Entropically-driven ring-opening polymerization of macrocyclic esters with up to 84-membered rings catalysed by polymer-supported Candida antarctica lipase B

Abstract

Macrocyclic esters with 24- to 84-ring atoms can be polymerized successfully using PS Candida antarctica lipase B as the catalyst. This method allows the polymerization of macrocycles containing substantial in-chain functional moieties, such as monomers incorporating steroid residues. It also affords metal-free products.

---

Introduction

In recent years there has been considerable interest in the lipase-catalyzed ring-opening polymerization (ROP) of lactones because they can afford polymers of substantial molecular weight in good yields and under mild reaction conditions.¹ The substrates are usually strained rings such as β-butyrolactone,² ε-caprolactone,^3,4 and D,L-2,6-dimethyl-1,4-dioxane-2,5-dione,⁵ and the polymerizations are driven by the relief of the ring strain. In a few cases, however, such as the polymerizations of cyclopentadecanolactone (16-ring atoms),^4,6 and the cyclic dimers of butylene succinate and adipate (20- and 24-ring atoms, respectively),^7,8 the lactones are strainless. These latter polymerizations are actually examples of a different type of ring-opening polymerization, namely entropically-driven ROPs (ED-ROPs).⁹ These exploit classical ring-chain equilibria, i.e. the equilibria that can exist, in the presence of a suitable catalyst, between a homologous family of macrocyclic oligomers and the corresponding condensation polymer, see Scheme 1.¹⁰ In a typical ED-ROP one macrocycle, or one or more members of a homologous family of oligomeric macrocycles, is equilibrated at high concentration, typically >15% w/v. Under these conditions the equilibrium shifts to give ca. 98% polymer and ca. 2% macrocycles. Polymerization is favoured because at high concentrations the macrocycles have relatively little translational entropy and the rings have only limited conformations: conformational flexibility increases greatly upon conversion into polymer. Since generally the macrocycles are more soluble than the polymer, they can be removed from the equilibrated mixture simply by precipitating the polymer into an appropriate solvent.

Scheme 1 Scheme showing ring–chain equilibria and the dependence on concentration (ED-ROP = entropically-driven ring-opening polymerization and CDP = cyclo-depolymerization).

A major attractive feature of ED-ROPs is that large strainless cyclics to be polymerized.⁹ The main aim of the present work was to extend the range of ED-ROPs achieved with polymer-supported (PS) Candida antarctica lipase B, the most commonly used lipase for ROPs,¹ to the polymerization of lactones with substantially more than 24-ring atoms. This is important because it would allow the macrocycle to contain large functional moieties and so would significantly increase the scope of enzyme-catalyzed polymerizations. Using the PS enzyme would not necessarily be successful, as large cyclics might not bind easily at the active site of the enzyme.¹¹ For example, the part of the ring opposite to the ester group might sterically inhibit binding to the active site, so that macrocycles might react only slowly or not at all. In this article we show that lactones with up to 84-membered rings can be polymerized successfully using PS C. antarctica. The examples include the synthesis of polyesters containing bile acid moieties, polymers that are expected to be biodegradable. The polymerizations are environmentally friendly, for example, the products are free of potential contaminants that might be present if traditional catalysts are used for the ED-ROPs such as tin^12,13 or ruthenium¹⁴ compounds.

Results and discussion

Initially three readily available mixtures of homologous cyclic oligomers, each containing a substantial fraction by weight of macrocycles with >24-ring atoms, were studied. In each case an 18% w/v solution of the mixture in toluene at 70 °C was treated with PS C. antarctica lipase B (7.5% of the weight of the substrate) for 20 h. At the end of the reaction period the polymer beads were filtered off, most of the solvent removed from the filtrate under vacuum and the concentrate added to cold methanol. This precipitated the polymeric product. It was collected, dried and characterized by ¹H NMR and FT-IR spectroscopies and by size exclusion chromatography (SEC). The results are summarized in Table 1.

Table 1 Entropically-driven ring-opening polymerizations of macrocycles catalyzed by PS C. antarctica^a

Entry	Monomer	Ring atoms in monomer	Concentration(%w/v)	Polymeric product	Yield^b (%)	Molecular weights^c		PDI^d
						M n × 10^−3	M w × 10^−3
a Unless indicated otherwise reactions were carried out for 20 h at the indicated concentration of substrate in toluene at 70 °C with 7.5% w/w of PS C. antarctica. b The final reaction mixture was filtered, concentrated, and then precipitated into methanol. Yields quoted are of dried precipitated polymer. c Determined by SEC relative to polystyrene standards. d PDI = polydispersity index = Mw/Mn. e After a reaction time of 4 h, by SEC the yield of polymer was 98% and it had Mn 42 800 and Mw 81 300. f By DSC the product had Tm 74 °C. g After a reaction time of 6 h, by SEC the yield of polymer was 95% and it had Mn 38 500 and Mw 64 300. h By DSC the product had Tm 65 °C. i Lower concentration was used so as to give sufficient solvent to allow efficient stirring. j After a reaction time of 4 h, by SEC the yield of polymer was 95% and it had Mn 12 200 and Mw 28 200. k By DSC the product had Tm 74 °C. l n.d. = no polymer detected by SEC. m By DSC the product had Tg 33 °C. n By DSC the product had Tg 20 °C. o Mole ratio 50 : 50. p The product had Tg < 18 °C and Tm 65 °C and by ^1H NMR spectroscopy monomers 16 and 1a were incorporated in the mole ratio 47 : 53.
1	1	12 to >72	18	2 ^,	93	63.4	120.5	1.9
2	3	13 to >65	18	4	97	55.0	110.0	2.0
3	5	28 to >84	18	7	95	23.5	44.6	1.9
4	1a	12	118	2	93	49.0	97.7	2.0
5	1b	24	18	2	99	59.4	93.4	1.6
6	1c	36	18	2	98	16.8	28.6	1.7
7	1d	48	13^i	2	97	16.6	31.5	1.9
8	1e	60	6^i	2	98	13.0	26.2	2.0
9	9	38	23	10 ^,	91	17.9	32.2	1.8
10	11	28	7^i	14	n.d.^l	—	—	—
11	12	42	7^i	14	n.d.^l	—	—	—
12	15	29	22	17	92	25.4	49.5	1.9
13	16	35	22	18	88	18.2	32.6	1.8
14	16 + 1a^o	—	48	22	87	9.1	14.5	1.6

---

The first mixture studied consisted of cyclic oligoundecanoates 1: see Chart 1. It was easily prepared by the cyclo-depolymerization (CDP, see Scheme 1) of polyundecanoate 2.¹² By SEC it contained cyclic monomer 1a, 1%; cyclic dimer 1b, 58%; cyclic trimer 1c, 22%; cyclic tetramer 1d, 9%; cyclic pentamer 1e, 6%; with cyclic hexamer and higher 4%. When subjected to the standard polymerization conditions polymer 2 was formed. It had, relative to polystyrene standards, M_n 63 400 and M_w 120 500: see Table 1, entry 1. The former corresponds to an average degree of polymerization of >300. Given the composition of the feedstock, the high yield (93%) implies that at least the cyclic dimer 1b (24-membered ring) and cyclic trimer 1c (36-membered ring) underwent ED-ROP successfully (Chart 1).

Chart 1

The second mixture consisted of cyclic oligododecanoates 3 prepared by a PS cyclo-oligomerization.¹³ The mixture was polymerized successfully using PS C. antarctica under the standard conditions: see Table 1, entry 2. Since the yield of polymer 4 was 97% and the composition of the starting mixture 3 was cyclic monomer 3a, 5%; cyclic dimer 3b, 71%; cyclic trimer 3c, 15%; cyclic tetramer 3d, 4%; cyclic pentamer 3e and higher oligomers, 5%, it is evident that at least the cyclic dimer 3b (26-membered ring) and cyclic trimer 3c (39-membered ring) polymerized successfully.

The final mixture studied consisted of macrocycles 5. This was obtained by hydrogenating a mixture of macrocycles 6, available from a previous project.¹⁵ The mixture had the composition: cyclic monomer 5a, 42%; cyclic dimer 5b, 20%; cyclic trimer 5c, 14%; and higher cyclics 24%. When treated with PS C. antarctica under the usual reaction conditions polymer 7 was formed in 95%: see Table 1, entry 3. It is thus evident that the monomer 5a (28-membered ring), dimer 5b (56-membered ring) and trimer 5c (84-membered ring) all polymerized successfully.

It is clear from these simple initial experiments that 24-, 26-, 28-, 36-, 39-, 56- and 84-membered macrocycles undergo ED-ROP satisfactorily on treatment with PS C. antarctica. A reaction time of 20 h is relatively long for an ROP catalyzed by a lipase, for example, cyclopentanoloactone typically needs only 2–4 h for a high yield of polymer.⁴ However, we found that whilst reaction times of 4–6 h were sufficient to afford good yields of polymer, reaction times of 20 h resulted in significantly higher molecular weights: compare Table 1 entries 1, 2 and 9 with footnotes e, g and j.

Several experiments were then carried out using pure individual cyclics. Monomers 1b–1e were obtained in a pure form by careful flash column chromatography of mixture 1. They were fully characterized by SEC, ¹H NMR spectroscopy and elemental analysis and/or mass spectrometry. Monomer 1a was obtained commercially. Cyclics 1a–1e were treated separately with PS C. antarctica under the usual reaction conditions, except that substrate concentrations sometimes differed. For instance, due to the small amounts of 1d and 1e available they were lower than usual. In all cases polymerizations occurred smoothly in high yields to give polymer 2: see Table 1, entries 4–8. All the products had significant molecular weights. Thus, not only the oligomers with 24- and 36-ring atoms polymerize successfully, but also those with 12-, 48- and 60-ring atoms.

Pure macrocycle 8 was available from a previous project.¹⁵ Hydrogenation gave macrocycle 9. The latter was treated with PS C. antarctica using the usual reaction conditions: see Table 1, entry 9. Polymerization occurred satisfactorily to give polymer 10. Thus, 38-membered lactone 9 underwent polymerization successfully.

Chart 2

As noted earlier, ED-ROP provides an opportunity to polymerize macrocycles that contain substantial functional moieties. In order to test this with PS C. antarctica, the cyclic dimer 11 (see Chart 2) and cyclic trimer 12 of lithocholic acid 13 were prepared.¹⁶ Each was treated with the supported enzyme under the usual reaction conditions: Table 1, entries 10 and 11. Surprisingly no polymer 14 was detected by SEC in either case. This suggests that here, as in some other transesterifications catalyzed by lipases,¹⁷ esters of secondary-alcohols are significantly less reactive than those of primary-alcohols.

Following this failure, lithocholic acid derivatives 15 and 16 were prepared. These incorporate spacers and more reactive ester groups. In the former case lithocholic acid 13 was esterified with pent-4-en-1-ol to give the ester and this was then treated with undec-10-enoyl chloride and triethylamine to give the diester. Ring-closing metathesis of the latter using Grubbs' first-generation catalyst gave monomer 15. Monomer 16 was prepared analogously, as described in the literature.¹⁴ These two monomers were treated separately with PS C. antarctica in the usual manner. Polymers 17 and 18 were obtained in high yields and with substantial molecular weights: see Table 1, entries 12 and 13. Thus, these 29- and 35-membered rings polymerize satisfactorily. These polymers contain lithocholic acid 11 plus hydroxy-acid units 19 and 20, respectively, the sequence depending on the relative reactivities of the ester groups A and B in formulae 15 and 16. It is expected that ester group B will be the more reactive because it is the ester of a primary-alcohol. Unfortunately, the ¹H NMR and ¹³C NMR spectra of the polymers do not allow conclusions to be drawn on such details of their structures. Cyclic 16 has previously been subjected to ED-ROP via olefin metathesis.¹⁴ It should be noted that this product, polymer 21, is different from polymer 18. Thus, it can have repeat units in head–head, head–tail and tail–tail arrangements. Consistent with this, polymers 18 and 21 have different T_gs, viz. 20 °C and 15 °C, respectively.

Polymer 21 is expected to biodegrade to products that are compatible with many biological systems.^14,18 Moreover, it has been shown to have a rubber-like elasticity with elongation moduli that are similar to those of certain tissues and elastin.¹⁴ Polymers 17 and 18 may well have similar properties. If, in an application, it is desirable to raise the T_g, it should be noted that polymer 17, with the shorter spacers, has a higher T_g (33 °C) than polymer 18 (20 °C). If it is necessary to lower the T_g, this might be achieved by copolymerization. To demonstrate this equimolar amounts of monomers 16 and 1a were copolymerized using PS C. antarctica under the usual conditions but with a longer reaction time to allow a random copolymer to form.¹⁹ Copolymer 22 was formed in high yield: see Table 1, entry 14. By ¹H NMR spectroscopy the monomers 16 and 1a were incorporated in the molar ratio 47 : 53 respectively. The T_g was now <15 °C. Unfortunately, it was not possible to confirm by the ¹H and ¹³C NMR spectroscopy whether a random copolymer formed.¹⁹

Conclusions

It has been shown here that macrocyclic esters with 24- to 84-ring atoms can be polymerized successfully using PS C. antarctica lipase B. The molecular weights obtained suggest that larger substrates tend to react more slowly. Examples of what are expected to be biodegradable polymers with T_gs up to 33 °C have been obtained. These polymerizations are particularly attractive in a biological context because they can be achieved easily (i) on a small scale, (ii) at moderate temperatures, (iii) without the elimination of any small molecules (and hence without the emission of any volatiles), (iv) with little or no generation of heat, and (v) catalyzed by an enzyme that is also biodegradable.

Experimental

Materials and methods

DCM = dichloromethane; THF = tetrahydrofuran.

Unless stated otherwise, starting materials were purchased from the Aldrich Chemical Company or Lancaster Synthesis Ltd and were used without further purification. Organic solutions were dried using magnesium sulfate. C. antarctica lipase B (CALB), physically immobilized on Lewatit OC VOC 1600, macroporous cross-linked poly(methyl methacrylate) resin beads, had specified activity ≥10 000 units per gram. It was purchased from Sigma. It was stored in a vacuum desiccator over P₂O₅ and was kept there for at least 24 h before use.

Mps were determined using an electrothermal apparatus and are uncorrected. Unless indicated otherwise FT-IR spectra were measured using an ATI Mattson Genesis series FT-IR spectrometer for samples cast as thin films from chloroform solutions onto NaCl plates. ¹H or ¹³C NMR spectra were recorded on a Varian INOVA 300 MHz Athos instrument for sample solutions in CDCl₃. Other ¹H NMR spectra were recorded on a Varian Unity 500 MHz spectrometer for samples in CDCl₃ solutions. Chemical shifts are quoted in parts per million (ppm) downfield from tetramethylsilane. Signals are described as singlet (s), triplet (t) or multiplet (m). J values are given in MHz. The integrals and assignments given for the ¹H NMR spectra of mixtures of oligomers are given per repeat unit. Optical rotations were measured using an Optical Activity Ltd AA-100 Digital Polarimeter with a cell path length of 10 cm. The [α]_D values are given in 10⁻¹ deg cm² g⁻¹. Matrix assisted laser desorption ionization time of flight mass spectra (MALDI-ToF MS) were obtained using a Micromass ToF Spec 2E spectrometer equipped with a nitrogen laser operating at 337 nm with a 4 ns pulse width. The matrix employed was dithranol doped with sodium bromide. SEC analyses were carried out using one of two instruments. Analyses of polymers were carried out using an in-house-assembled instrument equipped with a Knauer 64 pump operating at a flow rate of 1 mL of THF (NB: naming solvent is later used to indicate which SEC instrument was used) per minute through a PL gel 30 cm 10 µ mixed-B (×2) and 500 Å (×1) three-column set followed by a Gilson 132 differential refractometer for detection. The system was calibrated using a series of polystyrene standards, each with a narrow dispersity. SEC analyses of oligomers were carried out using an in-house-assembled instrument equipped with a Waters M45 pump operating at a flow rate of 1 mL of chloroform per minute through a Waters Styragel HR1, three-column set followed by a Gilson 132 differential refractometer for detection. Differential scanning calorimetry measurements were carried out on a Seiko SSC/5200 machine operating at a heating rate of 10 °C min⁻¹ in a nitrogen atmosphere. T_g and T_m values were taken on second heating scans, and were taken as the midpoints of the transitions.

Synthesis of the mixture 1

(a) Polyundecanoate 2. Polyundecanoate 2 was prepared from 11-bromoundecanoic acid as described in the literature.²⁰ The FT-IR and ¹H NMR spectra were satisfactory. By SEC (THF) it had M_n 26 000 and M_w 43 000.

(b) CDP of polyundecanoate 2 to give mixture 1 using di-n-butyltin oxide as catalyst¹². The above polyundecanoate 2 (5.20 g, 28.3 mmol) and di-n-butyltin oxide (2% w/v, 0.55 mmol) were dissolved in chlorobenzene (250 mL) and the mixture heated under reflux. After 1 week the mixture was cooled and the solvent evaporated off under reduced pressure. This gave the crude product (5.10 g). It had δ_H: 3.98 (2H; t, J = 6.5; CH₂OCO), 2.22 (2H; t, J = 6.5; CH₂CO₂), 1.55 (4H; br m; CH₂CH₂OCO and CH₂CH₂CO₂) and 1.25 ppm (12H; br s; 6 × CH₂). By SEC (chloroform) analysis it consisted of cyclic monomer 1a, 1%; cyclic dimer 1b, 58%; cyclic trimer 1c, 22%; cyclic tetramer 1d, 9%; cyclic pentamer 1e, 6%; other higher cyclics, 4%. No polymer was detected.

Synthesis of the mixture 3¹³

(a) Preparation of polymer-supported 12-hydroxydodecanoic acid. Using the procedure given in the literature,¹³ 12-hydroxydodecanoic acid was attached to chloromethylated (2.4 mmol g⁻¹) 1% cross-linked gel type 200–400 mesh beads. The loading achieved was 1.23 mmol of acid per gram.

(b) Cyclo-oligomerization of polymer-supported 12-hydroxydodecanoic acid. The beads prepared in (a) were cyclo-oligomerized using di-n-butyltin oxide as the catalyst.¹³ The yield of cyclic oligomers was 70%. The mixture had δ_H: 3.98 (2H; t, J = 6.5; CH₂OCO), 2.22 (2H; t, J = 6.5; CH₂CO₂), 1.55 (4H; br m; CH₂CH₂OCO and CH₂CH₂CO₂) and 1.25 ppm (14H; br s; 7 × CH₂); and by SEC (chloroform) the composition of mixture 3 was cyclic monomer 3a, 5%; cyclic dimer 3b, 71%; cyclic trimer 3c, 15%; cyclic tetramer 3d, 4%; cyclic pentamer 3e and higher oligomers, 5%.

Synthesis of mixture 5

A mixture of cyclic oligomers 6 was available from a previous project.¹⁵ The mixture 6 (300 mg, 0.70 mmol) in degassed THF (60 mL) was hydrogenated in a Parr hydrogenation apparatus with 10% Pd on charcoal (90 mg) as the catalyst, hydrogen at 1 atm pressure and the temperature at 20 °C. The final reaction mixture was filtered through a bed of Celite™, the filtrate reduced to about 20 mL under vaccuo, and then poured into hexane (100 mL). The precipitated solid was filtered off, washed with hexane and dried to give mixture 5 as a white powder (yield 273 mg, 91%). It had mp 75–80 °C. ν_max/cm⁻¹ (KBr): 2924, 2852, 1735, 1462, 1176, 802 and 724; δ_H: (500 Hz) 4.08 (4H; t; 2 × CH₂OCO), 2.36 (4H; t; 2 × CH₂COO) and 1.55 ppm (40H; br m; 20 × CH₂). Elemental analysis: found C, 73.32; H, 11.50. Calc. for C₂₆H₄₈O₄: C, 73.58; H, 11.32%. By SEC (chloroform) the mixture had the composition: cyclic monomer 5a, 42%; cyclic dimer 5b, 20%; cyclic trimer 5c, 14%; and higher cyclics 24%.

Isolation of pure cyclic oligomers 1b–1e

A portion of mixture 1 (5.02 g), obtained as above, was subjected to flash chromatography over silica gel [elution gradient: 35 (hexane) : 1 (ethyl acetate); then 30 (hexane) : 1 (ethyl acetate); then ethyl acetate]. This gave two main fractions: cyclic dimer 1b (1.41 g, 28%) and larger cyclics (2.02 g, 40%). The latter fraction (630 mg) was subjected to further flash chromatography over silica gel [elution gradient: 35 (hexane) : 2 (ethyl acetate); then 10 (hexane) : 1 (ethyl acetate)]. Overall this gave the following.

Dimer 1b. White powder (1.41 g), mp 72–73 °C (lit.,²¹ mp 73–74 °C); ν_max/cm⁻¹ 1735; SEC (CHCl₃) t_R = 28.8 min; δ_H: 4.07 (2H; t, J = 6.5; CH₂OCO), 2.34 (2H; t, J = 6.5; CH₂COO), 1.65 (4H; br m; CH₂CH₂OCO and CH₂CH₂CO₂) and 1.34 ppm (12H; br s; 12 × CH₂).

Trimer 1c. Colourless oil (180 mg); ν_max/cm⁻¹ 1737; SEC (CHCl₃) t_R = 27.4 min; δ_H: 4.04 (2H; t, J = 6.5; CH₂OCO), 2.31 (2H; t, J = 6.5; CH₂COO), 1.65 (4H; br m; CH₂CH₂OCO and CH₂CH₂CO₂) and 1.34 ppm (12H; br s; 18 × CH₂); ESI-MS (+): 575.6 [M + Na]⁺, (C₁₁H₂₀O₂)₃ requires 575.

Tetramer 1d. White powder (70 mg), mp 77–78 °C; ν_max/cm⁻¹ 1734; SEC (CHCl₃) t_R = 26.9 min; δ_H: 4.04 (2H; t, J = 6.5; CH₂OCO), 2.31 (2H; t, J = 6.5; CH₂COO), 1.65 (4H; br m; CH₂CH₂OCO and CH₂CH₂CO₂) and 1.34 ppm (12H; br s; 24 × CH₂); ESI-MS (+): 759.7 [M + Na]⁺, (C₁₁H₂₀O₂)₄ requires 759.

Pentamer 1e. White powder (50 mg), mp 50–52 °C; ν_max/cm⁻¹ 1735; SEC (CHCl₃) t_R = 23.3 min. δ_H: 4.04 (2H; t, J = 6.5; CH₂OCO), 2.31 (2H; t, J = 6.5; CH₂COO), 1.65 (4H; br s; CH₂CH₂OCO and CH₂CH₂CO₂) and 1.34 ppm (12H; br s; 30 × CH₂); ESI-MS (+): 943.8 [M + Na]⁺, (C₁₁H₂₀O₂)₅ requires 943.

Preparation of macrocycle 9

Macrocycle 8 was available from an earlier project.¹⁵ Macrocycle 8 (300 mg, 0.53 mmol) in degassed THF (60 mL) was hydrogenated using the same procedure as described in the preparation of mixture 5. This gave macrocycle 9 as a white powder (284 mg, 95% yield). It had mp 95–97 °C; ν_max/cm⁻¹ (KBr): 2919, 2857, 1737, 1472, 1267, 806 and 719; δ_H: (500 Hz) 4.09 (4H; t; 2 × CH₂OCO), 2.36 (4H; t; 2 × CH₂COO) and 1.54 ppm (60H; m; 30 × CH₂); m/z (CI) 565 (100%), C₃₆H₆₈O₄ requires 565 g mol⁻¹ for ([M + H]⁺). Elemental analysis: found: C, 92.12; H, 14.39. Calc. for expected C, 92.38; H, 14.64%.

Preparation of macrocycles 11 and 12

Following the experimental procedure described by Li and Dias,²² 2-chlorobenzoyl chloride (240 mg, 1.37 mmol) was added to a solution of lithocholic acid 11 (400 mg, 1.06 mmol) and 4-N,N-dimethylaminopyridine (540 mg, 4.40 mmol) in anhydrous toluene (90 mL). After stirring at 110 °C for 32 h the mixture was concentrated under reduced pressure and the concentrate filtered and washed successively with 5% aqueous hydrochloric acid (50 mL), saturated sodium bicarbonate (50 mL) and water (100 mL), then dried. Removal of the solvent gave the crude product. When subjected to column chromatography over silica gel, the major fraction was a white powder (204 mg). This was recrystallized from ethyl acetate–chloroform. The product (155 mg, 39%) had mp 290–300 °C and by electrospray MS m/z 425.4 (100), 740.2 (C2 + Na)⁺ and 1098.1 (C3 + Na)⁺. By SEC (CHCl₃) the mixture consisted of the cyclic dimer 11, 30%, and the cyclic trimer 12, 70%. By careful flash column chromatography pure samples of each component were obtained.

Dimer 11 had mp 292–295 °C (lit.,²² 298–300 °C), ν_max/cm⁻¹ 1719; δ_C: 175.26, 74.00, 56.51, 52.61, 42.63, 41.84, 40.30, 39.67, 35.99, 35.18, 34.61, 34.40, 32.05, 39.11, 28.98, 28.01, 26.99, 26.65, 26.34, 26.11, 24.16, 23.30, 20.75, 18.37 and 11.94 ppm.

Trimer 13 had mp 235–237 °C (lit.,²² 238–240 °C), ν_max/cm⁻¹ 1721; δ_C: 173.90, 74.00, 56.50, 54.61, 42.62, 41.83, 40.38, 40.28, 35.78, 35.07, 34.58, 34.44 32.34, 30.47, 30.05, 28.05, 26.99, 26.65, 26.34, 24.16, 23.29, 20.75, 18.37 and 11.93 ppm. These ¹³C NMR shifts are very similar to those reported in the literature.²²

Preparation of monomer 15

(a) Preparation of ω-pentenyl lithocholate. A mixture of lithocholic acid (4.00 g, 10.62 mmol), pent-4-en-1-ol (10 mL) and concentrated sulfuric acid (0.02 mL) was stirred and heated at 85 °C under nitrogen overnight. The mixture was then cooled to room temperature and diluted with DCM (100 mL). The solution was washed with water (2 × 100 mL), then dried over magnesium sulfate. The solvent was evaporated off and the residue recrystallized from methanol. This gave the desired ester as a white powder (3.77 g, 80%). It had mp 175–178 °C, ν_max/cm⁻¹ 1735; δ_H: (500 Hz) 5.65 (1H; m; CH=C), 4.85 (2H; m; CH₂=C), 3.85 (2H; t; CH₂OCO), 3.45 (1H; m), 2.25 (2H; m; CH₂COO), 2.00–0.80 (31H; m), 0.77 (6H; s; CH₃), 0.73 (6H; s; CH₃) and 0.45 ppm (3H; s; CH₃), and MS (EI) (M + H⁺, 100%), C₂₉H₄₈O₃ requires 444. Elemental analysis: found: C, 78.01; H, 9.81. Expected: C, 78.31; H, 9.98%.

(b) Preparation of diester. The above product (2.64 g, 5.95 mmol), chloroform (anhydrous, 50 mL) and triethylamine (3 mL) were mixed together in a round-bottom flask fitted with a nitrogen inlet and a pressure equalizing dropping funnel. The mixture was cooled to 0 °C using an ice bath prior to the dropwise addition of undec-10-enoyl chloride (1.32 g, 6.95 mmol). A precipitate (triethylammonium salt) formed immediately. The mixture was stirred overnight at room temperature and then poured into dilute aqueous hydrochloric acid (0.1 M, 100 mL). The obtained solution was extracted with water, washed with a saturated aqueous potassium hydrogen carbonate solution, and then with saturated brine. The organic phase was dried over magnesium sulfate prior to filtration through a plug of Celite™. Evaporation of the filtrate gave the crude product. Recrystallization of this from hexane gave the required diester as a white powder (3.04 g, 84%). It had mp 62–64 °C; δ_H: (500 Hz) 5.75 (2H; m; CH=C), 4.95 (4H; m; C=CH₂), 4.00 (2H; t; CH₂OCO), 2.35–2.15 (4H; m; CH₂COO), 2.00–0.93 (34H; m), 0.90–0.80 (6H; 2 × s; CH₃) and 0.60 ppm (3H; s; CH₃); MS (electrospray) 611 (M + H⁺, 100%), C₄₀H₆₆O₄ requires 610. Elemental analysis: found C, 78.39; H, 10.70. Expected: C, 78.62; H, 10.89%.

(c) Ring-closing metathesis of the diester to give monomer 15. The above diester (1.87 g, 3.08 mmol) in dry DCM (300 mL) at 40 °C was treated with Grubbs' ‘first-generation catalyst’ (127 mg, 0.154 mmol). The progress of the reaction was monitored by ¹H NMR spectroscopy. After three days the reaction was stopped, the solvent evaporated off and the residue subjected to column chromatography (silica gel, petroleum ether (b.p. 40–60 °C)–ethyl acetate, 90 : 10). Monomer 15 was obtained as a white solid (1.11 g, 62%). Analysis by SEC (CHCl₃) showed only one peak and the MALDI-ToF (dithranol, NaBr) mass spectrum showed a major mass peak corresponding to the cyclic monomer: 618 (M₁ + Na)⁺. The solid had mp 132–134 °C; ν_max/cm⁻¹ 1740; δ_H: (500 Hz) 5.39 (2H; m; CH=C, cis and trans), 4.70 (1H; m), 4.07 (2H; m; CH₂OCO), 2.36–2.15 (4H; m; CH₂COO), 2.08–0.95 (34H; m), 0.91–0.84 (6H; 2 × s; 2 × CH₃) and 0.65 ppm (3H; s; CH₃).

Preparation of monomer 16

This monomer was prepared by a literature method that is analogous to that used to prepare monomer 15.¹⁴ It had mp 120–122 °C (lit.,¹⁴ 120 °C); ν_max/cm⁻¹ 1736; δ_H: (500 Hz) 5.38 (2H; m; CH=C, cis and trans), 4.72 (1H; m), 4.06 (2H; m; CH₂OCO), 2.37–2.12 (4H; m; CH₂COO), 2.08–0.96 (46H; m), 0.90–0.85 (6H; 2 × s; 2 × CH₃) and 0.62 ppm (3H; s; CH₃). Analysis by SEC (CHCl₃) showed only one peak, and the MALDI-ToF (dithranol, NaBr) mass spectrum of the product showed a mass peak at m/z 690 corresponding to (M₁ + Na)⁺.

ED-ROP of mixture 1 using PS C. antarctica

This experiment is summarized in Table 1, entry 1.

Polymer-supported C. antarctica lipase B (75 mg) was weighed into an oven-dried Pyrex tube and then dried in a vacuum oven in the presence of P₂O₅ in a separate vessel. A solution of mixture 1 (180 mg, 0.98 mmol) in anhydrous toluene (1.00 mL) was added into the reaction vial under nitrogen using a syringe. The mixture was heated at 70 °C with stirring. After 20 h the solution had become very viscous. The mixture was cooled and chloroform (100 mL) was added. The supported enzyme was immediately removed by filtration. Most of the chloroform was removed using a rotary evaporator. The concentrate (ca. 8 mL) was then added to methanol (80 mL). The white precipitate that formed was filtered off and dried in a vacuum oven. This gave polymer 2 as a white powdery material (176 mg, 98%). It had T_m 74 °C (lit.,^6b,12 83 °C); δ_H: (500 Hz) 3.98 (2H; t, J = 6.5; CH₂OCO), 2.24 (2H; t, J = 6.5; CH₂COO), 1.54 (4H; br m; CH₂CH₂OCO and CH₂CH₂CO₂) and 1.21 ppm (12H; br s; 6 × CH₂). SEC (THF) of the product gave M_n 63 400 and M_w 120 500.

ED-ROP of mixture 3 using PS C. antarctica

This experiment, summarized in Table 1, entry 2, was carried out using the experimental procedure described above for the polymerization of mixture 1. Polymer 4 was obtained in 97% yield. It had δ_H: (500 Hz) 3.98 (2H; t, J = 6.5; CH₂OCO), 2.24 (2H; t, J = 6.5; CH₂COO), 1.54 (4H; br m; CH₂CH₂OCO and CH₂CH₂CO₂) and 1.21 ppm (14H; br s; 7 × CH₂). SEC (THF) analysis of the product indicated M_n 55 000 and M_w 110 000.

ED-ROP of mixture 5 using PS C. antarctica

This experiment, summarized in Table 1, entry 3, was carried out using the procedure described above for the polymerization of mixture 1. Polymer 7 was obtained as a white solid in 95% yield. It had T_g 65 °C; δ_H: (500 Hz) 4.11 (4H; t, J = 6.5; CH₂OCO), 2.38–2.24 (4H; t, J = 6.5; CH₂COO), 1.54 (8H; br m; CH₂CH₂OCO and CH₂CH₂CO₂) and 1.76–1.24 ppm (32H; m; 16 × CH₂). SEC (THF) of the product indicated M_n 23 500 and M_w 44 600.

ED-ROP of 12-oxa-2-ketocyclododecane 1a using PS C. antarctica

This experiment, summarized in Table 1, entry 4, was carried out on a larger scale principally as a convenient way to obtain a sample of polymer 2 with hydroxyl and carboxylic acid end groups.

Polymer-supported C. antarctica (50 mg) was weighed into an oven-dried Pyrex tube and dried in a vacuum oven. A solution of 12-oxacyclododecan-2-one 1a (530 mg, 2.87 mmol) in anhydrous toluene (0.45 mL) was added under nitrogen into the reaction vessel via a syringe. The mixture was heated to 70 °C, with stirring, and kept at that temperature. After 6 h the solution had become extremely viscous. The reaction was continued for another 24 h to ensure completion. The cooled mixture was then dissolved in chloroform (20 mL) and the supported enzyme was immediately removed by filtration. The filtrate was concentrated using a rotatory evaporator and the concentrate (ca. 10 mL) added to methanol (60 mL). The white precipitate that formed was filtered off and dried in a vacuum oven. This gave polyundecanote 2 as a white powdery solid (493 mg; 93%). It had δ_H: (500 MHz) 3.97 (2H; t, J = 6.5; CH₂OCO), 2.22 (2H; t, J = 6.5; CH₂COO), 1.55 (4H; br m; CH₂CH₂OCO and CH₂CH₂CO₂) and 1.22 ppm (12H; br s; 6 × CH₂). SEC (THF) of the product indicated M_n 49 000 and M_w 97 700.

ED-ROP of monomers 1b–1e using PS C. antarctica

These experiments, summarized in Table 1, entries 5–8, were carried out using the procedure described above for the polymerization of mixture 1. Polymer 2 was obtained in the yields and with the molecular weights given in the Table. For monomers 1d and 1e reduced concentrations of reaction mixtures were used on account of the limited amounts available of these monomers.

ED-ROP of monomer 9 using PS C. antarctica

The experiment, summarized in Table 1, entry 9, was carried out using the procedure described above for the polymerization of mixture 1. Polymer 10 was obtained as a white powder in 91% yield. It had T_m 74 °C; δ_H: (500 Hz) 4.08 (4H; t, J = 6.5; CH₂OCO), 2.30 (4H; t, J = 6.5; CH₂COO), 1.56 (8H; br m; CH₂CH₂OCO and CH₂CH₂CO₂) and 1.76–1.24 ppm (52H; m; 26 × CH₂). SEC (THF) analysis of the product indicated M_n 17 900 and M_w 32 220.

ED-ROP of monomer 15 using PS C. antarctica

The experiment, summarized in Table 1, entry 12, was carried out using the procedure described above for the polymerization of mixture 1. Polymer 17 was recovered as a gum. It had T_g 33 °C; [α]_D²⁵ +65 (c = 1 in THF); ν_max/cm⁻¹ 1738; δ_H: (500 Hz) 5.39 (2H; m; CH=C, cis and trans), 4.72 (1H; m; 3b-H), 4.05 (2H; m; CH₂OCO), 2.36–2.11 (4H; m; CH₂COO), 2.07–0.96 (46H; m), 0.91–0.84 (6H; 2 × s; 2 × CH₃) and 0.63 ppm (3H; s; CH₃); δ_C: 174.69, 173.71, 128.70, 74.33, 64.68, 56.73, 56.31, 42.98, 42.16, 40.66, 40.39, 36.04, 35.60, 35.30, 35.02, 34.84, 32.86, 31.62, 31.29, 29.50, 28.89, 28.43, 27.27, 26.94, 26.59, 26.19, 25.31, 24.44, 23.58, 21.08, 18.51 and 12.28 ppm (32 signals observed, 38 required). SEC (THF) analysis of the product indicated M_n 25 400 and M_w 49 500.

ED-ROP of monomer 16 using PS C. antarctica

This experiment, summarized in Table 1, entry 13, was carried out using the procedure described above for the polymerization of mixture 1. Polymer 20 was obtained in 88% yield as a waxy solid. It had T_g 20 °C; [α]_D²⁵ +68 (c = 1 in THF); ν_max/cm⁻¹ 1736; δ_H: (500 Hz) 5.43–5.30 (2H; m; CH=, cis and trans), 4.80–4.68 (1H; m; 3b-H), 4.10–4.02 (2H; m; CH₂OCO), 2.39–2.12 (4H; m; CH₂COO), 2.06–1.02 (56H; m), 0.94–0.84 (6H; 2 × s; 2 × CH₃) and 0.66 ppm (3H; s; CH₃); δ_C: (500 Hz) 174.66, 1.73.68, 130.57, 74.33, 64.68, 56.73, 56.33, 42.99, 42.16, 40.66, 40.43, 36.05, 35.61, 35.30, 35.01, 34.85, 32.85, 32.56, 31.63, 31.29, 29.88, 29.75, 29.68, 29.56, 29.50, 29.38, 2, 23.59, 21.09, 28.91, 28.43, 27.29, 26.95, 26.59, 26.19, 25.32, 24.44, 18.51 and 12.28 ppm (39 observed, 44 required). SEC analysis of the product indicated M_n 18 200, M_w 32 600.

ED-ROP of a mixture of monomers 16 and 1a using PS C. antarctica

This experiment is summarized in Table 1, entry 14.

PS C. antarctica lipase B (20.00 mg) was weighed into a dried Pyrex tube. A solution of cyclic 16 (150 mg, 0.22 mmol) and cyclic 1a (40.50 mg, 0.22 mmol) in anhydrous toluene (0.40 mL) was added via a syringe. The mixture was heated at 70 °C with stirring for 7 days. The resultant viscous mixture was then diluted with chloroform (100 mL) and the enzyme removed by filtration. The filtrate was concentrated using a rotary evaporator. The concentrate was added dropwise to methanol (80 mL). The precipitated material was filtered off and dried in a vacuum oven at 20 °C. This gave copolymer 24 (165 mg, 87%) as a pale brown gum. It had T_g < 18 °C and T_m 65 °C; M_n 9100 and M_w 14 500. By ¹H NMR spectroscopy (500 Hz) the copolymer consisted of 47% lithocholic acid repeat units and 53% of undecanoate repeat units.

Acknowledgements

The authors would like to thank Dr Giudi Bartalucci and Dr Arantxa Rodriguez-Menendez for assistance with the synthesis of the steroidal substrates, and to acknowledge financial support from the EPSRC and PH Ltd.

References and notes

1. For reviews see: (a) S. Kobayashi, H. Uyama and S. Kimura, Chem. Rev., 2001, 101, 3793–3818; (b) R. A. Gross, A. Kumar and B. Kalra, Chem. Rev., 2001, 101, 2097–2124; (c) S. Matsumura, Adv. Polym. Sci., 2006, 194, 95–132; (d) I. K. Varma, A.-C. Albertsson, R. Rajkhowa and R. K. Srivastava, Prog. Polym. Sci., 2005, 30, 949–981.
2. (a) G. A. R. Nobes, R. J. Kazlauskas and R. H. Marchessault, Macromolecules, 1996, 29, 4829–4833; (b) S. Matsumura, Y. Suzuki, K. Tsukada, K. Toshima, Y. Doi and K. Kasuya, Macromolecules, 1998, 31, 6444–6449.
3. (a) A. Kumar and R. A. Gross, Biomacromolecules, 2000, 1, 133–138; (b) F. He, S. Li, H. Garreau, M. Vert and R. Zhuo, Polymer, 2005, 46, 12682–12688; (c) M. Hans, P. Gasteier, H. Keul and M. Moeller, Macromolecules, 2006, 39, 3184–3193; (d) K. J. Thurecht, A. Heise, M. deGeus, S. Villarroya, Z. Zhou, M. F. Wyatt and S. M. Howdle, Macromolecules, 2006, 39, 7967–7972; (e) B. Chen, E. M. Miller, L. Miller, J. K. Maikner and R. A. Gross, Langmuir, 2007, 23, 1381–1387.
4. A. Kumar, B. Kalra, A. Dekhterman and R. A. Gross, Macromolecules, 2000, 33, 6303–6309.
5. S. Matsumura, K. Mabuchi and K. Toshima, Macromol. Rapid Commun., 1997, 18, 477–482.
6. (a) K. S. Bisht, L. A. Henderson, R. A. Gross, D. L. Kaplan and G. Swift, Macromolecules, 1997, 30, 2705–2711; (b) L. van der Mee, F. Helmich, R. de Brijn, J. A. J. M. Vekemans, A. R. A. Palmans and E. W. Meijer, Macromolecules, 2006, 39, 5021–5027.
7. S. Sugihara, K. Toshima and S. Matsumura, Macromol. Rapid Commun., 2006, 27, 203–207.
8. S. Okajima, R. Kondo, K. Toshima and S. Matsumura, Biomacromolecules, 2003, 4, 1514–1519.
9. (a) P. Hodge and H. M. Colquhoun, Polym. Adv. Technol., 2005, 16, 84–94; (b) P. Hodge, React. Funct. Polym., 2001, 48, 15–23.
10. (a) H. Jacobson and W. H. Stockmayer, J. Chem. Phys., 1950, 18, 1600–1606; (b) J. A. Semlyen, Adv. Polym. Sci., 1976, 21, 41–75; (c) U. W. Suter, in Comprehensive Polymer Science, ed. G. Allen and J. C. Bevington, Pergamon Press, Oxford, 1989, vol. 5, pp. 91–96; (d) G. Ercolani, L. Mandolini, P. Mencarali and S. Roelens, J. Am. Chem. Soc., 1993, 115, 3901–3908.
11. J. Uppenberg, N. Öhrner, M. Norin, K. Hult, G. J. Kleywegt, S. Patkar, V. Waagen, T. Anthonsen and T. A. Jones, Biochemistry, 1995, 34, 16838–16851.
12. C. L. Ruddick, P. Hodge, Y. Zhuo, R. L. Beddoes and M. Helliwell, J. Mater. Chem., 1999, 9, 2399–2405.
13. C. L. Ruddick, P. Hodge, A. Cook and A. J. McRiner, J. Chem. Soc., Perkin Trans. 1, 2002, 629–637.
14. J. E. Gautrot and X. X. Zhu, Angew. Chem., Int. Ed., 2006, 45, 6872–687.
15. (a) P. Hodge and S. D. Kamau, Angew. Chem., Int. Ed., 2003, 42, 2412–2414; (b) S. D. Kamau, P. Hodge, A. J. Hall, S. Dad and A. Ben-Haida, Polymer, 2007, 48, 6808–6822.
16. M. Hikota, Y. Sakurai and K. Horita, Tetrahedron Lett., 1990, 31, 6367–6370.
17. H. Kikuchi, H. Uyama and S. Kobayashi, Polym. J. (Tokyo), 2002, 34, 835–840.
18. J. E. Gautrot and X. X. Zhu, J. Biomater. Sci., Polym. Ed., 2006, 17, 1123–1139.
19. S. D. Kamau, P. Hodge, R. T. Williams, P. Stagnaro and L. Conzatti, J. Comb. Chem., 2008, 10, 644–654.
20. P. Hodge, R. O'Dell, M. S. K. Lee and J. R. Ebdon, Polymer, 1996, 37, 1267–1271.
21. E. J. Corey and K. C. Nicolau, J. Am. Chem. Soc., 1974, 96, 5614–5616.
22. (a) Y. Li and J. R. Diaz, Synthesis, 1997, 425–430; (b) H. Gao and J. R. Dias, Croat. Chim. Acta, 1998, 71, 827–831.

---