定义：
应用于激光器中来提高激光器输出光功率、频率或其它量的稳定性。

激光器存在各种激光器噪声，在实际应用中很不利，因此需要采用一些技术抑制噪声并且稳定激光器的参数。下面讨论主动和被动激光稳定机制。 

主动激光稳定 
主动稳定机制包含一些电子反馈系统（或者正反馈系统），参数的涨落转化为电子信号，然后在以某种形式作用于激光器。 

图1：二极管泵浦的固态激光器，采用了反馈系统来稳定输出功率。 
主动稳定机制有以下例子： 

1. 可以采用如图1所示的示意图来稳定激光器的输出功率。采用光电二极管控制激光器功率，然后通过控制激光谐振腔中或者外部的泵浦功率和损耗来校正功率。采用这种方法，可以同时减小开通电源产生的尖峰效应以及稳态时的强度噪声。这里需要注意的是，作用于输出光束而不是激光器本身可以减小强度噪声，参阅词条激光功率稳定系统（noise eaters）。 
2. 通过控制谐振腔长度可以稳定单频激光器的光频率，或者锁模激光器频率梳中的频率。获得反馈信号有几种方式，例如，通过记录第二个激光器的拍音，或者测量一个非常稳定的参考腔或者干涉仪的透射和反射，或者采用无多普勒效应光谱学方法测量吸收池的透射率等。常用的采用参考强产生信号的方法是Pound-Drever-Hall方法，它将弱相位调制的光传送到参考腔。另一个不需要调制的机制为Hansch-Couillaud方法。 
3. 稳定载波包络相位偏移或者锁模激光器频率可以采用，例如，先用f-2f干涉仪测量相位，然后通过激光器谐振腔中的楔形物或者倾斜镜子产生反馈光。这种稳定方法在频率测量中非常重要。 
4. 锁模激光器的脉冲计时可以通过对比光二极管信号与参考的电子振荡器来控制，然后通过控制腔长来稳定脉冲。 
5. 通过测量光束位置（例如，采用四象限的光电二极管）可以稳定输出光束的指向，然后采用压电控制的谐振腔镜纠正它。 

主动系统能得到的稳定性由光电探测噪声，控制器件的带宽，反馈电路的设计和参考物的稳定性（例如，光参考腔）等因素决定。 

被动激光稳定 
被动稳定机制不需要电子学器件，只是采用纯光学效应。例如： 

• 可以采用稳定的参考腔的反馈光来稳定激光器的频率。（即可以认为是采用了另一个激光谐振腔，组成一个复合腔） 
• 可以通过克尔介质的交叉相位调制效应使两个锁模激光器同步，腔内的两束脉冲在介质中相遇。 

也可以采用注入锁定来稳定激光器的频率，也就是将另一个激光器中产生的具有非常稳定频率的光束注入到谐振腔中。

Definition: measures applied to lasers in order to improve their stability in terms of output power, optical frequency, or other quantities

As lasers exhibit various kinds of laser noise, which can be detrimental in applications, it is sometimes necessary to use techniques for suppressing noise and stabilizing certain laser parameters. There are active and passive stabilization schemes, as discussed in the following.

See also the article on synchronization of lasers, which treats both timing and phase synchronization.

Active Laser Stabilization

Active stabilization schemes usually involve some kind of electronic feedback (or sometimes feedforward) system, where fluctuations of some parameters are converted to an electronic signal, which is then used to act on the laser in some way.

Examples are:

• The output power of a laser may be stabilized with a scheme as shown in Figure 1. The laser power is monitored with a photodiode and corrected e.g. via control of the pump power or the losses in or outside the laser resonator. In this way, both spiking after turn-on and the intensity noise under steady-state conditions can be reduced.
• Note that it is also possible to reduce intensity noise by acting on the output beam instead of the laser itself; see the article on noise eaters.

Figure 1: Diode-pumped solid-state laser with a feedback system stabilizing the output power.

• The optical frequency of a single-frequency laser, or the frequency of one line of the frequency comb from a mode-locked laser, can be stabilized via resonator length control. The feedback signal can be obtained e.g. by recording a beat note with a second laser, by measuring the power which is transmitted or reflected at a very stable reference cavity or another kind of interferometer, or by measuring the transmission of a gas cell (e.g. an iodine cell), possibly using Doppler-free laser absorption spectroscopy. A frequently used technique for generating an error signal with a reference cavity is the Pound–Drever–Hall method [3, 4], using a weak phase modulation of the light which is sent to the reference cavity. A scheme not requiring such modulation is the Hänsch–Couillaud method [2]. Another method is tilt locking, where spatial mode interference is utilized [15, 29].
• The stabilization of the carrier–envelope offset phase or frequency of a mode-locked laser (CEO stabilization) can be based on, e.g., a phase measurement with an f−2f interferometer and feedback via some wedge or tilted mirror in the laser resonator. This kind of stabilization is important for frequency metrology.
• The timing of the pulses (→ timing jitter) from a mode-locked laser can be monitored by comparing the phases of a photodiode signal and of an electronic reference oscillator, and stabilized e.g. via cavity length control.
• Stabilization of the pointing direction of the output beam is possible via a beam position measurement (e.g. with a four-quadrant photodiode) and correction via piezo-controlled resonator mirrors.

The stability which is achieved with such active systems is determined by factors such as photodetection noise, the bandwidth of control elements, the design of the feedback electronics, and the stability of the reference standards (e.g. optical reference cavities).

Passive Laser Stabilization

Passive schemes do not involve electronics and are based on purely optical effects. Examples are:

• The frequency of a laser can be stabilized via optical feedback from a stable reference cavity. (This may also be considered as using an extended laser resonator, being a kind of composite cavity.)
• Synchronization of two mode-locked lasers is possible via cross-phase modulation in a Kerr medium, in which the intracavity pulses of both lasers meet.

The optical frequency of a laser may also be stabilized by injection locking, i.e., injecting a beam with a highly stable optical frequency from another laser.

Bibliography