Triplet pathways in diarylethene photochromism: Photophysical and computational study of dyads containing ruthenium(II) polypyridine and 1,2-bis(2-methylbenzothiophene-3-yl)maleimide units
Triplet pathways in diarylethene photochromism: Photophysical and computational study of dyads containing ruthenium(II) polypyridine and 1,2-bis(2-methylbenzothiophene-3-yl)maleimide units
复制标题
DOI：
10.1021/ja711173z
复制
发表时间：
2008-06-11
影响因子：
15
通讯作者：
Scandola, Franco
中科院分区：
文献类型：
--
作者：
Indelli, Maria Teresa;Carli, Stefano;Scandola, Franco
合成了 1,2-双（2-甲基苯并噻吩-3-基）马来酰亚胺模型（DAE）和两个二元体，其中该光致变色单元通过直接氮碳键（Ru-DAE）或通过插入的亚甲基（Ru-CH2-DAE）与钌聚吡啶发色团偶联。这些系统的光化学和光物理学已经通过固定和时间分辨（纳秒和飞秒）光谱方法的组合在乙腈中得到了彻底的表征。二芳基乙烯模型 DAE 在 448 nm 激发下发生光环化，稳定状态下光转化率为 35%。脱气后量子产率从 0.22 增加到 0.33。光化学环化回复（量子产率，0.51）可以在 lambda > 500 nm 激发时完成。光环化既可以从激发单线态 (Si) 发生，作为超快（约 0.5 ps）过程，也可以从三线态 (T-1) 在微秒时间尺度内发生。在Ru-DAE和Ru-CH2-DAE二元组中，钌发色团吸收光后发生有效的光环化，并具有氧敏感的量子产率（在脱气和充气溶液中分别为0.44和0.22）。光转换效率几乎是单一的（90%），比单独的光致变色 DAE 高得多。观察到基于 Ru 的 MLCT 磷光和 DAE 荧光的有效猝灭。通过皮纳秒时间分辨光谱法获得了完整的动力学表征。除了快速光环化（0.5 ps）之外，从激发的二芳基乙烯到 Ru(II) 发色团还发生快速单线态能量转移（Ru-DAE 中为 30 ps，Ru-CH2-DAE 中为 150 ps）。在 Ru(II) 发色团中，迅速系间窜越至 MLCT 三重态，随后三重态能量转移至二芳基乙烯（Ru-DAE 中为 1.5 ns，Ru-CH2-DAE 中为 40 ns）。二芳基乙烯部分的三线态在微秒时间内发生环化。实验结果得到了从头算和 DFT 组合计算研究的补充，其中沿着光环化/环化回复的反应坐标研究了二芳基乙烯的基态 (S-0) 和最低三重态 (T-1) 的势能面 (PES)。在 DFT 理论水平上，So 和 T-1 上的过渡态结构相似，并且沿着反应坐标倾向于闭环形式。在过渡态几何结构中，S-0 和 T-1 PES 几乎是简并的。而在 S-0 上，一个大的势垒（约 45 kcal mol(-1)）将开环和闭环最小值分开，而在 T-1 上，异构化势垒是适度的，环化势垒（约 8 kcal mol(-1)）小于环化势垒（约 14 kcal mol(-1)）。这些特征解释了用 Ru-DAE 观察到的高效敏化光环化和低效敏化环化回复。三重态环化被视为一种非绝热过程，起始于开环几何的 T-1，通过过渡态几何的系间交叉进行，并在闭环几何的 S-0 上完成。对原型模型 1,2-双（3-噻吩基）乙烯的计算研究用于将 DFT 结果与从头算 CASSCF//CASPT2 结果进行基准比较，并证明 DAE 获得的 S-0 和 T-1 PES 主要拓扑特征的普遍性。总之，这些结果为光致变色二芳基乙烯的光环化中的三重态途径提供了强有力的实验证据和理论依据。
A 1,2-bis(2-methylbenzothiophene-3-yl)maleimide model (DAE) and two dyads in which this photochromic unit is coupled, via a direct nitrogen-carbon bond (Ru-DAE) or through an intervening methylene group (Ru-CH2-DAE), to a ruthenium polypyridine chromophore have been synthesized. The photochemistry and photophysics of these systems have been thoroughly characterized in acetonitrile by a combination of stationary and time-resolved (nano- and femtosecond) spectroscopic methods. The diarylethene model DAE undergoes photocyclization by excitation at 448 nm, with 35% photoconversion at stationary state. The quantum yield increases from 0.22 to 0.33 upon deaeration. Photochemical cycloreversion (quantum yield, 0.51) can be carried out to completion upon excitation at lambda > 500 nm. Photocyclization takes place both from the excited singlet state (Si), as an ultrafast (ca. 0.5 ps) process, and from the triplet state (T-1) in the microsecond time scale. In Ru-DAE and Ru-CH2-DAE dyads, efficient photocyclization following light absorption by the ruthenium chromophore occurs with oxygen-sensitive quantum yield (0.44 and 0.22, in deaerated and aerated solution, respectively). The photoconversion efficiency is almost unitary (90%), much higher than for the photochromic DAE alone. Efficient quenching of both Ru-based MLCT phosphorescence and DAE fluorescence is observed. A complet e kinetic characterization has been obtained by ps-ns time-resolved spectroscopy. Besides prompt photocyclization (0.5 ps), fast singlet energy transfer takes place from the excited diarylethene to the Ru(II) chromophore (30 ps in Ru-DAE, 150 ps in Ru-CH2-DAE). In the Ru(II) chromophore, prompt intersystem crossing to the MLCT triplet state is followed by triplet energy transfer to the diarylethene (1.5 ns in Ru-DAE, 40 ns in Ru-CH2-DAE). The triplet state of the diarylethene moiety undergoes cyclization in a microsecond time scale. The experimental results are complemented with a combined ab initio and DFT computational study whereby the potential energy surfaces (PES) for ground state (S-0) and lowest triplet state (T-1) of the diarylethene are investigated along the reaction coordinate for photocyclization/cycloreversion. At the DFT level of theory, the transition-state structures on So and T-1 are similar and lean, along the reaction coordinate, toward the closed-ring form. At the transition-state geometry, the S-0 and T-1 PES are almost degenerate. Whereas on S-0 a large barrier (ca. 45 kcal mol(-1)) separates the open- and closed-ring minima, on T-1 the barriers to isomerization are modest, cyclization barrier (ca. 8 kcal mol(-1)) being smaller than cycloreversion barrier (ca. 14 kcal mol(-1)). These features account for the efficient sensitized photocyclization and inefficient sensitized cycloreversion observed with Ru-DAE. Triplet cyclization is viewed as a nonadiabatic process originating on T-1 at open-ring geometry, proceeding via intersystem crossing at transition-state geometry, and completing on S-0 at close-dring geometry. A computational study of the prototypical model 1,2-bis(3-thienyl)ethene is used to benchmark DFT results against ab initio CASSCF//CASPT2 results and to demonstrate the generality of the main topological features of the S-0 and T-1 PES obtained for DAE.Altogether, the results provide strong experimental evidence and theoretical rationale for the triplet pathway in the photocylization of photochromic diarylethenes.