Iron-Catalyzed Cross-Electrophile Coupling of Inert C–O Bonds with Alkyl Bromides

Open AccessCCS ChemistryRESEARCH ARTICLES6 Oct 2022Iron-Catalyzed Cross-Electrophile Coupling of Inert C–O Bonds with Alkyl Bromides Shuo Chen†, Zijian Wang†, Shasha Geng, Hongdan Zhu, Zhengli Liu, Yun He, Qian Peng and Zhang Feng Shuo Chen† Chongqing Key Laboratory of Natural Product Synthesis and Drug Research, School of Pharmaceutical Sciences, Chongqing University, Chongqing 401331 , Zijian Wang† State Key Laboratory of Elemento-Organic Chemistry, Tianjin Key Laboratory of Biosensing and Molecular Recognition, College of Chemistry, Nankai University, Tianjin 300071 , Shasha Geng Chongqing Key Laboratory of Natural Product Synthesis and Drug Research, School of Pharmaceutical Sciences, Chongqing University, Chongqing 401331 , Hongdan Zhu State Key Laboratory of Elemento-Organic Chemistry, Tianjin Key Laboratory of Biosensing and Molecular Recognition, College of Chemistry, Nankai University, Tianjin 300071 , Zhengli Liu Chongqing Key Laboratory of Natural Product Synthesis and Drug Research, School of Pharmaceutical Sciences, Chongqing University, Chongqing 401331 , Yun He Chongqing Key Laboratory of Natural Product Synthesis and Drug Research, School of Pharmaceutical Sciences, Chongqing University, Chongqing 401331 , Qian Peng *Corresponding authors: E-mail Address: email protected E-mail Address: email protected State Key Laboratory of Elemento-Organic Chemistry, Tianjin Key Laboratory of Biosensing and Molecular Recognition, College of Chemistry, Nankai University, Tianjin 300071 Haihe Laboratory of Sustainable Chemical Transformation, Tianjin 300192 and Zhang Feng *Corresponding authors: E-mail Address: email protected E-mail Address: email protected Chongqing Key Laboratory of Natural Product Synthesis and Drug Research, School of Pharmaceutical Sciences, Chongqing University, Chongqing 401331 Medical Imaging Key Laboratory of Sichuan Province, Affiliated Hospital of North Sichuan Medical College, Nanchong 637000 https://doi.org/10.31635/ccschem.022.202202234 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail An example of iron-catalyzed cross-electrophile coupling of inert C–O bonds with alkyl bromides via an iron/B2pin2 catalytic system has been developed. Aryl and heteroaryl carbamates can smoothly undergo this transformation under mild conditions, delivering the alkylated products with good efficiency. This protocol exhibits good functional group compatibility and enables the late-stage functionalization of biorelevant compounds, thus providing excellent opportunities for applications in medicinal chemistry. Control experiments and computational studies reveal that a high spin Fe(I/II/III) catalytic mechanism might be involved in this reaction through single electron transfer to activate alkyl bromides, oxidative addition of aryl carbamates, and reductive elimination to form Csp2–Csp3 bonds. Download figure Download PowerPoint Introduction Iron catalysis has drawn much attention in recent years and provides great potential in organic chemistry because iron is abundant, nontoxic, and relatively inexpensive.1–5 To date, most iron-catalyzed reactions rely on preformed organometallic reagents such as organoborons, organomagnesiums, and organozincs (Scheme 1a and 1b).6–17 However, these organometallic reagents are sometimes unstable and incompatible with sensitive functional groups and often require preparation, thus restricting their widespread applications. To address these issues, in 2009, the Jacobi von Wangelin group18 disclosed an impressive iron-catalyzed Kumada cross-coupling reaction in which the organomagnesium reagents were generated in situ. There is no doubt that the development of operationally simple cross-coupling reactions that avoid the use of organometallic reagents is of great importance (Scheme 1c and 1d). To this end, significant achievements have been made in nickel-catalyzed cross-electrophile coupling reactions.19–35 Very recently, the interesting iron-catalyzed alkyl radical addition reactions of aryl alkenes, allyl carbonates, and vinyl halides were reported by the Lipshutz, Gong, and Koh groups,36–38 in which oxidative addition was not involved. To the best of our knowledge, the iron-catalyzed cross-electrophile coupling reaction involving oxidative addition and reductive elimination has never been documented. This may be because the low-valent iron species are unstable and difficult to obtain and there exist challenges associated with understanding elementary reaction steps. With our continuing interest in iron catalysis,39–43 we set out to develop an iron-catalyzed cross-electrophile coupling reaction to construct C–C bonds. Scheme 1 | Iron-catalyzed cross-coupling reactions. Download figure Download PowerPoint In the previous studies on the cross-electrophile coupling reactions, aryl halides were always employed as coupling partners.44–48 Compared with aryl halides, phenol derivatives are more attractive because of their abundance and avoidance of the halide-containing wastes.49 Thus far, C–O electrophiles are less common than aryl halides as oxygen-based electrophiles in cross-electrophile coupling reactions.50–55 Moreover, the use of unactivated C–O bonds as coupling partners has seldom been documented because the C–O bonds with strong bond dissociation energy are challenging for oxidative addition.56 At present, the greatest challenge in iron catalysis is the generation of reactive low-valent iron species to facilitate the oxidative addition of the inert bonds. Therefore, organometallic reagents are usually employed as additives to afford the low-valent iron catalyst via reductive elimination.57–60 During our investigation on iron-catalyzed borylation and silylation of phenol derivatives, interestingly, we found that iron catalysts might possess some unique reactivity in the functionalization of inert bonds.40–43 Inspired by these findings, we envisioned that when special oxygen-based electrophiles, such as aryl carbamates, are used as coupling partners, the iron-catalyzed cross-electrophile coupling reactions might proceed under a suitable catalytic system. Herein, we disclose the example of iron-catalyzed cross-electrophile coupling reaction between aryl carbamates and alkyl bromides via an iron/B2pin2 catalytic system (Scheme 1e). Experimental Methods A 25 mL flame-dried Schlenk tube equipped with a magnetic stir bar was charged with FeI2 (0.02 mmol, 6.2 mg, 0.1 equiv), bis(pinacolato)diboron (0.5 mmol, 127 mg, 2.5 equiv), MeOLi (1.2 mmol, 45.6 mg, 6.0 equiv), and aryl carbamates (0.2 mmol, 1.0 equiv) in a glove box. Then, the tube was evacuated and backfilled with nitrogen three times; N,N,N′,N′-tetramethylethylenediamine (TMEDA) (0.02 mmol, 0.1 equiv), alkyl bromides (0.3 mmol, 1.5 equiv), and freshly distilled tert-butyl methyl ether (MTBE) (1.0 mL) were added under nitrogen atmosphere. The reaction mixture was stirred at 70 °C for 16 h. The cooled solution was quenched with saturated NH4Cl aqueous solution, then diluted with ethyl acetate, and washed with saturated brine. The organic phase was dried over Na2SO4 and concentrated under vacuum. The residue was purified by silica gel flash chromatography to afford the corresponding compounds. Computational Methods Theoretical calculations were performed with the Gaussian 09 program using the unrestricted dispersion-corrected ω-B97XD functional without any symmetry constraints. Spin states of iron complexes were considered for all calculated intermediates and transition states. Structural optimizations were performed using the effective core potentials of Hay and Wadt with a double-ζ basis set (LanL2DZ) for Br and I, the TZVP basis set for Fe, and the def2SVP basis set for H, Li, B, C, N, and O. The single point energies were further estimated using a larger basis set def2TZVPP for all atoms with the solvation model at the solute electron density (SMD) level. Calculated intermediates and transition states were verified as either minima or transition state structures by the presence of zero or a single imaginary vibrational frequency, respectively. Free energies were evaluated at 298 K using harmonic vibrational frequencies. The calculated structures were displayed with the CYLview software. The Mayer bond order was calculated by Multiwfn. Spin density iso-surfaces were shown by the visual molecular dynamics (VMD) 1.9.3 program to gain more insights. The details of experimental and computational methods are available in the Supporting Information. Results and Discussion With these considerations in mind, we began our investigation on the iron-catalyzed cross-electrophile coupling reaction (for details, see Supporting Information Tables S1–S12). After our extensive investigations, we found that oxygen-based groups, such as OAc, OTs, OTf, and OPiv, could not undergo the cross-electrophile coupling reaction with cyclopentyl bromide. Gratifyingly, using B2pin2 as a reductant, aryl carbamate 1a reacted smoothly, providing the desired product 1 in 24% yield (Table 1, entry 3). However, when Zn, Mn, or tetrakis(dimethylamino)ethylene (TDAE) was used as the reductant, alkylation did not occur (for details, see Supporting Information). The inorganic bases were very crucial to this reaction. MeOLi successfully promoted this transformation, but other bases, such as MeOK and t-BuONa could not provide the desired product (Table 1, entries 1–3). After evaluating various solvents, this reaction was found to proceed well in ethers, and MTBE stood out, affording 1 in 56% yield (Table 1, entry 4; for details, see Supporting Information). This reaction would not take place in strong polar solvents such as dimethylacetamide, N-methyl-2-pyrrolidone, or dimethyl sulfoxide. Then, iron sources were tested, and Fe(OTf)3 was found to promote this reaction as well, albeit in low yield (Table 1, entry 5). FeI2 exhibited good efficiency, providing the alkylated product in 53% yield (Table 1, entry 6). To our delight, by increasing the base loading, the reaction proceeded with high efficiency, and 1 in 78% yield was provided (Table 1, entry 7). After examining various ligands, moderate yields (24%–36%) were obtained using L1 ((1R,2R)-N1,N1,N2-trimethyl-1,2-diphenylethane-1,2-diamine) and L2 (2,2,6,6-tetramethylheptane-3,5-dione), and TMEDA proved to be the best choice (for details, see the Supporting Information). Phosphine ligands could also promote this transformation but resulted in a poor yield (Table 1, entry 8; for details, see the Supporting Information). Interestingly, upon increasing the concentration of this reaction, 84% yield was obtained (Table 1, entry 9). In addition, the borylated byproducts were also be observed in this reaction (for details, see the Supporting Information). Control experiments demonstrated the necessity of the iron catalyst. No desired product was observed in the absence of an iron catalyst, and only 40% yield was obtained without ligand. The high-purity iron catalyst was examined as well, and a comparable yield was afforded (Table 1, entry 12). The evaluation of trace-metal effect was also investigated, and no desired product was observed under copper and nickel catalysis (Table 1, entries 13 and 14). This transformation was slightly sensitive to water and air. A 66% yield was obtained using reagent grade MTBE, and 37% of the desired product was generated under air condition (Table 1, entries 15 and 16). Table 1 | Optimization of the Reaction Conditionsa Entry Fe Ligand Solvent Base Yield (%)b 1 Fe(OTf)2 TMEDA 1,4-dioxane MeOK 0 2 Fe(OTf)2 TMEDA 1,4-dioxane t-BuONa 0 3 Fe(OTf)2 TMEDA 1,4-dioxane MeOLi 24 4 Fe(OTf)2 TMEDA MTBE MeOLi 56 5 Fe(OTf)3 TMEDA MTBE MeOLi 29 6 FeI2 TMEDA MTBE MeOLi 53 7c FeI2 TMEDA MTBE MeOLi 78 8c FeI2 dppe MTBE MeOLi 20 9c,d FeI2 TMEDA MTBE MeOLi 84 10c,d FeI2 — MTBE MeOLi 40 11c,d — TMEDA MTBE MeOLi 0 12c,d FeI2 (99.99%) TMEDA MTBE MeOLi 76 13c,d CuI TMEDA MTBE MeOLi 0 14c,d NiBr2 TMEDA MTBE MeOLi 0 15c,d,e FeI2 TMEDA MTBE MeOLi 66 16c,d,f FeI2 TMEDA MTBE MeOLi 37 aReaction conditions (unless otherwise specified): Aryl carbamate 1a (0.2 mmol, 1.0 equiv), cyclopentyl bromide 1b (0.3 mmol, 1.5 equiv), Fe (0.02 mmol, 0.1 equiv), Ligand (0.1 equiv), B2pin2 (0.5 mmol, 2.5 equiv), base (1.0 mmol, 5.0 equiv), solvent (2.0 mL), 16 h. bDetermined by 1H NMR using mesitylene as an internal standard. The isolated yield is shown in parentheses. cMeOLi (1.2 mmol, 6.0 equiv) was used. dMTBE (1.0 mL) was used. eReagent grade MTBE (1.0 mL) was used. fThis reaction was carried out under air condition. After the optimal reaction conditions were established, we next evaluated the scope of this cross-electrophile coupling reaction (Table 2). This protocol exhibited good efficiency using naphthyl carbamates as substrates and offered the corresponding products in moderate to good yields ( 1– 36, 35%–87%) (Some negative results are shown in Supporting Information). Some functional groups were well-tolerated, such as amine, tert-butyldimethylsilyl)oxy- (OTBS), F, OCF3, ketal, carboxylate, Bpin, CF3, and carbamate ( 6, 11, 12, 15, 16, 17, 18, 20, 22, 25, 32, and 51). The free NH group was also tolerated, and a reasonable yield was obtained ( 30, 52%). Numerous primary and secondary alkyl bromides smoothly underwent this transformation, while tertiary alkyl bromides did not react. Cyclobutyl bromide bearing a strained ring performed well, delivering the desired product in acceptable yield ( 3, 47%). The alkyl bromide with the base-sensitive fluoro group, reacted well, and 38% yield was provided ( 15). Next, we examined the electronic effect of substituents on the aryl carbamates and found that substrates containing electron-rich groups underwent this transformation well. The substrate with the electron-deficient carboxylate performed this reaction as well, affording the corresponding product in accepted yield ( 20, 35%). Importantly, this reaction was not sensitive to steric effect. Substrates bearing an ortho-MeO or Me group delivered the desired products in reasonable yields ( 24, 47%; 54, 43%). Moreover, heteroaromatic derivatives, such as furan and thiophene, provided the alkylated products in moderate yields ( 27– 28, 53%–57%). However, this transformation proceeded with low efficiency using biphenyl carbamates as substrates under standard conditions. After extensive investigations, we found that ligand L1 containing an unprotected amine group smoothly promoted this reaction. Biphenyl carbamates reacted well with cyclopentyl bromide in the presence of ligand L1, furnishing the corresponding products in moderate yields ( 37– 39, 50%–52%). The more inert substrates, monophenyl carbamates, could also undergo this transformation under the modified conditions, albeit with low efficiency ( 40– 43, 20%–30%), which may be because monophenyl carbamates without a π-conjugated skeleton show low reactivity for oxidative addition of iron species, resulting in low efficiency. Subsequently, heteroaryl carbamates were studied (for details, see Supporting Information Tables S13–S17), and the pyridine and quinine derivatives reacted well. Interestingly, the carbamate group at different positions of pyridine exhibited different reactivity. Using the 1,3-dicarbonyl compound ( L2) as a ligand, pyridines with a carbamate group at the para-position provided the desired products in moderate yields ( 44– 48, 43%–70%). Pyridines containing a carbamate group at the meta-position underwent this transformation well with the aid of ligand L1, and moderate yields were obtained ( 49– 54, 40%–51%). Unfortunately, phenyl carbamate and tert-butyl bromide were not suitable for this reaction (see Supporting Information Figure S16). Table 2 | Scope of the Iron-Catalyzed Reductive Cross-Coupling Reaction of Aryl Carbamates with Alkyl Bromidesa aReaction conditions: Aryl carbamates (0.2 mmol, 1.0 equiv), alkyl bromides (0.3 mmol, 1.5 equiv), FeI2 (0.02 mmol, 0.1 equiv), TMEDA (0.02 mmol, 0.1 equiv), MeOLi (1.2 mmol, 6.0 equiv), B2pin2 (0.5 mmol, 2.5 equiv), MTBE (1.0 mL), 70 °C, 16 h. bFeCl2 (0.02 mmol, 0.1 equiv), L1 (0.02 mmol, 0.1 equiv), MeOLi (1.4 mmol, 7.0 equiv), B2pin2 (0.5 mmol, 2.5 equiv), MTBE (1.0 mL), 80 °C, 16 h. cFeBr2 (0.02 mmol, 0.1 equiv), L2 (0.02 mmol, 0.1 equiv), MeOLi (1.2 mmol, 6.0 equiv), B2pin2 (0.5 mmol, 2.5 equiv), MTBE (1.5 mL), 90 °C, 16 h. dFeBr2 (0.02 mmol, 0.1 equiv), L1 (0.02 mmol, 0.1 equiv), MeOLi (1.4 mmol, 7.0 equiv), B2pin2 (0.5 mmol, 2.5 equiv), MTBE (1.5 mL), alkyl bromides (1.2 equiv), 80 °C, 16 h. The rapid one-pot synthesis of 1 was carried out to validate the utility of this iron-catalyzed cross-electrophile coupling reaction (Scheme 2a). The unseparated in situ-generated aryl carbamate 1a was used as the substrate, and a moderate yield was obtained. Moreover, 1-naphthol could be directly employed as a coupling partner, and the corresponding product was afforded in 35% yield (Scheme 2a). Furthermore, the late-stage functionalization of biomolecules was conducted (Scheme 2b). The biorelevant molecule derived from cholesterol with an alkenyl group underwent this transformation smoothly, delivering the desired product in moderate yield ( 56, 43%). The diacetone-d-glucose derivative also performed well, and the corresponding product was obtained in acceptable yield ( 57, 41%). Moreover, biomolecules derived from estradiol, estrone, and (+)-dehydroabietylamine also performed this transformation, providing the desired products in acceptable yields ( 58– 60; 29%–55%). Then, the gram-scale synthesis of 1 was realized, and 69% yield was obtained (Scheme 2c). The carbamate group has been widely applied in the functionalization of C–H bonds.61,62 Therefore, a sequential synthesis of the multisubstituted arene through a carbamate directing group was conducted. Compound 61 was obtained in an acceptable yield through the C–H borylation followed by an iron-catalyzed cross-electrophile coupling reaction (Scheme 2d; for details, see Supporting Information). Scheme 2 | One-pot synthesis and applications. Download figure Download PowerPoint To gain further insight into the mechanism of this iron-catalyzed cross-electrophile coupling reaction, the corresponding arylboronic ester 62 and alkylboronic ester 63 were first synthesized, but they did not undergo this transformation under standard reaction conditions, which revealed that the Suzuki reaction pathway of this reaction could be ruled out (Scheme 3a, for details, see Supporting Information Table S18 and Figures S1 and S2). Radical-trapping experiments were also conducted (Scheme 3b); when a radical scavenger, 2,2,6,6-tetramethylpiperidinyloxy (TEMPO) (100 mol %), was added under the standard reaction conditions, the adduct 64 between TEMPO and alkyl bromide 13b was isolated in 20% yield, while the desired product 1 and aryl-TEMPO adduct were not observed, suggesting that an alkyl radical was involved in this reaction (Scheme 3b1, for details, see Supporting Information Table S19 and Figure S3). Moreover, the alkyl radical was further trapped by 1,1-diphenylethylene with low efficiency, indicating that the alkyl radical showed poor reactivity with alkenes in this catalytic system (Scheme 3b2, for details, see Supporting Information Figure S4). Scheme 3 | Mechanistic studies. Download figure Download PowerPoint Then, experiments were performed to study how this reaction was induced. We monitored this reaction through NMR spectroscopy and found that this reaction had a long induction period (see the Supporting Information Table S20 and Figure S5). In the initial stage of the reaction, a large amount of alkyl bromide was consumed, while the aryl carbamate remained unchanged. Moreover, when 1b was first used as the substrate to trigger this reaction, a higher yield was obtained (Scheme 3c1 vs c2; see Supporting Information Figure S6). After the induction period, this transformation could proceed well when 1a was added to the mixture of 1b, TMEDA, FeI2, B2pin2, and MeOLi (Scheme 3c2), suggesting that this transformation might be first induced from alkyl bromides. Encouraged by these findings, we thought that the diboron reagent might reduce the iron catalyst to form a low-valent iron-boryl complex during the induction period. After many attempts to determine the key reaction intermediates, we were unable to observe the TMEDA-iron complexes using TMEDA as the ligand, possibly due to the instability of such complexes. Next, we used 1,2-bis(diphenylphosphino)benzene (dppbz) as a ligand to capture the possible intermediates due to its strong coordination to the iron center. A cationic (dppbz)2-Fe(II)-Bpin complex was observed by high-resolution mass spectrometry (see Supporting Information Figures S7 and S8). In Bedford's work,60 the diboron reagent could reduce the Fe(II) catalyst to an Fe(I)-boryl species. Therefore, a known Fe(I)-boron complex was synthesized to promote this transformation (see Supporting Information Figure S9). To our delight, the coupling product was obtained in 5% yield in the presence of Fe(I)-boron complex 1 (Scheme 3d1 and Supporting Information Table S21). The very low efficiency of this reaction may be caused by the high stability of dppbz-ligated iron(I)-Bpin complex 1.60 Moreover, an important dppe-ligated Fe(I)-Cl complex was synthesized to perform this reaction, and 22% yield of the coupling product 1 was provided (Scheme 3d2). Additionally, Fe(0) complexes such as N-Heterocyclic Carbene (NHC)-ligated Fe(0) complex63 and Fe(PMe3)4 did not perform this transformation (see Supporting Information Figures S10 and S11). Based on these results, initiation of this reaction by an Fe(0) species was unlikely, and the Fe(I) species might be involved in the catalytic cycle. Next, when the standard reaction mixture was analyzed by X-ray photoelectron spectroscopy, the peaks corresponding to FeII2p3/2 (710.4 eV, compared with FeCO3),64 FeIII2p3/2 (712.9 eV, compared with Fe2(SO4)3),65 and a suspected Fe(I) species were found with the binding energy at 711.0 eV,66 suggesting that the Fe(I) and Fe(III) species are involved in this catalytic cycle (see the Supporting Information Figure S12). Additionally, a kinetic study indicates that the rate of this reaction tends to be first order with respect to the iron catalyst (see the Supporting Information Tables S22 and and Figures To further the calculations were performed at the of with the solvation model (Scheme The calculated results the from FeI2 to Fe(I)-boryl is by energy on the 4 of high spin state under the MeOLi This was calculated to be more than the species ( Supporting Information Figure The Fe(I)-boryl complex 4 without MeOLi would to an energy suggesting the important of MeOLi to the 4 (Scheme To the the 4 the bond in 4 through single electron transfer than the C–O with the relatively high energy (Scheme which our in Scheme of electronic 4 at the state the effect of and to the bond ( Supporting Information Figure by the molecular and bond with Mayer bond the resulting 5 was by in the MeOLi had to and to with species 5 which the coordination of for an by alkyl radical (Scheme This transformation may also the of due to the center. The radical via 6 a coordination was an suggesting the alkyl radical might not from the iron which may provide a for the very radical by in Scheme The electronic energy of transition state 6 at the state is more than that at the the energy in this which is with the energy in our energy ( Supporting Information Figure The reductive elimination is a and low via 4 to complex and species 4 (see the Supporting Information Figure The relatively strong binding of 1a to the species could form a key of C–O oxidative addition 4 and the coordination of MeOLi which the MeOLi may not be involved in the oxidative addition of the C–O bond and reductive elimination of the C–C bond (see Supporting Information Figure The transition state 4 displayed a oxidative addition with in to the common oxidative The reductive elimination with low energy 4 would afford the product 1, a Csp2–Csp3 bond by the elimination of alkyl and aryl groups from the species 4 Then, 4 would be under the condition for the next catalytic cycle. Scheme 4 | Theoretical studies. Download figure Download PowerPoint Based on our experiments and computational a mechanism via Fe(I/II/III) intermediates is in Scheme FeI2 is by B2pin2 and MeOLi to Fe(I)-boron complex I, which with alkyl bromides through the pathway to afford Fe(II) complex and Fe(III) complex An complex is then delivered via reductive The is through the oxidative addition of aryl carbamates with the Fe(I) complex the desired product is provided via the reductive elimination of the Fe(I) complex the of iron Scheme 5 | Download figure Download PowerPoint We have an example of iron-catalyzed cross-electrophile coupling reaction of aryl carbamates with alkyl bromides via C–O bond This reaction under mild conditions with good functional delivering the alkylated and pyridines with good efficiency. This protocol late-stage functionalization of compounds, alkylation of and the synthesis of multisubstituted A mechanism via a Fe(I/II/III) catalytic cycle was revealed by experimental and computational studies. In addition, this transformation not only nickel-catalyzed cross-electrophile coupling reactions, but also a understanding of the unique reactivity of iron Supporting Information Supporting Information is available and experimental and of 1H NMR and NMR for all compounds. of and have a other no of Information We the Key and of Natural of and Natural of Sichuan the Natural of Tianjin Chongqing and Affiliated Hospital of North Sichuan Medical College for We also the and of Chongqing University, for of Nankai University, and Haihe Laboratory of Sustainable Chemical of We