Reoptimization of the Organocatalyzed Double Aldol Domino Process to a Key Enal Intermediate and Its Application to the Total Synthesis of Δ12‐Prostaglandin J3

Abstract Re‐investigation of the l‐proline catalyzed double aldol cascade dimerization of succinaldehyde for the synthesis of a key bicyclic enal intermediate, pertinent in the field of stereoselective prostaglandin synthesis, is reported. The yield of this process has been more than doubled, from 14 % to a 29 % isolated yield on a multi‐gram scale (32 % NMR yield), through conducting a detailed study of the reaction solvent, temperature, and concentration, as well as a catalyst screen. The synthetic utility of this enal intermediate has been further demonstrated through the total synthesis of Δ12‐prostaglandin J3, a compound with known anti‐leukemic properties.


General Information
Anhydrous solvents were either dried using an Anhydrous Engineering alumina column drying system (THF,toluene,CH2Cl2) or obtained as Acroseal bottles and used directly (acetone), all other solvents used were reagent grade solvents and were used directly. Petroleum ether refers to the fraction collected between 40 -60 °C. Reactions requiring anhydrous conditions (where specified) were conducted under a N2 atmosphere using standard Schlenk techniques unless otherwise stated. All reagents were purchased from commercial sources and used as received, unless otherwise stated. 2,5-Dimethoxytetrahydrofuran (mixture of cis-and trans-isomers) was purchased from Acros Organics and was used as received. Morpholinium trifluoroacetate 6 (and all other analogous trifluoroacetate salts) was synthesised according to the literature procedure and the spectroscopic data were in agreement with the literature. [1] Methanesulfonyl chloride was purchased from Sigma-Aldrich and was distilled (Hickman apparatus) prior to use. Flash column chromatography was carried out using either Aldrich silica gel (40-63 μm) in a glass column, or using a Biotage Isolera™ Prime automated flash column chromatography system. All reactions were followed by thin-layer chromatography (TLC) when practical, using Merck Kieselgel 60 F254 fluorescent treated silica which was visualised under UV light (254 nm) or by staining with an aqueous basic potassium permanganate solution. A Mettler-Toledo OptiMax Synthesis Workstation was used for reactions conducted in a jacketed vessel. 1 H NMR spectra were recorded using either Jeol ECS 400 MHz, Bruker 400 MHz, Bruker Cryo 500 MHz, or Varian VNMR (400 MHz or 500 MHz) spectrometers. Chemical shifts (δ) are given in parts per million (ppm) and coupling constants (J) are given in Hertz (Hz). 13 C NMR spectra were recorded using either Varian VNMR 400 (101 MHz) or Bruker Cryo 500 (126 Hz) spectrometers. High resolution mass spectra (HRMS) were recorded on a Bruker Daltonics Apex IV by Electrospray Ionisation (ESI). IR spectra were recorded on a Perkin Elmer Spectrum One FT-IR as a thin film. Only selected absorption maxima (ν max ) are reported in wavenumbers (cm -1 ). Melting points (MP) were recorded in degrees Celsius (°C), using a were measured on a Bellingham and Stanley Ltd. ADO220 polarimeter, where c is given in g/100 mL. Chiral supercritical fluid chromatography (SFC) was performed using a Chiralpak® IA column (4.6 × 250 mm × 5μm) using a Waters TharSFC system and was monitored using a diode array detector (DAD). Chiral HPLC (HPLC) was performed on a HP Agilent 1100 using a Chiralpak® IC column (4.6 × 250 mm × 5μm) and was monitored using a diode array detector (DAD).

Optimization Studies.
We began our re-optimization by conducting a small solvent screen, employing the same reaction conditions as developed in a previous publication (Table S1, see entry 1 for our previously reported conditions). [2] Acetonitrile was identified as a slightly superior reaction solvent, providing enal 1 in a 16% NMR yield. [a] Yield determined by 1 H NMR spectroscopy using 1,3,5-trimethoxybenzene as an internal standard.
We found that decreasing the reaction concentration for the first step from 2.0 M to 1.0 M further improved the NMR yield to 19% (Table S2, entry 1). However, due to the dibenzylamine catalyst 5 reacting with 1 during the isolation process, only a 9% isolated yield was achieved.
This issue was alleviated by changing to thiomorpholine catalyst 6 for the second step of the reactionthis resulted in a 20% NMR and isolated yield (entry 2). Employing three further amine trifluoroacetate salts A-C did not improve the yield (entries 3-5). We then identified that the reaction can be improved by heating the reaction to 65 °C for the second step of the reactionthis allowed for a shorter reaction time for the second step (2 h) and a 23% NMR yield for 1 could be achieved when using acetonitrile as the reaction solvent (Table S3, entry 1). We conducted an extensive solvent screen with a second step temperature of 65 °C. Acetonitrile, THF, and 1,4-dioxane gave identical yields (Table S3, entries 3, 8 and   11). Using ethyl acetate as the reaction solvent provided a similar NMR yield (21%, entry 1), and a noticeable visual difference of the reaction mixture was observed. Typically large precipitation of oily oligomers is observed after the first step. However, less oligomeric precipitation was observed when using ethyl acetate as the solvent (see Figure S1). Notably, the oligomeric precipitate was a crystalline, water-soluble solidwe reasoned that this could allow for a simpler purification process at the end of the reaction. We therefore decided to use ethyl acetate as the reaction solvent in future optimization experiments. [a] Yield determined by 1 H NMR spectroscopy using 1,3,5-trimethoxybenzene as an internal standard. [b] Remaining succinaldehyde at the end of the reaction was determined by 1 H NMR spectroscopy using 1,3,5trimethoxybenzene as an internal standard. Figure S1. Images of inverted reaction vessels for the extended solvent screen after 20 h (first step)the label numbers correspond to the entry numbers in Table S3.

S5
The concentration of the first and second step of the reaction was explored next. The concentration of the first step was varied from 0.25 to 4.0 M using ethyl acetate as the reaction solvent, whilst using the same concentration for the second step (Table S4, entries 1-8). In general, the reaction yield improved as the reaction concentration was decreased, culminating in a 28% NMR yield being achieved with a 0.5 M reaction concentration (entry 2). However, further reducing the reaction concentration to 0.25 M only provided a 12% yield due to a poor conversion of succinaldehyde (entry 1). Again, there was a striking visual difference to the reaction mixtures at varying concentrations, with only minor yellow precipitate formation being observed at low concentrations, and large amounts of dark pink oligomeric precipitates forming at higher concentrations (see Figure S2). We saw further improvements to the reaction yield through dilution of the reaction mixture before the addition of the second catalyst for the second step from 0.5 M to 0.35 M (31%, entry 9). Finally, our optimal reaction conditions were obtained through increasing the concentration of the first step to 0.75 M, and decreasing the concentration of the second step to 0.2 M. This allowed for the formation enal 1 in a 33% NMR yield (entry 11).
[a] Yield determined by 1 H NMR spectroscopy using 1,3,5-trimethoxybenzene as an internal standard. [b] Remaining succinaldehyde at the end of the reaction was determined by 1 H NMR spectroscopy using 1,3,5trimethoxybenzene as an internal standard. [c] A second portion of L-proline (2 mol%) was added after 24 h. Figure S2. Images of reaction vessels for the concentration screen after 24 h (first step)the label numbers correspond to the entry numbers in Table S4.  Our final optimization studies were aimed at determining the optimal time for the second step (Table S5). An aliquot of the reaction conducted under the optimized reaction conditions was removed at hourly time points for four hours, and then analyzed by 1H NMR. This revealed that the optimal reaction time for the second step is 2 hours, and that the reaction yield begins to drop after 3 hours.

Detailed Synthetic Procedures for the Synthesis of Enal 1.
We have previously reported the preparation of succinaldehyde on scales ranging from 5g-200g using standard lab glassware. [2] The procedure below was carried out in a jacketed Optimax vessel which enables greater control and monitoring of internal reaction temperature.

Succinaldehyde synthesis
To a 1 L jacketed vessel, equipped with an overhead stirrer and a temperature probe, was added 2,5-dimethoxytetrahydrofuran (mixture of cis and trans-isomers, 200 mL, 1.55 mol), and H2O (400 mL). The reaction mixture was stirred at 300 RPM and the jacket temperature was warmed to 115 °C, over a 10 min periodthe reaction mixture (internal temperature) increased to 90 °C over a 40 min period, and the reaction was held i at this temperature for a further 2 h. The reflux condenser was moved into a distillation position, a 500 mL receiving flask was added and the flask was placed into an ice-bath. ii The jacket temperature was increased to 135 °C over a 15 min period, and the distillate iii was collected for 90 min. iv The jacket temperature was decreased to 90 °C and the pressure of the reaction vessel was reduced to ca. 100 mbar. v The distillation was continued until a further 190 mL of distillate was collected (430 mL in total). vi At this point, the pressure was returned to atmospheric pressure, toluene (200 mL) was added and the reaction temperature was warmed to 50 °C. The pressure was reduced to ca. 100 mbar and the reaction mixture was concentrated to remove the toluene (with concomitant azeotropic removal of water). vii This azeotropic removal of water with toluene was repeated twice more (2 x 200 mL), and the resultant yellow oil (ca. 100 g) was transferred to a 500 mL round bottom i The jacket temperature was maintained at 115 °C. ii All exposed (i.e. not covered by the heated jacket) hot surfaces were insulated with cotton wool and aluminium foil to ensure a consistent distillation. iii This distillate contains a mixture of H2O and MeOH, generated from the reactionsee Figure S4 for set-up. iv The temperature of the reaction slowly increased to ca. 100 °C during this time, and a minimum of 240 mL distillate was collectedif this volume was not achieved, then wait for a further 30 min (or until this volume has been achieved). v The reduction of pressure should be conducted in a controlled and gradual manner to ensure the reaction mixture does not excessively bumponce the pressure has been reduced to ca. 100 mbar, the reaction temperature will drop to ca. Analytical data were consistent with those reported in the literature. [2] Figure S4. Images from the succinaldehyde synthetic procedure: A) Jacketed vessel equipped with overhead stirrer and temperature probe (no reaction present in image). B) Jacketed vessel with distillation set-up (minus cotton wool and foil insulation). C) Succinaldehyde distillation set-up (minus oil bath and foil). D) Freshly distilled succinaldehyde appearance.
viii The crude reaction mixture containing succinaldehyde can be stored in a freezer as a solution in CH2Cl2 (ca. 4 mL g -1 ) for 2-3 weeks. The CH2Cl2 solution should be concentrated to dryness on a rotary evaporator prior to distillation of the succinaldehyde. ix The oil bath temperature was set to 75 °C. The hot surfaces were covered with aluminium foil to ensure an efficient distillation. The receiving flask was cooled in a dry ice/acetone bath. At this temperature the succinaldehyde is collected as a colourless solidat the end of the distillation the ice bath was removed, and the receiving flask was allowed to naturally warm to RT under reduced pressure (0.1 mbar).
A B D C S11 Figure S5. 1 H NMR spectrum of freshly distilled succinaldehyde (good quality).

(3aR,6aS)-2-hydroxy-3,3a,6,6a-tetrahydro-2H-cyclopenta[b]furan-5-carbaldehyde (1)
To a 50 mL round bottom flask equipped with a small (12 × 4.5 mm) stirrer bar was added freshly distilled succinaldehyde (500 mg, 5.81 mmol). Ethyl acetate ( 2 mol%) was added to the reaction, the reaction vessel was placed into a pre-heated (65 °C) oil bath and the reaction was stirred at 700 rpm for 2 h at this temperature. xiii An aliquot of the reaction mixture obtained at this point showed a 33% NMR yield (as an approximately 2:1 mixture of diastereomers) for enal 1. xii The reaction mixture was removed from the oil bath and the reaction was allowed to cool to RT. An aq. Na2SO4 (15 mL, 17% w/w) solution was added and the biphasic mixture was stirred vigorously for 10 min. The biphasic mixture was transferred to a 100 mL separating funnel and the phases were separated. The aqueous phase was extracted with EtOAc (2 × 30 mL) and the combined organic layers (29 mL + 2 × 30 mL) were dried (MgSO4), filtered and concentrated in vacuo to dryness. The crude reaction mixture was purified by flash column chromatography using a Biotage automated chromatography x It is essential to ensure full dissolution of succinaldehyde before the addition of L-prolinethe formation of pink/purple oligomers is rapidly observed if L-proline is added to a non-homogeneous mixture of ethyl acetate and succinaldehyde. xi Typically, 23-25 °C. xii An aliquot of ca. 50 μL was added to a standard NMR tube and was diluted with CDCl3 (0.5 mL). A 1 H NMR spectrum of this solution was obtained using a Varian 400 MHz NMR spectrometer (25 °C, 8 scans, 30 ° pulse angle, 30 s relaxation delay). The integration of the aromatic signal corresponding to 1,3,5-trimethoxybenzene [6.05 (3H, s)] was compared with the integration of either the aldehydic signal corresponding to succinaldehyde [9.78 (2H, s)] or the combined integration of the alkenyl signals corresponding to the two diastereoisomers of enal 1 [6.79 (1H, q) 6.64 (1H, q)], to quantify the amount of succinaldehyde and enal 1, respectively. xiii During this heating period, the reaction will turn from light pink, through to purple, and will end as a dark brown heterogeneous mixturesee Figure S10. Analytical data were consistent with those reported in the literature. [2][3] S14

g Scale Enal 1 Synthesis
To a 2 L round bottom flask equipped with a large (40 × 8 mm) stirrer bar was added freshly distilled succinaldehyde (50.0 g, 581 mmol). Ethyl acetate (774 mL, 0.75 M) was added and the mixture was briefly stirred (ca. 30 seconds) to ensure full dissolution of the succinaldehyde. x To the reaction mixture was added 1,3,5-trimethoxybenzene (internal standard, 2.44 g, 14.5 mmol, 2.5 mol%) followed by L-proline (1.34 g, 11.6 mmol, 2 mol%). succinaldehyde remaining) for enal 1. xii The reaction mixture was removed from the hot oil bath and was quickly cooled to 30 °C with an ice bath. The ice bath was removed, and the reaction was transferred to a 3 L conical flask equipped with a large stirrer bar. An aq. Na2SO4 (1.0 L, 17% w/w) xiv solution was added to the reaction mixture, the biphasic mixture was stirred vigorously for 10 min, and the mixture was transferred to a 5 L separating funnel. The phases were separated, and the aqueous phase was returned to the 3 L conical flask. This extraction process was repeated twice more (each with 10 min vigorous stirring) to further extract the aqueous phase with with EtOAc (2 x 1.0 L), and the combined organic layers (1.66 L + 2 × 1.0 L) were dried (MgSO4), filtered and concentrated in vacuo to a ca. 1 L volume. xv Pre-treated wet silica xvi (100 g SiO2 + 75 mL H2O) was added, and the mixture was stirred vigorously for 30 mins. The mixture was filtered through a Buchner funnel equipped with filter paperthe silica was washed with EtOAc (2 x 250 mL) and the combined mother liquors were concentrated in vacuo to dryness. xvii The crude reaction mixture was further xiv The aq. Na2SO4 (17% w/w) solution was prepared by adding Na2SO4•10H2O (385 g) portionwise to vigorously stirred H2O (785 mL). The stirring was continued until complete dissolution of Na2SO4 was observed gentle heating was required. xv An aliquot of the combined organic phases was analysed by 1 H NMR to reveal that 7% succinaldehyde was remaining following the aqueous work up. xvi The pre-treated wet silica was prepared by adding water to the dry silica in a sealable containerthe container was closed with a lid, and the mixture was shaken vigorously until the consistency of the wet silica became uniform, and the silica was then left to stand for 1 h. xvii An aliquot of the combined organic phases was analysed by 1 H NMR to reveal that 4% succinaldehyde was remaining following the silica filtration.
xviii The pre-treated wet silica was manually packed into an emptied 340 g Biotage SNAP® KP-SIL cartridge. S16 Figure S7. 1 H NMR of Enal 1 isolated from a 50 g scale reaction.  Figure S11. Images illustrating the purification process for the synthesis of enal 1. A) Large sinter funnel containing silica (yellow/ brown colorization due to adherance of oligomers) from the filtration of the crude reaction mixture following treatment with wet silica and B) the resultant viscous dark brown oil obtained after concentrating the mother liquour to dryness. C) Visual appearance of the silica gel column after being air-purged following purification of the crude reaction mixturea clear visible dark brown band caused by adherance of oligomeric material can be observed at the top of the column. D) Visual appearnce of enal 1 (brown solid).
xx The substrate begins to precipitate out of solution at −78 °C if the addition of DIBAL-H is not performed immediately. However, in such case addition of DIBAL-H gradually redissolves the precipitated substrate.
(iii). Elimination: 1,.0]undec-7-ene (DBU, 457 mg, 3.00 mmol, 6.0 equiv.) was added to a solution of the crude mesylated product in CH2Cl2 (10 mL) in a round bottom flask at −10 °C under an argon atmosphere and stirred at 0 °C. After 4 h, the reaction was quenched by adding brine (3 mL) and the reaction mixture was extracted with dichloromethane (3 × 3 mL). The combined organic layers were dried over MgSO4, evaporated under reduced pressure. The crude eliminated product was directly taken into the next step without further purification.
Analytical data were consistent with those reported in the literature. [5]