Synthesis of Naphthoxazinones in a One-Pot Two-Step Manner by the Application of Propylphosphonic Anhydride (T3P ® )

: A sequential one-pot two-step protocol has been elaborated for the synthesis of naphthoxazinones from 2-naphthol, methyl carbamate, and aromatic aldehydes. First, a three-component reaction was optimized with the dehydrating additive propylphosphonic anhydride (T3P ® ), resulting in 1-carbamatoalkyl 2-naphthols in good to excellent yields. Following the successful multicomponent approach, intramolecular acylation was performed at high temperature, again with the contribution of T3P ® , resulting in naphthoxazinone derivatives in moderate yields. These two steps were optimized together in one-pot as well, and the sequential rise in the requisite temperature eventuated the optimal procedure for the multistep cascade.


Introduction
Multicomponent reactions play a significant role in organic synthesis [1], medicinal chemistry [2], and materials science [3]. In this highly atom-efficient process, three or more starting materials react in a single operational step to form the product. Therefore, large libraries of organic molecules can be readily produced in this way. Hitherto, a vast amount of significant multicomponent reactions have been discovered from the Strecker α-aminonitrile synthesis to the Groebke-Blackburn-Bienaymé reaction.
Propylphosphonic anhydride (T3P ® ) was discovered and first used in peptide chemistry as a coupling agent by Wissmann and Kleiner [26]. Later, this effective water scavenger reagent was applied in several functional group transformations, rearrangements, carbon-carbon bond formations, the synthesis of various heterocycles, and organophosphorus compounds [27][28][29].
Furthermore, one should note that T3P ® has several advantageous properties including good solubility in organic solvents, low toxicity, and broad functional group tolerance. The forming byproducts are water-soluble and can be easily removed by alkaline extraction. Our research group has so far successfully applied the T3P ® -reagent in multicomponent and one-pot cascade reactions [30][31][32][33][34]. By continuing our work in this area, we herein report a T3P ® -mediated three-component Bettitype reaction of 2-naphthol, methyl carbamate, and aromatic aldehydes. Scheme 1. Schematic illustration of the Betti-type three-component reaction of 2-naphthol, methyl carbamate, and aldehydes.

General
Melting points were determined on a Büchi B-540 capillary melting point apparatus (Essen, Germany) or on a Jasco SRS OptiMelt melting point apparatus (Sunnyvale, CA, USA) and are uncorrected. 1 H NMR and 13 C NMR spectra were recorded at 303 K on a Bruker Avance III HD (600 and 150 MHz for 1 H and 13 C NMR spectra, respectively, a Bruker Avance III (400 and 100 MHz for 1 H and 13 C NMR spectra, respectively) (Billerica, MA, USA), or a Varian Unity Inova 300 MHz spectrometer (300 and 75 MHz for 1 H and 13 C NMR spectra, respectively) (Oxford, UK). DMSO-d6 and/or CDCl3 was used as the solvent and tetramethylsilane (TMS) as the internal standard. Chemical shifts (δ) and coupling constants (J) are given in ppm and in Hz, respectively. Mass spectra were recorded on a Bruker O-TOF MAXIS Impact mass spectrometer coupled to a Dionex Ultimate 3000 RS HPLC system (Thermo Fischer Scientific, Waltham, MA, USA) with a diode array detector. The reactions were followed by analytical thin-layer chromatography on silica gel 60 F254 and HPLC-MS chromatography on a Shimadzu LC-MS-2020 apparatus. Purifications by flash chromatography were carried out using Merck 107736 silica gel 60 H (Darmstadt, Germany) using a hexane-ethyl acetate, hexane-dichloromethane, or dichloromethane-methanol solvent system. All reagents were purchased from commercial sources (Merck, Darmstadt, Germany, Fluorochem, Hadfield, UK, Combi-Blocks, San Diego, CA, USA). Analytical samples of new compounds were obtained by recrystallization from the solvents or solvent mixtures given below in parentheses. The 1 H and 13 C NMR spectra of the compounds can be found in the Supplementary Material.

General
Procedure for the Synthesis of naphta [1,2]oxazinones (5a-f) We have investigated six reaction pathways (A-F) to the one-pot ring-closing synthesis of naphtha[1,2-e]oxazinone derivatives. It was found that the most effective pathway was E1, E2 under microwave (MW) conditions. In regard to the optimization results of the one-pot synthesis, we used pathway E1, E2 for the synthesis of naphta[1,2-e]oxazinones.

Results and Discussion
The T3P ® -mediated model reaction of 2-naphthol (1), methyl carbamate (2), and 3fluorobenzaldehyde (3) (1 equiv. of each) was optimized by varying different experimental parameters (Scheme 2, Table 1). Initially, the effect of the reaction time and the molar excess of the components were investigated. Based on preliminary experiments, the reaction was performed in ethyl acetate (EtOAc) at room temperature for 26 h using 1 equiv. of T3P ® ( Table 1, entry 1), resulting in the desired racemic product (4a) in 32% yield. The progress of the reactions was followed by TLC and HPLC-MS. Next, the temperature was raised to 77 °C. It was found that 2 h of reaction time in EtOAc under reflux is sufficient for full conversion, and the yield increased to 53% (Table 1, entry 2). In the absence of T3P ® , the product (4a) was not detected (Table 1, entry 3). At the next stage of the optimization, the amount of the reactants was screened. Adding methyl carbamate (2), 3fluorobenzaldehyde (3), and T3P ® in 1.1 equiv. or 1.3 equiv. excess to 2-naphthol (Table 1, entries 4,5), the yield increased to 83% for both reactions that might be explained with the mode of action of T3P ® , i.e., sometimes, it can react not only in the cyclic, but also in the open-chain form after the addition of one water molecule multiplying the activity [33]. In the case of applying 1.5 equiv. and 5.0 equiv. T3P ® , a slight decrease in the yields was observed with the higher amount of the additive (81% and 73%, respectively) ( Table 1, entries 6,7) possibly due to degradation. The last reaction was refluxed for 5 h, suggesting that the larger amount of the dehydrating agent and the longer reaction time do not influence the reaction further. After the amount of the components was optimized, the proper solvent of the reaction was screened for. The products were obtained in high yields (73-91%, With the optimized reaction conditions in hand, the scope and limitations of the T3P ® -assisted synthesis of 1-carbamatoalkyl 2-naphthols (4) were examined. The reaction of 2-naphthol (1) and methyl carbamate (2) with various aromatic aldehydes (3a-i) was studied in toluene at 80 °C for 30 min (Scheme 3). The products 4a-i were obtained in good to high yields (60-90%). Aromatic aldehydes containing electron-withdrawing groups (EWG, e.g., chlorine and/or bromine) were compatible with the reaction, providing 4a, 4c, and 4f in 71-89% yields. However, the phenoxy derivatives (4b and 4e) were isolated in different, 60% and 90% yields, respectively. The position of the electron-donating phenoxy group seemed to be limiting the efficacy; particularly, in the para position, it strongly hindered the reaction (4b, 60% yield) possibly by slowing the formation and weakening the reactivity of the electrophilic imine intermediate. Heteroaromatic aldehydes (3g-i) provided good yields (4g-i, 66-77%). The application of benzaldehyde (3d) gave product 4d in 81% yield after 30 min. When the reaction mixture forming 4d was heated under microwave (MW) conditions, the corresponding product (4d) was obtained in 76% yield after 15 min, suggesting that the MW has some rate-enhancing effect for the reaction. To the best of our knowledge, naphthol derivatives 4a,c,d,e,f,h are new compounds with no precedence in the scientific literature. In some cases, more by-products could be detected by HPLC-MS. The side products were isolated, and the structures were elucidated by NMR spectroscopy. It was found that the side product is a naphta [1,2]oxazinone derivative resulting from the intramolecular ring closure of compounds 4. Based on this finding, the reaction conditions for the cyclization were optimized starting from 4d (Scheme 4, Table 2). Executing the reaction at lower temperatures did not result in the desired product even after longer reaction times ( Table 2, entries 1-4). After raising the excess of T3P ® to 2 equiv. and elevating the temperature to 160 °C, the naphthoxazinone could be isolated in 75% yield after 1 h (Table 2, entry 5). Using a lower quantity of T3P ® did not lead to full conversion ( Table 2, entry 6), while in a larger excess, the formation of unidentified decomposition products was observed with a significant decrease in the yield (Table 2, entry 7). Heating the reaction mixture under microwave (MW) conditions caused decomposition similarly, and 5d could be isolated only in low yields ( Table  2, entries 8,9).  After the optimization of the two separate reactions, we aimed to develop a one-pot protocol for the synthesis of naphta[1,2-e]oxazinone derivatives. It should be noted that the two separate steps with double work-up and purification resulted in 81% × 75% = 61% (Δ) and 75% × 75% = 56% (MW) total yields for the model compound 5d. Thus, we challenged to reach at least similar efficacy. Three different synthesis pathways were elaborated to find the efficient method, performed with conventional heating in a sealed tube (A-C) and under MW conditions (D-F), as well (Scheme 5, Table 3). First, we have challenged the one-pot procedure by applying 2-naphthol (1), methyl carbamate (2) (1.1 equiv.), benzaldehyde (3d) (1.1 equiv.), and T3P ® (3.1 equiv.) and heating the reaction at 160 °C for 90 min in toluene (Scheme 5 A,D). The conventional heating (A) resulted in a yield of 26%, while on the contrary, in the MW procedure, (D) 5d was obtained in 38%. This result led us to the conclusion that the two reaction steps should be treated differently, either by increasing the reaction temperature sequentially (B and E), or together with that by adding the T3P ® reagent in two portions (C and F). In pathway B and E, all the necessary T3P ® was added at the beginning of the reaction, and the reaction times and temperatures were taken from the separately optimized steps. First, the temperature was held at 80 °C for the first step (30 min at B1 and 15 min at E1), and then raised to 160 °C for the ring closure (90 min at B2 and 35 min at E2). Naphthoxazinone 5d was isolated in 28% in the conventional way (B), while in this case, the MW conditions (E) led to 52% yield. The most complicated pathways C and F resulted in moderate to good yields. If the T3P ® was added in two portions (1.1 equiv. at 80 °C and 2.0 equiv. at 160 °C), the expected product 5d was formed in 51% yield under conventional heating (C) and in 45% yield under MW conditions (F). It should be noted that there was no significant remaining starting material in any cases, but different amounts of small degradation products were detected in the HPLC-MS. In summary, pathway E was found to be the most efficient for the one-pot synthesis of 5d from 2-naphthol (1), methyl carbamate (2), and benzaldehyde (3d). Although the total yield is lower than those for the two separate reactions, the total work time and costs for the reactions have been reduced, as only one work-up and purification procedure is required.  After finding the suitable method for the T3P ® -assisted one-pot synthesis of naphthoxazinones, the scope and limitations of the reaction were examined with aldehydes 3a-i (Scheme 6). Scheme 6. Scope and limitations for the ring-closing reaction with T3P ® for the synthesis of naphta[1,2-e]oxazinone derivatives.
The reactions with aldehydes 3a,c,e,f went smoothly, resulting in moderate yields (41-53%) similarly to benzaldehyde 3a. Aldehyde 3b equipped with the phenoxy group in the para position gave product 5b in a low yield (23%), supposedly due to the weak efficacy observed earlier for the first step. Notably, in the case of the heterocyclic aldehydes, we were able to isolate a side product (6g-i) in very poor yield and observe unidentified decomposition products. The desired 5g-i could only be detected by HPLC-MS. The side product has been identified as the adduct of T3P ® to the Betti-product (Figure 1).

Figure 1.
The identified side products in the one-pot procedure derived from 4g-i.
To the best of our knowledge, the four synthesized oxazinones 5b,c,e,f have not been precedented in the scientific literature yet. The new derivatives have been characterized by 1 H and 13 C-NMR, HRMS, and melting point (see Supplementary Material).
A plausible mechanism for the formation of carbamates 4 via the Betti reaction in the presence of T3P ® and the T3P ® -mediated synthesis of naphtha[1,2-e]oxazinones (5) is outlined in Scheme 7 A and B, respectively. The first step is the T3P ® -promoted condensation of aldehyde 3 and methyl carbamate (2), forming an imine-type intermediate (7) along with 1,5-dihydroxy-1,3,5tripropyltriphosphoxane 1,3,5-trioxide (T3P ® + H2O abbreviated as QOH, 8). In the next step, the nitrogen atom of the C=N bond is protonated, and the iminium ion (9) reacts with naphthalen-2-olate (10) in a nucleophilic addition, forming 2-oxo-1,2-dihydronaphtalen-1-yl intermediate 11. Finally, the aromatic system is stabilized again by a proton exchange leading to 4. The cyclization-forming heterocycle 5 presumably begins with the deprotonation of the 1-carbamatoalkyl naphthol derivative 4. This is followed by the nucleophilic addition of the anion to the carbonyl group and the subsequent MeOH elimination, resulting in the tricyclic product 5. From the proposed intermediates, imine 7 and Betti-product 4 could be detected by HPLC-MS.

Conclusions
A three-component reaction has been optimized between 2-naphthol, methyl carbamate, and aromatic aldehydes in the presence of the dehydrating additive T3P ® . The formed 1-carbamatoalkyl 2-naphthols have been transformed, further leading to naphthoxazinones. The successful multicomponent reaction and the subsequent ring closure have been optimized in a one-pot manner, resulting in a three-component sequential multistep cascade.