Arene C–H Iodination Using Aryl Iodides

Abstract

Open AccessCCS ChemistryCOMMUNICATION6 Jun 2022Arene C–H Iodination Using Aryl Iodides Shangda Li†, Chunhui Zhang†, Lei Fu, Hang Wang, Lei Cai, Xiaoxi Chen, Xinchao Wang and Gang Li Shangda Li† Key Laboratory of Coal to Ethylene Glycol and Its Related Technology, Center for Excellence in Molecular Synthesis, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, Fuzhou, Fujian 350002 , Chunhui Zhang† Key Laboratory of Coal to Ethylene Glycol and Its Related Technology, Center for Excellence in Molecular Synthesis, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, Fuzhou, Fujian 350002 , Lei Fu Key Laboratory of Coal to Ethylene Glycol and Its Related Technology, Center for Excellence in Molecular Synthesis, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, Fuzhou, Fujian 350002 , Hang Wang Key Laboratory of Coal to Ethylene Glycol and Its Related Technology, Center for Excellence in Molecular Synthesis, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, Fuzhou, Fujian 350002 , Lei Cai Key Laboratory of Coal to Ethylene Glycol and Its Related Technology, Center for Excellence in Molecular Synthesis, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, Fuzhou, Fujian 350002 , Xiaoxi Chen Key Laboratory of Coal to Ethylene Glycol and Its Related Technology, Center for Excellence in Molecular Synthesis, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, Fuzhou, Fujian 350002 , Xinchao Wang Key Laboratory of Coal to Ethylene Glycol and Its Related Technology, Center for Excellence in Molecular Synthesis, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, Fuzhou, Fujian 350002 and Gang Li *Corresponding author: E-mail Address: [email protected] Key Laboratory of Coal to Ethylene Glycol and Its Related Technology, Center for Excellence in Molecular Synthesis, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, Fuzhou, Fujian 350002 Frontiers Science Center for Transformative Molecules, Shanghai Jiao Tong University, Shanghai 200240 https://doi.org/10.31635/ccschem.021.202101156 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Metathesis reactions represent powerful synthetic tools that have been used in a number of fields from the synthesis of natural product to functional material preparation. However, the C–H metathesis reaction is extremely rare. Herein, we report the first Pd(II)-catalyzed C–H iodination of arenes using 2-nitrophenyl iodides as the mild iodinating reagents via a formal metathesis reaction. Unusual C–I bond formation occurred with aryl iodides in preference to competing C–C coupling in this reaction. Assisted by aliphatic carboxyl directing groups, a range of hydrocinnamic acids and related arenes could be selectively iodinated at either meta- or ortho-positions of the phenyl ring. Remote diastereoselective C–H activation was also promising. This method might unfold a novel approach to iodinate challenging substrates. Download figure Download PowerPoint Introduction The exploration of a novel method to cleave and reorganize chemical bonds is the continuous pursuit of organic chemists.1,2 In recent years, significant advances have been achieved in the study of metathesis or isodesmic reactions, which often use user-friendly reagents and exhibit good functional group tolerance.3–29 Of particular note, Morandi group3 and Arndtsen group4 independently reported a functional group metathesis between aryl iodides and aroyl chlorides via a Pd(0)/Pd(II) catalysis (Scheme 1a),30–34 enabling mild iodination of aroyl chlorides. However, catalytic C–H iodination via metathesis between two arenes is unknown.1 Importantly, C–H functionalization via metathesis is extremely rare.22–29 To the best of our knowledge, transition-metal-catalyzed intermolecular aryl C–H functionalization via metathesis has not been realized to date.22–24 Thus, the development of a C–H iodination via (formal) metathesis using aryl iodides is highly attractive, as it could utilize readily available iodinating reagents and offer a novel strategy to generate sophisticated aryl iodides, not readily attainable via conventional methods. Scheme 1 | (a–d) Aryl C–H iodination via formal metathesis. Download figure Download PowerPoint Aryl iodides are extensively used as arylating reagents through exclusive C–C reductive elimination (RE), favored over C–I RE at the metal-center in transition-metal-catalyzed C–H activation reactions (Scheme 1b).35–45 Notably, Whitfield and Sanford46 reported the first carbon–halogen bond-forming RE that occurred in preference to aryl C–C coupling with a Pd(IV) complex to yield aryl chloride in 2007 (Scheme 1c). However, such preference has not been reported in a catalytic reaction.47–60 During our previous study of Pd-catalyzed remote meta-C–H arylation using 2-nitrophenyl iodide, we detected ∼10% of meta-C–H iodination side product.61 Inspired by this unexpected discovery, we envisioned that the successful development of C–H iodination reactions using aryl iodides, which have been used in carboiodination reactions34 and oxidants in C–H activation,62–67 would introduce a mechanistically distinct pathway for catalytic halogenation reactions. In the past decade, site-selective C–H iodination reactions have become an essential strategy for synthesizing aryl iodides, which are versatile, valuable chemicals to use in cross-coupling reactions.68–88 Notably, Yu's group70,71 and Dai's group72 realized few meta-C–H iodination of arenes using pyridine-based templates. Nonetheless, the classes of iodinating reagents for such reactions are still limited,68–88 and most of them are electrophilic such as IOAc generated from I2 with PhI(OAc)2 or AgOAc,89 which might lead to unwanted electrophilic iodination that reduces the site-selectivity of the overall reaction. Therefore, exploring a complementary mild iodinating reagent able to eliminate unwanted side reactions is desirable. Herein, we report an unprecedented Pd(II)-catalyzed C–H iodination reaction of arenes via formal metathesis using aryl iodides as mild iodinating reagents (Scheme 1d). Assisted by the aliphatic carboxyl groups, site-selective ortho- and meta-C–H iodination of hydrocinnamic acids and related arenes have been achieved using commercially available 2-nitrophenyl iodides. Notably, challenging remote diastereoselective C–H activation was also possible. Results and Discussion Initially, we used the hydrocinnamic amide 1a′ (Table 1) bearing an aryl carboxyl meta-directing template as the substrate to investigate the C–H iodination using an aryl iodide, since the desired meta-C–H iodinated product had been obtained as a side product with 1a′ in our previous study.61 Moreover, meta-C–H iodination70–72 of arenes is still very limited to narrow substrate scope,90–107 and hydrocinnamic acids are a class of important core structure of biologically active molecules such as the drug, Baclofen. However, we encountered difficulties in eliminating the undesired meta-C–H arylation product using 1a′. Therefore, substrate 1a was designed as the new substrate, the directing group of which could be prepared on a large scale from known β-amino acid (see Supporting Information). Pleasingly, C–H arylation side product was almost eliminated while using 1a to optimize the reaction, possibly due to the better chelating ability of the aliphatic carboxyl (see Supporting Information for comparison). After extensive tuning of the reaction conditions (see Supporting Information), the desired meta-C–H iodination products 2a was obtained, with 85% combined yield in a reaction involving 2-nitrophenyl iodide using Pd(OAc)2 and pyridine-type L1 as the ligand, in the presence of AgOAc (0.5 equiv) and K2HPO4 (0.5 equiv) in hexafluoroisopropanol (HFIP), set at 100 °C for 24 h (entry 1). Note that 2a was the methylation product from the original acid product, enabling easier isolation. Approximately 0.1 mmol nitrobenzene was also detected by proton nuclear magnetic resonance (1H NMR). This represented the first example that suggests carbon–halogen formation is favored over competing C–C formation at the Pd center in a catalytic reaction using aryl iodides. The yield decreased dramatically without the pyridone ligand, L1, indicating L1 played a crucial role in the reaction (entry 2). L1 might have a dual role: (1) it might accelerate the C–H bond cleavage step by acting as an internal base like an acetate surrogate, (2) it might stabilize the Pd catalyst effectively by forming a stable complex with Pd without reducing Pd's reactivity in the meantime.108 Other ligands such as electron-deficient ligand L2 were also evaluated that led to comparable overall yield with a bit higher turnover number than L1 (entry 3), but a lower yield was obtained with pyridin-2-ol (entry 4). N-mono-protected amino acid ligands such as N-Ac-L-Phe-OH could also promote the reaction but were less effective (entry 5). The addition of silver salt was necessary, and the catalytic amount was feasible (entries 6–10), but the exact role of the silver salt in the reaction is not clear at present. Surprisingly, although it was believed that silver salt was crucial in promoting the C–H arylation for iodide removal,47,48 C–H arylation was not detectable with 1 equiv of AgOAc (entry 8) though trace C–H arylation product was detected using 2 equiv (entry 9). Utilizing a base like Na2CO3 in the reaction was beneficial, but other bases such as K2CO3 also gave comparable good yields (entries 11–13). Solvents were also evaluated, and HFIP proved to be the most suitable. Subsequently, Pd(OAc)2 was superior to other Pd catalysts tested (entries 16 and 17). The reaction was also sensitive to temperature, as the yield decreased markedly at 90 °C, compared with 100 °C under parallel reaction conditions (entries 1 and 18). In addition, reducing the loading of 2-nitrophenyl iodide decreased the yield (entry 19). Notably, meta-selectivity of the reaction was generally excellent, and only very trace regioisomers were detected during the optimization of the reaction conditions. Evaluation of other iodinating reagents indicated electron-withdrawing ortho-substitution of the phenyl iodide was critical, but no better one than 2-nitrophenyl iodide was identified (bottom). The electron-withdrawing ortho-substitution could promote oxidative addition of aryl iodide to the Pd center by a conjoint electronic and weakly coordinating effect.109–111 Moreover, the weak coordination of the nitro group rather than strong coordination (compared with using 2-iodobenzoic acid) to the Pd center, might be critical in preventing C–C RE from Pd(IV).62,65–68 Although the rationale for the preference of this C–I RE is not clear at present, the above reaction conditions optimization suggested that the new carboxyl-based directing group was important in dominating the preference, which might also relate to the steric/electronic properties of the aryl iodide. Other reaction conditions such as ligand and the base additive were also critical in achieving a good yield (see Supporting Information Table S1 for comparison). In contrast, mainly ortho- and para-iodination products, together with trace meta-isomer, were observed with IOAc that might lead to direct electrophilic iodination, and only trace ortho- and para-isomers were detected with N-Iodosuccinimide (NIS) and I2, while 1,3-diiodo-5,5-dimethylhydantoin (DIH) decomposed the substrate. Table 1 | Optimization of the Arene C–H Iodination Reaction Conditionsa Entry Deviation from Standard Conditions Yield (%) [mono/di] 1 None 85 [3.3/1] 2 Without L1 16 [1/0] 3 L2 instead of L1 83 [1.9/1] 4 Pyridin-2-ol instead of L1 70 [2.7/1] 5 N-Ac-L-Phe-OH instead of L1 78 [2.9/1] 6 Without AgOAc 45 [1/0] 7 0.25 equiv of AgOAc 80 [3/1] 8 1.0 equiv of AgOAc 75 [2.9/1] 9 2.0 equiv of AgOAc 74 [3.6/1] 10 Ag2CO3 instead of AgOAc 82 [3.1/1] 11 Without K2HPO4 39 [1/0] 12 K2CO3 instead of K2HPO4 81 [2.9/1] 13 Na2HPO4 instead of K2HPO4 76 [2.8/1] 14 t-Amyl-OH instead of HFIP N.D. 15 TFE instead of HFIP 43 [1/0] 16 Pd(TFA)2 instead of Pd(OAc)2 73 [4.2/1] 17 PdCl2(MeCN)2 instead of Pd(OAc)2 61 [5.8/1] 18 90 °C instead of 100 °C 57 [1/0] 19 1.5 equiv of 2-nitrophenyl iodide 66 [5.6/1] Note: N.D., no product detected. aReaction conditions: (1) 0.1 mmol scale, HFIP (1 mL), under air; (2) MeI (0.2 mmol), K2CO3 (0.3 mmol). The yield of 2a was determined by 1H NMR with CH2Br2 as an internal standard. Part of nitrobenzene by-product removed by rotavapor (entry 1). Unless otherwise noted, both C–H arylation side product and regioisomers were traces determined by gas chromatography–mass spectrometry (GC-MS) with a flame ionization detector (FID). bA little arylation and iodination regioisomers detected. cDetected with GC-MS, IOAc (from I2/PhIOAc). With the optimized conditions in hand, we tested this protocol with a series of hydrocinnamic acids and related arenes (Scheme 2). The combined yield of isolated 2amonoand 2adi was high (81%); ligand L2 led to a higher turnover and was employed for other substrates. To our delight, generally, average to good yields of desired products were obtained with a range of mono-substituted substrates bearing either electron-withdrawing or -donating groups ( 2b–2l; 49–82%). Importantly, halides such as chloride ( 2f and 2k) and bromide ( 2g and 2l) were tolerated, providing the opportunity to synthesize diversely substituted arenes. However, para-substituted substrates only gave low yields of desired products. Furthermore, di-substitution ( 2m) and substituting the alkyl chain such as 3-phenyllactic acid derivative ( 2n; 52%) were allowed. Finally, structurally related biphenylcarboxylic acids ( 2o– 2q; 65–71%) and benzyl alcohol ( 2r; 50%) derivatives could also be iodinated at the desired meta-positions. The selectivity of the reactions was excellent with trace amounts of regioisomers, and generally no detectable arylation side product was observed. The directing group could be smoothly cleaved under acidic conditions (see Supporting Information). Scheme 2 | Scope of meta-C–H iodination. Reaction conditions: standard conditions, deviation: L2 as the ligand, 48 h. Isolated yields. aL1 used. b24 h. cOptical pure (<99% ee) directing group was used for 1n; the yield of 2nmono was calculated after hydrolysis. dAbout 10% di-product, but it could not be isolated. Download figure Download PowerPoint Since ortho-iodinated hydrocinnamic acids are also valuable compounds, we moved on to test the generality of this method through ortho-C–H iodination of hydrocinnamic acids. Importantly, ortho-C–H functionalization of hydrocinnamic acids using their native free carboxyl as the chelating group is exceptionally scarce, and the scope of reported examples was very limited,112–114 possibly due to the requirement of the formation of challenging 7-membered metallacycle. Based on the above reaction conditions, and after careful investigation (see Supporting Information Tables S2–S4 for details), we obtained the desired ortho-C–H iodination products after methylation (Scheme 3, 4a) in excellent combined yield (93%) with 1-iodo-4-methoxy-2-nitrobenzene, which is commercially available and could be prepared readily using N-Formyl-Gly-OH as the ligand in the presence of AgOBz (0.1 equiv) and NaOAc (0.5 equiv). This protocol proved robust, leading to generally high yields of desired products with a broad range of hydrocinnamic acids ( 4a– 4r; 62–93%). More complicated 3-phenyllactic acid ( 4s; 87%), phenylalanine ( 4t; 68%), and drug Baclofen ( 4u; 51%) derivatives could also be iodinated to give valuable products. Scheme 3 | Scope of ortho-C–H iodination. Reaction conditions: 3 (0.2 mmol), 1-iodo-4-methoxy-2-nitrobenzene (0.4 mmol), Pd(OAc)2 (0.02 mmol), N-Formyl-Gly-OH (0.04 mmol), AgOBz (0.02 mmol), NaOAc (0.1 mmol), HFIP (2 mL), 80 °C, 24 h, under air. Isolated yields. ayield of 7 mmol scale. b10% (o,m)-di-product was isolated, see Supporting Information. cSOCl2/MeOH was used for methylation, see Supporting Information. Download figure Download PowerPoint As remote asymmetric meta-C–H functionalization is still extremely rare and challenging,115,116 we were curious to use an optical pure directing group to induce diastereoselective remote meta-C–H iodination via desymmetrization. In our preliminary study (Scheme 4), good diastereoselectivity (up to d.r. = 90.5/9.5, 6bmono) could be achieved with a 5-bromopyridin-2-ol ligand. The absolute configuration of 6amono after removal of the directing group was determined by X-ray crystallography ( 7). However, higher diastereoselectivity could not be obtained at present even after extensive study; thus, further investigation is required. Scheme 4 | Diastereoselective remote meta-C–H iodination. Download figure Download PowerPoint In addition, the synthetic potential of the methods was briefly evaluated (Schemes 5a–5c). Cross-coupling reactions proceeded smoothly with product 2amono to afford meta-substituted derivatives ( 8– 10). Besides, a synthetic chiral amino acid derivative ( 11) could also be efficiently produced with an ortho-iodinated phenylalanine derivative. The meta-directing group could be removed under acidic conditions to give a high yield of iodide 12. Scheme 5 | (a–c) Synthetic elaborations of the arene C–H iodination. Download figure Download PowerPoint To gain insight into the reaction mechanism, preliminary mechanistic studies were conducted. First, no C–H/C–I exchange was detected using an excess of arene or other substrates (Schemes 6a and 6b), suggesting the reaction was not reversible (see Supporting Information for more control reactions). Second, inspired by Daugulis's work,47 we performed the reaction with Pd(0) as the catalyst; this did not lead to any iodination product, suggesting a C–I RE from a Pd(II) intermediate was unlikely (Scheme 6c). Notably, C(aryl)–I RE from a Pd(II) complex generally required the use of a bulky ligand, which was not present in our reaction.30–34 Moreover, a good yield was also obtained without using silver salt.65–68 Third, although it was hard for us to detect much arylation product with substrate 1a and 2-nitrophenyl iodide by modifying the reaction conditions, much arylation product could be observed when using methyl 2-iodobenzoate as the iodinating reagent (Scheme 6d, see Supporting Information for more variations). Of note, a Pd(IV) complex is often believed to be involved in C–H arylation with aryl iodide.47–53 Moreover, three types of products could be generated by subjecting 2-phenyl phenol to our standard conditions (Scheme 6e), where related dibenzofuran formation often required a strong oxidant that could enable a C–O RE from a Pd(IV) complex if no bulky ligand was utilized.117–119 These results in Schemes 6c–6e might support that a Pd(IV) intermediate was possibly involved in a selective C–I RE competing with C–C RE in current iodination reactions. Finally, control experiments involving excluding the slow release of iodine from aryl iodide were also carried out (see Supporting Information). Scheme 6 | (a–e) Mechanistic studies to gain insight into the arene C–H iodination. Download figure Download PowerPoint Based on previous works47–53,61 and the above mechanistic studies, the proposed catalytic cycle for the above meta-C–H iodination is outlined in Scheme 7, as follows: First, active Pd catalyst A is generated through ligand exchange. Subsequently, the substrate 1a might coordinate to Pd in a κ1 or κ2 coordination mode, but the latter mode is believed to facilitate the approaching of the Pd center to the remote phenyl ring giving complex B. Then the C–H bond at the meta-position of the phenyl ring is selectively cleaved via a potential electrophilic metalation deprotonation process,120 possibly due to its best-matched distance and geometry, affording palladacycle C. Oxidative addition of C with 2-nitrophenyl iodide gives a Pd(IV) intermediate D. In complex D, the nitro group might act as a weak ligand to coordinate to the Pd center, which might play an important role to inhibit C–C RE; thus, facilitating C–I RE to afford product 2amono-H together with an arylated Pd(II) complex E. Such an effect of weak coordination of the ortho-substituent was also observed by Chen's group while using 2-methoxy phenyl iodides as the oxidants to promote C–N RE in the enantioselective intramolecular C(sp3)–H amidation reaction.62 Finally, protonolysis of complex E regenerates active Pd(II) catalyst A and nitrobenzene. In addition, the reduction of complex E to generate nitrobenzene and Pd(0) by oxidizing HFIP might be a minor pathway. Notably, the generation of nitrobenzene is consistent with the metathesis reaction. However, since the proton for the protonation of 2-nitrophenyl iodide was predominantly from other sources such as the solvent HFIP rather than the substrate (see Supporting Information for reaction with deuterated substrate); this reaction could be viewed as a formal metathesis reaction. For the ortho-C–H iodination, the catalytic cycle is similar, except that κ1 coordination of the substrate to Pd center is better to facilitate cyclopalladation at the ortho-position ( F). Moreover, other reaction pathways such as Pd(II)/Pd(III) pathway could not be excluded at this stage;57,58 hence, further mechanistic studies are being actively carried out, and the results would be reported in due course. Scheme 7 | Proposed catalytic cycle for the meta-C–H iodination. Download figure Download PowerPoint Conclusion We have developed the first Pd(II)-catalyzed C–H iodination reaction of arenes using aryl iodides via formal metathesis. Two 2-nitrophenyl iodides were identified as the mild iodinating reagents for meta- and ortho-C–H iodination of a range of hydrocinnamic acids and related arenes assisted by the carboxyl directing groups. In addition, remote diastereoselective C–H activation was also proved to be possible. This method might stimulate the study on developing C–H activation reactions via (formal) metathesis and uncover a mild way to iodinate challenging substrates. Mechanistic study and further application of this method are currently underway in our laboratory. Supporting Information Supporting Information is available and includes general experimental procedures and characterization spectra. Conflict of Interest The authors declare no competing financial interest. Preprint Statement An early version of the manuscript was posted on the preprint server ChemRxiv before publication in CCS Chemistry. The corresponding preprint article can be found here: http://dx.doi.org/10.26434/chemrxiv.14449668 Acknowledgments The authors gratefully thank the financial supports from NSFC (grant nos. 22022111 and 22071248), the Natural Science Foundation of Fujian Province (grant nos. 2020J02008 and 2020J01108), the Youth Innovation Promotion Association of the Chinese Academy of Sciences (grant no. 2020306), and the Strategic Priority Research Program of the Chinese Academy of Sciences (grant no. XDB20000000). References 1. Bhawal B. N.; Morandi B.Catalytic Isofunctional Reactions-Expanding the Repertoire of Shuttle and Metathesis Reactions.Angew. Chem. Int. Ed.2019, 58, 10074–10103. Google Scholar 2. Bhawal B. N.; Morandi B.Catalytic Transfer Functionalization through Shuttle Catalysis.ACS Catal.2016, 6, Google Scholar Morandi a Metathesis between and Aryl Google Scholar Arndtsen B. A Metathesis of and Chem. Google Scholar Morandi B.Catalytic through Transfer Google Scholar Bhawal B. N.; Morandi or Metathesis by Google Scholar Morandi Shuttle and for Google Scholar Li Reaction between and of by and Google Scholar bond by Transfer Google Scholar of Chem. Google Scholar N.; of to via the of Two Google Scholar 12. Transfer of and Using Chem. Google Scholar Chen C. Transfer of Chem. Google Scholar of and by Chem. Int. Ed.2019, 58, Google Scholar Transfer of to Chem. Int. Ed.2019, 58, Google Scholar N.; Transfer Reaction of with and by a Google Scholar Li of Aryl with and Google Scholar as of via in Google Scholar Transfer of of and Google Scholar Reaction via Metathesis for of Chem. Google Scholar Reaction for the of with and Google Scholar and of of Transfer Chem. Google Scholar C. in the Transfer C–H of and Google Scholar of from through of and Chem. Int. Google Scholar in and 7, Google Scholar by Assisted by a Chem. Int. Google Scholar of

Topics & Concepts