掺镱光纤放大器及其在ICF前级系统中的应用
文献类型：学位论文
| 作者 | 卢秀权 |
| 学位类别 | 博士 |
| 答辩日期 | 2000 |
| 授予单位 | 中国科学院上海光学精密机械研究所 |
| 导师 | 陈绍和 |
| 关键词 | ICF前级系统 掺镱光纤放大器（YDFA） 增益 饱和输出功率 |
| 中文摘要 | 惯性约束核聚变(Inertial Confinement Fusion-ICF)是当今国际上的重大科研领域，有极为重要的科学意义和新能源应用前景。ICF实验研究的关键是高功率固体激光驱动器，它不仅要求激光脉冲的能量很高，而且对脉冲的频域宽度、时间、空间分布形状也有十分苛刻而又特殊的要求。因而用于ICF的激光驱动器往往是一个规模较大、结构复杂的系统，它一般可分为三部分，即前端、主放和靶室。其中，前端要完成多种重要功能，包括激光种子脉中的产生、频谱展宽、放大、分束及时间、空间整形。历史上，围绕着提高前端性能问题，人们作了大量的研究工作。进入九十年代后，传统前端中闪灯泵浦的固体激光器及高压驱动的块状调制器因稳定性和可靠性问题，逐渐被最新发展起来的集成光学技术取代了。用光纤系统来完成种子脉冲产生及传输的功能，而脉冲的频谱展宽和时间整形则由波导调制器来实现。在前级系统集成化的过程中，必然引入的一个技术单元是光纤放大器，它可用来对低功率激光脉冲进行高增益放大，或者在分束前作为信号的功率放大器。作为我国将兴建的“神光－III”前级方案预研工作的一部分，本论文首次从理论和实验两方面开展了掺镱光纤放大器（YDFA）的研究，取得了重要成果。论文第一章主要由四部分组成。第一部分简述了著名的美国利弗莫尔实验室及中国的高功率激光物理国家实验室的ICF前级系统发展和未来，阐明了现代的集成光学技术确实为前端性能的提高带来了新的机遇。第二部分介绍了YDFA在ICF前端中的应用，重点解释了选择掺镱光纤（YDF） 作为激活光纤的理由。第三部分说明了YDFA的工作原理及构成的各要素，包括：a)增益介质YDF的一般制作程序、光谱特性和一些特殊的效应；b)LD泵浦源的工作原理、性能及其与单模光纤的耦合；c)放大器的其它常用元件。第四部分综述了国际上YDFA的发展概况，指出了虽然YDF在目前主要用于激光的产生，但是它在将来有可能被广泛用于特殊波长(如1083nm)的功率放大、光纤传感器中的小信号放大、自由空间激光通信以及超短脉冲的啁啾脉冲放大。第二章理论分析了光纤放大器中导模的空间分布。在“弱导波近似”条件下，对阶跃型折射率光纤求解波动议程，导得各阶LP模的横向功率密度归一化分布，首次发现YDFA内各波长LP_(01)模的空间分布均可近似用同一个半径为W的高斯函数来表示。第三章首次将YDFA作为一个简单的二能级系统，得到了粒子数速率议方程，泵浦光、信号光及放大的自发辐射传播方程，并画出了用计算机数值求解的流程图。在处理单模光纤放大器过程中，引入了基模的高斯函数近似。考虑到基本方程中有许多物理量难以直接测量，提出了工种化的模拟方法，其中的小信号吸收系数及饱和光功就绪参数可以通过小功率单光束传输实验测得。增益和噪声指数是用来评价YDFA的两个重要性能的参数，一般情况下，都要通过复杂的数值计算才能看出它们与光纤参数之间的关系，这不利于具体YDFA的设计。忽略信号和ASE功率对离子反转数的影响之后，得到了小信号增益的简单解析解。如果只忽略ASE，就可得到描述信号增益和输入泵浦功率、信号功率之间关系的超越方程。在非饱和增益区，正向和反向ASE功率及噪声指数可有解析表达式。注意到在YDFA作为功率放大器时通常不存在ASE的自饱和，因此，计算时先不考虑ASE的影响，利用龙格－库塔法得到信号功率沿光纤的分布，然后利用ASE和信号的关系求得噪声指数。第四章运用前面关于YDFA的基本理论和概念首次对YDFA的各种特性进行了数值计算。4.2节研究了在不同泵浦功率、不同激活光纤长度和不同泵浦波长条件下，掺镱单模石英光纤中ASE谱的变化，描述了正向和反向ASE的差异，这对选择激光器振荡波长以及提高放大器增益和改善其噪声性能有重要意义。4.3节数值计算了在910nm泵浦下，YDFA的增益、饱和与噪声特性，结果将对YDFA的设计提供有益的参考。4.4节分别计算了在不同波长泵浦下YDFA的增益、饱和与噪声特性，结果表明在975nm泵浦时，由于不存在975nm本身的ASE问题，此时放大器的增益、饱和等特性甚至优于910nm泵浦时的情况，根据该分析，就结合具体情况，给一个实际的YDFA选择合适的泵浦波长。4.5节首次提出了YDFA的分段模拟方法，并分析了内插特殊光学元件的掺镱光纤放大器。表明，隔离器、滤波器和光纤光栅的使用不但可以提高放大器增益，而且同时能够使噪声特性得到优化，文中还讨论了隔离器、滤波器在光纤内的最佳位置。第五章理论研究了YDFA对脉冲的放大。5.2节写出了YDFA的二能级瞬态基本方程，基本方程求解复杂。为了直观和定性的解释YDFA的瞬态增益和饱和现象，5.3节首次导得了YDFA输入端粒子反转数的解析解。5.4节把YDFA瞬态理论模型进一步化简，首次得到一个用来描述放大器瞬态动力学行为的、简单的、而物理意义又非常明确的单个常微分方程，并数值计算了方波脉冲的放大。第六章实验研究了YDFA。首先，我们选择了技术较成熟的980nm LD作为泵浦源(6.2节)。为了进一步了解YDF的特性，分别在6.3和6.4节作了两个测试性实验，即YDF中放大的自发辐射及YDFA对1030nm连续小信号的放大。在6.5节我们根据现有的技术条件，研制了一个两级结构的高增益脉冲YDFA，在国内首次实现了大于500倍的增益。在6.6节，我们指出了将来在设计Yb光纤功率放大时需注意的要点。 |
| 英文摘要 | Inertial Confinement Fusion (ICF) is an important scientific research field today, which is of extreme importance for science and bears a farreaching application foreground for the new energy source. The key for ICF experiment is the laser driver, where not only the large energy, but also a given amount of broadened bandwidth and the special spatio-temporal shape of the laser pulses are required. Generally, the laser driver is a large-scale and complicated system, composed of three major subsystems: front end, main amplifier chains and target room. The front-end system is the portion of the laser driver where a single seed pulse is produced, bandwidth broadened, amplified and multiplexed, then temporally and spatially shaped to feed the main amplifier chains. In the history, many efforts have been made to improve the performance of the front-end system. In the 1990s, because of their poor stability and low reliability, the flashlamp-pumped lasers and the highvoltage bulk modulators in the traditional front end were gradually replaced by the modern integrated optics. Accordingly, the laser pulse is generated and transported within the fiber system, and the bandwidth broadening and temporally shaping of the pulses are accomplished by the highly efficient, low-power waveguide modulators. With the development of the front end's integration, the fiber amplifier will necessarily be introduced. It can provide high gain for the low power laser pulse, or be used as a power amplifier before the fiber splitter. As part of the research for the front end of the proposed "SHENGUANG-III", in this dissertation, we have studied for the first time the Yb-doped fiber amplifier (YDFA) both theoretically and experimentally and made some significant achievements. Chapter one consists of four major subparts. In the first part, narrated are the front end's developments and futures in both LLNL in America and NLHPLP in China, from which the capability improvements of the front end by the modern integrated optics are obvious. Part two explains the reasons for choosing the Yb-doped fiber (YDF) as the activated fiber, and then sums up the possible applications of the YDFA in the front end. Part three details the YDFA's three major elements: a) the YDF and its fabrication procedure, spectroscopic properties, and some nonlinear effects; b) the LD pumping sources and the coupling technologies of LD to single-mode fiber; c) the other amplifier's components in common use. Part four summarizes the research status quo of the YDFA in the world, and it is pointed out that although the YDFs have so far been used mainly in lasers, the possible applications in amplification of the power at special wavelengths(e.g., 1083nm and 1053nm), amplification of the small signal in fiber sensors, free-space laser communications and chirped-pulse amplification of the ultrashort pulses have attracted great interest. In the second chapter, we theoretically investigate the spatial distribution of the guided optical modes in the amplifier. By introducing the "weak-guidance" approximation, the normalized radial distribution of each LP mode in the step-index fiber is derived from Maxwell's equations. For the first time, we find that the distribution of the LP_(01) modes corresponding to various wavelengths in the YDFA can be well approximated by the Gaussian function with the same radius of w. In the third chapter, the YDFA is simplified as a two-level system. We write out the equations for the Yb populations, and the propagation of signal, pump and (amplified spontaneous emission) ASE, then the flow chart for their computation. For the single-mode fiber amplifier, we can use the Gaussian approximation. To get around the difficulty for measurement of some fiber parameters in the basic equations, we for the first time, propose the modal method suitable for the engineering design of the YDFA, where the required small-signal-absorption coefficient and saturation power can be directly determined from one-beam fiber-transmission experiment. The gain and noise figure are the two major parameters used to evaluate the performance of the YDFA. As a rule, they can be determined for a given YDF, and power condition, through the complicated and time consuming numerical computation, which may be inconvenient for the practical design. However, when the ASE power can be neglected, we obtain a single transcendental equation for the relation between the gain and the input pump, signal power. In the case of both the ASE and the signal being negligible, the gain has a simple analytical expression. In the unsaturated regime, the analytical solutions for the gain, the forward and backward ASE power spectra and the noise figure exist. If the YDFA is used as a power amplifier, there usually isn't the problem of self-saturation induced by the ASE. In this case, we can solve for the pump and signal power distribution along the YDF first, using the Runge-Kutta method, then we can find the solutions for the noise figure with the relation between the ASE and the signal. In the forth chapter, we utilize the foregoing theories and basic concepts to calculate the various characteristics of the YDFA numerically, for the first time. In section 4.2, the differences between the forward and backward ASE in ytterbium-doped monomode silica fiber are described. The changes in ASE spectral shapes under different pump power, different activated fiber length and different pump wavelength conditions are also detailed. The results could be important for the design of the YDF devices, such as the superfluorescent laser sources, tunable lasers and traveling-wave amplifiers. In section 4.3, we predict the gain, saturation and noise characteristics of YDFAs. It is shown that when pumped at λp=910nm, the strong backward propagating ASE around 975nm could dramatically increase the NF and limit the gain available at longer wavelengths. In section 4.4, the gain, saturation and noise characteristics of the YDFAs with different pump wavelengths are discussed. It is pointed out that the YDFAs pumped at 975nm, free from self-saturation induced by the 975-nm ASE, have some advantages, as compared with YDFAs pumped at 910nm. In section 4.5, we for the first time, propose the segmentation theory to investigate the YDFA, which incorporates fiber isolator, fiber filter or fiber grating. The optimum placement of the filters and isolators are also calculated so that the gain and noise performance can be improved simultaneously. In chapter five, we theoretically investigate the amplification of the laser pulse by the YDFA. In section 5.2, the complicated basic equations describing the transient behavior of the YDFA are listed. To physically interpret the observed transient dynamics, we solve for the first time the basic system of equations for the exact solutions of the Yb populations at the fiber input end, (see section 5.3). In section 5.4, we succeed in further simplifying the transient model of the YDFAA, and for the first time reducing the set of basic equations into a simple ordinary differential equation that has explicit physical meaning for the amplification process of laser pulse. Using this equation, we calculate the amplification of a rectangular pulse numerically. In chapter six, the YDFA is studied experimentally. First, we choose the commercial 980-nm LD for EDFA as the pumping source (see section 6.2). Then, to further understand the properties of the YDF, in section 6.3 and 6.4 respectively, we do two fundamental experiments: the output ASE and the amplification of the 1030-nm-CW small signal. In section 6.5, we develop a high-gain two-stage YDFA for laser pulse, with the existing technical condition. The gain is higher than 500, which is to our knowledge the maximum gain obtained with YDFA in the country. In section 6.6, we made some suggestions for the design of the power YDFA. |
| 语种 | 中文 |
| 源URL | [http://ir.siom.ac.cn/handle/181231/15306]  |
| 专题 | 上海光学精密机械研究所_学位论文 |
| 推荐引用方式 GB/T 7714 | 卢秀权. 掺镱光纤放大器及其在ICF前级系统中的应用[D]. 中国科学院上海光学精密机械研究所. 2000. |
入库方式： OAI收割
来源：上海光学精密机械研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。