Methyl glycosides via Fischer glycosylation: translation from batch microwave to continuous flow processing
A continuous flow procedure for the synthesis of methyl glycosides (Fischer glycosylation) of various monosaccharides using a heterogenous catalyst has been developed. In-depth analysis of the isomeric composition was undertaken and high consistency with corresponding results observed under microwave heating was obtained. Even in cases where addition of water was needed to achieve homogeneity—a prerequisite for the flow experiments—no detrimental effect on the conversion was found. The scalability was demonstrated on a model case (mannose) and as part of the target-oriented synthesis of d-glycero-d-manno heptose, both performed on multigram scale.
KeywordsGlycosides Heterogeneous catalysis Flow chemistry Carbohydrates
Although it had been mentioned as a promising scale-up option for corresponding microwave protocols , to the best of our knowledge, there has never been a study to test and demonstrate its feasibility. In contrast, more complex glycosylations with glycosyl donors and promotors have already been studied under conditions allowing rapid mixing and efficient removal of heat [28, 29, 30, 31, 32].
Expanding on our results in this successful case study, we set out to thoroughly and systematically investigate the methyl glycoside formation under Fischer glycosylation conditions in a flow regime.
Results and discussion
Screening of reaction conditions for the Fischer glycosylation of d-mannose in continuous flow
For the optimization of the flow process, d-mannose was selected as it features a strong preference for one isomer, the methyl α-pyranoside, under equilibrium conditions  which allows for an easier interpretation of how close to the equilibrium conditions a specific data point is. Further, it showed sufficient solubility in pure MeOH. We performed a screen of temperature and residence time by injecting plugs of a mannose stock solution into a bulk MeOH stream passing through a heated column reactor filled with QuadraPure™ sulfonic acid beads (QP-SA). Throughout this study, isomer analysis of the evaporated product streams was performed by 1H NMR through integration of diagnostic signals that had prior been assigned via 1H, 13C, and 2D-NMR experiments and/or comparison to relevant literature data (the diagnostic signals used are compiled in the supporting information).
It is noteworthy that at lower temperatures and/or shorter residence times small amounts of the reducing sugar were still observed (Table 1, entries 1–6). As expected, the composition is shifted towards higher percentage of the pyranosides, particularly the α-pyranoside α-4, when employing higher temperature and longer residence times; this correlates with the equilibrium conditions reported in the literature (Table 1, entry 13). However, only a small increase in the α-4 content was observed at 120 °C with prolonged residence times exceeding 4 min (Table 1, entries 9–12), which were the conditions chosen for the comparative experiments with the other sugars and for comparison with the corresponding microwave (µW) experiments (4 min, 120 °C).
Comparison of the Fischer glycosylation of various monosaccharides under continuous flow and microwave conditions
Demonstration of successful upscaling to multigram quantities
In the described work, we successfully demonstrated the Fischer glycosylation of various monosaccharides as a continuous flow process with a heterogenous acidic catalyst (QP-SA). High consistency between the ratios of formed products under continuous flow and the related batch-wise microwave conditions was shown. Under the optimized conditions, the addition of water for otherwise insoluble starting materials was tolerated and without detrimental effect on the observed product ratios. The confinement of the catalyst inside the reactor column simplifies downstream processing and allows for increasingly (with scale) better substrate/catalyst ratio in preparative experiments. The developed continuous flow setup offers the possibility of scale-up without any re-optimization which was demonstrated on selected examples.
All starting materials were purchased from commercial sources and used as received. A Biotage Initiator EXP EU Microwave Synthesizer was used for microwave-assisted synthesis. Thin-layer chromatography (TLC) was performed on aluminum sheets precoated with 60 F254 silica gel; visualization was accomplished by dipping with anisaldehyde/sulfuric acid and heating. 1H and 13C NMR spectra were recorded on a Bruker Avance 400 spectrometer or a Varian VNMRS-600. All spectra were recorded at ambient temperature (25 °C). Chemical shifts (δ) are quoted in ppm relative to tetramethylsilane and are referenced internally to the residual solvent peaks (CDCl3: δ = 7.26 ppm; D2O: δ = 4.79 ppm; MeOH-d4: δ = 3.31 ppm) . Assignments are based on APT, COSY, HSQC, and HMBC spectra. Melting points were measured in open ended glass capillary tubes with a Büchi B-540 melting point apparatus.
General procedure for the microwave-mediated Fischer glycosylation
A microwave-vial was charged with 30 mg of the sugar, 300 mg QP-SA, and 3 cm3 MeOH. The vial was capped and the sample was subjected to microwave irradiation (pre-stirring: 30 s, absorption level setting high) to 80–120 °C for 1–20 min. The reaction mixture was allowed to cool to ambient temperature, was filtered through a small pad of cotton wool and the solvent evaporated. Analysis was performed by 1H NMR spectroscopy.
Flow reactor setup and assembly
For continuous flow reactions, a Syrris Africa flow chemistry module was used for fluid management, fitted with an Omnifit® glass column reactor (100 mm × 10 mm) filled with acidic ion-exchange resin QuadraPure™ SA 450-800 micron (QP-SA; 2.5 g) which was heated in a dedicated aluminum block heated by a stirrer hot plate with associated thermosensor. The outlet flow from the reactor column was connected to a 5 bar back pressure regulator. The void volume of the reactor was measured via differential weighing of the reactor filled with dry QP-SA beads and the reactor filled with QP-SA beads and flooded with MeOH at 22 °C and determined to be 4.0 cm3. Consequently, an exemplary flow rate of 1 cm3/min equates to a theoretical residence time of 4 min.
General procedure for the Fischer glycosylation under continuous flow conditions
A methanolic sugar stock solution (2% w/v) was prepared—in case of residual insoluble material, the minimum amount of H2O was added and is indicated as v/v % in Table 2 (e.g., 5% H2O addition refers to addition of 200 mm3 H2O to 4 cm3 MeOH solution). For the optimization experiments, aliquots of this solution were filled into loops of 1 cm3 and were injected at flow rates corresponding to residence times of 2–10 min at 120 °C using methanol as the bulk solvent. The reactor was equilibrated to the conditions by flushing at least three reactor volumes with bulk solvent at the specific conditions, prior to injection. The outlet flow was collected (monitored by TLC) for subsequent NMR analysis.
Large scale continuous flow preparation of methyl α-d-mannopyranoside (4)
A stock solution of 12 g d-mannose (67 mmol) in MeOH (2% w/v) was pumped through a reactor column at a flow rate of 1 cm3/min at 120 °C for 10 h and the outlet flow was fed into a collecting vessel. Every hour of operation a sample of the flow output was evaluated by TLC (CHCl3/MeOH/H2O 7:3:0.5) and 1H NMR analysis. Evaporation of the bulk collected solution yielded 12.4 g of crude solid product (α-4/β-4 = 91:9 + 8% furanosides 2). Recrystallization from super-heated (100 °C) MeOH in the microwave oven gave colorless needle-shaped crystals upon cooling. After allowing to cool to ambient temperature and storing in the refrigerator (0 °C) overnight the colorless crystalline solid was filtered, washed with a small amount of cold MeOH, and dried in air to give pure methyl α-d-mannopyranoside 4 (9.4 g, 72%). M.p.: 191.1–192.8 °C (MeOH) (lit. 193 °C (EtOH) ). Spectral data matched those previously reported .
d-Glycero-α-d-manno-heptose hexaacetate (9)
Fischer trans-glycosylation in flow. The crude mixture of dd-manno and ll-gulo triols (maximum total content 40 mmol, 6:7 ~ 6:1)  was dissolved in 250 cm3 MeOH, filtered through a filter paper, and pumped through a packed bed reactor (15 g of QP-SA) at 90 °C with 1 cm3/min flow rate. The product solution was evaporated taken up in water and washed with DCM and Et2O until all apolar impurities were extracted from the aqueous layer (monitored by TLC). The aqueous layer was evaporated and analyzed by 1H NMR indicating a small proportion of remaining acetonide protection. Therefore, the material was taken up in 200 cm3 MeOH and passed through the same reactor under identical conditions as before achieving full cleavage of acetonides to methyl heptosides 8.
Acetylation and acetolysis. To the methyl heptoside mixture 8 first 100 cm3 Ac2O were added and the mixture was stirred for several minutes before 1 g H2SO4–SiO2  was added at rt. The reaction mixture started to warm and within 1 h the reaction mixture turned homogenous. When all material had dissolved, stirring was continued for an additional 30 min to allow the mixture to cool to rt. Then, 3 cm3 concentrated H2SO4 were added dropwise at rt and the reaction mixture was stirred at rt overnight. The reaction mixture was cooled with an ice bath and treated with 32 cm3 DIPEA (a change of color from violet to orange, pH ~ 5–7) and was stirred for 10 min before being diluted with EtOAc (200 cm3 in total) and washed with water (2 × 200 cm3), 1 M HCl (100 cm3, pH acidic) and water, NaHCO3, and brine, dried over Na2SO4 and evaporated, co-evaporated from toluene twice and once from EtOH and dried in vacuo to leave a crude material of 19 g of a sticky solid. The material was recrystallized from boiling EtOH (~ 20 cm3), crystallization while stirring furnished a colorless solid that was collected by filtration, washed with fresh cold EtOH and hexane to yield, the pure target compound 9 (12.2 g), according to 1H NMR with minor amounts of the β-anomer but without any indication for l-glycero-l-gulo isomers. An analytical sample was prepared by a second recrystallization (500 mg) from boiling EtOH (2 cm3, µW, 3 min, 100 °C) to yield large colorless crystals (430 mg) after filtration and washing with fresh cold EtOH. M.p.: 137.1–137.7 °C (EtOH) (lit. 138–139 °C (CHCl3) ); [α] D 20 = + 71 (c = 1.0, CHCl3) (lit. +66.5 (c = 2.5, CHCl3) ). Spectral data are consistent with those reported . 1H NMR (600 MHz, CDCl3): δ = 6.05 (d, J = 2.1 Hz, 1H, H1), 5.36–5.29 (m, 2H, H4, H3), 5.25–5.22 (m, 1H, H2), 5.18 (dt, J = 7.0, 3.4 Hz, 1H, H6), 4.41 (dd, J = 12.1, 3.6 Hz, 1H, H7a), 4.22 (dd, J = 12.1, 7.2 Hz, 1H, H7b), 4.11–4.03 (m, 1H, H5), 2.17, 2.16, 2.10, 2.07, 2.05, 2.01 (6 × s, 6 × 3H, 6 × COCH3) ppm; 13C NMR (151 MHz, CDCl3): δ = 170.6, 170.1, 170.0, 169.8 (2 ×), 168.1 (6 × COCH3), 90.5 (C1), 72.0 (C5), 70.2 (C6), 68.9 (C3), 68.3 (C2), 66.4 (C4), 61.6 (C7), 21.0, 20.94, 20.88, 20.86 (2 ×), 20.77 (6 × COCH3) ppm; HRMS (+ESI–TOF): m/z calcd. for C19H26NaO13 ([M +Na]+) 485.1271, found 485.1282.
Open access funding provided by Austrian Science Fund (FWF). Technical support by Alexander Pomberger and Markus Draskovits and financial support by the Austrian Science Fund FWF (J 3449-N28, P 29138-N34) is gratefully acknowledged.
- 7.Bohé L, Crich D (2014) In: Knochel P (ed) Comprehensive organic synthesis II, 2nd edn. Elsevier, AmsterdamGoogle Scholar
- 11.Collins PM, Ferrier RJ (1995) Monosaccharides: their chemistry and their roles in natural products. Wiley, ChichesterGoogle Scholar
- 14.Dean GR, Pyle RE (1952) Preparation of methyl glucoside. US Patent 2606186, Aug 5, 1952; (1953) Chem Abstr 47:839Google Scholar
- 15.Florent J-C, Monneret C (1987) J Chem Soc Chem Commun:1171Google Scholar
- 30.Ratner DM, Murphy ER, Jhunjhunwala M, Snyder DA, Jensen KF, Seeberger PH (2005) Chem Commun:578Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.