Nanoporous Films with Photoswitchable Absorption Kinetics Based on Polymerizable Columnar Discotic Liquid Crystals.

A photoresponsive nanoporous polymer film has been produced from the templated self-assembly of a columnar liquid crystal containing azo units. A liquid crystalline complex of polymerizable azobenzoic acid and a tris-benzimidazolyl benzene template molecule was cross-linked via thiol-ene radical copolymerization with dodecanedithiol. Subsequent removal of the template yielded nanoporous polymer films with pores of approximately 1 nm in diameter. Both trans-cis and cis-trans photoisomerizations of azobenzoic acid took place in the porous films. At room temperature, the cis isomer was sufficiently long-lived to establish a difference in dye absorption kinetics of the two isomers. The cationic dye rhodamine 6G was bound to both isomers, but the rate of binding to films enriched in the cis isomer was 8 times faster.

Introduction

An intensive study of nanoporosity in different types of materials over the last decade has led to improved performance in applications such as separation membranes, catalysis, drug delivery, etc.[1−3] The use of self-assembling organic building blocks to develop nanoporous materials is an elegant approach to nanoporosity that gives well-defined structures with uniform pore distribution.[4] Selectivity, absorption kinetics, and absorption capacity are often better in inorganic nanoporous materials like silica or zeolites, which have better long-term stability but are more difficult to process.[5,6] The versatility of soft organic nanostructured materials permits tuning of features like absorption kinetics and selectivity using external stimuli such as light.[7]

Incorporating light-responsive molecular moieties in the pores that can photoswitch between two isomers provides a method to alter pore dimension and polarity.[8,9] Azobenzenes, stilbenes, and acylhydrazones are light-responsive compounds that isomerize reversibly when irradiated with light.[10−12] Among these units, azobenzenes are the most convenient building blocks, since their synthesis is relatively straightforward, and photoisomerization is brought about by convenient light sources with wavelengths of ∼400 nm, whereas stilbene and acylhydrazone are switched by light of shorter wavelengths.[13]

Azobenzenes usually isomerize from trans to cis when irradiated with 300–400 nm UV light, while cis-to-trans isomerization is stimulated by irradiation at >400 nm wavelengths.[14,15] The trans isomer is also formed by thermal relaxation of the cis form, since this isomer is thermodynamically more stable.[16,17] Reversible isomerization of azobenzene compounds can lead to photochromic pKa changes and changes in binding strength to other molecules.[18] When the azo moiety is part of a receptor molecule, reversible isomerization may change the size of the binding site, affecting the selectivity toward molecules of different sizes.[19]

Azobenzenes have also been applied as photoresponsive units in smectic liquid crystalline supramolecular networks. In nanoporous films of these networks, the number of binding sites and pore dimensions increased upon irradiation with UV light due to trans–cis isomerization of the azo groups.[20]

Well-defined nanoporous polymer membranes have been made by self-assembly of block copolymers and liquid crystals, providing diverse pore sizes and shapes.[4,21] Moreover, the density of their nanometer-sized pores can be increased considerably using columnar discotic liquid crystals (CLCs).[22−24] In earlier work by our group, nanoporous films have been prepared with CLCs from a 1:3 supramolecular complex between a tris-benzimidazolyl benzene core (TB), as a template molecule, and three benzoic acids giving C3-symmetric supramolecular disks, which stack into columns.[25] A nanoporous material was obtained when the columnar morphology was fixated by cross-linking the double bonds of the acids, and the TB template molecules were subsequently removed.

This paper describes the synthesis and characterization of photoswitchable nanoporous films based on a wedge-shaped azo derivative 3, Azoac, which forms a disk-shaped 3:1 supramolecular complex with TB (Figure ) and self-assembles into a columnar mesophase. The columnar morphology was fixated by cross-linking with a thiol-ene reaction, and removal of the template led to the formation of a nanoporous material. We have established previously that the pores in similar materials without photoswitchable moieties have a fixed selectivity for binding of charged dyes.[25] In the current work, we establish that photoswitching between trans and cis isomeric states in the porous material influences the absorption kinetics of the cationic dye rhodamine 6G.

Figure 1

Isomerization of azo moieties with UV light in a porous material changes the pore size and the absorption rate.

Results and Discussion

Synthesis and Characterization of Azoac

The synthesis of 4-((2,3,4-tris(undec-10-en-1-yloxy)phenyl)diazenyl)benzoic acid (3) was performed in three steps, as shown in Scheme .

Scheme 1

Synthesis of Wedge-Shaped Azo Derivative 3 (Azoac)

First, ethyl 4-aminobenzoate was diazotized with sodium nitrite to give the corresponding diazonium salt. The azo product, 4-((3,4,5-trihydroxyphenyl)diazenyl)benzoate (1), was obtained by reacting the freshly prepared diazonium salt with 1,2,3-tris(hydroxybenzene) (pyrogallol).[26]1H nuclear magnetic resonance (NMR) showed that substitution predominantly took place at the 4-position of pyrogallol due to the electron-donating nature of the hydroxy groups, favoring substitution at the ortho and para positions.[27] Alkylation of the phenolic hydroxy groups with 11-bromo-1-undecene yielded tris-alkene ester 2, which was hydrolyzed with aqueous potassium hydroxide, followed by acidification with aqueous hydrochloric acid to give photoswitchable acid Azoac. 1H NMR showed that only the thermodynamically more stable trans isomer was present. In bulk, molecules of Azoac are present as hydrogen-bonded dimers, as indicated by the broadened C=O stretch band at 1686 cm–1 (Figure S11).[28] Dimerization of the fan-shaped molecule promotes liquid crystallinity, and polarized optical microscopy (POM) and differential scanning calorimetry (DSC) showed that pure Azoac is liquid crystalline between 25 and 130 °C (Figure S12).

Formation of TB·Azoac complex

Mixing TB with 3 equiv of Azoac afforded a hydrogen-bonded 1:3 C3-symmetric complex (Figure A) as is evident from the Fourier-transform infrared (FT-IR) spectrum in which the C=O band shifted from 1686 to 1674 cm–1, and a new band appeared at 3250 cm–1 (Figure B), which was assigned to the imidazolium N+–H stretching vibration. The presence of a mesophase was established with POM and DSC. However, at a 3:1 ratio of Azoac and TB, needle-shaped crystals were observed after isotropization of the mesophase, indicating phase separation of the template. Using 3.2 equiv of Azoac prevented crystallization of TB and resulted in a clean Colhex LC phase with fan-shaped texture in POM (Figure C) below 50 °C in DSC (Figure S13).

Figure 2

(A) Formation of supramolecular discotic complex TB·Azoac via H-bonding. (B) IR spectra of Azoac, TB, and the 3:1 mixture, indicating the formation of a H-bonded complex. (C) POM image of a 3.2:1 mixture of Azoac and TB at room temperature.

The X-ray diffractogram showed peaks with q-ratios of 1:√3:√4:√7 (Table and Figure S14), confirming the presence of a Colhex mesophase with a lattice parameter of 3.76 nm. The wide-angle X-ray scattering (WAXS) diffractogram shows a diffraction peak corresponding to an interdisk distance c of 0.38 nm, identifying the mesophase as ordered columnar hexagonal (Colho).

Table 1

XRD Data of TB·Azoac at 298 K after Annealing

peak	q (nm–1)	hkl	dobs (nm)	dcalc (nm)	structural parameter
1	1.93	100	3.25	3.25	a = 3.76 nm
2	3.36	110	1.87	1.88
3	3.87	200	1.62	1.62
4	5.02	210	1.25	1.23
5	5.79	220	1.08	1.08
6	19.04	001	0.33		c = 0.38 nm

Formation of Nanoporous Films by Photopolymerization

Polymer films obtained from the liquid crystal phase must be highly cross-linked to obtain nanoporosity after template removal. Colhex phases have previously been cross-linked with acyclic diene metathesis (ADMET) polymerization of tris-alkenyl benzoic acids using Grubbs’ 2nd generation catalyst.[25,29] A similar approach with the complex of TB with Azoac was not successful. Drop-casting a solution of the azo-based LC complex with 2 mol % Grubbs’ 1st, 2nd, or 3rd generation catalyst and heating the film under vacuum at 100 °C overnight yielded an orange-brown film. However, FT-IR showed that with each catalyst, polymerization was incomplete, since the =C–H bending peak at 906 cm–1 was still present. The low degrees of polymerization are likely caused by the interaction of the azo group with Grubbs’ catalyst.[30]

As an alternative cross-linking method, the photochemical thiol-ene click reaction (Figure A) was more successful. Drop-cast films (thickness of 0.5–1 μm) of the complex containing 4.5 equiv of 10-decanedithiol (DDT) and 2 wt % Irgacure 819 photoinitiator[31,32] were irradiated with UV light for 15 min, using a 405 nm cutoff filter to prevent trans–cis isomerization of the azo group. Alkene conversion was estimated to be 82%, based on the intensity of the =C–H bending band at 907 cm–1 in the FT-IR spectrum (Figure B).[33] Finally, a porous polymer was obtained after removal of the TB template by soaking the film in dimethyl sulfoxide (DMSO) for 3 h, followed by two additional washing steps with water and extensive drying in vacuo. FT-IR showed full disappearance of the N+–H stretching band at 3250 cm–1 and a shift of the C=O stretching vibration of carboxylic acid from 1676 to 1689 cm–1, confirming template removal (Figure C).

Figure 3

(A) Cross-linking of the film by photopolymerization and consecutive removal of the TB template molecule. (B) FT-IR spectra of the TB·Azoac complex before and after polymerization. The arrow indicates nearly complete disappearance of the =C–H stretching band of the double bond (C) FT-IR spectra of a thin film of the polymerized complex before and after removal of the template. The arrows indicate the disappearance of the N–H band and a shift of the carbonyl band. (D) 1D wide-angle X-ray diffractograms of the polymerization mixture (red), after polymerization (black), and after the removal of TB (gray).

X-ray diffractograms were taken to verify retention of the columnar order after cross-linking and after removal of the template (Figure D). Before polymerization, the mixture of TB·Azoac with 4.8 equiv of DDT and photoinitiator had a Colhex LC phase with an increased lattice spacing d(100) of 4.65 versus 3.76 nm compared to pure TB·Azoac (Figure S14). Polymerization did not affect the Colhex order, but the lattice spacing decreased to 4.33 nm (Figure D). When the TB template was removed, the spacing increased to 4.52 nm (Figure D, the gray line). A reflection from the interdisk distance at wide angles is absent in the diffractogram, indicating disordered stacking in the polymerized material.[34]

Photoisomerization of TB·Azoac in Solution and in Films

Trans-to-cis photoisomerization of Azoac in solution was monitored with UV–vis spectroscopy and the slower cis-to-trans thermal relaxation was monitored with both UV–vis and 1H NMR spectroscopy. The UV–vis spectrum of a 93.0 μM chloroform solution of Azoac shows an intense π→π* absorption band at 377 nm, corresponding to the trans isomer and a much weaker n→π* band at 465 nm.[17] Upon irradiation of the solution with 365 nm UV light (0.7 mW cm–2), the intensity of the π–π* transition band at 377 nm decreased strongly, indicating trans-to-cis isomerization. After approximately 1 min, a photostationary state was reached (Figure ). Irradiation of the solution with 450 nm blue light (350 mA) resulted in the increase of the 377 nm absorption band and a slight decrease at 465 nm as the result of cis-to-trans photoisomerization (Figure ), showing that isomerization is photochemically reversible with a photostationary state that depends on the irradiation wavelength.[35]

Figure 4

UV–vis spectra of Azoac in CHCl3 after UV irradiation for various times. (A) Irradiation at 365 nm and (B) irradiation at 450 nm.

Thermal relaxation of azobenzene was investigated to obtain the half-life of the cis isomer. A solution of Azoac in CDCl3 (100 mM) was irradiated at 365 nm (0.7 mW cm–2) for 10 min. By monitoring the absorption on UV–vis of the recovery of the band at 377 nm (Figure S15), the half-life of the cis isomer was calculated. The data were fitted with an exponential model yielding a half-life for the cis isomer of 13.8 h at room temperature (Figure S16), in line with lifetimes for cis–trans isomerization of similar azo compounds.[36] Excitingly, photoisomerization studies on thin films of the porous polymer revealed that azobenzene still switched from trans to cis under these conditions.

Upon irradiation of a thin film (≈0.5 μm) at 365 nm, the intensity of the band at 377 nm decreased by 40% in 1.5 h (Figure ), at which time a photostationary state had been reached. When this cis-enriched film was irradiated at 450 nm light (Figure S17), conversion to the trans isomer only took a few minutes. When the cis-enriched film was allowed to relax back thermally at room temperature in the dark, the absorption band at 377 nm increased with a half-life of 8.6 h, significantly faster than in solution (Figure S16).

Figure 5

Photoisomerization of the azo groups in the polymerized film. (A) Absorbance spectra at different times of irradiation at 365 nm for 3 h and (B) thermal relaxation curve measured at room temperature at 377 nm and its exponential fitting curve.

The shorter half-life for relaxation of the cis-to-trans isomerization in the polymer is related to the restriction of free volume. Initially, the network is disposed to accommodate the trans form. Hence, photoisomerization to the cis form is slower than in solution because of the restricted movement. The cis form is in a sterically strained environment, favoring thermal relaxation to the more stable trans form.[37−39]

Although the half-life of the cis form is significantly lower in the polymer, the films are sufficiently stable to study dye absorption in pure trans- and cis-enriched films, separately.

Uptake of Rhodamine 6G in Porous Films

Trans-to-cis isomerization in the porous polymer is expected to result in increased pore size, as shown in Figure . Isomerization does not only result in larger pores, but the pores are also expected to become more polar because molecular symmetry is reduced.[17,40] To verify these predictions, absorption of rhodamine 6G was tested, which is a cationic dye with a size (1.35 nm × 1.06 nm) similar to the pore diameter. cis-Enriched films were obtained by irradiation at 365 nm for 1.5 h. Samples of 0.5 mg of trans- or cis-enriched porous polymer films (a thickness of 1 μm) were immersed in a 10 μM rhodamine 6G solution in a 1.0 cm cuvette. Dye absorption was followed by monitoring the decrease of light absorption at 474 nm, the rhodamine 6G absorption maximum (Figure S19), while taking care to keep the films away from the light path. To analyze the absorption rates, the concentrations were plotted with a smoothing function over 20 points (Figure A, dotted lines). Both types of films absorbed the dye but with significantly different uptake rates, with initial absorption in the cis-enriched film approximately 2 times higher after 10 h. Reversibility of the uptake behavior was tested by comparing dye absorption in trans films with absorption in films that were irradiated to create the cis-enriched state and then were thermally relaxed for 10 h prior to measuring dye uptake. Figure B shows that the absorption rates for both routes are equal.

Figure 6

(A) Overlay of absorption measurements of rhodamine 6G versus time in porous films of trans- and cis-enriched isomers with the fitted Fickian transport equation using the optimized DF and C0 at t = 0. (B) Absorption of rhodamine 6G versus time in the porous film in its native trans state and after irradiation + relaxation.

After 45 h, the cis-enriched film had bound 0.036 mmol dye per mg of the film, corresponding to approximately 1 dye molecule per 6 Azoac units, which is the same as 2 TB·Azoac complexes, while for the trans polymer, approximately half of that amount was adsorbed. The maximum amount of the dye absorbed, based on the size of the molecule, may be assumed to correspond to one rhodamine cation per TB·Azoac complex.

The absorption rate of rhodamine 6G was fitted using a Fickian diffusion model[41] (see the Supporting Information for details) with the diffusion coefficient of the adsorbate in the film as the single fitting parameter.

The data of the initial 45 min (2600 s) of the absorption measurements, when little isomerization has taken place, was used for the fitting. The decreasing concentration of the dye in the solution was taken into account for the calculation of the diffusion coefficients of rhodamine 6G for each isomer.

The best fit for the diffusion coefficients DF (Table ) gave an approximately 8 times higher absorption rate for the cis-enriched material than for the trans polymer (Figure A). Since the cis isomer has a limited lifetime and absorption equilibrium had not been reached after 48 h, it is not feasible to determine the equilibrium absorptions of cis and trans isomers individually.

Table 2

Optimal Parameters from Fitting with the Fickian Transport Equationa

isomer	time (s)	DF (10–5·μm2·s–1)
trans	0–2600	0.290
cis	0–2600	2.278

Pore length: 0.5 μm, DF: diffusion coefficient of the adsorbate.

Conclusions

A light-responsive nanoporous polymer film was prepared based on a columnar discotic liquid crystal formed by a new 3:1 hydrogen-bonded complex containing azobenzene groups. The columnar morphology was retained after thiol-ene polymerization, and nanometer-sized pores were created by removal of the template. Trans-to-cis photoisomerization in the solid was slower than in solution, while thermal and photochemical cis–trans isomerizations were faster than in solution. Absorption of rhodamine 6G was faster in the cis-enriched films than in the trans films, confirming the initial hypothesis that pore size increases upon irradiation due to the isomerization of the azo moieties in the nanoporous structure. The change in absorption kinetics opens the possibility to apply these materials as light-controlled filtration membranes.

Experimental Section

General

Chemicals and solvents were purchased from Sigma-Aldrich or BioSolve and used as received unless stated otherwise. NMR spectra were recorded at room temperature on a Bruker FT-NMR spectrometer AVANCE III HD-NanoBay (400 MHz, Bruker UltraShield magnet, BBFO Probehead, BOSS1 shim assembly) in either acetone-d6 or CDCl3. The chemical shifts are given in ppm and the coupling constants as J in Hz. FT-IR spectra are measured on a Perkin Elmer Spectrum One spectrometer equipped with an attenuated total reflectance (ATR) sampling accessory. Matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectra were obtained on a PerSeptive Biosystems Voyager-DE PRO spectrometer with α-cyano-4-hydroxycinnamic acid as a matrix. Flash column chromatography was carried out using silica gel. Polarized optical microscopy (POM) images were recorded on a Jeneval microscope equipped with crossed polarizers and on a Polaroid CCD camera equipped with a Linkam THMS 600 hot stage. DSC measurements were performed in aluminum sample pans using a TA Instruments Q1000. All runs were performed with heating and cooling rates of 10 K min–1. X-ray scattering measurements were performed on a Ganesha lab instrument equipped with a Genix-Cu ultra-low divergence source producing X-ray photons with a wavelength of 0.154 nm and a flux of 1 × 108 photons s–1. Diffraction patterns were collected using a Pilatus 300 K silicon pixel detector with 487 × 619 pixels of 172 μm2 in size, placed from the sample at a detector distance of 91 mm (wide angle, WAXS) or 500 mm (medium angle, MAXS). On the obtained diffraction patterns, an azimuthal integration was performed, using SAXSGUI software, to calculate the intensity against the scattering vector q, where q = (4π/λ) sin ϑ (ϑ is the angle of incidence and λ is the wavelength). The beam center and the q-range were calibrated using silver behenate (d(100) = 1.076 nm–1; 5.839 nm) as a reference. The d(300) was used for calibration. The temperature was controlled with a Linkam HFSX350 heating stage and cooling unit. Measurements were performed on bulk samples sealed in 1.0 mm diameter glass capillaries (Hilgenberg) or directly on the polymer film. UV–vis experiments were performed on a Jasco V-750 spectrophotometer. All experiments were performed in a 10 × 10 mm2 quartz cuvette at 20 °C. The photopolymerization was performed for 15 min with a mercury lamp (Omnicure s1000, emitting at 320–500 nm), which was provided with a 400 nm cutoff filter (Newport) to avoid the isomerization from trans to cis of the azobenzene moiety during polymerization. The intensity was approximately 22 mW cm–2 at the sample surface.

Synthetic Procedures

Compound 1

The synthesis and characterization of the tris-benzimidazolyl benzene core (TB) are already reported.[22] To a stirred solution of ethylaminobenzoate (5.71 g; 34.5 mmol; 1.0 equiv) in H2O (17 mL), 23 mL of 3 M HCl (2.0 equiv) was added. A solution of NaNO2 (2.87 g; 41.6 mmol; 1.2 equiv) in H2O (15 mL) was added dropwise to the solution. The mixture was stirred for 30 min at a temperature below 5 °C. Then, pyrogallol (4.44 g; 35.2 mmol; 1.0 equiv) was added and reacted for 1 h (<5 °C). After completion of the reaction, saturated NaHCO3 was added to reach a pH of ∼6 and the suspension was filtered to obtain a solid. The solids were washed 3 times with 50 mL of water and then freeze-dried. The crude compound was purified by flash column chromatography (chloroform/EtOH, 99:1 to 9:1, in 20 CV) to obtain a dark brown solid (938 mg, 10%). 1H NMR (acetone-d6, 400 MHz): δ = 8.18 (d, 2H, J = 8.8 Hz, 2 × CH arom.), 7.95 (d, 2H, J = 8.7 Hz, 2 × CH arom.), 7.39 (d, 1H, J = 9.0 Hz, CH arom.) 6.68 (d, 1H, J = 9.0 Hz, CH arom.), 4.39 (q, 2H, J = 7.1 Hz and 14.3 Hz, O–CH2–CH3), 1.40 (t, 3H, J = 7.1 Hz, CH3–CH2).

Compound 2

To a stirred solution of 1 (400 mg; 1.32 mmol; 1.0 equiv) in dimethylformamide (DMF) (10.5 mL), K2CO3 (1.927 g; 13.9 mmol; 10.0 equiv) and 11-bromo-1-undecene (1.008 g; 4.32 mmol; 3.3 equiv) were added. The resulting mixture was refluxed (110 °C) overnight, cooled to room temperature, and then purified by flash column chromatography (heptane/ethyl acetate 97:3) to obtain an orange oil (569 mg, 57%) 1H NMR (CDCl3, 400 MHz): δ = 8.17 (d, 2H, J = 8.8 Hz, 2 × CH arom.), 7.91 (d, 2H, J = 8.6 Hz, 2 × CH arom.), 7.50 (d, 1H, J = 9.2 Hz, CH arom.), 6.70 (d, 1H, J = 9.3 Hz, CH arom.), 5.81 (m, 3H, CH2=CH–CH2), 4.95 (dd, 6H, J = 10.3 Hz and 17.1 Hz, 3 × CH2=CH), 4.40 (q, 2H, J = 7.1 Hz and 14.2 Hz, O–CH2–CH3), 4.25 (t, 2H, J = 65 Hz, O–CH2–CH2), 4.05 (m, 4H, O–CH2–CH2), 2.04 (m, 6H, 3 × O–CH2–CH2–CH2), 1.83 (m, 6H, 3 × CH–CH2–CH2), 1.51 (m, 6H, 3 × CH2–CH2–CH2), 1.42 (t, 3H, J = 7.2 Hz, CH3–CH2), 1.32 (m, 30H, 15 × CH2–CH2–CH2). 13C NMR (CDCl3, 400 MHz): δ = 166.1, 157.0, 155.6, 153.2, 142.2, 141.0, 139.2, 131.5, 130.5, 122.5, 114.2, 111.6, 107.7, 76.3, 74.0, 68.9, 61.1, 33.8, 30.4, 29.7, 29.6, 29.5, 29.4, 29.3, 29.2, 29.1, 29.0, 28.9, 28.7, 28.5, 26.2, 25.8. ATR FT-IR: 3077, 2978, 2924, 2853, 1718, 1639, 1582, 1490, 1466, 1438, 1404, 1365, 1270, 1242, 1196, 1174, 1148, 1091, 1016, 992, 908, 862, 801, 772, 722, 696, 630, 599, 559, 522 cm–1. MALDI-TOF MS: m/z calculated for C48H74N2O5 (M + H)+: 759.57 found: 759.59, (M + Na)+: 781.55, found: 781.57.

Compound 3 (Azoac)

A solution of ethyl-4-((2,3,4-tris-(undec-10-en-1-yloxy)phenyl)diazenyl)benzoate 2 (470 mg; 0.62 mmol; 1.0 equiv) in EtOH/MeOH/H2O (6:3:1 v/v, 30 mL) was treated with KOH (311 mg; 5.54 mmol; 8.9 equiv) and heated to reflux (70 °C) for 5 h. After completion of the hydrolysis, 5.5 mL of 1 M HCl was added to reach a pH of ∼3.0. The resulting solids were washed with H2O (3 × 10 mL) and then freeze-dried to obtain the final compound as an orange powder (314 mg, 69%). 1H NMR (CDCl3, 400 MHz): δ = 8.24 (d, 2H, J = 8.6 Hz, 2 × CH arom.), 7.94 (d, 2H, J = 8.6 Hz, 2 × CH arom.), 7.52 (d, 1H, J = 9.2 Hz, CH arom.), 6.71 (d, 1H, J = 9.3 Hz, CH arom.), 5.81 (m, 3H, CH2=CH–CH2), 4.96 (m, 6H, 3 × CH2=CH), 4.25 (t, 2H, J = 6.5 Hz, O–CH2–CH2), 4.06 (m, 4H, O–CH2–CH2), 2.04 (m, 6H, CH2–CH2–CH2), 1.83 (m, 6H, CH–CH2–CH2), 1.58 (m, 6H, CH2–CH2–CH2), 1.32 (m, 30H, CH2–CH2–CH2). 13C NMR (CDCl3, 400 MHz): δ = 171.2, 157.1, 156.1, 153.3, 142.2, 141.1, 139.2, 131.2, 130.4, 122.6, 114.1, 111.7, 107,8, 74.0, 69.0, 33.8, 30.4, 30.3, 29.7, 29.6, 29.5, 29.4, 29.3, 29.2, 29.1, 29.0, 26.2, 26.1. ATR FT-IR: 3079, 2977, 2921, 2850, 2659, 2540, 1684, 1640, 1600, 1586, 1491, 1466, 1443, 1419, 1382, 1282, 1248, 1197, 1152, 1095, 1039, 1011, 990, 964, 909, 863, 832, 800, 772, 756, 719, 697, 672, 656, 630, 549, 524, 508, 482 cm–1. MALDI-TOF MS: m/z calculated for C46H70N2O5 (M + H)+: 731.54, found: 731.55.