One-Pot Difunctionalization of Aryldiazonium Salts for Synthesis of para-Azophenols

A novel difunctionalization of aryldiazonium salts was realized for the one-step generation of symmetric and asymmetric p-azophenols. This approach is proceeded by the sequentially regioselective aromatic C-O and C-N bond construction under mild reaction conditions, unlocking a new reaction strategy to facilitate the synthesis of p-azophenols.


INTRODUCTION
Azoaromatic compounds are significant core motifs and found the prevalence in a variety of applications due to their unique properties of the azo unit (Pang et al., 2020). They are universally applied in synthetic chemistry (Nguyen et al., 2018), sensing probes (Dong et al., 2015), organic dyes (Bafana et al., 2011), smart materials (Mohamed et al., 2015;Gerkman et al., 2020), and therapeutic agents (Lizard et al., 2020;Babamale et al., 2021). Considering the widespread utility of aromatic azo scaffolds, a handful of methods for their formation were established by adding an external oxidant or noble metal catalyst (Grirrane et al., 2008;Dai et al., 2018;Han et al., 2021;Sitter and Vannucci, 2021). Therefore, the synthesis route to achieve azobenzene derivatives in an efficient and handy manner continues to be keenly pursued.
Aryldiazonium salts (Griess, 1858), abundant and inexpensive building blocks, have stimulated a series of remarkable name reactions as useful and efficient synthons over the next hundred years (Sandmeyer, 1884;Hodgson, 1947;Gomberg and Bachmann, 1924;Balzand Schiemann, 1927;Meerwein et al., 1939;Kikukawa and Matsuda, 1977;Kikukawa et al., 1979). Even to this day, the aryldiazonium compounds still receive much attention and constantly result in plenty of new functionalization conversions. For example, the diazo-oriented substitutions of aryldiazonium salts with various coupling partners via the denitrogenation step have been successfully explored ( Figure 1A) (Mo et al., 2018;Felpin and Sengupta, 2019;Yuan et al., 2021). Aryldiazonium salts by the nucleophilic addition-type reaction have also been extensively used for inserting the azo group into the target molecules ( Figure 1A) Liu Z et al., 2017;Abrams et al., 2018;Liu et al., 2018). Recently, cyclization by retaining the privileged N 2 fragment has been a thriving pattern in heterocyclic chemistry ( Figure 1B) (Mo et al., 2013;Zhang et al., 2020). Apart from the familiar aryl or/and aryl azo precursors described above, the aryl-based difunctionalization pattern of aryldiazonium salts has seriously remained in the shadow of underdevelopment. Herein, we exposed a first example utilizing this aryl-based difunctionalization tactics of aryldiazonium salts with H 2 O to directly assemble versatile p-azophenols in good yields under mild reaction conditions ( Figure 1C).

RESULTS AND DISCUSSION
To optimize our aryl difunctionalization scenario, we first employed phenyldiazonium salt 1a as the standard substrate to evaluate reaction conditions with the high-yielding product 4-(phenyldiazenyl)phenol 2a in mind. When MeOH was conducted as a solvent in the difunctionalization reaction of the phenyldiazonium salt 1a, an unprecedented preference was smoothly delivered with the yielding product 2a in 46% yield ( Table 1, entry 1). The absolute structure of 2a was also confirmed by X-ray diffraction analysis (CCDC No. 2070994, see details in SI). In light of this initial experimental result, we further investigated other solvents for this aryl-based cascade reaction, while keeping the moderate yield permanent ( Table 1, entries 2-5). With the consideration of a hydroxyl source in the target product, we next screened the amount of H 2 O in the reaction ( Table 1, entries 6-8). The optimal ratio of MeOH/H 2 O was found to be 3:1 with a superior increase in the yield to 89%. Then, other bases were probed in the reaction conditions and gave somewhat unsatisfied yields of the desired product (Supplementary Table S1, entries 1-8 in SI). Similarly, adjusting the amount of the base or reaction temperature was proved to impede the desired conversion (Supplementary Table  S1, entries 9-11 in SI).
Having the optimized reaction conditions in hand, we began examining the reactivity of various aryldiazonium salts to give the corresponding p-azophenol derivatives (Scheme 1).
Aryldiazonium salts with ortho-and meta-substituents including electron-donating or electron-withdrawing groups (1b-1g) smoothly reacted with H 2 O to produce the symmetrical products (2b-2g) in good yields. In addition, poly-substituents with different positions on the aryldiazonium salts were fully compatible in the standard conditions, delivering azo-containing phenolic compounds (2h-2l) in 73-82% yields.
Regrettably, when the substrate bearing the meta-trifluoromethyl group was performed in the standard tandem reaction, no desired difunctionalized product 2m was observed. A cross reaction between two different diazonium salts by using this efficient difunctionalization procedure is undoubtedly an attractive extension (Scheme 2). Just as we speculated, the reaction with the phenyldiazonium salt 1a and 2-chlorobenzene diazonium salt 1dwas viable in this cascade conversion and provided the unsymmetrical p-azophenol 3a in 86% yield. Then, we further examined other cross-coupling partners with the 2chlorobenzene diazo salt as one model reactant. To our delight, aryldiazonium salts attached to the ortho-or meta-group were smoothly subjected into the reaction, yielding the aimed unsymmetrical products (3b-3e) in 82-87% yields. Similarly, the substrate with multiple substituents at different positions was also found to successfully deliver the target products (3f-3g). The X-ray structure of products 3f and 3g (CCDC No. 2070995; 2070996, see details in SI) also validated the asymmetrical structure of the assumed product. Note that the 3-chlorobenzene diazonium salt by combining with electron-neutral or electron-donating ones was well suitable in the reaction, producing the unsymmetrical p-azophenol (3i-3l) in 63-80% yields.
To improve the scalability and practicality of this difunctionalization approach, large-scale synthesis was applied with the substrate 1b in the standard one-pot reaction, leading to the p-azophenol derivative 2b with a satisfied yield (Scheme 3). Regarding the possibilities for further decoration of the hydroxyl group, this difunctionalized product 2b was readily introduced   a Conditions: 1a (0.5 mmol, 1.0 eq.), Base (0.5 mmol, 1 eq.), in 2 ml solvent at 0°C-r.t. for 3 h; yields of isolated products.
To obtain the information of this difunctionalization pathway, free radical trap experiments (Liu et al., 2019;Liu et al., 2020) with using 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO) or 2,6-di-tert-butyl-4-methylphenol (BHT) did not give the target product 2a or the corresponding radical coupling byproduct with the assistance of HRMS (Supplementary Scheme S1A in SI). Those inhibited results might ascribe to the acidity of BHT (Huang and SCHEME 1 | Synthesis of symmetrical p-azophenol. a Conditions: 1 (0.5 mmol), NaOAc (0.5 mmol, 1 eq.), in 2 ml of the mixed solvent (V MeOH /V H2O = 3:1) at 0°C-r.t. for 3 h; yields of isolated products. b Without 1.0 eq. NaOAc. Lumb, 2019) or oxidability of TEMPO (Ryan et al., 2019) influencing on the stability of diazonium salts in the standard reaction. So an ionic process was still heavily weighted in the optimal process. To examine the sequence of hydroxylation and azo-modification, a hypothesized PhOH intermediate led to the para-functionalization product 2b' rather than the target product 2b (Supplementary Scheme S1B in SI). Diphenylazo 5a (Okumura et al., 2013) replacing the diazonium salt was used as the reactant whereas the desired formation of 2a was completely shut down (Supplementary Scheme S1C in SI). Those observations were more likely to highlight that the diazo-localized hydroxylation of the aryldiazonium salt preceded over the para-selective azo reaction in the standard reaction. In order to further help understand the reaction mechanism, the charge densities of the carbon adjacent to the azo group in the aryldiazonium cation were calculated by the natural population analysis (NPA) charge analysis (see Supplementary Scheme S2 in SI).
In view of the above experimental observations and the results of calculated NPA charges for selective carbon in aryl azo ions (see Supplementary Scheme S2 in SI), a proposed pathway is displayed in Scheme 4. Initially, the releasing nitrogen of substrate 1a produced the active phenyl cation A (Canning et al., 1999;Canning et al., 2002), which quickly underwent the electrophilic substitution reaction with H 2 O by forming the phenol intermediate B (Ando, 1978). If different diazonium salts were employed, the substrate bearing more positive charges at the carbon adjacent to the azo group could be preferentially converted to phenol species. Sequentially, the in situ para-selective bias of species B with another diazonium salt through electrophilic aromatic substitution (EArS) occurred (Fiege et al., 2000;Xu and Zeng, 2010;Huang and Lumb, 2019), offering the final product 2/3 in one-pot.

CONCLUSION
In summary, we have successfully demonstrated a novel and practical aryl-based difunctionalization strategy of aryldiazonium salts with H 2 O, providing a convenient avenue to valuable symmetric/asymmetric p-azophenols in a one-pot operation. Obviously, this tandem method directly forming aromatic C-O and C-N bonds results in tolerance of readily available substrates, step efficiency, and practical reaction operations. Mechanistically, an ionic mechanism was suggested by relying on the preliminary findings of our designed experiments and the corresponding calculated NPA charges. As far as we know, this is the first example by achieving the aryl-based difunctionalization strategy of aryldiazonium salts with H 2 O to straightforwardly construct versatile p-azophenol derivatives in one-pot. Our difunctionalization model of diazosalts undoubtedly presents a seductive blueprint to assess the significant skeleton in organic synthetic chemistry.

General Information
All reagents were purchased from commercial sources and used without treatment, unless otherwise indicated. The products were SCHEME 4 | Possible mechanism for the reaction. SCHEME 3 | Large-scale reaction and derivatizations.
Frontiers in Chemistry | www.frontiersin.org January 2022 | Volume 10 | Article 818627 purified by column chromatography over silica gel. 1 H NMR and 13 C NMR spectra were recorded at 25°C on a Varian 400 and 100 MHz, respectively, and TMS was used as the internal standard. Mass spectra were recorded on a BRUKER AutoflexIII Smartbeam MS-spectrometer. High-resolution mass spectra (HRMS) were recorded on Bruck microTof by using the ESI method. General procedure for the synthesis of 2/3: Phenyldiazonium tetrafluoroborate 1 (0.5 mmol, 1.0 eq.) and sodium acetate (0.5 mmol, 1.0 eq.) were added in an oven-dried 15-ml Schlenk tube with 2 ml of mixed solvent (MeOH/H 2 O = 3/1) at 0°C; the reaction mixture was then stirred at r.t. for 3 h. The resulting reaction mixture was taken up by dichloromethane (3 × 15 ml). The organic layer was washed with brine (3 × 40 ml), dried over MgSO 4 , and concentrated under reduced pressure. The residue was purified by a silica gel column chromatography (petroleum ether/ethyl acetate = 5:2), and the target product two or three was afforded.
for 3 h. The mixture was extracted with DCM (3 × 30 ml), and the combined organic layers were dried over Na 2 SO 4 and concentrated under reduced pressure. The residue was purified by preparative TLC (PE/EA = 5/1, v/v) to give 4c.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material; further inquiries can be directed to the corresponding authors.

AUTHOR CONTRIBUTIONS
ZL and YF designed the study, performed the experiments, and collected the data. WF performed theoretical calculations and edited the manuscript. YL and XG prepared some substrates (40%). ZL, WG, and BT conceived this project, gave financial support, wrote, and revised the manuscript.