Radiative heat transfer across nanoscale-thick layers attracts considerable attention because of its importance for modern technology, and also because of the evidence that conventional methods of radiative heat transfer fail at such small scales. This paper analyzes several approaches to this problem that have been proposed since the late 1960s when the first adaptations of the classical Stefan–Boltzmann law of radiative heat transfer to smaller scales were presented. It is shown that while all authors agree that thermal radiation is described by Maxwell’s equations for electrodynamics, the methods of these studies drifted from deductive reasoning based on these equations toward heuristic guesses. This paper identifies critical points of several of these previous studies that are responsible for the lack of a suitable theory of radiative heat transfer across nanoscale layers despite almost 50 years of effort. Among these points are: (1) attempts to describe heat transfer using statistical distributions that are limited to equilibrium systems that cannot produce any heat flux; (2) application of the so-called fluctuation–dissipation theorem when its conditions are not satisfied; (3) the failure to distinguish between different kinds of evanescent fields; (4) an unjustified assumption that resonant surface waves can transfer heat by tunneling.

跨纳米级厚层的辐射传热引起了极大的关注，因为它对现代技术非常重要，并且还因为有证据表明常规的辐射传热方法在如此小的规模下会失败。本文分析了自1960年代末提出的解决该问题的几种方法，当时经典的Stefan-Boltzmann辐射热传递定律首次进行了改编。结果表明，尽管所有作者都同意麦克斯韦的电动力学方程式描述了热辐射，但这些研究方法已从基于这些方程式的演绎推理转向启发式猜测。尽管有近50年的努力，但本文确定了其中一些先前研究的关键点，这些关键点导致缺乏合适的跨纳米层辐射热传递的理论。这些要点包括：（1）尝试使用统计分布来描述传热，而统计分布仅限于不能产生任何热通量的平衡系统；（2）在不满足其条件的情况下应用所谓的波动耗散定理；（3）无法区分不同类型的渐逝场；（4）一个不合理的假设，即谐振表面波可以通过隧穿传递热量。（1）尝试使用统计分布来描述传热，而统计分布仅限于不能产生任何热通量的平衡系统；（2）在不满足其条件的情况下应用所谓的波动耗散定理；（3）无法区分不同类型的渐逝场；（4）一个不合理的假设，即谐振表面波可以通过隧穿传递热量。（1）尝试使用统计分布来描述传热，而统计分布仅限于不能产生任何热通量的平衡系统；（2）在不满足其条件的情况下应用所谓的波动耗散定理；（3）无法区分不同类型的渐逝场；（4）一个不合理的假设，即谐振表面波可以通过隧穿传递热量。