Control of thermal properties and hydrolytic degradation in poly(lactic acid) polymer stars through control of isospecificity of polymer arms**

This is the pre-peer reviewed version of the following article:

Cameron, D. J. A., & Shaver, M. P. (2012). Control of thermal properties and hydrolytic degradation in poly(lactic acid) polymer stars through control of isospecificity of polymer arms. Journal of Polymer Science Part A - Polymer Chemistry, 50(8), 1477-1484.

which has been published in final form at http://dx.doi.org/10.1002/pola.25927

This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for self-archiving (http://olabout.wiley.com/WileyCDA/Section/id-817011.html).

Manuscript received: 15/09/2011; Accepted: 10.01/2012; Article published: 01/02/2012

Donald J. A. Cameron1 and Michael P. Shaver1,*

[1]Department of Chemistry, University of Prince Edward Island, 550 University Avenue, Charlottetown, PE, C1A 4P3, Canada.

[*]Corresponding author; (current address): Michael.Shaver@ed.ac.uk, EaStCHEM, School of Chemistry, Joseph Black Building, University of Edinburgh, West Mains Road, Edinburgh, EH9 3JJ, UK.

[**]The authors thank the Natural Sciences and Engineering Research Council of Canada, the Canada Foundation for Innovation and the Atlantic Canada Opportunities Agency for funding and Innovation PEI for a student fellowship to DC.

Synopsis:

Dipentaerythritol polymer stars with poly(lactic acid) arms of varied tacticity are generated using aluminum and tin catalysts. The stereospecificity of the arms controls the thermal properties of the materials and the length of crystalline domains thus influencing hydrolytic degradation.

Keywords:

Ring-opening polymerization; stereospecific polymers; biodegradable; drug delivery systems; isotactic; star polymers

Abstract

The synthesis of a family of polymer stars with arms of varied tacticities is discussed. The effect of polymer tacticity on the physical properties of these polymer stars is dramatic. Dipentaerythritol cores support six poly(lactic acid) arms. Lewis acidic tin and/or aluminum catalysts control the polymerization to afford polymer stars of variable tacticity. Analysis of these polymers by 1H NMR spectroscopy, thermogravimetric analysis, powder X-ray diffraction and differential scanning calorimetry reveals the effects of tacticity control on the physical properties of the polymer stars. Hydrolytic decomposition studies suggest that the degradation profile of a polymer star may also be tuned by stereochemical control. Differences between isotactic samples derived from rac-lactide and l-lactide are heightened by longer arms of 50 and 100 monomer units. Control of polymer isospecificity shows that a ~70% isotacticity bias is necessary to induce crystallinity and alter the thermal and degradation properties of the material. Above 70% isotacticity the degradation properties and thermal transitions can be further tuned across a relatively wide range. This technique allows for significant tunability to the physical properties of aliphatic polyester polymer stars.

Introduction

The development of biodegradable polymers derived from cyclic esters has particularly focused on poly(lactic acid).1 Fuelling the growth in PLA research is its decreased environmental footprint, as the polymerizable monomer 3,6-dimethyl-1,4-dioxan-2,5-dione (lactide or LA) is derived from lactic acid, a product that may be fermented from the sugars of various cultivable plants. As the lactide monomer possesses two chiral centers, several stereoregular forms are accessible with judicious choice of catalyst including atactic, heterotactic and isotactic polymers synthesized from a 1:1 mixture of d- and l-lactide (rac-lactide, rac-LA). While tin(II) 2-ethylhexanoate (Sn(Oct)2) is used industrially to catalyze the controlled ring-opening polymerization of lactide to give atactic PLA, catalyst design has led to increased activity, improved control over molecular weights and stereoregular polymers.2-4 Aluminum salen and salan catalysts have shown particular utility in synthesizing stereoregular polymers (Figure 1).5-6

While polymeric microstructures can have wide ranging effects on the macroscopic properties of a material, these studies have been predominantly limited to linear polymer chains.7 Star polymers, multi-armed polymeric materials in which arms radiate from a central core, are of particular interest as they possess rheological, mechanical and biomedical properties that are not accessible in traditional linear polymers.8 Although the polymer star architecture can be built with a wide range of monomers,9 much work has focused on aliphatic polyesters like PLA.8 These systems usually employ a core first approach and initiate the polymerization from a multi-functional alcohol. In general, PLA stars exhibit lower melting temperatures (Tm), glass transtion temperatures (Tg) and crystallization temperatures (Tc) than their linear counterparts. The effect of core structure has also been studied, especially the role of the number of arms, arm length and rigidity on polymer properties.8,10 Prior to our entry into this field, only one stereocontrolled star had been reported,11 where an aluminum salan complex was used to prepare a six-armed heterotactic PLA star which was unfortunately not thermally characterized. Polymer stars, while being a versatile and unique macromolecular scaffold with important applications, are plagued by non-optimal physical, thermal and solution properties.8 Our goal was to exploit catalyst design and tacticity control to tune the physical properties and hydrolytic degradation of these macromolecules.

Figure 1. Tacticity control by aluminum and tin complexes in the polymerization of rac-lactide.

We recently reported our preliminary study on the synthesis of a family of polymer stars with arms of varied tacticities, including the effect of polymer tacticity on the physical properties of the materials.12 The role of polymer tacticity and monomer feedstock on the properties of DPE-based PLA star polymers is significant.12 Stars with oligomeric arms were be prepared using Sn(Oct)2 and rac-LA to produce atactic stars, Sn(Oct)2 and l-LA to produce isotactic-l stars, Cl[salan]AlMe (Cl[salan] = N,N-ethylenebis(benzyl)bis(3,5-di-chlorosalicylamine)) and rac-LA to produce heterotactic stars and tBu[salen]AlMe (tBu[salen] = N,N-ethylenebis(3,5-di-tert-butylsalicylimine)) and rac-LA to produce isotactic-rac star polymers (Figure 2). Thermal stability increased dramatically (>50 °C) for all samples with stereoregularily, regardless of the specific stereoregular form. Thermal transitions were controlled by the relative isospecificity of the polymer samples. Tg values varied over 11 °C between atactic, heterotactic, isotactic(rac) and isotactic(l) samples, while Tm values for isotactic-biased samples showed significant differences between isopure stars and those with stereoerrors, also correlating to the presence of crystalline domains within the samples.12 Importantly, the relative isospecificity also changed the rate of hydrolytic degradation for these stars, suggesting a tunability that warranted further investigation. We now present an expansion of this work, further examining the oligomeric stars, detailing how the physical properties of the PLA stars change with longer star arms and show that the relative isospecificity of the arms can further tune the polymer properties.

Figure 2. Dipentaerythritol poly(lactic acid) polymer stars.

Results and Discussion

Building from these initial studies, we hoped to better understand the nature of these materials and the source of the resistance to hydrolytic degradation. While PLA stars have previously been investigated for resistance to hydrolytic degradation,8 a systematic investigation of the effect of stereocontrol on the degradation profiles of stereocontrolled stars has not been reported. We investigated the base-promoted hydrolysis of the ester linkages to form soluble oligolactide units and lactic acid, promoted by 1,5,7-triazabicyclo[4.4.0]dec-5-ene (TBD) in methanol (23 °C, pH = 12.8). These results strengthen previous research which shows that the crystallinity of linear PLA plays a key role in hydrolytic degradation rates.13-14 This is usually linked to higher stereospecificity, as linear isotactic polymers show resistance to hydrolytic degradation. For DPE-based polymer stars with oligomeric arm lengths, degradation varied significantly with tacticity bias (Figure 3). The amorphous nature of atactic and heterotactic samples improves TBD solution diffusion into the polymer matrix, which may increase COOH auto-catalyzed scission rates at the core of the polymer samples.13 The morphology of heterotactic PLA does not differ greatly from atactic samples, and while the heterotactic PLAs possessed a strong bias for the mrm and rmr tetrads and stronger thermal stability, these stereosequences did not afford the sample any great enhancement of stability against hydrolytic degradation. Long sequences of isotactic blocks produce crystalline domains into which solution diffusion may be difficult. These results indicate an increased susceptibility of racemic linkages to the scission mechanism and the importance of the crystalline domains in isotactic samples.

Figure 3. Polymer lifetimes of stereocontrolled samples 1 (atactic), 2 (heterotactic), 3 (isotactic-rac), 4 (isotactic-l). n = 4.

Small sample molecular weight differences lead to greater standard deviations for isospecific polymer stars. While known to affect the degradation of PLA samples, this trend has not been quantified for stereocontrolled star PLAs. Little effect of molecular weight on atactic and heterotactic samples is observed. This is attributed to the amorphous character of these samples dominating the effects of Mn. The morphology of these samples allows for such a rapid uptake of the degradation solution that slight differences in Mn were not discernable by these methods. However an effect of molecular weight on degradation lifetimes for the isotactic samples was observed. By plotting the molecular weight of the stars vs. time, strong correlations (R2 = 0.97, 0.99) between molecular weight and polymer lifetimes were observed. While these trends are significant, the effect of stereoregularity dwarfs these changes and provides a much more powerful method to alter the hydrolytic degradation in solution. These results for isotactic-rac and isotactic-l samples are illustrated in Figure 4.

The effect of molecular weight on these polymer star properties is greater at low molecular weights. Controversy exists over whether a typical Mn of ~8700 Da could be considered too low to properly assess the materials properties. To alleviate these concerns, isotactic-rac and isotactic-l samples with arms of 50 and 100 monomer units were prepared. These experiments proved an interesting study in the initiation efficiency of our system and arm length uniformity it provided. These results are presented in Table 1.

Figure 4. Relationship between Mn and polymer lifetimes for isotactic rac () and l () DPE-PLA stars.

Table 1. Polymer Properties of Isotactic-rac50 and Isotactic-L50 DPE-PLA Starsa,b,c

Entry

Mn,GPC

Mn,th

PDI

Tg (oC)

Tm (oC)

Td (oC)

Xc (%)

HDT (min)d

I-r-50

41,403

43,038

1.16

61.3

185.8

348

38

475

I-L-50

40,726

42,754

1.08

63.4

165.1

322

45

1,266

I-r-100

76,723

84,504

1.15

62.3

190.2

370

39

950

I-L-100

78,493

84,504

1.10

64.5

167.5

320

44

2840

HDT, hydrolytic degradation time.

aPolymerization conditions: 300:6:1 and 600:6:1 monomer:catalyst: initiator ratio, 120 o C.

bGPC conditions: 1 mL min-1 in THF, 2 mg mL-1 sample against PS standards with 0.58 conversion factor.

cDSC conditions: 5 o C min-1 heating and cooling rates, 50 mL min-1.

dTGA conditions: 10 o C min-1 heating rate under N2, 60 mL min-1 sample purge flow. Decomposition temperature measured at the point of fastest weight loss.

Initial attempts at preparing these higher molecular weight stereocontrolled stars revealed the limitations of this solvent-free synthetic methodology. While highly efficient for the synthesis of stars with oligomeric arm lengths, this system proved incapable of synthesizing the longer armed stars. Preliminary results for neat polymerizations at 120°C revealed improper initiation of the DPE system, resulting in broad PDI values accompanied by inconsistent molecular weights. This was remedied by returning to a solvent-based polymerization, utilizing toluene at 70°C and preinitiation with 1. This proved successful in synthesizing DPE-centred isotactic stars with controlled polymer properties, albeit at longer reaction times for all catalysts.

A close correlation between GPC-determined Mn and the theoretical Mn was observed, especially considering the tendency for size exclusion chromatography techniques to underestimate the molecular weights of star polymers.15 While this effect becomes more pronounced at the higher molecular weights, there is still sufficiently good correlation between the values. The PDI values of these stars were also quite narrow, indicating uniformity of the molecular weight distribution. PDIarm values were also determined and have been shown to correlate to the uniformity of individual arms in star and dendritic systems.16 Values of 2.10 and 1.60 for isotactic-rac and isotactic-l stars respectively are an improvement in control over those observed for the polymer stars bearing arms of 10 monomeric units in length.12 All hydroxyl functionalities of DPE have been initiated as the DPE-core CH2 peaks in the 1H NMR spectrum of these star polymers shift (3.38ppm to 4.15ppm (OCH2C(CH2)3 and 3.34ppm to 3.31ppm (OCH2)). The isotacticity bias of the tBu[salen]AlMe catalyst (0.86) improved in solution conditions.

Physical characterization of these materials was also performed. The analysis of these samples further supported trends on oligomeric12 and stars with other cores.8 An increase in the molecular weight of the samples revealed an enhancement of the Tg values with a 5-fold increase in molecular weight resulting in an 18 °C increase in Tg for isotactic-rac stars. Similar enhancement was shown for isotactic-l stars, where a 5-fold increase in molecular weight increased the Tg values 15 °C. A similar trend is observed for the melting temperatures of these higher molecular weight stars. There is an increase of 25 °C in Tm for PlLA DPE-stars, attributable to the widely established correlation between Mn and melting temperatures in linear PLA.1 For isotactic-rac samples, however, an improvement of >50°C in melt stability is observed, which cannot be explained by simple increases in Mn alone. By utilizing rac-lactide, a racemic mixture of d- and l-lactide, it is possible that adjacent polymer star arms bearing strong RRRR and SSSS character are being synthesized, potentially creating stereocomplex behavior through intermolecular dipole-dipole interactions. Stereocomplex interactions are known to greatly enhance the thermal properties of linear PLA materials.17 While the greatest enhancement is shown to occur at 50:50 mixtures of PdLA and PlLA, a significant improvement in properties can be observed even at small PdLA incorporation. The enhanced properties that this intermolecular complex affords can reasonably result in the increased melt resistance that we have observed. Interestingly, no significant improvement in physical properties is noted for arms of 50 versus 100 monomer units, indicating that the transition from oligomeric to polymeric arms is inducing these changes.

This stereocomplex behaviour is also observed in the TGA analysis of these stars. An enhancement of the thermal stability of these larger stars is to be expected, and improved stability with an increase in Mn for the isotactic-l50 stars is shown. In agreement with our other results, a greater than expected enhancement of the thermal stability for the isotactic-rac stars is observed by TGA analyses (Figure 5).

Figure 5. A TGA overlay of isotactic-l50 (red) and isotactic-rac50 (blue).

Percent crystallinity calculations on these materials revealed an increase in crystallinity for isotactic-rac50 stars (38-39%) compared to their oligomeric analogues (~30%).12 Xc values of 45% were observed for the isotactic-l50 stars, an improvement of roughly 5% from the iso-l oligomeric stars. These results indicate that molecular weight increases have a positive effect on the percent crystallinity of DPE-PLA stars, which is to be expected. Longer chain arms should allow for larger crystalline domains to form resulting in increased overall crystallinity. Interestingly, while stereocomplex domains are only possible in rac-derived samples, these materials have lower percent crystallinity indicating that the stereoerrors have an overriding effect on the size of crystalline domains.

This was supported by hydrolytic degradation experiments. As expected for polymer stars of greater molecular weight, the polymer lifetimes of these stars exhibited significant enhancements for samples of similar isospecificities, increasing the polymer lifetimes from 58 to 475 to 950 minutes for isotactic-rac stars and from 175 to 1266 to 2840 minutes for the isotactic-l stars. Interestingly, a 5-fold increase in Mn (Table 1) correlates to a greater than 7-fold increase in polymer persistence, well beyond any observed experimental deviations. No changes in Tm or Td transitions were observed. This suggests that it is predominantly crystallinity, and specifically the size of crystalline domains, that is the overriding factor in protecting against hydrolysis in PLA polymer stars.

Controlling polymer properties by controlling isospecificity of DPE-PLA stars.

While stereocontrol over star arms has clear potential to alter polymer properties, the potential to further tune the polymer properties by controlling the percentage of isotactic linkages was of particular interest. Rather than utilize a series of catalysts, the results from which could not be considered consistent, a method using Sn(Oct)2, 3, coupled with a series of artificially generated monomer blends (l-lactide and rac-lactide) was employed to prepare polymers with oligomeric (10 monomer units) arms of various isotacticites. By employing a single catalyst, reproducibility of the results was assured, as well as consistent reaction times, kinetics and conversions. Polymer characterization and monomer details for this series of experiments are presented in Table 2.

Table 2. Polymerization Data for DPE-PLA Stars with Various Isotacticity Biases

Composition

(L:rac)

Mn,GPC

Mn,th

PDI

Tact. Bias (%)

Conv. (%)

100:0

8,771

8,781

1.18

100

99

90:10

8,552

8,930

1.16

93

92

80:20

8,405

8,645

1.15

83

92

70:30

8,300

8,645

1.12

74

93

60:40

8,473

9,065

1.11

62

90

50:50

8,223

8,787

1.11

51

90

0:100

8,873

9,464

1.14

0

94

The correlated molecular weights, high conversions, narrow PDIs and good isospecificity confirm the living nature of the polymerization and the potential for this simple methodology to control thermal and degradation properties for these polymer stars. Interestingly, there was an increase in isotacticity bias greater than the l-lactide percent content in all samples. As rac-lactide is a racemic mixture of d- and l-lactide, our contents should read, for example, 80:20 l:rac as 90:10 l:d. Due to the ability of the d-monomer to affect the linkage at both its ends, it should be projected as an isotacticity bias of 80% for this sample. The increase observed above this would presumably account for the statistical possibility of two d-lactide monomers being inserted adjacent to each other. Of greatest interest, however, is the affect that altering the polymer tacticity has on the properties of the resultant materials. Analysis by DSC, TGA and pXRD is summarized in Table 3.

Table 3. Thermal and Physical Properties of DPE-PLA Stars of Various Isospecificities

Composition (L:rac)

Tg (oC)

Tm (oC)

Td (oC)

d-spacing (Å)

Xc

(%)

100:0

47.7

127.4, 141.9

304.5

5.2

44

90:10

47.4

125.6, 138.5

314.6

5.2

31

80:20

44.7

121.6, 132.8

312.0

5.2

23

70:30

43.7

117.3, 128.3

298.8

5.2

15

60:40

43.5

–

294.3

–

–

50:50

43.2

–

289.9

–

–

0:100

39.9

–

255.1

–

–

Clear trends in the thermal stability of PLA polymer stars were noted with increasing isospecificity, as shown by TGA of the samples (Figure 6). DSC analysis showed little trending in the Tg values for samples between 51% and 74% isotacticity bias (Figure 7). However, a noticeable improvement of the Tg is observed for the 80:20 and 90:10 composite samples. This reinforces the assertion that tacticity bias, and the crystalline character of these stars, has a profound effect on the polymer properties. It also suggests that a clear switch may occur as the material changes from amorphous to semi-crystalline in nature.

Figure 6. A TGA overlay of mixed monomer DPE-PLA stars, blue-50:50, green-60:40, red-70:30, purple-80:20 and light blue-90:10.

Figure 7. A DSC overlay (only Tg region shown) of mixed monomer DPE-PLA stars, blue-50:50, red-60:40, green-70:30, purple-80:20, light blue-90:10.

Figure 8. pXRD patterns for DPE-PLA stars derived from mixed monomers and rac-lactide monomer.

The melting signals of these mixed monomer stars reinforce this trend. Distinct melting signals were observed for only the 90:10, 80:20 and 70:30 composites, which extends the isotacticity range we observed with isotactic-rac stars with 82% isotacticity bias.12 The Xc values of the mixed monomer stars were calculated and the observed trend indicates that the crystallinity of the samples bearing a distinct melting transition decreased with increasing content of rac-lactide. The propensity for l-lactide to form crystalline domains allows for the prediction of this trend. The observed Xc values were found to increase from the 70:30 composites, which had an initial crystallinity of 15%, to the 80:20 composites, which possessed an average Xc value of 23%. The 90:10 composites more closely resembled the iso-rac composites, possessing an average Xc value of 31% which indicated that the percent crystallinity can be tuned by this method. Additionally, 15% crystallinity seems to be near the threshold for these materials, below which no discernable melting transitions can be observed. To better understand the lack of true melting signals for some of these composites, pXRD analysis was performed on them (Figure 8).

While the isotacticity bias does not change the nature of the crystalline domains, as shown by equivalent diffraction angles and d-spacing across all crystalline samples, clear trends exist in crystallinity and crystallite size. Changes in isotacticity from 93% to 74% increases our average crystallite size from 110 to 230 nm, suggesting enhanced amorphous aggregation and shorter crystalline domains for samples with lower isospecificity. Decreases in the diffraction intensity (linear counts) occur when the content of l-lactide is decreased, until at 60% l-lactide content a loss of the semicrystalline nature of isotactically-enriched PLA is observed. Samples below this threshold show featureless pXRD patterns characteristic of amorphous materials.

As our interest in the area of polymer stars has focused on their biomedical applications, we were particularly interested in the control of hydrolytic degradation by variation of the isospecificity. As previously described, pressed PLA samples were exposed to a standard solution of TBD in MeOH. The results of this analysis are presented in Figure 9. A near linear relationship between the degradation time and the isotacticity bias exists for the semi-crystalline samples, with an R2 of 0.95. Shorter polymer lifetimes for the amorphous samples results from easier uptake of the basic solution.

Figure 9. Polymer lifetimes of mixed monomer pellets. Samples 1 (50:50), 2 (60:40), 3 (70:30) 4 (80:20) 5 (90:10), n = 4.

Experimental

General Procedures and Materials. All chemicals and solvents were obtained from Sigma Aldrich unless otherwise stated. Purasorb d,l-lactide and l-lactide (PURAC Biomaterials) were sublimed under vacuum three times prior to use. Technical grade dipentaerythritol (DPE) was recrystallized from ethyl acetate and dried in vacuo prior to use. 2,6-Diisopropylaniline (90%), benzyl alcohol (99%), cyclohexylamine and tin(II) bis(2-ethylhexanoate) were distilled under nitrogen prior to use. Reagent grade pentane, tetrahydrofuran and toluene were collected using an inline solvent purification system, consisting of alumina columns and a copper catalyst. The solvents were then degassed by three consecutive freeze-pump-thaw cycles. All other solvents were used as received.

All air-sensitive syntheses were performed in an MBraun Labmaster sp glovebox equipped with -35°C freezer, [O2] and [H2O] analyzers and built-in Siemens Simantic touch panel. All other inert atmosphere manipulations were performed under dinitrogen atmosphere on a dual manifold Schlenk line utilizing standard Schlenk techniques. 1H and 13C NMR spectra were collected on a 300 MHz Bruker Avance Spectrometer. GPC analyses were performed on a Polymer Laboratories PL-GPC 50 Plus system equipped with three 300 × 7.5 mm Resipore columns and a refractive index detector. Samples were dissolved and eluted in HPLC grade THF at a flow rate of 1 mL min-1 at 50°C. TGA analyses were carried out on a TA Instruments TGA Q500 under N2 atmosphere with balance and purge flow rates of 40 and 60 mL min-1 respectively and a heating rate of 10°C min-1. A furnace purging time of 15 minutes was employed prior to the start of data collection. DSC analyses were completed on a TA Instruments DSC Q100 in hermetically sealed aluminum pans. A flow rate of 50 mL min-1 and heating parameters of 5°C min-1 for heating and cooling were employed. Thermal properties were measured on both powder and pelleted forms of polymers with no noted deviations. Powder X-ray diffractograms were collected on a Bruker AXS Advance D8 diffractometer equipped with a graphite monochromator, variable divergence slit, variable antiscatter slit and a scintillation detector. Cu(Kα) radiation (λ = 1.542 Å), θ-2θ detection mode and 2-60° collection parameters were utilized. Samples were collected as loose powders on glass substrates adhered with double sided tape.

Synthesis of tBu[salen]AlMe (1). This compound was prepared according to a modified literature method.6 N,N'-bis(3,5-di-tert-butyl-2-hydroxybenzaldehyde)-1,3-propanediamine (2.50 g, 4.93 mmol) was dissolved in 10 mL toluene in a glass ampoule equipped with magnetic stir bar. To this stirred solution was added trimethylaluminum (2.47 mL, 2.0 M in heptane) dropwise. Heat and gas were evolved. The glass ampoule was sealed and the reaction contents were allowed to stir at 110°C for 12 h. A yellow solid was observed to precipitate from solution. The precipitate was collected by vacuum filtration and washed with cold toluene to give the desired yellow solid. Yield: 1.94 g, (72%). 1H NMR (300 MHz, C6D6, 25°C) δ: 7.75 (d, 2H, Ar-H, 2.4 Hz), 7.34 (s, 2H, HC=N), 6.89 (d, 2H, Ar-H, 2.4 Hz), 3.07 (m, 2H, N(CH'HCH2CH'H)N), 2.75 (m, 2H, N(CH'HCH2CH'H)N), 1.79 (s, 18H, ArC(CH3)3), 1.37 (m, 2H, N(CH2CH2CH2)N and s, 18H, ArC(CH3)3), -0.37 (s, 3H, AlCH3). 13C NMR (75 MHz, C6D6, 25°C) δ: 170.15, 164.12, 141.29, 137.48, 130.25, 127.45, 118.89, 54.94, 35.94, 34.16, 31.68, 30.18, 27.38.

Synthesis of Cl[salan]AlMe (2). This compound was prepared according to a modified literature method.5 N,N'-dibenzyl-N,N'-bis[(3,5-dichloro-2-hydroxyphenyl)methylene]-1,3-diaminopropane (3.06 g, 5.08 mmol) was dissolved in 15 mL of toluene in a glass ampoule equipped with magnetic stirring bar. To this stirred solution was added trimethylaluminum (2.54 mL, 2.0 M in heptane) dropwise. Heat and gas were evolved. The glass ampoule was sealed and the reaction contents were allowed to stir at 110°C for 12 h to yield a cloudly solution. Solvent was removed in vacuo to give the desired white solid. Yield: 2.24 g (70%). 1H NMR (300 MHz, C6D6, 25°C) δ: 7.39 (d, 2H, Ar-H, 2.6 Hz), 7.06-7.01 (m, 6H, Ar-H), 6.86-6.81 (m, 4H, Ar-H), 6.56 (d, 2H, Ar-H, 2.6 Hz), 5.1–3.2 (br m, 4H, NCH2Ph), 3.26 (br s, 4H, NCH2C6H5), 2.5–1.5 (br m, 4H, N(CH2CH2)N), -0.25 (s, 3H, AlCH3). 13C NMR (75 MHz, C6D6, 25°C) δ: 154.95, 132.23, 130.99, 129.89, 129.11, 128.71, 127.14, 126.37, 123.04, 120.79, 57.12, 56.51, 46.75, -8.19.

Synthesis of DPE-PLA stars under solvent-free reaction conditions. Rac and/or l-lactide (0.50 g, 3.52 mmol), DPE (1.50×10-2 g, 0.059 mmol) and Sn(Oct)2 (4.20×10-2 g, 0.354 mmol) were loaded into a glass ampoule equipped with a mechanical stir bar. The ampoule was sealed and heated to 120°C for the desired reaction time. The ampoule was removed from the heating bath and the contents were disolved through addition of a 10:1 v/v solution of CH2Cl2 and MeOH. The solution was allowed to stir at room temperature for 0.5 h, at which point it was precipitated by dropwise addition into chilled (-15°C) and stirring MeOH (100 mL). The supernatant was decanted to reveal the resulting polymer, which was dried in vacuo and weighed. The need for reprecipitation was determined by 1H NMR or TGA.

Synthesis of DPE-PLA stars in toluene. Rac and/or l-lactide (0.50 g, 3.52 mmol), DPE (1.50×10-2 g, 0.059 mmol) and Sn(Oct)2 (4.20×10-2 g, 0.354 mmol) were loaded into a glass ampoule equipped with a mechanical stir bar and dissolved in 5 mL of toluene. The ampoule was sealed and heated to 70°C for the desired reaction time. The ampoule was removed from the heating bath and the reaction was quenched by addition of 1 mL of MeOH. The solution was allowed to stir at room temperature for 0.5 h, at which point it was precipitated by dropwise addition into chilled (−15°C) and stirring MeOH (100 mL). The supernatant was decanted to reveal the resulting polymer, which was dried in vacuo and weighed. The need for reprecipitation was determined by 1H NMR or TGA. 1H NMR (300 MHz, CDCl3, 25°C) δ: 5.16 (m, PLA-CH), 4.35 (m, PLA-CHlast, 4.12 (dd, 6H, OCH2CCH2), 3.32 (s, 4H, OCH2), 2.66 (d, OH), 1.55 (m, PLA-CH3). 13C NMR (75 MHz, CDCl3, 25°C) δ: 169.62, 72.08, 71.72, 69.32, 66.53, 16.86.

Degradation of DPE-PLA polymer stars. Pellets of DPE-PLA stars of various tacticities were pressed under 2000 psi (d =1.3 cm, m = 200 mg). The pellets suspended in a solution of 1,5,7-triazabicyclo[4.4.0]dec-5-ene (TBD) in methanol at a concentration of 4.34×10-2 M. The samples were agitated at 23°C and monitored until all PLA had dissolved. The resultant solution contents were split into two portions, concentrated in vacuo and dissolved in either d4-MeOH or CHCl3. Analysis by 1H NMR spectroscopy (d4-MeOH) confirmed the presence of PLA oligomers and lactic acid. CHCl3 solutions were analyzed by GPC, however oligomers were too short to be quantified. 1H NMR spectra were further analyzed, integrating backbone and endgroup signals, suggesting an average oligomer length of 3.5 monomer units.

Conclusion

We have shown that variation of the stereospecificity of poly(lactic acid) arms in polymer stars can have a dramatic effect on polymer properties. Characterization of DPE-centred stars of PLA derived from rac-lactide shows that the thermal characteristics Tg, Tm and Tc are tuned by the tacticity of the polymer arms. An increase in the isotacticity bias leads to an increase in the Tg, Tm and, at high isospecificity, induces a Tc. While only isotacticity increased thermal transition temperatures significantly, any stereoregularity induced an improved thermal stability by >40°C. The solution stability of these materials was also investigated. When stereocontrol was introduced to these samples, lifetimes of pressed PLA star pellets were enhanced by upwards of a factor of 7 when exposed to a harsh basic solution. Investigations at higher Mn reveal that these trends are consistent across a wide molecular weight range. The tunable nature of these materials was further investigated by varying the monomer composition of these stars to prepare isotactically-biased stars with different Pm values. We determined that there was a direct correlation between the Pm, and the observed thermal, solution and X-ray properties of these materials. Additionally, the range at which these materials transition from amorphous to a more regular semi-crystalline polymeric material was identified. Current research in our group in this area is focused on the application of stereocontrol to other non-linear polymer architectures, the effect of core structure on polymer properties and kinetics and further characterization of the DPE-centred PLA stars to determine their true hydrodynamic volumes and solution shapes.

References

[1] Stanford, M. J.; Dove, A. P. Chem. Soc. Rev. 2010, 39, 486-494.

[2] Henton, D. E.; Gruber, P.; Lunt, J.; Randall J. in Natural Fibres, Biopolymers and Biocomposites, ed. Mohanty, A. K.; Misra, M.; Drzal, L. T. New York: CRC Press, 2005.

[3] O’Keefe, B. J.; Hillmyer, M. A.; Tolman, W. B. Dalton Trans. 2001, 2215-2224.

[4] Dechy-Cabaret, O.; Martin-Vaca, B.; Bourissou, D. Chem. Rev. 2004, 104, 6147-6176.

[5] Hormnirun, P.; Marshall, E. L.; Gibson, V. C.; White, A. J. P.; Williams, D. J. J. Am. Chem. Soc. 2004, 126, 2688-2689.

[6] Hormnirun, P.; Marshall, E. L.; Gibson, V. C.; Pugh, R. I.; White, A. J. P. Proc. Nat. Acad. Sci. 2006, 103, 15343-15348.

[7] Ovitt, T. M.; Coates, G. W. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 4686-4692.

[8] Cameron, D. J. A.; Shaver, M. P. Chem. Soc. Rev. 2011, 40, 1761-1776.

[9] Hsu, J.-C.; Sugiyama, K.; Chiu, Y.-C.; Hirao, A.; Chen, W.-C. Macromolecules 2010, 43, 7151-7158.

[10] Shaver, M. P.; Perry, M. R. Can. J. Chem. 2011, 89, 499-505.

[11] Stanford, M. J.; Dove, A. P. Macromolecules 2009, 42, 141-147.

[12] Shaver, M. P.; Cameron, D. J. A. Biomacromolecules 2010, 11, 3673-3679.

[13] de Jong, S. J.; Arias, E. R.; Rijkers, D. T. S.; van Nostrum, C. F.; Kettenes-van den Bosch, J. J.; Hennink, W. E. Polymer 2001, 42, 2795-2802.

[14] Xiao, Y.; Wang, Z.; Ding, K. Chem. Eur. J. 2005, 11, 3668-3678.

[15] Stenzel-Rosenbaum, M. H.; Davis, T. P.; Chen, V.; Fane, A. G. Macromolecules 1991, 34, 5433-5438.

[16] Matyjaszewski, K. Polym. Int. 2003, 52, 1559-1565.

[17] Drumright, R. E.; Gruber, P. R.; Henton, D. E. Adv. Mater. 2000, 12, 1841-1846.

Page 15 of 16