Harnessing a Continuous‐Flow Persulfuric Acid Generator for Direct Oxidative Aldehyde Esterifications

Abstract Persulfuric acid is a well‐known oxidant in various industrial‐scale purification procedures. However, due to its tendency toward explosive decomposition, its usefulness in organic synthesis remained largely underexplored. Herein, a continuous in situ persulfuric acid generator was developed and applied for oxidative esterification of aldehydes under flow conditions. Sulfuric acid served as a readily available and benign precursor to form persulfuric acid in situ. By taking advantage of the continuous‐flow generator concept, safety hazards were significantly reduced, whilst a robust and effective approach was ensured for direct transformations of aldehydes to valuable esters. The process proved useful for the transformation of diverse aliphatic as well as aromatic aldehydes, while its preparative capability was verified by the multigram‐scale synthesis of a pharmaceutically relevant key intermediate. The present flow protocol demonstrates the safe, sustainable, and scalable application of persulfuric acid in a manner that would not be amenable to conventional batch processing.


General information
All solvents and chemicals were obtained from typical commercial vendors and were used as received, without any further purification.
Column chromatographic purification was performed by using a Biotage Isolera automated flash chromatography system with cartridges packed with KP-SIL, 60 Å (32-63 μm particle size). Analytical thin-layer chromatography (TLC) was carried out using Merck silica gel 60 GF254 plates. Compounds were visualized by means of UV or by using KMnO4. 1 H-and 13 C-NMR spectra were recorded on a Bruker Avance III 300 MHz instrument at room temperature, in CDCl3 as solvent, at 300 MHz and 75 MHz, respectively. Chemical shifts (δ) are reported in ppm using TMS as internal standard. Coupling constants are given in Hz units.
Analytical HPLC measurements were carried out on a C18 reversed-phase column (150 × 4.6 mm, particle size 5 mm) at 37 °C using mobile phases A [H2O/CH3CN 90:10 (v/v) + 0.1% TFA] and B (CH3CN + 0.1% TFA) at a flow rate of 1.5 mL min -1 . The gradient applied was as follows: linear increase from 30% solution B to 100% B in 8 min, hold at 100% solution B for 2 min.
GC-FID analysis was performed on a Shimadzu GCFID 2030 instrument equipped with a flame ionization detector, using an RTX-5MS column (30 m × 0.25 mm ID × 0.25 μm) and helium as carrier gas (40 cm s -1 linear velocity). The injector temperature was set to 280 °C. After 1 min at 50 °C, the temperature was increased by 25 °C min −1 to reach 300 °C and then kept constant at 300 °C for 3 min. The detector gases used for flame ionization were hydrogen and synthetic air (5.0 purity).
Optical rotation was measured in CHCl3 (HPLC-grade) at 25 °C against the sodium D-line (λ= 589 nm) on a Perkin Elmer Polarimeter 341 using a 10-cm pathlength cell.
Equipment for the continuous flow reactions was assembled using commercially available components. Liquid streams were pumped by using Syrris ® Asia syringe pumps. Flow systems were pressurized by using an adjustable backpressure regulator (BPR) from Zaiput. Reaction coils were heated by means of a conventional oil bath. Reagent feeds were either streamed directly or by using injection valves and sample loops. Sample loops and reactor coils were made by using perfluoroalkoxy alkane (PFA) tubings (1/16" OD, 0.80 mm ID).
A micro reaction calorimeter (μRC) from Thermal Hazard Technology was used to study the thermal behavior of the H2SO4-H2O2 reaction system. Quantitative green metrics were calculated according to the literature. S1 CAUTION: Sulfuric acid is highly corrosive causing rapid tissue destruction and serious chemical burns. Persulfuric acid is one of the strongest oxidants known. It is unstable and potentially explosive, especially in mixtures with organic substances. Extreme care must therefore be taken when handling these substances! All equipment must be set up in a well-ventilated fume hood and personal protective equipment must be worn during experimentation. A thorough safety assessment should be made before conducting any experiments.

Titration tests at different temperatures
Solution preparation: 1.0 M H2O2: 0.486 g 35 wt% aq. H2O2 was diluted to 5 mL with MeOH or iPrOH. 1.0 M H2SO4: 0.516 g 95% H2SO4 was diluted to 5 mL with MeOH or iPrOH. NOTE: these concentrations were different from the ones used for the flow reactions.

Procedure:
The titration mode was selected and the official procedures for the calorimeter were followed. MeOH was used as the solvent for titration at 25 °C, 50 °C and 70 °C while iPrOH was used for titration at 80 °C. 300 μL H2SO4 solution was added to the sample vial and reference vial, respectively. A stirring bar was placed into the sample vial and the stirring speed was set to 200 rpm. H2O2 solution was loaded into a 100 μL syringe and used as titrant. After the baseline stabilized, 10 μL H2O2 was injected, and the power was recorded. The injection was repeated 9 times and the interval time between two injections was specified as 300 s. MeOH/iPrOH was also used as titrant to demonstrate the mixing process.

Results:
For the titrations at 25 °C, 50 °C and 70 °C, the detected heat was the same for the two processes, which meant that the reaction between H2O2 and H2SO4 did not happen and only the mixing heat was recorded. For the titration at 80 °C, the heat was much larger when H2O2 was used as titrant. This indicated the formation of persulfuric acid. Figure S1. Detected heat of titrations at different temperatures.

Procedure:
The scan mode was selected. 50 μL H2O2 solution and 50 μL H2SO4 solution were premixed in the sample vial at room temperature. To the reference vial, 100 μL iPrOH was added. A stirring bar was placed into both vials and the stirring speed was set to 200 rpm. Then the two vials were placed into the calorimeter. After the baseline stabilized, the sample was scanned from 25 °C to 150 °C at 1 °C min −1 and held at 150 °C for 30 min.

Results:
Only one exothermic peak, which started from around 75 °C, was observed during the scan test. As the formation of persulfuric acid cannot happen at room temperature, this peak undoubtedly included the formation heat. 8.50 M H2O2: 4.130 g 35 wt% aq. H2O2 was diluted to 5 mL with iPrOH. 1.89 M H2SO4: 0.976 g 95% H2SO4 was diluted to 5 mL with iPrOH.

Procedure:
The titration mode was selected, and the temperature was set to 80 °C. For the reaction, either 180, 270 or 360 μL H2SO4 solution was added to the sample vial and reference vial, respectively. A stirring bar was placed into the sample vials and the stirring speed was set to 200 rpm. H2O2 solution was loaded into a 100 μL syringe and used as titrant. After the baseline stabilized, either 20, 30, or 40 μL H2O2 was injected at once and the power was recorded. For each condition, the experiment was repeated twice in order to check the consistency. As H2SO4 was in excess, the calculation of reaction enthalpy was based on H2O2. For the mixing test, H2O was used as the titrant. For the blank test, iPrOH was used as a replacement for H2SO4. NOTE: The long-term exposure to high temperature does slowly cause damage to the syringe.

Iodometric peroxide determination
Total peroxide concentration was measured by iodometric titration. This method was based on two major steps, as follows.

Procedure:
H2O2 (1.70 M) and H2SO4 (3.40 M) solutions in the batch calorimetry scan test were used to prepare the reaction mixture. The scan mode of the calorimeter was adopted to provide the temperature for the formation of persulfuric acid. 100 µL of reaction solution was diluted with 1 mL of H2O. 200 µL of diluted H2SO4 solution, then 200 µL of the KI solution were added. One drop of the (NH4)6Mo7O24 solution was added. Titration with thiosulfate (0.1 M) was performed until a pale-yellow color was reached. 20 µL of starch indicator solution was added. Further titration with 0.1 M thiosulfate was performed until a colorless solution was observed. Results:

Oxidative esterifications with H 2 O 2 /HCl system
Substrate solution containing 1 equiv. of hydrocinnamaldehyde and 1.1-3.0 equiv. of HCl (added as 6 M aq. solution) in MeOH and a solution of H2O2 (prepared from 35 wt% aq. solution) in MeOH were pumped as separate streams (P1 and P2) by using Syrris ® Asia syringe pumps equipped with two injection valves and two sample loops (5 or 8 mL, each). With MeOH serving as carrier solvent, the feeds were combined in a Y-mixer. The resulting stream was directed through a 15-mL reaction coil (15-60 min residence time) which was heated at 50, 100 or 120 °C and pressurized at 5 bar. The reactor outlet was directed into a flask containing a stirred mixture of saturated aq. NaHCO3 and some MnO2 in order to quench any excess acid and/or oxidant. After reaching steady state, approx. 20 µL aliquots of crude material were collected. The samples were diluted with 1 mL of CH3CN/H2O 9:1 and was analyzed by analytical HPLC directly after the flow experiments. In order to assign HPLC signals of side product 3 and 4, reference samples were prepared according to a literature procedure. S2 Hydrocinnamic acid, 2 is commercially available. Side product si was isolated chromatographically and characterized by 1 H NMR. NMR data of the compound matched the reported literature. S3

Oxidative esterifications in the presence of various solid acids and oxidants
Substrate solution containing 0.5 M of hydrocinnamaldehyde in MeOH and a solution of UHP or TBHP or H2O2 (TBHP was added as 5.5 M decane solution, H2O2 was added as 35 wt% aq. solution) in MeOH were pumped as separate streams (P1 and P2) by using Syrris ® Asia syringe pumps equipped with two injection valves and two sample loops (6 mL, each). With MeOH serving as carrier solvent, the feeds were combined in a Y-mixer. The resulting stream was directed through an Omnifit ® glass column (10 mm ID, adjustable height) containing 1.5 g of Amberlyst 15 or 5.0 g of Dowex 50WX8 as solid acid. The column was heated at 80 or 100 °C by using a Syrris ® Asia column heater and was pressurized at 5 bar. The reactor outlet was directed into a flask containing a stirred mixture of some MnO2 in MeOH in order to quench any excess oxidant. After reaching steady state, approx. 20 µL aliquots of crude material were collected. The samples were diluted with 1 mL of CH3CN/H2O 9:1 and was analyzed by analytical HPLC directly after the flow experiments.  TBHP solution in MeOH (added as 5.5 M decane solution) was pumped by using a Syrris ® Asia syringe pump (P3) equipped with an injection valve and a sample loop (3 or 6 mL). With MeOH serving as carrier solvent, the substrate and H2SO4 feeds were combined in a Y-mixer, then the resulting stream was mixed up with the H2O2 feed through a second Y-mixer. The combined liquid stream was next directed through a heated reaction coil (15 or 1.5 mL, 1-20 min residence time) which was pressurized at 5 bar. The reactor outlet was directed into a flask containing a stirred mixture of saturated aq. NaHCO3 and some MnO2 in order to quench any excess acid and/or oxidant. After reaching steady state, approx. 20 µL aliquots of crude material were collected. The samples were diluted with 1 mL of CH3CN/H2O 9:1 and was analyzed by analytical HPLC directly after the flow experiments.
See Table 1 in the manuscript for the corresponding reaction data and also Scheme 1 for the setup.

Additional parameter screening with benzaldehyde
Substrate solution containing 1.0 or 0.5 M of benzaldehyde and a 4.0 or 8.0 M solution of H2SO4 (prepared from cc H2SO4) were pumped as separate streams (P1 and P2) by using Syrris ® Asia syringe pumps equipped with two injection valves and two sample loops (5 or 10 mL, each). As a third stream, 35 wt% aq. H2O2 solution (11.6 M) or 2.0 M UHP solution in MeOH was pumped by using a Syrris ® Asia syringe pump (P3) equipped with an injection valve and a sample loop (3 or 5 mL). With MeOH serving as carrier solvent, the substrate and H2SO4 feeds were combined in a Y-mixer, then the resulting stream was mixed up with the oxidant feed through a second Y-mixer. The combined liquid stream was next directed through a 1.5-mL reaction coil (1, 2 or 4 min residence time) which was heated at 80-140 °C and pressurized at 5 bar. The reactor outlet was directed into a flask containing a stirred mixture of saturated aq. NaHCO3 and some MnO2 in order to quench any excess acid and/or oxidant. The reactor outlet was directed into a flask containing a stirred mixture of saturated aq. NaHCO3 and some MnO2 in order to quench any excess acid and/or oxidant. After reaching steady state, approx. 20 µL aliquots of crude material were collected. The samples were diluted with 1 mL of CH3CN/H2O 9:1 and was analyzed by analytical HPLC directly after the flow experiments.

Investigation of the reaction scope
Substrate solution containing 1.0 or 0.5 M of the corresponding aldehyde and a 4.0 or 8.0 M solution of H2SO4 (prepared from cc H2SO4) were pumped as separate streams (P1 and P2) by using Syrris ® Asia syringe pumps equipped with two injection valves and two sample loops (10 mL, each). As a third stream, 35 wt% aq. H2O2 solution (11.6 M) or 2.0 M UHP solution in MeOH was pumped by using a Syrris ® Asia syringe pump (P3) equipped with an injection valve and a sample loop (3 or 6 mL). With MeOH serving as carrier solvent, the substrate and H2SO4 feeds were combined in a Y-mixer, then the resulting stream was mixed up with the oxidant feed through a second Y-mixer. (For some of the reactions, EtOH or iPrOH was used as solvent and also as carrier solvent.) The combined liquid stream was next directed through a 1.5-mL reaction coil (2 or 4 min residence time) which was heated at 120 °C and pressurized at 5 bar. The reactor outlet was directed into a flask containing a stirred mixture of saturated aq. NaHCO3 and some MnO2 in order to quench any excess acid and/or oxidant. After reaching steady state, the crude reactor outlet was collected for 2-10 min. A 20 µL aliquot of the crude material was diluted with 1 mL of CH3CN/H2O 9:1 and was immediately analyzed by analytical HPLC or GC-FID. The collected mixture was extracted with CH2Cl2, washed with brine and dried over MgSO4. After evaporation, the obtained material was analyzed by 1 H and 13 C NMR spectroscopy. When necessary, column chromatographic purification was performed using mixtures of cyclohexane and EtOAc as eluent.

S10
Continuous flow oxidative esterification of 5: Substrate solution containing 1.0 M of γ-nitroaldehyde 5 and an 8.0 M solution of H2SO4 (prepared from cc H2SO4) were pumped directly as separate streams (P1 and P2) by using Syrris ® Asia syringe pumps. As a third stream, 35 wt% aq. H2O2 solution (11.6 M) in MeOH was pumped by using a Syrris ® Asia syringe pump (P3) equipped with an injection valve and a sample loop (15 mL). With MeOH serving as carrier solvent for the H2O2 stream, the substrate and H2SO4 feeds were combined in a Y-mixer, then the resulting stream was mixed up with the oxidant feed through a second Y-mixer. The combined liquid stream was next directed through a 1.5-mL reaction coil (2 min residence time) which was heated at 120 °C and pressurized at 5 bar. The reactor outlet was directed into a flask containing a stirred mixture of saturated aq. NaHCO3 and some MnO2 in order to quench any excess acid and/or oxidant. After reaching steady state, the crude reactor outlet was collected for 45 min. 20 µL aliquots of the crude material was diluted with 1 mL of CH3CN/H2O 9:1 and was immediately analyzed by analytical HPLC. The collected mixture was extracted with CH2Cl2, washed with brine and dried over MgSO4. The obtained material was analyzed by 1 H and 13 C NMR spectroscopy.

Calculation of green metrics
E factor, process mass intensity (PMI), reaction mass efficiency (RME), atom economy (AE) and optimum efficiency (OE) were calculated using the following equations. S1 Green metrics were calculated for the oxidative esterification of γ-nitroaldehyde 5 yielding γ-nitroester 6. The in situ-generated persulfuric acid-mediated flow process ( The NBS-mediated reaction was reported as a one-pot/two-step process. In the first step, Michael addition between 4-fluorocinnamaldehyde and nitromethane in the presence of a diphenylprolinol-type organocatalyst resulted γ-nitroaldehyde 5 (similarly to the reaction shown in Scheme 1). This step was reported to be quantitative and selective, thus calculation of the green metrics exclusively for the subsequent oxidative esterification step was possible.
In the NBS-mediated process, reaction conditions and yield were specified for the preparation of compound ent-6 only ((S)-3-(4-fluorophenyl)-4-nitrobutanoate). The (R)-isomer (compound 6) was prepared under identical conditions but using the opposite enantiomer of the chiral catalyst. For the calculation of green metrics, the same yield and reaction conditions were thus assumed for compound 6 as those reported for compound ent-6.
According to the NBS-mediated process, γ-nitroester 6 was isolated by means flash chromatographic purification. The exact amounts of solvents used for the purification was not reported by the authors. In case of the persulfuric acid-mediated flow process, 6 was obtained in a pure form after extractive work-up. In order that the data can be compared directly, chemicals and solvents used for work-up and purification have been excluded from the calculations.
For the calculation of the E factor, the mass of H2O was excluded. For the calculation of PMI, the total mass used for the calculation included H2O as well. Table S8. Values used for the assessment of green metrics of the in situ-generated persulfuric acid-mediated continuous flow oxidative esterification of γ-nitroaldehyde 5.

MeO
Methyl 3-(4-methoxyphenyl)propanoate 1 H and 13 C NMR data of the compound matches the reported literature. S7

1
H and 13 C NMR data of the compound matches the reported literature. S6

COOMe Ph
Methyl 2-phenylacetate 1 H and 13 C NMR data of the compound matches the reported literature. S8

COOMe Ph
Methyl 2-phenylpropanoate 1 H and 13 C NMR data of the compound matches the reported literature. S9

COOMe
Methyl cinnamate 1 H and 13 C NMR data of the compound matches the reported literature. S8

Ph COOMe
Methyl 3-phenylpropiolate 1 H and 13 C NMR data of the compound matches the reported literature. S10

COOMe
Methyl cyclohexanecarboxylate 1 H and 13 C NMR data of the compound matches the reported literature. S8

COOMe
Methyl cyclopentanecarboxylate H and 13 C NMR data of the compound matches the reported literature. S18

COOMe F
Methyl 3-fluorobenzoate 379.1 mg (92%) was isolated after extractive work-up. 1 H and 13 C NMR data of the compound matches the reported literature. S19