Bioinspired All-Polyester Diblock Copolymers Made from Poly(pentadecalactone) and Poly(2-(2-hydroxyethoxy) benzoate): Synthesis and Polymer Film Properties

approach, end-functionalized PPDL and P2HEB are obtained separately by ROP with functional initiators, and connected by 1,3-dipolar Huisgen reaction (“click-chemistry”). Block copolymer compositions from 85:15 mass percent to 28:72 mass percent (PPDL:P2HEB) are synthesized, with M n of from about 30 000–50 000 g mol − 1 . The structure of the block copolymer is confirmed by proton NMR, Fourier-transform infrared spectroscopy, and gel permeation chromatography. Morphological studies by atomic force microscopy (AFM) further confirms the block copolymer structure, while quantitative nanomechanical AFM measurements reveal that the Derjaguin–Muller–Toporov moduli of the block copolymers range between 17.2 ± 1.8 and 62.3 ± 5.7 MPa, i.e., between the values of the parent P2HEB and PPDL homopolymers (7.6 ± 1.4 and 801 ± 42 MPa, respectively). Differential scanning calorimetry shows that the thermal properties of the homopolymers are retained by each of the copolymer blocks (melting temperature 90 ° C, glass transition temperature 36 ° C). work, it would be possible to conclude that all-polyester diblock copolymers without transesterification are available if at least one carbonyl group of the two blocks is sufficiently sterically hindered.While the sequential approach only led to block copolymers with limited molar masses of the P2HEB block, and thus to limited block ratios, the modular approach gave access to the whole range of desired molar fractions of PPDL-triazole-P2HEB . This enabled a systematic study of changes in the poly mer film properties with composition for this sample series. These studies revealed a distinct morphology at each composition, and nanomechanical moduli in between the limiting values of the parent homopolymers. In future work, the barrier properties of these thin films will be evaluated.


Introduction
Aliphatic and aromatic polyesters are not an invention of humankind-nature has made use of these highly hydrophobic polymers for a much longer time. For example, the biopolyesters cutin and suberin are found in many plant tissues that form barrier layers of the plant to its environment, e.g., leave Ring-opening polymerization (ROP) is a relatively recent method to obtain polyesters from cyclic lactones, and even large macrocycles like pentadecalactone can be polymerized via ROP with suitable catalysts to yield poly(pentadecalactone) (PPDL). [10][11][12][13][14][15][16][17] ROP is a much more elegant method to synthesize polyesters than polycondensation, as it yields products with precisely defined end groups, controlled molecular masses, and low polydispersity. [18,19] In the here presented context, the polyester PPDL is particularly interesting: first, because pentadecalactone can be obtained from renewable resources, [20] and second, because of the structural resemblance of PPDL to polyethylene, with a long aliphatic C 14 chain and only one heteroatom in the main chain per polymer repeat unit. PPDL is a semicrystalline polymer with about 60% crystallinity, and was reported to have a melting temperature T m near 100 °C, a glass transition temperature T g of −27 °C, and a crystal structure similar to both polyethylene and poly(ε-caprolactone). [21] It thus resembles low density polyethylene, [17] but has a slightly better solubility in organic solvents, thus allowing for solution processing. For example, it is soluble in chloroform at room temperature. Its hydrolytic degradation can be catalysed by metals, organocatalysts, and enzymes. [22] Overall, the literature on true all-polyester block copolymers containing PPDL is rather scarce, and mostly restricted to all-aliphatic one. The reason for this is transesterification problems during synthesis. If one of the blocks of an all-polyester block copolymer is not sufficiently stabilized against nucleophilic attack, repeat unit scrambling between the blocks can be observed during synthesis of the second block-something that is not possible in block copolymers with just one polyester block. Specifically for PPDL, a significant degree of transesterification within 1 h heating at 100 °C has been shown by model experiments. [17] However, PPDL-block-poly(ε-caprolactone), PPDL-block-poly(εdecalactone), and PPDL-block-poly(L-lactide) block copolymers could be realized with negligible or no transesterification between the blocks using sequential ROP. [16,23,24] Attempts to obtain aromatic-aliphatic block copolymers containing PPDL have been rather limited so far. [25] The suberin lamellae found in the natural material cork remind the polymer chemist of microphase-separated lamellar block copolymer structures. This has inspired us to synthesize aliphatic-aromatic block copolymers based on polyesters with an aliphatic PPDL block because the repeat units of PPDL are structurally similar to the fatty acid-derived aliphatic components of suberin. However, there is an important difference between the natural material and our synthetic model. In suberin, the aromatic part is crystalline, while the aliphatic part is not, due to numerous hydroxyl and epoxy substituents on the fatty acid repeat units, and due to branching via glycerol. In our model, the system is reversed, with a crystalline PPDL block and an amorphous aromatic block. In a previous piece of work, we synthesized the bioinspired all-polyester block copolymer poly(pentadecalactone-block-(3-hydroxy cinnamate)). [25] In this polymer, the PPDL block was combined with an amorphous aromatic poly(3-hydroxy cinnamate) (P3HCA) block. [25] PPDL could be obtained in relatively high molecular masses (number average molecular mass M n up to 43 000 g mol −1 ) and with a low polydispersity index (PDI in most cases < 1.8). The PDI of P3HCA, on the other hand, was broad (2-3), and its polycondensation yielded only a low M n of around 3000 g mol −1 . [25] For this reason, the PPDL-block-P3HCA system did not have a strongly phase separated microstructure, yet it had a structured fibrillar morphology that was distinctly different from either of its parent homopolymers. [25] In this paper, we describe the synthesis of poly(pentadecalactoneblock-(2-(2-hydroxethoxy)benzoate) (PPDL-block-P2HEB) polymers via a) a sequential approach, and b) a modular approach using 1,3-dipolar Huisgen addition ("click reaction") [26][27][28][29][30] to connect the  [1,2,9] Adapted with permission. b) Structure of the target block copolymer made from a poly(pentadecalactone) block and a poly(2-(2-hydroxyethoxy)benzoate) block.
www.advancedsciencenews.com www.mcp-journal.de two blocks. The aim was to obtain a series of block copolymers with different volume ratios, and to compare their morphology and film properties. P2HEB at first glance does not seem to be an obvious choice to build a suberin-inspired synthetic polymer, since other aromatic building blocks like hydroxycinnamic acids, phenols, or catechols have a much higher structural likeness to the aromatic domains of that polymer. However, these are notoriously difficult to polymerize with structural precision or even to high molecular masses. [31][32][33][34][35][36][37][38][39][40][41] Dihydro-5H-1,4-benzodioxepin-5-one, the monomer of P2HEB, has the advantage that it can undergo ring-opening polymerization. [42] It was therefore hoped that P2HEB blocks could be used to initiate pentadecalactone, or that PPDL blocks would initiate dihydro-5H-1,4-benzodioxepin-5-one, preferentially without transesterification. P2HEB has a glass temperature T g of 38 °C and has rubbery properties above T g . Thus, its combination with semicrystalline PPDL should give an interesting combination of properties.

Materials
All chemicals were obtained as reagent grade from Sigma-Aldrich (St. Louis, MO), Carl Roth (Karlsruhe, Germany), or Alfa Aeser (Haverhill, MA) and used as received. High performance liquid chromatography grade solvents were purchased dry from Carl Roth. Freshly distilled chloroform and toluene were stored over molecular sieves in the glovebox. 2,3-Dihydro-5H-1,4-benzodioxepin-5-one (2,3-DHB) and pentadecalactone (PDL) were sublimated and dried under high vacuum for 3 days. Propargyl alcohol (PA) and benzyl alcohol (BA) were dried over calcium hydride and distilled under reduced pressure. 3-azidopropanol (AP) was synthesized according to literature procedures and was distilled under reduced pressure as well. [43] Triazabicyclodecene (TBD) was recrystallized from diethyl ether and dried under high vacuum for 3 days. Toluene, chloroform, 2,3-DHB, PDL, AP, PA, BA, and TBD were stored and handled under nitrogen in a glovebox (MBRAUN, Garching, Germany). Benzyl-and alkyne-functionalized PPDL was synthesized according to the literature. [25]

Instrumentation
Gel permeation chromatography (GPC, in chloroform, calibrated with poly(methyl methacrylate) and polystyrene standards) was performed on PSS SDV 100, 1000, and 10 000 Å columns (PSS, Mainz, Germany) using a 1260 Infinity RI detector (Agilent Technologies, Santa Clara, CA). NMR spectra were recorded on a Bruker 250 MHz spectrometer (Madison, WI) using CDCl 3 as solvent and tetramethylsilane as internal reference. Fourier-transform infrared spectroscopy (FTIR) spectra were taken using a Cary 630 FTIR Spectrometer (Agilent Technologies, Santa Clara, CA). The glass transition temperature and melting temperature were measured by differential scanning calorimetry (DSC). The spectra were recorded on a Phönix DSC 204 F1 device (Netzsch, Selb, Germany) in a temperature range of −20-250 °C and with a speed of 5 K min −1 .
For the formation of polymeric layers on silicon wafers a SPIN150 spin coater (SPS-Europe, Putten, Netherlands) was used, with the spin-coating parameters 3000 rpm, 1000 rpm s −1 , and 30 s. The thickness of each polymer layer was measured with a SE400adv ellipsometer (Sentech Instruments GmbH, Berlin), and the static, advancing, and receding contact angles were measured using an OCA 20 set-up (Data Physics GmbH, Filderstadt, Germany). For each sample, the average values from measurements at three different positions were taken. Atomic force microscopy (AFM) was used in order to record surface topology images using a Dimension Icon AFM from Bruker (Karlsruhe, Germany). RFESP-75 cantilevers (width: 40 µm, length: 235 µm, spring constant: 4.18 N m −1 , resonance frequency: 76 kHz) and ScanAsyst-Air cantilevers (width: 25 µm, length: 115 µm, spring constant: 0.4 N m −1 , resonance frequency: 70 kHz) were used. Quantitative nanomechanical (PeakForce-QNM) measurements were performed as described in the literature. [25] The ScanAsyst-Air cantilever had a deflection sensitivity of 94.96 nm V −1 and the RFESP-75 cantilever had 125.92 nm V −1. The spring constant for the ScanAsyst-Air cantilever was calibrated using thermal tune and resulted in a value of 0.463 N m −1 . The spring constant of the RFESP-75 was calculated using the Sader method [44] and resulted in a value of 4.18 N m −1 . The tip radius was determined using an absolute method. This resulted in a RFESP-75 tip end radius of 6.1 nm for measuring the PPDL sample, and the following ScanAsyst-Air tip end radii: 8.5 nm for measuring the PPDL 21 -triazole-P2HEB 79 sample, 7.4 nm for the cantilever used to measure the PPDL 52 -triazole-P2HEB 48 sample, 9.4 nm for the cantilever used to measure the PPDL 66triazole-P2HEB 34 sample, and 8.5 nm for the cantilever used to measure the P2HEB sample.

Synthesis of Poly(2-(2-hydroxyethoxy)benzoate)
P2HEB was synthesized via ring-opening polymerization under inert gas. The amount of each reagent used, and the molecular weight of the resulting polymers, as determined by GPC, can be found in Table 1. In the glovebox, 2,3-DHB (2.00 g, 12.2 mmol), BA (6.19 mg, 0.057 mmol), and an equimolar amount of TBD (7.96 mg, 0.057 mmol) were put together in a Schlenk tube and dissolved in chloroform (0.7 mL). The reaction mixture was removed from the glovebox and stirred for 3 days at 40 °C. Afterward, 0.1 mL of methanol was added to quench polymerization. For all reactions, 1 H NMR of the crude product was measured to determine the monomer conversion. The crude product was precipitated twice with an excess of methanol. The resulting polymer was dried in vacuo at room temperature for 2 days. 1

Synthesis of Block Copolymers Using Poly(pentadecalactone) as Macroinitiator
The block copolymer PPDL-ester-P2HEB was synthesized using PPDL as the macroinitiator for polymerizing 2,3-DHB on the end of the aliphatic chain. The amount of each reagent used, and the molecular weight of the resulting block copolymers, as determined by GPC can be found in Table 2. In the glovebox, 2,3-DHB (500 mg, 3.05 mmol), PPDL (50.0 mg, M n = 2200 g mol −1 , 0.023 mmol) and an equimolar amount of TBD (3.16 mg, 0.023 mmol) were put together in a Schlenk tube and dissolved in chloroform (0.7 mL). The reaction mixture was removed from the glovebox and stirred for 3 days at 40 °C. Afterward, 0.1 mL of methanol was added to quench poly merization. The crude product was precipitated twice with an excess of methanol. The resulting polymer was dried in vacuo at room temperature for 2 days. 1

Synthesis of Poly(2-(2-hydroxyethoxy)benzoate Using Azidopropanol as Initiator
Azide-functionalized poly(2-(2-hydroxyethoxy)benzoate was synthesized via ring-opening polymerization under inert gas. The amount of each reagent used, and the molecular weight of the resulting polymers determined by GPC can be found in Table 1. In the glovebox, 2,3-DHB (1.60 g, 9.75 mmol), AP (5.39 mg, 0.053 mmol), and an equimolar amount of TBD (7.42 mg, 0.053 mmol) were put together in a Schlenk tube and dissolved in chloroform (0.5 mL). The reaction mixture was removed from the glovebox and stirred for 3 days at 40 °C. Afterward, 0.1 mL of methanol was added to quench polymerization. For all reactions, 1 H NMR of the crude product was measured to determine the monomer conversion. The crude product was precipitated twice with an excess of methanol. The resulting polymer was dried in vacuo at room temperature for 2 days. 1 Table 2. Ring-opening polymerization of 2,3-DHB using PPDL as a macroinitiator and TBD as catalyst at different reaction conditions. Reactions were performed for 3 days at 40 °C in chloroform. a) Number-average molecular mass M n and PDI were determined by GPC in chloroform using PMMA standards; b) The same molar amount was used for the catalyst; c) Calculated molecular mass based on the ratio of the reactants; d) Conversion (relative to 2,3-DHB) and molar ratio was determined by 1

Synthesis of the Block Copolymer PPDL-Triazole-P2HEB
The amount of each reagent used and the molecular weight of the resulting polymers as determined by GPC can be found in Table 3. In the glovebox, PPDL-alkyne (M n = 6.10 kg mol −1 , 76.3 mg, 0.013 mmol PPDL chains), P2HEB-azide (M n = 24.0 kg mol −1 , 100 mg, 0.004 mmol P3HCA chains), Cu I OAc (0.511 mg, 0.004 mmol), and an equimolar amount of TBD (0.580 mg, 0.004 mmol) were put together in a Schlenk flask and dissolved in dry chloroform (1 mL). The reaction mixture was stirred for 2 days at ambient temperature. Afterward, it was removed from the glovebox and the crude product was precipitated twice into an excess of ethanol. Reactions conducted with an excess of PPDL species were dissolved in dichloromethane and unreacted PPDL was filtered off. When the reaction was performed with an excess of P2HEB species, the resulting polymer mixture was dissolved in chloroform and precipitated in ice cold dichloromethane in order to remove unreacted P2HEB chains. The resulting block copolymers were dried in vacuo at room temperature for 2 days. 1

Polymer Film Formation
Polymer films were formed via spin coating of the respective polymer solution at a concentration of 20 mg mL −1 on silicon wafers prefunctionalized with triethoxy benzophenone silane (3EBP-silane), which has been synthesized as described in the literature. [45] The wafers were prefunctionalized as described in the literature. [46] The resulting polymer layers were put on a 120 °C hot plate for 5 min to delete the processing history of the sample. Afterward, the samples were heated at 90 °C for 30 min and stored overnight at room temperature.

Study Design
The aim of this work was to obtain bioinspired block copolymers with an aliphatic PPDL block and an aromatic P2HEB block via ROP (Scheme 1). In a sequential approach (Scheme 1b), a PPDL macroinitiator was used to initiate the ROP of the aromatic block. In a modular approach (Scheme 1c,d), the two polymer components were obtained separately using functionalized initiators (carrying an azide or alkyne group, respectively) and combined via 1,3-dipolar azide-alkyne addition (click reaction). Films cast from these polymers were then analyzed for their morphology and nanomechanical properties.

Sequential Approach
The synthesis of the aliphatic PPDL by ROP was reported previously. [25] Thus, PPDL with an OH end group was obtained, which is a suitable macroinitiator for ROP. 2,3-Dihydro-5H-1,4benzodioxepin-5-one (2,3-DHB) was chosen as a monomer for the aromatic polymer block since this lactone could also be polymerized by ROP and thus potentially initiated by PPDL. The reaction solvent for the sequential approach was chloroform, which was the only solvent found in which PPDL could be dissolved over a broad molecular mass range, which also dissolved 2,3-DHB. Prior to the block copolymer synthesis, the homopolymerization Table 3. Reaction parameters for the copper-catalyzed azide-alkyne cycloaddition reaction of P2HEB-azide and PPDL-alkyne using Cu I (OAc) as a catalyst and TBD as a base. Reactions were performed for 2 days at ambient temperature in chloroform (1-2 mL). www.advancedsciencenews.com www.mcp-journal.de of 2,3-DHB in chloroform was optimized (Scheme 1a; initiator: benzyl alcohol, catalyst: TBD, Table 1). The polymers thus obtained were isolated by precipitation into methanol. In these reactions, monomer conversion strongly depended on the concentration of the reaction mixture. At a higher concentration of 2.9 g mL −1 2,3-DHB in chloroform, a high conversion of 79% could be reached. With a concentration of only 1.0 g mL −1 , only 50% of the monomer was converted into polymeric species. The polydispersity index was relatively low (1.13-1.35) and did not seem to depend on the monomer concentration. The 1 H NMR spectrum of the obtained poly(2,3-dihydro-5H-1,4benzodioxepin-5-one) (P2HEB, Figure S1, Supporting Information) matched the literature data, [42] and the peaks found in the FTIR spectrum ( Figure S2, Supporting Information) were also Scheme 1. Polymer Synthesis: a) Homopolymerization of 2,3-dihydro-5H-1,4-benzodioxepin-5-one (2,3-DHB) via ring-opening polymerization in chloroform (TCM), catalyzed by triazabicyclodecene (TBD). b) Sequential approach: 2,3-DHB was polymerized using poly(pentadecalactone) (PPDL) with an OH end group as ROP macroinitiator. c) Homopolymerization of poly(2-(2-hydroxyethoxy)benzoate (P2HEB) using azido propanol (AP) as initiator, yielding azide-functionalized P2HEB (P2HEB-azide). d) Modular approach: alkyne-functionalized PPDL group (PPDL-alkyne, obtained after literature procedures) [25] was reacted with P2HEB-azide via 1,3-dipolar addition of azide to alkyne ("click"-reaction) using a copper catalyst.
in agreement with the expected polymer structure. The glass transition temperature of the aromatic homopolymer polymer (determined by DSC; Figure S3, Supporting Information) was at observed 38 °C. No additional melting temperature was found, indicating that the polymer is amorphous.
Based on these results, polymerization of 2,3-DHB was initiated with a PPDL macroinitiator haring a M n of 2200-8900 g mol −1 , respectively (Scheme 1b and Table 2). The lower M n macroinitiator was used to enable more detailed end group analytics. Since lower 2,3-DHB concentrations had to be used in these reactions to avoid precipitation of the PPDL macroinitiator, an overall low M n of aromatic block was also expected. GPC characterization of the reaction products indicated that the molecular mass of the obtained block copolymer had indeed increased compared to its macroinitiator, thus confirming a successful initiation of the aromatic monomer by PPDL. As expected, the conversion remained low (40% and 17%, respectively, depending on the monomer concentration). To remove the unconverted 2,3-DHB monomer, the crude product was precipitated twice with an excess of methanol. Afterward, the resulting polymer was dried in vacuo at room temperature for 2 days. In Figure 2a, the proton NMR spectrum of the PPDL macroinitiator (M n = 2200 g mol −1 ) and the resulting block copolymer (M n = 9800 g mol −1 ) are shown. The signals of the aliphatic macro initiator match the ones found in literature, [25] with additional end group signals (δ = 7.36 and 5.12 ppm: benzyl end group; δ = 3.65 ppm: CH 2 next to the OH end group). The end-group signal at 3.65 ppm vanished after copoly merization (Figure 2b), indicating the loss of the OH end group, as expected. Additional new signals characteristic for poly(2,3dihydro-5H-1,4-benzodioxepin-5-one) (P2HEB) repeat units appeared. The signal intensity of the P2HEB chain end matched the one from the PPDL chain end, which also indicated a successful block copolymer synthesis. Importantly, no additional signals due to undesired transesterification between the two polyester blocks appeared. In the proton NMR spectrum of the block copolymer initiated with the higher M n PPDL initiator, which had an overall M n of 16 000 g mol −1 , the end group signals were below the detection limit (using standard measurement protocols). Figure 2c,d shows the GPC elugrams of the two block copolyesters, together with their respective PPDL macroinitiators. The lower M n copoly mer has a relatively sharp maximum, with a slight shoulder on the high molecular mass flank of the elugrams (Figure 2c). Notably, the low molecular mass peaks originating from the PPDL initiator vanished, and the peak maximum shifted toward higher molecular masses. In the case of the higher M n copolymer, the peak maximum only slightly shifted to higher molecular masses, potentially due to the low 2,3-DHB conversion (17%). However, the oligomeric peaks of the higher M n PPDL macroinitiator also vanished, which is a strong indication that these peaks initiated P2HEB chains. This is plausible, as shorter polymer chains statistically have a higher probability to find a reaction partner than longer chains, where the reactive chain ends may be buried inside the polymer coil. Thus, for both GPC elugrams, the shift to higher molecular masses is more strongly observed at the lower M n flank than at the higher M n flank. Overall, both GPC and NMR results indicate the successful block copolymer formation. FTIR spectra ( Figure S4, Supporting Information) further confirm this result. The DSC curves of the block copolyesters are shown in Figure S5 (Supporting Information). The smaller molecular mass copolymer does not feature a sharp melting point, but a broad melting area between 40 and 80 °C with several peak maxima, each corresponding to crystalline domains of PPDL with different sizes, or possible structural rearrangements during melting. This can be attributed to the low molecular mass and relatively broad polydispersity of the PPDL macroinitiator used to synthesize this copolymer, where the short oligomers hamper the formation of crystallites of similar sizes. The DSC curve of the other block copolymer, with a higher M n of each block ( Figure S5b, Supporting Information) had a single melting temperature at around 89 °C, which is characteristic for the PPDL block. A very small glass transition temperature step can also be seen at around 22 °C, which is lower than the T g of the aromatic P2HEB homopolymer. Thus, overall, the thermal properties of each homopolymer could be retrieved in the block copolymer.
In summary, the sequential approach to obtain all-polyester block copolymers from PPDL and P2HEB worked successfully, and had the big advantage that the target block copolymer could be obtained in only 2 steps. On the other hand, the poor solubility of higher molecular weight aliphatic homopolymer prohibited the polymerization of 2,3-DHB at higher concentrations, thus the M n of the second block stayed low. A higher molecular mass PPDL macroinitiator would require even more solvent in the reaction mixture. Thus, both the M n of the PPDL block and the P2HEB block that can be obtained via the sequential approach are limited.

Modular Approach
An alternative reaction path to obtain the target block copolymers is shown in Scheme 1c,d. In this approach, both homopolymers-functionalized with suitable reactive end groups-are synthesized and characterized separately, and then connected by a copper-catalyzed 1,3-dipolar azide-alkyne cycloaddition ("click reaction"). With this approach, the molecular mass limit accessible with the sequential approach has been overcome. PDL polymerization was initiated with propargyl alcohol as described previously. [25] This yielded PPDL with an alkyne end group (PPDL-alkyne, Scheme 1d). Aromatic lactone 2,3-DHB was initiated with AP, yielding P2HEB with an azide end group (P2HEB-azide, Scheme 1c). The reaction parameters for these polymerizations have been included in Table 1. Again, conversion depended on the concentration of the reaction mixture and varied between 54% and 87%. Using AP as initiator, P2HEBazide polymers with a M n from 750 to 24 000 g mol −1 were accessible, with the usual decrease of PDI at higher molecular mass. In the proton NMR spectrum of P2HEB-azide (Figure 3a), all characteristic signals expected for a P2HEB homopolymer were found, in addition to a set of new signals originating from the azide end group (triplet at δ = 3.38 ppm from the α-CH 2 group next to the azide, quintet at δ = 1.93 ppm from the neighboring β-CH 2 group). In the corresponding FTIR spectrum ( Figure S6, Supporting Information), all characteristic signals of a P2HEB homopolymer were present, in addition to a strong band at 2097 cm −1 originating from the NNN stretching vibration.
The thus obtained P2HEB-azide and PPDL-alkyne homopolymers were connected by copper-catalyzed 1,3-dipolar addition (click-reaction). To obtain block copolymers with different block ratios, but similar overall molecular masses, various combinations of P2HEB-azide and PPDL-alkyne were allowed to react with each other (Table 3). In each case, an excess of one of the blocks was used. The homopolymers were dissolved in chloroform in a Schlenk flask inside the glove box, where the appropriate amounts of Cu I OAc and triazabicyclodecene were added. The reaction mixture was stirred for 2 days at ambient temperature. To purify the block copolymers, the crude product was precipitated into an excess of ethanol. Further work-up depended on the reaction stoichiometry: Reactions conducted with an excess of the PPDL block were suspended in dichloromethane. In this solvent, unreacted PPDL did not dissolve and could be removed by filtration. Reaction mixtures containing an excess of the P2HEB block were dissolved in chloroform and precipitated into ice cold dichloromethane in order to remove unreacted P2HEB.
In Figure 3b, the proton NMR spectrum of PPDL 79 -triazole-P2HEB 21 , which is the block copolymer with the lowest M n of all the copolymers synthesized, is shown as a representative for the block copolymers obtained by the modular approach. For this reaction, a low molecular weight P2HEB-azide (M n = 750 g mol −1 , PDI = 1.67) was used, consisting mainly of oligomeric species. Those smaller chains gave NMR signals at slightly higher ppm values than the known polymeric P2HEB signals. Therefore, the characteristic signals of the P2HEB block seem to be comparatively broad (see Figure 3a). Characteristic signals from each polyester block are visible in the spectrum, indicating that both species must be present. The end group signals of both the azide chain end of P2HEB-azide and the alkyne signal of PPDL-alkyne vanished, including the signal at 3.38 ppm from the CH 2 group next to N 3 and the signals at 4.68 and 2.47 ppm from protons of the alkyne group of PPDLalkyne. Instead, new signals from the newly formed triazole group appeared (singlet at δ = 7.59 ppm, corresponding to the CH proton in the triazole ring; singlet at δ = 5.13 ppm from the α-CH 2 group of PPDL next to the triazole group; triplet at δ = 4.43 ppm from the α-CH 2 group P2HEB next to the triazole group; triplet at δ = 2.04 ppm from the neighboring β-CH 2 group). The intensity of these signals matched the intensity of the end group signals of both the aromatic and the aliphatic block. The latter also corresponded to each other. Overall, this is substantial evidence of successful block copolymer formation and purification. The absence of any further signals indicated that no undesired transesterification took place during the cycloaddition reaction. Due to their higher molecular masses, the other block copolymer spectra did not feature the end group signals. The molar ratio of the two blocks was therefore calculated from the ratio of their characteristic peaks.
The GPC curves of PPDL 79 -triazole-P2HEB 21 and PPDL 21triazole-P2HEB 79 are shown in Figure 3c,d, together with the corresponding functionalized homopolymers. An oligomeric P2HEB-azide was used to synthesize PPDL 79 -triazole-P2HEB 21 , which is clearly visible by the characteristic oligomer peak maxima at the lower M n flank of the GPC curve. The PPDLalkyne used had a higher molecular mass, and thus fewer oligomeric signals in the GPC curve. In the block copolymer elugram, the oligomer signals disappeared, and the left flank of the curve shifted to higher M n , as did the peak maximum. These data indicate successful block copolymer formation. The GPC curve of the block copolymer PPDL 21 -triazole-P2HEB 79 also showed a peak maximum at a lower elution volume than the homopolymers, which indicates a shift toward higher molecular weights. Here, the GPC curve of the block copoly mer also was clearly unimodal and did not show any shoulders or oligomeric peaks, which indicates both the successful block copolymer formation and the absence of unreacted homopoly mer. In summary, the GPC data further confirms the NMR findings. In the FTIR spectrum of a representative PPDLtriazole-P2HEB block copolymer ( Figure S6, Supporting Information), all characteristic signals for the two polyester blocks were found. Additionally, the characteristic signals of the alkyne group at 3307 cm −1 as well as the signal for the azide group at 2097 cm −1 vanished, and new signals corresponding to the triazole groups were observed at 1636 and 1463 cm −1 (CC and CH vibrations, overlapping with signals from the homopolymers). The DSC curve of PPDL 66 -triazole-P2HEB 34 with a molecular weight of 32 000 g mol −1 ( Figure S7, Supporting Information) had a weakly visible glass transition temperature at 36 °C, which is characteristic for the aromatic P2HEB block. For the PPDL block, a sharp and intensive melting temperature at 90 °C was observed. Overall, the characterization data of the polymers obtained by the sequential and modular approach are rather similar, indicating that both methods yielded the desired all-polyester block copolymers.

Polymer Film Formation and Characterization
The film properties of the above described block copolymers were investigated and compared to the two homopolymers, as well as a 1:1 homopolymer blend. Polymer films were formed via spin coating using a polymer solution with a concentration of 20 mg mL −1 . The target substrates were silicon wafers prefunctionalized with the hydrophobic 3EBP-silane, [47] which was used to match the hydrophobicity of the substrate and the polymers, and thereby prevent surface-induced dewetting. After film formation, the layers underwent a heat treatment (5 min on a 120 °C on hot plate to delete the thermal history of the sample, 30 min on a 90 °C hot plate to enable equilibration the polymer chain conformations).
Contact angle measurements confirmed the hydrophobic character of both homopolymers (Table S1, Supporting Information), where PPDL (advancing contact angle 96°± 2°) was slightly more hydrophobic than the aromatic P2HEB (advancing contact angle 88°± 2°). The block copolymers varied between those values, with a slight trend to higher contact angles with increasing PPDL content. The polymer film thickness (measured by ellipsometry, Table S1, Supporting Information) was around 100 nm for most of the polymers used. Pure P2HEB yielded layers with a higher thickness of 135 ± 2 nm. This could be due to its aromatic structure and amorphous character. In agreement with general spin-coating theory, there was a trend toward higher film thickness with increasing molecular mass for the PPDL-triazole-P2HEB copolymers.
The surface morphology of the thin films was analyzed by AFM using the peak-force tapping mode (height images in Figures 4 and 5, Inphase and Quadrature images in Figure  S8, Supporting Information). Previously recorded PPDL images, showing the crystalline morphology of that polymer, are included as a reference (Figure 4a). [25] P2HEB had an amorphous morphology, with a roughness of only 0.3 nm on an area of 1 µm² (Figure 4b). Thus, the homopolymer morphology is in agreement with the thermal properties. The 1:1 blend of PPDL and P2HEB (Figure 4c) showed the expected phase separation of poorly miscible polymers, with spherical, unstructured P2HEB domains having a broad size distribution (height: 20-550 nm, width: 0.2-4.0 µm) embedded into a structured PPDL matrix. The overall sample roughness was 144 nm. Compared to the blend, the block copolymer films had much finer morphological features and an overall roughness of only 3-5 nm on a 100 µm² sample area. This absence of macrophase separation further confirms the postulated block copolymer structure. While the pure PPDL sample had a distinct spherulite morphology, the block copolymers with a lower PPDL content (PPDL 21 -triazole-P2HEB 79, PPDL 52 -triazole-P2HEB 48 , and PPDL 57 -triazole-P2HEB 43 , Figure 5a-e) had a slightly fibrillar nanostructure but no spherulite microstructure. Apparently, the higher P2HEB content prevented the formation of more ordered crystalline domains. PPDL 21 -triazole-P2HEB 79 (Figure 5a) with the highest P2HEB content even has a number of evenly distributed vacancies that were up to 5 nm deep and 120 nm wide. Its roughness was higher than for a pure P2HEB surface, but still as low as 3.0 ± 0.1 nm. The roughness of PPDL 52 -triazole-P2HEB 48 (Figure 5b) on the other hand was 5.4 ± 0.1 nm, a value closer to the roughness of the pure PPDL sample (8.6 ± 0.1 nm). Overall, the morphology of PPDL 52triazole-P2HEB 48 looks like lamellar crystals sticking together without a higher level of order. Furthermore, up to 10 nm deep and 150 nm wide, evenly distributed vacancies were also found on this material. The typical lamellar structure often found for block copolymers with a 1:1 volume fraction ratio was not observed here, most likely due to the higher PDI of these polymers (not close enough to 1), and the overall too low molecular masses. PPDL 57 -triazole-P2HEB 43 also showed vacancies which were up to 9 nm deep and 140 nm wide. The surface structure of this polymer layer was very simi lar to that of PPDL 52 -triazole-P2HEB 48 , only showing slightly thinner lamellar crystals due to the slightly higher amount of PPDL. PPDL 66 -triazole-P2HEB 34 and PPDL 79 -triazole-P2HEB 21 (Figure 5d,e), the two copolymers with the highest PPDL content, showed spherulite structures which are characteristic for a crystalline polymer. In contrast to PPDL, the diameter of the spherulites was smaller for both block copolymers, and their lamellar crystals seemed finer. The PPDL 79 -triazole-P2HEB 21 spherulite shown in Figure 4g is 44 × 29 µm 2 in size, and the PPDL 66 -triazole-P2HEB 34 spherulite in Figure 4g has a size of 27 × 50 µm 2 . Spherulites formed from pure PPDL typically had a diameter > 250 µm and consisted of densely packed lamellae. [25] The roughness of the spherulite-forming copolymers shown here was 4.1 ± 0.1 and 4.8 ± 0.1 nm, respectively, and thus lower than that of pure PPDL. The vacancies found in these samples were 120 nm wide/9 nm deep, and 160 nm wide/7 nm deep. In summary, all block copolymers showed defined microstructures characteristic for each composition of the copolymer.
AFM was further used to quantify the nanomechanical properties of the polymer films obtained using the AFM-QNM mode. [48] These QNM-AFM pictures can be found in Figures S9-S12, Supporting Information). From these, the elastic modulus of the samples was calculated using the Derjaguin-Muller-Toporov (DMT) model. P2HEB had an overall DMT modulus of 7.6 ± 1.4 MPa, which proves the softness of this amorphous polymer. In contrast, crystalline PPDL had a stiffer surface (801 ± 42 MPa), as previously reported. [25] The three block copolymers studied by QNM had DTM modulus values between those of the two reference homopolymers: PPDL 66 -triazole-P2HEB 34 , which consisted of a majority of aliphatic PPDL, had a DMT modulus of 62.3 ± 5.7 MPa. The DMT modulus of PPDL 21 -triazole-P2HEB 79 , with only 21 mol% of the aliphatic block, was only 25.6 ± 2.9 MPa. Interestingly, the block copoly mer PPDL 52 -triazole-P2HEB 48 , with an almost equimolar ratio of both components, showed a DMT modulus of 17.2 ± 1.8 MPa, which is lower than that of PPDL 21 -triazole-P2HEB 79 . This demonstrates that the surface properties can strongly vary with the block ratio due to the specific morphology formed at each ratio, for example, by preferential segregation of one blocks to the air-polymer interface.

Conclusion
In this work, the bioinspired diblock copolyesters PPDL-ester-P2HEB and PPDL-triazole-P2HEB were synthesized-the former by sequential ring-opening polymerization, the latter by a modular approach in which functionalized homopolymers were connected via alkyne-azide 1,3-dipolar cycloaddition reaction. Importantly, in both cases, no signs of transesterification between the two ester blocks were observed under the given reaction conditions. Reasons for this were the mild ROP conditions for the synthesis of the P2HEB block in the sequential approach, and for the alkyne-azide addition in the modular approach, respectively. However, it was not studied if transesterification would occur upon heating to higher temperatures. However, previous work with PPDL shows that mild conditions alone are not sufficient to prevent transesterification. [25,37,49] Additionally, it seems that the benzene ring in α position to the carbonyl group of P2HEB renders a nucleophilic attack on this repeat unit more difficult. If this can be verified by further  work, it would be possible to conclude that all-polyester diblock copolymers without transesterification are available if at least one carbonyl group of the two blocks is sufficiently sterically hindered.
While the sequential approach only led to block copolymers with limited molar masses of the P2HEB block, and thus to limited block ratios, the modular approach gave access to the whole range of desired molar fractions of PPDL-triazole-P2HEB. This enabled a systematic study of changes in the poly mer film properties with composition for this sample series. These studies revealed a distinct morphology at each composition, and nanomechanical moduli in between the limiting values of the parent homopolymers. In future work, the barrier properties of these thin films will be evaluated.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.