Aromatic Functionalized Indanones and Indanols: Broad Spectrum Intermediates for Drug Candidate Diversification

Abstract

A series of new aromatic substituted indanone and indanol building blocks have been prepared and are anticipated to aid future drug discovery studies. In total, seven compounds (7, 12–17) are expounded on, and all have been fully characterized. In doing so, we have shown multiple examples of highly chemoselective reactions. One example employed an adaptation of Fujioka’s chemoselective reduction methodology, allowing an ester to be reduced in the presence of a ketone. In another example, an uncommon benzylic methyl group to aldehyde oxidation was demonstrated for two different compounds. These and other chemoselective interconversions allowed us to identify compound (12) as a remarkably flexible springboard for accessing a diverse array of indan-based building blocks (13–17).

1. Introduction

The medicinal potential of compounds 1–3 (Figure 1) is diverse. They range from the treatment of diabetes (1) [1] to serotonin receptor agonist probes (2) [2] to analogs (3) of carbovir [3,4], the HIV reverse transcriptase drug. Common to all those compounds is the indanone structural motif or the need for an indanone or indanol building block to synthesize them. While the shown drug candidates were ultimately superseded by superior ones, this has not dissuaded medicinal chemists from continuing to explore indanone and indanol intermediates as medicinal chemistry building blocks [5,6].

Indanone 4 (Figure 2) is the main indan building block used for the syntheses of 13 and was always converted into either 5 or 6 to reach those targets. Examples of indanone 4 with aromatic -OH, -OR, and/or alkyl group substituents are limited in number for representative but not exhaustive examples (see [7,8,9,10]), while 5 and 6 have not been reported with those aromatic substituents. Bearing those facts in mind and a 2015 patent claiming the usefulness of aromatic substituted indanones within drug discovery [1] (pp. 1, 42), we were drawn to the potential of indanone 7 (Figure 2) for accessing new indan-based medicinal chemistry building blocks. We surmised that 7 could serve that role, despite its dense and rich number of functional groups, by undergoing chemoselective electrophile trappings and/or functional group interconversions. Accordingly, this manuscript outlines our first achievements toward that ambition by providing substituted aromatic analogs of 4–6.

2. Materials and Methods

2.1. General Experimental Details

The experimental descriptions of seven previously uncharacterized compounds are provided here. For the corresponding spectra (¹H and ¹³C NMR, IR, and HRMS), please refer to the Supplementary Materials document submitted with this work.

Commercial reagents were used as received, unless otherwise stated, from their commercial source, usually Merck. Routine monitoring of reactions was performed by thin-layer chromatography (TLC) using precoated plates of silica gel 60 F254 and visualized under ultraviolet irradiation (254 nm) and CAM (ceric ammonium molybdate) staining solution. Column chromatography was performed using silica gel 60 (0.040–0.063 mm), and EtOAc and petroleum ether (pet ether), with a boiling point range of 40–60 °C, were used. A JEOL NMR (ECX) 400 MHz instrument was operated at 400 MHz (¹H) and 100 MHz (¹³C) for NMR analysis. Chemical shifts (δ) were reported in parts per million (ppm) relative to CHCl₃ (7.26 ppm) for ¹H NMR and relative to CHCl₃ (77.16 ppm) for ¹³C NMR. Multiplicities are abbreviated as (s = singlet, d = doublet, t = triplet, q = quartet, bs = broad singlet, m = multiplet).

2.2. Synthesis of 4-Hydroxy-7-methyl-3-oxo-2,3-dihydro-1H-indene-1-carboxylic acid (7)

A positive flow of nitrogen was directed into one neck of a double-neck 1.0 L round-bottom flask containing a stir bar (4 × 1.5 cm). In the other neck was placed a glass funnel, and aluminum chloride (98%, MW = 133.34 g/mol, 8.00 equivalent, 740 mmol, 98.63 g) and sodium chloride (99%, MW = 58.44 g/mol, 4.00 equivalent, 370 mmol, 21.62 g) were added. The addition neck was then sealed with a glass stopper, and a positive nitrogen atmosphere was maintained while the mixture was heated to 140 °C with rapid stirring of the solids. Note 1: After ≈ 30 min at 140 °C, the aluminum chloride and sodium chloride melt to form a semi-clear, uniform, viscous liquid. The stoppered neck was then freed, the nitrogen pressure increased to maintain a good flow out of the opened neck, and a glass funnel was inserted. p-Cresol (99%, MW = 108.14 g/mol, 1.00 equivalent, 92.5 mmol, 10.00 g, d = 1.03 g/mL, 9.67 mL) was added, followed by the slow addition of maleic anhydride (≥99%, MW = 98.06 g/mol, 1.00 equivalent, 92.5 mmol, 9.07 g). Note 2: Using either maleic anhydride or p-cresol in excess, specifically 1.3 equivalent, did not result in increased yield. Note 3: Caution: Maleic anhydride was added slowly in small portions over 10 min to suppress small but erratic bursts/eruptions of material from the stirred reaction mass. The addition neck was again stoppered. The reaction mixture turned black in color and became homogenous soon after the addition of maleic anhydride. The oil bath temperature was elevated to 180 °C, and stirring was continued at that temperature for a total of 30 min. The reaction mixture continues to be homogenous and black in color. The completion of the reaction was monitored using TLC. To do that, a small portion of the reaction mixture was taken out using a disposable glass pipette, quenched in aqueous HCl (2.0 N, 2 mL), and extracted with EtOAc (≈0.1 mL). The disappearance of both starting materials was observed on TLC. The reaction was performed no less than six times at this scale and was always complete after 30 min. Note 4: Attempts to scale up the reaction, e.g., doubling or tripling the scale, resulted in approximately half the yield. We surmise this is due to inefficient stirring due to equipment limitations in larger round-bottom flasks.

Work-up: The reaction mixture was brought to room temperature over approximately 45 min. After the mixture dipped below 100 °C, a hard, rock-like black mass ensued, and stirring ceased. The mixture was quenched at 25 °C by the slow addition of ice-cold water (250 mL). This process was strongly exothermic. After cooling to 25 °C, 2.0 N HCl (≈50 mL) was slowly added. The pH of the solution was measured to be ≈2. The solid black mass became a heterogeneous solution in which sticky chunks of black material floated or sat at the bottom of the flask when no stirring was provided. Celite (10 g) was added, which improved the stirring and reduced the sticky chunks. The mixture was further stirred for 10 min at 25 °C and was then filtered through a celite bed (2 cm in height on Whatman filter paper, 110 mm diameter, pore size = 11 μm) using a Büchner funnel (15 cm diameter). The celite bed was washed with EtOAc (500 mL) until the desired compound was fully eluted from the celite bed. This was confirmed by TLC analysis of the filtrate. After filtration, a black-colored, free-flowing solid material remained on the celite bed. The organic layer and aqueous layer were separated from the filtrate using a separatory funnel (1.0 L). The aqueous layer was again extracted with EtOAc (3 × 100 mL) until no product remained in the aqueous layer (confirmed by TLC). The combined organic extracts were dried (Na₂SO₄) and concentrated under rotary evaporation (bath temperature 35–40 °C) to obtain a viscous black crude product (15.6 g).

Purification: The crude product (15.6 g), with EtOAc added, was uniformly slurried with approximately two times its weight of silica gel (230–400 mesh) and concentrated using rotary evaporation (recommended bath temperature: <35 °C). The resulting free-flowing material was added on top of a silica gel (230–400 mesh) column (16 cm in height, 4.5 cm in diameter). Mobile-phase elution began with 5% EtOAc in petroleum ether doped with acetic acid (1 vol%). This solvent ratio was maintained until all colored impurities were removed from the column. Product elution began when using 10% EtOAc in petroleum ether doped with acetic acid (1 vol%). The concentration of the pure fractions provided 7 as a yellow solid (MW = 206.20 g/mol, 7.70 g, 37.3 mmol, 40% yield). Despite the low yield, the reaction was reliable.

• R_f = 0.40, EtOAc/petroleum ether (30:70) doped with acetic acid.
• Melting Point: 149–150 °C.
• ¹H NMR (CDCl₃, 400 MHz) δ 7.33 (d, J = 8.3 Hz, 1H), 6.81 (d, J = 8.3 Hz, 1H), 4.23 (dd, J = 7.2, 3.4 Hz, 1H), 3.05–2.90 (m, 2H), 2.29 (s, 3H).
• ¹³C NMR (CDCl₃, 100 MHz) δ 207.1, 178.0, 155.8, 148.2, 139.5, 127.2, 121.8, 115.9, 43.2, 40.8, 17.2.
• IR (ATR, cm⁻¹) ν = 3400, 1686, 1603.
• HRMS (ESI-QTOF) m/z: [M − H⁺]⁻ Calculated for C₁₁H₉O₄: 205.0495; Found: 205.0506; Error: 5.4 ppm.

2.3. Synthesis of Methyl 4-Methoxy-7-methyl-3-oxo-2,3-dihydro-1H-indene-1-carboxylate (12)

A positive flow of nitrogen was directed into one neck of a double-neck round-bottom flask (500 mL) containing a stir bar (3 × 1.5 cm). In the other neck, a glass funnel was inserted, and compound 7 (MW = 206.20 g/mol, 1.00 equiv, 33.9 mmol, 7.00 g) and acetone (≥99.5%, 0.1 M, 339 mL), via a graduated cylinder, were added to the flask. Note that while anhydrous methods were not used, e.g., no precautions were taken to keep the acetone anhydrous (screw cap storage), there were also no extended periods of atmospheric air exposure. The addition neck was then closed using a natural rubber septum. After stirring for 2 min at 25 °C, the mixture became a clear, brown-colored solution. Through the addition neck, potassium carbonate (≥99.0%, MW = 138.21 g/mol, 4.00 equiv, 136 mmol, 18.74 g) was added to the flask while stirring, followed by the addition of dimethyl sulfate (≥99.5%, MW = 126.13 g/mol, d = 1.33 g/mL, 4.00 equiv, 136 mmol, 12.9 mL) at 25 °C. The addition neck was sealed. The reaction mixture was brown in color and heterogeneous in nature. A reflux condenser was added, and gentle reflux (55–60 °C) was maintained for 8 h. For TLC reaction monitoring, we recommend adding one drop (disposal glass pipette) of the reaction medium to a vial (2 mL), followed by evaporation of the acetone with pressurized nitrogen. EtOAc (≈0.1 mL) and water (≈1.0 mL) were then added. TLC of the EtOAc showed the reaction was only complete at 8 h. The reaction was performed no less than six times at this scale. This reaction was performed once at double the scale (15 g), and the yield and reaction time remained the same.

Work-up: The reaction mixture was brought to 25 °C and concentrated using rotary evaporation (recommended bath temperature: 25 °C or lower) until the acetone was completely removed. Water (100 mL) was added to the mixture to dissolve the potassium carbonate and stirred for 10 min. The aqueous phase was extracted with EtOAc (2 × 100 mL). The combined organic extracts were dried over Na₂SO₄, filtered, and concentrated using rotary evaporation to give the crude product (12.0 g) as a viscous dark brown oil.

Purification: The crude product (12.0 g), with EtOAc added, was uniformly slurried with approximately two times its weight of silica gel (230–400 mesh) and concentrated using rotary evaporation (recommended bath temperature <35 °C). The resulting free-flowing material was added on top of a silica gel (230–400 mesh) column (16 cm in height, 4.5 cm in diameter). Mobile phase elution began with 10% EtOAc in petroleum ether and was gradually increased up to 40% EtOAc/petroleum ether, at which point the product started eluting out. The concentration of the pure fractions provided 12 as a yellow solid (MW = 234.25 g/mol, 7.16 g, 30.6 mmol, 90% yield).

• R_f = 0.30, EtOAc/petroleum ether (60:40)
• Melting Point: 103–104 °C
• ¹H NMR (CDCl₃, 400 MHz) δ 7.35 (d, J = 8.4 Hz, 1H), 6.81 (d, J = 8.4 Hz, 1H), 4.17 (dd, J = 8.4, 3.0 Hz, 1H), 3.93 (s, 3H), 3.69 (s, 3H), 2.93 (dd, J = 18.6, 8.4 Hz, 1H), 2.80 (dd, J = 18.6, 3.0 Hz, 1H), 2.25 (s, 3H).
• ¹³C NMR (CDCl₃, 100 MHz) δ 202.5, 173.5, 156.9, 152.5, 138.3, 128.0, 125.0, 111.2, 56.4, 53.0, 43.4, 42.1, 17.6.
• IR (ATR, cm⁻¹) ν = 2962, 1732, 1697, 1605, 1582.
• HRMS (ESI-QTOF) m/z: [M + H⁺]⁺ Calculated for C₁₃H₁₅O₄: 235.096485; Found: 235.095968; Error: 2.2 ppm.

2.4. Synthesis of 3-(Hydroxymethyl)-7-methoxy-4-methyl-2,3-dihydro-1H-inden-1-one (13)

A positive flow of nitrogen was delivered via one neck of a three-neck round-bottom flask (250 mL) containing a stir bar (3 × 1.5 cm). In another neck, a graduated addition funnel (closed with a natural rubber septum) was added, and the joint was sealed with Teflon tape. In the third neck, a glass funnel was inserted, and compound 12 (MW = 234.25 g/mol, 1.00 equiv, 17.1 mmol, 4.00 g) was added. The neck was stoppered using a natural rubber septum. CH₂Cl₂ (≥99.5%, purchased with ≤0.02% water content (no sure-seal cap was present, the bottle was capped with a natural rubber septum and stored under nitrogen), 0.4 M, 42.75 mL) was added via glass syringe through the third neck. Within 2 min of stirring at 25 °C, the starting material dissolved. Dry THF (sodium and benzophenone distilled, 0.8 M, 21.4 mL) was added via a glass syringe through the third neck. The mixture was stirred for 2 min at 25 °C. Tributylphosphine (nBu₃P) (≥93.5%, neat, MW = 202.32 g/mol, 1.20 equiv, 20.52 mmol, d = 0.82 g/mL, 5.13 mL) was slowly added over 1 min via syringe at 25 °C. Approximately 2 min after the complete addition of the nBu₃P, a homogenous brown-colored solution ensued. Trimethylsilyl trifluoromethanesulfonate (TMSOTf) (≥98.0%, MW = 222.26 g/mol, 1.20 equiv, 20.5 mmol, d = 1.228 g/mL, 3.71 mL) was slowly added via glass syringe over 2 min due to the exothermic nature of the addition at 25 °C. White fumes were observed during TMSOTf addition. The reaction mixture was stirred for 1 h at 25 °C. Diisobutylaluminium hydride (DIBAL-H, 1.0 M in hexanes, MW = 142.22 g/mol, 3.00 equiv, 51.3 mmol, 51.3 mL) was transferred to the graduated addition funnel via cannula. The reaction flask was first cooled in an ice-water bath, and only then was DIBAL-H added dropwise from the graduated addition funnel over 20 min. The reaction was a brown-colored homogenous solution. To monitor consumption of the starting material (12), a small aliquot (≈50 μL) was taken by disposable syringe into a sample vial (2.0 mL), 2 drops of tetrabutylammonium fluoride (TBAF, 1.0 M in THF) were added, and this mixture was directly spotted onto the TLC plate. After complete DIBAL-H addition, the reaction was stirred for 30 min at 0 °C and then worked up. This reaction was also performed on a larger scale (10 g) and was complete at the same time, with a very similar yield.

Work-up: Aqueous NaHCO₃ (saturated solution, 40 mL) was slowly added to the reaction; effervescence was observed and ceased after the complete addition of the aqueous NaHCO₃. The mixture became a heterogeneous solution in which sticky chunks of white material floated or sat at the bottom of the flask when no stirring was provided. Celite (10 g) was added to improve stirring and reduce the agglomeration of sticky material into balls. Methanol (20 mL) was added, and after adding a reflux condenser, the mixture was heated at 40 °C for 2 h. After 2 h, the mixture was cooled down to 25 °C and filtered through a celite bed (2 cm height on Whatman filter paper, 110 mm diameter, pore size = 11 μm) using a Büchner funnel (diameter 15 cm). The celite bed was washed portion-wise with EtOAc (≈200 mL) until the product was fully eluted from the celite bed (confirmed by TLC analysis of the last celite bed rinse). After filtration, a white-colored, free-flowing solid remained on the celite bed. The filtrated aqueous layer was extracted with EtOAc (4 × 50 mL). The combined organic phase was dried over Na₂SO₄, filtered, and rotary evaporated (recommended bath temperature <35 °C). Crude product weight: 6.25 g.

Purification: The crude product (6.25 g), with added EtOAc, was uniformly slurried with approximately two times its weight of silica gel (230–400 mesh) under a rotary evaporator (recommended bath temperature < 35 °C). The resulting free-flowing material was added on top of a silica gel (230–400 mesh) column (16 cm in height, 4.5 cm in diameter). The mobile phase elution began with 10% EtOAc in petroleum ether and was gradually increased up to 60% EtOAc/petroleum ether, at which point the product started eluting out. The concentration of the pure fractions provided 13 as a white solid (MW = 206.24 g/mol, 2.46 g, 11.93 mmol, 70% yield).

• R_f = 0.20, EtOAc/petroleum ether (80:20)
• Melting Point: 133–134 °C
• ¹H NMR (CDCl₃, 400 MHz) δ 7.34 (d, J = 8.4 Hz, 1H), 6.76 (d, J = 8.4 Hz, 1H), 3.98 (m, 1H), 3.91 (s, 3H), 3.64 (m, 1H), 3.55 (m, 1H), 2.80 (dd, J = 18.5, 7.4 Hz, 1H), 2.68 (dd, J = 18.5, 1.8 Hz, 1H), 2.35 (s, 3H).
• ¹³C NMR (CDCl₃, 100 MHz) δ 204.2, 156.2, 154.7, 137.7, 126.9, 125.4, 110.0, 64.8, 55.7, 41.9, 40.2, 17.3.
• IR (ATR, cm⁻¹) ν = 3360, 2932, 1677, 1585, 1493.
• HRMS (ESI-QTOF) m/z: [M+H⁺]⁺ Calculated for C₁₂H₁₅O₃: 207.101571; Found: 207.101448; Error: 0.6 ppm.

2.5. Synthesis of 3-(Hydroxymethyl)-7-methoxy-1-oxo-2,3-dihydro-1H-indene-4-carbaldehyde (14)

In a single-neck round-bottom flask (100 mL) containing a stir bar (1 × 0.5 cm), potassium persulfate (≥99.0%, MW = 270.32 g/mol, 2.24 equiv, 5.43 mmol, 1.47 g) was added, followed by copper(II) sulfate pentahydrate (≥98.0%, MW = 249.69 g/mol, 0.23 equiv, 0.55 mmol, 139 mg). Deionized water (0.2 M, 12.1 mL) was added, and despite stirring, the materials did not fully dissolve. In another flask, compound 13 (MW = 206.24 g/mol, 1.00 equiv, 2.42 mmol, 500 mg) was dissolved in acetonitrile (≥99.8%, 0.16 M, 14.7 mL), and this solution was added over 1 min to the aqueous-containing reaction flask. The reaction mixture remained heterogeneous. Pyridine (MW = 79.10 g/mol, 2.0 equiv, 4.84 mmol, d = 0.982 g/mL, 0.39 mL) was added, and the mixture turned blue in color and remained heterogeneous. A reflux condenser was added, and the mixture was stirred for 3 h at a gentle reflux (75–80 °C) under an open atmosphere (the reflux condenser was not stoppered). As the temperature reached 75–80 °C, the mixture became a homogenous blue-colored solution. After 2 h, the reaction color turned green. TLC monitoring at 2 h showed persistence of the starting material (13), but at 3 h the homogenous solution turned yellow in color, and TLC analysis showed no 13 remained. To monitor the reaction, a small portion (≈25 μL) was taken and quenched by adding it to an aqueous NaHCO₃ solution (saturated, ≈1.0 mL) with EtOAc (≈0.1 mL) present. The EtOAc was sampled and co-spotted against 13 on a TLC plate.

Work-up: Aqueous NaHCO₃ (saturated solution, 10 mL) was slowly added to the reaction mixture. Effervescence was observed while adding the NaHCO₃ solution and ceased after complete addition. The reaction mixture was extracted with EtOAc (4 × 50 mL). The aqueous layer was checked for residual products using TLC. If aqueous dissolved product remained, it was extracted using EtOAc (2 × 50 mL). The combined organic phase was dried over Na₂SO₄, filtered, and rotary evaporated (recommended bath temperature: ≤25 °C). Crude product weight: 420 mg.

Purification: The crude product (420 mg), with added EtOAc, was uniformly slurried with approximately two times its weight of silica gel (230–400 mesh). Under rotary evaporation (recommended bath temperature: 25 °C), the solvent was removed. The resulting free-flowing material was added on top of a silica gel (230–400 mesh) column (16 cm in height, 2 cm in diameter). Mobile phase elution began with 10% EtOAc in petroleum ether and was gradually increased up to 80% EtOAc/petroleum ether, at which point the product eluted out. The concentration of the pure fractions provided 14 as a light brown solid (MW = 220.22 g/mol, 283 mg, 1.29 mmol, 53% yield). This reaction was examined on a larger scale (1 g) on two occasions but resulted in a lower yield (30%). In these situations, the persistence of the starting material was noted, and extended reaction times did not overcome the problem and may have oxidized the desired aldehyde to the corresponding carboxylic acid.

• R_f = 0.15, EtOAc/petroleum ether (90:10)
• Melting Point: 141–142 °C
• ¹H NMR (CDCl₃, 400 MHz) δ 10.07 (s, 1H), 8.05 (d, J = 8.6 Hz, 1H), 7.01 (d, J = 8.6 Hz, 1H), 4.16 (q, J = 5.7 Hz, 1H), 4.05 (s, 3H), 3.88 (m, 1H), 3.80 (m, 1H), 2.85 (dd, J = 18.6, 7.6 Hz, 1H), 2.68 (dd, J = 18.6, 1.6 Hz, 1H), 1.78 (t, 1H).
• ¹³C NMR (CDCl₃, 100 MHz) δ 202.7, 190.5, 162.2, 159.8, 141.5, 126.7, 126.3, 110.4, 66.4, 56.6, 41.7, 40.0
• IR (ATR, cm⁻¹) ν = 3389, 2964, 1677, 1574.
• HRMS (ESI-QTOF) m/z: [M + H⁺]⁺ Calculated for C₁₂H₁₃O₄: 221.0808; Found: 221.0816; Error: −3.4 ppm.

2.6. Synthesis of 4-Methoxy-3-oxo-2,3-dihydro-1H-indene-1,7-dicarbaldehyde (15)

Method A (from compound 14). A positive flow of nitrogen was directed into one neck of a double-neck round-bottom flask (50 mL) containing a stir bar (1 × 0.5 cm). In the other neck, a glass funnel was inserted, and Dess–Martin periodinane (89.99% pure by GC, 99.45% pure by titration, MW = 424.14 g/mol, 1.75 equiv, 1.59 mmol, 674 mg) and sodium hydrogen carbonate (≥99.5%, MW = 84.01 g/mol, 5.00 equiv, 4.55 mmol, 382 mg) were added. The neck was then stoppered using a natural rubber septum. CH₂Cl₂ (≥99.5%, purchased with ≤0.02% water content [no sure-seal cap was present, the bottle was capped with a natural rubber septum and stored under nitrogen], 0.05 M, 18.2 mL) was added via glass syringe. The mixture was cooled using an ice-water bath. The mixture was a heterogeneous, colorless solution. In another flask, compound 14 (MW = 220.22 g/mol, 1.00 equiv, 0.91 mmol, 200 mg) was fully dissolved in acetonitrile (≥99.8%, 0.5 M, 1.8 mL), and this solution was added (over 1 min) to the reaction flask. The final molarity is 0.045 M. The reaction mixture remains heterogeneous throughout the entire reaction. The ice bath was removed within 5 min of adding 14, and the reaction was complete after 5 h of stirring at 25 °C. After 1 h of stirring at 25 °C, a white suspended material was noted and stayed for the remainder of the reaction. To monitor consumption of the starting material (14), an aliquot containing the reaction liquid (≈25 μL) was removed (disposable syringe) and added to a vial (2 mL) containing CH₂Cl₂ (≈100 μL), an aqueous saturated sodium thiosulfate solution (≈0.2 mL), and a saturated aqueous sodium bicarbonate solution (≈0.2 mL). The capped vial was shaken vigorously. TLC of the CH₂Cl₂ layer was required.

Method B (from compound cis-17). A positive flow of nitrogen was directed into one neck of a double-neck round-bottom flask (50 mL) containing a stir bar (1 × 0.5 cm). In the other neck, a glass funnel was inserted, and Dess–Martin periodinane (89.99% pure by GC, 99.45% pure by titration, MW = 424.14 g/mol, 3.00 equiv, 2.69 mmol, 1.14 g) and sodium hydrogen carbonate (≥99.5%, MW = 84.01 g/mol, 5.00 equiv, 4.50 mmol, 378 mg) were added. The neck was stoppered using a natural rubber septum. CH₂Cl₂ (≥99.5%, purchased with ≤0.02% water content (no sure-seal cap was present, the bottle was capped with a natural rubber septum and stored under nitrogen), 0.05 M, 18.0 mL) was added via glass syringe. The mixture was cooled using an ice-water bath. The reaction mixture was a heterogeneous, colorless solution. In another flask, compound cis-17 (MW = 222.22 g/mol, 1.00 equiv, 0.90 mmol, 200 mg) was fully dissolved in acetonitrile (≥99.8%, 0.5 M, 1.8 mL), and this solution was added (over 1 min) to the reaction flask. The final molarity is 0.045 M. The reaction mixture remains heterogeneous throughout the entire reaction. The ice bath was removed within 5 min of adding cis-17, and the reaction was complete after 90 min of stirring at 25 °C. After 1 h of stirring at 25 °C, a suspended white material was noted and stayed for the remainder of the reaction. To monitor consumption of the starting material (cis-17), an aliquot containing the reaction liquid (≈25 μL) was removed (disposable syringe) and added to a vial (2 mL) containing CH₂Cl₂ (≈100 μL), an aqueous saturated sodium thiosulfate solution (≈0.2 mL), and a saturated aqueous sodium bicarbonate solution (≈0.2 mL). The capped vial was shaken vigorously. TLC of the CH₂Cl₂ layer was required.

Work-up and Purification (Methods A and B): Aqueous Na₂S₂O₃ (saturated solution, 10 mL) and aqueous NaHCO₃ (saturated solution, 10 mL) were added to the reaction mixture and stirred for 5 min at 25 °C. The solution was biphasic, and the organic phase was brown-colored while the aqueous phase was red-colored. The organic phase was separated from the aqueous phase using a separatory funnel, and the organic phase was returned to the round-bottom flask. Fresh aqueous Na₂S₂O₃ (saturated solution, 10 mL) was added, and the biphasic solution was stirred for 5 min. The aqueous phase was slightly red in color, while the organic phase turned orange in color. The organic phase was separated from the aqueous phase. The organic phase was washed for a third time with a fresh aqueous Na₂S₂O₃ solution (saturated solution, 10 mL) with 5 min of stirring. After phase separation, the aqueous layer was colorless, and the organic phase remained orange in color. The lack of the red color in the aqueous phase (third aqueous washing) appears to indicate when complete removal of excess DMP and/or reduced DMP by-products from the organic phase has occurred. All three aqueous phases were combined and back-extracted with CH₂Cl₂ (15 mL), and we recommend that this CH₂Cl₂ extract be washed for 5 min with fresh aqueous saturated sodium thiosulfate before adding it to the main organic extract. Lastly, the combined organic extracts were washed with water, dried over Na₂SO₄, filtered, and concentrated under rotary evaporation (recommended bath temperature: ≤25 °C) to give dialdehyde 15 as a yellow solid. For method A: 180 mg (MW = 218.21 g/mol, 0.83 mmol, 91% yield). For method B: 177 mg (MW = 218.21 g/mol, 0.81 mmol, 90% yield).

The purity of the produced dialdehyde (15), after work-up (no chromatography), was high and required no further purification. The method A material was used to collect all characterization data (see Supplementary Materials). A ¹H NMR was also collected for the method B material (see Supplementary Materials). The TLC examination of the dialdehyde did not provide a single spot. Furthermore, attempts at silica gel chromatography using mixtures of EtOAc/petroleum either with or without a dopant (Et₃N or AcOH) always resulted in a greatly reduced yield of the dialdehyde, still impure, and with new impurities.

• All data and spectroscopy were collected for the Method A dialdehyde product (15).
• R_f = 0.20, EtOAc/petroleum ether (90:10)
• Melting Point: 75–76 °C.
• ¹H NMR (CDCl₃, 400 MHz) δ 10.00 (s, 1H), 9.94 (d, J = 1.3 Hz, 1H), 8.06 (d, J = 8.5 Hz, 1H), 7.08 (d, J = 8.6 Hz, 1H), 4.93 (dd, J = 8.5, 3.1 Hz, 1H), 4.06 (s, 3H), 2.97 (dd, J = 19.0, 3.1 Hz, 1H), 2.85 (dd, J = 19.0, 8.5 Hz, 1H).
• ¹³C NMR (CDCl₃, 100 MHz) δ 200.5, 197.4, 190.3, 162.5, 153.8, 142.5, 126.2, 126.1, 111.1, 56.7, 49.5, 37.7.
• IR (ATR, cm⁻¹) ν = 3412, 2947, 2849, 1700, 1683, 1591, 1574.
• HRMS (ESI-QTOF) m/z: [M − H⁺]⁻ Calculated for C₁₂H₉O₄: 217.0506; Found: 217.0507, Error: −0.3 ppm.

2.7. Synthesis of 3-(Hydroxymethyl)-7-methoxy-4-methyl-2,3-dihydro-1H-inden-1-ol (cis-16)

A positive flow of nitrogen was directed into one neck of a three-neck round-bottom flask (500 mL) containing a stir bar (3 × 1.5 cm). In one of the other two necks, a graduated addition funnel (closed with a natural rubber septum) was added, and the joint was sealed with teflon tape. In the third neck, a glass funnel was placed, and compound 12 (MW = 234.25 g/mol, 1.00 equiv, 21.3 mmol, 5.0 g) was added. This neck was then sealed using a natural rubber septum. Dry THF (≥99.5% and distilled from sodium and benzophenone, 0.1 M, 213.40 mL) was added to a single-neck round-bottom flask (500 mL) and subsequently transferred to the reaction flask via cannula. After stirring for 2 min at 25 °C, a light yellow-colored homogenous solution ensued. Lithium borohydride (MW = 21.78 g/mol, 5.0 equiv, 106.7 mmol, 2.0 M in THF, 53.36 mL) was transferred to the graduated addition funnel via cannula. The reaction mixture was cooled using an ice bath, and then the LiBH₄ solution was added dropwise over 20 min. After complete addition, the mixture became brown in color and remained homogenous. The ice bath was removed, and the reaction mixture was allowed to stir at 25 °C for 3 h. To monitor consumption of the starting material, a reaction aliquot was taken using a disposable syringe (≈50 μL) and quenched in a vial (2 mL) containing an aqueous saturated ammonium chloride solution (≈1.0 mL) and EtOAc (≈100 μL). The organic layer was co-spotted with 12 on a TLC plate. This reaction was also performed at a larger scale (10 g) and was complete at the same reaction time with a very similar yield.

Work-up: Aqueous NH₄Cl (saturated solution, 30 mL) was slowly added to the reaction; effervescence was observed and ceased after addition was complete. The solution was monophasic and colorless, but upon addition of EtOAc (≈75 mL), a biphasic solution ensued. The organic phase was removed, and the aqueous phase was further extracted with EtOAc (3 × 50 mL). The aqueous phase was checked for residual products (TLC analysis). If aqueous dissolved product remained, it was extracted using EtOAc (2 × 30 mL). The combined organic phases were dried over Na₂SO₄, filtered, and rotary evaporated. Crude product weight: 6.50 g.

Purification: The crude product (6.50 g), with added EtOAc, was uniformly slurried with approximately two times its weight of silica gel (230–400 mesh). Under rotary evaporation (recommended bath temperature: <30 °C), the solvent was removed. The resulting free-flowing material was then added on top of a silica gel (230–400 mesh) column (16 cm in height, 4.5 cm in diameter). Mobile phase elution was performed with EtOAc/petroleum ether and began with a 1:9 ratio, which was gradually increased to a 2:3 ratio, at which point the product started eluting out. Cis-16 elutes out slightly before trans-16. The cis and trans product fractions were combined and concentrated to provide cis/trans-16 as a white solid (MW = 208.26 g/mol, 3.91 g, 18.77 mmol, 88% yield). A ¹H NMR was recorded (see Supplementary Materials), and a cis/trans ratio of 14.5:1 was recorded (see ¹H NMR expansion at 5.2 to 5.6 ppm). However, other resonance comparisons imply a cis/trans ratio of no better than >10:1. A small portion of the cis/trans-16 product was exposed to a second round of chromatography, wherein the cis-16 (major product) was carefully isolated free of trans-16. Note: We do not have definitive proof for the structure of trans-16 (minor product).

• All data and spectroscopy were collected for cis-16 (racemic).
• R_f = 0.40, EtOAc/petroleum ether (80:40).
• Melting Point: 113–114 °C.
• ¹H NMR (CDCl₃, 400 MHz) δ 7.07 (d, J = 8.2 Hz, 1H), 6.68 (d, J = 8.2 Hz, 1H), 5.25 (d, J = 6.6 Hz, 1H), 3.97 (dd, J = 10.6, 3.5 Hz, 1H), 3.84 (s, 3H), 3.81 (dd, J = 10.5, 2.6 Hz, 1H), 3.41 (dt, J = 8.8, 2.9 Hz, 1H), 2.56 (ddd, J = 14.4, 8.9, 6.9 Hz, 1H), 2.27 (s, 3H), 1.96 (d, J = 14.5 Hz, 1H).
• ¹³C NMR (CDCl₃, 100 MHz) δ 154.3, 143.4, 133.1, 131.5, 125.9, 109.3, 71.6, 63.4, 55.3, 45.4, 39.1, 17.8.
• IR (ATR, cm⁻¹) ν = 3308, 3239, 2926, 1595, 1495.
• HRMS (ESI-QTOF) m/z: [M + H⁺]⁺ Calculated for C₁₂H₁₅O₃: 207.101571; Found: 207.101821; Error: −1.2 ppm.

2.8. Synthesis of 1-Hydroxy-3-(hydroxymethyl)-7-methoxy-2,3-dihydro-1H-indene-4-carbaldehyde (cis-17)

In a single-neck round-bottom flask (100 mL) containing a stir bar (1 × 0.5 cm), was added potassium persulfate (≥99.0%, MW = 270.32 g/mol, 2.24 equiv, 5.37 mmol, 1.45 g), followed by copper(II) sulfate pentahydrate (≥98.0%, MW = 249.69 g/mol, 0.23 equiv, 0.55 mmol, 138 mg). Deionized water (0.2 M, 12.0 mL) was added to the flask, and despite being stirred, it did not fully dissolve. In another flask, cis/trans-16 (MW = 208.26 g/mol, 1.00 equiv, 2.40 mmol, 500 mg) was fully dissolved in acetonitrile (≥99.8%, 0.17 M, 14.5 mL), and this solution was added (over 1 min) to the aqueous K₂S₂O₈ medium. The reaction mixture was heterogeneous. Pyridine (≥99.0%, MW = 79.10 g/mol, 2.0 equiv, 4.80 mmol, d = 0.982 g/mL, 0.4 mL) was added, and the mixture turned blue and remained heterogeneous. A reflux condenser was added, and the mixture was stirred for 3 h at gentle reflux (75–80 °C) under an open atmosphere of air by not stoppering the reflux condenser. As the temperature reached 75–80 °C, the mixture became homogenous, and the color remained blue. After approximately 2 h, the color of the mixture turned green. TLC monitoring at 2 h showed persistence of the starting material (cis/trans-16), but at 3 h the homogenous solution turned yellow in color, and TLC analysis showed no more cis/trans-16. To monitor the reaction, an aliquot (≈25 μL) was taken and quenched by adding it to a vial (2 mL) containing an aqueous saturated solution of NaHCO₃ (≈1.0 mL) and EtOAc (≈0.1 mL). The EtOAc was sampled and co-spotted against cis/trans-16 on a TLC plate. This reaction was examined on a larger scale (1 g) on two occasions but resulted in a lower yield. During those reactions, the persistence of the starting material was noted, and extended reaction times did not overcome the problem and may have oxidized the desired aldehyde to the corresponding carboxylic acid.

Work-up: Aqueous NaHCO₃ (saturated solution, 10 mL) was slowly added to the reaction mixture. Effervescence was observed and ceased after complete addition. The reaction mixture was extracted with EtOAc (4 × 50 mL). The aqueous layer was checked for residual products using TLC. If aqueous dissolved product remained, it was extracted using EtOAc (2 × 50 mL). The combined organic phases were dried over Na₂SO₄, filtered, and rotary evaporated (recommended bath temperature: ≤25 °C). Crude product weight: 425 mg.

Purification: The crude product (425 mg), with added EtOAc, was uniformly slurried with approximately two times its weight of silica gel (230–400 mesh). Under rotary evaporation (recommended bath temperature: ≤25 °C), the solvent was removed. The resulting free-flowing material was added on top of a silica gel (230–400 mesh) column (16 cm in height, 2 cm in diameter). Mobile phase elution was performed with EtOAc/petroleum ether, beginning with a 10:90 ratio and gradually increasing to an 85:15 ratio, at which time the product eluted out. Concentration of the pure fractions provided cis/trans-17 (10:1 dr) as a white solid (MW = 222.24 g/mol, 283 mg, 1.27 mmol, 53% yield). A smaller portion of the isolated cis/trans-17 product was exposed to a second round of chromatography, wherein cis-17 (major diastereomer) eluted off the column before trans-17 and was isolated free of trans-17.

• All data and spectroscopy were collected for cis-17 (racemic).
• R_f = 0.10, EtOAc/petroleum ether (90:10)
• Melting Point: 151–152 °C.
• ¹H NMR (CDCl₃, 400 MHz) δ 9.95 (s, 1H), 7.80 (d, J = 8.3 Hz, 1H), 6.92 (d, J = 8.5 Hz, 1H), 5.21 (d, J = 7.1 Hz, 1H), 4.13 (dd, J = 10.5, 2.9 Hz, 1H), 4.00 (dt, J = 9.1, 3.1 Hz, 1H), 3.97 (s, 3H), 3.86 (dd, J = 10.3, 3.1 Hz, 1H), 3.79 (bs, 1H, -OH), 2.58 (ddd, J = 14.4, 9.0, 7.0 Hz, 1H), 2.27 (s, 1H, -OH), 2.03 (d, J = 14.3 Hz, 1H).
• ¹³C NMR (CDCl₃, 100 MHz) δ 191.4, 160.9, 147.8, 138.0, 135.4, 126.2, 109.5, 70.6, 64.4, 55.9, 45.3, 39.4.
• IR (ATR, cm⁻¹) ν = 3203, 2943, 2846, 2742, 1679, 1601, 1577, 1498.
• HRMS (ESI-TOF) m/z: [M − H⁺]⁻ Calculated for C₁₂H₁₄O₄: 221.068014; Found: 221.068863. Error: −3.8 ppm.

3. Discussion/Results

3.1. Initial Indanone Product Formation (7) and Possibilities Therefrom

In 1953, Baddeley reported the synthesis of 7 from the reaction of maleic anhydride and p-cresol in an AlCl₃/NaCl melt (180 °C) [11], but no subsequent report on this one-pot multi-step synthesis has since come forth. However, related, albeit stepwise longer, approaches to similar compounds have been described [8]. We began our synthesis of 7, guided by Baddeley, and, after some trial and error with this unorthodox reaction, established a robust procedure (Scheme 1). Baddeley reported no yield data, and our yield is low (40%), but nevertheless, rapid access to a highly functionalized building block (7) from inexpensive starting materials and reagents was achieved. We have fully characterized compound 7 (Supplementary Materials) and agree with Baddeley’s structural assignment.

Several chemoselective operations are potentially applicable to keto-acid 7, e.g., related compounds have undergone selective ketone reduction or hydrogenolysis [12]. Such operations would respectively lead to alcohol-acid 8 or carboxylic acid 9 (Scheme 1), but the higher-level chemoselective challenge of carboxylic acid reduction drew our attention. This was especially so because the conversion of indanone 4 to 5 (Figure 2) has never been reported using one reaction step. Instead, two-step [1] (p. 49) and four-step [5] (Scheme 4 therein) methods have been elaborated on. Although exceptions are known, borane can chemoselectively reduce carboxylic acids in the presence of ketones, and we examined BH₃٠DMS [12,13], BH₃٠DMS/B(OMe)₃ [14], and BH₃٠THF [15] for that purpose. In the event, BH₃٠DMS was the best reagent, but only provided a 36% yield of keto-alcohol 10 (Scheme 1). This compound was only analyzed by ¹H NMR analysis and not further pursued due to its low yield. We then examined the benzylic methyl group oxidation of 7 to aldehyde 11, but only obtained a 30% crude weight of a mixture of compounds, albeit with the requisite aldehydic resonance in observance (¹H NMR analysis). This was disappointing, but in the ensuing paragraphs, we report on the successful application of the same benzylic oxidation conditions to other compounds. At this juncture, we turned our attention to accessing keto-ester 12, and in doing so, we discovered a surprisingly diversifiable springboard from which five useful indanone and indanol building blocks could be synthesized. Keto-ester 12 was reliably obtained in 90% yield after exposing 7 to dimethylsulfate (Scheme 1) [16].

3.2. Synthesis of Aromatic Substituted Indanones

To access our first indanone analog of 5 (Figure 2), we pursued a report on the reversal of chemoselectivity for carbonyl reduction by Fujioka [17,18,19,20]. The concept is realized in a one-pot operation wherein in situ protection of the more reactive carbonyl-based functional group is followed by reduction (hydride or carbanion addition) of the less reactive carbonyl-based functional group. As stipulated by Fujioka, treatment of keto-ester 12 with Et₃P, followed by TMSOTf, forms an O,P-acetal phosphonium salt intermediate (not schematically shown) within 1 h, and its exposure to DIBAL provides the corresponding keto-alcohol 13 (Scheme 2). This was a valuable result but not practical due to the exceedingly high cost of Et₃P (purchased as a 1.0 M solution in THF). This led us to investigate the use of inexpensive, neat n-Bu₃P (≥93.5% chemical purity), which consistently resulted in the formation of keto-alcohol 13 with the same yield (70%) as for Et₃P. Although Fujioka did not previously explore the use of n-Bu₃P for his reversal of chemoselectivity methodology [17,19,20], he did use it in a related chemical study [18]. Attempts to arrive at the corresponding keto-aldehyde (not schematically shown), e.g., by lowering the temperature to −78 °C before DIBAL addition, only provided keto-alcohol 13 and with no advantage. Regrettably, the use of fewer equiv of DIBAL at low temperature, e.g., 1.1 equiv at −78 °C, was not investigated. The inability to directly access the keto-aldehyde from 12 was unfortunate but also consistent with Fujioka’s reports, wherein access to aldehyde products from ester DIBAL reductions was not reported on.

With keto-alcohol 13 secured, we next investigated the possibility of forming aldehyde 14 via benzylic methyl group oxidation. This was affected using a stoichiometric excess of potassium persulfate in combination with catalytic quantities of copper sulfate (Scheme 2). At times, pyridine is included [21,22] or excluded [23,24] during these types of oxidations, and we investigated its influence (0, 1.0, 2.0, or 3.0 equiv) and found 2.0 equiv to be optimal. Despite the mediocre yield of keto-aldehyde 14 (53%), it represents a higher-than-average yield outcome for this category of benzylic oxidations and provides a functional group capable of undergoing a wide variety of reactions. Attempts to optimize the K₂S₂O₈ oxidation yield wherein CuSO₄ wasis replaced with NiCl₂ [25] or a ferrocene/ferrous phthalocyanine/polymethylhydrosiloxane system [26], respectively, provided no or significantly lower yields of product 14. Finally, a hypervalent iodine reagent system (PhI(OAc)₂/TFA) [27] was investigated, but 14 was not observed (TLC analysis).

As a final diversification, we treated 14 with PCC and obtained dialdehyde 15 (Scheme 2), but with a low yield (<20%). We then examined the Dess–Martin periodinane (DMP) oxidizing reagent, and after careful optimization, we noted a greatly improved yield. Initially, we used DMP (3.0 equiv) in the commonly prescribed solvent of CH₂Cl₂ (0.1 M), which provided 15 albeit in 30–40% yield. However, chemical purity remained a challenge because all attempts to chromatograph the crude dialdehyde product (15), with or without a dopant (Et₃N and AcOH were individually examined), in an EtOAc/petroleum ether eluent system resulted in new impurities and <75% chemical purity. Using more concentrated solutions and/or fewer equivalents ofless DMP resulted in a lower yield. Given that 14 was not fully soluble in the reaction solvent (CH₂Cl₂), we pursued the use of an acetonitrile cosolvent; for a previous example, see [28]. With some trial and error, we found DMP (1.75 equiv) in a 0.05 M solution of CH₂Cl₂/acetonitrile (10:1) allowed full consumption of 14 within 5 h. Careful workup with aqueous saturated sodium thiosulfate provided dialdehyde 15 in high purity and 91% yield (see ¹H NMR, Supplementary Materials) without the need for further purification. This was an especially important finding after our earlier observation that 15 decomposes on exposure to silica gel.

Despite having identified the satisfying two-step sequence converting 13 to 15 (Scheme 2), we also examined the reverse oxidative sequence in which the primary alcohol of 13 was first oxidized, followed by benzylic methyl group oxidation (schematic not shown). This sequence provided 15 but with a lower overall yield, specifically because the benzylic methyl group oxidation was lower yielding (30%).

The use of dialdehydes is not common, but there is good precedent for the chemoselective α-functionalization of benzylic aldehydes, as found in 15, under mild reaction conditions and enantioselectively, e.g., via fluorination [29,30]. Alternatively, both aldehydes within 15 can be used to extend the existing aromatic ring to a substituted naphthyl ring under mild, one-pot reaction conditions [31,32]. Those reaction possibilities endow dialdehyde 15 with meaningful diversification potential.

3.3. Synthesis of Aromatic Substituted Indanols

To pursue the corresponding indanol-based compounds 6 (Figure 2), we exposed keto-ester 12 to LiBH₄ [33] and arrived at diol 16 (Scheme 3) in 88% yield (cis/trans > 10:1). Although challenging, cis-16 could be chromatographically isolated in pure form and was fully characterized. Subsequent oxidation of the benzylic methyl groups in cis/trans-16 provided a cis/trans mixture of aromatic aldehyde 17 in 53% yield (Scheme 3). Interestingly, this is the same yield that was noted when performing the benzylic methyl group oxidation of compound 13 to aldehyde 14 (Scheme 2). cis-17 was isolated in its pure form and fully characterized. Curious if dialdehyde 15 could be accessed via a double alcohol oxidation, we treated cis/trans-17 under the earlier noted DMP conditions, albeit now with 3.0 equiv of DMP. In the event, careful aqueous sodium thiosulfate work-up provided dialdehyde 15, but with unacceptable amounts of unreacted trans-17 in coexistence. Because chromatographic purification of 15 is not possible, we performed another DMP oxidation, albeit now with only the cis-17 diastereomer. After 1.5 h, the reaction was complete, and careful work-up with aqueous saturated thiosulfate washings permitted isolation of 15 in high purity (see ¹H NMR, Supplementary Materials) and a 90% yield. We also examined the reverse oxidative sequence, wherein 16 was first exposed to DMP and, in a second step, applied the benzylic oxidation (not schematically shown). In the event, the overall yield for that approach was unfortunately lower, specifically because the benzylic methyl group oxidation was particularly low yielding (25%).

4. Conclusions

Our focus was to identify efficient synthetic routes to new aromatic substituted indanones and indanols that could serve as medicinal chemistry building blocks, and to a large degree, we have met that goal and fully characterized seven compounds. In doing so, we have overcome several challenging chemoselective reactions, e.g., our Fujioka reduction was successfully implemented, and we used inexpensive nBu₃P instead of expensive Et₃P. A current limitation is the racemic nature of the synthesized compounds; however, resolution of similar compounds has been reported [33], and good precedent exists for converting racemic compounds very similar to 15 to enantioenriched compounds under mild conditions [29,30]. In addition, two routes to dialdehyde 15 have been elaborated on, with the second one (Scheme 3) allowing an experimentally simplified route to 15. Finally, further possibilities exist for expanding on the number and type of building blocks shown here, e.g., as shown in Scheme 1, and will be part of a future program to further diversify indanone and indanol-based compounds.