Bio-based polycarbonate as synthetic toolbox

Completely bio-based poly(limonene carbonate) is a thermoplastic polymer, which can be synthesized by copolymerization of limonene oxide (derived from limonene, which is found in orange peel) and CO2. Poly(limonene carbonate) has one double bond per repeating unit that can be exploited for further chemical modifications. These chemical modifications allow the tuning of the properties of the aliphatic polycarbonate in nearly any direction. Here we show synthetic routes to demonstrate that poly(limonene carbonate) is the perfect green platform polymer, from which many functional materials can be derived. The relevant examples presented in this study are the transformation from an engineering thermoplastic into a rubber, addition of permanent antibacterial activity, hydrophilization and even pH-dependent water solubility of the polycarbonate. Finally, we show a synthetic route to yield the completely saturated counterpart that exhibits improved heat processability due to lower reactivity.

in N 2 atmosphere. The glass transition temperature is -as for PLimC -found at 130 °C and also the 5% decomposition temperature remains at 240 °C. Hence the hydrogenation has no influence on the thermal properties of this aliphatic polycarbonate.    a) estimates are given, as the number of colonies even for the 10-fold dilution was above 300.   The prolonged reaction times lead to cross-linking, which also influenced surface roughness of cast films and thus the contact angle (higher for increased roughness).

GPC analysis of PLimC-MAc
GPC data of the acid-functionalized PLimC sample (PLimC-MAc) is not given as it is not soluble in CHCl 3 and THF (only swells the polymer). The polymer is soluble in DMSO and DMF, though the sample elutes within the exclusion volume of the GPC column set due to the strong interaction with the eluent and weak interaction with the hydrophobic column packing material.

Curing procedure and tensile tests for PLimC rubber
The mechanical properties listed in Supplementary Table 2 were measured on samples that had been cured at 100 °C for 5 h in air. The curing procedure was applied in order to induce cross-linking of the residual double bonds in the polymer. After curing, the samples were assessed towards their solubility in chloroform and except for PLimC-B3MP 0.0 (no crosslinking is possible even when exposed to 100 °C for 24 h) none of them dissolved but only swelling was observed.
To elucidate the mechanism of heat-induced cross-linking, a solution of PLimC-B3MP 2.0 in chloroform was mixed with 1 wt% butylated hydroxytoluene (BHT, radical inhibitor) and cast into a film. After exposing the dry film to the same curing procedure (100 °C for 5 h) -in contrast to the BHT-free samples -no cross-linking was observable, indicating a free radical cross-linking mechanism between adjacent double bonds of the partially unsaturated samples.
The heat-induced cross-linking was proven indirectly by assessing the solubility and the mechanical properties of the cured samples. A direct analysis of residual double bonds by IR spectroscopy is not applicable because the low degree of unsaturation (< 3%) lies beyond the detection limit of the spectrometer. The effect of curing (time) is shown in Supplementary   Fig. 5 for PLimC-B3MP 2.0 . The curing temperature of 100 °C was chosen due to the trade-off between decomposition of the backbone of the rather labile polycarbonate (as discussed in our previous article on PLimC) 3 and the rate of thermally induced cross-linking (the higher the faster). On the one hand, a curing time of 5 h was found to be sufficient, since longer curing would not change the tensile properties any further. On the other hand, a curing time of 5 h was necessary, since samples heated for only 3 h were still soluble in chloroform. As an initial statement regarding the shelf life of PLimC-B3MP 2.0 , we can state that a 100 µm film exposed to ambient conditions (21 °C, air with 50% relative humidity, laboratory lighting) for 20 days is still soluble in chloroform.
The mechanical properties of the cured samples of PLimC-B3MP are shown in Supplementary Fig. 6. As expected, the introduction of cross-links leads to an increase of Young's modulus and tensile strength while the strain at break is reduced. With increasing cross-linking density, i.e. higher amount of double bonds in the uncured sample, the strain at break is decreasing while the Young's modulus is increasing. The tensile strength σ s is running through a maximum for low cross-linking density (PLimC-B3MP 0.5 ) but remains higher than in the saturated sample (PLimC-B3MP 0.0 ). Furthermore, a study of elasticity is shown in Supplementary Fig. 7, where a cured PLimC-B3MP 2.0 sample was strained by 20% in 40 cycles (5 s strain, 10 s relaxation, strain rate 10 mm min -1 ). The strain is not completely reversible (maximum stress drops from 0.08 MPa for each consecutive cycle down to 0.06 MPa for the last cycle), as the T g (5 °C) of the polymer is very close to the testing temperature of 21 °C, i.e. the slow dynamics can be assigned to the slow segmental motion of the polymer backbone. To improve reversibility in the strain experiment, the modification of PLimC with thiols that lower the T g well below 0 °C should be used.

Degradation tests in composting environment
The PLLA samples readily disintegrated within the first two weeks (holes appeared after 8 days), whereas PLimC samples did not show any traces of degradation even after 60 °C days under the conditions mentioned, i.e. no holes, no surface changes overserved by SEM imaging, no change of molecular weight (distribution) measured by GPC. From those observations we concluded, that PLimC possesses a rather good bio-stability against the industrial composting environment.

Degradation tests in enzymatic environment
The enzyme (13 000 units mL -1 , substrate: glyceryl tributyrate) was chosen because of its high activity in the cleavage of ester linkages of condensed matter like water-insoluble polyesters that have been synthesized and readily degraded with this enzyme in our group. 4 The change of hydrophilicity and T g do not promote mass loss of the polymer samples PLimC, PLimC-ME7/46/82 or BPA-PC respectively, for the conditions tested (see Supplementary Table 8).
Furthermore, for none of the samples a significant change in molar mass was observed (see GPC data in Supplementary Fig. 22), from which was deduced that the investigated polymers are stable under those conditions within 21 days. For further studies either the testing time should be increased or harsher conditions have to be applied. BPA-PC was employed as reference material and so far it can be stated, that PLimC and its modifications exhibit similar stability under those conditions.

The saturation of PLimC
The conversion of the monomer MenO and pre-monomers menth-1-ene and MenBrOH, respectively, were monitored by GC analysis. The chromatograms combined with the peak information are shown in Supplementary Figs. 23-25, respectively. The GC analysis of the precipitation bath of PMenC ( Supplementary Fig. 26) is added to prove the preferential incorporation of trans-MenO into the polymer chain. The accumulation of cis-MenO after polymerization is obvious, rising from 8% before to 56% after copolymerization with CO 2 .
The preferential incorporation is also represented in the NMR spectra of PMenC. The 1 H-NMR spectrum ( Supplementary Fig. 27) shows a single peak at 5.00 ppm without any downfield shoulder, which would be an indication of incorporation of the cis-isomer into the backbone. This argument is further supported by the 13 C-NMR spectrum of PMenC, which shows only one carbonyl resonance at 152.2 ppm i.e. no stereo-irregularities are present. GPC analysis revealed a relative M n of 61.3 kDa and Ð of 1.14 ( Supplementary Fig. 28).

Thiol-ene chemistry on PLimC
PLimC was dissolved in degassed chloroform to produce a 2 wt% solution. After addition of 5 to 40 eq. of the desired thiol, 0.3 eq of AIBN were added. The solution was kept at 60 °C for the desired time, before the solution was concentrated and precipitated in an adequate nonsolvent, corresponding to the functionalization, washed and reprecipitated when necessary.
The resulting colorless samples were dried at 60 °C in vacuo (except for butyl 3-mercaptopropionate functionalized PLimC that was dried at 20 °C and 0.02 mbar and stored in argon atmosphere).

PLimC-MAc: Functionalization with mercaptoacetic acid:
The solvent was removed in vacuo and the polymer redissolved in acetone before it was precipitated in water several times. The product was functionalized with 100% mercaptoethanol. 1
Synthesis of trans-menth-1-ene oxide. The procedure for the stereoselective epoxidation of menth-1-ene is analogue to the stereoselective epoxidation of (R)-limonene described elsewere. 3,7 The epoxidation of menth-1-ene has also been subject in patent literature. 8 The Masking of hydroxyl impurities in menth-1-ene oxide. Hydroxyl-containing impurities were masked according to a procedure previously described. 3 The product was purified by vacuum distillation. The purified product used for polymerization consisted of 0.9 % cismenthane, 1.0 % trans-menthane, 9.0% cis-MenO and 88.9 % trans-MenO.

Synthesis of poly(menthene carbonate).
The polymerization was carried out in accordance to the copolymerization of LO and CO 2 that was described elsewhere. 3 The product was characterized by 1 H and 13 C NMR spectroscopy. The precipitation bath was concentrated in vacuo and the residue analyzed by GC, whereas the chromatogram revealed an accumulation of cis-MenO, supporting the expectation that only the trans-isomer would be incorporated into the polymer.