In most applications in industry involving catalysis, heterogeneous catalysts are preferred over their homogeneous counterparts when available. However, the former cannot be easily made as selective for complex chemical conversions as the latter. Selectivity requires control of the structural and electronic properties of the catalytic sites at a molecular level, and that is in general quite difficult to accomplish in solids. Nevertheless, a number of novel approaches have been advanced in recent years toward that goal. In this brief Perspective we provide a personal view of some of the avenues available to make highly selective heterogeneous catalysts. First, we discuss the possibility of tethering or otherwise immobilizing homogeneous catalysts on solid surfaces. Alternatively, molecular complexity can be added to solid surfaces via the adsorption of discrete modifiers. Isolated catalytic sites with unique molecular characteristics can also be created during or after the synthesis of heterogeneous catalysts. The structure of the exposed facets of solids can themselves be exploited to control catalytic performance. Finally, complex nanostructures such as Janus and core-shell nanoparticles can be synthesized and employed as scaffolds for multiple catalytic functionalities. A few examples of all of these methodologies, mainly from our own laboratory, are provided, and a discussion of the pros and cons of each approach is provided.

在涉及催化的工业中的大多数应用中，多相催化剂在可用时优于其均相催化剂。然而，前者不容易像后者那样对复杂的化学转化具有选择性。选择性需要在分子水平上控制催化位点的结构和电子特性，这通常很难在固体中实现。尽管如此，近年来已经为实现这一目标提出了许多新颖的方法。在这个简短的观点中，我们提供了一些可用于制造高选择性多相催化剂的途径的个人观点。首先，我们讨论将均相催化剂束缚或以其他方式固定在固体表面上的可能性。或者，分子复杂性可以通过吸附离散改性剂添加到固体表面。在多相催化剂合成期间或之后，也可以创建具有独特分子特征的孤立催化位点。固体暴露面的结构本身可以用来控制催化性能。最后，可以合成复杂的纳米结构，如 Janus 和核壳纳米粒子，并将其用作多种催化功能的支架。提供了所有这些方法的几个例子，主要来自我们自己的实验室，并提供了对每种方法的利弊的讨论。固体暴露面的结构本身可以用来控制催化性能。最后，可以合成复杂的纳米结构，如 Janus 和核壳纳米粒子，并将其用作多种催化功能的支架。提供了所有这些方法的几个例子，主要来自我们自己的实验室，并提供了对每种方法的利弊的讨论。固体暴露面的结构本身可以用来控制催化性能。最后，可以合成复杂的纳米结构，如 Janus 和核壳纳米粒子，并将其用作多种催化功能的支架。提供了所有这些方法的几个例子，主要来自我们自己的实验室，并提供了对每种方法的利弊的讨论。