size Sizedistribution distributionof ofD D MCs. MCs. Figure 7. Size distribution of D MCs.

PolymersPolymers2018 2018, 10, ,10 825, x FOR PEER REVIEW 9 of9 of 16 16 Polymers 2018, 10, x FOR PEER REVIEW 9 of 16

I MCs I MCs

Frequency (a.u.) Frequency (a.u.)

0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 μ Diameterμ ( m) Diameter ( m)

FigureFigure 8. 8.Size Size distribution distribution of of I MCs. I MCs. Figure 8. Size distribution of I MCs.

1.0 1.0

0.8 0.8

0.6 0.6

0.4 0.4

Transmittance (a.u.) N C O Transmittance (a.u.) N C O 0.2 0.2

AT MCs AT MCs 0.0 D MCs 0.0 D MCs I MCs I MCs 4000 3500 3000 2500 2000 1500 1000 4000 3500 3000 2500 2000 1500 1000 -1 Wavenumber (cm-1 ) Wavenumber (cm )

Figure 9. FTIR spectra of MCs obtained in this work. The inset shows a magnification of the peaks FigureFigure 9.9.FTIR FTIR spectraspectra ofofMCs MCsobtained obtainedin inthis thiswork. work. TheThe insetinset showsshows aamagniﬁcation magnification ofof thethepeaks peaks assigned to isocyanate NCO bond stretching and to the carbonyl (C=O) group present in urea and assignedassigned toto isocyanateisocyanate NCONCO bondbond stretchingstretching andand toto thethecarbonyl carbonyl (C=O)(C=O) groupgroup presentpresent inin ureaurea andand urethane moieties. urethaneurethane moieties. moieties. The peak at 1070 cm−1, typical of Si–O–Si asymmetric stretching vibrations, confirms the presence The peak at 1070 cm−−1, typical of Si–O–Si asymmetric stretching vibrations, confirms the presence of siloxaneThe peak Si–O–Si at 1070 cmmoieties, typical in the of shell, Si–O–Si derived asymmetric from the stretching silanes vibrations,employed conﬁrmsin the synthesis, the presence in the of siloxane Si–O–Si moieties in the shell, derived from the silanes employed in the synthesis, in the ofcases siloxane of AT Si–O–Si MCs moieties and I MCs. in the During shell, derived the synthesi from thes of silanes AT MCs, employed the aminosilane in the synthesis, and inTEOS, the cases when cases of AT MCs and I MCs. During the synthesis of AT MCs, the aminosilane and TEOS, when ofincorporated AT MCs and Iinto MCs. the During aqueous the synthesis phase, oftend AT MCs,to hydrolyze, the aminosilane forming and TEOS,silanol whengroups incorporated which, by incorporated into the aqueous phase, tend to hydrolyze, forming silanol groups which, by intopolycondensation the aqueous phase, reactions tend form to hydrolyze, siloxane mo formingieties (FTIR silanol peak groups assignment which, by at polycondensationca. 1070 cm−1), and polycondensation reactions form siloxane moieties (FTIR peak assignment−1 at ca. 1070 cm−1), and reactionswhen located form siloxaneat the droplets’ moieties surface, (FTIR peakcan also assignment react with at NCO ca. 1070 groups cm genera), andting when urethane located moieties. at the when located at the droplets’ surface, can also react with NCO groups generating urethane moieties. droplets’Moreover, surface, the NH can groups also react from with the NCO aminosilane groups generatingtend to react urethane with the moieties. NCO groups Moreover, at thethe interface NH Moreover, the NH groups from the aminosilane tend to react with the NCO groups at the interface groupsof the from aqueous/organic the aminosilane phas tendes, originating to react with urea the NCOmoieties. groups The at presen the interfacece of silanes, of the aqueous/organicas active H source, of the aqueous/organic phases, originating urea moieties. The presence of silanes, as active H source, phases,is found originating to result urea in a moieties. PUa/PU–silica The presence hybrid ofshell silanes, material. as active TheH siloxane source, ismoieties found toare result known ina to is found to result in a PUa/PU–silica hybrid shell material. The siloxane moieties are known to PUa/PU–silicacontribute to an hybrid improved shell material.mechanical The strength siloxane and moieties thermal are resistance known to of contribute the resulting to an shell. improved contribute to an improved mechanical strength and thermal resistance of the resulting shell. mechanical strength and thermal resistance of the resulting shell.

Polymers 2018, 10, 825 10 of 16

Regarding the synthesis of D MCs, at which DETA was added as an active H source, the success of the IPDI´s encapsulation can be explained by the reaction of the amine groups of DETA with –NCO groups of the Ongronat® 2500, leading to a fast formation of a PUa shell. For the I MCs synthesis, IPES was added to the organic phase. IPES contains a double organic functionality, namely –NCO and ethoxy (OCH2CH3) groups, and can be used as a crosslinker, contributing to the maturity and cohesiveness of the shell. In fact, once the NCO groups from IPES react with the OH groups from the aqueous phase, this molecule will be kept at the interface of the organic/aqueous phases, where the IPES´s alkoxide groups (OCH2CH3) suffer hydrolysis, forming silanol groups, which will react with the NCO groups from Ongronat® 2500. The fact makes IPES to be considered a “latent” active H source, which also displays NCO functionality. In this sense, one can state that IPES acts as an active H source, which can be added to the organic (isocyanate containing) phase, contrary to most of the active H sources, establishing links with either the organic or the aqueous phases, due to its double organic functionality. The presence of IPES inside the tiny droplets of the emulsion, at a concentration of 25 wt.% in the O phase, results in a significant concentration at the interface of the organic/aqueous phases, rather than if it had been added to the aqueous phase, which accounts for 93.3 wt.% of the total emulsion mass. The NCO groups from Ongronat® 2500 also react with the aqueous phase, contributing to the formation of the shell material. Further calculation was carried out to estimate a relative measure of the isocyanate encapsulation efficiency, by considering the peak at ca. 2260 cm−1 assigned to NCO stretching (area in the range of 1926 to 2444 cm−1) and the area of the peak at 1300 cm−1 assigned to C–O stretching, typically related to the PUa/PU shell material, which does not tend to suffer significant changes over time after synthesis of the MCs. The peaks area was calculated using OriginPro 9 software, and the relative encapsulation yields were calculated based on the equation below:

Area −1 Y = 2260 cm Area 1300 cm−1 where Y can be considered as a relative encapsulation yield, which represents an indirect measure of the encapsulation efficiency. Area 2260 cm−1 and Area 1300 cm−1 represent the areas of the peak related to the NCO group and the area of the peak related to C-O stretch from the shell material, respectively. The obtained Y values are 13.75, 5.84 and 18.48 for AT MCs, D MCs and I MCs, respectively, as reported in Table3. The MCs synthesized using IPES as the active H source are the ones shown to have more IPDI encapsulated, which might be correlated with a fast reaction between the active H source and the NCO groups of Ongronat® 2500, leading to a faster shell formation which creates a barrier between the NCO groups from IPDI and the surrounding water medium. Indeed, the incorporation of IPES, as a “latent” active H source, inside the tiny droplets of the microemulsion, is shown to be an effective strategy for the desired high encapsulation yield. Its NCO group is responsible for its concentration at the interface of the organic/aqueous phases, being therefore highly available to react through its double organic functionalities, establishing links with both the isocyanate phase and the aqueous phase. On the other hand, the remaining active H sources, employed in the synthesis of AT MCs and D MCs, were added to the dispersant (W) phase, which creates a “dilution” effect (since the W phase consists of 93.3 wt.% of the total emulsion), requiring some time for them to migrate to the emulsion droplet surface to react with the NCO groups. The aminosilane and TEOS have the advantage, in comparison to DETA, of having more affinity to the oil phase, before the hydrolysis of the alkoxy groups occur, leading to a faster migration to the oil droplets and consequent reaction with the NCO groups, despite the slower reaction kinetics of the silanol groups with the NCO groups in comparison with NH (from DETA). DETA, on the other hand, has an affinity to the aqueous phase and not so much tendency to move toward the isocyanate droplets surface. Shelf-life studies were conducted on the I MCs (Figure 10) to confirm their moisture barrier capability. FTIR spectra of the MCs were taken at three different times after the MCs’ synthesis: right after the synthesis, and ten days and three months later. As it is possible to observe, the intensity Polymers 2018, 10, 825 11 of 16 ofPolymers the NCO 2018, 10 peak, x FOR remains PEER REVIEW nearly constant over time, indicating a similar amount of liquid isocyanate11 of 16 inside the MCs and a good barrier effect by the shell. The NCO peak shape and wavenumber, observedobserved forfor thethe II MCs spectra, when compared to that of thethe isocyanateisocyanate species species employed employed in thein the synthesis, synthesi revealss, reveals that thethat encapsulated the encapsulated isocyanate isocyanate is indeed is ® IPDI,indeed since IPDI, the since NCO the peak NCO of peak the Ongronat of the Ongronat® 2500 is2500 much is widermuch thanwider the than one the observed one observed in the Iin MCs the spectra,I MCs spectra, and the and NCO the peak NCO from peak IPES from presents IPES pr aesents deviation a deviation to larger to wavenumbers.larger wavenumbers. TGA studies enabled the quantificationquantification of the liquidliquid isocyanate in the MCs’ samples. Generally, Generally, waterwater andand otherother solventssolvents evaporationevaporation occuroccur duringduring aa firstfirst step,step, belowbelow 100100 ◦°C.C. The second degradation stepstep is ascribed ascribed to to the the decomposition decomposition of of encapsulated encapsulated monomeric monomeric isocyanate, isocyanate, followed followed by a by third a third and andlast step last related step related to the topolymeric, the polymeric, or hybrid, or shell hybrid, decomposition. shell decomposition. It should be It stressed should that be stressed there is thatan overlap, there is resulting an overlap, from resulting the decomposition from the decomposition of isocyanate compounds of isocyanate (monomers, compounds oligomers, (monomers, and oligomers,pre-polymer and species), pre-polymer PU and species), PUa shell PU material, and PUa so shell there material, is not an soexact there borderline is not an to exact separate borderline them. toThere separate are references them. There which are mention references that whichthe PU mention decomposition that the starts PU decompositionaround 200 °C [20], starts however, around 200it was◦C[ observed20], however, that the it wasdecomposition observed that of isocyanate the decomposition also occurs of around isocyanate this alsotemperature occurs around [5,9,21]. this In temperatureaddition, it [should5,9,21]. Inbe addition,noted that it should the decomp be notedosition that the temperature decomposition range temperature for the encapsulated range for the encapsulatedisocyanate compounds isocyanate is compounds slightly different is slightly from different that of fromnon-encapsulated that of non-encapsulated ones, and such ones, difference and such differencewill depend will on depend the shell on thickness, the shell thickness,composition, composition, and structure and [20,22]. structure [20,22].

Ongronat(R) 2500

IPDI

Isocyanato triethoxysilane (IPES)

I MCs (after 3 months)

I MCs (after 10 days)

Transmittance (a.u.) I MCs (as prepared)

N=C=O stretching

4000 3500 3000 2500 2000 1500 1000 -1 Wavenumber (cm )

Figure 10. FTIR spectra spectra of of the the I IMCs MCs as as prepared, prepared, and and 10 10 da daysys and and three three months months after after their their synthesis, synthesis, to ® tostudy study the the MCs MCs shelf shelf life. life. The The spec spectratra of ofIPES, IPES, IPDI IPDI and and Ongronat Ongronat2500® 2500 are arealso also represented represented for forcomparison. comparison.

Figures 11 and 12 show the thermograms of the MCs, of the isocyanate species used in the Figures 11 and 12 show the thermograms of the MCs, of the isocyanate species used in the syntheses and of the respective shells, as well as the corresponding first derivative curves. Indeed, it syntheses and of the respective shells, as well as the corresponding ﬁrst derivative curves. Indeed, can be observed that, in the MCs’ shell thermograms, the PUa/PU decomposition tends to initiate at it can be observed that, in the MCs’ shell thermograms, the PUa/PU decomposition tends to initiate around 200 °C, as referred in the literature; however, it is only at 300 °C that a major mass loss is at around 200 ◦C, as referred in the literature; however, it is only at 300 ◦C that a major mass loss is observed. The step of weight loss, in the range of ca. 150–300 °C, only present in the MCs observed. The step of weight loss, in the range of ca. 150–300 ◦C, only present in the MCs thermograms, thermograms, can be referred to the decomposition of encapsulated isocyanate compounds, but some can be referred to the decomposition of encapsulated isocyanate compounds, but some care in the care in the interpretation of the results should be taken because there starts to be an overlap with the interpretation of the results should be taken because there starts to be an overlap with the shell material shell material decomposition for temperatures above 200 °C, as observed in the first derivative decomposition for temperatures above 200 ◦C, as observed in the ﬁrst derivative curves. Contrary to curves. Contrary to some of the references found in the literature [8], the organic phase in the present study has no solvent at all, so that one can state that the mass loss observed above ca. 150 °C and below 300 °C is mainly due to pure encapsulated isocyanate. In fact, the decomposition temperature

Polymers 2018, 10, 825 12 of 16 somePolymers of 2018 the, references10, x FOR PEER found REVIEW in the literature [8], the organic phase in the present study has no solvent12 of 16 at all, so that one can state that the mass loss observed above ca. 150 ◦C and below 300 ◦C is mainly duefor pure to pure IPDI encapsulated is found to isocyanate.start around In 100 fact, °C, the and decomposition it has been reported temperature to occur for pure in the IPDI range is found of 120– to start250 °C around [5], which 100 ◦ C,is found and it hasto happen been reported in the thermograms to occur in the of range Figures of 120–25011 and 12.◦C[ 5], which is found to happenIt should in the thermograms be noted that of the Figures thermograms 11 and 12. exhibit different slopes between 150 and 300 °C, indicatingIt should that bepossibly noted there that theare thermogramsisocyanates in exhibitdifferent different stages of slopes polymerization, between 150 i.e., and monomeric 300 ◦C, indicating(IPDI and MDI) that possibly species and there oligom are isocyanateseric (pre-polymeric) in different species. stages of polymerization, i.e., monomeric (IPDIRegarding and MDI) speciesthe shell, and its oligomeric decomposition (pre-polymeric) starts around species. 300 °C and two steps can be observed in the thermograms:Regarding the a shell, first step its decomposition (300–370 °C) related startsaround to the decomposition 300 ◦C and two of steps soft segments can be observed and a second in the thermograms:step (above 370 a first°C) related step (300–370 to the decomposition◦C) related to the of decompositionhard polymeric of segments soft segments from PU and and a second PUa parts step (aboveof the 370shell◦C) [22]. related These to thesteps decomposition correspond to of hardfirst polymericderivative segmentscurves peaked from PU at andca. 345 PUa and parts 440 of the°C. shellConsidering [22]. These these steps findings, correspond it was to possible first derivative to calcul curvesate the peakedamount at of ca. encapsulated 345 and 440 isocyanate◦C. Considering in the theseAT MCs, findings, D MCs it and was I possibleMCs, which to calculate correspond the to amount 52.8%, of13.95%, encapsulated and 63.3%, isocyanate respectively, in the as ATreported MCs, Din MCsTable and 3. I MCs, which correspond to 52.8%, 13.95%, and 63.3%, respectively, as reported in Table3.

FigureFigure 11.11. (a) ThermogramsThermograms of thethe ATAT MCs,MCs, thethe respectiverespective shellshell andand thethe isocyanatesisocyanates usedused inin thethe synthesissynthesis ((toptop)) andand respectiverespective ﬁrstfirst derivativederivative curvescurves ((bottom);); andand ((bb)) thermogramsthermograms ofof thethe DD MCs,MCs, thethe respectiverespective shellshell andand thethe isocyanatesisocyanates usedused inin thethe synthesissynthesis ((toptop)) andand respectiverespective ﬁrstfirst derivativederivative curvescurves ((bottombottom).). As shown in Table3, the trend obtained through the FTIR spectra analysis is corroborated by As shown in Table 3, the trend obtained through the FTIR spectra analysis is corroborated by the TGA results, regarding the amount of encapsulated isocyanate, being I MCs the ones with higher the TGA results, regarding the amount of encapsulated isocyanate, being I MCs the ones with higher IPDI encapsulation. IPDI encapsulation.

Polymers 2018, 10, 825 13 of 16 Polymers 2018, 10, x FOR PEER REVIEW 13 of 16

Table 3. Relative encapsulation yield (Y) and mass loss (%) of encapsulated IPDIIPDI forfor eacheach synthesis.synthesis.

Mass Loss (%)(%) fromfrom TGATGA MCsMCs Relative Relative Encapsulation Encapsulation Yield Yield (Y) from (Y) from FTIR FTIR (Amount(Amount of Pure EncapsulatedEncapsulated Isocyanate) Isocyanate) AT AT13.75 13.75 52.80 52.80 D D5.84 5.84 13.95 13.95 I I18.48 18.48 63.30 63.30

Figure 12. Thermogram of the I MCs, the respective shellshell and the isocyanates used in the synthesis (top) and respective ﬁrstfirst derivativederivative curvescurves ((bottombottom).).

As discussed before, the peeling strength test was performed as an indirect way to evaluate the As discussed before, the peeling strength test was performed as an indirect way to evaluate the behavior of the MCs in terms of isocyanate release, crosslinking effectiveness and resulting adhesion behavior of the MCs in terms of isocyanate release, crosslinking effectiveness and resulting adhesion capacity. The adhesive formulation base component (OH pre-polymer), when tested without any capacity. The adhesive formulation base component (OH pre-polymer), when tested without any isocyanate compound, exhibits an adhesion performance by itself, but it leads to an adhesive type of isocyanate compound, exhibits an adhesion performance by itself, but it leads to an adhesive type failure when performing the peeling test. It should be noted that an effective crosslinking, by the reaction with NCO groups, is expected to lead to a structural type of failure of the adhesive joint, which is the one that denotes a stronger adhesion performance. This is critical for the high adhesive

Polymers 2018, 10, 825 14 of 16 of failure when performing the peeling test. It should be noted that an effective crosslinking, by the reactionPolymers 2018 with, 10 NCO, x FOR groups,PEER REVIEW is expected to lead to a structural type of failure of the adhesive14 joint, of 16 which is the one that denotes a stronger adhesion performance. This is critical for the high adhesive performanceperformance andand durability,durability, desireddesired inin thethe footwearfootwear industry.industry. TheThe averageaverage loadload measuredmeasured perper unitunit widthwidth ofof thethe bond,bond, atat thethe peelingpeeling strengthstrength tests,tests, isis listedlisted inin TableTable4 ,4, together together with with the the type type of of failure failure observedobserved during during the the tests.tests. TheThe samesame quantityquantity ofof isocyanate,isocyanate, eithereither encapsulatedencapsulated(in (in thethe formform ofof II MCs)MCs) oror non-encapsulated,non-encapsulated, was was addedadded toto thethe OHOH pre-polymer,pre-polymer, and and thethe resultingresulting adhesiveadhesive jointsjoints werewere testedtested andand comparedcompared withwith thethe OHOH pre-polymerpre-polymer (base(base component)component) withoutwithout anyany isocyanateisocyanate (crosslinker).(crosslinker). ItIt isis shown,shown, inin FigureFigure 13 13,, thatthat bothboth adhesiveadhesive jointsjoints havinghaving IPDI,IPDI, eithereither encapsulatedencapsulated oror non-encapsulated,non-encapsulated, exhibitexhibit thethe same peeling peeling strength, strength, i.e., i.e., an an average average load load per per unit unit width width of bond of bond of ca. of 3 ca.N/mm, 3 N/mm, which whichdenotes denotes that the that MCs the are MCs releasing are releasing the IPDI the under IPDI underthe thermal the thermal and pressure and pressure stimuli, stimuli, employed employed during duringthe sample the sample preparation, preparation, following following the procedure the procedure used used in inthe the industry. industry. According According to internalinternal informationinformation fromfrom CIPADECIPADE S.A.,S.A., anan adhesive,adhesive, toto bebe allowedallowed forfor commercialization,commercialization, mustmust exhibitexhibit aa peelingpeeling strength strength value value equal equal or or superior superior to 3to N/mm. 3 N/mm. Moreover, Moreover, a structural a structural and and cohesive cohesive failure failure of the of adhesivethe adhesive joint joint was observedwas observed for the for sample the sample with I MCs,with whichI MCs, is which an indicator is an indicator of these MCs of these potential MCs forpotential the current for the application, current application, i.e., for the i.e., next for generation the next generation of mono-component of mono-component adhesives. adhesives.

140

120

100

60 Load (N) Load

20 adhesive joint with IPDI adhesive joint with microencapsulated IPDI (I MCs) 0 0 20406080100120 Propagation Extension (mm)

FigureFigure 13.13.Peeling Peeling strengthstrength teststestsresult. result. LoadLoad versusversus propagationpropagation extensionextension forfor adhesiveadhesive jointsjoints withwith IPDIIPDI andand microencapsulated microencapsulated IPDI IPDI (I (I MCs), MCs), as as crosslinkers. crosslinkers.

Table 4. Peeling strength tests’ results. Table 4. Peeling strength tests’ results. Crosslinker Added to the Average Load Per Unit Type of Failure Observed in the CrosslinkerOH Pre-Polymer Added to the OH AverageWidth Load of Per the Unit Bond Width Type ofPeeling Failure ObservedStrength inTest the Peeling Pre-Polymernone of <2 the N/mm Bond Adhesive, at the Strengthsubstrate/adhesive Test interface noneIPDI 2.97 <2 N/mm N/mm Adhesive, Cohesive at the, through substrate/adhesive the adhesive interface IPDI 2.97 N/mm Cohesive, through the adhesive Microencapsulated IPDI (I MCs) 2.99 N/mm Structural and cohesive rupture Microencapsulated IPDI (I MCs) 2.99 N/mm Structural and cohesive rupture

4. Conclusions 4. Conclusions Perfectly spherical, core–shell and free-flowing MCs, of PUa/PU polymeric shell or PUa/PU- Perfectly spherical, core–shell and free-ﬂowing MCs, of PUa/PU polymeric shell or PUa/PU-silica silica hybrid shell, with average diameters ranging from 81 ± 4 to 95 ± 7 μm, were synthesized using hybrid shell, with average diameters ranging from 81 ± 4 to 95 ± 7 µm, were synthesized using an an O/W emulsion combined with interfacial polymerization and exhibit a high loading of pure IPDI O/W emulsion combined with interfacial polymerization and exhibit a high loading of pure IPDI in in the core. It was concluded that an enhanced encapsulation of pure IPDI isocyanate (above 60 wt.% the core. It was concluded that an enhanced encapsulation of pure IPDI isocyanate (above 60 wt.% of the MCs’ weight) occurs when the following strategies are used together: (a) mixture of IPDI with a small amount of commercially available Ongronat® 2500, an isocyanate source that consists of monomeric and polymeric MDI species, and exhibits a significantly higher reactivity than IPDI, contributing for the fast shell formation; and (b) incorporation of a “latent” H active source, isocyanato triethoxysilane (IPES), in the organic phase (tiny droplets of the emulsion), which tends

Polymers 2018, 10, 825 15 of 16

of the MCs’ weight) occurs when the following strategies are used together: (a) mixture of IPDI with a small amount of commercially available Ongronat® 2500, an isocyanate source that consists of monomeric and polymeric MDI species, and exhibits a signiﬁcantly higher reactivity than IPDI, contributing for the fast shell formation; and (b) incorporation of a “latent” H active source, isocyanato triethoxysilane (IPES), in the organic phase (tiny droplets of the emulsion), which tends to be mainly concentrated at the interface of the organic/aqueous phases, and establishes links either with the aqueous phase (through its NCO groups) or with the organic phase (through its Si-OH groups). These PUa/PU-silica hybrid shell MCs exhibited approximately the same isocyanate content three months after their synthesis, demonstrating an effective barrier capacity of the shell. Their performance as crosslinkers in adhesive formulations was tested through peeling strength analyses and compared with adhesive joints having the same amount of non-encapsulated IPDI. Samples containing the I MCs were found to exhibit the same peeling strength as the samples with non-encapsulated IPDI, which reveals the release and crosslinking ability of the encapsulated IPDI. Additionally, the I MCs were responsible for a structural failure of the adhesive joint, which illustrates the potential of these MCs as crosslinkers in mono-component, self-reactive, safer and eco-innovative adhesives for highly demanding applications.

Author Contributions: Conceptualization, M.A. and A.C.M.; Formal analysis, M.A., M.V.L., M.d.V., J.A.D.C., I.P. and A.C.M.; Funding acquisition, I.P., J.C.B. and A.C.M.; Investigation, M.A., M.V.L. and A.C.M.; Methodology, M.A., M.V.L., I.P., J.C.B. and A.C.M.; Project administration, I.P., J.C.B. and A.C.M.; Resources, I.P., J.C.B. and A.C.M.; Supervision, A.C.M.; Validation, M.A., M.V.L. and A.C.M.; Visualization, M.A., M.V.L., M.d.V. and J.A.D. Condeço; Writing—original draft, M.A., M.V.L. and M.d.V.; and Writing—review and editing, M.V.L., M.d.V., I.P., J.C.B. and A.C.M. Funding: This research was developed in the framework of the Co-Promotion Project # 17930, “ECOBOND— Development of new ecological, self-reactive, monocomponent adhesives”, which is funded by the Regional Operational Program of Lisbon—LISBOA2020, in the scope of Portugal2020, through the European Regional Development Fund—FEDER. Acknowledgments: This research was developed in the framework of the Co-Promotion Project # 17930, “ECOBOND—Development of new ecological, self-reactive, monocomponent adhesives”, which is funded by the Regional Operational Program of Lisbon—LISBOA2020, in the scope of Portugal2020, through the European Regional Development Fund—FEDER. The authors would like to thank Prof. Ricardo Simões for fruitful discussions on this topic. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Langenberg, K.V.; Warden, P.; Adam, C.; Milner, H.R. The Durability of Isocyanate-Based Adhesives under Service in Australian Conditions. The Results from a 3 Year Exposure Study Accelerated Testing Regime (Literature Review); Forest & Wood Products Australia Limited: Melbourne, Australia, 2010. 2. Hu, M.; Guo, J.; Yu, Y.; Cao, L.; Xu, Y. Research Advances of Microencapsulation and Its Prospects in the Petroleum Industry. Materials 2017, 10, 369. [CrossRef][PubMed] 3. Duan, B. Microencapsulation via In Situ Polymerization. In Handbook of Encapsulation and Controlled Release; Karsa, D.R., Stephenson, R.A., Eds.; Woodhead Publishing Limited, Taylor & Francis Group, LLC: Cambridge, UK, 2016; pp. 307–314. 4. Poland, I.; Datta, S.S.; Weitz, D.A. Controlling the morphology of polyurea microcapsules using microfluidics. Langmuir 2014, 44, 13405–13410. [CrossRef] 5. Yang, J.; Keller, M.W.; Moore, J.S.; White, S.R.; Sottos, N.R. Microencapsulation of Isocyanates for Self-Healing Polymers. Macromolecules 2008, 41, 9650–9655. [CrossRef] 6. Nguyen, L.T.T.; Hillewaere, X.K.; Teixeira, R.F.A.; Berg, O.V.D.; Prez, F.E.D. Efficient microencapsulation of a liquid isocyanate with in situ shell functionalization. Polym. Chem. 2014, 6, 1159–1170. [CrossRef] 7. Credico, B.D.; Turri, S. An efficient method for the output of new self-repairing materials through a reactive isocyanate encapsulation. Eur. Polym. J. 2013, 49, 2467–2476. [CrossRef] Polymers 2018, 10, 825 16 of 16

8. Ming, Y.; Hu, J.; Xing, J.; Wu, M.; Qu, J. Preparation of polyurea/melamine formaldehyde double-layered self-healing microcapsules and investigation on core fraction. J. Microencapsul. 2016, 33, 307–314. [CrossRef] [PubMed] 9. Li, J.; Mazumder, M.A.J.; Stover, H.D.H.; Sherly, I.M.; Hitchcock, A.P. Polyurea microcapsules: Surface modiﬁcation and capsule size control. J. Polym. Sci. A Polym. Chem. 2011, 49, 3038–3047. [CrossRef] 10. Sondari, D.; Septevani, A.A.; Randy, A.; Triwulandari, E. Polyurethane microcapsule with glycerol as the polyol component for encapsulated self-healing agent. IJET 2010, 2, 446–471. 11. Kardar, P. Preparation of polyurethane microcapsules with different polyols component for encapsulation of isophorone diisocyanate healing agent. Prog. Org. Coat. 2015, 89, 271–276. [CrossRef] 12. Loureiro, M.V.; Vale, M.; Schrijver, A.D.; Bordado, J.C.; Silva, E.; Marques, A.C. Hybrid custom-tailored sol-gel derived microscaffolds for biocides immobilization. Microporous Mesoporous Mater. 2018, 261, 252–258. [CrossRef] 13. Loureiro, M.V.; Lourenço, M.J.; Schrijver, A.D.; Santos, L.F.; Bordado, J.C.; Marques, A.C. Amino-silica microcapsules as effective curing agents for polyurethane foams. J. Mater. Sci. 2017, 52, 5380–5389. [CrossRef] 14. Loureiro, M.V.; Ciriminna, R.; Lourenço, M.J.; Santos, L.F.; Schrijver, A.D.; Bordado, J.C.; Pagliaro, M.; Marques, A.C. Organically-modiﬁed silica based microspheres for self-curing polyurethane one component foams. Microporous Mesoporous Mater. 2017, 244, 244–250. [CrossRef] 15. Ciriminna, R.; Marques, A.C.; Bordado, J.C.; Schrijver, A.; Pagliaro, M. Green-Caps: Towards solid curing agents for sustainable polyurethane foams. Sustain. Chem. Process. 2014, 2, 24. [CrossRef] 16. Almeida, R.M.; Marques, A.C. Characterization of Sol–Gel Materials by Infrared Spectroscopy. In Handbook of Sol-Gel Science and Technology; Klein, L., Aparicio, M., Jitianu, A., Eds.; Springer: Cham, Switzerland, 2016. [CrossRef] 17. Hano, N.; Takafuji, M.; Ihara, H. One-pot preparation of polymer microspheres having wrinkled hard surfaces through self-assembly of silica nanoparticles. Chem. Commun. 2017, 53, 9147–9150. [CrossRef] [PubMed] 18. Huang, M.; Yang, J. Facile microencapsulation of HDI for self-healing anticorrosion coatings. J. Mater. Chem. 2011, 21, 11123–11130. [CrossRef] 19. Tatiya, P.D.; Mahulikar, P.P.; Gite, V.V. Designing of polyamidoamine-based polyurea microcapsules containing tung oil for anticorrosive coating applications. JCT Res. 2016, 13, 715–726. [CrossRef] 20. Brochu, A.B.; Chyan, W.J.; Reichert, W.M. Microencapsulation of 2-octylcyanoacrylate tissue adhesive for self-healing acrylic bone cement. J. Biomed. Mater. Res. B Appl. Biomater. 2012, 7, 1764–1772. [CrossRef] [PubMed] 21. Yi, H.; Yang, Y.; Gu, X.; Huang, J.; Huan, C.W. Multilayer composite microcapsule synthesized by Pickering emulsion templates and its application in self-healing coating. J. Mater. Chem. A 2015, 3, 13749–13757. [CrossRef] 22. Koh, E.; Kim, N.K.; Shin, J.; Kim, Y.W. Polyurethane Microcapsules for Self-Healing Paint Coatings. RSC Adv. 2014, 4, 16214–16223. [CrossRef]

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).