Hyperbranched 5,6-glucan as reducing sugar ball

Abstract

The ring-opening polymerization of 5,6-anhydro-1,2-O-isopropylidene-α-D-glucofuranose (1) as a latent cyclic AB₂-type monomer was carried out using potassium tert-butoxide (t-BuOK) or boron trifluoride diethyletherate (BF₃·OEt₂) as an initiator in order to synthesize a novel hyperbranched glycopolymer. The anionic and cationic polymerizations proceeded via the proton-transfer reaction mechanism to produce the hyperbranched poly(5,6-anhydro-1,2-O-isopropylidene-α-D-glucofuranose) (2). In particular, the cationic polymerization with the slow-monomer-addition strategy is a facile method leading to the hyperbranched glycopolymers with high molecular weights and highly branched structures. The weight-average molecular weight (M_w,SEC-MALLS) values of 2 measured by multi-angle laser light scattering (MALLS) varied in the range from 7400 to 122 400, which were significantly higher than the weight-average molecular weight (M_w,SEC) values determined by size exclusion chromatography (SEC). The intrinsic viscosities ([η]) of these polymers were very low in the range of 3.3–4.6 mL g⁻¹ and the Mark–Houwink–Sakurada exponents, α, were calculated to be 0.08–0.27. These results of the MALLS, SEC, and viscosity measurements suggested that these polymers exist in a compact spherical conformation in solution because of their highly branched structure. The synthesis of the hyperbranched 5,6-glucan (3) by hydrolysis of polymer 2 was also demonstrated. Polymer 3 is a novel water-soluble hyperbranched glycopolymer arranged with numerous reducing D-glucose units on the peripheries of the polymer, and has a higher reducing ability than D-glucose because of the glyco-cluster effect or the multivalent effect of the reducing D-glucose units. Therefore, polymer 3 should be called a “reducing sugar ball”.

---

Introduction

The spherical architectures of highly branched macromolecules, such as dendrimers, star-shaped polymers, and hyperbranched polymers, have attracted much attention from the viewpoint of nanotechnology, because their numerous terminal units can be converted into various functional groups leading to novel nanomaterials.^1–32 Thus, various types of dendrimers, star polymers, and hyperbranched polymers have been synthesized and their properties were compared to the linear analogues.^23,33–46

Spherical macromolecular architectures with their surfaces covered with sugars are of increasing interest due to their unique structures along with specific properties, such as the glycol-cluster or multivalent effect on the carbohydrate-protein interactions. For a spherical polymer with many terminal sugar residues, there are a few types of glycodendrimers, e.g., dendrimers prepared from the reaction of the dendrimer surface with mono- and disaccharides (sugar ball),^47–59 and dendrimers prepared through the convergent method using semi-dendritic branched oligosaccharides.^60–64 Although the synthesis of such glycodendrimer types was accomplished, dendrimers essentially consisting of sugar units, i.e., dendritic carbohydrate polymers, have not been prepared until now due to the complicated and time-consuming synthetic process involving numerous protection, deprotection, and purification procedures. Star-shaped glycopolymers were also prepared using the “core-first” method based on the RAFT polymerization^65,66 and the “arm-first” method based on the TEMPO-mediated radical polymerization.^67–70 Hyperbranched glycopolymers from the polymerization of AB_m-type monomers can be prepared more easily than glycodendrimers and star-shaped glycopolymers.^71–76 Recently, we proposed that the ring-opening multibranching polymerization of anhydro and dianhydro sugars as latent AB_m-type monomers should be a convenient method of synthesizing hyperbranched glycopolymers with a novel spherical macromolecular architecture,^77–83 for example, the hyperbranched polysaccharides from the cationic polymerization of 1,6-anhydro-β-D-hexopyranose, the hyperbranched poly(2,5-anhydro-D-glucitol) from 1,2:5,6-dianhydro-D-mannitol, and the hyperbranched polytetritols from 1,4-anhydrotetritol and 2,3-anhydrotetritol. For our strategy for synthesising hyperbranched glycopolymers, it is important to elucidate the effect of the structure of the sugar monomers on the ring-opening multibranching polymerization tendency and to clarify the relationship between the structure and properties of the hyperbranched glycopolymers.

In this article, we report the ring-opening multibranching polymerization of 5,6-anhydro-1,2-O-isopropylidene-α-D-glucofuranose (1) as a latent cyclic AB₂-type monomer using anionic or cationic initiators, as shown in Fig. 1, leading to the hyperbranched poly(5,6-anhydro-1,2-O-isopropylidene-α-D-glucofuranose) (2). In fact, Uryu et al. previously reported pioneering research on the ring-opening polymerization of 1,^84,85 although the polymer structure and the solution properties were insufficiently characterized. Therefore, the characteristics of the polymerization, the polymer structure, and the solution characteristics were discussed by the NMR measurements and the size exclusion chromatography (SEC) with a multi-angle laser light scattering (MALLS) detector and a viscosity detector. In addition, the synthesis of a hyperbranched 5,6-glucan (3) by hydrolysis of polymer 2 was demonstrated. Since polymer 3 is a novel water-soluble hyperbranched glycopolymer arranged with numerous reducing D-glucose units on the peripheries of the polymer, the reduction property of 3 is discussed from the viewpoint of the physical property as a “reducing sugar ball”.

Fig. 1 Polymerization of 5,6-anhydro-1,2-O-isopropylidene-α-D-glucofuranose (1) and synthesis of hyperbranched 5,6-glucan (3) by hydrolysis of polymer 2.

Experimental

Materials

5,6-Anhydro-1,2-O-isopropylidene-α-D-glucofuranose (1) was synthesized from 1,2-O-isopropylidene-α-D-glucofuranose using a procedure similar to that reported by Yus et al.⁸⁶1 was purified by column chromatography on silica gel with n-hexane/ethyl acetate (3/2, v/v, R_f 0.15) and dried in a vacuum oven at 50 °C for 2 days. ¹H NMR (CDCl₃):δ (ppm) 1.32 and 1.48 (s, 6H, 2CH₃), 2.88 and 3.00 (m, 2H, 2H-6), 3.11 (m, 1H, OH), 3.43 (m, 1H, H-5), 4.07 (m, 1H, H-4), 4.26 (m, 1H, H-3), 4.52 (m, 1H, H-2), 5.99 (d, 1H, H-1). ¹³C NMR (CDCl₃):δ (ppm) 26.1 and 26.7 (2C, 2CH₃), 45.9 (C-6), 50.2 (C-5), 75.1 (C-3), 79.2 (C-4), 85.0 (C-2), 104.9 (C-1), 111.8 (C). Potassium tert-butoxide (t-BuOK)(1.0 mol L⁻¹ in THF) was obtained from the Sigma-Aldrich Co., and used as received. Boron trifluoride diethyletherate (BF₃·OEt₂) was purchased from Kanto Chemical Co., Ltd (Tokyo, Japan) and distilled over CaH₂ under reduced pressure. Silica gel 60N (Kanto Chemical Co., Ltd; spherical shape; particle size, 40–50 μm; neutral) was used for the column chromatography. Methyl iodide (>99.5%), silver oxide (>99.0%), an ammonia aqueous solution (28.0–30.0%), tetrachloroauric (III) acid (HAuCl₄) tetrahydrate (>99.0%), trifluoroacetic acid (>99.0%), dry dichloromethane (dry CH₂Cl₂) (>99.5%; water content, <0.001%), dry tetrahydrofuran (THF) (>99.5%; water content, <0.001%), dry acetonitrile (CH₃CN) (>99.5%; water content, <0.001%), methanol (>99.5%), acetone (>99.0%), n-hexane (>95.0%), ethyl acetate (>99.5%), dimethylformamide (DMF) (>99.5%), and chloroform (CHCl₃) (>99.0%) were obtained from Kanto Chemical Co., Ltd, and used without further purification.

Measurements

The spectra of ¹H (400 MHz), ¹³C NMR (100 MHz), distortionless enhancement by polarization transfer (DEPT), and ¹³C-¹H correlation spectroscopy (COSY) were recorded using a JEOL JNM-A400II instrument. The molecular weight values of polymers 2, 4 and 6 were determined by size exclusion chromatography (SEC) in THF (1.0 mL min⁻¹) at 40 °C using an Agilent 1100 series instrumentation equipped with two Shodex KF-804L columns (linear, 8.0 mm × 300 mm; exclusion limit, 4 × 10⁵; bead size, 7 μm), a DAWN 8 multi-angle laser light scattering (MALLS) detector (Wyatt Technology, Santa Barbara, CA), a Viscostar viscosity detector (Wyatt Technology), and an Optilab rEX refractive index detector (Wyatt Technology). The absolute molecular weights (M_w,SEC-MALLS), polydispersity (M_w,SEC-MALLS/M_n,SEC-MALLS), intrinsic viscosity ([η]), refractive index change (dn·dc⁻¹), and Mark–Houwink–Sakurada constants, α and K, were estimated by ASTRA 5.1.6.0 software (Wyatt Technology). The relative molecular weight (M_w,SEC) and the polydispersity (M_w,SEC/M_n,SEC) in THF were calculated on the basis of a polystyrene calibration. The molecular weight values of polymer 3 were determined by size exclusion chromatography (SEC) in a 0.2 mol L⁻¹ NaNO₃ aqueous solution (1.0 mL min⁻¹) at 40 °C using an Agilent 1100 series instrumentation equipped with two Tosoh TSKgel GMPW_XL columns (linear, 7.8 mm × 300 mm; exclusion limit, 5 × 10⁷; bead size, 13 μm), a DAWN 8 multi-angle laser light scattering (MALLS) detector (Wyatt Technology), a Viscostar viscosity detector (Wyatt Technology), and an Optilab rEX refractive index detector (Wyatt Technology). The M_w,SEC-MALLS, M_w,SEC-MALLS/M_n,SEC-MALLS, [η], dn·dc⁻¹, and Mark–Houwink–Sakurada constants, α and K, were estimated by ASTRA 5.1.6.0 software (Wyatt Technology). Preparative SEC for the chloroform-soluble polymers was performed in chloroform (3.5 mL min⁻¹) at 23 °C using a JAI LC-9201 (Japan Analytical Industry Co., Ltd, Japan) equipped with a JAI JAIGEL-3H column (20 mm × 600 mm; exclusion limit, 7 × 10⁴) and a JAI RI-50s refractive index detector. The preparative SEC for the water-soluble polymers was performed in water (14 mL min⁻¹) at 23 °C using a JAI LC-928 (Japan Analytical Industry Co., Ltd, Japan) equipped with a JAI JAIGEL-W253-40 column (40 mm × 500 mm; exclusion limit, 5 × 10⁴) and a JAI RI-50s refractive index detector. The elemental analysis was performed using a MICRO CORDER JM10 and Dionex DX-500 (Instrumental Analysis Division, Management Center, Creative Research Instaction Sousei, Hokkaido University). The measurement of the optical rotations was performed in CHCl₃ or H₂O using a Jasco DIP-1000 digital polarimeter (λ = 589 nm). The FI-MASS spectroscopy measurement was performed using a JEOL JMS-SX102A mass spectrometer (GC-MS and NMR Laboratory, Graduate School of Agriculture, Hokkaido University). The UV-vis spectra were measured at 70 °C with a 10 mm path length using a JASCO V-550 spectrometer with a deuterium lamp as the light source for the UV range (190–350 nm) and a halogen lamp for the visible range (330–900 nm). The dynamic light scattering (DLS) measurement of the obtained Au particles was performed at 25 °C using an Otsuka Electronics FDLS-3000 light scattering spectrophotometer equipped with a solid state laser (λ = 532 nm, scattering angle = 90°). Before the measurement, the samples were filtered using 0.45 μm PTFE membrane filters to eliminate any dust particles. The data analysis was carried out using the histogram methods including the Marquadt analysis.

Anionic polymerization of 1

All the anionic polymerizations of 1 were carried out using a storage flask with a needle valve under an argon atmosphere. A typical polymerization procedure (entry 2 in Table 1) is as follows: 1.0 mol L⁻¹t-BuOK in a THF solution (0.50 mL, 0.50 mmol, [1]/[t-BuOK] = 20) was added to a solution of the monomer 1 (2.02 g, 10.0 mmol) in dry THF (4.50 mL, [1] = 2.0 mol L⁻¹) at 50 °C using a microsyringe. After 600 h, the reaction mixture was poured into a large amount of methanol, then the solution was neutralized with dry ice. After evaporating the solvent, the residue was purified by reprecipitation from methanol–water to yield the crude polymer. The precipitate was redissolved in CHCl₃ (15 mL) and applied to the preparative SEC to remove the unreacted monomer 1. The concentrated polymer solution was dried in vacuo to give a white solid polymer 2 in a 68.4% yield (1.38 g); M_w,SEC (THF) = 1400; M_w,SEC/M_n,SEC (THF) = 1.28; M_w,SEC-MALLS (THF) = 7400; M_w,SEC-MALLS/M_n,SEC-MALLS (THF) = 1.37; dn·dc⁻¹ (THF) = 0.080 mL g⁻¹; [α]_D = −23.1° (c 1.00, CHCl₃, 24 °C). ¹H NMR (CD₂Cl₂):δ 5.87 (br.s, H-1), 4.80–3.20 (m, including peaks at 4.59, 4.48, 4.26, 4.01, 3.73, 3.60, and 3.47 ppm), 3.06 (br.s), 2.93–2.70 (m, epoxy group), 1.44 (s, =C(CH₃)₂) and 1.28 ppm (s, =C(CH₃)₂). ¹³C NMR (CD₂Cl₂):δ 112.14 (=C(CH₃)₂), 105.74 (C-1), 105.44 (C-1), 85.55, 84.20, 82.31, 80.46, 74.65, 73.45, 68.60, 48.49 (epoxy group), 46.98 (epoxy group), 26.96 (=C(CH₃)₂), and 26.35 ppm (=C(CH₃)₂). It is hard to estimate the degree of branching value because the NMR peaks due to each branch structure were broadened and overlapped.

Table 1 The polymerization results of 1

Entry	Cat.	Temp/°C	Polymerization time/h	Addition time/h	yield (%)	M w,SEC (Mw/Mn) ^c	M w,SEC-MALLS (Mw/Mn) ^d	α	[η] ^d/mL g^−1	K ^d/mL g^−1
a Conditions: Ar atmosphere; [M]/[I] = 20; solvent, THF; [M] = 2.0 mol L^−1. b Condition: Ar atmosphere; [M]/[I] = 40; solvent, CH2Cl2; [M] = 0.5 mol L^−1. c Determined by SEC using polystyrene standards in THF. d The weight average molecular weight Mw,SEC-MALLS, polydispersity Mw,SEC-MALLS/Mn,SEC-MALLS, intrinsic viscosity [η], and Mark–Houwink–Sakurada constants, α and K ([η] = KMw,SEC-MALLS^α), were determined by SEC in THF equipped with MALLS and viscosity detectors.
1	t-BuOK^a	60	160	—	52.9	2000 (1.26)	7700 (1.22)	0.27	4.6	0.40
2	t-BuOK^a	50	600	—	68.4	1400 (1.28)	7400 (1.37)	0.25	4.0	0.48
3	t-BuOK^a	50	770	—	69.5	1600 (1.34)	8900 (1.33)	0.27	4.3	0.38
4	BF3·OEt2^b	10	24	—	68.5	4100 (1.29)	15 000 (1.45)	0.11	3.4	1.16
5	BF3·OEt2^b	10	24	4.4	73.4	5600 (1.32)	37 000 (1.19)	0.09	3.3	1.22
6	BF3·OEt2^b	10	24	13.3	75.5	6500 (1.33)	81 300 (1.23)	0.09	3.4	1.26
7	BF3·OEt2^b	10	24	17.8	66.1	8600 (1.26)	122 400 (1.70)	0.08	3.8	1.42

---

Cationic polymerization of 1

All procedures were performed using a Schlenk flask under an argon atmosphere. A typical procedure for the cationic polymerization (entry 4 in Table 1) is as follows: BF₃·OEt₂ (8.0 μL, 62.5 μmol, [1]/[BF₃·OEt₂] = 40) was added to a solution of 1 (0.505 g, 2.50 mmol) in dry CH₂Cl₂ (5.0 mL, [1] = 0.5 mol L⁻¹) at 10 °C using a microsyringe. After 24 h, the polymerization was terminated by adding methanol containing a small portion of an aqueous ammonia solution. After evaporating the solvent, the residue was purified by reprecipitation from methanol–water to yield the crude polymer. The precipitate was redissolved in CHCl₃ (4 mL) and applied to the preparative SEC to remove the unreacted monomer 1. The concentrated polymer solution was dried in vacuo to give a white solid polymer 2 in a 68.5% yield (346 mg); M_w,SEC (THF) = 4100; M_w,SEC/M_n,SEC (THF) = 1.29; M_w,SEC-MALLS (THF) = 15 000; M_w,SEC-MALLS/M_n,SEC-MALLS (THF) = 1.45; dn·dc⁻¹ (THF) = 0.079 mL g⁻¹; [α]_D = −50.6° (c 1.00, CHCl₃, 24 °C). The ¹H and ¹³C NMR spectra of 2 (entry 4) are almost similar to those of entry 7.

Cationic polymerization of 1 using slow-monomer-addition strategy

A typical procedure for the cationic polymerization (entry 7 in Table 1) using the slow-monomer-addition strategy⁸⁷ is as follows: BF₃·OEt₂ (32.0 μL, 250 μmol, [1]/[BF₃·OEt₂] = 40) was added to a solution of 1 (0.404 g, 2.00 mmol) in dry CH₂Cl₂ (4.0 mL, [1] = 0.5 mol L⁻¹) at 10 °C using a microsyringe. A dry CH₂Cl₂ (16.0 mL, [1] = 0.5 mol L⁻¹) solution of 1 (1.62 g, 8.00 mmol) in a microsyringe was then slowly added to the mixture at 10 °C using a dosing pump. The drip rate was 15.0 μL min⁻¹, i.e., the addition time of 1 was 17.8 h. After 24 h, the polymerization was terminated by adding 400 mL of methanol containing a small portion of an aqueous ammonia solution. After evaporating the solvent, the residue was purified by reprecipitation from methanol–water to yield the crude polymer. The precipitate was redissolved in CHCl₃ (15 mL) and applied to the preparative SEC to remove the unreacted monomer 1. The concentrated polymer solution was dried in vacuo to give a white solid polymer 2 in a 66.1% yield (1.34 g); M_w,SEC (THF) = 8,600; M_w,SEC/M_n,SEC (THF) = 1.26; M_w,SEC-MALLS (THF) = 122,400; M_w,SEC-MALLS/M_n,SEC-MALLS (THF) = 1.70; dn·dc⁻¹ (THF) = 0.074 mL g⁻¹; [α]_D = −22.2° (c 1.00, CHCl₃, 25 °C). ¹H NMR (CD₂Cl₂):δ 5.87 (br.s, H-1), 5.20–2.40 (br, including peaks at 4.47, 4.28, 4.18, 3.99, 2.91(epoxy group), 2.81 (epoxy group), 2.66 and 2.63 ppm), 1.44 (s, =C(CH₃)₂), 1.28 ppm (s, =C(CH₃)₂), and 0.89 ppm. ¹³C NMR (CD₂Cl₂):δ 111.95 (=C(CH₃)₂), 106.83 (C-1), 105.40 (C-1), 101.49, 86.00–63.00 (m, including peaks at 85.62, 84.24, 82.16, 79.80, 75.35, 72.03, 69.39, 68.38), 27.05 (=C(CH₃)₂), and 26.42 ppm (=C(CH₃)₂). It is hard to estimate the degree of branching value because the NMR peaks due to each branch structure were broadened and overlapped.

Methylation of polymer 2

Silver oxide (0.50 g, 2.20 mmol) and methyl iodide (3.00 mL, 48.6 mmol) were added to polymer 2 (78.2 mg, 0.39 unit-mmol, entry 2 in Table 1) in 5 mL of dry acetonitrile. After the mixture was refluxed for 21 h, the insoluble parts were removed by filtration and the solvent was evaporated. The residue was redissolved in CHCl₃ (1 mL) and applied to the preparative SEC to prepare the resulting polymer. The concentrated polymer solution was dried in vacuo to give a white solid polymer 4 in an 80.1% yield (67.0 mg); The degree of methylation, which was estimated from the ¹H NMR measurement, was ca. 100%. M_w,SEC-MALLS (THF) = 4200; M_w,SEC-MALLS/M_n,SECMALLS (THF) = 1.49; dn·dc⁻¹ (THF) = 0.076 mL g⁻¹; [η] = 4.7 mL g⁻¹; the Mark–Houwink–Sakurada exponents, α and K, are 0.22 and 0.90 mL g⁻¹, respectively. ¹H NMR (CD₂Cl₂):δ 5.83 (br.s, H-1), 4.55 (br.s), 4.30–3.20 (m, including peaks at 4.11, 3.93, 3.89, 3.77, 3.61, 3.60, 3.48 (OCH₃), 3.46 (OCH₃) and 3.44 ppm (OCH₃)), 3.00–2.70 (m, including br.s at 2.91, 2.86, and 2.76 ppm, epoxy group) 1.45 (s, =C(CH₃)₂), and 1.29 ppm (s, =C(CH₃)₂). ¹³C NMR (CD₂Cl₂):δ 111.60 (=C(CH₃)₂), 105.06 (C-1), 83.60–80.50 (m, including peaks at 83.51, 82.05, and 81.28 ppm), 78.50, 76.90–75.00 (m, including peak at 76.75 ppm), 74.00–67.00 (m, including peaks at 72.00, 69.97, and 67.50 ppm), 59.50–57.00 (m, including peaks at 58.43 and 57.64 ppm, OCH₃), 48.15, 46.49, 29.68, 26.77 (=C(CH₃)₂), and 26.28 ppm (=C(CH₃)₂).

Synthesis of 5,6-anhydro-1,2-O-isopropylidene-3-O-methyl-α-D-glucofuranose (5)

Silver oxide (4.60 g, 19.8 mmol) and methyl iodide (1.20 mL, 19.8 mmol) were added to 1 (1.95 g, 9.64 mmol) in 5 mL of dry acetonitrile. After the mixture was refluxed for 20 h, the insoluble parts were removed by filtration and the solvent was evaporated. The residue was purified by column chromatography on silica gel with n-hexane/ethyl acetate (2/1, v/v; R_f 0.31) to afford colorless liquid 5. Yield: 1.92 g (92.1%). ¹H NMR (CDCl₃):δ 5.88 (d, J = 3.2 Hz, 1H, H-1), 4.58 (m, 1H, H-2), 3.82 (m, 1H, H-4), 3.68 (m, 1H, H-3), (s, 3H, OCH₃), 3.46, 3.20 (m, 1H, H-5), 2.88 (m, 1H, H-6), 2.74 (m, 1H, H-6), 1.43 (s, 3H, C(CH₃)₂) and 1.29 ppm (s, 3H, C(CH₃)₂). ¹³C NMR (CDCl₃):δ 111.9 (C(CH₃)₂), 105.3 (C-1), 84.4 (C-2), 82.0 (C-3), 81.6 (C-4), 58.4 (OCH₃), 48.1 (C-5), 46.9 (C-6), 26.9 (C(CH₃)₂) and 26.3 ppm (C(CH₃)₂). Anal. Calcd for C₁₀H₁₆O₅ (216.23): C, 55.55; H, 7.46. Found: C, 55.24; H, 7.31. FI-MS (m/z): 216 (M⁺).

Anionic polymerization of 5

The anionic polymerizations of 5 were carried out using a storage flask with a needle valve under an argon atmosphere. 1.0 mol L⁻¹t-BuOK in a THF solution (0.30 mL, 0.30 mmol, [1]/[t-BuOK] = 20) was added to a solution of 5 (1.30 g, 6.00 mmol) in dry THF (2.70 mL, [1] = 2.0 mol L⁻¹) at 60 °C using a microsyringe. After 180 h, the reaction mixture was poured into a large amount of methanol, and the solution was neutralized with dry ice. After evaporating the solvent, the residue was purified by reprecipitation from chloroform–n-hexane to yield the crude polymer. The precipitate was redissolved in CHCl₃ (9 mL) and applied to the preparative SEC to remove the unreacted monomer. The concentrated polymer solution was dried in vacuo to give a white solid polymer, poly(5,6-anhydro-1,2-O-isopropylidene-3-O-methyl-α-D-glucofuranose) (6), in a 65.9% yield (857 mg); M_w,SEC-MALLS (THF) = 11 100; M_w,SEC-MALLS/M_n,SEC-MALLS (THF) = 1.26; dn·dc⁻¹ (THF) = 0.062 mL g⁻¹; [η] = 6.8 mL g⁻¹; Mark–Houwink–Sakurada exponents α and K are 0.50 and 0.06 mL g⁻¹, respectively. [α]_D = −51.0° (c 1.00, CHCl₃, 25 °C). ¹H NMR (CDCl₃):δ 5.93 (d, J = 3.2 Hz, H-1), 5.83, 4.54, 4.49 (d, J = 3.7 Hz, H-2), 4.27 (d, J = 8.7 Hz, H-6), 4.01 (br.s, H-3 and H-4), 3.87, 3.76 (br.s, H-5), 3.53 (br.s, H-6) 3.46 (s, 3H, OCH₃), 1.45 (s, C(CH₃)₂), 1.27 (s, C(CH₃)₂) and 1.17 ppm. ¹³C NMR (CDCl₃):δ 111.03, 110.95 (C(CH₃)₂), 105.01 (C-1), 82.95 (C-4), 81.76 (C-2), 79.10 (C-3), 75.65 (C-5), 74.37 (C-6), 72.63, 57.61 (OCH₃), 27.58, 26.76 (C(CH₃)₂) and 26.30 (C(CH₃)₂).

Hydrolysis of polymer 2

Polymer 2 (0.60 g, 2.97 unit-mmol, entry 7) was dissolved in 5 mL of a 85/15 (v/v) trifluoroacetic acid/water mixture. The solution was magnetically stirred at room temperature for 18 h, and diluted with 40 mL of water. The solution was neutralized using an aqueous ammonia solution, then dialyzed in water for several days using a Spectra/Por^® CE (cellulose Ester) membrane (MWCO: 1000 Dalton, SPECTRUM). After freeze-drying from water, the residue was redissolved in water (5 mL) and applied to the preparative SEC to purify the resulting polymer. After freeze-drying from water, the hyperbranched 5,6-glucan (3) was isolated as a white powder. The yield was 62.7% (335 mg, entry 9). M_w,SEC-MALLS (0.2 mol L⁻¹ NaNO₃ aq.) = 74 400; M_w,SEC-MALLS/M_n,SEC-MALLS (0.2 mol L⁻¹ NaNO₃ aq.) = 2.74; dn·dc⁻¹ (0.2 mol L⁻¹ NaNO₃ aq.) = 0.135 mL g⁻¹; [α]_D = +17.3° (c 1.00, H₂O, 25 °C). ¹H NMR (entry 9, D₂O including acetone as internal reference):δ 6.05 (br.s), 5.49 (br.s, α-franosyl unit, H-1), 5.23 (br.s, β-franosyl and α-pyranosyl units, H-1), and 5.00–3.20 ppm (m, including peaks at 4.64 (br.s, β-pyranosyl unit, H-1), 4.28, 4.12, 3.96, 3.71, 3.56, 3.48, 3.26), 1.44, and 1.23 ppm. ¹³C NMR (entry 9, D₂O including acetone as an internal reference): δ 102.94 (β-franosyl unit, C-1), 97.30 (α-franosyl unit, C-1), 96.58 (β-pyranosyl unit, C-1), 92.74 (α-pyranosyl unit, C-1), and 88.00–62.00 ppm (m, including signals at 80.92, 77.83, 76.28, 75.48, 74.68, 73.33, 72.04, and 70.45 ppm). For entry 8: [α]_D + 53.6° (c 1.00, H₂O, 25 °C). ¹H NMR (entry 8, D₂O including acetone as internal reference):δ 6.01 (br.s), 5.48 (br.s, α-franosyl unit, H-1), 5.22 (br.s, β-franosyl and α-pyranosyl units, H-1), 4.65 (br.s, β-pyranosyl unit, H-1) and 4.50–3.20 ppm (m, including peaks at 4.33, 4.28, 4.12, 4.08, 3.99, 3.60, 3.52, 3.49, 3.47, 3.44, 3.40, 3.34, 3.33, 3.25), 1.52, and 1.36 ppm. ¹³C NMR (entry 8, D₂O including acetone as internal reference): δ 103.24 (C-1), 102.80 (β-franosyl unit, C-1), 97.31 (α-franosyl unit, C-1), 96.56 (β-pyranosyl unit,C-1), 92.72 (α-pyranosyl unit, C-1), 85.62, 82.97, 80.89, 80.38, and 79.00–68.00 ppm (m, including signals at 77.65, 76.43, 76.25, 76.05, 75.68, 75.49, 74.62, 74.24, 74.15, 73.27, 72.21, 72.01, 71.68, 71.60, 71.37, 71.03, 70.56, and 69.90 ppm).

Formation of gold nanoparticles

A typical procedure for the formation of the gold particles is as follows (run 1 in Fig. 15): polymer 3 (37.5 mg, 0.231 unit-mmol, entry 8) was first dissolved in deionized (Mill-Q) water (5 mL). The polymer solution (46 unit-mmol L⁻¹, 1.8 mL) was filtered, then kept at 70 °C for 30 min in the UV-cell equipped with a stirring bar. The tetrachloroauric acid (HAuCl₄) aqueous solution (0.01 mol L⁻¹, 6.67 μL) was added to the polymer solution. The UV-vis spectra of the polymer–metal solution were measured in the wavelength range of 450–600 nm at 1 min intervals with the polymer solution as a reference. The obtained absorbance data were corrected by removing the absorption of the HAuCl₄ aqueous solution.

Results and discussion

Polymerization

The ring-opening polymerizations of 5,6-anhydro-1,2-O-isopropylidene-α-D-glucofuranose (1) were carried out using potassium tert-butoxide (t-BuOK) as the anionic initiator or boron trifluoride diethyletherate (BF₃·OEt₂) as the cationic initiator. The typical results are summarized in Table 1. When t-BuOK was used in dry tetrahydrofuran (THF) as the solvent, the reaction mixtures immediately changed from colorless to dark brown as the reaction progressed, and the polymerization homogeneously proceeded until the end of the polymerization, while the cationic polymerization using BF₃·OEt₂ in dry methylene chloride (CH₂Cl₂) homogeneously proceeded without any color change. The polymers obtained using t-BuOK or BF₃·OEt₂ were a light-yellow solid or a white solid, respectively, which were soluble in chloroform, THF, DMF and methanol, but insoluble in water. The polymer yields for the anionic polymerization (entries 1–3) were in the range of 52.9 to 69.5% and the increase in conversion was very slow, e.g., 52.9% at 60 °C for 160 h (entry 1), 68.4% at 50 °C for 600 h (entry 2) and 69.5% at 50 °C for 770 h (entry 3). The weight-average molecular weights (M_w,SEC) measured by size exclusion chromatography (SEC) of the resulting polymers ranged from approximately 1400 to 2000. In contrast, the polymer yield for the cationic polymerization (entry 4) was 68.5% for 24 h and the M_w,SEC was 4100, indicating that the cationic polymerization is much better than the anionic one.⁸⁸

In general, highly branched polymers are known to have spherical conformations in a solution, and the M_w,SECs of these polymers were often claimed to be too low because the hydrodynamic volumes of these polymers are smaller than the corresponding linear polymers used for the calibration. Therefore, the absolute weight-average molecular weight (M_w,SEC-MALLS) of the resulting polymers was also measured using SEC equipped with a multiangle laser light scattering instrument (MALLS), and was compared with the M_w,SEC values in Table 1. The M_w,SEC-MALLS values for entries 1–4 ranged from 7400 to 15 000, which were ca. 1.6–5.5 those of the M_w,SEC values. These results suggest that the resulting polymer has a compact form in solution because of its branched structure.

From the viscosity measurement in THF using SEC equipped with the viscosity detector, the intrinsic viscosities ([η]) of the resulting polymers were observed. The solution viscosities of the polymers obtained using t-BuOK were very low in the range from 4.0 to 4.6 mL g⁻¹. The M_w,SEC-MALLS dependence of [η] for the polymer (entry 2) is shown in Fig. 2. On the basis of these results, the exponents (α) of the Mark–Houwink–Sakurada equations ([η] = KM^α) of the polymers obtained using t-BuOK were determined to be 0.25–0.27. In general, α = 0.5 suggests that the polymer behaves as dense spheres, α = 0.6–0.8 for a flexible chain and α > 1 for an elongated rod, whereas it is well-known that the α value was less than 0.5 for the various hyperbranched polymers. Our experimental α values were noticeably low, which are ascribed to the compact spherical structure in solution. Therefore, the SEC, MALLS, and viscosity measurement results suggested that the polymer obtained using t-BuOK was a branched spherical molecule, i.e., a hyperbranched poly(5,6-anhydro-1,2-O-isopropylidene-α-D-glucofuranose) (2).

Fig. 2 Double logarithmic plots of [η] vs.M_w,SEC-MALLS for polymer 2 in THF (entry 2).

The solution viscosity of the polymer (entry 4) obtained using BF₃·OEt₂ was 3.4 mL g⁻¹, which was lower than those of the polymers obtained using t-BuOK, regardless of the higher molecular weight polymer. In addition, the α value was 0.11, which was less than those of the polymers obtained using t-BuOK. These results indicate that the polymer obtained using BF₃·OEt₂ should have a more compact spherical structure in solution than the polymer obtained using t-BuOK, i.e., the polymer obtained using BF₃·OEt₂ should have a higher degree of branching value than the polymer obtained using t-BuOK.

As an efficient preparation method for hyperbranched polymers with both a high molecular weight and a high branching value degree, the slow-monomer-addition strategy⁸⁷ was used for the cationic polymerization of 1. According to the established time of monomer addition, 16 mL of the 0.5 mol L⁻¹ monomer solution was slowly added to the mixture of 4 mL of monomer solution and BF₃·OEt₂ at 10 °C using a dosing pump. These results are summarized in entries 5–7 of Table 1. The slow-monomer-addition strategy results in a very low monomer concentration actually present in the reaction mixture. Thus, the exclusive reaction of the monomer with the growing polyfunctional macromolecules occurs, resulting in a high molecular weight and a high degree of branching. Based on a longer addition time, i.e., a decrease in the drop rate, the M_w,SEC-MALLS values dramatically increase up to 122,400 (entry 7). Interestingly, regardless of the higher molecular weight polymer, the [η] values were in the range from 3.3 to 3.8 mL g⁻¹, which was similar to the value for entry 4. The α values of the Mark–Houwink–Sakurada equations were determined to be very low in range of 0.08 to 0.09, suggesting that the resulting polymers have a highly branched spherical structure. In addition, the polydispersities (M_w,SEC/M_n,SEC) were found to be in the relatively narrow range of 1.26–1.33. These results indicate that the slow-monomer-addition strategy is a facile method for preparing the hyperbranched polymer 2 with a controlled molecular weight and highly branched structure.

Polymer structure

In order to investigate the polymer structure, the ¹H and ¹³C NMR measurements were carried out in CD₂Cl₂. Parts a and b of Fig. 3 show the ¹H NMR spectra of the polymers obtained using t-BuOK (entry 2) and BF₃·OEt₂ (entry 7), respectively. Though the unreacted monomer was removed using the preparative SEC, in both spectra, the peaks due to the epoxy groups slightly exist in the range of 2.6–3.0 ppm as the starting polymer end, which is a characteristic of a hyperbranched polymer. From a comparison of both spectra, the spectrum of the polymer obtained using t-BuOK is obviously different from that of the polymer using BF₃·OEt₂, recognizing that the structure of the former is more regular than that of the latter because of distinct splitting of the peaks and narrow peak width for the anionically obtained polymer. This trend is also recognized by the comparison of the ¹³C NMR spectra in Fig. 4. Ring-opening of the monosubstituted epoxide occurs in two ways, i.e., by α- or β-scission. In general, the polymerization using an anionic catalyst predominantly cleaves the CH₂–O bond (β-scission) via an SN₂ displacement to form the regular head-to-tail linkage. Therefore, the difference in the spectra is mainly caused by the stereoregularity of the C-5 carbon in the polymer. Though the anionically obtained polymer has higher stereoregularity, the ¹³C NMR spectrum contains many peaks due to the C-6 methylene carbons, as shown in Fig. 4c, indicating that the polymer obviously contains some branched structures as irregular structure.

Fig. 3 ¹H NMR spectra of polymer 2 in CD₂Cl₂ (TMS as the internal standard), (a) entry 2 and (b) entry 7.

Fig. 4 ¹³C NMR spectra of polymer 2 in CD₂Cl₂ (TMS as internal standard), (a) entry 2, (b) entry 7, and (c) DEPT-135 spectrum of entry 2.

In order to investigate the branching structure in polymer 2, the ¹H and ¹³C NMR spectra of the methylated hyperbranched polymer (4) were compared with those of the linear analog, poly(5,6-anhydro-1,2-O-isopropylidene-3-O-methyl-α-D-glucofuranose) (6). The polymer 4 was prepared by the methylation of the anionically obtained polymer 2 (entry 2) using CH₃I and Ag₂O in dry CH₃CN, as shown in Fig. 5. On the other hand, the linear polymer 6 was obtained by the anionic polymerization of 5,6-anhydro-1,2-O-isopropylidene-3-O-methyl-α-D-glucofuranose (5) using t-BuOK, as shown in Fig. 6. Parts a and b of Fig. 7 show the ¹H NMR spectra of 4 and 6, respectively. The ¹H NMR spectrum of 4 consists of some broad peaks, which is obviously different from that of 6. In particular, the methoxy peaks of 4 were split into three or more, indicating that there are some methoxy groups in different surroundings, as shown in Fig. S1 (see ESI†). A comparison of the ¹³C NMR spectra in Fig. S2 (see ESI†) also shows the obvious difference. The ¹³C NMR spectrum of 6 mainly consists of 10 sharp peaks corresponding to the repeating unit, while the spectrum of 4 includes some broad peaks which are due to the irregular structure. In addition, the Mark–Houwink–Sakurada exponent α of 4 was 0.22, whereas the α value of 6 was 0.50. The NMR and viscosity results strongly support the fact that polymer 4 obviously had some branched structure as an irregular structure, i.e., the polymerization of 1 proceeded through the ring-opening multibranching mechanism to produce the hyperbranched poly(5,6-anhydro-1,2-O-isopropylidene-α-D-glucofuranose) 2.

Fig. 5 Methylation of polymer 2.

Fig. 6 Anionic polymerization of 5,6-anhydro-1,2-O-isopropylidene-3-O-methyl-α-D-glucofuranose (5).

Fig. 7 ¹H NMR spectra of (a) polymer 4 and (b) polymer 6 in CDCl₃ (TMS as the internal standard).

Polymerization mechanism

Fig. 8 and 9 represent the proposed anionic and cationic polymerization mechanisms, respectively. During the anionic polymerization, the ring-opening occurred through only β-scission to afford the linear unit with the regular head-to-tail linkage. The branches of the polymers should be produced by the proton transfer from the hydroxy group at C-3 to the other alkoxide and the following ring opening of 1 by the alkoxide at C-3. On the other hand, for the cationic polymerization, the protonated 1 reacted with the epoxy group of a second monomer, and two kinds of linkages through the α- or β-scission were formed as a linear structure. The branched structure should be produced by proton transfer from the hydroxyl groups in the polymer chains to the oxygen atom of the epoxy group. These reactions mentioned above simultaneously occurred in the polymerization systems and consequently produced the hyperbranched polymer 2.

Fig. 8 Proposed mechanism for the anionic polymerization of 1.

Fig. 9 Proposed mechanism for the cationic polymerization of 1.

Hydrolysis of 2

The hydrolysis of polymer 2 was carried out using a mixture of trifluoroacetic acid and water (85/15, v/v) to produce the hyperbranched poly(5,6-anhydro-α-D-glucofuranose) (3), e.g., the hyperbranched 5,6-glucan. The obtained polymers 3 were white powders which were soluble in water, and insoluble in chloroform, THF, DMF and methanol, which remarkably differed from the solubility of 2. Fig. 10 and 11 show the ¹H and ¹³C NMR spectra of 3 (entries 8 and 9) obtained from 2 (entries 2 and 7), respectively. Based on these spectra, it is concluded that the isopropylidene groups were mostly hydrolyzed. The physical data of the obtained 3 are listed in Table 2. The M_w,SEC-MALLS values for entries 8 and 9 are 3740 and 74 400, respectively, which were lower than the calculated M_w values. This indicates that some linkage may break under the hydrolysis conditions. However, the exponents, α, of the Mark–Houwink–Sakurada equations were determined to be 0.36 for entry 8 and 0.26 for entry 9, as shown in Fig. 12, indicating that the obtained polymer still have the branched structure. The ¹H and ¹³C NMR spectra of 3 consisted of very complicated peaks, which were constructed by the difference sequences of the D-glucose units. In the ¹³C NMR spectra, four distinct signals exist in the C-1 region. The big signals at 92.74 and 96.58 ppm in Fig. 11a, which are very similar to the chemical shifts of the carbons for α- and β-D-glucopyranoses,⁸⁹ should be assigned to the carbons of the α- and β-D-glucopyranose units in 3, respectively. In addition, the small signals at 97.30 and 102.94 ppm should be assigned to the carbons of the α- and β-D-glucofranose units in 3, respectively, since these signals are very similar to the chemical shifts of 3,5,6-tri-O-methyl α- and β-D-glucofranoses.⁹⁰ This result indicates that after hydrolysis, most of the D-glucofranose units in 2 changes to the D-glucopyranose unit in 3. As shown in Fig. 13, the D-glucofranose units with a hydroxyl group at C-5 could isomerize to the D-glucopyranose units, e.g., the 6-linked terminal and 3,6-linked linear D-glucofranose units isomerize to the corresponding D-glucopyranose units, while the 5,6-linked linear and 3,5,6-linked dendritic units do not change. Since the intensities of the D-glucopyranose units in Fig. 11 are stronger than those of D-glucofranose, polymer 3 includes numerous 6-linked terminal and 3,6-linked linear D-glucopyranose units, which is a characteristic of a hyperbranched polymer.

Table 2 The results of hydrolysis of 2 to prepare 3

Entry	Starting material 2	M w,SEC-MALLS of 3^a (Mw,SEC-MALLS/Mn,SEC-MALLS)	α	[η] ^a/mL g^−1	K /mL g^−1
a The weight average molecular weight Mw,SEC-MALLS, polydispersity Mw,SEC-MALLS/Mn,SEC-MALLS, intrinsic viscosity [η], and Mark–Houwink–Sakurada constants, α and K ([η] = KMw,SEC-MALLS^α), were determined by SEC in 0.2 mol L^−1 NaNO3 aqueous solution equipped with MALLS and viscosity detectors.
8	entry 2	3740 (1.26)	0.36	4.4	0.22
9	entry 7	74 400 (2.74)	0.26	5.0	0.29

---

Fig. 10 ¹H NMR spectra of polymer 3 in D₂O (acetone as internal standard), (a) entry 8 and (b) entry 9.

Fig. 11 ¹³C NMR spectra of polymer 3 in D₂O (acetone as internal standard), (a) entry 8 and (b) entry 9.

Fig. 12 Double logarithmic plots of [η] vs.M_w,SEC-MALLS for polymer 3 in 0.2 mol L⁻¹ NaNO₃ aqueous solution, (a) entry 8 and (b) entry 9.

Fig. 13 Possible isomerizations of polymer 3 to produce the D-glucopyranose units.

Formation of gold nanoparticles using 3

As mentioned in the previous section, polymer 3 is a novel hyperbranched glycopolymer arranged with numerous reducing D-glucose units on the peripheries of the hyperbranched polymer, whereas almost all of the glycodendrimers, star-shaped glycopolymers, and hyperbranched glycopolymers consist only of nonreducing sugar ends. Therefore, polymer 3 is a particularly interesting material, which could be called a “reducing sugar ball”. Polymer 3 probably has a higher reducing ability than D-glucose because of the glyco-cluster effect or the multivalent effect of the reducing D-glucose units. Such a hyperbranched glycopolymer with numerous reducing sugar units is also of considerable interest in a wide variety of fields such as biochemical and medicinal applications.

To estimate the reducing ability of 3, the formation of gold particles, which is derived from the reduction of HAuCl₄, is demonstrated in the aqueous solution of 3. Fig. 14 shows the UV-vis spectra after 150 min mixing of 3 and HAuCl₄ aqueous solutions at 70 °C (runs 1–3). As a comparative experiment, the UV-vis spectrum of the mixture of D-glucose and HAuCl₄ aqueous solutions (run 4) is also shown in Fig. 14.⁹¹ After mixing of the 3 and HAuCl₄ aqueous solutions, the color of the system changed from colorless to light red with time, as shown in Fig. S3 (see ESI†). The UV-vis spectra of the 3-HAuCl₄ system have a λ_max that ranged from 512 nm to 514 nm, which due to the formation of the gold particles. From previous studies,⁹² the average gold particle diameter can be estimated from the λ_max, i.e., gold particles of ca. 8–12 nm were obtained by the mixing of 3 and the HAuCl₄ aqueous solutions. Thus, the average hydrodynamic diameter of the resulting Au particles (run 2 in Fig. 14), which was measured by a dynamic light scattering (DLS) technique, was ca. 11.3 nm, as shown in Fig. S4 (see ESI†).

Fig. 14 UV-vis spectra after 150 min mixing of polymer 3 (or D-glucose) and HAuCl₄ at 70 °C. Run 1: polymer 3 (46 unit-mmol L⁻¹ of entry 8, 1.8 mL) and HAuCl₄ (0.01 mol L⁻¹, 6.67 μL) aqueous solutions; run 2: polymer 3 (25 unit-mmol L⁻¹ of entry 8, 1.8 mL) and HAuCl₄ (0.01 mol L⁻¹, 6.67 μL) aqueous solutions; run 3: polymer 3 (46 unit-mmol L⁻¹ of entry 9, 1.8 mL) and HAuCl₄ (0.01 mol L⁻¹, 6.67 μL) aqueous solutions; and run 4: D-glucose (46 unit-mmol L⁻¹, 1.8 mL) and HAuCl₄ (0.01 mol L⁻¹, 6.67 μL) aqueous solutions. The maximal wavelengths (λ_max) are 512 nm for run 1, 514 nm for run 2, 514 nm for run 3 and 526 nm for run 4.

To obtain the relationship between the formation of the gold particles and the reaction time, the increasing trend in the UV-vis absorbance at λ_max is presented in Fig. 15. For all the 3-HAuCl₄ systems (runs 1–3), the UV-vis absorbance sharply increases without an induction time and finally became a constant value within 10 min.^93,94 On the other hand, the D-glucose-HAuCl₄ system (run 4) shows a long induction time, then the absorbance sharply increases as an autocatalytic system. A similar autocatalytic tendency with a long induction time was also reported using the sugar-persubstituted poly(amidoamine) dendrimer, i.e., a sugar ball with numerous nonreducing sugar units.⁹⁵ The results indicate that 3 has a high reducing ability that originated from the high density of the reducing sugars in 3, i.e., a glyco-cluster effect or multivalent effect of the reducing sugars. Thus, the reaction rates within 10 min for the 3-HAuCl₄ systems were much faster than that for the D-glucose-HAuCl₄ system, as shown in Fig. 15b. This indicates that “reducing sugar ball” 3 effectively act as a sugar based reductant.

Fig. 15 The change in the UV absorption at λ_max for mixtures of polymer 3 (or D-glucose) and HAuCl₄ as a function of reaction time at 70 °C. Run 1 (○): polymer 3 (46 unit-mmol L⁻¹ of entry 8, 1.8 mL) and HAuCl₄ (0.01 mol L⁻¹, 6.67 μL) aqueous solutions at 512 nm, run 2 (●): polymer 3 (25 unit-mmol L⁻¹ of entry 8, 1.8 mL) and HAuCl₄ (0.01 mol L⁻¹, 6.67 μL) aqueous solutions 514 nm, run 3 (□): polymer 3 (46 unit-mmol L⁻¹ of entry 9, 1.8 mL) and HAuCl₄ (0.01 mol L⁻¹, 6.67 μL) aqueous solutions 514 nm, and run 4 (▲): D-glucose (46 unit-mmol L⁻¹, 1.8 mL) and HAuCl₄ (0.01 mol L⁻¹, 6.67 μL) aqueous solutions 526 nm. (a) Overall figure and (b) the expanded figure in the range of 0 to 10 min. The obtained absorbance data were corrected by removing the absorption of the HAuCl₄ aqueous solution.

Conclusions

In this study, we performed the anionic or cationic ring-opening multibranching polymerization of 5,6-anhydro-1,2-O-isopropylidene-α-D-glucofuranose (1) as a latent cyclic AB₂-type monomer. The polymerizations proceeded through the proton-transfer reaction mechanism to produce hyperbranched poly(5,6-anhydro-1,2-O-isopropylidene-α-D-glucofuranose) (2). In particular, the cationic polymerization with the slow-monomer-addition strategy is a facile method leading to the polymers with a high molecular weight and highly-branched structure. The weight-average molecular weight (M_w,SEC-MALLS) values of 2 measured by multi-angle laser light scattering (MALLS) varied in the range of 7400 to 122 400, which were significantly higher than the weight-average molecular weight (M_w,SEC) values by size exclusion chromatography (SEC). The intrinsic viscosities ([η]) of these polymers were very low in the range of 3.3–4.6 mL g⁻¹ and the Mark–Houwink–Sakurada exponent α was calculated to be 0.08–0.28. These results of the MALLS, SEC, and viscosity measurements suggested that these polymers exist in a compact spherical conformation in solution because of their highly-branched structure. A water-soluble hyperbranched 5,6-glucan (3) was also synthesized by the hydrolysis of polymer 2. Polymer 3 is a novel hyperbranched glycopolymer arranged with numerous reducing D-glucose units on the peripheries of the polymer, and has a high reducing ability derived from the glycol-cluster effect or the multivalent effect of the reducing D-glucose units, i.e., the “reducing sugar ball” 3 effectively acts as a sugar-based reductant. Such a hyperbranched glycopolymer with numerous reducing sugar units can be expected to be useful in a wide variety of fields such as biochemical and medicinal applications; for example, a biocompatible reductant, antioxidant, and a part of the carrier for a drug delivery system, etc.

Acknowledgements

This study was supported by the Industrial Technology Research Grant Program in 2005 from the New Energy and Industrial Technology Development Organization (NEDO) of Japan and the Research Fellowships of the Japan Society for the Promotion of Science (JSPS) for Young Scientists. The authors thank the OPEN FACILITY (Hokkaido University Sousei Hall) for the MALLS (THF) and DLS measurements.

References

1. C. J. Hawker and W. Devonport, ACS Symp. Ser., 1996, 624, 186.
2. J. M. J. Fréchet, C. J. Hawker, I. Gitsov and J. W. Leon, J. Macromol. Sci., Part A: Pure Appl. Chem., 1996, 33, 1399.
3. F. Zeng and S. C. Zimmerman, Chem. Rev., 1997, 97, 1681.
4. O. A. Matthews, A. N. Shipway and J. F. Stoddart, Prog. Polym. Sci., 1998, 23, 1.
5. Y. H. Kim, J. Polym. Sci., Part A: Polym. Chem., 1998, 36, 1685.
6. A. Hult, M. Johansson and E. Malmström, Adv. Polym. Sci., 1999, 143, 1.
7. M. Fischer and F. Vogtle, Angew. Chem., Int. Ed., 1999, 38, 884.
8. B. Voit, J. Polym. Sci., Part A: Polym. Chem., 2000, 38, 2505.
9. F. Vogtle, S. Gestermann, R. Hesse, H. Schwierz and B. Windisch, Prog. Polym. Sci., 2000, 25, 987.
10. K. Inoue, Prog. Polym. Sci., 2000, 25, 453.
11. N. Hadjichristidis, M. Pitsikalis, S. Pispas and H. Iatrou, Chem. Rev., 2001, 101, 3747.
12. M. Jikei and M.-A. Kakimoto, Prog. Polym. Sci., 2001, 26, 1233.
13. H. Frey and R. Haag, Rev. Mol. Biotechnol., 2002, 90, 257.
14. S.-E. Stiriba, H. Frey and R. Haag, Angew. Chem., Int. Ed., 2002, 41, 1329.
15. F. O. H. Pirrung, E. M. Loen and A. Noordam, Macromol. Symp., 2002, 187, 683.
16. W. B. Turnbull and J. F. Stoddart, Rev. Mol. Biotechnol., 2002, 90, 231.
17. H. Frey and R. Haag, Rev. Mol. Biotechnol., 2002, 90, 257.
18. D. A. Tomalia and J. M. J. Fréchet, J. Polym. Sci., Part A: Polym. Chem., 2002, 40, 2719.
19. I. Gitsov and K. R. Lambrych, Dendrimers. Synthesis and Applications, in Microspheres, Microcapsules & Liposomes, ed. R. Arshady, Citus Books, London, 2002, vol. 5, pp. 31–68.
20. S. Hecht, J. Polym. Sci., Part A: Polym. Chem., 2003, 41, 1047.
21. K. Ishizu, K. Tsubaki, A. Mori and S. Uchida, Prog. Polym. Sci., 2003, 28, 27.
22. C. Gao and D. Yan, Prog. Polym. Sci., 2004, 29, 183.
23. A. Hirao, M. Hayashi, S. Loykulnant, K. Sugiyama, S. W. Ryu, N. Haraguchi, A. Matsuo and T. Higashihara, Prog. Polym. Sci., 2005, 30, 111.
24. B. D. Mather, K. Viswanathan, K. M. Miller and T. E. Long, Prog. Polym. Sci., 2006, 31, 487.
25. X. Feng, D. Taton, R. Borsali, E. L. Chaikof and Y. Gnanou, J. Am. Chem. Soc., 2006, 128, 11551.
26. N. Hadjichristidis, H. Iatrou, M. Pitsikalis and J. Mays, Prog. Polym. Sci., 2006, 31, 1068.
27. T. Satoh, M. Tamaki, Y. Kitajyo, T. Maeda, H. Ishihara, T. Imai, H. Kaga and T. Kakuchi, J. Polym. Sci., Part A: Polym. Chem., 2006, 44, 406.
28. A. Hirao, K. Sugiyama and H. Yokoyama, Prog. Polym. Sci., 2007, 32, 1393.
29. Y. Kitajyo, T. Imai, Y. Sakai, M. Tamaki, H. Tani, K. Takahashi, A. Narumi, H. Kaga, N. Kaneko, T. Satoh and T. Kakuchi, Polymer, 2007, 48, 1237.
30. Y. Kitajyo, Y. Nawa, M. Tamaki, H. Tani, K. Takahashi, H. Kaga, T. Satoh and T. Kakuchi, Polymer, 2007, 48, 4683.
31. S. Peleshanko and V. V. Tsukruk, Prog. Polym. Sci., 2008, 33, 523.
32. T. Satoh, Soft Matter, 2009, 5, 1972.
33. K. L. Wooley, J. M. J. Fréchet and C. J. Hawker, Polymer, 1994, 35, 4489.
34. C. Schipp, M. J. Hill, P. J. Barham, V. M. Cloke, J. S. Higgins and L. Oiarzabalt, Polymer, 1996, 37, 2291.
35. C. J. Hawker, E. Malmström, C. W. Frank and J. P. Kampf, J. Am. Chem. Soc., 1997, 119, 9903.
36. M. W. Weimer, J. M. J. Fréchet and I. Gitsov, J. Polym. Sci., Part A: Polym. Chem., 1998, 36, 955.
37. P. G. Santangelo and C. M. Roland, Macromolecules, 1999, 32, 1972.
38. J.-I. Kadokawa, H. Tagaya and K. Chiba, Polym. Drugs Drug Delivery Syst., 2001, 251.
39. S.-E. Stiriba, H. Kautz and H. Frey, J. Am. Chem. Soc., 2002, 124, 9698.
40. M. Ballauff and C. N. Likos, Angew. Chem., Int. Ed., 2004, 43, 2998.
41. K. R. Kumar and D. E. Brooks, Macromol. Rapid Commun., 2005, 26, 155.
42. F. Samadi, J. Eckelt, B. A. Wolf, F.-J. Lopez-Villanueva and H. Frey, Eur. Polym. J., 2007, 43, 4236.
43. Y. Kitajyo, Y. Kinugawa, M. Tamaki, H. Kaga, N. Kaneko, T. Satoh and T. Kakuchi, Macromolecules, 2007, 40, 9313.
44. Y. Haba, C. Kojima, A. Harada and K. Kono, Angew. Chem., Int. Ed., 2007, 46, 234.
45. Y. Chi, S. T. Scroggins and J. M. J. Fréchet, J. Am. Chem. Soc., 2008, 130, 6322.
46. S. Javor, A. Natalello, S. M. Doglia and J.-L. Reymond, J. Am. Chem. Soc., 2008, 130, 17248.
47. K. Aoi, K. Itoh and M. Okada, Macromolecules, 1995, 28, 5391.
48. K. Aoi, K. Tsutsumiuchi, A. Yamamoto and M. Okada, Tetrahedron, 1997, 53, 15415.
49. K. Aoi, K. Tsutsumiuchi, A. Yamamoto and M. Okada, Macromol. Rapid Commun., 1998, 19, 5.
50. D. Zanini and R. Roy, J. Org. Chem., 1998, 63, 3486.
51. P. R. Ashton, E. F. Hounsell, N. Jayaraman, T. M. Nilsen, N. Spencer, J. F. Stoddart and M. Young, J. Org. Chem., 1998, 63, 3429.
52. H. W. I. Peerlings, S. A. Nepogodiev, J. F. Stoddart and E. W. Maijer, Eur. J. Org. Chem., 1998, 1879.
53. K. Tsutsumiuchi, K. Aoi and M. Okada, Polym. J., 1999, 31, 935.
54. S. André, P. J. C. Ortega, M. A. Perez, R. Roy and H.-J. Gabius, Glycobiology, 1999, 9, 1253.
55. J. J. Lundquist and E. J. Toone, Chem. Rev., 2002, 102, 555.
56. S. M. Rele, W. Cui, L. Wang, S. Hou, G. Barr-Zarse, D. Tatton, Y. Gnanou, J. D. Esko and E. L. Chaikof, J. Am. Chem. Soc., 2005, 127, 10132.
57. H. Baigude, K. Katsuraya, S. Tokunaga, N. Fujiwara, M. Satoyama, T. Magome, K. Okuyama, G. Borjihan and T. Uryu, J. Polym. Sci., Part A: Polym. Chem., 2005, 43, 2195.
58. A. Takasu, T. Makino and T. Hirabayashi, J. Polym. Sci., Part A: Polym. Chem., 2009, 47, 310.
59. B. W. Greatrex, S. J. Brodie, R. H. Furneaux, S. M. Hook, W. T. McBurney, G. F. Painter, T. Rades and P. M. Rendle, Tetrahedron, 2009, 65, 2939.
60. N. Jayaraman, S. A. Nepogodiev and J. F. Stoddart, Chem.–Eur. J., 1997, 3, 1193.
61. B. Colonna, V. D. Harding, S. A. Nepogodiev, F. M. Raymo, N. Spencer and J. F. Stoddart, Chem.–Eur. J., 1998, 4, 1244.
62. W. B. Turnbull, A. R. Pease and J. F. Stoddart, ChemBioChem, 2000, 1, 70.
63. W. B. Turnbull, S. A. Kalovidouris and J. F. Stoddart, Chem.–Eur. J., 2002, 8, 2988.
64. C. D. Heidecke and T. K. Lindhorst, Chem.–Eur. J., 2007, 13, 9056.
65. J. Bernard, A. Favier, L. Zhang, A. Nilasaroya, T. P. Davis, C. Barner-Kowollik and M. H. Stenzel, Macromolecules, 2005, 38, 5475.
66. J. Bernard, X. Hao, T. P. Davis, C. Barner-Kowollik and M. H. Stenzel, Biomacromolecules, 2006, 7, 232.
67. A. Narumi, S. Yamane, Y. Miura, H. Kaga, T. Satoh and T. Kakuchi, J. Polym. Sci., Part A: Polym. Chem., 2005, 43, 4373.
68. A. Narumi, H. Kaga, Y. Miura, T. Satoh, N. Kaneko and T. Kakuchi, Polymer, 2006, 47, 2269.
69. A. Narumi, H. Kaga, Y. Miura, I. Otsuka, T. Satoh, N. Kaneko and T. Kakuchi, Biomacromolecules, 2006, 7, 1496.
70. A. Narumi and T. Kakuchi, Polym. J., 2008, 40, 383.
71. J.-I. Kadokawa, M. Sato, M. Karasu, H. Tagaya and K. Chiba, Angew. Chem., Int. Ed., 1998, 37, 2373.
72. A. Kanazawa, S. Okumura and M. Suzuki, Org. Biomol. Chem., 2005, 3, 1746.
73. S. Muthukrishnan, G. Jutz, X. Andre, H. Mori and A. H. E. Müller, Macromolecules, 2005, 38, 9.
74. S. Muthukrishnan, H. Mori and A. H. E. Müller, Macromolecules, 2005, 38, 3108.
75. S. Muthukrishnan, F. Plamper, H. Mori and A. H. E. Müller, Macromolecules, 2005, 38, 10631.
76. A. Kanazawa and M. Suzuki, Polymer, 2006, 47, 176.
77. T. Satoh, T. Imai, H. Ishihara, T. Maeda, Y. Kitajyo, A. Narumi, H. Kaga, N. Kaneko and T. Kakuchi, Macromolecules, 2003, 36, 6364.
78. T. Imai, T. Satoh, H. Kaga, N. Kaneko and T. Kakuchi, Macromolecules, 2003, 36, 6359.
79. T. Imai, T. Satoh, H. Kaga, N. Kaneko and T. Kakuchi, Macromolecules, 2004, 37, 3113.
80. T. Satoh, T. Imai, Y. Kitajyo, T. Maeda, A. Narumi, H. Kaga, N. Kaneko and T. Kakuchi, Macromol. Symp., 2004, 217, 39.
81. T. Imai, Y. Nawa, Y. Kitajyo, T. Satoh, H. Kaga, N. Kaneko and T. Kakuchi, Macromolecules, 2005, 38, 1648.
82. T. Satoh, T. Imai, H. Ishihara, T. Maeda, Y. Kitajyo, Y. Sakai, H. Kaga, N. Kaneko, F. Ishii and T. Kakuchi, Macromolecules, 2005, 38, 4202.
83. T. Satoh and T. Kakuchi, Macromol. Biosci., 2007, 7, 999.
84. T. Uryu, K. Kitano and K. Matsuzaki, Makromol. Chem., 1979, 180, 1135.
85. T. Uryu, K. Kitano, H. Tachikawa, K. Ito and K. Matsuzaki, Makromol. Chem., 1978, 179, 1773.
86. T. Soler, A. Bachki, L. R. Falvello, F. Foubelo and M. Yus, Tetrahedron: Asymmetry, 2000, 11, 493.
87. A. Sunder, R. Hanselmann and H. Frey, Macromolecules, 1999, 32, 4240.
88. A similar polymerization tendency was reported by Uryu et al.^84,85.
89. The chemical shifts of the carbons for α- and β-D-glucopyranoses in D₂O are 93.18 and 97.00 ppm, respectively.
90. G. Just and D. Crosilla, Can. J. Chem., 1980, 58, 2349 The chemical shifts of the protons for 3,5,6-tri- O-methyl α- and β-D-glucofranoses in CDCl₃ are 5.40 and 5.10 ppm, respectively. Based on the ¹H NMR data of 3,5,6-tri- O-methyl α- and β-D-glucofranoses and CH-COSY spectrum of polymer 3 (entry 8) in Fig. S5 (see ESI†), the signals at 97.30 and 102.94 ppm in Fig. 11a are assigned to the carbons of the α- and β-D-glucofranose units in 3, respectively.
91. J. Liu, G. Qin, P. Raveendran and Y. Ikushima, Chem.–Eur. J., 2006, 12, 2131.
92. K. Nakamura, T. Kawabata and Y. Mori, Powder Technol., 2003, 131, 120.
93. In run 3 using the high molecular weight polymer 3 (entry 9), the intensity at λ_max is consistently rather weaker compared to run 1 using the low molecular weight sample (entry 8), probably because the sugar units on the peripheries of polymer 3 at the same unit-mol concentration is less for the high molecular weight polymer than the low molecular weight polymer, i.e., the inner sugar units of the high molecular weight polymer cannot effectively be utilized as a reductant because of the sugar units in the enclosed space.
94. The final intensity at λ_max for the D-glucose-HAuCl₄ system was greater than those for the 3-HAuCl₄ system at the same mol concentration. The main factor is attributed to the links between the sugar units in 3, which cannot act as a reductant.
95. K. Esumi, T. Hosoya, A. Suzuki and K. Torigoe, J. Colloid Interface Sci., 2000, 226, 346.

---

Footnote

† Electronic supplementary information (ESI) available: Possible repeating units of polymer 4; ¹³C NMR spectra of polymer 4 and polymer 6 in CDCl₃; demonstration pictures of the formation of gold particles using polymer 3 (entry 8); dynamic light scattering (DLS) measurements of Au particles; C–H COSY spectrum of polymer 3 (entry 8) in D₂O. See DOI: 10.1039/b9py00223e

---