Studies on Comonomer Compositional Distribution of Bacterial Poly (3

Jul 19, 2002 - Two factions were obtained from each sample, the precipitated fraction had a lower 3HH unit content while the supernatant fraction had ...
0 downloads 0 Views 255KB Size
Biomacromolecules 2002, 3, 1071-1077

1071

Studies on Comonomer Compositional Distribution of Bacterial Poly(3-hydroxybutyrate-co-3-hydroxyhexanoate)s and Thermal Characteristics of Their Factions Lidan Feng,† Takumi Watanabe,† Yi Wang,† Tomoyasu Kichise,‡ Takeshi Fukuchi,§ Guo-Qiang Chen,| Yoshiharu Doi,†,‡ and Yoshio Inoue*,† Department of Biomolecular Engineering, Tokyo Institute of Technology, Nagatsuta 4259, Midori-ku, Yokohama 226-8501, Japan, Polymer Chemistry Laboratory, RIKEN Institute, 2-1 Hirosawa, Wako-shi, Saitama 351-0198, Japan, Takasago Research Laboratories, Kaneka Corporation, Takasago, Hyogo 678-8688 Japan, and Department of Biological Science & Biotechnology, Tsinghua University, Beijing 100084, China Received May 6, 2002

The comonomer-unit compositional distributions have been investigated for bacterial poly(3-hydroxybutyrateco-3-hydroxyhexanoate) [P(3HB-co-3HH)] samples with 3HH unit content of 13.8, 18.0, 22.0, and 54.0 mol %. They were comonomer compositionally fractionated using chloroform/n-heptane mixed solvent at ambient temperature. The fractionation of P(3HB-co-18.0 mol %3HH) and P(3HB-co-22.0 mol % 3HH), which could not be carried out effectively at room temperature, were refractionated at 70 °C in the mixed solvent. Fractions with different 3HH unit content in a wide range (from 4.4 to 80.7 mol %) were obtained. By use of these fractions with narrow compositional distribution, the comonomer composition dependence of thermal properties was investigated by differential scanning calorimetry. The melting point (Tm) and heat of fusion (∆H) decreased as the 3HH unit content increased in the range of low 3HH content (70 mol %). The minimum Tm and ∆H values were found to exist at 3HH unit content of about 60 mol %. The glass transition temperature (Tg) decreased linearly with the increase of 3HH unit content. The values of Tm, ∆H, and Tg of P(3HB-co-3HH)s were compared with those of poly(3-hydroxybutyrate-co-3-hydroxyvalerate), poly(3hydroxybutyrate-co-3-hydroxypropionate), and poly(3-hydroxybutyrate-co-4-hydroxybutyrate), and the effects of comonomer types on the thermal properties were revealed. Introduction Recently natural polyester poly(3-hydroxybutyrate) [P(3HB)], which is biosynthesized by various kinds of bacteria in the environment as carbon and energy storage compounds, has attracted much attention from the viewpoints of science and industry, due to its inherent biodegradability and the serious environmental problem caused by plastics wastes. P(3HB) possesses a melting point close to that of polypropylene (PP), better oxygen barrier property than those of poly(ethylene terephthalate) and PP, and mechanical properties resembling those of polystyrene and PP. However, its brittleness and narrow processing temperature window limit its application.1 To overcome the inferior properties of P(3HB), a variety of P(3HB) copolymers have been synthesized and investigated, such as, poly(3-hydroxybutyrate-co-3-hydroxyvalerate) [P(3HB-co-3HV)],2 poly(3-hydroxybutyrate-co-3-hydroxypropionate) [P(3HB-co-3HP)],3-5 poly(3-hydroxybutyrate* To whom correspondence should be addressed. E-mail: yinoue@ bio.titech.ac.jp. Tel.: +81-45-924-5794. Fax: +81-45-924-5827. † Tokyo Institute of Technology. ‡ RIKEN Institute. § Kaneka Corporation. | Tsinghua University.

co-4-hydroxybutyrate) [P(3HB-co-4HB)],6,7 poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) [P(3HB-co-3HH)],8-12 and poly(3-hydroxybutyrate-co-3-hydroxyoctanoate) [P(3HH-co3HO)].13 It has been known that some of the bacterially synthesized copolymers, such as, P(3HB-co-3HV),14-17 P(3HB-co3HP),18-22 and P(3HB-co-4HB)23 were mixtures of random copolymers having different comonomer unit compositions. These copolymers have been comonomer compositionally fractionated through a process of solvent/nonsolvent precipitation, such as chloroform/n-heptane mixed solvent16-22 or acetone/water mixed solvent,14 and a series of fractions with different comonomer unit composition have been obtained. The relationship between the compositional distribution and the physical properties, such as melting temperature, crystallization behavior, thermal properties, biodegradability, etc., have been studied.14-23 Recently, the comonomer unit composition and distribution of bacterial P(3HB-co-3HH)s with average 3HH contents of 7.4, 18.0, and 22.0 mol % have been investigated.24 They were fractionated by chloroform/n-heptane mixed solvent at ambient temperature and found to have a wide compositional distribution. The P(3HB-co-3HH) with 7.4 mol % 3HH can be fractionated effectively in the studied condition. However,

10.1021/bm0200581 CCC: $22.00 © 2002 American Chemical Society Published on Web 07/19/2002

1072

Biomacromolecules, Vol. 3, No. 5, 2002

the fractionation of P(3HB-co-3HH)s with 18.0 and 22.0 mol % 3HH unit was difficult under these conditions. Two melting peaks appeared in the differential scanning calorimetry (DSC) thermograms. The DSC measurements at different heating rates indicated that neither of these two melting peaks was caused by recrystallization during DSC heating process. The appearance of two melting peaks in the same DSC thermogram suggested the ineffectiveness of compositional fractionation. In this paper, we report the results of continuous work on the fractionation of P(3HB-co-3HH)s. The P(3HB-co-3HH)s with 18.0 and 22.0 mol % 3HH unit are successfully fractionated at 70 °C using chloroform/n-heptane mixed solvent, although these two samples had failed to fractionate effectively at ambient temperature as reported in our previous paper.24 In addition, in this work, two new P(3HB-co-3HH)s with 13.8 and 54.0 mol % 3HH unit content, which were biosynthesized by Aeromonas hydrophila 4AK4 and recombinant strains of Ralstonia eutropha PHB-4, respectively, were fractionated using chloroform/n-heptane mixed solvent at ambient temperature. Further, the physical properties of the as-produced original and all fractionated samples were investigated and the relationships between the properties and comonomer unit composition will be discussed. At last, the values of Tm, ∆H, and Tg of P(3HB-co-3HH) will be compared with those of P(3HB-co-3HV), P(3HB-co-3HP), and P(3HB-co-4HB) in order to reveal the effects of comonomer unit types on the properties. Experimental Part Materials. P(3HB-co-13.8 mol % 3HH) was biosynthesized by Aeromonas hydrophila 4AK4 using lauric acid as a sole carbon source.12 P(3HB-co-18.0 mol % 3HH) and P(3HB-co-22.0 mol % 3HH) were biosynthesized with Ralstonia eutropha PHB-4 pJRDEE32d13 using coconut oil as the carbon source.24 P(3HB-co-54.0 mol % 3HH) was biosynthesized by recombinant strains of Ralstonia eutropha PHB-4 fed hexanoate and octanoate. 9 The details of biosyntheses of these P(3HB-co-3HH) samples have been reported elsewhere.9,12,24 Fractionation of As-Produced P(3HB-co-3HH)s. Fractionation of the as-produced P(3HB-co-3HH) samples were carried out in chloroform/n-heptane mixed solvents as follows: P(3HB-co-3HH) was dissolved in chloroform at a polymer concentration of 10 g/L, and a predetermined amount of n-heptane was carefully added to the solution under gentle agitation at ambient temperature. When the precipitate was visually deposited, the mixed solution was kept at ambient temperature for 24 h, and the precipitated fraction was obtained by centrifugation. The precipitate was dried in vacuo at ambient temperature. This procedure was repeated until the addition of a large amount of n-heptane caused no appreciable precipitation. The residual copolyester in the supernatant solution was recovered by vaporization of the solvent. As referred to in the previous work,24 the fractionation at ambient temperature is not effective for samples P(3HB-co18.0 mol % 3HH) and P(3HB-co-22.0 mol % 3HH),

Feng et al.

evidenced by the observation of two melting points for some fractions. Here, these fractions, which possessed two melting points, were merged again and refractionated. The merged fractions were initially dissolved in hot chloroform at a polymer concentration of 5 g/L. Then, the solution was carefully added to a large amount of n-heptane (300 mL of n-heptane per 100 mL of solution) at 70 °C. The solution was kept at 70 °C under gentle agitation for 24 h. As a result, some polymer precipitated from the solution. The precipitate was dissolved again in chloroform, and the solution (1 wt %) was cast in a Teflon dish. The supernatant was concentrated and cast to film. All of the films were dried in the vacuum oven at ambient temperature for 48 h before thermal analysis. 1 H NMR Measurement. To determine the 3HH-unit composition of P(3HB-co-3HH)s, 270 MHz 1H NMR spectra were recorded on a JEOL GSX-270 spectrometer at 30 °C in CDCl3 solutions with a 4.5 µs pulse width (45° pulse angle), a 5 s pulse repetition time, 2500 Hz spectral width, 32K data points, and 16 FID accumulations. Characterization by Gel Permeation Chromatography (GPC). Molecular weights were characterized by a Tosoh HLC-8020 GPC system with TSK GEL G2000Hxl and GMHxl columns. The concentration of copolymer in chloroform is 1 mg/mL. The injection volume is 100 µL. Chloroform was used as the eluent at a flow rate of 1.0 mL/ min. Polystyrenes with narrow molecular-weight distribution were used as standards to calibrate the GPC elution curve. Number-average molecular weight (Mn), weight-average molecular weight (Mw), and polydispersity (Mw/Mn) were calculated from the GPC data using a SC-8010 data processor. Measurement of DSC. Calorimetric measurements of the copolyesters were carried out on a Seiko DSC-220 assembled with SSC-580 thermal controller. A 3-5 mg sample was encapsulated in an aluminum pan. Before measurement, the samples were kept at ambient temperature for more than 3 weeks. The DSC thermogram was recorded from -50 to 200 °C at a scanning rate of 20 °C/min (first heating run). Tm was taken as the summit of melting peak and ∆H was calculated from the area of the endotherm peak. After the first heating, the sample was quenched to -100 °C with liquid N2. Then the sample was heated again from -100 to 200 °C (second heating run) and Tg was taken as the summit of the peak of the DDSC curve (the differentiated DSC curves). Results and Discussion The samples of P(3HB-co-3HH)s with 3HH-unit content of 13.8, 18.0, 22.0, and 54.0 mol % were denoted as A, B, C, and D, respectively. The composition of comonomer unit in the copolymer was determined from the relative integrated intensities of the proton resonances of the 3HB and 3HH repeating units in the 1H NMR spectrum. The as-biosynthesized P(3HB-co-3HH)s, which are denoted “original” samples, were fractionated with a chloroform/n-heptane mixed solvent at ambient temperature. The characteristics of the original and the obtained fractions from samples A and D are listed in Tables 1 and 2, respectively.

Biomacromolecules, Vol. 3, No. 5, 2002 1073

Comonomer Compositional Distribution Table 1. Results of Fractionation at 25 °C for P(3HB-co-13.8 mol % 3HH) Sample A concn of fraction n-heptane, no. % wt % original a1 a2 a3 a4 a5 a6 a7 a8 a9a

54.5 56.5 57.5 58.5 59.5 60.5 61.5 62.5

100 6.9 7.5 21.7 30.4 16.0 7.7 5.1 3.1 1.6

3HH content, mol % 10-5Mn Mw/Mn 13.8 9.2 10.1 12.3 14.3 13.9 13.8 14.9 14.7 17.1

2.85 2.95 4.13 5.33 4.45 2.44 1.71 1.22 1.00 0.80

2.01 2.05 1.77 1.44 1.30 1.51 1.42 1.45 1.54 1.64

Tm, °C

Tg, °C

110.7 130.3 125.9 118.5 111.0 106.9 109.4 107.5 107.0 105.6

1.1 2.5 4.4 4.7 3.5 3.3 3.0 3.0 2.9 2.8

a Final fraction remained in the supernatant after separation of a8 fraction.

Table 2. Results of Fractionation at 25 °C for P(3HB-co-54.0 mol % 3HH) Sample D concn of 3HH fraction n-heptane, content, Tm, no. % wt % mol % 10-4Mn Mw/Mn °C original d1 d2 d3 d4b

68.0 72.0 76.0

100 9.9 39.6 32.4 12.0

54.0 50.5 53.1 56.3 66.3

3.6 3.4 1.9 4.3 3.4

4.3 5.1 1.6 1.7 2.1

41.2 44.4 a a 26.7

Figure 1. Weight percent of fractions vs 3HH mol % for P(3HB-co13.8 mol % 3HH) (sample A).

Tg, °C -12.7 -12.8 -10.6 -13.8 -15.5

a Not detected. b Final fraction remained in the supernatant after separation of d3 fraction.

The data shown in Table 1 indicate that the fractionation of sample A initially depended on the 3HH unit content, which gradually increased with an increasing concentration of heptane in the mixed solvent. From the fourth fraction, it seemed that the copolymer was also fractionated based on the molecular weight as well as 3HH content. The molecular weight of the following fractionated P(3HB-co-3HH) gradually decreased with an increasing concentration of n-heptane in the mixed solvent. On the other hand, the data shown in Table 2 revealed that the fractionation of sample D proceeded mainly as a result of difference in 3HH unit content and the 3HH content of the fractionated P(3HB-co-3HH) gradually increased with the fractionation process. According to a previous report,18 the comonomer-unit compositional fractionation of P(3HB-co-3HV) and P(3HBco-3HP) depends on the 3HV or 3HP unit content. The solubility of the P(3HB) copolymers depends on both the length of side chain and composition of comonomer units. The higher the content of the monomer units with longer side chain is, the higher is the solubility of copolyesters in the mixed solvent with high n-heptane content. In our cases, the fractionation of P(3HB-co-3HH) basically followed this regulation. However, Table 1 indicated that when the distribution of comonomer-unit composition of P(3HB-co3HH) was too narrow to fractionate by 3HH unit content (fractions a4-a8 of sample A), it can also be fractionated by molecular weight. That is, sample A with 13.8 mol % 3HH content was initially fractionated by the 3HH content and then by the molecular weight. Figures 1 and 2 showed the comonomer compositional distribution of samples A and D. Sample A was separated

Figure 2. Weight percent of fractions vs 3HH mol % for P(3HB-co54.0 mol % 3HH) (sample D).

into eight fractions, and sample D was only separated into three fractions. Samples A and D had a compositional distribution from 9.2 to 17.1 mol % and from 50.5 to 56.3 mol %, respectively. DSC thermograms were used to investigate the effects of compositional distribution on the physical properties of P(3HB-co-3HH). Figures 3 and 4 depict the DSC results of the fractions as well as their original samples A and D, respectively. For the fractions of sample A, the melting point from fraction a1 to a4 shows a tendency to shift to lower temperature with an increase in the 3HH content (Figure 3). However, little change was observed after the fraction a5 because they had almost the same 3HH content (Figure 3). This result indicated that the melting point of P(3HB-co3HH) was mainly decided by the comonomer composition but not by molecular weight. Figure 4 shows the DSC results of sample D, P(3HB-co54.0 mol %3HH), and their fractions, which have higher 3HH content than the samples A-C. Only the first fraction had a sharp melting point, which was almost the same as that of the original. No melting peak was observed for other fractions with 3HH unit content ranging from about 55 to 65 mol %, indicating fractions d2-d4 were amorphous copolymers. Therefore, the melting behavior of the original

1074

Biomacromolecules, Vol. 3, No. 5, 2002

Figure 3. DSC thermograms of original P(3HB-co-13.8 mol % 3HH) (sample A) and its series of fractions observed at a heating rate 20 °C/min (first heating scan).

Figure 4. DSC thermograms of original P(3HB-co-54.0 mol % 3HH) (sample D) and its series of fractions observed at a heating rate 20 °C/min (first heating scan).

was mainly determined by the d1 fraction, which contributed to only about 10 wt % of original as-produced sample D. In the previous paper,24 samples B [P(3HB-co-18.0 mol %3HH)] and C [P(3HB-co-22.0 mol %3HH)] had been fractionated using chloroform/n-heptane mixed solvent at ambient temperature. The DSC thermograms of some fractions showed two melting peaks, and these peaks hardly changed with 3HH content. None of these two peaks were found to be caused by recrystallization. The appearance of the two peaks may come from the following reasons. One is that samples B and C cannot be fractionated well by the solvent/nonsolvent fractionation process. The obtained fractions are still mixtures of P(3HB-co-3HH) copolymers with different 3HH content. The second is two types of crystal lattices (P(3HB) type and P(3HH) type) are formed simultaneously in P(3HB-co-3HH) with 3HH content about 20 mol % leading to the emergence of two separate melting peaks in the DSC endothermic curve. To clear these possibilities, the fractions with two peaks of fusion were combined again and refractionated at 70 °C. As the two melting peaks appeared at about 50 and 110 °C in the DSC

Feng et al.

traces, a temperature of 70 °C was taken as the fractionation temperature. The results of refractionations are listed in Tables 3 and 4. Two factions were obtained from each sample, the precipitated fraction had a lower 3HH unit content while the supernatant fraction had a higher 3HH unit content than the original before refractionation. Their DSC thermograms are shown in Figures 5 and 6. Before refractionation, the samples showed broad melting endotherms that had the peak tops at 54.3 and 116.2 °C for sample B, 60.6 and 109.3 °C for sample C. After refractionation, the precipitated fraction from sample B had a single peak at 117.9 °C and that from sample C at 113.3 °C. The supernatant fraction from sample B possessed a single peak at 53.7 °C and that from sample C at 42.8 °C. This result confirmed that the samples before refractionation were mixture of the copolymers with different 3HH unit content. It also indicated that a fractionation procedure conducted at higher temperature was effective to fractionate these P(3HB-co-3HH) samples. Thus far, five P(3HB-co-3HH) samples with 3HH content of 7.4, 13.8, 18.0, 22.0, and 54.0 mol % have been compositionally fractionated.24 Now, P(3HB-co-3HH) samples in a broad range of 3HH content from 4.4 to 80.7 mol %, but with narrow compositional distribution, were studied. On the basis of such a number of P(3HB-co-3HH) fractions, it becomes possible to get the full image of the relationship between the comonomer composition and its properties. Thus, we can investigate the effect of 3HH unit content on melting point (Tm), glass transition point (Tg), and heat of fusion (∆H) in a wide composition range as shown in Figures 7, 8, and 9, respectively. Figure 7 shows that the value of Tm significantly decreased at first with the increase of 3HH content. At 3HH content of about 60 mol %, a minimum melting temperature was observed. Then, it remained constant or increased a little with the increase of 3HH content. Figure 8 shows the relationship between ∆H and 3HH unit content. The tendency of ∆H change was similar to that of Tm. It suggested that minimum crystallinity was a result of the disturbance of the crystalline lattice caused by the second repeating unit. Figure 9 depicted the glass transition temperature as a function of 3HH content. The value of Tg decreased linearly with the increase of 3HH unit content. In the previous papers, the results on the thermal properties have been reported for fractionated samples of P(3HB-co3HV),17,25 P(3HB-co-3HP),18,21 and P(3HB-co-4HB).26 To clarify the effects of the second monomer unit on the structure-property relationships for a series of P(3HB) copolymers, that is, P(3HB-co-3HV), P(3HB-co-3HP), P(3HBco-4HB), and P(3HB-co-3HH), the values of Tm, ∆H, and Tg plotted against the content of the second monomer units (3HV, 3HP, 4HB, 3HH) for these copolymers are shown in Figures 10, 11, and 12, respectively. As shown in Figure 10, each P(3HB-co-3HV) fraction has melting point despite its comonomer composition. As we know, the 3HB-rich and 3HV-rich P(3HB-co-3HV)s form P(3HB)- and P(3HV)-type crystalline lattices, respectively.27,28 Also, P(3HB-co-3HV) with the 3HV unit content between ca. 45 and 55 mol % was considered to be the

Biomacromolecules, Vol. 3, No. 5, 2002 1075

Comonomer Compositional Distribution Table 3. Results of Refractionated at 70 °C for P(3HB-co-35.2 mol % 3HH) Sample B

before refractionation after refractionation a

precipitation supernatant

wt %a

3HH content, mol %

10-4Mn

Mw/Mn

Tm, °C

Tm, °C

Tg, °C

100 81.9 1.2

35.2 29.8 69.8

1.78 1.90 1.74

2.09 2.07 2.16

54.3 53.7

116.2 117.9

-4.6 -2.2 -6.3

The total weight of samples after refractionation was not 100% due to loss during the refractionation and recovery processes.

Table 4. Results of Refractionated at 70 °C for P(3HB-co-24.8 mol % 3HH) Sample C

before refractionation after refractionation a

precipitation supernatant

wt %a

3HH content, mol %

10-4Mn

Mw/Mn

Tm, °C

Tm, °C

Tg, °C

100 86.6 2.1

24.8 19.5 80.7

1.62 1.91 b

1.96 2.41 b

60.6

109.3 113.3

0.2 1.2 -18.6

42.8

See the footnote of Table 3. b Not detected.

Figure 5. DSC thermograms of before and after refractionation of P(3HB-co-35.2 mol % 3HH) observed at a heating rate 20 °C/min (first heating scan).

Figure 7. Plots of melting temperature (Tm) vs 3HH content (mol %) for fractionated P(3HB-co-3HH)s.

Figure 8. Plots of heat of fusion (∆H) vs 3HH content (mol %) for fractionated P(3HB-co-3HH)s. Figure 6. DSC thermograms of before and after refractionation of P(3HB-co-24.8 mol % 3HH) observed at a heating rate 20 °C/min (first heating scan).

intermediate point of critical 3HV content range for crystalline structure change from the P(3HB) to the P(3HV) homopolymer type.29 In this range, two crystalline lattice types coexist.17 On the other hand, none of P(3HB-co-3HP), P(3HB-co-4HB), and P(3HB-co-3HH) has a melting point

in the intermediate content range of the second monomer. At about 40-60 mol % of the second monomer content, a minimum melting temperature exists, which is consistent with the relationship between ∆H and the content of the second monomer unit (Figure 11). It seems that minimum crystallinity exists due to the disturbance to the crystalline lattice caused by the second repeating unit. The Tm values of all of these four copolymers show a similar tendency to decrease as the content of the second monomer unit increases,

1076

Biomacromolecules, Vol. 3, No. 5, 2002

Figure 9. Plots of glass transition temperature (Tg) vs 3HH content (mol %) for fractionated P(3HB-co-3HH)s.

Figure 10. Plots of melting temperature (Tm) vs second monomer (3HH, 3HV, 3HP, 4HB) content (mol %) for fractionated copolymers: b, P(3HB-co-3HH) fraction; [, P(3HB) homopolymer; ], P(3HB-co4HB) fraction; 0, P(3HB-co-3HP) fraction; 4, P(3HB-co-3HV) fraction.

and then increase when the content of the second monomer is higher than 60 mol %. The similar change of the Tm value of P(3HB-co-3HP), P(3HB-co-4HB), and P(3HB-co-3HH) implies that the effect of the chemical structure (main and side chain structure) of the second monomer unit on the Tm values of uncocrystalline copolymer is not very obvious. P(3HB-co-3HV) can cocrystallize in the intermediate comonomer composition range.17 Therefore, the value of ∆H of P(3HB-co-3HV) is quite high even in the intermediate 3HV unit content range. However, the values of ∆H of P(3HB-co-3HP), P(3HB-co-4HB), and P(3HB-co-3HH) in the intermediate second unit content range are almost zero, and so they are amorphous in this content range, no cocrystallization occurs. Figure 12 shows that the Tg values of all of the four copolymers decrease linearly with the increase of the second monomer unit content, indicating that the segmental mobilities of copolymer chains are directly related to their comonomer compositions. The slopes of line of Tg-comonomer content plots of P(3HB-co-3HV), P(3HB-co-3HP), and P(3HB-co-3HH) are similar, though the Tg values of P(3HBco-3HP) and P(3HB-co-3HH) are slightly lower than those

Feng et al.

Figure 11. Plots of heat of fusion (∆H) vs second monomer (3HH, 3HV, 3HP, 4HB) content (mol %) for fractionated copolymers: b, P(3HB-co-3HH) fraction; [, P(3HB) homopolymer; ], P(3HB-co-4HB) fraction; 0, P(3HB-co-3HP) fraction; 4, P(3HB-co-3HV) fraction.

Figure 12. Plots of glass transition temperature (Tg) vs second monomer (3HH, 3HV, 3HP, 4HB) content (mol %) for fractionated copolymers: b, P(3HB-co-3HH) fraction; [, P(3HB) homopolymer; ], P(3HB-co-4HB) fraction; 0, P(3HB-co-3HP) fraction; 4, P(3HBco-3HV) fraction.

of P(3HB-co-3HV) at the same comonomer content. However, the change of the Tg values with the 4HB content is larger than that of the other three copolymers. It can be explained by the differences in structures of main chain and side chain of these copolymers. P(3HB-co-3HV), P(3HBco-3HP), and P(3HB-co-3HH) have the same main chain structure and different side chains. Compared with the side chain of the second monomer unit of P(3HB-co-3HV), that of P(3HB-co-3HP) has no side chain. Therefore, the segment mobility of main chain of P(3HB-co-3HP) should be higher than that of P(3HB-co-3HV), inducing the Tg values of P(3HB-co-3HP)s lower than those of P(3HB-co-3HV)s. On the other hand, the side chain of the second monomer unit, namely, 3HH, of P(3HB-co-3HH) is longer than that of P(3HB-co-3HV), and the distance between molecules of P(3HB-co-3HH) is farther than that of P(3HB-co-3HV). These differences result in the increase of segmental mobility of P(3HB-co-3HH) molecules, inducing the Tg values of P(3HB-co-3HH) lower than that of P(3HB-co-3HV) at the

Comonomer Compositional Distribution

same comonomer content. The second monomer unit, the 4HB unit, of P(3HB-co-4HB) has no side chain and its main chain is one carbon more than those of other monomer units. Due to these facts resulting in the good segmental mobility of P(3HB-co-4HB) molecular chain, the Tg values of P(3HBco-4HB) are lower than those of other copolymers at the same comonomer content.

Biomacromolecules, Vol. 3, No. 5, 2002 1077

of the comonomer, the values of Tg of all of the four copolymers decrease linearly with the increase of the second monomer unit content. However, there is cocrystallization in intermediate comonomer composition range for P(3HBco-3HV), while P(3HB-co-3HP), P(3HB-co-4HB), and P(3HBco-3HH) are amorphous in this range. References and Notes

Conclusion Bacterial poly(3hydroxybutyrate-co-3-hydroxyhexanoate) [P(3HB-co-3HH)] with 3HH unit content of 13.8, 18.0, 22.0, and 54.0 mol % was fractionated into a number of fractions of P(3HB-co-3HH) with quite different comonomer composition using chloroform (solvent)/n-heptane (nonsolvent) mixed solvent at ambient temperature. Some fractions of P(3HB-co-18.0 mol % 3HH) and P(3HB-co-22.0 mol % 3HH) that could not be fractionated effectively at ambient temperature, were successfully refractionated at 70 °C, and two fractions with quite different 3HH unit content were obtained. It was found that fractionation mainly depended on the 3HH unit content. When the comonomer compositional distribution of P(3HB-co-3HH) sample was narrow, the fractionation by molecular weight difference was dominated. The results also suggested that as-produced P(3HBco-13.8 mol % 3HH) and P(3HB-co-54.0 mol % 3HH) had relatively narrow 3HH compositional distribution, while P(3HB-co-18.0 mol % 3HH) and P(3HB-co-22.0 mol % 3HH) had broad 3HH compositional distribution. However, it is true that all of the as-produced samples are mixtures or blends of P(3HB-co-3HH)s with different comonomer unit compositions. On the basis of these fractions with relatively narrow compositional distribution, the relationship between composition and thermal properties was revealed for the first time in a wide composition range. The melting behavior of fractions of P(3HB-co-3HH) was remarkably different from that of an original unfractionated P(3HB-co-3HH) sample. A minimum in Tm and ∆H vs 3HH content plots was observed at 3HH content about 60 mol %, while the value of Tg decreased linearly with increase of 3HH unit content. Combined with the results of other P(3HB) copolymers P(3HB-co-3HV), P(3HB-co-3HP), and P(3HB-co-4HB), it is clear that the values of Tm and ∆H of P(3HB) copolymers decrease at first with the increase of the content of the second monomer unit and then increase when the content of the second monomer is higher than 60 mol %. Despite the type

(1) Nayak, P. L. J. Macromol. Sci., ReV. Macromol. Chem. Phys. 1999, C39, 481. (2) Doi, Y.; Tamaki, A.; Kunioka, M.; Soga, K. Appl. Microbiol. Biotechnol. 1988, 28, 330. (3) Nakamura, S.; Kunioka, M.; Doi, Y. Macromol. Rep. 1991, A28 (suppl. 1), 15. (4) Hiramitsu, M.; Doi, Y. Polymer 1993, 34, 4782. (5) Shimamura, E.; Scandola, M.; Doi, Y. Macromolecles 1994, 27, 4429. (6) Kunioka, Y.; Nakamura, Y.; Doi, Y. Polym. Commun. 1989, 22, 694. (7) Doi, Y.; Kunioka, Y.; Nakamura, Y.; Soga, K. Macromolecules 1988, 21, 2722. (8) Doi, Y.; Kitamura, S.; Abe, H. Macromolecules 1995, 28, 4822. (9) Kichise, T.; Fukui, T.; Yoshida, Y.; Doi, Y. Biol. Macromol. 1999, 25, 69. (10) Park, S. J.; Ahn, W. S.; Green, P. R.; Lee, S. Y. Biomacromolecules 2001, 2, 248. (11) Lee, S. H.; Oh, D. H.; Ahn, W. S.; Lee, Y.; Choi, J.; Lee, S. Y.; Biotechnol. Bioeng. 2000, 67, 240. (12) Chen, G. Q.; Zhang, G.; Park, S. J.; Lee, S. Y. Appl. Microbiol. Biotechnol. 2001, 57, 50. (13) Timm, A.; Byrom, D.; Steinbu¨chel, A. Appl. Microbiol. Biotechnol. 1990, 33, 296. (14) Mitomo, H.; Morishita, N.; Doi, Y. Macromolecules 1993, 26, 5809. (15) Mitomo, H.; Morishita, N.; Doi, Y. Polymer 1995, 36, 2573. (16) Yoshie, N.; Fujiwara, M.; Kasuya, K.; Abe, H.; Doi, Y.; Inoue, Y. Macromol. Chem. Phys. 1999, 200, 977. (17) Wang, Y.; Yamada, S.; Asakawa, N.; Yamane, T.; Yoshie, N.; Inoue, Y. Biomacromolecules 2001, 2, 1315. (18) Cao, A.; Ichikawa, M.; Kasuya, K.-I.; Yoshie, N.; Asakawa, N.; Inoue, Y.; Doi, Y.; Abe, H. Polym. J. 1996, 28, 1096. (19) Cao, A.; Ichikawa, M.; Kasuya, K.-I.; Yoshie, N.; Inoue, Y. Macromol. Chem. Phys. 1997, 198, 3539. (20) Cao, A.; Y Arai,; Yoshie, N.; Kasuya, K.-I.; Doi, Y.; Inoue, Y. Polymer 1999, 40, 6821. (21) Cao, A. K Kasuya, -I.; Abe, H.; Doi, Y.; Inoue, Y. Polymer 1998, 39, 4801. (22) Arai, Y.; Cao, A.; Yoshie, N.; Inoue, Y. Polym. Int. 1999, 48, 1219. (23) Shi, F. Y.; Ashby, R. D.; Gross, R. A. Macromolecules 1997, 30, 2521. (24) Watanabe, T.; He, Y.; Fukuchi, T.; Inoue, Y. Macromol. Biosci. 2001, 1, 75. (25) Kamiya, N.; Yamamoto, Y.; Inoue, Y.; Chujo, R.; Doi, Y. Macromolecules 1989, 22, 1676. (26) Ishida, K.; Wang, Y.; Inoue, Y. Biomacromolecules 2001, 2, 1285. (27) Kunioka, M.; Tamaki, A.; Doi, Y. Macromolecules 1989, 22, 694. (28) Kamiya, N.; Sakurai, M.; Inoue, Y.; Chujo, R.; Doi, Y. Macromolecules 1991, 24, 2178. (29) Yamada, S.; Wang, Y.; Asakawa, N.; Yoshie, N.; Inoue, Y. Macromolecules 2001, 34, 4659.

BM0200581