Self-assembly behaviour of telechelic polyethylene glycol with triptycene termini capable of two-dimensional ordering

Abstract

The self-assembly of polymers into well-defined structures is of great interest in the design of functional materials. We have previously shown that telechelic polydimethylsiloxanes bearing 1,8,(13)-substituted triptycene termini form highly ordered “2D + 1D” structures, which significantly improve their rheological and thermal properties. In this study, to gain insight into the scope and limitations of this terminal-triptycene-driven polymer ordering, we investigated a new system based on crystalline polyethylene glycol (PEG). We synthesised telechelic PEGs with 1,4-, 1,8- and 1,8,13-substituted triptycenes (i.e., 1,4-, 1,8- and 1,8,13-Trip-PEGs) to examine how the substitution pattern of the triptycene termini influences polymer self-assembly. In water, 1,4- and 1,8-Trip-PEGs form hydrogels without long-range ordering, while 1,8,13-Trip-PEG forms a hydrogel with a well-defined “2D + 1D” structure. The critical gelation concentration decreases as the self-assembly ability of the terminal groups increases. In the solid state, the structures of 1,4- and 1,8-Trip-PEGs are dominated by PEG crystallisation. In contrast, 1,8,13-Trip-PEG forms a distinct ordered structure regardless of whether the PEG chains are melted or crystallised. These results demonstrate the strong ability of 1,8,13-substituted triptycene termini to induce structural ordering, even in crystalline polymers.

---

Introduction

The self-assembly of polymers into well-defined structures is of great interest in both fundamental research and polymer design for applications such as nanopatterning, directional transport and optics.^1–5 Approaches based on block copolymers with phase-separating segments have been widely used for this purpose, as they can provide periodic hetero-domains at the nanoscale.^1–5 Meanwhile, we have shown that 1,8- and 1,8,13-substituted triptycenes serve as a supramolecular scaffold^6–9 that promotes a wide variety of functional groups^6–9 and polymers^10–13 to align in a well-defined “two-dimensional (2D) + one-dimensional (1D)” order, which is due to the nested 2D hexagonal packing of the triptycene units. An unusual terminal-group effect for polymer ordering was observed when incorporating 1,8-substituted triptycene units at both termini of amorphous polydimethylsiloxane (PDMS):¹⁰ a telechelic PDMS (Fig. 1, 1,8-Trip-PDMS) forms a highly ordered “2D + 1D” structure, which greatly improves the rheological properties of the original PDMS. Moreover, we found that when one methoxy group is attached to the triptycene termini, the resulting telechelic PDMS (Fig. 1a, 1,8,13-Trip-PDMS) exhibits a further increase in the rheological and thermal properties, allowing for the formation of a free-standing film, which can retain its shape at high temperatures (ca. below 110 °C) without any chemical crosslinking.¹¹ To gain insight into the scope and limitations of the terminal-triptycene-driven polymer ordering, we investigated a new system with polyethylene glycol (PEG), a well-known hydrophilic crystalline polymer, as the main chain.^14,15 PEG is one of the most widely used water-soluble polymers, with a broad range of potential applications. In this context, the chemical modification of PEG and the molecular designs that promote structuring on a larger length scale have been extensively studied. Here we report the synthesis and characterization of a series of 1,8,(13)-triptycene-appended PEGs (Fig. 1b and c), and their self-assembly behaviour in water and in the solid state.

Fig. 1 (a) Schematic structure of previously reported 1,8,(13)-Trip-PDMS and (b) telechelic PEGs synthesised in this work. (c) Typical crystal structure of PEG.

Results and discussion

Synthesis and self-assembly behaviour of terminal modifiers

For the synthesis of the telechelic PEGs, we first prepared alkyne-appended 1,8-substituted and 1,8,13-substituted triptycenes (Fig. 2, 1,8-Trip and 1,8,13-Trip) by procedures similar to those used in our previous work on telechelic PDMSs (Scheme S1, SI),^10,11 and unambiguously characterized their chemical structures (Fig. S1–S20, SI). Cu-catalysed Huisgen cycloaddition reaction¹⁶ of 1,8-Trip or 1,8,13-Trip with azide-terminated PEG (M_n = ca. 10 kDa) (Fig. S21–S24, SI) in N,N-dimethylformamide (DMF) in the presence of CuSO₄·5H₂O and sodium ascorbate, gave the corresponding 1,8-Trip-PEG (M_n = 11 kDa, n = ca. 230) or 1,8,3-Trip-PEG (M_n = 11 kDa, n = ca. 230) (Fig. 2). In addition, 1,4-substituted triptycene-terminated PEG (Fig. 2, 1,4-Trip-PEG, M_n = 11 kDa, n = ca. 230) was prepared as a reference polymer, which lacks the specific “2D + 1D” structure-forming ability. These polymers were unambiguously characterized by ¹H NMR and IR spectroscopy as well as SEC analysis (Fig. S25–S33, SI).

Fig. 2 Synthesis of Trip-PEGs.

In differential scanning calorimetry (DSC), 1,4-Trip, upon heating, displays a broad endothermic peak at 36 °C and two exothermic peaks at 54 and 59 °C (Fig. 3a). The endotherm and exotherm are due to crystallisation and melting, respectively. Upon cooling from its isotropic melt, no crystallisation feature was observed. In contrast, 1,8-Trip and 1,8,13-Trip each clearly shows a pair of endothermic and exothermic peaks upon heating and cooling, respectively, which are due to the formation and melting of “2D + 1D” structures. Similar to our previous results,¹¹ the higher self-assembling ability of 1,8,13-Trip (Fig. 3c) should lead to it exhibiting higher melting (T_m = 227 °C) and crystallisation points (T_c = 218 °C) than 1,8-Trip (Fig. 3b, T_m = 123 °C, T_c = 116 °C).

Fig. 3 DSC profiles of (a) 1,4-Trip, (b) 1,8-Trip and (c) 1,8,13-Trip during a second heating/cooling cycle, measured at a scan rate of 10 °C min⁻¹ under N₂ flow (50 mL min⁻¹).

Fig. 4 shows the powder X-ray diffraction (XRD) profiles of the triptycene derivatives at 30 °C after cooling from their isotropic melts. 1,4-Trip shows multiple diffractions (Fig. 4a) typical of low molecular weight compounds. On the contrary, the XRD profiles of 1,8-Trip and 1,8,13-Trip (Fig. 4b and c) can be fully assigned as “2D + 1D” structures comprised of 2D hexagonal packing of the triptycene units with a lattice parameter of a = 0.80 nm, and 1D lamellar stacking with c = 1.71 nm for 1,8-Trip and 1.79 nm for 1,8,13-Trip (Fig. 4d). These values are typical for 1,8- and 1,8,13-alkoxylated triptycenes.^6–11 Notably, the diffractions of 1,8,13-Trip are narrower and extend up to the 005 (Fig. 4c), whereas those of 1,8-Trip are broader and observed only up to 002 (Fig. 4b), indicating the higher structural integrity of the 1,8,13-Trip assembly.¹¹

Fig. 4 Powder XRD patterns of (a) 1,4-Trip, (b) 1,8-Trip and (c) 1,8,13-Trip at 30 °C measured after cooling from their isotropic melts in glass capillaries with diameters of 1.5 mm. (d) A schematic illustration of the 2D hexagonal array and 1D lamellar structure formed during the assembly of 1,8-Trip and 1,8,13-Trip.

Aggregation behaviour of Trip-PEGs in water

As the structure of the Trip-PEGs is similar to that of other associative polymers consisting of a hydrophilic main chain with hydrophobic termini,^17,18 it is expected that they exhibit clouding behaviour in water on heating due to the dehydration of the PEG chains. Using aqueous solutions of the triptycene-terminated PEGs at a concentration of 0.1 wt%, we monitored the change in their transmittance at 500 nm upon heating (Fig. 5), the temperature at which the transmittance reached 50% was determined as the clouding point, i.e. the lower critical solution temperature (LCST).^19,20 Although the hydrophobicity of these triptycene units does not appear to differ significantly, the clouding points of 1,8-Trip-PEG (63 °C) and 1,8,13-Trip-PEG (60 °C) are much lower than that of 1,4-Trip-PEG (76 °C). This observation clearly indicates that the presence of terminal groups with high self-assembling ability can induce the aggregation of the PEG chains by dehydration, which is far superior to the terminal group association caused by hydrophobic effects. Interestingly, a difference in the terminal-group effect is also observed for 1,8-Trip-PEG and 1,8,13-Trip-PEG, with the latter being more sensitive to temperature changes.

Fig. 5 Transmittance–temperature curves of the aqueous solutions of (a) 1,4-Trip-PEG, (b) 1,8-Trip-PEG and (c) 1,8,13-Trip-PEG during heating (Conc.: 0.1 wt%).

We found that Trip-PEGs at a higher concentration form hydrogels (Fig. 6). Inversion tests allowed to determine the critical gelation concentration (CGC) of 1,4-, 1,8- and 1,8,13-Trip-PEGs to be ca. 13, 11 and 8 wt%, respectively (Fig. 6). Thus, the CGC of each polymer tends to be lower the higher the self-assembling ability of the terminal triptycene unit. While 1,4-Trip and 1,8-Trip-PEGs give transparent gels (Fig. 6a and b), the gel formed from 1,8,13-Trip-PEG is turbid (Fig. 6c). This indicates that aggregate structures large enough to scatter visible light are developed inside the 1,8,13-Trip-PEG gel. Powder XRD measurements (Fig. 7) confirmed the increased self-assembling ability of 1,8,13-Trip-PEG compared to 1,8-Trip-PEG, which is consistent with the observations seen for their respective terminal units. While no diffraction peaks could be observed for the 1,4-Trip-PEG gel (Fig. 7a), the 1,8,13-Trip-PEG gel showed a set of diffractions that can be assigned to a “2D + 1D” structure with a hexagonal lattice parameter of 0.80 nm and a layer spacing of 21.0 nm (Fig. 7c and d). For the 1,8-Trip-PEG gel, shoulder-like weak peaks can be barely detected, which may be due to diffractions from the (110) and (200) planes of 2D hexagonally assembled triptycene units (Fig. 7b).

Fig. 6 Critical gelation concentration and photographs of hydrogels (20 wt%) of (a) 1,4-Trip-PEG, (b) 1,8-Trip-PEG and (c) 1,8,13-Trip-PEG at 25 °C.

Fig. 7 Powder XRD patterns of 20 w% hydrogels of (a) 1,4-Trip-PEG, (b) 1,8-Trip-PEG and (c) 1,8,13-Trip-PEG at 25 °C in glass capillaries with diameters of 1.5 mm. The diffraction peaks are marked with asterisks (*). (d) A schematic illustration showing the 2D hexagonal array and 1D lamellar structure of a 1,8,13-Trip-PEG hydrogel.

Self-assembly behaviour of Trip-PEGs in the solid state

In DSC (Fig. 8), all Trip-PEGs show a similar phase transition behaviour with crystallisation (T_c) and melting (T_m) of the PEG chains at around 35 and 55 °C, respectively. The enthalpy changes associated with crystallisation (ΔH_c) and melting (ΔH_m) of 1,4-Trip-PEG and 1,8-Trip-PEG are comparable (Fig. 8a and b), while 1,8,13-Trip-PEG (Fig. 8c) gives relatively smaller values. Comparing the reported phase transition temperatures and enthalpy changes of a pristine PEG (T_c = 41 °C, ΔH_c = 180 J g⁻¹; T_m = 63 °C, ΔH_m = 180 J g⁻¹)^21,22 with those observed for Trip-PEGs, the presence of bulky triptycene units at the termini is considered to inhibit to some extent the crystallisation of the PEG chain. During the DSC measurements, no particular thermal events due to the assembly of the terminal triptycene units can be observed.

Fig. 8 DSC profiles of (a) 1,4-Trip-PEG, (b) 1,8-Trip-PEG and (c) 1,8,13-Trip-PEG during a second heating/cooling cycle, measured at a scan rate of 10 °C min⁻¹ under N₂ flow (50 mL min⁻¹).

Fig. 9 shows variable temperature (VT) XRD profiles of Trip-PEGs upon heating. Each sample was prepared by cooling from its isotropic melt in a glass capillary. The VT-XRD profiles are reversible upon heating and cooling. All polymers below their melting temperatures exhibit multiple diffractions, including strong peaks at d₁₂₀ = 0.45 nm and d₀₃₂ = 0.38 nm for monoclinic crystals of the PEG chain with a 7/2 helical structure (e.g., Fig. 1c).^14,15 The XRD patterns agree well with the simulated XRD patterns for monoclinic crystals of the PEG main chain with 7/2 helical structure (Fig. 1c and Table S1, SI).¹⁴ In the small angle region, 1,4-Trip-PEG and 1,8-Trip-PEG exhibit three peaks with q values of 1 : 2 : 3 arising from a periodic lamellar structure, along with amorphous halo of the PEG crystals.^23,24 The lamellar thickness is estimated to be 14.4 nm for both 1,4-Trip-PEG and 1,8-Trip-PEG (Fig. 9a and b). Although 1,8,13-Trip-PEG also forms a lamellar structure in the solid state, only up to second-order diffractions can be observed, and the layer spacing (15.5 nm) is relatively large (Fig. 9c and 10a). On the other hand, the occurrence of 2D hexagonal packing of the triptycene termini is unclear due to the overlap with strong diffractions from the PEG crystal.

Fig. 9 VT-XRD profiles of (a) 1,4-Trip-PEG, (b) 1,8-Trip-PEG and (c) 1,8,13-Trip-PEG measured after cooling from their isotropic melts in glass capillaries with diameters of 1.5 mm. Simulated XRD patterns of the monoclinic crystal of PEG are shown as red curves.

Fig. 10 Schematic illustration of the structure of 1,8,13-Trip-PEG (a) below T_m of the PEG main chain, (b) at 60 °C and (c) above the T_m of the triptycene termini.

Above the temperature at which 1,4-Trip-PEG and 1,8-Trip-PEG melt judging from the DSC profile, e.g., 60 °C, the X-ray diffraction peaks disappear for both polymers (Fig. 9a and b). In contrast, for 1,8,13-Trip-PEG (Fig. 9c), diffractions from the (100), (110) and (200) planes of a 2D hexagonal packing of the triptycene units (a = 0.80 nm), as well as diffractions from the (001) and (002) planes of a 1D lamellar structure (layer spacing = 16.1 nm) can be clearly seen. This diffraction feature is similar to that observed for 1,8,(13)-Trip-PDMS at 25 °C,^10,11 where the rubbery, amorphous polymer chains are accommodated between 2D triptycene arrays and organise into a 1D multilayer structure (Fig. 10b). Upon further heating, diffractions due to the triptycene assembly become weaker and completely disappear above 90 °C, where the entire system is in its isotropic melt (Fig. 10c). This transition is not observable in DSC, likely due to the fact that the weight fraction of the triptycene termini is very small, and also the transition is gradual.

Considering that the layer spacing between the triptycene arrays at 60 °C is similar to the lamellar thickness of the PEG chain at lower temperatures, it is suggested that the lamellar structure is defined by the 1D layer spacing within the “2D + 1D” assembly (Fig. 10a and b). To evaluate this, we simulated a 231-mer model of crystalline 1,8,13-Trip-PEG (M_n = 11 502 g mol⁻¹) (Fig. 11). In its 7/2 helical conformation (Fig. 1c), the polymer chain extends approximately 64.3 nm, which is roughly four times longer than the observed lamellar thickness (15.5 nm), indicating that each chain likely folds into three turns within a single lamella (Fig. 11). In this structural model, two triptycene termini are associated with four PEG chains. The estimated cross-sectional area of these coiled PEG chains is approximately 1.05 nm² (Fig. 1c and 11a), while that of the two triptycene units arranged in a hexagonal array (a = 0.8 nm) is approximately 1.10 nm², suggesting a good geometric match (Fig. 10a and 11). This close agreement supports the feasibility of such packing, although the slight mismatch may partly hinder crystallinity of PEG. While diffraction peaks from the triptycene units are overlapped with intense diffractions from the PEG crystals, the appearance of ordered structures upon cooling from the isotropic melt (Fig. 10) supports the formation of “2D + 1D” assemblies in the solid state.

Fig. 11 Molecular model of 1,8,13-Trip-PEG (M_n ≈ 12 kDa) showing lamellar folding. (a) Triptycene termini arranged in a 2D hexagonal array with a lattice parameter of 0.8 nm. (b) PEG 231-mer with a 7/2 helical conformation forming approximately three turns per lamella, yielding an observed lamellar thickness of ca. 15.5 nm.

Table 1 summarises the self-assembling behaviours of the Trip-PEGs. In water, differences in the self-assembling ability among 1,4-, 1,8- and 1,8,13-Trip-PEGs are evident, leading to distinct LCST and gelation behaviours. In the solid state, 1,4- and 1,8-Trip-PEGs exhibit essentially identical crystallisation behaviours, suggesting that the self-assembling ability of 1,8-Trip is insufficient to direct the crystallisation of PEG. In contrast, 1,8,13-Trip-PEG enables structural ordering even in the crystalline state of PEG, owing to the strong self-assembling ability of its 1,8,13-substituted triptycene termini. These findings demonstrate that robust terminal-group assembly can dominate over main-chain crystallisation, offering a powerful route to structural control in crystalline polymers.

Table 1 Summary of properties of Trip-PEGs in aqueous and solid states

			1,4-Trip-PEG		1,8-Trip-PEG		1,8,13-Trip-PEG
In H2O	Solution	LCST	76 °C	>	63 °C	>	60 °C
		%T at 90 °C	12%	>	9%	>	2%
	Gel	CGC	12.8 wt%	>	10.9 wt%	>	7.6 wt%
		2D + 1D structure	No		No		Yes
Solid state	DSC	Tc of PEG	38 °C	≒	35 °C	≒	33 °C
		ΔHc of PEG	123 J g^−1	≒	125 J g^−1	>	92 J g^−1
		Tm of PEG	55 °C	≒	55 °C	≒	56 °C
		ΔHm of PEG	123 J g^−1	≒	124 J g^−1	>	98 J g^−1
	XRD	Lamella thickness	14.4 nm	≒	14.4 nm	<	15.5 nm
		2D + 1D structure	No		No		Yes

---

Conclusions

In this study, we synthesised telechelic PEGs with 1,4-, 1,8- and 1,8,13-substituted triptycene termini to investigate how terminal-group-driven polymer ordering affects the structuring of crystalline polymers. In water, 1,8,13-Trip-PEG formed a hydrogel with a well-defined “2D + 1D” structure, whereas 1,4- and 1,8-Trip-PEGs produced hydrogels without long-range ordering. In the solid state, XRD and DSC analyses revealed that 1,4- and 1,8-Trip-PEGs exhibit thermal and structural behaviour dominated by PEG crystallisation. In contrast, 1,8,13-Trip-PEG forms a well-defined “2D + 1D” structure, regardless of whether the PEG chains are melted or crystallised. This demonstrates that the terminal triptycene-driven self-assembly is strong enough to induce polymer ordering even in crystalline systems. We believe that the approach presented here will be broadly applicable to a variety of polymer systems, irrespective of the backbone crystallinity.

Author contributions

Ta. F. conceived the project; F. I., To. F. and Ta. F. designed the experiments; Y. C., To. F. and F. I. carried out the synthesis and characterisation of the materials; Y. C., F. I., To. F. and T. K. performed the X-ray diffraction experiments and analysed the data; Y. C., F. I. and To. F. performed the LCST experiments and analysed the data; F. I., Y. C., To. F. and Ta. F. co-wrote the manuscript.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the SI. Supplementary information: Experimental procedures, NMR and IR spectral data, SEC profiles, and crystallographic data for PEG. See DOI: https://doi.org/10.1039/d5tc02652k

Acknowledgements

This work was supported by JSPS KAKENHI (JP21H05024 and JP21H04690 for Ta. F.). This work was also supported in part by the Research Program of “Five-Star Alliance” in “NJRC Mater. & Dev.”. We thank the Materials Analysis Division, Core Facility Center, Research Infrastructure Management Center, Institute of Science Tokyo, for their support with the NMR measurements.

Notes and references

1. I. W. Hamley, The Physics of Block Copolymers, Oxford University Press, Oxford, 1998.
2. R. Gronheid and P. Nealey, Directed Self-assembly of Block Copolymers for Nano-manufacturing, Woodhead Publishing, Cambridge, 1st edn, 2015.
3. H.-C. Kim, S.-M. Park and W. D. Hinsberg, Chem. Rev., 2010, 110, 146–177.
4. Y. Mai and A. Eisenberg, Chem. Soc. Rev., 2012, 41, 5969–5985.
5. S. Singh and B. Nandan, Macromolecules, 2025, 58, 2807–2828.
6. N. Seiki, Y. Shoji, T. Kajitani, F. Ishiwari, A. Kosaka, T. Hikima, M. Takata, T. Someya and T. Fukushima, Science, 2015, 348, 1122–1126.
7. F. Ishiwari, Y. Shoji and T. Fukushima, Chem. Sci., 2018, 9, 2028–2041.
8. F. Ishiwari, Y. Shoji, C. J. Martin and T. Fukushima, Polym. J., 2024, 56, 791–818.
9. M. Zharnikov, Y. Shoji and T. Fukushima, Acc. Chem. Res., 2025, 58, 312–324.
10. F. Ishiwari, G. Okabe, H. Ogiwara, T. Kajitani, M. Tokita, M. Takata and T. Fukushima, J. Am. Chem. Soc., 2018, 140, 13497–13502.
11. Y. Chen, F. Ishiwari, T. Fukui, T. Kajitani, H. Liu, X. Liang, K. Nakajima, M. Tokita and T. Fukushima, Chem. Sci., 2023, 14, 2431–2440.
12. F. Ishiwari, G. Okabe, T. Kajitani and T. Fukushima, ACS Macro Lett., 2021, 10, 1529–1534.
13. J. Yu, A. Itagaki, Y. Chen, T. Fukui, F. Ishiwari, T. Kajitani and T. Fukushima, Macromolecules, 2023, 56, 4556–4565.
14. Y. Takahashi and H. Tadokoro, Macromolecules, 1973, 6, 672–675.
15. Y. Takahashi, I. Sumita and H. Tadokoro, J. Polym. Sci., Polym. Phys. Ed., 1973, 11, 2113–2122.
16. Z. Geng, J. J. Shin, Y. Xi and C. J. Hawker, J. Polym. Sci., 2021, 59, 963–1042.
17. L. Voorhaar and R. Hoogenboom, Chem. Soc. Rev., 2016, 45, 4013–4031.
18. Z. Zhang, Q. Chen and R. H. Colby, Soft Matter, 2018, 14, 2961–2977.
19. M. T. Cook, P. Haddow, S. B. Kirton and W. J. McAuley, Adv. Funct. Mater., 2021, 31, 2008123.
20. G. Pasparakis and C. Tsitsilianis, Polymer, 2020, 211, 123146.
21. R. Paberit, E. Rilby, J. Göhl, J. Swenson, Z. Refaa, P. Johansson and H. Jansson, ACS Appl. Energy Mater., 2020, 3, 10578–10589.
22. B. Liang, X. Lu, R. Li, W. Tu, Z. Yang and T. Yuan, Sol. Energy Mater. Sol. Cells, 2019, 200, 110037.
23. H. Sun, D. M. Yu, S. Shi, Q. Yuan, S. Fujinami, X. Sun, D. Wang and T. P. Russell, Macromolecules, 2019, 52, 592–600.
24. R.-J. Roe, Methods of X-ray and Neutron Scattering in Polymer Science, Oxford University Press, Oxford, 2000.

---

Footnote

† Present address: Department of Applied Chemistry for Environment, Graduate School of Urban Environmental Sciences, Tokyo Metropolitan University, 1-1 Minami-Osawa, Hachioji, Tokyo 192-0397, Japan.

---