创新背景

尽管激光的发明带来了大量的应用：从外科手术工具到条形码扫描仪再到精密蚀刻，但光学研究人员有一个必须克服的持久限制。相干的、单波长的定向光是激光的一个决定性特征，随着激光腔的增大，这种光开始分解。标准的解决方法是使用外部机制，如波导，来放大光束。

自从1960年第一台激光器被制造出来以来，提高单模激光器的尺寸和功率一直是光学领域的挑战。

创新过程

伯克利的工程师们创造了一种新型半导体激光器，它实现了光学领域一个难以实现的目标:发射单模光的能力，同时保持放大尺寸和功率的能力。

研究团队展示了一种具有均匀间距和大小相同的孔穿孔的半导体膜作为一个完美的、可伸缩的激光腔。他们证明，无论腔体的大小，激光都能发出一致的单一波长。这项发明被称为伯克利表面发射激光器(BerkSELs)。

伯克利表面发射激光器（BerkSEL）的扫描电子显微照片的顶视图。六方晶格光子晶体（PhC）形成电磁腔。

要使用另一种介质来放大激光会占用大量空间，因此，通过消除外部放大的需要，研究人员缩小计算机芯片和其他依赖激光的组件的尺寸，提高效率。

这项研究的结果与垂直腔面发射激光器(VCSELs)尤其相关，在这种激光器中，激光垂直地从芯片中发射出去。这种激光器被广泛应用，包括光纤通信、计算机鼠标、激光打印机和生物识别系统。

VCSEL通常很小，只有几微米宽。目前用来提高其功率的策略是将数百个独立的VCSEL聚集在一起。因为激光是独立的，它们的相位和波长不同，所以它们的功率不能一致地结合。在面部识别等应用中，这是可以容忍的，但在通信或外科手术等精度至关重要的应用中，这是不可接受的。

多模激光不仅会降低效率，甚至会产生反效果。BerkSELs中的单模激光这比现有激光器的效率高得多。

伯克利表面发射激光器（BerkSEL）示意图，说明了泵浦光束（蓝色）和激光束（红色）。半导体膜的非常规设计使所有晶胞（或谐振器）同步相位，以便它们都参与激光模式。

研究发现，BerkSELs的设计能够实现单模光发射，因为光通过膜上的孔的物理特性，膜是一层200纳米厚的铟镓砷化磷化物，一种通常用于光纤和电信技术的半导体。这些孔是用光刻技术蚀刻出来的，必须有固定的大小、形状和间距。

研究人员表示，膜上的周期性孔变成了狄拉克点，这是基于能量线性色散的二维材料的拓扑特征。它们是以英国物理学家、诺贝尔奖得主保罗·狄拉克的名字命名的，他因对量子力学和量子电动力学的早期贡献而闻名。

显示“狄拉克锥体”的示意图。由于狄拉克点奇点，光从整个半导体腔同步发射。

研究人员指出，从一个点传播到另一个点的光的相位等于折射率乘以传播的距离。由于在狄拉克点的折射率为零，从半导体的不同部分发出的光完全处于相位，因此在光学上是相同的。研究中的薄膜大约有3000个孔，但从理论上讲，它可能有100万个或10亿个孔，结果应该是一样的。

创新关键点

研究人员使用高能脉冲激光光泵并为BerkSELs设备提供能量。他们使用近红外光谱优化的共聚焦显微镜测量每个孔径的发射。

创新价值

这项新发明意味着尺寸不必以牺牲相干性为代价，它使激光更强大，在许多应用中覆盖更长的距离。

The innovative development of new semiconductor laser can realize the emission of single mode light

Berkeley engineers have created a new type of semiconductor laser that achieves an elusive goal in optics: the ability to emit single-mode light while maintaining the ability to amplify its size and power.

The research team demonstrated a semiconductor film with evenly spaced and identically sized holes perforated as a perfect, retractable laser cavity. They showed that the laser emits a consistent single wavelength regardless of the size of the cavity. The invention is called the Berkeley surface-emitting Laser (BerkSELs).

Using another medium to amplify the laser would take up a lot of space, so by eliminating the need for external amplification, the researchers reduced the size of computer chips and other laser-dependent components, increasing efficiency.

The results of this research are particularly relevant to vertical-cavity surface-emitting lasers (VCSELs), in which lasers are emitted vertically from a chip. The lasers are used in a wide range of applications, including fiber optic communications, computer mice, laser printers and biometric systems.

Vcsels are usually tiny, just a few micrometers wide. The current strategy used to increase its power is to cluster hundreds of individual VCsels together. Because lasers are independent, their phases and wavelengths differ, so their powers do not combine consistently. This is tolerable in applications such as facial recognition, but unacceptable in applications such as communications or surgery where precision is crucial.

Multimode lasers are not only inefficient, but even counterproductive. Single-mode lasers in BerkSELs This is much more efficient than existing lasers.

The study found that BerkSELs were designed to enable single-mode light emission because of the physical properties of the light passing through holes in the film, which is a 200-nanometry-thick layer of indium gallium arsenide phosphide, a semiconductor commonly used in fiber optics and telecommunications technology. The holes are etched using lithography and must have a fixed size, shape and spacing.

The researchers say periodic holes in the membrane become Dirac points, a topological feature of two-dimensional materials based on linear dispersion of energy. They are named after British physicist and Nobel laureate Paul Dirac, who is known for his early contributions to quantum mechanics and quantum electrodynamics.

The researchers show that the phase of light propagating from one point to another is equal to the refractive index multiplied by the distance travelled. Since the refractive index at the Dirac point is zero, the light coming from different parts of the semiconductor is exactly in phase and is therefore optically identical. The film in the study had about 3,000 holes, but theoretically it could have a million or a billion holes, and the result should be the same.

The researchers used high-energy pulsed laser light to pump and power the BerkSELs device. They measured the emission from each aperture using a confocal microscope optimized for near infrared spectroscopy.