Comonomer Unit Composition and Thermal Properties of Poly(3

Biomacromolecules 2001, 2, 1285-1293

1285

Comonomer Unit Composition and Thermal Properties of Poly(3-hydroxybutyrate-co-4-hydroxybutyrate)s Biosynthesized by Ralstonia eutropha Kazuki Ishida, Yi Wang, and Yoshio Inoue* Department of Biomolecular Engineering, Tokyo Institute of Technology, Nagatsuta 4259, Midori-ku, Yokohama 226-8501, Japan Received July 16, 2001; Revised Manuscript Received September 12, 2001

A series of poly(3-hydroxybutyrate-co-4-hydroxybutyrate)s [P(3HB-co-4HB)s] with different 4HB content, biosynthesized by Ralstonia eutropha H16 with mixed carbon sources of 4-hydroxybutyric acid (4HBA) and butyric acid, were fractionated by solvent/nonsolvent fractionation into copolyester fractions with different 4HB content and narrower compositional distribution. The fractions obtained were classified into two groups, 3HB- and 4HB-rich P(3HB-co-4HB)s. The thermal properties were investigated for these fractionated copolyesters. With increasing 4HB content, the melting temperature at first decreased while 3HB content was rich, and then increased while 4HB content was rich. The glass transition temperature decreased linearly with increasing 4HB content. The 4HB-rich P(3HB-co-4HB) was found to be immiscible with the 3HBrich P(3HB-co-4HB), as two glass transitions corresponding to those of respective P(3HB-co-4HB)s were observed by DSC. It was concluded that as-produced bacterial P(3HB-co-4HB) samples used in this study should be considered as immiscible polymer blends. Introduction A wide variety of bacteria synthesize an optically active polyester, poly(3-hydroxybutyrate) [P(3HB)], and accumulate it in their cells as carbon and energy sources.1,2 P(3HB) has attracted much industrial attention as a biodegradable and biocompatible material for agricultural, marine, and medical applications.2 However, P(3HB) is rather brittle compared with common chemosynthesized plastics. In purpose of improving the physical properties of P(3HB), its copolymers have been widely studied. Poly(3-hydroxybutyrate-co-4hydroxybutyrate) [P(3HB-co-4HB)] is a biodegradable thermoplastic, and many kinds of studies have been reported about this copolyester produced by Ralstonia eutropha,3-9 Alcaligenes latus,7,10 and Comamonas acidoVorans.7,11-13 Recently, it has been found that the range of compositional distribution in several bacterial copolymers was very broad and complex.5,8,9,13-21 In general, compositional distribution as well as comonomer unit composition influence various properties of copolyesters. So the compositional fractionation has been carried out in order to obtain copolyester fractions having a narrower compositional distribution and to examine the precise relationships between the comonomer unit composition and several properties including biodegradability of bacterial copolyesters.5,7-9,11-20 For example, Doi et al.5 have performed fractionation of P(3HB-co-24mol %4HB) produced by R. eutropha with hot acetone, and they found that the soluble and insoluble fractions in hot acetone had 86 and 7 mol % 4HB, respectively. Shi et al.8 have carried out fractionation of P(3HB-co-28mol %4HB) produced by R. eutropha with chloroform/acetone and acetone/methanol * Corresponding author. E-mail: [email protected].

mixed solvent systems. Acetone-insoluble (AIS) and acetonesoluble fractions (AS) had 1 and 52 mol % 4HB, respectively. The AS fraction was further fractionated with acetone/ methanol, and the soluble (AS-MAS) and insoluble (ASMAIS) fractions in this mixed solvent were found to have 28 and 82 mol % 4HB, respectively. Kimura et al.9 have adopted chloroform/n-hexane solvent/nonsolvent system for the compositional fractionation of a series of P(3HB-co4HB)s produced by R. eutropha. They found that asproduced P(3HB-co-4HB)s were fractionated to 3HB-rich and 4HB-rich fractions. Recently, Mitomo et al.13 have used acetone/water fractionation procedure for the fractionation of P(3HB-co-4HB)s produced by C. acidoVorans. The 4HB content of fractions decreased with increasing acetone concentration. The comonomer unit composition, the compositional distribution, and their effects on the morphology, the physical properties, and biodegradability have been well investigated for a series of bacterial poly(3-hydroxybutyrate-co-3-hydroxyvalerate)s [P(3HB-co-3HV)s],14-17 and poly(3-hydroxybutyrate-co-3-hydroxypropionate)s [P(3HB-co-3HP)s].18-21 It has been found that some physical properties and biodegradability of well-fractionated bacterial P(3HB-co3HP)s were different from those of as-produced unfractionated bacterial ones having similar comonomer unit composition but having broad compositional distribution.18-21 The same results have been also found for P(3HB-co-3HV)s.15,16 As the monomeric unit of the 4HB has one more backbone carbon atom than those of 3HB, 3HV, and 3HP units, P(3HBco-4HB) shows unique crystallization behavior and physical properties different from those of P(3HB-co-3HV) and P(3HB-co-3HP).2-6,22 It is very interesting and important to

10.1021/bm010115a CCC: $20.00 © 2001 American Chemical Society Published on Web 10/26/2001

1286

Biomacromolecules, Vol. 2, No. 4, 2001

Ishida et al.

Figure 1. 270 MHz 1H NMR spectrum of fraction7 (47.3% 4HB) of sample 2 in CDCl3. Table 1. Biosyntheses of Copolyesters from 4HBA and Butyric Acid in the Presence of Ammonium Sulfate (2 g/L) by R. eutropha at 30 °C sample 1 2 3 4 a

carbon sources (g/L) 4HBAa butyric acid 9 10 10 11

6 5 5 4

time (h)

cell dry wt (g/L)

polyester contentb (wt %)

48 48 78 48

7.42 6.98 6.89 6.92

17.7 21.5 10.1 13.7

compositionc (mol %) 3HB 4HB 83.3 49.1 30.2 34.9

av mol wtd 10-5Mn Mw/Mn

16.7 50.9 69.8 65.1

2.6 1.0 1.9 4.8

3.4 6.5 3.9 2.1

4-Hydroxybutyric acid. b Polyester content in cell dry weight. c Determined by 1H NMR. d Determined by GPC.

Table 2. Fractionation of Bacterial P(3HB-co-16.7%4HB) with Chloroform/n-Hexane sample fraction sample 1

a

original 1 2 3 4 5 6 7

av mol wtb

concn of n-hexane (vol %)

amt of sample in fraction (wt %)

4HB contenta (mol %)

10-5Mn

Mw/Mn

50 52 54 56 58 60 62

100 5.5 16.9 51.5 7.2 9.0 2.0 1.0

16.7 46.4 6.3 9.1 26.4 16.6 28.8 26.6

2.6 5.5 3.0 5.0 3.7 1.8 1.0 0.5

3.4 1.7 3.2 2.1 2.3 3.5 3.8 3.3

Tm (°C)c 140.6 41.7 143.2 140.5 134.8 133.1 49.2 52.1

142.1

124.9 123.5

Tg (°C)c 2.5 -33.8 2.4 2.1 -25.1 0.1 -28.1 -26.7

3.0

1.5 -6.6 -7.6

Determined by 1H NMR. b Determined by GPC. c Measured by DSC.

investigate the comonomer unit composition-property relations for P(3HB-co-4HB)s by using well-fractionated samples having different 4HB content and having narrow comonomer unit compositional distribution. In this study, the comonomer unit composition and its distribution of P(3HB-co-4HB) samples, biosynthesized by R. eutropha H16, are investigated by fractionation with chloroform/n-hexane and chloroform/ethanol solvent/nonsolvent systems. Their comonomer unit composition, molecular weight, and thermal properties are characterized by 1H and 13C nuclear magnetic resonance (NMR) spectroscopy, gel permeation chromatography (GPC), and differential scanning calorimetry (DSC).

Experimental Section Biosyntheses of Copolyesters. R. eutropha H16 (ATCC17699) was used for biosyntheses of P(3HB-co-4HB) samples in this study. The biosyntheses of P(3HB-co-4HB)s was carried out by the two-stage cultivation of R. eutropha. At the first stage, R. eutropha cells were grown under an aerobic condition at 30 °C for 24 h on a shaker in a 500 mL Sakaguchi flask (Shaking flask with flat bottom: Iwaki Co., Tokyo) with 100 mL of nutrient-rich medium containing 1 g of yeast extract, 1 g of polypepton (Nihon Seiyaku Co., Tokyo), 0.5 g of meat extract, and 0.5 g of (NH4)2SO4. The cells were harvested by centrifugation at 6000g for 15min. The centrifuged cells were transferred into the 500 mL flask

Biosynthesized [P(3HB-co-4HB)s]

Biomacromolecules, Vol. 2, No. 4, 2001 1287

Figure 2. DSC thermograms of the original and fractionated samples during the first heating scan: (a) sample 1 (16.7% 4HB); (b) sample 2 (50.9% 4HB); (c) sample 3 (69.8% 4HB).

with 100 mL of mineral medium containing 0.265 g of KH2PO4, 0.958 g of Na2HPO4‚12H2O, 0.02 g of MgSO4, and 1 mL of a microelement solution. The microelement solution contained 0.217 g of CoCl2‚6H2O, 16.2 g of FeCl3‚6H2O, 7.8 g of CaCl2, 0.118 g of NiCl2‚6H2O, 0.135 g of CrCl3‚ 6H2O, and 0.156 g of CuSO4‚5H2O (per liter of 0.1 M HCl). The prescribed amounts of 4-hydroxybutyric acid (4HBA), butyric acid, and ammonium sulfate were added to the mineral medium. The cells were cultivated in these media for 48 or 78 h at 30 °C, and then harvested by centrifugation, washed with distilled water, and lyophilizized. Copolyesters were extracted from the lyophilized cells with hot chloroform using a Soxhlet apparatus, and purified by reprecipitation with n-hexane. The precipitated mass was isolated by centrifugation (8000g, 15 min). The solvent-cast film was prepared from chloroform solution of each precipitate, and the film was dried under vacuum at 40 °C for 24 h. Fractionation Procedures. The bacterially as-produced original copolyesters were compositionally fractionated by

repeated solvent/nonsolvent fractionation procedures. At first, the original copolyester sample (1.0 g) was dissolved in chloroform (100 mL), and then predetermined constant amount of n-hexane was added slowly to this solution with stirring at room temperature. The precipitated mass was isolated by centrifugation (8000g, 15min). After this, the same amount of n-hexane was again added to the remained solution and the precipitated mass was isolated. This procedure was repeated until adding any amount of n-hexane could not cause appreciable precipitation. The solvent-cast film was prepared from chloroform solution of each precipitate. The film was dried under vacuum at 40 °C for 24 h. It was found that some of the precipitated fractions obtained by fractionation with chloroform/n-hexane could not be fractionated appropriately in comonomer unit compositional distribution, so these fractions were combined and were dissolved again in chloroform (40 mL). Ethanol was added slowly to this solution with stirring at 0 °C until turbid. The mixture was stored at -20 °C overnight, and the

1288


Ishida et al.

Figure 3. DSC first-scan thermograms of the fractions obtained by chloroform/n-hexane fractionation and chloroform/ethanol soluble and insoluble fractions: (a) fraction 1 of sample 1; (b) fractions 7-10 of sample 2; (c) fractions 4 and 5 of sample 3; (d) fractions 7-9 of sample 3. Table 3. Fractionation of Bacterial P(3HB-co-50.9%4HB) with Chloroform/n-Hexane concn of n-hexane (vol %)



original 1 2 3 4 5

44 46 48 50 52

100 0.4 13.8 16.4 5.8 2.5

50.9 95.0 96.1 96.4 89.2 80.8

6

54

36.2

7.1

7

56

3.3

47.3

8

58

1.8

50.9

9

60

2.1

31.1

10

62

1.0

46.8

sample fraction sample 2

a

av mol wtb 10-5Mn

Mw/Mn

1.0 n.d.d 4.7 2.9 1.6 6.8 0.3 8.2 0.4 0.3 0.01 6.8 0.3 4.2 0.2 1.0 0.1

6.5 n.d. 1.7 2.1 2.4 1.1 1.9 1.8 1.8 2.0 1.3 1.9 1.3 1.9 1.2 1.8 1.1

Tm (°C)c 54.0 n.d. 50.8 51.1 51.0 50.7

140.7

140.5

142.3

Tg (°C)c -43.4 n.d. -42.3 -43.3 -42.7 -42.9 2.4

49.5

135.5

-45.0

52.4

132.3

-44.5

50.8

129.3

-45.8

n.d.

n.d.


resulting white precipitated mass was isolated by centrifugation (8000g, 15min). The solvent (chloroform/ethanol) was then removed by rotoevaporation. By this way, two P(3HBco-4HB) fractions, soluble and insoluble in chloroform/ ethanol mixed solvent, were obtained. Analytical Procedures. 1H NMR and 13C NMR analyses were performed on a JEOL GSX-270 spectrometer. The 270 MHz 1H NMR spectra were recorded at 18 °C in CDCl3 solution of polyester (3 mg mL-1) with 5.5 µs pulse width

(π/4 pulse angle), 4s pulse repetition time, 5400 Hz spectral width, 16 000 data points, and 32 accumulations. The 67.8 MHz 13C NMR spectra were recorded at 18 °C in CDCl3 solution of polyester (10 mg mL-1) with 4.2 µs pulse width (π/4 pulse angle), 1s pulse repetition time, 18000 Hz spectral width, 32 000 data points, and 10 000 accumulations. Tetramethylsilane (SiMe4, δ ) 0 ppm) was used as an internal chemical shift standard. The 4HB mole fractions of the fractionated samples along with the original products were


Biosynthesized [P(3HB-co-4HB)s] Table 4. Fractionation of Bacterial P(3HB-co-69.8%4HB) with Chloroform/n-Hexane concn of n-hexane (vol %)



original 1 2 3 4 5

44 46 48 50 52

100 0.2 32.2 21.0 3.7 3.8

69.8 94.9 96.2 94.4 93.2 54.0

6

54

20.3

11.0

7

56

1.2

57.7

8

58

3.4

20.5

9

60

1.0

38.5

sample fraction sample 3

a

av mol wtb 10-5Mn

Mw/Mn

1.9 n.d.d 4.2 3.0 2.0 8.0 0.5 7.3 0.4 5.8 0.2 6.8 0.2 0.8 1.2

3.9 n.d. 2.0 2.2 2.7 1.5 1.3 1.7 1.2 1.8 1.4 1.5 1.3 0.2 1.2

Tm (°C)c 50.7 n.d. 52.7 51.8 49.5 51.4

Tg (°C)c

140.9

144.1 143.8

142.5

-43.1 n.d. -42.9 -42.9 -42.1 -45.8 1.0

53.3

133.8

-44.7

53.9

130.6

-46.9

52.7

127.3

-44.8

Determined by 1H NMR. b Determined by GPC. c Measured by DSC. d Not determined.

Table 5. Results of Refractionation with Chloroform/Ethanol of the Fractionated Samples with Chloroform/n-Hexane composition (mol %)a 3HB 4HB

sample fraction

concn of ethanol (vol %)

fraction by chloroform/ethanol

wt ratio (wt %)

fraction 1 of sample 1 combined fractions 7 + 8 + 9 + 10 of sample 2

56

soluble insoluble soluble insoluble

55.8 44.2 36.2 63.8

27.3 92.1 7.9 83.2

72.7 7.9 92.1 16.8

soluble insoluble soluble insoluble

75.4 24.6 21.4 78.6

6.6 93.0 11.0 81.9

93.4 7.0 89.0 18.1

combined fractions 4 + 5 of Sample 3 combined fractions 7 + 8 + 9 of sample 3 a

65

54 68

av mol wtb Mw/Mn

10-5Mn 4.8 7.0 0.2 5.2 0.1 0.8 8.2 0.2 6.1 0.3

1.7 1.6 1.2 1.9 1.2 2.3 1.5 1.6 1.7 1.5

Tm (°C)c

Tg (°C)c

40.7 142.5 51.3 135.5

-34.1 1.8 -43.4 1.2

49.6 144.9 49.7 131.2

-42.9 2.4 -44.5 0.5


estimated from the relative integrated intensities of CH2(8) and CH(3) proton resonances (Figure 1). Average molecular weight of the polyester samples was measured by a Tosoh HLC-8020 GPC system with a Tosoh SC-8010 controller and a refractometer. Chloroform was used as an eluent at a flowing rate of 1.0 mL/min and at 40 °C. Polystyrene standards of low polydispersity were used to construct the calibration curve. The number-average (Mn) and weight-average molecular weight (Mw) and the polydispersity index (Mw/Mn) were calculated through a SC-8010 data processor. Differential scanning calorimetry (DSC) measurements were performed on a SEIKO DSC220U instrument. Samples (3-5 mg) encapsulated in aluminum pans were heated from -40 to +200 °C at a rate of 10 °C/min (the first heating scan), and held for 2 min at 200 °C. After that, samples were rapidly quenched to -100 °C with liquid nitrogen and then heated to 200 °C at a rate of 10 °C/min (the second heating scan). The melting temperature (Tm) was taken as the peak temperature of the melting endotherm in the first heating scan. The glass transition temperature (Tg) was taken as the midpoint of the heat capacity change in the second heating scan. Results and Discussion Microbial Synthesis of P(3HB-co-4HB). In Table 1 are listed the results of biosyntheses of copolyesters by R.

eutropha at 30 °C using 4-hydroxybutyric acid (4HBA) and butyric acid as the mixed carbon source in the presence of ammonium sulfate. The comonomer unit compositions of copolyesters were widely changed depending on both the combination of carbon sources and incubation time. 4HB contents increased with increasing 4HBA concentration. The number-average molecular weight (Mn) and the polydispersity index (Mw/Mn) were comparatively large. Fractionation of P(3HB-co-4HB). Of four samples listed in Table 1, three samples (samples 1-3) were fractionated by chloroform/n-hexane, and the results were listed in Tables 2-4, respectively. Except for some fractions, the higher chloroform volumetric content in the mixed solvent at which the polymer fraction was sampled, the higher was the 4HB monomer composition obtained. These tendencies are well correspondent with the fractionation results obtained for P(3HB-co-3HP).18-20 The 4HB repeating unit has one more carbon atom than the 3HP unit in the main chain, but both have no side chain. Bacterially synthesized P(3HB-co-3HP) samples have been compositionally fractionated by using mixed chloroform/n-heptane solvent.18-20 The P(3HB-co3HP) fractions with higher 3HP comonomer unit content were separated by the fractionation with mixed solvent of higher chloroform content. Sample 1 with 16.7 mol % 4HB was fractionated into 7 fractions by changing the n-hexane concentration from 50 vol % to 62 vol %. The 4HB unit content of these fractions ranged from 9.1 to 46.4 mol %. Thus, the bacterially

1290


Ishida et al.

Figure 4. DSC second-scan thermograms of the fractions obtained by chloroform/n-hexane fractionation and chloroform/ethanol soluble and insoluble fractions: (a) fraction 1 of sample 1; (b) fractions 7-10 of sample 2; (c) fractions 4 and 5 of sample 3; (d) fractions 7-9 of sample 3.

synthesized P(3HB-co-16.7mol %4HB) sample was found to have very broad comonomer unit compositional distribution as found previously for P(3HB-co-3HV)14,15,17 and P(3HB-co-3HP).18-20 The 4HB unit content of the main fraction, that is, the fraction 3 (51.5%), is 9.1 mol %, but not the average 4HB content of 16.7 mol % for original sample 1. Sample 2 with 50.9 mol % 4HB was fractionated into 10 fractions by 44-62 vol % n-hexane. The 4HB unit content of the fractions ranged from 7.1 to 96.4 mol %. The compositional distribution of sample 2 is much broader than that of sample 1. The main fraction (fraction 6) has a 4HB unit content of 7.1 mol %, which is very smaller than that of original sample (50.9 mol %). Sample 3 with 69.8 mol % 4HB was fractionated into nine fractions by 44-60 vol % n-hexane. The 4HB unit content of fractions is again very broad, that is, from 11.0 to 96.2 mol %. In this sample, the main fraction (fraction 2) has a 4HB unit content of 96.2 mol %, which is very much larger than that of original sample (69.8 mol %). This result is contrary to those found for samples 1 and 2. The 4HB contents of these fractions first decreased with increasing n-hexane concentration and then showed variable changes. On the whole, the fractions obtained by the first half of fractionation processes were rich in the 4HB unit, and those obtained by the last half of the processes were generally rich in the 3HB unit.

For the latter half fractions of samples 2 and 3, coexistence of two types of P(3HB-co-4HB) copolymers with distinctively different molecular weights were observed by GPC analysis, as shown in Tables 3 and 4. The one type of these has one-order larger molecular weight than the other. Figure 2 shows the results of DSC curves observed by the first heating scan of sample 1 (a), sample 2 (b), and sample 3 (c), consisting of those for original samples and their fractions. Original samples and many fractions had two melting peaks at high-temperature region (over 120 °C). One of them, which appeared at higher temperature, should be the melting peak due to the recrystallized phase in the case the lattice of copolyester had enough time to be rearranged at a given heating rate, because the intensity of this peak decreased with increasing the heating rate, which was relative to the peak at lower temperature.15,23 The melting peaks owing to the recrystallization were disregarded, hereafter. For the unfractionated original samples 2 and 3, two endothermic peaks (except for the peaks due to the recrystallization) were observed at low and high-temperature regions. The melting peaks (Tm) of the 4HB-rich fractions (89-96 mol % 4HB) and those of the 3HB-rich fractions (6-17 mol % 4HB) were observed at low (50-53 °C) and high temperature (133-143 °C), respectively. However, the other fractions showed clearly two melting peaks, so it was supposed that the fractions which showed two endothermic



Figure 5. GPC curves of the P(3HB-co-4HB) samples obtained by chloroform/n-hexane fractionation and chloroform/ethanol soluble and insoluble fractions: (a) fraction 1 of sample 1; (b) fractions 8-10 of sample 2; (c) fractions 4 and 5 of sample 3; (d) fractions 7-9 of sample 3.

Figure 6. Melting temperature (Tm) of the fractionated P(3HB-co4HB) samples against 4HB content.

Figure 7. Glass transition temperature (Tg) of the fractionated P(3HBco-4HB) samples against 4HB content.

peaks were not fractionated well by comonomer unit compositional difference and still contained two main components with very different comonomer unit compositions. We needed to fractionate again the fractions, which showed two DSC melting peaks. Therefore, the second fractionation was carried out by chloroform/ethanol for several fractions obtained by the first chloroform/n-hexane fractionation processes. The fraction 1 from the original sample 1 and the combined mixture of two, three, or four fractions of original samples 2 and 3, obtained by the first fractionation with chloroform/n-hexane, were further fractionated into two fractions, that is, the fraction soluble and that insoluble in chloroform/ethanol mixed solvent. The

results were listed in Table 5. All of the fractions soluble and insoluble in chloroform/ethanol mixtures showed, respectively, one main melting peak (Figure 3). The Tm values of 73-93 mol % 4HB and 7-18 mol % 4HB fractions were 41-51 and 131-145 °C, respectively. Figure 4 shows the DSC curves of samples obtained by first fractionation and refractionation recorded during the second heating scan. The fraction 1 with 46.4 mol % 4HB unit content of original sample 1 shows two glass transitions at -33.8 and +3.0 °C as shown in Table 2 and Figure 4a. Each of two fractions, that is, one soluble (72.7 mol % 4HB) and the other insoluble (7.9 mol % 4HB) in chloroform/ ethanol mixed solvent, obtained from the fraction 1 of original sample 1 shows each single glass transitions at -34.1

1292


Ishida et al.

Figure 8. 67.8 MHz 13C NMR spectrum of (a) chloroform/ethanol insoluble part (16.8% 4HB) of combined fractions 7-10 of sample 2 and (b) enlarged views of the part of carbonyl carbon resonance region of part a.

and +1.8 °C, respectively. The glass transitions observed for refractionated P(3HB-co-72.7mol %4HB) and P(3HBco-7.9mol %4HB) samples correspond to those observed for the parent fraction 1 of original sample 1. Both the refractionated P(3HB-co-72.7mol %- and 7.9mol %4HB) samples also show respective melting peaks at 40.7 and 142.5 °C, which are again very close to the two melting peaks observed for the fraction 1 of original sample 1 (Tables 2, 5). The weight fractions of refractionated P(3HB-co-72.7mol %4HB) and P(3HB-co-7.9mol %4HB) are 55.8% and 44.2%, respectively. These results clearly indicate that the fraction 1 of the original sample 1 should be considered to be a immiscible P(3HB-co-4HB) blends composed of 4HB-rich and 3HB-rich component copolyesters. Similar DSC results were observed for the fractions obtained by refractionation of the fractions of samples 2 and 3, except for the appearance of a large exothermic peak at about -5∼0 °C, under which the glass transition at hightemperature region was hidden (Figure 4, parts b-d). The chloroform/ethanol soluble fractions of samples 2 and 3 also showed the exothermic peak in the low-temperature region of about -5 to 0 °C, and the chloroform/ethanol insoluble fractions of samples 1-3 showed an exothermic peak in the temperature region higher than 50 °C (Figure 4). The exothermic peaks at low and high-temperature regions corresponded to the crystallization of 4HB-rich and 3HBrich P(3HB-co-4HB) fractions, respectively. The fraction

soluble in chloroform/ethanol mixture of the fraction 1 of the original sample 1, whose 4HB content was 72.7 mol %, did not show such DSC exothermic peak, as seen in Figure 4a. These results indicated that P(3HB-co-4HB) with 4HB content higher than ca. 90 mol % has higher tendency to crystallize than those with lower 4HB content. The higher crystallization tendency of 4HB-rich P(3HB-co-4HB) fractions should contribute, at least partly, to precipitation of these fractions in the early stage of solvent/nonsolvent fractionation. As found for original samples 2 and 3, the P(3HB-co-4HB) fractions with 4HB content larger than ca. 90 mol % precipitated by chloroform/n-hexane mixed solvent with lower n-hexane composition. Figure 5 shows the GPC curves of these fractions. In general, each of the 4HB-rich fractions soluble in chloroform/ ethanol had lower molecular weight than that of the 3HBrich fractions insoluble in chloroform/ethanol. So, it turned out that there were two components in the fractions (Tables 3 and 4), one with high 4HB unit content and small molecular weight and another with high 3HB unit content and large molecular weight. Figures 6 and 7 show the melting temperature (Tm) and the glass transition temperature (Tg) plotted against the 4HB content, respectively, obtained for the well fractionated and refractionated samples. With increasing 4HB content, the Tm value decreases while the 4HB content is low, and increases while the 4HB content is high. The Tg value decreases with


increasing 4HB content continuously. As we know from Figure 6, it was found that P(3HB-co-4HB) biosynthesized by R. eutropha using 4HBA and butyric acid in the presence of ammonium sulfate is a mixture of 4HB- and 3HB-rich copolyesters with distinct physical properties. Figure 8a shows the 13C NMR spectrum of chloroform/ ethanol insoluble fraction (16.8 mol % 4HB) of sample 2, and Figure 8b shows an enlarged view of carbonyl carbon resonances. The signals assigned to the 4HB-3HB and 3HB-4HB heterodiad sequences as well as 4HB-4HB and 3HB-3HB homodiad sequences3,5,10,13 are clearly observed, indicating that this P(3HB-co-4HB) is a true copolymer but not the blend of P(3HB) and P(4HB) homopolymers. Conclusion The comonomer unit compositional distribution and thermal properties were investigated for a series of P(3HBco-4HB)s, biosynthesized by R. eutropha H16 with the mixed carbon source of 4HBA and butyric acid in the presence of ammonium sulfate. By repeated compositional fractionation with chloroform/n-hexane and chloroform/ethanol solvent/ nonsolvent mixtures, bacterial P(3HB-co-4HB)s were found to have broad comonomer unit compositional distributions. Bacterially as-produced original P(3HB-co-4HB)s were composed of two main component copolymers, that is, 4HBrich and 3HB-rich ones. With increasing 4HB content, the melting temperatures of the 3HB-rich P(3HB-co-4HB) fractions decreased and those of the 4HB-rich fractions increased. The glass transition temperature of P(3HB-co-4HB) decreased linearly with increasing 4HB content. Molecular weights of some 4HB-rich fractions were much smaller than those of the other 4HB-rich ones, though the amount of the fractions whose molecular weights were small was a little. The 4HB-rich P(3HB-co-4HB) was immiscible with the 3HB-rich one. Thus, as-produced original P(3HB-co-4HB) samples should be considered as immiscible polymer blend systems. This conclusion is very important in the studies on relationships between comonomer unit composition and several properties, such as thermal properties, mechanical properties, biodegradability, and so on.


Acknowledgment. This work was partially supported by a Grant-in-Aid for Scientific Research on Priority Area, “Sustainable Biodegradable Plastics”, No.11217205 (2001) from the Ministry of Education, Culture, Sports, Science, and Technology (Japan). References and Notes (1) Doi, Y. Microbial polyester; VCH Publishers: New York, 1990; Microorganisms and poly(3-hydroxyalkanoates), pp 33-62. (2) Inoue, Y.; Yoshie, N. Prog. Polym. Sci. 1992, 17, 571-610. (3) Doi, Y.; Kunioka, M.; Nakamura, Y.; Soga, K. Macromolecules 1988, 21, 2722-2727. (4) Kunioka, M.; Nakamura, Y.; Doi, Y. Polym. Commun. 1988, 29, 174-176. (5) Doi, Y.; Segawa, A.; Kunioka, M. Int. J. Biol. Macromol. 1990, 12, 106-111. (6) Nakamura, S.; Doi, Y.; Scandola, M. Macromolecules 1992, 25, 4237-4241. (7) Saito, Y.; Nakamura, S.; Hiramitsu, M.; Doi, Y. Polym. Int. 1996, 39, 169-174. (8) Shi, F.; Ashby, R. D.; Gross, R. A. Macromolecules 1997, 30, 25212523. (9) Kimura, H.; Ohura, T.; Takeishi, M.; Nakamura, S.; Doi, Y. Polym. Int. 1999, 48, 1073-1079. (10) Kang, CK.; Kusaka, S.; Doi, Y. Biotechnol. Lett. 1995, 17, 583588. (11) Saito, Y.; Doi, Y. Int. J. Biol. Macromol. 1994, 16, 99-104. (12) Mitomo, H.; Morishita, N.; Doi, Y. Rep. Prog. Polym. Phys. Jpn. 1994, 37, 128-129. (13) Mitomo, H.; Hsieh, W.-C.; Nishiwaki, K.; Kasuya, K.; Doi, Y. Polymer 2001, 42, 3455-3461. (14) Mitomo, H.; Morishita, N.; Doi, Y. Polymer 1995, 36, 2573-2578. (15) Yoshie, N.; Menju, H.; Sato, H.; Inoue, Y. Macromolecules 1995, 28, 6516-6521. (16) Yoshie, N.; Menju, H.; Sato, H.; Inoue, Y. Polym. J. 1996, 28, 4550. (17) Yamada, S.; Wang, Y.; Asakawa, N.; Yoshie, N.; Inoue, Y. Macromolecules 2001, 34, 4659-4661. (18) Cao, A.; Ichikawa, M.; Ikejima, T.; Yoshie, N.; Inoue, Y. Macromol. Chem. Phys. 1997, 198, 3539-3557 (19) Cao, A.; Kasuya, K.; Abe, H.; Doi, Y.; Inoue, Y. Polymer 1998, 39, 4801-4816. (20) Wang, Y.; Ichikawa, M.; Cao, A.; Yoshie, N.; Inoue, Y. Macromol. Chem. Phys. 1999, 200, 1047-1053. (21) Na, Y.-H.; Arai, Y.; Asakawa, N.; Yoshie, N.; Inoue, Y. Macromolecules 2001, 34, 4834-4841. (22) Nakamura, K.; Yoshie, N.; Sakurai, M.; Inoue, Y. Polymer 1994, 35, 193-197. (23) Cao, A.; Ichikawa, M.; Kasuya, K.; Yoshie, N.; Asakawa, N.; Inoue, Y.; Doi, Y.; Abe, H. Polym. J. 1996, 28, 1096-1102.

BM010115A

Comonomer Unit Composition and Thermal Properties of Poly(3

Recommend Documents