五甲基茂取代硼陽離子的合成與反應性

Abstract

硼化合物在有空的P軌域的時候通常會表現出路易士酸性，有的人就想推展酸性的界線看看會部會有什麼新奇的反應性，除了在取代基上氟化，還有讓分子帶正電。我們用可以提供彈性配位數的五甲基茂來建構正二價的硼化合物。除了零碎的反應性測試，最多的篇幅聚焦在氫負離子和硼二價陽離子反應時並沒有攻擊硼原子，而反應五甲基茂的反面，經過結構的重排最後形成攤平的硼苯化合物。但我們發現兩個正電沒有帶來預期的路易士酸性，反而犧牲了溶解度和合成難易度。帶有五甲基茂的一價硼陽離子溶解度有明顯的改善，它可以催化醛的醚化或酮的矽氫化及脫氧反應，在苯甲醛的例子中萬分之一的硼陽離子就可以很有效的在半小時內完成催化。而二苯基酮的還原反應則連續做了兩個當量，把氧原子直接拔除，產生二苯基甲烷。其他不同的酮類則隨取代基的推拉電子性質改變得到的產物，缺電子的酮傾向轉變成矽氫化產物，電子較豐沛者則傾向脫氧產物。我們同時提出了簡單的反應機制佐以幾個間接的實驗證據，為之後新催化劑的分子設計提供了一定的基礎。

To increase the Lewis acidity of cationic boron compounds, we tried to obtain a reactive boron dication. We hope to overcome the chemical inertness of those reported di- or tri-cationic species by choosing the substituents carefully. In our molecular design, Cp* acted as a flexible substituent which might bring us some interesting reactivity. On the other hand, the neutral NHC served as a potent sigma-donating ligand. Fortunately, Cp* behaved as expected to form the eta5-Cp*B fragment and eventually led us to the boron dication [Cp*-B-IMes][AlCl4]2 ([3]2+). After the synthesis of [3]2+, we sought to answer two important questions: Is it reactive enough? Is it more acidic than mono-cations? Acidity of [3]2+ was evaluated by CV and calculated hydride ion affinity (HIA). The accumulation of positive charges on the Cp*-substituted boron dication indeed increases the electron deficiency of the system. However, the Lewis acidity of [3]2+ did not display the same trend. The estimated HIA values of the boron dication did not exceed that of the perfluorinated borane or boreniums. Although [3]2+ was not that acidic as expected, it displayed diverse reactivity toward different reagents that has never been reported for other multi-cationic boron derivatives. CN- and N3- tended to reopen the eta5 Cp*-B cluster by attacking the boron center. However, the reaction with [H-BEt3]- led to the planarized C5B ring of borabenzene (6). This significant structural transformation drew our attention, and we sought to clarify the mechanism of this hydride-induced transformation. The reaction mechanism of the Li[HBEt3] induced transformation of [3]2+ to 6 has been investigated. Although the boron center was predicted to be a stronger Lewis acidic site, steric crowdedness around the boron atom forces the relative bulky [HBEt3]- to attack the more exposed carbon atom of Cp*. The resulting borabicycle[2.1.1]hex-2-ene borenium ([7-CH]+) was identified and characterized using solution NMR spectroscopic methods at 243 K. Subsequent skeletal rearrangement of [7-CH]+ to a planar boracyclohexa-2,5-diene borenium ([9]+) was observed upon raising the temperature. The [7-CH]+ to [9]+ transformation process follows a first-order kinetic with a energy barrier of 20.2 kcal/mol at 298 K and a KIE value of 0.87. A sequential pericyclic reactions was identified for the rearrangement of [7-CH]+ using DFT method. [7-CH]+ undergoes a [1,3]-sigmatropic migration of the B-C1 bond, then an electrocyclic CBC ring opening, and then a 1,2-hydrogen migration to generate [9]+, which can then be transformed into borabenzene 6 via deprotonation. In addition to dicationic system, we have recently developed the related mono-cationic molecules. Two Cp*-coordinated boron mono-cations have been successfully synthesized. The mesityl substituted one ([Cp*-B-Mes][B(C6F5)4], [11]+) showed excellent solubility in DCM which facilitates the later on reactivity studies of [11]+. We demonstrated that even at low catalyst loading (0.01 mol%), [11]+ was able to complete the etherification of PhCHO. Furthermore, [11]+-catalyzed hydrosilylation and deoxygenation of various ketones suggested that the selectivity of the products was potentially attributed to the stability of the carbenium intermediate. Although we could not rule out the possible influence of Brønsted–Lowry acid in the etherification of PhCHO, we were able to confirm that [11]+ was the actual catalyst responsible for the deoxygenation of benzophenone. Encouraged by these observations, preliminary research on hydrosilylation of other unsaturated bonds including alkyne, alkene, amide and nitrile or defluorination of C-F bonds were conducted. Unfortunately, no positive result was obtained thus far. Optimization of reaction condition and tuning the stereoelectronic property of the catalyst might be required for further investigation.