Regulation of Hard Segment Cluster Structures for High‐performance Poly(urethane‐urea) Elastomers

Abstract Elastomers are widely used in daily life; however, the preparation of degradable and recyclable elastomers with high strength, high toughness, and excellent crack resistance remains a challenging task. In this report, a polycaprolactone‐based poly(urethane‐urea) elastomer is presented with excellent mechanical properties by optimizing the arrangement of hard segment clusters. It is found that long alkyl chains of the chain extenders lead to small and evenly distributed hard segment clusters, which is beneficial for improving mechanical properties. Together with the multiple hydrogen bond structure and stress‐induced crystallization, the obtained elastomer exhibits a high strength of 63.3 MPa, an excellent toughness of 431 MJ m−3 and an outstanding fracture energy of 489 kJ m−2, while maintaining good recyclability and degradability. It is believed that the obtained elastomer holds great promise in various application fields and it contributes to the development of a sustainable society.


Fabrication of the elastomer
PCL diol (12 g) was added to the three-neck round bottom flask and heated at 80 ℃ to melt.N 2 was injected into the three-neck flask to purge the air for 30 minutes to remove moist air.HDI (2.42 g) and DBTDL (0.025 g) dissolved in 3 mL DMAc were then added into the flask.The obtained mixture was further heated at 80 °C under N 2 atmosphere and stirring for 3 h and then cooled to 40 °C.Consequently, chain extender (1.05 g AD for PCL-AD and 1.55 g DD for PCL-DD) dissolved in 100 mL DMAc was added into the reaction system, which was further kept at 40 ℃ for 18 h under stirring and N 2 atmosphere.The as-obtained viscous polymer solution was next poured into homemade glass dish and then heated on a hot plate at 70 ℃ overnight to obtain the final elastomer.

General characterization
Attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectroscopy (Nicolet iS10, Thermo Fisher Scientific, USA), proton-1 and carbon-13 nuclear magnetic resonance ( 1 H-NMR and 13 C-NMR) (600 MHz JNM-ECZ600R spectrometer, JEOL, Japan) were used to determine the chemical composition of the obtained elastomers.Differential scanning calorimetry (DSC) of samples were recorded on a TA Discovery DSC 2500 analyzer under N 2 atmosphere.The samples were heated from -80 °C to 150 °C at 20 °C min -1 and cooled at 20 °C min -1 from 150 °C to -80 °C to eliminate thermal history.Final data were collected during the second heating run from -80 °C to 150 °C at a scan rate of 10 °C min -1 .Wide-angle Xray diffraction (WAXD) measurements were conducted on a Bruker D8 diffractometer using Cu Kα radiation (λ = 1.5406Å).Thermogravimetric analysis (TGA) was conducted on a TGA 55 (TA Instruments) from 30 to 600 °C at a heating rate of 10 °C min -1 under air atmosphere.Dynamic mechanical analyses (DMA) were performed on the Dynamic Mechanical Analyzer Q800 (TA Instruments) under the tension mode in the temperature range from -80 to 200 °C at a heating rate of 3 °C min -1 .Gel permeation chromatography (GPC; Waters 1515 GPC system, Waters Corporation, US) was used to measure the molecular weight of the synthesized elastomer.Scanning electron microscope (SEM, TESCAN Mira3) was used to analyze the micromorphology of elastomer samples before and after degradation.Transmission electron microscope (TEM; Talos F200X, America) was used to observe microphase separation structure of elastomer samples.Thin slices of PCL-AD and PCL-DD elastomers were prepared by cryo-ultrathin sectioning (Leica EM UC7 & FC7, America) and stained with phosphotungstic acid before TEM tests.

Tensile tests
The mechanical properties of the samples were tested using a universal tensile machine (Instron 68TM-5, 5kN) at a speed of 50 mm min -1 at room temperature.
Uniaxial tensile test samples were cut to a standard dumbbell shape and the size of the middle rectangular area of the sample was 4 mm wide and 17 mm long.Test length of the sample between two clamps was controlled at ~10 mm.At least five individual tensile tests were conducted for each sample.
Tensile toughness () of the samples can be calculated by integrating the area under the engineering stress-strain curves, using the following equation: where  is the engineering stress,  is the engineering strain,  is the elongation-atbreak of the sample.
True stress () and true strain () are calculated based on the engineering stressstrain curves by the following equation [1] : where  is the engineering stress,  is the engineering strain,  is the instant length of the deformed sample,  0 is the original length of the sample.
For the cyclic tensile test, the loading and unloading processes of the samples at 100% strain were performed continuously for 100 times at room temperature and at a speed of 50 mm min -1 .Next, the samples were placed in an oven at 80°C and heated for 5 min to restore the original shape.The heated samples were loaded and unloaded again at the same rate for the second cyclic tensile test.
Fracture energy (Γ) is the energy per unit area required for crack propagation in a material and can be tested through pure-shear test and trouser tearing test. [2]For pureshear test, the width, thickness, length between two clamps and notch size of the sample were ~25 mm, ~0.35 mm, ~5 mm and ~10 mm, respectively.The pure-shear test was performed at a speed of 50 mm min -1 , and the fracture energy was calculated by the following equation [2] : where H is the distance between the two clamps when the precut sample is undeformed, λ c is the critical strain when the crack begins to propagate noticeably, W(λ c ) is the energy per volume of the uncut sample while stretched at the critical strain λ c for crack propagation.
For trouser tearing test, the samples were prepared with a width of 50 mm, a length of 80 mm, a thickness of ~0.35 mm and an initial notch of 30 mm in the middle along the length direction.The trouser tearing test was conducted at room temperature at a stretching speed of 50 mm min -1 , and the tearing force-distance curve was obtained.
Fracture energy was calculated based on the tearing force-distance curve, by the following equation [3] : where P is the force on the legs of the sample, λ is the extension ratio in the legs (ratio of length of deformed to undeformed leg), h is the sample thickness, b is the width of the legs and w the strain energy density in the legs.During the test, no significant changes were observed in the legs of the samples, so λ=1 and w=0 were used.

Recycling and degradation tests
The recycling test was conducted by re-dissolving the elastomer fragments in DMF at 120 ℃.The solution was then poured into a homemade glass dish and dried overnight in a 70 ℃ oven.
For the degradation test, dried samples were weighed and placed in ~5 mL PBS (pH=7.2,Guangzhou Hewei Pharmaceutical Technology Co., Ltd.) buffer solution containing ~0.1 g lipase at room temperature.The samples were taken out every 24 hours, and the surface of the samples was cleaned with deionized water.The cleaned samples were then dried and weighed.The degradation solution was replaced every 3 days to maintain the enzyme activity.Three sets of experiments were performed simultaneously to calculate the average mass reduction rate.

All-atom molecular dynamics simulations
All molecular dynamics simulations were performed through Materials Studio.The amorphous cell of the elastomer was first constructed using the Amorphous Cell module.Each cell included 3 molecular chains consisting of 4 hard segments and 5 soft segments.Then the Forcite module was used to perform geometry optimization and annealing operations on the amorphous cells with COMPASSII force field.The calculated quality was always selected as fine and the charges were assigned by the force field.Each amorphous cell experienced 10 times of gradual heating processes from 300 K to 800 K and gradual cooling processes from 800 K to 300 K under NPT ensemble at 1 atm pressure, with a total simulation time of 5 ns.Then the NVT ensemble was selected to perform a 1 ns dynamic simulation at a temperature of 298 K and a pressure of 1 atm.The Velocity Scale algorithm was used to control temperature, and the Berendsen algorithm was used to control pressure.All other parameters during the simulation were default parameters.The cohesive energy density (CED) was calculated directly using the Forcite module on the amorphous cell after completing annealing and dynamics simulations.N atoms and O atoms were selected as donors and acceptors of hydrogen bonds, respectively.The criterion for the hydrogen bond in cells was that the length of H--O bond was smaller than 2.5 Å and the angle of O--H-N was greater than 120 degrees.All hydrogen bonds were generated from the N-H structure on the hard segments, and the three polymer chains in the simulations had a total of 12 hard segments, so theoretically up to 96 hydrogen bonds could be generated.After optimization, there were 62 and 56 hydrogen bonds in the cells of PCL-AD and PCL-DD, respectively.

Density functional theory (DFT) calculations
Density functional theory (DFT) calculations were performed by Gaussian 16. [4] the B3LYP-D3(BJ) functional and 6-311 G** basis sets were used for geometry optimization of two AD hard segments or two DD hard segments.The optimized hard segment structures were imported into Materials Studio to more clearly display the generation of hydrogen bonds.The criterion for the hydrogen bond was that the length of H--O bond was smaller than 2.5 Å and the angle of O--H-N was greater than 120 degrees.
In order to obtain the bonding energy between two hard segments, single-point energy calculations were performed at the optimized geometries using the B3LYP-D3(BJ) functional and def2-TZVP basis set.All the binding energies were corrected by the basis set superposition error (BSSE) using counterpoise command.

Figure S6 .
Figure S6.FTIR spectra of (a) PCL-AD and (b) PCL-DD in the C=O stretching vibration region.

Figure S7 .
Figure S7.Temperature dependence of the storage modulus G' and loss modulus G'' of the (a) PCL-AD and (b) PCL-DD elastomers.(c) Tan δ as functions of temperature for PCL-AD and PCL-DD elastomers.

Figure S8 .
Figure S8.TGA curves of the PCL-AD and PCL-DD elastomers.

Figure S9 .
Figure S9.MD simulations of the structures of the (a) PCL-AD and (b) PCL-DD elastomers.For clearer display, parts of the soft segments were hidden.An enlarged view of the local area can be seen on the right, and the angles of these areas have been adjusted appropriately for better display.

Figure S11 .
Figure S11.WAXD results of the PCL-AD and PCL-DD elastomers at initial condition.

Figure S13 .
Figure S13.Cyclic loading-unloading tensile curves of the original and recovered (a) PCL-AD and (b) PCL-DD elastomers subjected to the first cycle at a strain of 100% under a deformation rate of 50 mm min -1 .

Figure
Figure S14.a) Schematic of the samples (80 mm long, 50 mm wide) for the trouser tearing test.b) Force-displacement curve of the PCL-DD elastomer (80 mm long, 50 mm wide, ≈0.42 mm thick, and 30 mm notch) measured by the trouser tearing test.

Figure S15 .
Figure S15.a) Tensile curves of intact and notched PCL-AD measured by the pure shear test.b) Force-displacement curve of the PCL-AD elastomer (80 mm long, 50 mm wide, ≈0.35 mm thick, and 30 mm notch) measured by the trouser tearing test.

Figure S17 .
Figure S17.Solvent recycling process of the PCL-DD elastomer.

Figure S19 .
Figure S19.Tensile curves of intact and recycled PCL-AD elastomer.

Figure S20 .
Figure S20.Degradation of PCL-AD elastomer in lipase PBS solution.

Table S1 .
Summary of the molecular weights (Mn and Mw) and polydispersity index (PDI) of the PCL-AD and PCL-DD elastomers.

Table S2 .
Summary of the assignment of the deconvoluted subpeaks in the FTIR C=O absorption bands for the PCL-AD and PCL-DD elastomers.

Table S3 .
Summary of the bond lengths and bond angles of hydrogen bonds in AD and DD hard segments.