Elsevier

Polymer

Volume 53, Issue 5, 28 February 2012, Pages 1049-1053
Polymer

Polymer communication
Effect of the degree of branching on the glass transition temperature of polyesters

https://doi.org/10.1016/j.polymer.2012.01.020Get rights and content

Abstract

Glass transition temperature dependence on the branching degree can be empirically estimated by excluding additional effects on this parameter as molecular weight distribution, end group interactions or crystallization. In this communication aliphatic–aromatic polyesters with a well defined degree of branching between 0% (linear) and 50% (hyperbranched) are investigated by differential scanning calorimetry. The hydrogen bonding effect of the OH-terminal groups was successfully extracted from the pure branching effect by protection of the end-functionalities. Fractionation of samples with variation of the branching degree and end-functionalities led to series of narrowly distributed molar masses. The dependence of the molecular weight on the glass transition temperature for different branching degrees was calculated and compared for polar and non-polar end groups.

Introduction

Hyperbranched (hb) polymers are functional materials, which can be easily prepared by “one pot” polymerization, which, in turn, is accompanied by reduced costs as compared to the dendrimers [1], [2], [3]. However, despite the fact that many hb structures were synthesized and comprehensively characterized, some uncertainties in their structure-property relations still exist. One of those is the relation between the branching and glass transition temperature, which is an essential parameter influencing the material properties.

To clarify this relation several investigations [2], [4], [5], [6] on differently branched polymers were performed but no apparent correlation between the branching and the glass transition temperature was obtained. Moreover, the experimental observations showed two different tendencies. The first tendency states that the glass transition temperature of highly branched polymers only slightly depends on the branching. This conclusion is supported by results of Wooley et al. [7], where similar glass transition temperatures were found for the linear, hb and dendritic aromatic polyesters having identical repeating unit, as well as number and nature of functional groups. Similar observation was done by two other groups. Ramakrishnan et al. [8] compared the thermal behavior of linear, branched and “kinked” PET and indicated similar glass transition temperatures for all three types of architectures. Furthermore, Finelli et al. [9] reported on the comparison between the linear and branched poly(butylenes isophthalates), confirming the lack of the effect of branching on the glass transition temperature. However, it should be mentioned that the glass transition temperature behavior of those polyesters was additionally influenced by their partially crystalline nature where the crystallinity acts similarly to the crosslinking and causes a raise of the glass transition temperature through its restrictive effect on the segmental motion of the polymer chains.

The second tendency implies that the branching causes a decrease of the glass transition temperature. This trend was observed in hb polyetherimides [10], indicating a significantly enhanced segmental mobility with branching. A comparative study on hb and linear polyimides [11] showed, that the introduction of branching leads to glass transition temperature decrease from 266 °C for linear polyimide down to 222 °C for amino-terminated- and 231 °C for the anhydride-terminated hb polyimides.

Additionally, besides those described theories there is an interesting study made by Zhu and co-workers [4] on the amorphous poly[3-methyl-3-(hydroxymethyl)oxetane] revealing an increase of the glass transition temperature with branching, which passes through a maximum at DB = 27% and then decreases sharply. The branching leads actually to two main effects – restriction of the segmental mobility, which leads to increasing glass transition temperatures and simultaneous increase of end groups number and free volume respectively which decreases the Tg.

The observation of these controversial tendencies gave us motivation to search for the pure effect of branching on Tg minimizing overlapping effects. In the above mentioned studies the branching effect is accompanied by the effect of end group interactions [6], different amount of end groups per monomer unit [10], [11] or crystalzation [4], [9]. In order to extract the branching influence on Tg we focused on amorphous aromatic-aliphatic polyesters with hb, gradually branched and linear architecture and controlled amount of functional groups. These polymers are of identical chemical origin possessing two types of functional groups with different polarity. Additionally, fractionation was carried out in order to clarify the molar mass effect on Tg at different degree of branching.

Section snippets

Samples

The synthesis of the samples is described in a previous work and schematically shown in Scheme 1 [12]. The following sample abbreviation was used: OH- and SY- corresponds to the type of functional groups (OH = hydroxyl group and SY = t-butyldimethylsilyl group). The number indicates the degree of branching (DB) which is calculated to be 0 for linear and 50 for the hb samples.

Fractionation

Fractions of narrow polydispersity were obtained using preparative elution fractionation similar to previously described

Influence of the branching degree

The DSC analysis for both OH- and SY-terminated samples was performed showing no crystallinity and, thus, a fully amorphous nature of the polymers. This fact allows obtaining the pure influence of branching on glass transition temperature.

The glass transition temperature data of SY-functionalized polyesters were listed in our previous work [12]. Fig. 1 illustrates a gradual decrease of the glass transition temperature with the branching of the SY-terminated polymers.

Due to the identical

Conclusions

It was demonstrated that the glass transition temperature depends clearly on the polymer branching architecture but it is strongly influenced by the type of the functional groups. In case of non-polar functional polymer, 16 °C difference in the Tg value between hb and linear sample was found, whereas the dependence of the Tg from DB has a linear character. For polymers, capable of hydrogen bonding the impact of architecture becomes negligibly small and is completely overlapped by the influence

Acknowledgements

This work was supported by the Dutch Polymer Institute, Eindhoven, The Netherlands.

References (19)

  • M.G. McKee et al.

    Prog Polym Sci

    (2005)
  • K.L. Wooley et al.

    Polymer

    (1994)
  • J. Peter et al.

    Eur Polym J

    (2009)
  • A. Lederer et al.

    J Chromatogr A

    (2002)
  • H.F. Gao et al.

    Macromolecules

    (2006)
  • B.I. Voit et al.

    Chem Rev

    (2009)
  • Q. Zhu et al.

    J Phys Chem B

    (2009)
  • D. Foix et al.

    Macromol Mat Eng

    (2012)
  • K.L. Wooley et al.

    Macromolecules

    (1993)
There are more references available in the full text version of this article.

Cited by (62)

  • On the impact of the type of anion on the properties of solid-state electrolytes

    2022, Polymer
    Citation Excerpt :

    The Tg of the native polymer and all polymer complexes obtained from the thermal analysis is summarized in Table 2. While the presence of the sodium salts as additives in the polymer is expected to have an effect on the Tg [34], differences in the degree of polymerization and/or cross-linking of the polymers will also affect the obtained values [35–38], making a fair comparison based on absolute values difficult. For this reason, rather than evaluating the magnitude of the different Tg values found, it is only relevant to acknowledge that all the systems studied are in their rubbery state at room temperature.

View all citing articles on Scopus
View full text