Conversion of Sugars to Lactic Acid using Homogeneous Niobium-Substituted Polyoxometalate Catalysts - Supporting Information

Chemistry—A European Journal Supporting Information Conversion of Sugars to Lactic Acid using Homogeneous Niobium-Substituted Polyoxometalate Catalysts Anne Wesner, Jan-Christian Raabe, Maximilian J. Poller, Sebastian Meier, Anders Riisager, and Jakob Albert* Wiley VCH Dienstag, 03.12.2024 2469 / 378523 [S. 263/263] 1 Supporting Information Conversion of Sugars to Lactic Acid using homogeneous Niobiumsubstituted Polyoxometalate Catalysts Anne Wesnera, Jan-Christian Raabea, Maximilian J. Pollera, Sebastian Meierb, Anders Riisagerb and Jakob Alberta* a Institute of Technical and Macromolecular Chemistry, University of Hamburg, Bundesstraße 45, 20146 Hamburg, Germany b Department of Chemistry, Technical University of Denmark, Kemitorvet, 2800 Kgs. Lyngby, Denmark * Contact details of the corresponding author: Phone: +49-(0)40 42838 4209; e-mail: [email protected] Table of contents Reagents and Materials 2 Catalyst preparation and characterization 3 Analytical Methods 11 Supplementary catalytic results 15 Screening HPSs for conversion of dihydroxyacetone (DHA) into LA Fehler! Textmarke nicht definiert. Substrate screening of different sugars for conversion into LA Fehler! Textmarke nicht definiert. Post-reaction catalyst analysis 19 References 20 Reagents and Materials For the catalyst synthesis the chemicals were obtained from the following suppliers: sodium molybdate dihydrate (99.5 %, Carl Roth), disodium hydrogen phosphate (95 %, Merck), sodium vanadium oxide (96 %, Alfa Aesar), niobium(V) oxide (99.5 %, Alfa Aesar), potassium hydroxide (85 %, Chemsolute), hydrochloric acid (37 % solution in water, VWR chemicals), hydrogen peroxide (30 % solution in water, VWR chemicals). For catalyst and substrate screening, the following substrates were used: dihydroxyacetone (85 %, Merck, 820482), glucose (≥ 99.5 %, Sigma-Aldrich, G8270), fructose (≥ 99 %, Sigma-Aldrich, X3877), sucrose (≥ 99.5 %, Sigma-Aldrich, S9378), xylose (≥99 %, Thermo Scientific, X3877), mannose (≥ 99 %, Fluka, 63582), cellobiose (≥99 %, Fluka, C7252). HPLC determination of retention time and calibration was performed using the following chemicals: glycolic acid (50 %, Merck, 814084), glyceraldehyde (≥ 90 %, Sigma-Aldrich, G5001), pyruvaldehyde (40 %, EGA-CHEMIE, 17,733-4), lactic acid (85 %, Sigma-Aldrich, W261106), dihydroxyacetone (85 %, Merck, 820482), formic acid (≥ 99 %, AnalaR NORMAPUR, 20318.30), acetic acid (97 %, ThermoFischer, A/0400/PB15), acetaldehyde (40 % solution in water, Merck, W200379). Catalyst screening was carried out using the following compounds: Na3[PMo12O40] (NaMo), Na6[PMo9V3O40] (NaV3), Na6[PV2NbMo9O40] (NaV2Nb), Na6[PVNb2Mo9O40] (NaVNb2), Na4[PNbMo11O40] (NaNb), Na5[PNb2Mo10O40] (NaNb2), Na6[PNb3Mo9O40] (NaNb3). All chemicals were obtained and utilized without further purification. Catalyst preparation and characterization The original synthesis procedures are published in a previous study.[1] Synthesis of potassium hexaniobate K8Nb6O19 as a precursor compound In a typical procedure, diniobium pentoxide (22.63 g, 85.13 mmol, 1 equivalent) and potassium hydroxide (71.04 g, 1.27 mol, 14.87 equivalents) were mixed and added in small portions to a nickel crucible heated over a gas burner flame. The mixture was bubbling and the diniobium pentoxide was dissolved in the potassium hydroxide melt. After complete addition of both precursors, the reaction melt was heated until bubbling has stopped. The melt was dissolved in water, filtered and reduced to one-eighth of its volume under reduced pressure, resulting in a precipitation of a colorless solid. It was cooled to 4 °C overnight and the precipitate was collected by filtration and washed several times with cooled ethanol. After washing the solid was dried at 60 °C in an oven. A colorless solid (25.25 g) was obtained. The NbV content of the potassium hexaniobate was first determined using ICP-OES and the weights were calculated for the respective reaction procedures based on the Nb content, considering the stoichiometric ratios. Synthesis and Characterization of PNbxMo12-xO40(3+x)-: (Na4[PNbMo11O40] (NaNb), Na5[PNb2Mo10O40] (NaNb2), Na6[PNb3Mo9O40] (NaNb3)) Synthesis of PNbxMo12-xO40(3+x)-: (Na4[PNbMo11O40] (NaNb), Na5[PNb2Mo10O40] (NaNb2), Na6[PNb3Mo9O40] (NaNb3)) The stoichiometry of P/Nb/Mo was ensured by weighing the precursors in the correct molar ratios. Sodium molybdate dihydrate (10.01 g for Na4[PNbMo11O40], 10.00 g for Na5[PNb2Mo10O40] and 10.01 g for Na6[PNb3Mo9O40]) and disodium hydrogen phosphate (0.65 g for Na4[PNbMo11O40], 0.65 g for Na5[PNb2Mo10O40] and 0.66 g for Na6[PNb3Mo9O40]) (stoichiometry 9:1) were dissolved in water (50 mL) and the pH was adjusted to ~1, forming a yellow, clear solution. To this solution, a solution of potassium hexaniobate (1.12 g for Na4[PNbMo11O40], 2.16 g for Na5[PNb2Mo10O40] and 3.36 g for Na6[PNb3Mo9O40]) in a diluted aqueous hydrogen peroxide solution (1.5 %) (20 mL) was added and refluxed to form an orange or yellow reaction solution. For substitution degrees x = 1 and 2, sodium molybdate dihydrate (2.22 g for Na4[PNbMo11O40] and 1.11 g for Na5[PNb2Mo10O40]) was added in the required stoichiometry to fill the remaining vacancies of the lacunary species. The pH value was 5.142 (in the experiment for Na4[PNbMo11O40]), 4.789 (in the experiment for Na5[PNb2Mo10O40]) and 5.450 (in the experiment for Na6[PNb3Mo9O40]) A clear, yellow reaction solution was then formed. The pH was adjusted to ~1.6 by adding a 37 % hydrochloric acid solution in water. The reaction solution was filtered and desalted using an established nanofiltration method. The solution of potassium hexaniobate must be added directly to the reaction solution without delay, otherwise niobium(V) oxide may precipitate. The solution should therefore be freshly prepared and used immediately. After the above described solution has been added to the lacunary solution, the reaction mixture must be heated directly to prevent precipitation of niobium(V) oxide. It is recommended to use a preheated oil bath. Sometimes direct precipitate formation occurs after the addition of the potassium hexaniobate solution. This is not always a problem if the heating is fast enough, and a clear reaction solution form Characterization of Na4[PNbMo11O40] (NaNb) 31P-NMR: (242.9 MHz, H2O/D2O, 20 °C):δ (ppm) = -1.28, -1.37, -1.43, -2.99, -3.29, -3 .30, -3.36, -3.38, -3.66 IR (ATR): ṽ (cm-1) = 3396 (w, O-H, H2O), 1615 (O-H, hydration H2O), 1058, 1034 (w, P-O), 944 (me, M=Ot), 859 (me, (M-O-M)vertex), 741 (st, (M-O-M)edge). ICP-OES: Calculated for Na4[PNbMo11O40] ∙ 6 H2O: 52.264 % Mo, 4.601 % Nb, 1.534 % P, 0.00 % K, 4.554 % Na. Found for Na4PNbMo11O40 ∙ 6 H2O: 44.77 % Mo, 4.33 % Nb, 1.17 % P 0.75 % K, 4.84 % Na. Data normalized to molybdenum. Na/K/P/Nb/Mo ratio: 4.96/0.45/0.893/1.10/11. TGA: 4.933 % weight loss upon drying, this corresponds to 6 mol lattice water per mol of the POM. Characterization of Na5[PNb2Mo10O40] (NaNb2) 31P-NMR: (242.9 MHz, H2O/D2O, 20 °C): δ (ppm) = -0.28, -0.36, -1.26, -1.33, -1.40, -2.45, -2.51, -2.63, -2.92, -2.96, -3.04, -3.0 9, -3.19, -3.27, -3.28, -3.33, -3.37, -3.46, -2.64. IR (ATR): ṽ (cm-1) = 3412 (w, O-H, H2O), 1613 (O-H, hydration H2O), 1049 (w, P-O), 947 (me, M=Ot), 856 (me, (M-O-M)vertex), 757 (st, (M-O-M)edge). ICP-OES: Calculated for Na5[PNb2Mo10O40] ∙ 5 H2O: 47.467 % Mo, 9.193 % Nb, 1.532 % P, 0.00 % K, 5.687 % Na. Found for Na5PNb2Mo10O40 ∙ 5 H2O: 47.74 % Mo, 8.456 % Nb, 1.50 % P 1.36 % K, 4.565 % Na. Data normalized to molybdenum. Na/K/P/Nb/Mo ratio: 3.99/0.70/0.98/1.83/10. TGA: 4.652 % weight loss upon drying, this corresponds to 5 mol lattice water per mol of the POM. Characterization Na6[PNb3Mo9O40] (NaNb3) 31P-NMR: (242.9 MHz, H2O/D2O, 20 °C): δ [ppm] = -2.36, -2.48, -2.52, -2.61, -2.73, -2.78, -2.91, -2.96, -3.04, -3.09, -3.18, -3.26 , -3.28, -3.33, -3.37, -3.63. IR (ATR): ṽ (cm-1) = 3388 (w, O-H, H2O), 1613 (O-H, hydration H2O), 1047 (w, P-O), 944 (me, M=Ot), 857 (me, (M-O-M)vertex), 760 (st, (M-O-M)edge). ICP-OES: Calculated for Na6[PNb3Mo9O40] ∙ 7 H2O: 41.569 % Mo, 13.418 % Nb, 1.491 % P, 0.00 % K, 6.641 % Na. Found for Na6PNb3Mo9O40 ∙ 7 H2O: 38.54 % Mo, 12.03 % Nb, 1.26 % P 2.37 % K, 4.30 % Na. Data normalized to molybdenum. Na/K/P/Nb/Mo ratio: 4.19/1.36/0.91/2.90/9. TGA: 6.297 % weight loss upon drying, this corresponds to 7 mol lattice water per mol of the POM. Synthesis and Characterization of NaPVxNbyMo9O40(3+x): (Na6[PV2NbMo9O40] (NaV2Nb), Na6[PVNb2Mo9O40] (NaVNb2)) Synthesis of Na6[PV2NbMo9O40] (NaV2Nb), Na6[PVNb2Mo9O40] (NaVNb2) The stoichiometry of P/V/Mo was ensured by weighing the precursors in the correct molar ratios. Sodium molybdate dihydrate (15.00 g for Na6[PV2NbMo9O40] and 10.00 g for Na6[PVNb2Mo9O40]) and disodium hydrogen phosphate (0.98 g for Na6[PV2NbMo9O40] and 0.66 g for Na6[PVNb2Mo9O40]) (stoichiometry 9 : 1) were dissolved in water and the pH was adjusted to ~1, forming a yellow, clear solution. To this solution, a solution of sodium vanadate in water (1.68 g for Na6[PV2NbMo9O40] and 0.56 g Na6[PVNb2Mo9O40] ) was added and refluxed to form a red reaction solution, directly after the vanadate addition. A pH value of 5.351 (in the experiment for Na6[PV2NbMo9O40]) and 5.235 (in the experiment for Na6[PVNb2Mo9O40]) was measured. The pH was adjusted to ~1.6 by adding a 37 % hydrochloric acid solution in water. In the last step, the reaction solution was filtered and desalted using a nanofiltration approach. Sodium vanadium oxide dissolves in water after the aqueous suspension has been heated briefly. Characterization of Na6[PV2NbMo9O40] (NaV2Nb) 31P-NMR: (242.9 MHz, H2O/D2O, 20 °C): δ (ppm) = -2.50 to -4.30. 51V-NMR (157.8 MHz, H2O/D2O, 20 °C): (ppm) = -517.7, -525.9, -531.3, -532.8, -533 .2, -533.8, -534.3, -535.7, -538.5, -540 to -546. IR (ATR): ṽ [cm-1] = 3374 (w, O-H, H2O), 1610 (O-H, hydration H2O), 1046 (w, P-O), 941 (me, M=Ot), 847 (me, (M-O-M)vertex), 750 (st, (M-O-M)edge). ICP-OES: Calculated for Na6[PV2NbMo9O40] ∙ 7 H2O: 43.319 % Mo, 4.661 % Nb, 5.111 % V, 1.554 % P, 0.00 % K, 6.92 % Na. Found for Na6PV2NbMo9O40 ∙ 7 H2O: 38.50 % Mo, 4.38 % Nb, 5.225 % V 1.52 % P 1.38 % K, 6.49 % Na. Data normalized to molybdenum. Na/K/P/V/Nb/Mo ratio: 6.33/0.792/1.10/2.3/1.06/9. TGA: 5.901 % weight loss upon drying, this corresponds to 7 mol lattice water per mol of the POM. Characterization of Na6[PVNb2Mo9O40] (NaVNb2) 31P-NMR: (242.9 MHz, H2O/D2O, 20 °C): δ (ppm) = -4.92, -5.08, -5.27 to -5.36, -5.49, -5.62, -5.69, -5.93, -5.96, -5.99, -6.04, - 6.20, -6.30, -6.42, -6.43, -6.49, -6.54, -6.73, -6.77, -6.86, -6.88, -6.89, -6.90. 51V-NMR (157.8 MHz, H2O/D2O, 20 °C): δ (ppm) = -519.8, -527.9, -532.7, -533.7, -534.4, -535.4, -537.1, -545.1. IR (ATR): ṽ (cm-1) = 3391 (w, O-H, H2O), 1615 (O-H, hydration H2O), 1048 (w, P-O), 944 (me, M=Ot), 859 (me, (M-O-M)vertex), 762 (st, (M-O-M)edge). ICP-OES: Calculated for Na6[PVNb2Mo9O40] ∙ 5 H2O: 43.191 % Mo, 9.294 % Nb, 2.548 % V, 1.549 % P, 0.00 % K, 6.90 % Na. Found for Na6PVNb2Mo9O40 ∙ 5 H2O: 42.28 % Mo, 8.915 % Nb, 2.19 % V 1.46 % P 1.545 % K, 4.645 % Na. Data normalized to molybdenum. Na/K/P/V/Nb/Mo ratio: 4.13/0.81/0.964/0.880/1.96/9. TGA: 4.077 % weight loss upon drying, this corresponds to 5 mol lattice water per mol of the POM. Supplementary catalytic results Screening HPSs for conversion of dihydroxyacetone (DHA) into lactic acid (LA) Figure S1: HPLC chromatogram displaying retention times of products from the catalytic conversion of DHA into LA with NaNb2. Reaction conditions: 42 mg catalyst, 100 mg substrate, 4 g H2O, 160 °C, 20 bar N2, 1000 rpm, 1 h. Table S1: Conversion and product yields for the conversion of DHA into LA.a Catalyst X (%) Yields (%) YGA YGly YGlycA YPA YLA YFA YAA YAcA Control (blank) 23.7 0 0 0. 20.7 2.6 0 0 0 NaMo 75.2 6.4 2.3 3.8 4.0 8.3 1.8 7.9 10.6 NaV3 85.0 4.4 1.7 3.1 2.9 8.3 1.7 7.4 19.8 NaV2Nb 92.5 3.4 1.8 6.2 2.9 10.2 2.0 11.7 18.0 NaVNb2 96.0 3.4 1.6 7.3 1.8 12.1 1.5 11.0 17.7 NaNb 97.5 2.4 1.3 10.1 1.7 15.4 1.7 11.0 12.7 NaNb2 97.4 2.7 1.8 8.0 1.6 20.9 1.5 10.3 10.6 NaNb3 97.2 3.3 0.6 6.6 1.7 20.7 1.6 11.6 11.2 a Reaction conditions: 42 mg catalyst, 100 mg substrate, 4 g H2O, 160 °C, 20 bar N2, 1000 rpm, 1 h. GA: glycolic acid, Gly: glyoxal, GlycA: glyceraldehyde, PA: pyruvaldehyde, LA: lactic acid, FA: formic acid, AA: acetic acid, AcA: acetaldehyde. Table S2: Carbon balances and product selectivities for the conversion of DHA into LA.a Catalyst Cb (%) Selectivities (%) SGA SGly SGlycA SPA SLA SFA SAA SAcA Control (blank) 100 0.0 0.0 0.0 87.3 11.0 0.0 0.0 0.0 NaMo 70 8.6 3.0 5.1 5.4 11.0 2.3 10.5 14.1 NaV3 65 5.2 2.0 3.6 3.4 9.8 2.0 8.7 23.3 NaV2 64 3.7 1.9 6.7 3.1 11.1 2.2 12.7 19.4 NaVNb2 61 3.5 1.6 7.6 1.9 12.6 1.5 11.5 18.5 NaNb 59 2.5 1.4 10.4 1.8 15.9 1.7 11.3 13.0 NaNb2 62 2.8 1.8 8.2 1.6 21.5 1.5 10.6 10.8 NaNb3 60 3.4 0.7 6.8 1.8 21.3 1.6 12.0 11.5 a Reaction conditions: 42 mg catalyst, 100 mg substrate, 4 g H2O, 160 °C, 20 bar N2, 1000 rpm, 1 h. GA: glycolic acid, Gly: glyoxal, GlycA: glyceraldehyde, PA: pyruvaldehyde, LA: lactic acid, FA: formic acid, AA: acetic acid, AcA: acetaldehyde. Substrate scope for production of LA using NaNb2 as a catalyst Figure S2: HPLC chromatogram displaying retention times of products from the catalytic conversion of glucose into LA with NaNb2. Reaction conditions: 42 mg catalyst, 100 mg substrate, 4 g H2O, 160 °C, 20 bar N2, 1000 rpm, 1 h. Table S3: Conversion and product yields for the conversion of different sugars into LA without catalyst (first entries) and with NaNb2 (second entries).a Substrate X (%) Yields (%) YDHA YGlycA YPA YLA YFA YAA YAcA YGlu YMan Glucose 0 0 0 0 0 0 0.0 0.0 - 0.0 Glucose 76.9 3.6 4.3 4.4 9.8 5.3 4.1 5.1 - 13.6 Fructose 0 0 0 0 0 0.0 0.0 0.0 0.0 0.0 Fructose 97.4 2.3 5.3 2.1 10.0 5.0 6.6 5.6 4.2 3.4 Sucrose 0 0 0 0 0 0.0 0 0.0 0.0 0.0 Sucrose 100 2.1 10.7 3.4 11.3 5.6 6.2 6.0 16.1 14.8 Xylose 0.0 0 0 0 0 0.0 0.0 0.0 0.0 0.0 Xylose 91.5 3.1 8.7 3.1 12.1 6.3 5.6 6.4 0.0 0.0 Mannose 0 0 0 0 0 0.0 0.0 0.0 0.0 - Mannose 77.8 3.9 4.2 3.9 9.2 5.0 4.7 3.6 25.8 - Cellobiose 2.9 0 0 0.0 0 0.0 0.0 0.0 0.0 0.0 Cellobiose 36.2 0 0.53 3.5 2.4 0.0 0.0 0.0 2.6 2.2 a Reaction conditions: 42 mg catalyst, 100 mg substrate, 4 g H2O, 160 °C, 20 bar N2, 1000 rpm, 1 h. DHA: dihydroxyacetone, GlycA: glyceraldehyde, PA: pyruvaldehyde, LA: lactic acid, FA: formic acid, AA: acetic acid, AcA: acetaldehyde, Glu: glucose, Man: mannose. Table S4: Carbon balances and product selectivities for the conversion of different sugars into LA without catalyst (first entries) and with NaNb2 (second entries).a Substrate Cb (%) Selectivities (%) SGlycA SPA SLA SFA SAA SAcA SGlu SMan Glucose 100 0.0 0.0 0.0 0.0 0.0 0.0 - 0.0 Glucose 83.4 5.6 5.8 12.7 6.9 5.4 6.7 - 17.7 Fructose 95.0 0.0 0.0 0.0 0.0 0.0 0.0 4.3 0.0 Fructose 44.1 5.5 2.2 10.3 5.1 6.8 5.7 1.9 3.5 Sucrose 100 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Sucrose 73.2 10.7 4.0 11.3 5.6 6.2 5.9 16.1 14.8 Xylose 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Xylose 49.3 9.5 3.4 13.2 6.9 6.1 7.0 0.0 0.0 Mannose 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 - Mannose 52.8 5.4 5.1 11.8 6.4 6.1 4.7 33.2 - Cellobiose 97.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Cellobiose 74.6 1.5 9.6 6.6 6.6 0.0 0.0 7.2 6.1 a Reaction conditions: 42 mg catalyst, 100 mg substrate, 4 g H2O, 160 °C, 20 bar N2, 1000 rpm, 1 h. GlycA: glyceraldehyde, PA: pyruvaldehyde, LA: lactic acid, FA: formic acid, AA: acetic acid, AcA: acetaldehyde, Glu: glucose, Man: mannose. Table S5: IR spectra band assignments.[2] Wave number (cm-1) Assignment 750 + 795 C-H Out of plane vibration substituted furan ring 965 C-H vibration furan ring 1020 C=C stretch vibrations 1090 C-O-C ether vibration 1160 + 1200 C-O-C deformation vibration furan ring 1295 C-H rocking vibration 1360 C-C framework vibration (furan) C6 sugars 1395 C-C framework vibration (furan) C5 sugars 1460 C-H aliphatic chain vibration 1510 C=C vibration aromatic couple bonds of polysubstituted furans 1600 C=C stretch vibration 1670 C=O carbonyl, aldehyde vibrations 1700 C=O stretch of acids, aldehydes and ketones Figure S3: IR spectra of solid residues after the substrate screening experiments. Table S6: Yields of selected by-products determined via NMR for the conversion of different sugars with NaNb2.a Substrate Yields (%) YAll YAlt YXyl YAra YLyx YRib Yother disacharides Glucose 3.1 2.2 - - - - - Xylose - - - 7.0 5.5 2.4 - Cellobiose 0.4 0.3 - - - - 12.8 a Reaction conditions: 42 mg catalyst, 100 mg substrate, 4 g H2O, 160 °C, 20 bar N2, 1000 rpm, 1 h. All: allose, Alt: Altrose, Xyl: xylose, Ara: arabinose, Lux: lyxose, Rib: ribose. Figure S4: Structural formula of glucose and its isomers. Figure S5: Structural formula of xylose and its isomers. Post-reaction catalyst analysis Table S7: pH-values for the conversion of different sugars into LA with NaNb2 before (pHinitial) and after reaction (pHfinal).a Substrate pHinitial pHfinal Glucose 3.4 1.9 Fructose 3.5 2.1 Sucrose 3.6 2.4 Xylose 3.4 2.4 Mannose 3.4 1.9 Cellobiose 3.4 2.8 aReaction conditions: 42 mg catalyst, 100 mg substrate, 4 g H2O, 160 °C, 20 bar N2, 1000 rpm, 1 h. Figure S6: 31P NMR spectra of NaPNb2 in aqueous solution (bottom), NaPNb2 in aqueous solution acidified to the pH value of the reaction solutions and of NaPNb2 in the reaction solution, oxidized with elemental bromine. References [1] J.-C. Raabe, T. Esser, F. Jameel, M. Stein, J. Albert, M. J. Poller, Inorg. Chem. Front. 2023, 10, 4854. [2] A. Wassenberg, T. Esser, M. J. Poller, J. Albert, Materials 2023, 16, e202300072.