可见光氧化还原催化具有反应条件温和、催化剂用量较少、官能团耐受性较好、反应物适用范围广、反应器易于设计等优势，过去的十几年内在有机化学领域得到了蓬勃发展。

β-胺基酮在天然产物和药物合成方面具有极其重要的价值。Mannich反应是合成β-胺基酮的重要方法，但仍然存在局限性：无法合成–NH₂未被修饰的β-氨基酮或–NH₂连在β-季碳上的酮。硝基作为氨基的前体，合成β-硝基酮就可以很容易解决这个问题。

论文的主要研究内容是在可见光诱导下由酮的烯醇三甲基硅醚与偕溴代硝基化合物高效制备β-硝基酮。以fac-Ir(ppy)₃为可见光催化剂、乙腈为溶剂，在蓝光LED 灯照射的温和条件下，以中等至优秀的收率制备了β-硝基酮。为了探索新合成方法的普适性，我们分别尝试了以酮肟和硝基烷烃为原料制备的对称或不对称、环状或非环状、四级或三级/二级偕溴代硝基化合物，它们都表现出了良好的反应性。以烷基芳基酮、烯基烷基酮或简单酮为原料制备出相应的烯醇硅醚，它们同样都表现出了良好的光反应活性。通过条件控制实验、自由基捕获实验、荧光猝灭实验以及¹H NMR监测实验，证实了我们新方法的反应机理。对照实验表明，可见光和光氧化还原催化剂都是偶联反应所必需的。自由基捕获实验表明，α-硝基烷基自由基是可能的中间体。¹H NMR监测2-溴-2-硝基丙烷与1-苯基-1-三甲基硅氧基乙烯的反应清楚地表明，三甲基溴硅烷是另一产物。由Stern–Volmer猝灭图的分析可知，偶联反应过程中发生了激发态*fac-Ir(ppy)₃向偕溴代硝基化合物的单电子转移，而*fac-Ir(ppy)₃与烯醇三甲基硅醚之间没有单电子转移过程。基于上述事实，我们提出了合理的反应机理。我们的新方法具有原料容易制备、操作简单、产物易分离、普适性好等优点，而且还可以直接利用太阳光进行反应，是一种合成β-硝基酮的高效、快捷的方法。β-硝基酮很容易转化为β-NH₂酮、1, 3-氨基醇、α, β-不饱和酮、β-氰基酮和γ-硝基酮。

论文的另一部分研究内容是利用可见光氧化还原催化制备3-噁唑啉-5-酮。在溶剂乙腈中，以N-羟基邻苯二甲酰亚胺的氧化还原活性酯和二级硝基化合物为原料、TBD作为碱、fac-Ir(ppy)₃作为可见光催化剂，在蓝光LED灯照射下高效制备出3-噁唑啉-5-酮。此项研究中我的主要工作是：合成了氧化还原活性酯、硝基化合物和噁唑啉酮氮氧化物。我们推测反应存在3-噁唑啉-5-酮类氮氧化物中间体，通过条件控制实验证明了其脱氧过程，成功地合成了3-噁唑啉-5-酮，并验证了其实用性。

Visible light photoredox catalysis is characteristic of mild reaction conditions, low catalyst loading, good functional group tolerance, wide scope of reactants and facile set-ups. In the past ten years, it has experienced a rapid development in the field of organic chemistry.

β-Amino ketones are invaluable for the synthesis of natural products and pharmaceutical agents. Mannich reaction is an important method for the synthesis of β-amino ketones, but still has some limits: it is inaccessible to a ketone bearing a nude β-NH₂ group or a β-quaternary carbon with nude -NH₂ group. Accomplishment of β-nitro ketone syntheis can easily solve this problem since nitro group is a precursor to amino group.

The major part of my thesis work is visible light induced efficient synthesis of β-nitro ketones from TMS enol ethers of ketones and geminal bromonitroalkanes. In the presence of fac-Ir(ppy)₃, β-nitro ketones were prepared in CH₃CN at rt in moderate to high yields under blue LEDs irradiation. In order to explore the universality of our new method, symmetrical or asymmetrical, cyclic or acyclic, quaternary or tertiary/secondary gem-bromonitroalkanes were examined, which were prepared from ketoximes or nitroalkanes and all exhibited good reactivities. The TMS enol ethers of alkyl aryl ketones, alkenyl alkyl ketones or unfunctionalized ketones were prepared and exhibited high reactivities as well. We demonstrated the reaction mechanism through control experiments, radical-trapping, phosphorescence quenching and ¹H NMR monitoring experiments. Control experiments verified that both visible light and the photoredox catalyst were necessary for the coupling reaction to proceed. Radical-trapping experiments indicated that α-nitroalkyl radicals were possible intermediates. ¹H NMR monitoring the reaction between 2-bromo-2-nitropropane and 1-phenyl-1-trimethylsiloxyethylene clearly showed that trimethylsilyl bromide was the by-product. Stern–Volmer quenching analysis supported that single electron transfer from photoexcited fac-Ir(ppy)₃ to gem-bromonitroalkane should take place during the coupling reactions, and no such process was involved between fac-Ir(ppy)₃ and TMS enol ether. Based on the above facts, we proposed a reasonable reaction mechanism. Our new method is an efficient and mild way to synthesize β-nitro ketones due to the following advantages: good availability of starting materials, ease of operation and separation, wide scope of substrates. Further, the reaction can proceed smoothly under irradiation of sunlight. These β-nitro ketones are easily converted to β-NH₂ ketones, 1,3-amino alcohols, α, β-unsaturated ketones, β-cyanoketones and γ-nitroketones.

Another part of my thesis work is the preparation of 3-oxazolin-5-ones under visible light photoredox catalysis. Using redox-active esters and secondary nitro compounds as the starting materials, TBD as the base, fac-Ir(ppy)₃ as the catalyst and CH₃CN as the solvent, 3-oxazolin-5-ones were obtained under irradiation of blue LEDs. In this section, my work focused on the synthesis of redox active esters, nitro compounds and 3-oxazolin-5-one-N-oxides. We speculated the presence of 3-oxazolin-5-one-N-oxdies in the photocatalytic processes and verified the deoxygenation process through control experiments. We prepared 3-oxazolin-5-ones successfully and proved the practicability of these compounds.