Synthesis and Electrospinning of Polycaprolactone from An

polymers

Article Synthesis and Electrospinning of Polycaprolactone from an Aluminium-Based Catalyst: Inﬂuence of the Ancillary Ligand and Initiators on Catalytic Eﬃciency and Fibre Structure

Ioannis K. Kouparitsas , Elisa Mele and Sara Ronca *

Department of Materials, Loughborough University, Leicestershire LE11 3TU, UK; [email protected] (I.K.K.); [email protected] (E.M.) * Correspondence: [email protected]; Tel.: +44-0-1509-564-848

Received: 29 March 2019; Accepted: 11 April 2019; Published: 13 April 2019

Abstract: In the present study, we investigated the catalytic performance of a 2,20-methylenebis(6- tert-butyl-4-methylphenol) (MDBP)–aluminium complex for the ring-opening polymerisation (ROP) of ε-caprolactone in combination with various alcohols as initiators. Three different alcohols were investigated: 1-adamantanemethanol (A), 1H,1H,2H,2H-perfluoro-1-octanol (F) and isopropanol (I). Samplings of polycaprolactone (PCL) at various reaction times showed a linear increase in the polymer molecular weight with time, with very narrow polydispersity, confirming the living nature of the catalytic system. Scanning electron microscope (SEM) images of electrospun PCL fibre mats produced from 30 wt % dichloromethane/dimethyl sulfoxide solutions showed a high level of surface porosity with a reasonable homogeneity of fibre diameters. The values of the liquid absorption and water contact angle were measured for the electrospun mats, with the F-capped PCL consistently showing absorption values up to three times higher than those of PCL samples capped with the other two alcohols, as well as increased hydrophobicity. The nature of the alcohol can influence the surface hydrophobicity and absorption ability of electrospun fibres, demonstrating the possibility of tailoring material properties through controlled polymerisation.

Keywords: ring-opening polymerisation; functionalised polycaprolactone; electrospinning; porous fibres

1. Introduction The possibility of fine-tuning the composition and molecular architecture of polymers through synthesis and their precise assembly and organisation through various types of processing is the main reason for their incredibly successful use in such a wide range of applications [1,2]. The field of biomedical sciences has benefitted considerably from this opportunity, as demonstrated by the fast development of novel polymer-based solutions in tissue engineering and drug delivery, with the ability to replicate and/or enhance biological structures and functions [3]. In this respect, the development of biodegradable polymers with fine-tuned composition and their controlled assembly is of substantial interest, whether the material needs to offer either mechanical support in a timeframe that is compatible with the restoration of injured tissue [4], or the controlled release of compounds [5]. Polycaprolactone (PCL) has been the preferred choice for many of these applications thanks to its biocompatibility, favourable mechanical properties and degradation kinetics that can be tailored by varying the molecular weight and/or crystallinity of the polymer [6]. PCL is usually synthesised by the ring-opening polymerisation (ROP) of ε-caprolactone (ε-CL) [7], a process that can be mediated by a metal catalyst, according to the mechanism illustrated in Figure1.

Polymers 2019, 11, 677; doi:10.3390/polym11040677 www.mdpi.com/journal/polymers Polymers 2019, 11, x FOR PEER REVIEW 2 of 11

PolymersPCL2019, 11is ,usually 677 synthesised by the ring-opening polymerisation (ROP) of ε-caprolactone (ε-CL)2 of 11 [7], a process that can be mediated by a metal catalyst, according to the mechanism illustrated in Figure 1.

FigureFigure 1. 1.Ring-opening Ring-opening polymerisation polymerisation mechanism of of εε-caprolactone-caprolactone mediated mediated by by a metal a metal catalyst. catalyst.

TinTin octoate octoate is is certainly certainly the the most most commoncommon catalyst for for the the ROP ROP of of εε-CL-CL thanks thanks to to its its commercial commercial availability,availability, easier easier handling handling and and solubility solubility inin mostmost commonly used used organic organic solvents. solvents. An An alcohol alcohol is is requiredrequired to to initiate initiate the the reaction. reaction. TheThe mainmain issue when when using using tin tin octoate octoate is is that that it itnecessitates necessitates high high temperatures,temperatures, which which may may favour favour intermolecularintermolecular an andd intramolecular esterification, esterification, broadening broadening the the polydispersitypolydispersity of of the the resulting resulting polymer. polymer. Moreover,Moreover, for biomedical applications, applications, the the use use of oftin tin may may be be ofof concern concern due due to to the the cytotoxicity cytotoxicity of of thisthis metalmetal inin some forms [8,9]. [8,9]. AA possible possible alternative alternative catalystcatalyst was was described described by by Hamitou Hamitou et al. et [10]. al. [ 10The]. authors The authors elucidated elucidated the thereaction reaction mechanism mechanism of of aluminium-based aluminium-based catalysts catalysts activated activated by byalcohols alcohols in inthe the polymerisation polymerisation of of lactones, showing that these species can produce almost mono-dispersed polymers at room lactones, showing that these species can produce almost mono-dispersed polymers at room temperature temperature and with high conversion. Interestingly, the reaction mechanism implies that the and with high conversion. Interestingly, the reaction mechanism implies that the resulting polyester resulting polyester will always have one hydroxyl end group resulting from the hydrolysis of the will always have one hydroxyl end group resulting from the hydrolysis of the catalytic mixture, while catalytic mixture, while the nature of the other end group will be fixed by the initial structure of the the nature of the other end group will be fixed by the initial structure of the catalyst alkoxide groups, catalyst alkoxide groups, i.e. by the choice of the alcohol/initiator. For these reasons, and due to the i.e.,inherent by the low choice toxicity of the of alcoholaluminium/initiator. oxide Forcompound these reasons,s [11], we and decided due toto thefocus inherent our attention low toxicity on this of aluminiumclass of catalysts. oxide compounds We have reported [11], we previously decided to [12] focus on ourthe attentionpossibility on to this synthesise class of PCL catalysts. with photo- We have reportedcurable previously vinyl groups [12] onusing the an possibility aluminium to synthesise catalyst with PCL 2,6-di- with photo-curabletert-butyl-4-methylphenol vinyl groups (BHT), using an aluminiumsimilar to catalystthat reported with 2,6-di-by Akatsukatert-butyl-4-methylphenol et al. in 1995 [13]. (BHT), similar to that reported by Akatsuka et al. inFor 1995 biomedical [13]. applications, electrospinning is widely used to process PCL in the form of micro- or Fornano-fibres, biomedical showing applications, potential electrospinning as scaffolds isfo widelyr tissue used engineering. to process The PCL high in the surface form of area micro- of or nano-fibres,electrospun showing fibres is potential advantageous as sca ffinolds increasing for tissue load engineering. capacity [14], The even high more surface if the area fibres of electrospunexhibit a fibresporous is advantageous surface instead in of increasing a smooth loadone. capacity [14], even more if the fibres exhibit a porous surface insteadThe of a possibility smooth one. to electrospin PCL into porous fibres holds considerable interest for such applications,The possibility as surface to electrospinporosity or internal PCL into channels porous can fibres promote holds cell considerableadhesion and the interest transport for suchof applications,nutrients in as tissue-engineering surface porosity applications or internal channels [15]. Most can electrospinning promote cell adhesionapproaches and to theachieve transport fibre of nutrientsporosity in involve tissue-engineering the use of systems applications composed [15 of]. po Mostlymers electrospinning or solvent blends approaches [16], sometimes to achieve in the fibre porositypresence involve of specific the useenvironmental of systems conditions composed [17], of polymers such as increased or solvent humidity blends [18], [16], or sometimes with the help in the presenceof additional of specific post-processing environmental [19]. conditions [17], such as increased humidity [18], or with the help of additional post-processing [19]. Recently, Katsogiannis et al. reported on the possibility to produce porous electrospun PCL fibres by the careful selection of electrospinning parameters and solvent composition [20]. The authors showed that electrospinning 15 w/v % solutions of PCL in mixtures of chloroform and dimethyl Polymers 2019, 11, 677 3 of 11 sulfoxide could directly originate porous, bead-free fibres, but at the expense of fibre homogeneity and dimensions. Huang et al. reported on the possibility of generating either surface or internal porosity when electrospinning systems based on polylactic acid (PLA) in concentrations of 15 w/v % in various solvent combinations that included ethanol, dimethyl sulfoxide and chloroform [17]. The authors showed that porosity played a more important role in oil sorption than fibre diameter. In the present paper, we report a combined study on the synthesis and processing of polycaprolactone to elucidate: (i) the influence of the ancillary ligand structure and nature of the initiator on the catalytic performance of the chosen aluminium catalyst, and (ii) the effect that the functional groups introduced via the initiators have on the structure, hydrophobicity and liquid-loading ability of resulting electrospun fibres.

2. Materials and Methods All reagents, except for anhydrous toluene (purchased from Sigma-Aldrich, Gillingham, UK), were supplied by Fisher Scientific (Loughborough, UK), Acros Organics (Loughborough, UK) and used as received unless otherwise stated. All glassware used for the preparation of catalysts and for polymer synthesis was dried overnight in an oven at 125 ◦C and purged with 3 vacuum-nitrogen cycles before use. Synthesis: The catalyst preparation and polymerisation reactions were performed under a controlled atmosphere (either nitrogen or argon). Catalyst preparation was conducted in a glove box under argon atmosphere, by reacting equimolar amounts of tri-isobutyl aluminium (TibAl, solution in toluene 1.1 M) and 2,20-methylenebis(6-tert-butyl-4-methylphenol) (MDBP) dissolved in anhydrous toluene under stirring for 10 min at room temperature in a Schlenk tube. Subsequently the desired alcohol (1-adamantanemethanol, 1H,1H,2H,2H-perfluoro-1-octanol or isopropanol, indicated respectively as the A, F and I initiator) was added in an equimolar amount to the Al-MDBP solution, under an additional 10 min of stirring. After that time, the solution was injected into a 0.5 L jacketed Pyrex reactor containing a solution of ε-caprolactone (typically 20 g, dried over molecular sieves, 3 Å) in anhydrous toluene (250 mL) at 40 ◦C under a nitrogen atmosphere and the reaction was kept under magnetic stirring for up to 5 h. The reaction was quenched by slowly pouring the contents of the reactor into a beaker containing 500 mL of cold methanol. The resulting polymers were filtered and dried overnight in a vacuum oven at 30 ◦C. Samplings (~10 mL solution per sampling) for the kinetic study were taken using a syringe with needles that had been oven-dried and purged with nitrogen beforehand. Samplings were immediately quenched in an excess of methanol and the precipitated polymers were dried and analysed as described below. Electrospinning: Dichloromethane (DCM) and dimethyl sulfoxide (DMSO) were used as solvents in a ratio of 6:1 v/v to prepare the electrospinning PCL solutions at 10, 20 and 30 wt %. The electrospinning parameters used were: flow rate of 4.5 mL/h, voltage of 8–10 kV and tip-collector distance of 15 cm. All mats obtained were coherent and mechanically stable enough to be easily detached from the aluminium collector and cut in desired shapes/sizes for further testing (contact angle, liquid absorption). Characterisation: Differential scanning calorimetry (DSC), gel permeation chromatography (GPC), scanning electron microscopy (SEM), liquid absorption and water contact angle (WCA) studies were employed to characterise the materials used. DSC: The thermal properties of the polymer powders were analysed using a TA Instruments Q2000 DSC (TA instruments, Elstree, UK). The samples were weighed and placed in Tzero aluminium pans with Tzero lids. Between 4 and 6 mg of polymer were weighted on a high-precision balance ( 0.001 mg) and analysed from 90 C to 130 C with a heating/cooling/heating ramp at a rate of ± − ◦ ◦ 10 ◦C/min. From the resulting data the peak produced at the first melting temperature (Tm1) and enthalpy of the first heating cycle (∆Hm1 in J/g) as well as the peak crystallisation temperature (Tc) of the cooling cycle were recorded. Using the enthalpy of fusion for the samples along with the enthalpy Polymers 2019, 11, 677 4 of 11 of fusion for 100% crystalline PCL, 135.6 J/g [21], the % crystallinity of each sample was calculated using Equation (1):

Xc % = (sample enthalpy of fusion/100% crystalline PCL enthalpy of fusion) 100 (1) × GPC: Measurements were conducted on an Agilent 1260 Inﬁnity (Agilent, Cheshire, UK) equipped with a refraction index (RI) detector and calibrated with GPC PS standard provided by Agilent.

Solutions of 1Polymers mg/ mL2019, 11 of, xpolymer FOR PEER REVIEW in chloroform with added 3% v/v TEA (triethylamine)4 of were11 prepared. Analysis of the results was conducted using Mark-Houwink constants for PCL in chloroform at 30 ◦C, K = 12.98 andtheα cooling= 0.828 cycle [22 were]. recorded. Using the enthalpy of fusion for the samples along with the enthalpy of fusion for 100% crystalline PCL, 135.6 J/g [21], the % crystallinity of each sample was calculated SEM: Inusing order Equation to assess (1): the porosity of the ﬁbres, the samples were observed using a Hitachi

TM3030 benchtop SEMXc % = (Hitachi, (sample enthalpy London, of fusion UK). / 100% Pieces crystalline of the PCL fibre enthalpy mats of werefusion) coated × 100 using (1) a Quorum Q150R S with AuGPC:/Pd Measurements (90 s, 80:20) were before conducted imaging. on an Agilent 1260 Infinity (Agilent, Cheshire, UK) Liquid absorption:equipped with a Forrefraction the absorptionindex (RI) detector tests, and olive calibrated oil, with manuka GPC PS essential standard provided oil, black by pepper oil, dodecane andAgilent. silicone Solutions oil of were 1 mg/mL employed. of polymer in chloroform A portion with of added electrospun 3% v/v TEA (triethylamine) mat was cut were with scissors, prepared. Analysis of the results was conducted using Mark-Houwink constants for PCL in weighed onchloroform a precision at 30°C, balance, K = 12.98 immersed and α = 0.828 in [22]. the liquids and then weighed again after immersion. The absorption abilitySEM: In wasorder reportedto assess the as porosity the weight of the fibres, of absorbed the samples liquid were /observedinitial weightusing a Hitachi of the mat. WCA: ContactTM3030 benchtop angle measurementsSEM (Hitachi, London, were UK). conducted Pieces of the usingfibre mats a Dataphysicswere coated using OCA a Quorum 20 contact angle Q150R S with Au/Pd (90 s, 80:20) before imaging. µ machine (Dataphysics,Liquid absorption: Filderstadt, For the Germany).absorption tests, A olive drop oil, of manuka 2 L essential of distilled oil, black water pepper was oil, placed onto flat pieces ofdodecane mats using and silicone the sessile oil were drop employed. method. A port Theion resultingof electrospun contact mat was angle cut with was scissors, measured via the accompaniedweighed camera on/ software.a precision balance, An average immersed was in the calculated liquids and fromthen weighed five measurements again after immersion. for each sample. The absorption ability was reported as the weight of absorbed liquid/initial weight of the mat. WCA: Contact angle measurements were conducted using a Dataphysics OCA 20 contact angle 3. Results andmachine Discussion (Dataphysics, Filderstadt, Germany). A drop of 2 μL of distilled water was placed onto flat We havepieces previously of mats using reported the sessile on the drop catalytic method. performanceThe resulting contact of an angle Al-(BHT) was measuredcatalyst via the in the presence accompanied camera/software. An average was calculated from five measurements for2 each sample. of suitable alcohol initiators [12]. In the present3. Results study, and discussion we wanted to assess the steric/electronic influence of the initiator on the catalyst performanceWe have and previously resulting reported polymer on the properties,catalytic performance especially of an inAl-(BHT) terms2 catalyst of hydrophobicity in the and liquid absorptionpresence ability of suitable when alcohol processed initiators [12]. in the form of electrospun fibres. For this purpose, three In the present study, we wanted to assess the steric/electronic influence of the initiator on the different initiatorscatalyst wereperformance selected: and resulting 1-adamantanemethanol polymer properties, especially (A), 1H,1H,2H,2H-perfluoro-1-octanolin terms of hydrophobicity and (F) and isopropanolliquid (absorptionI), shown ability in Figurewhen processed2. Isopropanol in the form isof aelectrospun common fibres. alcohol For this used purpose, as an three initiator with Al catalytic systemsdifferent initiators [7]; perfluoro-octanol were selected: 1-adamantanemethanol was chosen ( toA), potentially1H,1H,2H,2H-perf enhanceluoro-1-octanol the hydrophobicity (F) and isopropanol (I), shown in Figure 2. Isopropanol is a common alcohol used as an initiator with Al of the PCL chaincatalytic [ 23systems,24], [7]; while perfluoro-oc adamantane-terminatedtanol was chosen to potentially PCL enhance has been the hydrophobicity recently shown of the to produce non-covalentlyPCL bound chain [23,24], block-copolymers while adamantane-terminated with cyclodextrin-polyvinylpyrrolidone PCL has been recently shown to produce (CD-PVP) non- [25]. covalently bound block-copolymers with cyclodextrin-polyvinylpyrrolidone (CD-PVP) [25].

Figure 2. AlcoholFigure initiators2. Alcohol utilised initiators for theutilised activation for the of activation tri-isobutyl of aluminium:tri-isobutyl aluminium: 1-adamantanemethanol 1- (i), 1H,1H,2H,2H-perﬂuoro-1-octanoladamantanemethanol (i), 1H,1H,2H,2H-perflu (ii) and isopropanoloro-1-octanol (ii) (iii and). isopropanol (iii).

PreliminaryPreliminary attempts attempts at using at using the Al-(BHT)the Al-(BHT)2 catalyst in combination in combination with F were with notF successful,were not successful, possibly due to the high steric hindrance of the2 two BHT groups and the alcohol itself; moreover, possibly duealuminium to the high derivatives steric of hindrance monodentate ofbulky the phen twools BHT have been groups reported and tothe show alcohol a decrease itself; in moreover, aluminium derivativescatalytic activity of in the monodentate presence of excess bulky alcoho phenolsl (isopropanol have specifically), been reported due to the metathesis to show a decrease in catalytic activity in the presence of excess alcohol (isopropanol specifically), due to the metathesis occurring between the aluminium phenolate and isopropanol [26]. A more stable Polymers 2019, 11, 677 5 of 11 version of an Al-phenolate catalyst was reported by Ko et al. [27]. The authors tested the Polymers 2019, 11, x FOR PEER REVIEW 5 of 11 polymerisation performance of an organo-aluminium in combination with a bidentate ligand such as 2,2-ethylidenebis(4,6-di-occurring between thetert aluminium-butylphenol), phenolate demonstrating and isopropanol that [26]. this A catalytic more stable system version was of able an Al- to keep a livingphenolate character catalyst (i.e., was one reported catalytic by Ko site et producing al. [27]. The oneauthors polymeric tested the chain) polymerisation in the polymerisation performance of caprolactoneof an organo-aluminium for up to 32 h. in combination with a bidentate ligand such as 2,2-ethylidenebis(4,6-di-tert- Webutylphenol), decided to demonstrating test the catalytic that this effi ciencycatalytic of syst a systemem was similarable to keep to the a living one described character by(i.e. Ko,one with the usecatalytic of 2,2 site0-methylenebis(6- producing one polymerictert-butyl-4-methylphenol) chain) in the polymerisation (MDBP) of caprolactone as the ancillary for up to ligand 32 hours. for the aluminium.We Thedecided other to possibletest the catalytic advantage efficiency of using of a system a bidentate similar ligand to the insteadone described of two by monodentate Ko, with ones isthe the use increased of 2,2’-methylenebis(6- accessibility oftert the-butyl-4-methylphenol) metal’s frontal space (MDBP) (a well-established as the ancillary concept ligand in for the the design aluminium. The other possible advantage of using a bidentate ligand instead of two monodentate of metallocene catalysts for olefin polymerisation [28]), thus allowing an easier reaction with sterically ones is the increased accessibility of the metal’s frontal space (a well-established concept in the design hinderedof metallocene alcohols such catalysts as F andfor Aolefin. polymerisation [28]), thus allowing an easier reaction with Insterically Table1 ,hindered the results alcohols of a testsuch reaction as F and performedA. with the catalytic system Al-MDBP activated with isopropanolIn Table (Al-MDBP-1, the resultsI )of are a test reported. reaction Samplings performed ofwith the the catalytic catalytic solution system (~10Al-MDBP mL) wereactivated taken at variouswith times isopropanol up to 210 (Al-MDBP- min, quenchedI) are reported. in coldmethanol, Samplings andof the the catalytic resulting solution polymers (~10 mL) were were characterised taken via GPCat various to follow times the up progressive to 210 minutes, increase quenched in molecular in cold methanol, weight and (number the resulting average, polymersMn, and were weight average,characterisedMw) and via confirm GPC to the follow living the character progressive of theincrease system. in molecular weight (number average, Mn, and weight average, Mw) and confirm the living character of the system.

Table 1. Mn, Mw and polydispersity indices (PDI) of samplings taken at various times for the test Table 1. Mn, Mw and polydispersity indices (PDI) of samplings taken at various times for the test polymerisation of ε-CL with the catalytic system Al-2,20-methylenebis(6-tert-butyl-4-methylphenol)- polymerisation of ε-CL with the catalytic system Al-2,2’-methylenebis(6-tert-butyl-4-methylphenol)- isopropanol (Al-MDBP-I) in toluene a. isopropanol (Al-MDBP-I) in toluene a. ε Time Mn 1 Mw 1 PDI ε-CL -CL TheoreticalTheoretical Mw Time (min) Mn (g mol− ) Mw (g mol− ) PDI 1 (min) (g mol−1) (g mol−1) Conversion(%)Conversion (%) (g/mol)Mw (g mol− ) 3030 21062106 2481 2481 1.178 1.178 4 4 2200 2200 60 4051 4662 1.151 8 4300 60 4051 4662 1.151 8 4300 90 5646 6411 1.135 11 6000 12090 71205646 6411 8101 1.135 1.138 11 15 6000 8100 150120 84597120 8101 9467 1.138 1.119 15 18 8100 9700 180 9547 10,765 1.128 22 11,900 150 8459 9467 1.119 18 9700 210 10,688 11,782 1.102 25 13,500 a 180 9547 10765 1.128 22 11900 Reaction conditions: temperature = 40 ◦C; solvent = 250 mL toluene; Al-MDBP: 0.37 mmol; I: 0.37 mmol; ε-CL: 0.176 mol;210 ratio CL:Al:I = 10688480:1:1. 11782 1.102 25 13500 a Reaction conditions: temperature = 40 °C; solvent = 250 mL toluene; Al-MDBP: 0.37 mmol; I: 0.37 As evidentmmol; ε from-CL: 0.176 Figure mol;3 ,ratio the CL:Al:I molecular = 480:1:1. weight of PCL increased in a linear trend with time, with very narrowAs polydispersityevident from Figure indexes 3, the (PDI) molecular extending weight between of PCL 1.10increased and 1.18, in a thuslinear conﬁrming trend with thetime, living naturewith of thevery catalytic narrow polydispersity system. indexes (PDI) extending between 1.10 and 1.18, thus confirming the living nature of the catalytic system.

Figure 3. Values of Mn, Mw and PDI as a function of time in the polymerisation of ε-CL with the Al-MDBP-I system.

Polymers 2019, 11, 677 6 of 11

The graph in Figure4 shows the variation of ln([M] 0/[M]) with time, where [M]0 is the monomer concentration at the beginning of the reaction and [M] is the concentration at any given time t. Polymers 2019, 11, x FOR PEER REVIEW 6 of 11 The polymerisation rate showed a linear dependence of ln([M]0/[M]) versus time, according to Equation (2) [29]: Figure 3. Values of Mn, Mw and PDI as a function of time in the polymerisation of ε-CL with the Al- ln([M] /[M]) = kapp t (2) MDBP-I system. 0 ×

FigureFigure 4. Kinetics 4. Kinetics of theof theε-CL ε-CL polymerisation polymerisation in toluene in toluene at 40 at ◦40C, °C, initiated initiated by Al-MDBP-by Al-MDBP-I. I. The linearity confirmed that the reaction was first order with respect to the monomer, with a The graph in Figure 4 shows the3 variation1 of ln([M]0/[M]) with time, where [M]0 is the monomer kinetic rate constant kapp = 1.25 10 min . This value is close to those observed in similar systems concentration at the beginning× of− the reaction− and [M] is the concentration at any given time t. The by Dubois et al. [29]. polymerisation rate showed a linear dependence of ln([M]0/[M]) versus time, according to Equation In Table2, the reaction conditions used for the synthesis of three PCL samples in the presence (2) [29]: of the chosen initiators A, F and I and the resulting values of molecular weights and distributions as obtained from GPC measurements are reported.ln([M]0/[M]) = kapp × t (2) The linearity confirmed that the reaction was first order with respect to the monomer, with a Table 2. Reaction conditions and resulting molecular weights/distributions listed by initiator type a. kinetic rate constant kapp = 1.25 × 10−3 min−1. This value is close to those observed in similar systems by Dubois etAl-MDBP-In al. [29]. b CL Conv. Yield M M Al* c Sample CL:Al n w PDI In Table(mmol) 2, the reaction(mol) conditions used(%) for the synthesis(g) (kg of/mol) three(kg PCL/mol) samples(mmol) in the presence PCL_Aof the chosen initiators 0.37 A, F 0.355 and I and 960:1 the resulting 23 values 9 of molecular 17 weights 20 and ~0.4 distributions 1.14 as PCL_Fobtained from 0.74GPC measurements 0.176 are 240:1 reported. 40 8 18 21 ~0.4 1.12 PCL_I 0.74 0.176 240:1 ~100 20 22 27 ~0.7 1.19 a b AllTable reactions 2. Reaction performed conditions in 250 mL ofand toluene, resulting at 40 molecular◦C, for 5 h; weInights/distributions= Initiator: PCL_A: 1-adamantanemethanol,listed by initiator type a. PCL_F: perfluoro-octanol; PCL_I: isopropanol; c Al*: calculated active sites. Al-MDBP- CL Conv. Yield Mn Mw Al* c Sample In b CL:Al PDI Up to 20 g of PCL were(mol) synthesised in 5 h,(%) with a 0.5(g) L reactor;(kg/mol) we foresee (kg/mol) that the reaction(mmol) could be scaled up to(mmol) larger amounts by adjusting the reactor volume and corresponding reagents. PCL_A 0.37 0.355 960:1 23 9 17 20 ~0.4 1.14 It is interesting to note that for PCL_A and PCL_I, the moles of the active sites calculated from PCL_F 0.74 0.176 240:1 40 8 18 21 ~0.4 1.12 M M the ratioPCL_I Y / w, where0.74 Y is 0.176 polymer 240:1 yield (in~100 g) obtained20 after quenching, 22 and 27 w is the ~0.7 average 1.19 molecular weight (in g/mol), agreed rather closely with the moles of catalyst introduced into the system. a All reactions performed in 250 mL of toluene, at 40 °C, for 5h; b In = Initiator: PCL_A: 1- This confirms that all aluminium sites were correctly activated by the initiators and only one chain was adamantanemethanol, PCL_F: perfluoro-octanol; PCL_I: isopropanol; c Al*: calculated active sites. produced for each active site. In the case of PCL_F, the active sites were about half of the aluminium sites introduced,Up to 20 g however of PCL were we attributed synthesised this in result 5 h, with to the a 0.5 lower L reactor; solubility we foresee of F in toluene.that the reaction could beAnother scaled up interesting to larger amounts observation by adjusting is that the the molecular reactor volume weight ofand PCL_I corresponding was slightly reagents. higher than that of PCL_FIt is interesting and PCL_A, to note suggesting that for PCL_A either a and faster PCL_I, initiation the moles at the of less the sterically-hindered active sites calculated active from the ratio Y/Mw, where Y is polymer yield (in g) obtained after quenching, and Mw is the average molecular weight (in g/mol), agreed rather closely with the moles of catalyst introduced into the system. This confirms that all aluminium sites were correctly activated by the initiators and only one chain was produced for each active site. In the case of PCL_F, the active sites were about half of the aluminium sites introduced, however we attributed this result to the lower solubility of F in toluene.

Polymers 2019, 11, 677 7 of 11

Polymers 2019, 11, x FOR PEER REVIEW 7 of 11 sites, faster kinetics, or a combination of both. Studies of the kinetics of polymerisation as a function of the initiatorAnother are interesting underway observation to confirm theis that most the likely molecular reason, weight and toof PCL_I show howwas slightly it is possible higher to than target higherthat of or PCL_F lower molecularand PCL_A, weights suggesting by a either careful a choicefaster initiation of CL/initiator at the andless reactionsterically-hindered times. active sites,In Tablefaster3 ,kinetics, the melting or a /combinationcrystallisation of temperaturesboth. Studies andof the nascent kinetics crystallinities of polymerisation for the as three a function analysed samplesof the areinitiator reported. are underway All samples to showconfirm melting the most in the likely typical reason, range and of ~60to show◦C, with how highit is crystallinities,possible to reachingtarget higher over 77% or lower for PCL_F. molecular These weights values by agree a careful with choice previously of CL/initiator observed and values reaction for PCL times. samples of similar molecular weight [30]. Table 3. Differential scanning calorimetry (DSC) results showing melting/crystallisation Tabletemperatures 3. Differential and melting scanning enthalpies. calorimetry Nascent (DSC) crysta resultsllinity showing was calculated melting/ crystallisationfrom the melting temperatures enthalpy ° andusing melting ∆H m = enthalpies. 135.6 J/g [21]. Nascent crystallinity was calculated from the melting enthalpy using ∆H◦ m = 135.6 J/g [21]. Sample Tm1 (°C) Tc (°C) ΔHm1 (J/g) Xc % PCL_A 57.2 36.4 92 67.8 Sample Tm1 (◦C) Tc (◦C) ∆Hm1 (J/g) Xc % PCL_F 62.7 35.7 105.2 77.6 PCL_APCL_I 57.261.9 34.1 36.4 102.7 9275.7 67.8 In Table 3, the melting/crystallisationPCL_F 62.7 temperatures 35.7 and 105.2 nascent crystallinities 77.6 for the three PCL_I 61.9 34.1 102.7 75.7 analysed samples are reported. All samples show melting in the typical range of ~60 °C, with high crystallinities, reaching over 77% for PCL_F. These values agree with previously observed values for PCLThe samples three PCL of similar samples molecular were electrospun weight [30]. from solutions of DCM/DMSO (6:1 v/v) at concentrations rangingThe from three 10 wtPCL % tosamples 30 wt %, were and theelectrospu SEM imagesn from of solutions the resulting of fibresDCM/DMSO are shown (6:1 inv Figure/v) at 5. PCL_Aconcentrations and PCL_F ranging only from demonstrated 10 wt % to 30 bead wt %, formation and the SEM at images the lowest of the concentration,resulting fibres are while shown some proto-fibresin Figure 5. were PCL_A identified and PCL_F at 20only wt demonstrated %. Well-formed bead fibresformation could at the only lowest be obtained concentration, at the while highest concentrationsome proto-fibres of 30 wtwere %. identified For PCL_I, at fibre20 wt formation %. Well-formed started fibres at the could intermediate only be obtained concentration, at the highest possibly dueconcentration to the slightly of higher30 wt molecular%. For PCL_I, weight fibre of format the polymer.ion started at the intermediate concentration, possibly due to the slightly higher molecular weight of the polymer.

Figure 5. i vi Figure 5.Scanning Scanning electron electron microscope microscope (SEM)(SEM) micrographs of of electrospun electrospun samples samples of ofPCL_A PCL_A (i–vi), ( – ), PCL_FPCL_F (vii (vii–xii)–xii) and and PCL_IPCL_I ((xiii–xviii)xiii–xviii) at 10, 10, 20 20 and and 30 30 wt wt % % solutions solutions in inDCM/DMSO DCM/DMSO (6:1 (6:1 v/vv)./ v). Magniﬁcations at 500 (100 µm bar) and 5000 (10 µm bar). Magnifications at 500×× (100μm bar) and 5000×× (10μm bar).

AA close-up close-up of of the the ﬁbres fibres obtained obtained atat 3030 wtwt % is shown in in Figure Figure 6.6. It It is is interesting interesting to to note note that that the the ﬁbresfibres showed showed a higha high level level of surfaceof surface porosity, porosity, while while maintaining maintaining a reasonable a reasonable size size homogeneity. homogeneity. Fibre diametersFibre diameters were in were the range in the ofrange tenths of tenths of microns, of microns, with poreswith pores in the in range the range of microns. of microns.

Polymers 2019, 11, 677 8 of 11 Polymers 2019, 11, x FOR PEER REVIEW 8 of 11 Polymers 2019, 11, x FOR PEER REVIEW 8 of 11

FigureFigure 6. 6.SEM SEM micrographs micrographs at at 5000 5000×magnification magnification of PCL_A (i), (i), PCL_F PCL_F (ii) (ii )and and PCL_I PCL_I (iii) (iii electrospun) electrospun × fromfrom 30 30 wt wt % % solutions solutions in in DCM DCM/DMSO/DMSO (6:1 (6:1v v/v/v).). Figure 6. SEM micrographs at 5000× magnification of PCL_A (i), PCL_F (ii) and PCL_I (iii) electrospun FurtherFurtherfrom 30 to towt the the% discussion solutions discussion in in DCM/DMSOin the the introduction, introduction, (6:1 v/v the). the presence presence of of porosity porosity in in electrospun electrospun fibresfibres cancan be associatedbe associated with thewith fast the evaporation fast evaporation of the of solvent the so duringlvent during electrospinning electrospinning [31]. In[31]. our In case, our thecase, process the Further to the discussion in the introduction, the presence of porosity in electrospun fibres can mayprocess have beenmay morehave been pronounced more pronounced due to the usedue ofto morethe use concentrated of more concentrated solutions, thussolutions, reducing thus the be associated with the fast evaporation of the solvent during electrospinning [31]. In our case, the amountreducing of solvent the amount and consequentlyof solvent and theconseque time neededntly the fortime it needed to evaporate. for it to evaporate. process may have been more pronounced due to the use of more concentrated solutions, thus TheThe high high surface surface area area resultingresulting fromfrom thethe presencepresence of pores, associated associated with with the the inherent inherent reducing the amount of solvent and consequently the time needed for it to evaporate. hydrophobicityhydrophobicity of of PCL, PCL, make make these these electrospun electrospun matsmats ideal for for the the abso absorptionrption of of oils, oils, a acharacteristic characteristic The high surface area resulting from the presence of pores, associated with the inherent thatthat has has been been widely widely explored explored in in the the case case ofof biomedicalbiomedical applications applications when when antimicrobial antimicrobial essential essential oils oils hydrophobicity of PCL, make these electrospun mats ideal for the absorption of oils, a characteristic areare used used [32 [32,33].,33]. that Wehas beentherefore widely tested explored the ability in the of case the electrospunof biomedic samplesal applications to absorb when a variety antimicrobial of hydrophobic essential oils We therefore tested the ability of the electrospun samples to absorb a variety of hydrophobic liquidsare used and [32,33]. calculated their absorption ability Q according to Equation (3): liquids and calculated their absorption ability Q according to Equation (3): We therefore tested the ability of the electrospun samples to absorb a variety of hydrophobic Q = (W – W0) / W0 (3) liquids and calculated their absorption ability Q according to Equation (3): Q = (W W0)/W0 (3) where W0 is the initial mat weight and W the mat− weight after absorption. Figure 7 shows the absorption values forQ =the (W fib –re W mats0) / W produced0 from the three PCL samples, (3) where W0 is the initial mat weight and W the mat weight after absorption. with respect to five different hydrophobic liquids, namely olive oil, manuka essential oil, black whereFigure W70 isshows the initial the absorption mat weight values and W for the the mat fibre weight mats after produced absorption. from the three PCL samples, pepper oil, dodecane and silicone oil. The values of Q range between 3 and 15 g/g, with PCL_F with respectFigure to 7 five shows different the absorption hydrophobic values liquids, for namelythe fibre olive mats oil, produced manuka from essential the three oil, black PCL pepper samples, consistently showing absorption values that were double or even triple that of PCL_A and PCL_I. oil,with dodecane respect and to siliconefive different oil. The hydrophobic values of Q rangeliquids, between namely 3 andolive 15 oil, g/g, manuka with PCL_F essential consistently oil, black showingpepper absorption oil, dodecane values and that silicone were double oil. The or values even triple of Q that range of PCL_Abetween and 3 PCL_I.and 15 g/g, with PCL_F consistently showing absorption values that were double or even triple that of PCL_A and PCL_I.

Figure 7. Absorption of hydrophobic liquids, Q (g/g), for PCL_A, PCL_F and PCL_I.

Reshmi et al. have reported similar values of absorption for PCL fibres, both pure or with added beeswax to improve the efficiency of oil separation [34]. The authors mention the role of beeswax in FigureFigure 7. Absorption 7. Absorption of hydrophobicof hydrophobic liquids, liquids,Q (g Q/ g),(g/g), for for PCL_A, PCL_A, PCL_F PCL_F and and PCL_I. PCL_I.

Reshmi et al. have reported similar values of absorption for PCL fibres, both pure or with added beeswax to improve the efficiency of oil separation [34]. The authors mention the role of beeswax in

Polymers 2019, 11, 677 9 of 11

Reshmi et al. have reported similar values of absorption for PCL fibres, both pure or with added beeswaxPolymers to2019 improve, 11, x FOR thePEER e REVIEWfficiency of oil separation [34]. The authors mention the role of beeswax9 of 11 in increasing surface roughness and hydrophobicity, resulting in higher absorption capacities of the PCLincreasing/beeswax surface electrospun roughness blends and when hydrophobicity, compared to re puresulting PCL. in higher absorption capacities of the PCL/beeswaxIn our case, electrospun the advantage blends was when the possibility compared toto enhancepure PCL. the hydrophobicity of the material with the directIn our introduction case, the advantage of a hydrophobic was the possibility group into to the enhance polymeric the hydrophobicity chain itself, thus of the removing material the with need forthe additional direct introduction fillers or components. of a hydrophobic group into the polymeric chain itself, thus removing the needThe for enhanced additional hydrophobic fillers or components. character of PCL_F mats was confirmed by water contact angle measurements,The enhanced as shown hydrophobic in Table4 characterand Figure of8 .PCL_F mats was confirmed by water contact angle measurements, as shown in Table 4 and Figure 8. Table 4. Water contact angle for PCL mats electrospun from 30 wt % solutions. Table 4. Water contact angle for PCL mats electrospun from 30 wt % solutions. Sample Water Contact Angle ( ) Sample Water Contact Angle (°) ◦ PCL_APCL_A 131.9 131.9 ± 1.0 1.0 ± PCL_FPCL_F 137.6 137.6 ± 1.3 1.3 ± PCL_IPCL_I 129.6 129.6 ± 3.7 3.7 ±

FigureFigure 8. 8.Images Images of of water water contact contact angle angle (CA) (CA) measurements measurements for for each each sample. sample. Average Average values values from from five measurementsfive measurements are presented are presented in Table in 4Table. 4. 4. Conclusions 4. Conclusions InIn this this paper, paper, we we reported reported on on the the synthesissynthesis of polycaprolactone (PCL) (PCL) with with an an aluminium-based aluminium-based catalystcatalyst in in combination combination withwith various various alcohols alcohols asas initiators, initiators, followed followed by the by electrospinning the electrospinning of the of theresulting resulting polymers. polymers. The catalytic The catalytic system system showed showed a living character a living under character the experimental under the experimental conditions conditionsused (toluene used (toluenesolution, solution,40 °C, 5 h 40 reaction),◦C, 5 h reaction), allowing allowingfor the controlled for the controlled synthesis synthesisof almost ofmono- almost mono-modalmodal PCL PCL with with defined defined chain-end chain-end groups groups that that orig originatedinated from from the thechosen chosen initiator. initiator. Larger, Larger, more more sterically-hinderedsterically-hindered initiators, initiators, such such as as 1-adamantanemethanol1-adamantanemethanol and and 1H,1H,2H,2H-perfluoro-1-octanol 1H,1H,2H,2H-perfluoro-1-octanol resultresult in in lower lower molecular molecular weight weight polymers polymers forfor the same reaction reaction time time when when compared compared to tosmaller, smaller, less less stericallysterically hindered hindered initiators initiators such such as as isopropanol, isopropanol, suggesting an an effect effect of of the the initiator initiator on on the the induction induction timetime and and/or/or the the polymerisation polymerisation rate. rate. AllAll PCL PCL samples samples could could be electrospun be electrospun in homogeneous, in homogeneous, micro-sized micro-sized fibres from fibres rather from concentrated rather (30concentrated wt %) DCM (30/DSMO wt %) solutions.DCM/DSMO All solutions. fibres showed All fibres a highshowed level a high of surface level of porosity surface porosity that has that been exploitedhas been to exploited assess their to abilityassess totheir absorb ability hydrophobic to absorb liquids.hydrophobic The electrospunliquids. The mats electrospun produced mats from theproduced perfluoro-octanol-substituted from the perfluoro-octanol-substituted PCL showed the highestPCL showed hydrophobicity the highest from hydrophobicity water contact from angle water contact angle measurements, as well as the highest liquid absorption ability. measurements, as well as the highest liquid absorption ability. The results suggest that the chosen catalytic system could be efficiently used to produce PCL The results suggest that the chosen catalytic system could be efficiently used to produce PCL with with controlled molecular weights and with the desired functionalities successfully introduced controlled molecular weights and with the desired functionalities successfully introduced without the without the need for additional post-processing. need for additional post-processing. Author Contributions: Conceptualization, S.R. and E.M.; Methodology, S.R. and E.M.; Validation, S.R., E.M. and I.K.K.; Formal Analysis, I.K.K.; Investigation, I.K.K.; Resources, S.R. and E.M.; Data Curation, I.K.K.; Writing- Original Draft Preparation, S.R.; Writing-Review & Editing, S.R., E.M. and I.K.K.; Supervision, S.R. and E.M.; Project Administration, S.R.

Funding: This research received no external funding.

Polymers 2019, 11, 677 10 of 11

Author Contributions: Conceptualization, S.R. and E.M.; Methodology, S.R. and E.M.; Validation, S.R., E.M. and I.K.K.; Formal Analysis, I.K.K.; Investigation, I.K.K.; Resources, S.R. and E.M.; Data Curation, I.K.K.; Writing—Original Draft Preparation, S.R.; Writing—Review & Editing, S.R., E.M. and I.K.K.; Supervision, S.R. and E.M.; Project Administration, S.R. Funding: This research received no external funding. Acknowledgments: The authors wish to thank Noreen Thomas for useful discussions. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Sun, H.; Kabb, C.P.; Sims, M.B.; Sumerlin, B.S. Architecture-transformable polymers: Reshaping the future of stimuli-responsive polymers. Prog. Polym. Sci. 2019, 89, 61–75. [CrossRef] 2. Sun, H.; Kabb, C.P.; Dai, Y.; Hill, M.R.; Ghiviriga, I.; Bapat, A.P.; Sumerlin, B.S. Macromolecular metamorphosis via stimulus-induced transformations of polymer architecture. Nat. Chem. 2017, 9, 817–823. [CrossRef] 3. Shi, J.; Votruba, A.R.; Farokhzad, O.C.; Langer, R. Nanotechnology in Drug Delivery and Tissue Engineering: From Discovery to Applications. Nano Lett. 2010, 10, 3223–3230. [CrossRef][PubMed] 4. Martina, M.; Hutmacher, D.W. Biodegradable polymers applied in tissue engineering research: A review. Polym. Int. 2007, 56, 145–157. [CrossRef] 5. Kamaly, N.; Yameen, B.; Wu, J.; Farokhzad, O.C. Degradable Controlled-Release Polymers and Polymeric Nanoparticles: Mechanisms of Controlling Drug Release. Chem. Rev. 2016, 116, 2602–2663. [CrossRef] 6. Woodruﬀ, M.A.; Hutmacher, D.W. The return of a forgotten polymer—Polycaprolactone in the 21st century. Prog. Polym. Sci. 2010, 35, 1217–1256. [CrossRef] 7. Labet, M.; Thielemans, W. Synthesis of polycaprolactone: A review. Chem. Soc. Rev. 2009, 38, 3484–3504. [CrossRef] 8. Ahamed, M.; Akhtar, M.J.; Majeed Khan, M.A.; Alhadlaq, H.A. Oxidative stress mediated cytotoxicity of

tin (IV) oxide (SnO2) nanoparticles in human breast cancer (MCF-7) cells. Colloids Surf. B Biointerfaces 2018, 172, 152–160. [CrossRef] 9. Ganguly (Ghosh), B.B.; Talukdar, G.; Sharma, A. Cytotoxicity of tin on human peripheral lymphocytes in vitro. Mutat. Res. Lett. 1992, 282, 61–67. [CrossRef] 10. Hamitou, A.; Ouhadi, T.; Jerome, R.; Teyssié, P. Soluble bimetallic µ-oxoalkoxides. VII. Characteristics and mechanism of ring-opening polymerization of lactones. J. Polym. Sci. Polym. Chem. Ed. 1977, 15, 865–873. [CrossRef] 11. Takami, Y.; Nakazawa, T.; Makinouchi, K.; Glueck, J.; Nosé, Y. Biocompatibility of alumina ceramic and polyethylene as materials for pivot bearings of a centrifugal blood pump. J. Biomed. Mater. Res. 1997, 36, 381–386. [CrossRef] 12. Ronca, A.; Ronca, S.; Forte, G.; Zeppetelli, S.; Gloria, A.; De Santis, R.; Ambrosio, L. Synthesis and characterization of divinyl-fumarate poly-ε-caprolactone for scaffolds with controlled architectures. J. Tissue Eng. Regen. Med. 2018, 12, e523–e531. [CrossRef] 13. Akatsuka, M.; Aida, T.; Inoue, S. Alcohol/methylaluminum diphenolate systems as novel, versatile initiators for synthesis of narrow molecular weight distribution polyester and polycarbonate. Macromolecules 1995, 28, 1320–1322. [CrossRef] 14. Zhang, W.; Ronca, S.; Mele, E. Electrospun Nanofibres Containing Antimicrobial Plant Extracts. Nanomaterials 2017, 7, 42. [CrossRef] 15. Wu, J.; Wang, N.; Zhao, Y.; Jiang, L. Electrospinning of multilevel structured functional micro-/nanofibers and their applications. J. Mater. Chem. A 2013, 1, 7290. [CrossRef] 16. Zhang, W.; Mele, E. Phase separation events induce the coexistence of distinct nanofeatures in electrospun fibres of poly(ethyl cyanoacrylate) and polycaprolactone. Mater. Today Commun. 2018, 16, 135–141. [CrossRef] 17. Huang, C.; Thomas, N.L. Fabricating porous poly(lactic acid) fibres via electrospinning. Eur. Polym. J. 2018, 99, 464–476. [CrossRef] 18. Nezarati, R.M.; Eifert, M.B.; Cosgriff-Hernandez, E. Effects of Humidity and Solution Viscosity on Electrospun Fiber Morphology. Tissue Eng. Part C Methods 2013, 19, 810–819. [CrossRef] Polymers 2019, 11, 677 11 of 11

19. Pant, H.R.; Neupane, M.P.; Pant, B.; Panthi, G.; Oh, H.-J.; Lee, M.H.; Kim, H.Y. Fabrication of highly porous poly (ε-caprolactone) fibers for novel tissue scaffold via water-bath electrospinning. Colloids Surf. B Biointerfaces 2011, 88, 587–592. [CrossRef] 20. Katsogiannis, K.A.G.; Vladisavljević,G.T.; Georgiadou, S. Porous electrospun polycaprolactone (PCL) fibres by phase separation. Eur. Polym. J. 2015, 69, 284–295. [CrossRef] 21. Khambatta, F.B.; Warner, F.; Russell, T.; Stein, R.S. Small-angle X-ray and light scattering studies of the morphology of blends of poly(ε-caprolactone) with poly(vinyl chloride). J. Polym. Sci. Polym. Phys. Ed. 1976, 14, 1391–1424. [CrossRef] 22. Bohdanecký, M.; Tuzar, Z. Unperturbed dimensions of the molecules of poly-ε-caprolactam. Collect. Czechoslov. Chem. Commun. 1969, 34, 2589–2597. [CrossRef] 23. Desbief, S.; Grignard, B.; Detrembleur, C.; Rioboo, R.; Vaillant, A.; Seveno, D.; Voué, M.; De Coninck, J.; Jonas, A.M.; Jérôme, C.; et al. Superhydrophobic Aluminum Surfaces by Deposition of Micelles of Fluorinated Block Copolymers. Langmuir 2010, 26, 2057–2067. [CrossRef] 24. Dikić,T.; Ming, W.; van Benthem, R.A.T.M.; Esteves, A.C.C.; de With, G. Self-Replenishing Surfaces. Adv. Mater. 2012, 24, 3701–3704. [CrossRef] 25. Xie, C.; Zhang, P.; Zhang, Z.; Yang, C.; Zhang, J.; Wu, W.; Jiang, X. Drug-loaded pseudo-block copolymer micelles with a multi-armed star polymer as the micellar exterior. Nanoscale 2015, 7, 12572–12580. [CrossRef]

26. Hsueh, M.-L.; Huang, B.-H.; Lin, C.-C. Reactions of 2,20-(2-Methoxybenzylidene)bis(4-methyl-6-tert-butylphenol) with Trimethylaluminum: Novel Efficient Catalysts for “Living” and “Immortal” Polymerization of ε-Caprolactone. Macromolecules 2002, 35, 5763–5768. [CrossRef] 27. Ko, B.-T.; Lin, C.-C. Efficient “Living” and “Immortal” Polymerization of Lactones and Diblock Copolymer of ε-CL and δ-VL Catalyzed by Aluminum Alkoxides. Macromolecules 1999, 32, 8296–8300. [CrossRef] 28. Brintzinger, H.H.; Fischer, D.; Mülhaupt, R.; Rieger, B.; Waymouth, R.M. Stereospecific Olefin Polymerization with Chiral Metallocene Catalysts. Angew. Chem. Int. Ed. 1995, 34, 1143–1170. [CrossRef] 29. Dubois, P.; Ropson, N.; Jérôme, R.; Teyssié, P. Macromolecular Engineering of Polylactones and Polylactides. 19. Kinetics of Ring-Opening Polymerization of ε-Caprolactone Initiated with Functional Aluminum Alkoxides. Macromolecules 1996, 29, 1965–1975. [CrossRef] 30. Jiang, S.; Ji, X.; An, L.; Jiang, B. Crystallization behavior of PCL in hybrid confined environment. Polymer 2001, 42, 3901–3907. [CrossRef] 31. Casper, C.L.; Stephens, J.S.; Tassi, N.G.; Chase, D.B.; Rabolt, J.F. Controlling Surface Morphology of Electrospun Polystyrene Fibers: Effect of Humidity and Molecular Weight in the Electrospinning Process. Macromolecules 2004, 37, 573–578. [CrossRef] 32. Hajiali, H.; Summa, M.; Russo, D.; Armirotti, A.; Brunetti, V.; Bertorelli, R.; Athanassiou, A.; Mele, E. Alginate–lavender nanofibers with antibacterial and anti-inflammatory activity to effectively promote burn healing. J. Mater. Chem. B 2016, 4, 1686–1695. [CrossRef] 33. Zhang, W.; Huang, C.; Kusmartseva, O.; Thomas, N.L.; Mele, E. Electrospinning of polylactic acid fibres containing tea tree and manuka oil. React. Funct. Polym. 2017, 117, 106–111. [CrossRef] 34. Reshmi, C.R.; Sundaran, S.P.; Juraij, A.; Athiyanathil, S. Fabrication of superhydrophobic polycaprolactone/ beeswax electrospun membranes for high-efficiency oil/water separation. RSC Adv. 2017, 7, 2092–2102.

© 2019 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/).