In Situ Monitoring of Transmetallation in Electric Potential-Promoted Oxidative Coupling in a Single-Molecule Junction

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE2 Mar 2022In Situ Monitoring of Transmetallation in Electric Potential-Promoted Oxidative Coupling in a Single-Molecule Junction Yunpeng Li, Chengxi Zhao, Rui Wang, Ajun Tang, Wenjing Hong, Dahui Qu, He Tian and Hongxiang Li Yunpeng Li Key Laboratory for Advanced Materials, Feringa Nobel Prize Scientist Joint Research Center, Frontiers Science Center for Materiobiology and Dynamic Chemistry, Institute of Fine Chemicals, School of Chemistry and Molecular Engineering, East China University of Science and Technology, Shanghai 200237 Google Scholar More articles by this author , Chengxi Zhao Key Laboratory for Advanced Materials, Feringa Nobel Prize Scientist Joint Research Center, Frontiers Science Center for Materiobiology and Dynamic Chemistry, Institute of Fine Chemicals, School of Chemistry and Molecular Engineering, East China University of Science and Technology, Shanghai 200237 Google Scholar More articles by this author , Rui Wang Key Laboratory for Advanced Materials, Feringa Nobel Prize Scientist Joint Research Center, Frontiers Science Center for Materiobiology and Dynamic Chemistry, Institute of Fine Chemicals, School of Chemistry and Molecular Engineering, East China University of Science and Technology, Shanghai 200237 Google Scholar More articles by this author , Ajun Tang Key Laboratory for Advanced Materials, Feringa Nobel Prize Scientist Joint Research Center, Frontiers Science Center for Materiobiology and Dynamic Chemistry, Institute of Fine Chemicals, School of Chemistry and Molecular Engineering, East China University of Science and Technology, Shanghai 200237 Google Scholar More articles by this author , Wenjing Hong Key Laboratory of Physical Chemistry of Solid Surfaces, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005 Google Scholar More articles by this author , Dahui Qu *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] Key Laboratory for Advanced Materials, Feringa Nobel Prize Scientist Joint Research Center, Frontiers Science Center for Materiobiology and Dynamic Chemistry, Institute of Fine Chemicals, School of Chemistry and Molecular Engineering, East China University of Science and Technology, Shanghai 200237 Google Scholar More articles by this author , He Tian Key Laboratory for Advanced Materials, Feringa Nobel Prize Scientist Joint Research Center, Frontiers Science Center for Materiobiology and Dynamic Chemistry, Institute of Fine Chemicals, School of Chemistry and Molecular Engineering, East China University of Science and Technology, Shanghai 200237 Google Scholar More articles by this author and Hongxiang Li *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] Key Laboratory for Advanced Materials, Feringa Nobel Prize Scientist Joint Research Center, Frontiers Science Center for Materiobiology and Dynamic Chemistry, Institute of Fine Chemicals, School of Chemistry and Molecular Engineering, East China University of Science and Technology, Shanghai 200237 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.022.202101757 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail To monitor and investigate chemical reactions in real time and in situ is a long-standing, challenging goal in chemistry. Herein, an electric potential-promoted oxidative coupling reaction of organoboron compounds without the addition of base is reported, and the transmetallation process involved is monitored in real time and in situ with the scanning tunneling microscopy break single-molecule junctions (STMBJ) technique. We found that the electric potential applied determined the transmetallation. At low-bias voltage, the first-step transmetallation process occurred and afforded Au─ C-bonded aryl gold intermediates. The electronic properties of organoboron compounds have a strong influence on the transmetallation process, and electron-rich compounds facilitate this transformation. At high-bias voltage, the second-step transmetallation process took place, and the corresponding intermediate (highly reactive diaryl metal complex) was detected with the assistance of Pd(OAc)2. Our work demonstrates the applications of STMBJ on in situ monitoring and catalyzing of chemical reactions and provides a new methodology to fabricate single-molecule devices. Download figure Download PowerPoint Introduction Given the instability and high reactivity of reaction intermediates, cumbersome analytical methods, sophisticated instruments, and complicated sample pretreatment, the convenient monitoring of chemical reactions in real time and in situ is a long-standing challenge. The past half century has witnessed advances in molecular electronics, and many techniques that detect single molecules and visualize molecular behavior in real time and in situ have been developed, such as scanning tunneling microscopy break single molecule junctions (STMBJ)1–3 and graphene-molecule-graphene single-molcule junctions (GMG-SMJ).4–6 These techniques have been developed to measure the electrical properties of single molecules assembled into the gap of electrodes in solution. Since the conductive characteristics of molecules are correlated to their chemical structures, it is convenient to monitor the chemical structural change of individual molecules through these techniques. Moreover, it is reported that chemical reactions can occur in the molecule junctions.7–13 Thus, the single-molecule junction is an ideal platform to obtain real-time information about chemical reactions. For example, Guo and co-workers14 integrated a single-molecule Pd catalyst into GMG-SMJ and achieved the real-time and full description of the Suzuki–Miyaura cross-coupling reaction. With the assistance of STMBJ, Hong and co-workers15 investigated the oriented external electric field-induced selective catalysis in a two-step cascade reaction of the Diels–Alder addition followed by an aromatization process. As a fundamental process in organometallic chemistry,16 transmetallation participates in the progress of many important cross-coupling reactions.17 In general, transmetallation describes the transfer of ligands from main group metals or metalloids (including B and Si) to transition metals.18 Regardless of the long-standing interest in this reaction, the study of transmetallation has not been covered in the same depth as oxidative addition and other organometallic reactions. For example, the electronic influence of organic reagents on transmetallation is not well understood, and contradictory conclusions have been reported.19–22 Organoboron compounds are one of the most diverse classes of easily obtained reagents23 and provide access to a raft of indispensable transformations such as Suzuki–Miyaura coupling, Chan–Evans–Lam coupling,24 oxidative coupling,25 and so on.26 They undergo facile transmetallation with various transition metals in the presence of base. Metal-catalyzed oxidative coupling of arylboronic acids proceeds through two-step transmetallation, which furnishes a fascinating platform to investigate transmetallation.27–30 In this work, an electric potential-promoted oxidative coupling reaction of organoboron compounds is reported, and a real-time and in situ monitoring of the transmetallation involved is achieved with the STMBJ technique (Figure 1). We found that the transmetallation process was controlled by the electric potential applied. At low-bias voltage, the first-step transmetallation process occurred, and the corresponding intermediate, Au–C-bonded compound, formed. The electronic property of organoboron compounds has a strong impact on the first-step transmetallation process. When higher-bias voltage is applied, the second-step transmetallation process was induced, accompanied by the corresponding oxidative coupling products. Our work unveils the applications of STMBJ to real-time in situ monitoring and catalyzing chemical reactions, thereby promoting the advance of molecular electronics. Figure 1 | The processes of electric potential-promoted oxidative coupling reaction of aryl boronic acids. Download figure Download PowerPoint Experimental Methods Synthesis The organoboron compounds were synthesized via reported methods ( Supporting Information Schemes S2–S9). The synthetic procedures and characterizations ( Supporting Information Figures S1–S18) are listed in the Supporting Information. The chemical structures of the compounds studied in this paper are shown in Supporting Information Scheme S1. Single-molecule conductance measurement and data analysis The target molecules were dissolved (0.5 mM) in 1,2,4-trichlorobenzene (TCB) and dropped on the surface of a gold-plate silicon wafer washed by piranha solution. The single-molecule conductance measurements were carried out using the STMBJ technique, and the data were analyzed by XMe open-source code ( https://github.com/Pilab-XMU/XMe_DataAnalysis). Molecular junctions were formed during the stretch of electrodes. The experiments were conducted in air at room temperature. More than 3000 individual traces were recorded without data selection to obtain the most likely conductance values presented as one-dimensional (1D) and two-dimensional (2D) conductance-displacement histograms. Further details are provided in the Supporting Information. Results and Discussion The determination of first-step transmetallation In oxidative coupling, the first transmetallation step produced monoaryl-metal intermediates. We performed the single-molecule conductance measurements of B1 and B2 using STMBJ technique in TCB solution at a bias voltage of 100 mV. Besides the conductance quantum G0, molecular conductance peaks at 10−2.16G0 and 10−2.88G0 were clearly observed for B1 and B2 junctions respectively (Figures 2a–2d and Supporting Information Figures S19 and S20). These conductance peaks were subjected to (a) the direct anchoring of B1 and B2 with a Au tip through boron and oxygen atoms, and/or π systems of phenyl rings or (b) the chemical reactions between organoboron compounds and Au tips, that is, the formation of covalent bonds. It is well known that the boron atom is electron-deficient; it cannot donate electrons to link with the Au tip. And no junctions formed for Py-B (pyridin-4-ylboronic acid) in which the boron atom is more electron-deficient ( Supporting Information Scheme S1 and Figure S26), indicating that the boron atom cannot accept the electrons from the Au tip either. Additionally, control experiments showed that no conductance peaks were observed for C2 (2-(4′-(methylthio)-[1,1′-biphenyl]-4-yl)-1,3-dioxolane) ( Supporting Information Scheme S1 and Figure S23). These results excluded the possibility of direct anchoring of B1 and B2 with Au tips. Figure 2 | (a) 1D conductance histogram of B1 under 100 mV. Inset: the chemical structure of B1. (b) 1D conductance histogram of B2 under 100 mV. Inset: the chemical structure of B2. (c) Individual traces of B1 (orange) and B2 (blue). (d) 2D conductance histogram of B2. Download figure Download PowerPoint When a higher bias voltage was applied, low-G peaks, corresponding to the homocoupling compounds of B1 and B2, were observed (which will be discussed later). We therefore hypothesize that B1 and B2 undergo oxidative coupling reactions, and the conductance peaks around 10−2.0∼10−3.0G0 correspond to molecules binding through a covalent Au—C bond. In zero covalent Au (0) nanoparticles-catalyzed homocoupling reactions, a positively charged site formed by oxygen is a prerequisite.29 Our control experiments showed that when degassed solution of B1 was used in conductance measurement, no junctions formed in the initial period ( Supporting Information Figure S22). Please note that junctions formed immediately for the non-degassed B1 solution. We conclude that oxygen assists in the formation of junctions. It has been reported that Au—C covalent bonds can be generated through cleavage of trimethyltin groups31 and oxidative addition of aryl iodides.32 We measured the conductance of the molecules 2-Sn, I1, and I2, and their conductance values and junction elongation were in good agreement with our organoboron compounds (Figures 3a and 3b and Supporting Information Figures S23 and S24). These results confirm that B1 and B2 undergo a transmetallation process and form covalent Au─ C bonds in the junctions. In addition, for compound 2-Sn, no distinct conductance peak was observed within the first 10 min of the measurement, suggesting that the toxic trimethyltin group is more difficult to cleave. For I1 and I2, they can also bind to the Au tip through the dative interaction of I atoms, resulting in complex conductance signals. To further confirm the occurrence of transmetallation, we synthesized compounds Au1 and Au2 (for their chemical structures, see Supporting Information Scheme S1) by transmetallation of B1 and B2 with Au(I)PPh3Cl. Single-molecule conductance results showed that the conductance of Au1 and Au2 is close to that B1 and B2 ( Supporting Information Figures S23 and S24). Figure 3 | (a and b) 2D and 1D conductance histograms of 2-Sn under 100 mV. (c) 1D conductance histograms of 2-Bpin (violet), 2-Bnep (green), and 2-Bdan (black) under 100 mV. (d) The 1D conductance histogram of 2Py-B (navy) and 2NH2-B (olive) under 100 mV. For 2-PyB, obvious peak is observed after 30 min. (e) Semilogarithmic plot of the conductance value as a function of the number of phenylene units for B1, B2, and 3-Bpin. (f) 1D conductance histograms of B2 at different bias voltages: 200 mV (pink), 100 mV (cyan), −100 mV (royal), and −200 mV (wine). Inset: the JFP under different bias voltage. Download figure Download PowerPoint To more deeply understand the cleavage of the C—B bond and preclude the possibility that the B—O bond breaks to form the B—Au bond, theoretical calculation of the bond cleavage was performed based on the density functional theory calculation as detailed in the Supporting Information. Molecule B1 was chosen as the model to investigate whether the C—B or B—O is broken during the junction formation process by considering the bond energy ( Supporting Information Figure S31). The energy required to break the C—B bond in B1 is 5.27 eV, which is 0.87 eV lower than that of the B—O bond (6.14 eV). This result reveals that the formation of the C—Au bond is more energetically favorable than the B—Au bond, which is consistent with our experimental results mentioned above. The scope of organoboron for first-step transmetallation The success of transmetallation for arylboronic acid encourages us to expand the scope to other types of organoboron derivatives. We found that other types of common organoboron derivatives such as 2-Bpin, 2-Bdan, and 2-Bnep also can form a Au—C bond with a Au tip. 2-Bpin-, 2-Bdan-, and 2-Bnep-based junctions display the same molecular conductance and plateau length (Figure 3c and Supporting Information Figure S25), and these values are in good agreement with B2 junctions. For 2-Bdan, a low-G conductance peak was also observed in the 1D histogram. This peak has a remarkably longer plateau length than the high-G peak, indicating the formation of new molecule junctions. We believe that these junctions correspond to the molecules binding via Au-diaminonaphthalene interactions. Next, we studied the effect of electronic properties on the transmetallation process. For N-B1 ((6-(methylthio) pyridin-3-yl)boronic acid, Supporting Information Scheme S1), an analogue of B1, the junction formation probability (JFP) was much lower compared with B1 ( Supporting Information Figure S26). When the more electron-deficient molecules Py-B and CN-B ((4-cyanophenyl) boronic acid, Supporting Information Scheme S1) were tested, no junctions formed, even after 3 h of measurement ( Supporting Information Figure S26). 2Py-B with one more phenylene unit in the conjugation core than Py-B, displayed a weak conductance peak around 10−2.88G0 after 30 min of measurement (Figure 3d). In contrast, for molecules with electron donating groups, such as I-B ( Supporting Information Scheme S1 and Figure S26), 2NH2-B (Figure 3d), and 3-Bpin ( Supporting Information Scheme S1 and Figure S21), the conductance peaks were observed immediately during measurement. Based on these experimental results, we conclude that the electronic properties of organoboron compounds have a strong impact on transmetallation, and the more electron-deficient the molecules are, the more difficult the transmetallation is. This is because electron-withdrawing groups decrease the nucleophilicity of organoboron compounds and consequently hinder transmetallation. We have known that the Bpin unit undergoes the same reaction with boronic acid. Therefore, the dependence of conductance on molecular length was investigated with compounds B1, B2, and 3-Bpin. We found the conductance values decreased exponentially on the molecular length (Figure 3e), that was G = G0 exp (-β·Lm) where the β value was 1.3 per phenylene unit, which is consistent with the high-G state of oligophenylene terminated by iodide and thiomethyl (1.2 per phenylene unit) as reported by Venkataraman and co-workers.32 The influence of electric potential It is well known that for organoboron-based transmetallation, a base is required to increase the nucleophilicity of organoboron compounds. In our experiments, the transmetallation occurs in the absence of bases. We, therefore, suppose that the electric potential applied plays an important role in this transformation. To substantiate this assumption, we chose compound B2 as a model molecule to perform single-molecule conductance measurements. In the measurements, other experimental conditions remained unchanged, but the bias voltage changed from −200 to 200 mV. This setup allowed us to modulate the chemical potential of the electrode.33 Figure 3f illustrates the 1D conductance histograms under different bias voltages. We can clearly see conductance peaks around 10−2.88G0 in all curves (the slight position shift of conductance peaks is due to the gating effect of bias voltage change34). And the relative intensity of the conductance peak rose with the increase of the bias voltage, demonstrating that the positive bias voltage facilitated transmetallation. Further statistical analysis of JFP enables us to quantify the effect of the electric potential. As the bias voltage increased from −200 to 200 mV, the JFP gradually rose from 20% to 70%. The application of a positive bias voltage makes the Au tip more electron-deficient and therefore increases the electrophilicity of the Au tip, which enables the transmetallation of organoboron compounds without the addition of base. In situ homocoupling reaction At low-bias voltage, the conductance peaks of homocoupling products cannot be observed in a 1D histogram. This is because the Au tip does not have enough electrophilicity for the second-step transmetallation. When we increased the bias voltage to 400 mV, a conductance peak at 10−4.66G0 was observed 5 min later in the conductance histogram of B2 (Figures 4a and 4c). We assumed this peak corresponded to the homocoupling product of B2. To verify this assumption, we synthesized S4 (the homocoupling product of B2) ex situ, and compared its molecular conductance and plateau length with those of the low-G peak observed for B2 at high-bias voltage. The junctions of S4 exhibited a clear conductance peak at the same value as that for the low-G peaks of B2 junctions (Figure 4a). And the plateau length determined in the 2D histogram for S4 was essentially the same as that for the low-G peak of B2 junctions too (Figure 4b). It proved the formation of S4 in situ and the occurrence of homocoupling reaction at high-bias voltage. And then we turned to expand the scope of this reaction. We found arylboronic acid without electron-withdrawing substituents, such as B1, 2NH2-B, and I-B, can undergo the homocoupling reaction under higher-bias voltage ( Supporting Information Figures S27 and S28). Figure 4 | (a) 2D conductance histograms of B2 under 400 mV. (b) 2D conductance histograms of S4 synthesized ex situ under 400 mV. (c) 1D conductance histograms of B2 and S4 synthesized ex situ under 400 mV. Inset: chemical structure of S4. (d) Reaction kinetics of the homocoupling reaction of B2 in single-molecule junctions under different bias voltages. Inset: the JFP of product under different bias voltage 100 mV (cyan), 200 mV (green), 300 mV (blue), and 400 mV (violet). (e) The reaction procedure of electric potential-promoted oxidative coupling reaction. Download figure Download PowerPoint To quantify the influence of electric potential, the JFP values of S4 formed in situ under different bias voltages were calculated. As bias voltage increased from 100 to 400 mV,the JFP gradually rose from 12% to 68%, indicating that oxidative coupling reaction accelerated with the increase of electrical potential. Then the correlation between the proportion of homocoupling product and measurement time was investigated. The proportion of homocoupling product was obtained by counting the single traces at different periods. As shown in Figure 4d, the yield of homocoupling product significantly increased with the rise of bias voltage. These results reveal the dependence of reaction rate on electric potential. The determination of second-step transmetallation The formation of homocoupling products confirms the occurrence of second-step transmetallation. Unfortunately, it is impossible to directly monitor the second-step transmetallation intermediate, the diaryl-gold complex, because it displays the same conductance peak as the first-step transmetallation intermediate. It is well known that Pd (II) catalyzes the homocoupling reaction of organoboron via the similar mechanism.35 Hence, by using B1 as an example, we monitored the second-step transmetallation with the assistance of Pd(OAc)2 (Figures 5a and 5b). For the B1 solution containing Pd (OAc)2, an additional new conductance peak appeared at 10−3.7G0, this peak is denoted as LC. After 40 min measurement, another new conductance peak (denoted as MC) around 10−3.06G0 was observed. The conductance value and plateau length of MC peak were nearly the same as those of S2 (the homocoupling product of B1, Supporting Information Figure S28), indicating the occurrence of the Pd(OAc)2-catalyzed homocoupling reaction. The relative intensities of the conductance peak of B1 (HC), MC, and LC peaks changed along with measurement; that is, the relative intensity of the MC peak increased and the relative intensity of the HC and LC peaks decreased. After 140 min, the MC peak was much higher than the HC peak, and the LC peak nearly disappeared. The decrease of the LC peak along with measurement indicated that it was the conductance peak of an active intermediate. From the mechanism of the Pd-catalyzed oxidative coupling reaction ( Supporting Information Figure S30), we know that only the second-step transmetallation intermediate and the homocoupling product can form junctions. Combined with the high reactivity of the LC peak-related compound, we conclude that the LC peak is attributed to the second-step transmetallation intermediate, the diaryl palladium complex. Figure 5 | (a) The evolution of 1D conductance histograms of the Pd-catalyzed oxidative coupling of B1 under 200 mV with time, except the first histogram of 5 min. Other histograms are constructed with the same amounts of single traces. (b) Single trace with all three conductance states. Download figure Download PowerPoint Based on the above results and previous reports in the literature,28,29 we conclude that our electric potential-promoted oxidative coupling reaction undergoes the following procedure (Figure 4e). First, the Au tip readily absorbs O2 and generates a supero-like intermediate. Subsequent transmetallation between aryl boronic acids and a positively charged Au site occurs to furnish the first-step transmetallation intermediate. Due to the electron-donating carbon ligands, the second-step transmetallation is blocked under low-bias voltage. When higher-bias voltage is applied, the electrophilicity of the Au tip increases, and the second-step transmetallation intermediate thus forms. Finally, the reductive elimination affords the homocoupling product. Conclusion We have reported an electric potential-promoted oxidative coupling reaction of organoboron compounds and conducted real-time and in situ monitoring of the transmetallation involved. We find that the coupling reaction can be regulated via the electric potential applied. At low-bias voltage, the first step of the transmetallation process occurs and a Au─ C bonded transmetallation intermediate forms. When higher-bias voltage is applied, the second-step transmetallation process takes place, and the homocoupling product is observed. Additionally, the reaction scope and the effect of the electronic properties of organoboron compounds are investigated. This work provides a new platform for real-time and in situ monitoring of chemical transformations and offers exciting opportunities to probe the properties of transition states, which is important for investigating and understanding fundamental chemical processes. Morevoer, the in situ formation of the Au—C bond opens a new door for self-integrating molecular circuits. Supporting Information Supporting Information is available and includes detailed synthetic procedures and characterization data of compounds, 1D and 2D conductance histogarms, and theoretical data. Conflict of Interest There is no conflict of interest to report. Acknowledgments This work was supported by the National Natural Science Foundation of China (grant nos. 21875279, 21790362, and 22075080), the Shanghai Municipal Science and Technology Major Project (grant no. 2018SHZDZX03), the Fundamental Research Funds for the Central Universities, the Programme of Introducing Talents of Discipline to Universities (grant no. B16017), and the Program of Shanghai Academic/Technology Research Leader (grant no. 19XD1421100).