Open Access Article
Zhifei Chena,
Yibo Ninga,
Lei Lia,
Xueying Caoa,
Gaolei Xia,
Dongxu Chenga,
Qingfu Wang
*a,
Changtong Lu*a and
Kai Yang
*b
aTechnology Center, China Tobacco Henan Industrial Limited Company, Zhengzhou, 450002, P. R. China. E-mail: wqfhust@163.com; luchangtongfly@163.com
bSchool of Chemistry and Chemical Engineering, Shandong University of Technology, Zibo, 255049, P. R. China. E-mail: yangkai@sdut.edu.cn
First published on 10th February 2025
1,4:3,6-Dianhydro-α-D-glucopyranose (DGP) is one of important biomass pyrolysis anhydro sugar products that derive from the cellulose and hemicellulose components. There is no reliable method for the preparation of DGP at present, which contributed to its high cost with limited market supply and restricted applied research. In this study, we provided a novel method for the synthesis of DGP from methyl 3,6-anhydro-α-D-glucopyranoside for the first time. A mild and environmentally friendly synthetic approach for 3,6-anhydro glucopyranoside was developed via the intramolecular cyclization of 6-O-tosyl glucopyranoside, promoted by a catalytic amount of TBAF. And the preparation of DGP was achieved through the stabilization effect on carbocation intermediates by HFIP in the intramolecular cyclization of 3,6-anhydro glucopyranoside. Further sensory evaluation studies revealed that DGP had a sweetness similar to that of sucrose.
O or C
C unsaturated bonds (Scheme 1).1 This unique polyether cage structure endows DGP with good water solubility and potential applications in pharmaceutical chemistry and food.2,3 DGP was first isolated and identified by Tischenko et al. in 1948 from the gasification products of wood.4,5 And it is one of important biomass pyrolysis anhydro sugar products, such as levoglucosan (LG), 1-hydroxy-3,6-dioxabicyclo[3.2.1]octan-2-one (LAC), levoglucosenone (LGO), and DGP, that derived from the cellulose and hemicellulose components.6,7 The selective production and application in organic synthesis of LG, LAC, and LGO has been widely reported.8–15 The selective production of DGP from cellulose fast pyrolysis has recently been achieved for the first time by Lu et al., achieving a 14% yield at 400 °C.3,16 However, this process remained at the laboratory research stage and has not yet been applied in industry. Additionally, the synthesis of DGP through chemical synthesis method was only discovered by Wiersum et al. during the study of the high-temperature rearrangement reaction of isosorbide dinitrate, giving a mixture containing DGP with a yield of approximately 8% (Scheme 1a).17 Therefore, the current DGP in market is still produced by collecting and separating cellulose pyrolysis gas, contributing to the high cost of DGP with limited market supply (cost 5 mg > $200, through a comprehensive search on SciFinder), and thus its application has been rarely reported.3 While the long-term goal is to achieve large-scale, cost-effective production of DGP through cellulose fast pyrolysis, developing effective conventional chemical synthesis methods suitable for laboratory-scale preparation is crucial for promoting its application in food, pharmaceuticals, and other fields. Previously, our group discovered an unintentional side production of DGP with a 4% yield during the investigation of the acid-catalytic rearrangement of methyl 3,6-anhydro-α-D-glucopyranoside 1 for the conversion to its furanose form 2 (Scheme 1b). The structure of DGP was unambiguously confirmed by X-ray crystallographic analysis (ESI, Fig. S1†). Additionally, we have preliminarily determined that DGP possessed a certain degree of sweet taste. Inspired by our previous research on the taste identification of anhydro sugar 2,3-dihydro-3,5-dihydroxy-6-methyl-4H-pyran-4-one (DDMP),18,19 this work aimed to further optimize reaction conditions, develop an efficient chemical synthesis method for DGP and investigate the sweetness performance of DGP for the first time.
| Entry | Conditions | Time | 1 yieldb |
|---|---|---|---|
| a All reactions were performed on 0.2 mmol scale and using anhydrous TBAF unless otherwise stated.b Isolated yield.c Using TBAF·3H2O.d Yield < 5%. | |||
| 1 | TBAF (1.2 eq.), THF/DMF, 70 °C | 64 h | 62% |
| 2 | TBAF (0.1 eq.), Na2CO3(1.2 eq.), THF, reflux | 16 h | 86%/c85% |
| 3 | TBAF (0.1 eq.), K2CO3 (1.2 eq.), THF, reflux | 12 h | 83% |
| 4 | TBAF (0.1 eq.), NaHCO3 (1.2 eq.), THF, reflux | 30 h | 75% |
| 5 | TBAF (0.1 eq.), DIPEA (1.2 eq.), THF, reflux | 48 h | traced |
| 6 | TBAF (0.1 eq.), pyridine (1.2 eq.), THF, reflux | 48 h | traced |
| 7 | TBAF (0.1 eq.), Na2CO3 (1.2 eq.), DCE, reflux | 40 h | 63% |
| 8 | TBAF (0.1 eq.), Na2CO3 (1.2 eq.), toluene, reflux | 40 h | 50% |
| 9 | TBAF (0.1 eq.), Na2CO3 (1.2 eq.), MeCN, reflux | 16 h | 79% |
| 10 | Na2CO3 (1.2 equiv.), THF, reflux | 48 h | traced |
| 11 | TBAB (0.1 eq.), Na2CO3(1.2 eq.), THF, reflux | 48 h | traced |
| 12 | NaF (0.1 eq.), Na2CO3(1.2 eq.), THF, reflux | 48 h | traced |
6-O-Tosyl-glucopyranoside 3 was synthesized following methods described in the literature.21 Subsequently, the reaction conditions for the preparation of 1 were investigated in detail (Table 1). It was discovered that using 0.1 equivalents of TBAF in conjunction with bases like NaHCO3, Na2CO3, or K2CO3, efficiently afforded the target product in high yields (entries 2–4). Conversely, the use of organic bases such as DIPEA and pyridine resulted in negligible reaction progress (entries 5–6). Considering both reaction time and yield, Na2CO3 emerged as the most effective co-catalyst. Further investigation into the impact of the reaction solvent revealed THF as the optimal choice (entries 7–9). Additional experiments demonstrated that omitting TBAF or replacement by TBAB halted the reaction (entries 10–11), underscoring the crucial role of fluoride ion. TBAF exhibited superior catalytic activity in comparison of NaF (entries 12). The equivalent catalytic efficiency of the more economical TBAF.3H2O was also confirmed (entry 2c). In summary, the optimal reaction conditions were identified as using TBAF.3H2O (0.1 eq.), Na2CO3 (1.2 eq.), with reflux in THF under nitrogen atmosphere. The high tolerance for scale-up experiment was also further verified with starting material 4 of 50 grams, giving 27.2 grams 3,6-anhydro glucopyranoside 1 with similar yield, thus supporting subsequent research (Scheme 2).
Based on the aforementioned mechanistic analysis, the reaction conditions were further optimized to improve the selectivity of the DGP formation pathway and the results were summarized in Table 2. A series of Brønsted and Lewis acid catalysts in different solvents was investigated, while the yields were generally low disappointingly (entries 1–17). It was found that only product 2 was generated at room temperature, while the formation of DGP required higher temperatures and highly polar solvents. On the other hand, according to the transacetalization of 2-formylpyrrole acetals under strongly basic conditions in the literature,25 NaH was also explored as a potential catalyst. However, this investigation revealed no formation of the desired products DGP and 2 (entry 18).
| Entry | Conditions | Time | Yieldb: DGP/2 |
|---|---|---|---|
| a Reaction conditions: 3 (0.2 mmol), acid and solvent (2 mL) was stirred under argon atmosphere, and monitored by TLC until all 1 was consumed.b Isolated yield. | |||
| 1 | TFA (0.3 eq.), CHCl3, reflux | 1 h | 4%/85% |
| 2 | TFA (0.3 eq.), CHCl3, 25 °C | 48 h | 0/87% |
| 3 | TFA (0.3 eq.), MeCN, relux | 1 h | 10%/71% |
| 4 | TFA (0.3 eq.), DCE, relux | 1 h | Trace/67% |
| 5 | p-TsOH (0.3 eq.), MeCN, relux | 1 h | 9%/47% |
| 6 | p-TsOH (0.3 eq.), toluene, relux | 1 h | 5%/36% |
| 7 | p-TsOH (0.3 eq.), DCE, relux | 1 h | Trace/65% |
| 8 | p-TsOH (0.3 eq.), THF, relux | 1 h | Trace/44% |
| 9 | AcOH (1.0 eq.), CHCl3, reflux | 6 h | 0/Trace |
| 10 | HCOOH (0.3 eq.), CHCl3, reflux | 3 h | 0/55% |
| 11 | FeCl3 (0.3 eq.), MeCN, reflux | 8 h | 12%/49% |
| 12 | FeCl3 (0.5 eq.), MeCN, reflux | 8 h | 9%/52% |
| 13 | FeCl3 (0.1 eq.), MeCN, reflux | 16 h | 4%/66% |
| 14 | FeCl3 (0.3 eq.), DCE, reflux | 8 h | 0/60% |
| 15 | FeCl3 (0.3 eq.), THF, reflux | 8 h | 0/42% |
| 16 | BF3·Et2O (0.3 eq.), THF, 0 °C-reflux | 2 h | 0/21% |
| 17 | BF3·Et2O (0.3 eq.), MeCN, 0 °C-reflux | 2 h | 0/Trace |
| 18 | NaH (1.2 eq.), THF, reflux | 24 h | 0/0 |
| 19 | p-TsOH (0.3 eq.), TFE, reflux | 1 h | 9%/38% |
| 20 | p-TsOH (0.3 eq.), HFIP, reflux | 1 h | 33%/35% |
| 21 | TFA (0.3 eq.), HFIP, reflux | 1 h | 19%/40% |
| 22 | FeCl3 (0.3 eq.), HFIP, reflux | 1 h | Trace/trace |
| 23 | H2SO4 (0.3 eq.), HFIP, reflux | 1 h | 10%/25% |
| 24 | CF3SO3H (0.3 eq.), HFIP, reflux | 1 h | Trace/22% |
| 25 | p-TsOH (0.1 eq.), HFIP, reflux | 1 h | 12%/44% |
| 26 | p-TsOH (0.5 eq.), HFIP, reflux | 1 h | 17%/32% |
By analyzing the competition of the formation of 2 and DGP, we hypothesized that the stability of the carbocation intermediate B and C may be the key factor in enhancing the DGP formation pathway. In addition, strongly hydrogen bond-donating, polar solvents trifluoroethanol (TFE) and hexafluoroisopropanol (HFIP) have found numerous uses in organic synthesis due to its ability to stabilize ionic species, transfer protons, and engage in a range of other intermolecular interactions.26,27 For example, in the metal-free Markovnikov-type alkyne hydration reported by Li et al., the carbocation was strongly stabilized in TFE, which facilitated the progression of the reaction.26 Therefore, we investigated the reaction conditions using TFE and HFIP as solvent to stabilize carbocation intermediate B and C, promoted by different acid catalysts (entries 19–26). Pleasingly, when HFIP was used as the solvent and p-TsOH (0.3 eq.) as the acid catalyst, the reaction produced DGP with a 33% yield (entry 20). Although the yield was not high, it sufficed for the rapid preparation of DGP and the study of its taste identification.
To enhance the availability of high-purity materials for the taste evaluation, we have innovatively developed a vacuum sublimation purification technique for DGP, drawing upon previously reported thermogravimetric analysis.5 A sample of DGP, isolated through column separation, was subjected to sublimation using a glass sublimation apparatus under a controlled pressure of 0.15 mm Hg at 120 °C, yielding purified DGP.
| DGP | 1 | 2 | |
|---|---|---|---|
| a Taste quality and detection thresholds of taste compounds were determined in a triangle test.b Compare DGP aqueous solutions to a 5% sucrose aqueous solution to determine the concentration at which the sample's sweetness intensity matches that of the sucrose solution and divide this concentration by the sucrose solution concentration (5%). | |||
| Taste qualitya | Sweet taste | Tasteless | Tasteless |
| Relative sweetnessb (relative to sucrose/times) | 1.0 | — | — |
| Sweet taste thresholda (% or mmol L−1 in water) | 0.2% or 14 mmol L−1 | — | — |
:
MeOH = 50
:
1 to 20
:
1, v/v) afforded the product 3. (65.5 g, 73%). 1H NMR (400 MHz, CD3OD) δ 7.77 (d, J = 8.0 Hz, 2H), 7.42 (d, J = 8.0 Hz, 2H), 4.53 (d, J = 8.0 Hz, 1H), 4.28 (dd, J = 12.0, 4.0 Hz, 1H), 4.13 (dd, J = 12.0, 8.0 Hz, 1H), 3.79–3.66 (m, 1H), 3.66–3.58 (m, 1H), 3.35–3.21 (m, 3H), 3.13–3.18 (m, 1H), 2.43 (s, 3H). 13C NMR (100 MHz, CD3OD) δ 146.5, 134.5, 131.0, 129.1, 101.2, 74.9, 73.2, 71.3, 71.1, 71.0, 55.7, 21.6.
:
MeOH = 50
:
1 to 20
:
1, v/v) afforded the product 1. (27.2 g, 82%). 1H NMR (400 MHz, CD3OD) δ 4.93 (d, J = 3.0 Hz, 1H), 4.23 (t, J = 2.4 Hz, 1H), 4.17–4.12 (m, 2H), 4.12–4.08 (m, 1H), 3.93 (dd, J = 12.0, 3.2 Hz, 1H), 3.78 (s, 1H), 3.56 (s, 3H). 13C NMR (100 MHz, CD3OD) δ 99.78, 76.9, 73.3, 72.7, 71.7, 69.9, 57.5. HRMS (ESI-TOF): calcd. for C7H12NaO5 [M + Na]+ 199.0582; found 199.0580.
:
1 mixture of petroleum ether and ethyl acetate (20.0 mL) to slurry wash, then filter to obtain the product 2 (850 mg, 85%). 1H NMR (400 MHz, CDCl3) δ 5.08 (d, J = 4.4 Hz, 1H), 4.63 (t, J = 5.4 Hz, 1H), 4.45 (dd, J = 5.3, 2.8 Hz, 1H), 4.22 (dt, J = 6.9, 5.7 Hz, 1H), 4.17 (dd, J = 4.4, 2.8 Hz, 1H), 3.94 (dd, J = 9.3, 5.9 Hz, 1H), 3.61–3.54 (m, 1H), 3.52 (d, J = 3.5 Hz, 3H), 2.69–2.39 (m, 2H). 13C NMR (100 MHz, CDCl3) δ 104.9, 87.8, 79.4, 77.9, 72.0, 71.3, 56.0. HRMS (ESI-TOF): calcd. for C7H12NaO5 [M + Na]+ 199.0582; found 199.0584.
:
EA = 5
:
1 to 2
:
1, v/v) afforded the product DGP. (279 mg, 33%). 1H NMR (400 MHz, CDCl3) δ 5.43 (s, 1H), 5.17 (t, J = 4.0 Hz, 1H), 4.22 (s, 1H), 4.12 (d, J = 4.0 Hz, 1H), 4.03 (d, J = 12.0 Hz, 1H), 3.92 (d, J = 12.0 Hz, 1H), 3.76 (s, 1H), 1.86 (s, 1H). 13C NMR (100 MHz, CDCl3) δ 101.2, 82.9, 79.5, 79.1, 75.9, 71.8. HRMS (ESI-TOF): calcd. for C6H8NaO4 [M + Na]+ 167.0320; found 167.0321. A 1.0 g sample of DGP, isolated through column separation, was subjected to sublimation using a glass sublimation apparatus under a controlled pressure of 0.15 mm Hg and a temperature of 120 °C, yielding 850 mg purified DGP.
:
1 dilutions). Relative sweetness of taste compounds was determined as follows. Prepare a series of aqueous solutions with varying concentrations of the sample. Compare these to a 5% sucrose aqueous solution to determine the concentration at which the sample's sweetness intensity matches that of the sucrose solution. Then, divide this concentration by the sucrose solution concentration (5%) to obtain the relative sweetness of the sample.
Footnote |
| † Electronic supplementary information (ESI) available. CCDC 2416300. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d5ra00266d |
| This journal is © The Royal Society of Chemistry 2025 |