Non-reciprocal propagation versus non-reciprocal control
Non-reciprocal propagation versus non-reciprocal control
复制标题
非互易传播与非互易控制
DOI：
10.1038/s41566-020-00723-5
复制
发表时间：
2020
期刊：
影响因子：
35
通讯作者：
Khurgin, Jacob B.
中科院分区：
文献类型：
--
作者：
Khurgin, Jacob B.
致编辑——最近的一篇文章 1 报道了一种非线性、不可逆器件及其在信号隔离和路由方面的应用。文章最后明确指出，该设备仅对时间不一致的前向和后向信号启用非互易传播。当信号一致时，传输是完全互易的并且不会发生隔离。尽管如此，随附的新闻和观点文章3的标题不低于“硅中的低损耗非线性光学隔离器”，尽管文章正文包含有关上述限制的声明，但该设备在整篇文章中都被称为“隔离器”。在我看来，这对读者来说没有多大帮助，尤其是那些不经意的读者，他们可能会产生光隔离器已经开发出来的印象，而这种印象与事实相差甚远。随着许多非线性“隔离器”方案的激增，“什么是光隔离器，什么不是光隔离器”的问题已在同名评论文章4中得到解决，其中指出：“找到一种可以将功率从一侧传输到另一侧的状态以及另一种不能反向传输功率的状态是不够的。要使设备成为隔离器，它必须阻止或转移所有可能的状态以进行反向传播。”非线性方案1及其前身5都可以一般地呈现为光放大器、非线性开关元件和补偿增益的衰减器的序列，如图1所示。显然，前向信号首先被放大（或集中），并且能够将开关转至ON状态，而后向信号首先被衰减，并且在没有前向信号的情况下，开关保持在OFF状态。但无论开关在给定时间处于接通还是断开状态，对于前向和后向信号来说，它都处于相同的状态，即该方案在信号传播方面是完全可逆的。只是在两侧的可控性方面是不可逆的。它显然不是光隔离器，在我看来，它的正确名称应该是“非互易控制”设备。在这样的设备中，前向信号公平地控制前向信号和后向信号的传播，而后向信号不控制任一信号的传播。为了避免进一步混淆，应该指出的是，尽管非线性方案1及其许多祖先都未能满足光学隔离器的严格标准，但这并不意味着不能使用非线性光学（例如通过频率转换）来构建真正的光学隔离器6。总之，虽然不可逆控制器件不是光学隔离器，但它仍然是一项非常重要的成就，它展示了纳米光子设计和制造能力的极大提高，并且肯定会在实现硅兼容有源纳米光子电路的其他现有和尚未构想的功能方面找到应用。❐
To the Editor—A recent article1 reports a nonlinear, non-reciprocal device and its application to isolation and routing of signals. Towards the end, the article clearly specifies that the device enables non-reciprocal propagation only for the forward and backward signals that are not coincident in time2. When the signals are coincident, the transmission is fully reciprocal and no isolation takes place. Nevertheless, the accompanying News and Views article3 bears the title no less than ‘Low-loss nonlinear optical isolators in silicon’, and although the body of the article contains a statement about the aforementioned limitation, the device is referred to as an ‘isolator’throughout the article. This, in my view, is not very helpful to readers, especially casual ones who may get an impression that an optical isolator has been developed, and this impression cannot be further from the truth. As many nonlinear ‘isolator’schemes have proliferated, the issue of ‘What is—and what is not—an optical isolator’has been addressed in the namesake Commentary article4 stating:“It is insufficient to find a state in which power can be transmitted from one side to the other and another state in which the power is not transmitted in the reverse direction. For a device to be an isolator it must block or divert all possible states for backward propagation.” The nonlinear scheme1 and its predecessors5 can all be generically rendered as a sequence of an optical amplifier, a nonlinear switching element and an attenuator compensating the gain as shown in Fig. 1. Clearly the forward signal gets first amplified (or concentrated) and is capable of turning the switch to the ON state, while the backward signal is first attenuated and in the absence of a forward signal the switch remains in the OFF state. But whether the switch is in the ON or OFF state at a given time, it is in the same state for both forward and backward signals, that is, the scheme is fully reciprocal in terms of signal propagation. It is only non-reciprocal in terms of controllability from two sides. It is clearly not an optical isolator, and, in my opinion, a proper name for it should be a ‘non-reciprocally-controlled’device. In such a device the forward signal equitably controls the propagation of both the forward and backward signals, while the backward signal controls propagation of neither one. To avoid further confusion, it should be noted that although the nonlinear scheme1 as well as its many progenitors all fail to meet strict criteria for being an optical isolator, it does not mean that a true optical isolator cannot be built using nonlinear optics, for example, by frequency conversion6. In conclusion, while the non-reciprocally-controlled device is not an optical isolator it is still a very important accomplishment that demonstrates the greatly improved capabilities of nanophotonic design and fabrication, and will certainly find applications in enabling other existing and yet-to-be conceived functionalities of silicon-compatible active nanophotonic circuits.❐
影响因子：
35
作者：
Jalas, Dirk;Petrov, Alexander;Renner, Hagen
通讯作者：
Renner, Hagen