Introduction

Functional group interconversion is a beneficial synthetic transformation for everyday organic synthesis1. The development of environmentally benign and efficient synthetic methods for functional group interconversion is a central goal of current chemistry research. Phenolic compounds, the most abundant secondary metabolites in plants, are found ubiquitously in nature2. Phenolic compounds possess a standard chemical structure comprising an aromatic ring with one or more hydroxyl substituents3. Anilines are also cheap and abundant chemicals4. Carbonylation reaction represents an essential synthetic transformation for converting various cheap and readily available chemicals into a diverse set of valuable products in our daily lives5. Traditionally, carbon monoxide was used as a C1 building block under various transition metal-based catalysts6. Among different carbonylation methods developed7, Pd catalyzed carbonylation reactions have been widely applied to synthesize aldehyde, ester, amide, and other carbonyl derivatives in good yield8. In addition to aryl halides the aryl triflates and aryl diazonium salts are well-known coupling partners for oxidative addition with Pd-catalyst (Fig. 1d)9. Next CO-coordination, 1,2-migratory insertion followed by subsequent reductive elimination in the presence of various nucleophilic partners provided the corresponding carbonyl compounds. Despite its great success, reactions with carbon monoxide are less common in more complex organic syntheses due to the toxicity and risk associated with high-pressure CO cylinders10. Also, there are issues related to pharmaceuticals with permitted daily exposure to transition metals11. Hence, some alternative methods also have been developed using photoredox and Ni-chemistry but are constrained by limited scope12. Considering that a transition metal-free and CO-free carbonylation of phenols and aniline would be highly useful, but to date no report is available for the transition metal-free deoxygenetive/denitrogenetive conversion of phenol/anilines to aldehydes and ketones hence becoming a topic of prime interest.

Fig. 1: Synthesis of aldehydes, ketones from phenols and anilines and its diverse applications.
figure 1

a Use of phenols in organic synthesis. b Importance of deoxygenartive and deaminative transformations. c Bioactive molecules synthesized from aldehydes and ketones. d Aldehydes and ketones synthesis using phenols/anilines. e Our hypothesis. f Our work. g Regiospecific synthesis of acetophenone. h Short-step synthesis of bioactive molecule.

Full size image

Organoboronates have been identified as synthetic anchors owing to their versatile reactivity13. Recently, geminal polyboronates have emerged as a distinct class of organoborons for versatile C − C or C−heteroatom bond formations. The easy generation of α-bis(boryl)organometallic species from the corresponding geminal-bis-boron compounds allows diverse synthetic transformations14. The presence of empty p-orbital on boron stabilizes the adjacent negative charge. The α-bis(boryl)carbanion species reacts with aldehydes and ketones to yield alkenyl boronates15. Matteson established the addition of α-bis(boryl)carbanions to acid chloride, which resulted formation of ketones16. Further, this chemistry has been extended to alifatic acids and esters, resulting in alifatic ketones but till to date, it has been never applied for the conversion of phenols/anilines to aldehydes and ketones17,18,19,20,21. Therefore, we hypothesized that the α-bis(boryl)carbanions can react with in-situ generated quinketal/imine-ketal for the generation of vinyl boronic esters, and further in-situ oxidation of boronic esters will generate the corresponding aldehydes and ketones. Herein, we developed a methodology for the conversion of both the phenols and anilines to corresponding aldehydes and ketones in a one-pot manner. The reaction was carried out in a one-pot manner encompassing Csp2-O/N bond cleavage (Fig. 1f). This protocol provides access to the aldehydes and a wide variety of ketones such as acetophenones, aryl-alkyl, aryl-aryl, and conjugated ketones in a quantitative yield directly from easily available phenols and anilines. Further, we have demonstrated the utility of this strategy for the regiospecific synthesis of acetophenones (Fig. 1g), in the short synthesis of bioactive compounds (Fig. 1h) and the direct conversion of bioactive phenol and anilines to the corresponding ketones.

Results and discussion

With this hypothesis in hand, we have initiated our studies using phenol and homobenzyl geminal boronic ester. To establish the reactivity, we initially decided to isolate the quinketal after the oxidation of phenol using PIDA (phenyliodine(III) diacetate) and then reacted with lithiated geminal-(Bpin) (Supplementary Fig. 1). Next after successive oxidation using H2O2/NaOH the desired ketone was generated in moderate yield, which justified our hypothesis. We realized that there would be several challenges for conducting this reaction in one pot. We have three major concerns: (1) getting pure quinketal by removing the by-product AcOH coming from PIDA, (2) finding a suitable condition for the 1,2-addition step, and (3) oxidation of intermediate vinyl boronate avoiding competing Dakin oxidation. We initiated our studies using NaOMe as the AcOH quencher which although furnish NaOAc as the byproduct but should not hamper the next step. We have conducted several optimizations to improve the yield (Table 1). We observed that the yield of the 1,2-addition step was highly dependent on the reaction temperature (Table 1, entries 1–3). Further to improve the yield, we screened different bases, in which K2CO3 was found to be optimal (Table 1, entries 4). Further increase of geminal B(pin) equivalent improved the yield of corresponding ketone up to 80% in a one-pot manner with the formation of phenol from corresponding Dakin oxidation22 (Table 1, entry 10). A further variation of oxidation conditions from H2O2 to NaBO3 improved the yield with exclusive formation of the desired product (Table 1, entry 13).

Table 1 Optimization of reaction
Full size table

With the optimized conditions in hand, we next sought to determine the generality of the phenol group in this ketone-forming reaction (Fig. 2). First, we have explored the synthesis of various substituted benzaldehydes from the corresponding phenols. Examination of the scope of various phenols revealed that the reaction worked efficiently with various substituents present at the ortho, para, and meta-positions (Fig. 2a–s). Most importantly, this methodology allowed us to introduce a methoxy group at the para-position. Starting from monosubstituted phenols, we ended up isolating di-substituted benzaldehydes. Even, by using 2-bromo or 3-chloro phenols, we can isolate 4-methoxy-2-bromo 2 g or 4-methoxy-3-chloro benzaldehydes 2 h in great yield. We are surprised to see that phenol-containing free alcohol is equally reactive, affording the desired product in great yield. Next, we have efficiently converted the phenols to corresponding ketones by using substituted geminal boronic esters (Figs. 2, 3a–q). Further, the selective carbonylation for the phenols having free hydroxyl group, ketal, ester, and terminal alkynes is rewarding (Figs. 2, 3o, p, q). Besides, substituted phenols, several fused aryls also have been efficiently acylated with complete regiospecificity. Variation in the germinal B(pin) introduced several branch, acyclic, cyclic, and cyclopropylic ketone (Figs. 2, 4h) in good yields. Interestingly, we have also synthesized the diaryl ketone (4 g) by this single method. General methods such as Friedel–Crafts acylation, and Fries rearrangement for the synthesis of acetophenone is always a challenging task as it leads to non-separable regioisomeric products23. In-general methods require several steps or transition metal-mediated cross-coupling. Herein, we reported a one-pot method for the regioselective synthesis of acetophenones using methyl geminal B(pin) as the acyl unit (Figs. 2, 5a–e; and see below). Next, we envisioned that the engagement of homoallyl germinal B(pin) would furnish the allyl ketone which might undergo olefine isomerization and will result in α,β-unsaturated ketones. Gratifyingly, we have synthesized a variety of α,β-unsaturated ketone with 100% trans selectivity in good yield (Figs. 2, 6a–f). Anilines are also very cheap and abundant chemicals. We initiated our studies by converting protected aniline to the corresponding imine-ketal24 core following the reaction with α-bis(boryl)carbanions, but no desired product formation occurs. Further optimization revealed that the isolation in quinketal25,26 stage is necessary for good yield. A good number of anilines were converted to the corresponding aldehydes and ketones in good yields (Figs. 3, 7a–o).

Fig. 2: Scope for the aldehydes and ketones directly from phenols.
figure 2

a Optimized reaction condition. b Scope for Aldehydes. c Scope for Ketones (R = -CH2CH2Ph). d Ketone scope with B(pin) Variation. e Scope for the acetophenones. f Scope for the α,β- unsaturated ketones. All the reported yields are isolated yields and have been calculated from phenols. The yields with * marks have been calculated from corresponding quinketal.

Full size image
Fig. 3: Scope for the aldehydes and ketones starting from anilines.
figure 3

a Optimized reaction condition. b Scope from anilines. All the yields have been calculated from corresponding quinketals.

Full size image
Fig. 4: Bioactive molecules functionalization.
figure 4

a. Optimized reaction condition. b Scope for the bioactive aldehydes and ketones synthesis. All the reported yields are isolated yields and have been calculated from phenols. The yields with * marks have been calculated from corresponding quinketal.

Full size image
Fig. 5: Short step synthesis, regioselective synthesis of acetophenone and application to bioactive molecule synthesis.
figure 5

a Previous method for 5 f synthesis. b Our method for 5 f synthesis. c Use of 5 f for bioactive molecule synthesis. d Previous method and our method for 5 h synthesis. e Previous method and our method for 5j synthesis. f. Previous method and our method for 8b and 8d synthesis. g Previous method and our method for 10 g synthesis.

Full size image
Fig. 6: BINOL based photocatalysis; Synthesis, photophysical study and application of 11a as the photocatalyst under sunlight.
figure 6

a Optimized reaction condition. b Photophysical images and SC-XRD of 11a. c Numerical data of HOMO-LUMO of 11a and 11b. d Photophysical data of 11a and 11b. e Reactions using 11a as the photocatalyst. Here Fl means Fluorescence and Ph means Phosphorescence.

Full size image
Fig. 7: Probable mechanism, DFT calculation and 11B NMR study.
figure 7

a Proposed mechanism. b NBO calculation. c 11B NMR values. d Energy difference diagram. e Optimized intermediate structures IM-4. f Optimized intermediate structures of IM-4. g 11B NMR.

Full size image

Acyl is fairly common in bioactive structures. The group being planar, it can interact with a binding site as an H-bond acceptor through the two lone pairs of electrons on the carbonyl oxygen or via dipole moment. Besides, the acylation of the bioactive phenolic core also leads to APIs. Over the past decade, molecular synthesis has gained extensive impulse by the late-stage functionalization (LSF) methodologies. This method allowed us to build a technique in the diversification of pharmaceuticals towards medicinal chemists and in drug discovery programs via the controlled installation of small groups at diverse locations. Here we have subjected several bioactive molecules for LSF, which resulted in the desired product with excellent yield as well as high chemoselectivity (Fig. 4). This method was also found to be compatible with ester, olefin, etc, and stereoconvergent in nature. Let’s have a look, how efficient out strategy is compared to the reported method. We have elaborated a few examples which demonstrated the applicability of our method compared to the multi-step synthesis of the same material using traditional synthesis. The synthesis of 5 f from meta-cresol requires two or three steps which even resulted in o-,m- acylated product in the ratio of 3.2:2.1 regioisomeric mixture (Fig. 5a)23. However by our method, the regiospecific synthesis of 5 f has been achieved (Fig. 5b) and further utilized for the synthesis of CRAC inhibitor 10 c23,27,28,29. Next we resolved the regiospecific synthesis of 5 h while the reported procedures require three steps resulting with 1:2.2 regioisomeric mixtures (Fig. 5d)23. Compared to the general method, the synthesis of 5j (Fig. 5e) has been demonstrated in one pot manner and further engaged for the synthesis of bioactive compound 10d30,31 respectively. The antimalarial32 compound 10e and fungicidal33 10 f which are known to synthesize from the aldehydes 8d and 8b have been synthesized in one pot manner. Furthermore, the synthon 2 g for the synthesis of 10g34 is reported to be accessed by three steps, while we have reduced it to two steps (Fig. 5g).

While working with the binol system, we hypothesized that a similar reactivity could be explored toward BINOL, and in that case, we might end up with some sort of polycyclic hydrocarbon with the scaffold (Supplementary Fig. 2 and 4). Gratifyingly, we were surprised to see the formation of compounds 11a and 11b while using different germinal B(pin), which is a single step to access functionalizable ODA-analogous35 and we did not find any kind of photocatalyst having this core. By SEM study we found that both the compounds have a niddle-like planer shape indicating higher surface area compared to bulk counterparts, which provide higher active sites for catalytic reactions36. Their single crystalline nature eliminates the possibility of recombination at the grain boundary related defects zone, exhibiting the feasibility for efficient charge transport37,38 in solar cell applications (Supplementary Fig. 20). Further from the photophysical study, both of them exhibit fluorescence and phosphorescence emission in visible light. The HOMO-LUMO energy has been computed both by optically and by DFT (Fig. 6c). Cyclic voltammetry (CV) experiments demonstrate that 11a and 11b have an oxidation potential ∼ 0.9 V and ∼ 0.8 V (vs. SCE), respectively, suggesting the possibility of an efficient SET reduction of several organic substrates (Fig. 6d). The photocatalytic activity39 was elucidated for 11a under sunlight irradiation (Fig. 6e) which resulted in dehalogenetive arylation as well as decarbobylative and deborylative vinylation with 100% trans selectivity40,41. Besides their utility as an organic semiconductor and, organic field-effect transistors (OFETs), are under process42,43.

The mechanistic details of our proposed transformation are outlined in Fig. 7. We have done experimental and theoretical studies to establish the mechanism. Based on the HSAB44 concept, organolithium being a hard nucleophile, it only undergoes 1,2-addition. But there are some reports available on 1,4-addition using stabilized organlithium45. In the case of IM-2, the negative charge being stabilized by the boryl group, there might be a possibility for 1,4-addition. But we are getting the1,2-addition product exclusively. The NBO analysis of IM-1 revealed that C-1 is more positively charged (+0.480) compared to C-3 (−0.126), showing that C-1 is a more favored site for the nucleophilic attack (Supplementary Fig. 23). Further support came from the DFT calculation suggesting the formation of adduct IM-4 is exergonic and hence more favored compared to IM-3 adduct, which demands an additional 0.74 kcal/mol energy (Supplementary Fig. 24). We have also taken the 11B-NMR, which supports the probable mechanism by observing three different peaks (10.85, 33.04, 36.22) from the crude reaction mixture (Supplementary Fig. 25)46,47. The presence of IM-7 was further confirmed by HRMS data and 11B-NMR data. We are unable to isolate IM-7 for further characterization.

In conclusion, we have developed a transition metal-free pioneering strategy for the synthesis of a diverse range of aldehydes and ketones, starting from phenols in a one-pot manner and from protected anilines after the isolation of quinketal via C-O and C-N bond cleavage. A wide range of mono aryl, fused biaryl as well as substituted phenols and anilines have been successfully acylated with good to excellent yield. A variety of ketones, such as aryl-alkyl, aryl-aryl, acetophenones, and conjugated ketones, have been synthesized by this single protocol. This method has an excellent selectivity towards the keto and 1,2-addition over the ester and 1,4-addition. The regioselective synthesis of acetophenones has been developed with 100% selectivity. Several bioactive molecules and pharmaceutically active intermediates have been synthesized within short steps from the reported procedure. Further, we have designed a fused BINOL-based polycyclic compounds having a tuneable end and further exploited as a photocatalyst under sunlight irradiation.

Methods

To a flame-dried reaction tube the corresponding phenol (1 equiv, 0.5 mmol), was taken and dry MeOH (1 mL/0.5 mmol of phenol) was added. The reaction mixture was cooled to 0 oC and slowly PIDA (1.1 equiv) was added to it. Then the reaction mixture was allowed to stir at rt for 15 min. The reaction mixture was again cooled to 0 oC and to that 2.2 equivalent solid dry potassium carbonate was added. The reaction mixture was allowed to stir for 5 min at the same temperature. Then the MeOH was removed in a vacuum and it was azeotropically dried by dry toluene. Next dry THF was added (2 mL) and cooled to 0 oC. Further, 1.5 equiv of pre-prepared lithiated germinal B(pin) (pre-cooled at 0 oC) was added to it. The reaction mixture was allowed to stir for 5 min at the same temperature. Next 3 equiv of solid NaBO3.4H2O followed by 2 mL of water were added to it. The reaction mixture was allowed to stir at rt for 5 min. Next EA was added and the aqueous layer was extracted with EtOAc (3×10 mL). Then the solvent was removed under reduced pressure. The resultant crude product was purified by flash chromatography on silica using 10% to 15% EtOAc/Hexane.