Tunable Structure And Properties Of Segmented Thermoplastic .

Polymers 2019, 11, 1910MDI4 of 22250.12-402.2. Polyurethane SynthesisThermoplastic polyurethanes were obtained via a two-step, prepolymer synthesis method [16].In the first step, the macrodiol and an excess of diisocyanate were poured in a reactor at a temperatureof 70 C over 1 h in an argon atmosphere to form a prepolymer of polyol endcapped with diisocyanategroups. In a second step, butanodiol at a molar ratio of NCO/OH 1.03 was added to the prepolymerin a SpeedMixerTM Dac 600.1 FVZ mixer (Landrum, SC, USA) at room temperature for 1 min at 2250r.p.m. The subsequent PU solution was cast on aluminium moulds at 90 C, and was compressionmoulded at a pressure of 50 bars and temperature of 100 C for 24 h using a water-cooled hydraulicCarver press model 4128CE S/N 4128-220 (Wabash, IN, USA). The cooling procedure was keptuniform by carefully controlling the water flow rate. PU plaques were 2 mm thick.A molar ratio of 1:3:2 (polyol/MDI/BD) was used for the synthesis of all polyurethanes. Sampleswere tested in the as-moulded condition only. The experimental results are the mean value of at leastthree independent tests for every system.2.3. Characterization Techniques2.3.1. Differential Scanning Calorimetry (DSC)DSC scans were performed using a TA Instrument Q20 (New Castle, DE, USA) equipped witha refrigerated cooling system and nitrogen purge. Calibration was performed with indium accordingto the manufacturer’s recommended procedures. The uncertainty associated with each temperatureis approximately 2 C. About 4–6 mg of sample was sealed in an aluminium pan for every test.Thermal behavior was investigated by scanning the samples from –80 to 220 C at a heating rate of20 C·min–1. Previous thermogravimetric analysis results show that these samples are stable until 250 C [16]. After the first scan samples were cooled with liquid nitrogen, a second scan was immediatelyrecorded. The midpoint of the heat capacity change was chosen to represent Tg, Tm refers to theendotherm peak temperature, and the area of the endotherm peak to the melting enthalpy is H.2.3.2. Fourier Transform Infrared-Attenuated Total Reflection Spectroscopy (FTIR-ATR)FTIR-ATR measurements were performed with a Thermo Nicolet Nexus FTIR spectrometer(Waltham, MA, USA) equipped with a multiple internal reflection accessory ATR single bounce.Samples were pressed against ATR accessory diamond crystal by means of the fixing screw using aflat tip. Single beam spectra of the samples were obtained after averaging 128 scans between 4000and 400 cm–1 with a resolution of 4 cm–1. All spectra were obtained in the transmittance mode.2.3.3. Dynamic Mechanical Analysis (DMA)DMA was performed on a 2980 Dynamic Mechanic Analyzer (TA instruments) equipped withtensile head and reducing force option using the Custom Test and single cantilever geometry.Calibration was performed as per the manufacturer’s recommendations included in TA software,version 4.5A. The experiments were carried out on rectangular samples of dimensions close to (18.000 6.000 2.000) mm3. The experimental conditions employed were frequency of 1 Hz and amplitudeof 30 m (linear viscoelastic region) with a temperature ramp of 3 C·min–1 and a scanningtemperature range from –100 to 180 C. These experiments yield the storage modulus (E′), the lossmodulus (E′′), and the damping factor tan δ ( E′′/E′). The glass transition temperature wasdetermined from the peak of the tan δ curve.

Polymers 2019, 11, 19105 of 222.3.4. Synchrotron Small-Angle X-Ray Scattering (SAXS)SAXS experiments (samples with 10 mm of diameter and 1 mm thickness) were done on theSAXS1 beamline of the Brazilian Synchrotron Light Laboratory (LNLS). The X-ray was monitoredwith a photomultiplier and detected on a Pilatus (300 k, 84 mm 107 mm) positioned at 1000 mm,generating scattering wave vectors, q, from 0.12 to 4.0 nm –1. The wavelength of the incident X-raybeam, λ, was 0.155 nm. Silver behenate (AgBH) was used to calibrate the diffraction angle.Polyurethane samples were placed in perpendicular position regarding the X-ray beam at roomtemperature. The background and parasitic scattering were determined by separate measurementson an empty holder and subtracted.The PU morphology can be explained through the use of a pseudo two-phase systemconsidering that the copolymer structure is composed of periodical stacks of alternate lamellarcrystals and amorphous layers [18,19]. The long period (Lp), amorphous thickness (La), and crystallinethickness (Lc) were determined using Lorentz-corrected plots and the one-dimensional correlationfunction, (r) [19–21]. The (r) function was calculated according to a procedure given in the literature[18,22] with the following equation: (r ) I (q)q02 cos(qr )dq q I (q)dq21 Q q I (q) cos(qr )dq2(1)00where r is the real space direction perpendicular to the lamellar surfaces, and Q is the invariant andrepresents the electron density difference between the hard and soft phases. In this work, theinterdomain distance (bLp) obtained by the correlation function corresponds to the r value of the firstmaximum of the (r) data. The r value at the first zero (r0) is defined to be r0 H(1–Vh), where H is thethickness of hard domain and Vh corresponds to its volume fraction. The H value is determined basedon the right triangle whose hypotenuse passes through at (r) 1 and (r) 0, and whose baseline istangent to the (r) curve at its minimum [23]. By the combination of r 0 and H data, it is possible tocalculate Vh 1–(r0/H). The soft domain thickness, S, is defined as the difference between H andinterdomain distance: S bLp–H. Another important structure parameter in the PU copolymer is theaverage interface thickness between rigid and flexible segments, IT, which is obtained from the ratioof the hard thickness to the first minimum long period: IT H2/Lpmin [22].2.3.5. Shore D hardnessShore D hardness was measured at room temperature using a Zwick Roell (Ulm, Germany)analogical hardness testing apparatus following “UNE-EN ISO 868:1998: Plastics and ebonite.Determination of indentation hardness by means of a durometer (Shore hardness)” standardprocedure at (23 2) C and 50% relative humidity.2.3.6. Tensile propertiesTensile properties were measured at 23 C on five replicates of each material with an InstronModel 5582 Universal Testing machine (Grove City, PA, USA) according to “ISO 527-3 Testingmethod for thermoplastic polyurethane elastomers”. A 100 kN load cell was used and the cross headspeed was 200 mm/min. Pneumatic grips were required to hold the test specimens.2.3.7. Durability testsThe polyurethanes were subjected to different durability tests according to a method based onthe international standard ISO 13206, ‘Thermoplastic covering films for use in agriculture andhorticulture.’ These studies are as follows: (a) heating resistance: test pieces were heated in Gear oven

Polymers 2019, 11, 19106 of 22P Selecta at 120 C for 15, 30, and 50 days; (b) hydrolytic resistance: test pieces were immersed inwater at 80 C for 20, 40, and 60 days; (c) oil resistance: test pieces were immersed in BP Oil CS 150 at100 C for 10, 20, and 30 days; (d) weather resistance: test pieces were exposed in sunshineweatherometer for 200 h. Weather conditions: λ 340 nm borosilicate filters, radiation of 35 Wm2 nm,T (65 3) C; and relative humidity of 65% 5%. A dried cycle of 102 min was followed by 1 minof spray water (raining simulation).After the degradations, the material retention of tensile properties was measured following theequation:𝑝𝑟𝑜𝑝𝑒𝑟𝑡𝑦 𝑟𝑒𝑡𝑒𝑛𝑡𝑖𝑜𝑛 (%) 𝑣𝑎𝑙𝑢𝑒 𝑎𝑓𝑡𝑒𝑟 𝑑𝑒𝑔𝑟𝑎𝑑𝑎𝑡𝑖𝑜𝑛 𝑡𝑒𝑠𝑡 100𝑣𝑎𝑙𝑢𝑒 𝑏𝑒𝑓𝑜𝑟𝑒 𝑑𝑒𝑔𝑟𝑎𝑑𝑎𝑡𝑖𝑜𝑛 𝑡𝑒𝑠𝑡(2)3. Results and DiscussionThe current study gives valuable information about the influence of the flexible segmentmolecular structure and chain length on the morphology, thermal, and mechanical properties andresistance to external agents like weather, water, oil, and heat of thermoplastic polyurethanes. Themain task is to evaluate the influence of the soft phase in order to tailor a polyurethane with selectedproperties for a specific application. So, segmented thermoplastic polyurethanes with five differentflexible segment molecular structures and two molar masses (1000 and 2000 g/mol) were synthesizedwithout solvent by the two-step method. Butanediol (BD) was used as chain extender and 4,4’diphenylmethane diisocyanate (MDI) is the diisocyanate that was used to react with the OH groupsof the polyols. The notation used in this article is PU-XY. The letter X denotes the flexible segmenttype, that is, X UH, PH, UHC, PCL, or PTMG (see Table 1), and the letter Y 100 or 200 refers to themacrodiol molar mass, that is, 1000 or 2000 g/mol. Different techniques were used to determine thedegree of phase mixing rigid-flexible segments that determine polyurethanes properties, thusrelating with mechanical properties and resistance to degradation under different agents.Differential scanning calorimetry (DSC) curves were obtained from the as-casted systems todetermine the behavior under heat flow. Figure 1 depicts the first and the second scans of all thesamples assayed as a function of temperature. All the systems show two temperature regions, withthe one at a low temperature showing a glass transition temperature and the one at a hightemperature as an endothermic peak [24–27]. The glass transition temperature observed at lowtemperature values is related to the amorphous part of the flexible segment. Values of Tg of thepolyurethane strongly depend on the type of macrodiol employed in the synthesis. This value isindicative of the soft and rigid segment mixing degree. The higher the difference between the puremacrodiol or flexible segment, Tg,s (see Table 1), and the Tg of the final polyurethane, the highermiscibility or compatibility degree rigid-flexible segments [10]. In order to determine the degree ofmixing macrodiol or flexible segment with rigid segment, values of the difference between the glasstransition temperature of the polyurethane (Tg in Table 2) and the glass transition temperature of themacrodiol (Tg,s in Table 1), (Tg – Tg,s), are compiled in Table 2. These values decrease in the followingorder: PH UH UHC PCL PTMG. That is, polyurethanes based on macrodiols with carbonategroups show higher Tg and Tg increment than when an ester and a more extended ether group areincluded in the structure. The highest phase mixing is found with PH macrodiol with odd and evencarbonate groups repeating units, which increases interactions and phase mixing. UH-basedpolyurethanes with a homogeneous structure with six methylene repeating units decrease urethaneto carbonate geometric fit and phase mixing. PU-UHC with ester linkages in addition to carbonateones impedes geometrical fit with urethane groups, thus decreasing interactions and, consequently,phase mixing. Thus, polycarbonatediol polyurethanes depict the highest Tg values, that is, the highestflexible segment-rigid segment interaction and phase mixing. Also, the Tg increments are higher forthe PCD-based PUs caused by an increase in partial mixing owing to an increase in molecular

Polymers 2019, 11, 19107 of 22interactions, thus reflecting the best miscibility of flexible segment phase with the amorphous phaseof the rigid segments [27,28]. The ester group imparts similar phase mixing characteristics. Theopposite situation occurs for PTMG as macrodiol with an ether functionality that impairs molecularinteraction, thus decreasing phase mixing and lowering Tg. This trend is supported by looking on themolar attraction constants for ether, ester, and carbonate groups that are 256, 512, and 767J3/2cm3/2mol 1, respectively. [29] Higher molar attraction and dipolar moment values of carbonategroups result, in general, in higher overall phase mixing.At higher temperatures, 140–170 C, there is an endotherm peak corresponding to the meltingof the rigid segment phase. The value of the endotherm shifts to higher temperatures and the valueof the associated enthalpy increases with the decrease in molar mass of macrodiol from 2000 to 1000g/mol. This can be related to the formation of longer microdomains of rigid segment or structureswith a greater degree of organization. This trend agrees with the results obtained in the literature[7,8,15,30–32].In the second DSC scan (Figure 1), the glass transition temperature of all systems increases tohigher values, indicating that the heating process increases miscibility between domains. Values ofTg and, consequently, (Tg–Tg,s) increase in the second scan, causing phase mixing. Neithercrystallization exotherms nor melting endotherms are observed. The chains are not allowed tocrystallize under the cooling process applied to the samples. Between the first and second scan, rapidcooling is carried out after reaching the melting point of the materials. This causes a rearrangementof the chains and facilitates a greater interaction between chains, increasing the miscibility betweenrigid and flexible segment, as can be seen in the increases in Tg for all materials. This cooling alsocauses the disappearance of the melting point in some of the materials owing to the short time thatcooling lasts, which makes material recrystallization impossible.Polyurethanes with macrodiol of lower molar mass (1000 g/mol) show higher values of T g and(Tg – Tg,s) than PUs with 2000 macrodiol molar mass (Figure 1, Table 2). This implies higher miscibilityor rigid-flexible segments’ phase mixing. Furthermore, there is a gradual broadening of the glasstransition with decreasing flexible segment molar mass, which is associated with increasedheterogeneity in the flexible segment microdomains, as observed by other authors [29]. Increasingthe molar mass from 1000 to 2000 is reflected in a decrease of the polyurethane Tg, an increase in themelting temperature, and a decrease in the enthalpy (see Table 2). Lower flexible segment molar massimplies better geometrical fit owing to lower steric hindrance with the urethane groups of the rigidsegments, a stronger interaction, and thus better flexible segment than with higher flexible segmentmolar mass because the stoichiometric relation (1:2:3) is kept constant.

Polymers 2019, 11, 19108 of 22Figure 1. Differential scanning calorimetry (DSC) curves for the different polyurethanes (PUs) studiedshowing the transitions characteristics of first (continuous line) and second scans (dashed line). (a)system with macrodiol of 1000 g/mol molar mass and (b) system with macrodiol of 2000 g/mol. UH,poly(hexamethylene) carbonate diol; PH, poly(hexamethylene–pentamethylene) carbonate diol;UHC, poly(hexamethylene–caprolactone) carbonate diol; PCL, polycaprolactone polyester diol;PTMG, poly(tetramethylene ether) glycol; MDI, 4,4’-diphenylmethane diisocyanate; BD, 1,4butanediol.The fraction of rigid segment present in the soft phase, wH, can be determined from values ofPU’s Tg. Thus, keeping in mind that the change in a thermal property of a single phase twocomponent system is the linear weight addition of the two individual component changes in thatproperty, the following relation holds [33]:Tg (1 wH ,mix )Tg ,s wH ,mixTg , H(3)where Tg,s is the Tg of the soft phase and Tg,H is the Tg of the hard phase formed by MDI BD, andwH,mix is the corresponding weight fraction of rigid segment in the amorphous soft phase accordingto the mixing rule. The value of Tg,H 110 C was determined in this article, in good agreement withthe value from the literature [33].Values of wH are depicted in Table 2 for all systems according to the first and second scan. Thesevalues decrease in the following order: UH PH UHC PCL PTMG, with the increase in molarmass of the flexible segment from 1000 to 2000. This trend is in good agreement with that observedby the Tg trends.Table 2. Values of the glass transition temperature (Tg); temperature difference (Tg – Tg,S); meltingtemperature (Tm); melting enthalpy (ΔH); fraction of rigid segment in the soft phase, wH,DSC; fractionof hydrogen-bonded urethane carbonyl groups in the rigid segment region, Xb; and maximum rigidsegment–flexible segment mixing determined by FTIR analysis, wH,FTIR. DSC, differential scanningcalorimetry; FTIR, Fourier transform infrared.SystemTg ( C)(Tg – Tg,S)Tm ( UH100–6.4( TMG200–69.76.5162.237.840.040.780.09Dynamic mechanical analysis (DMA) was used to obtain information on viscoelastic properties,which can be related to soft and hard microdomain thermal transitions [9,26,34–36]. Figure 2 showsplots of the storage modulus (E′) and the dissipation factor or tan delta (tan δ) as a function oftemperature for the different polyurethanes. At a low temperature, the storage modulus shows a highand constant value characteristic of its glassy state, higher for the system with macrodiol of lowermolar mass. A drop in stiffness is observed accompanying the soft domain glass transition as thetemperature increases, sharper for the system with the highest polyol molar mass [9,26,37,36]. The

Polymers 2019, 11, 19109 of 22onset of E’ decay occurs at a higher temperature for the polycarbonate-based polyols, indicating ahigher ability to restore the energy supplied mechanically to the system, and is related to rigidity.The existence of an elastomeric plateau after the abrupt decrease in modulus indicates the existenceof physical crosslinks because of the increase in the size of macrodiol and interconnectivity with rigidsegment domains acting as physical cross-linking points. Thus, a continuous microdomain structureis developed, providing significant reinforcement and elastomeric behavior. [9,36] All the samplesdepict an elastomeric plateau that widens with the increase in the molar mass of the polyol [16,38].The increase of polyol molar mass or flexible chain length in polyurethanes favors its elastomericbehavior less than for the polycarbonate-based ones.When the molar mass or chain length of the flexible segment decreases from 2000 to 1000 g/mol,the damping peak broadens and flattens related to the degree of ordered and the freedom of motionof molecules in the soft domains, that is, an inhibiting effect on molecular motion of the amorphousregion [39,40]. Tan δ describes the ratio of storage to loss modulus and is a measure of the energydissipation of the material. The drop in storage modulus (E′is because of the glass transition of the amorphous polymer in the semi-crystalline material. As shownin Figure 2, the highest values of maximum temperature of tan δ are observed for the polycarbonatebased polyurethanes, similarly to DSC measurements. So, the trend of DMA data is consistent withthat of DSC data and suggest that the miscibility between the flexible and rigid segments follows thefollowing order: PH UH UHC PCL PTMG. This trend is the result of the greater flexibility ofthe etheric bond and higher phase separation in comparison with esteric and carbonate bonds. ThePU-PTMG and PU-PCL systems, without polycarbonate groups, depict a higher phase segregation,and thus a higher elastomeric plateau [26,41].Figure 2. Storage modulus (E’) and tan δ of materials based on polyols with molecular mass equal to1000 (part a) and 2000 (part b) g/mol.Infrared spectroscopy was used to study morphology aspects in TPUs. Figure 3a shows the FTIRspectra of the systems with macrodiol 1000 g/mol molar mass, as an example. Different absorptionpeaks can be used to characterize these materials, that is, the OH absorption at 3500 cm –1, the NHstretching vibration at 3500–3000 cm–1, and the absorption of NCO groups at 2260 cm–1 and CO groupof urethane at 1800–1640 cm–1. A small broad peak at 3276 cm–1 appears, resulting from the formationof NH of the urethane linkage, while the –OH band at 3470 cm–1 of the corresponding polyolsdisappears. Simultaneously, the disappearance of the NCO stretching band at 2273 cm –1 as a

Polymers 2019, 11, 191010 of 22consequence of the reaction between OH and NCO

and 2000 g/mol, and 4,4'-diphenylmethane diisocyanate (MDI) and 1,4-butanediol as a rigid segment. The polyurethanes obtained reveal a wide variation of microphase separation degree that is correlated with mechanical properties and retention of tensile properties under degradation by heat, oil, weather, and water.