Fund Project:Project supported by the Basic Research Program of the National Major Project of China (Grant Nos. G2017149, JG2017029, JG2018115) and the China Innovative Talent Promotion Plans for Innovation Team in Priority Fields (Grant No. 2014RA4051).
Received Date:26 November 2018
Accepted Date:16 January 2019
Available Online:01 April 2019
Published Online:20 April 2019
Abstract:In the laser-driven inertial confinement fusion facilities, the irradiation uniformity of the laser beams on the target is a key factor affecting the effective compression of the target. At present, a variety of beam-smoothing techniques have been developed to control the spatiotemporal characteristics of the focal spots. However, many optical components involved in optical transmission links and complex transmission transformations often lead to complex optical transmission. Moreover, when using the diffraction optical method to analyze the shape and characteristics of the focal spots, a lot of data are needed to be processed and calculated, resulting in large calculation and low computational efficiency. It is urgent to find a new and fast method to describe the statistical properties of the focal spots. In addition, in the beam-smoothing technique, since the phase distribution of the continuous phase plate is obtained by multiple iterations of random numbers, although the details of focal spots obtained by different continuous phase plates are not the same, they all have similar statistical properties. Therefore, the modulation of the laser beam by the continuous phase plate can be regarded as the transmission process of the laser beam through a random surface. Although the intensities of the speckle within the focal spot at different locations have the strong randomness, and the random distributions of the target speckles obtained by different beam-smoothing methods are different, the overall distribution satisfies a certain statistical law. In this paper, the light-field properties of the focal spot are described by the statistical characterization method. The circular complex Gaussian random variables are used to directly describe the statistical properties of the target surface light field, and the far-field focal spots obtained by the diffractive optical method and those by the statistical characterization method are compared with each other and analyzed based on the typical focal spot evaluation parameters. The results show that the instantaneous properties of the focal spots obtained by the diffractive optical method and those obtained by the statistical characterization method are basically identical, but their time-integrated far-field focal spots are different. The correlation coefficient can be further used to describe the time-varying properties of the far-field focal spots. Compared with the diffractive optical method, in the numerical calculation process, the statistical characterization method of light field properties can directly obtain the analytical expression of the statistical distribution of the light field according to the statistical properties of the continuous phase plate surface shape. Secondly, this method can avoid the numerical calculation process from near field to far field. Last but not least, there is no need to perform data processing on each point of the light field, which makes things simple and effective and does not require large-scale data storage and processing. Keywords:statistical optics/ inertial confinement fusion/ beam smoothing/ focal spot
在ICF装置中, 常采用CPP对激光束的焦斑进行空间整形[20]. CPP的随机性主要体现在随机种子数上, 而确定性主要体现在相位滤波函数上. 相位滤波过程中通过改变滤波截止频率可以获得不同最小空间周期的CPP[21]. CPP的位相分布由随机数多次迭代获得, 保留了一定的随机特征, 同时其位相梯度也具有一定的确定性. 采用最小空间周期一定而随机数种子不同的CPP, 用相同的统计方法提取不同CPP的位相统计特征, 其位相的统计分布如图2所示. 图 2 不同随机数种子得到的CPP的位相统计分布 Figure2. Statistical distribution of the phase of CPP obtained from different random number seeds.
式中$\bar g(x,y)$为位相梯度的平均值; ${g_{i,j}}\left( {x,y} \right)$为离散点对应的位相梯度值; n, m分别为离散点对应的行数和列数. 计算最小空间周期相同而随机数种子不同的CPP的GRMS值, 得到3个CPP的GRMS值分别为0.4540, 0.4480和0.4479, 即采用最小空间周期相同而随机数种子不同的CPP得到的GRMS值基本一致. 由此可见, 不同随机数种子设计得到不同CPP的位相分布在一定误差范围内满足相同的统计规律. 这说明了CPP位相分布的随机性与确定性并存. 对于这种不完全随机的统计分布特征, 更有利于我们从其中提取出规律性, 可进一步根据CPP面形的统计特征推导出靶面光场的统计特征. CPP的面形分布是连续且随机的, 因而可以将其看作一个表面高度为随机函数的衍射光学元件[23]. 因此, 激光束经过CPP汇聚至靶面可视为光源照明粗糙表面产生散射光的过程, 可以采用统计光学的理论模型对焦斑特性进行统计分析. 如图3所示, 将CPP看成许多小单元构成的位相元件, 且对激光束的附加位相满足某一类统计分布. 当对位相板划分的单元数足够多时, 可用圆型复数高斯随机变量对瞬时焦平面进行描述: 图 3 经CPP调制后激光束汇聚至靶面的过程 Figure3. The process of the laser beam converged to the target plane after the modulation of CPP.
为了验证采用圆型复数高斯随机变量描述靶面光场的可行性, 先对靶面光场的统计特性进行分析. 输入光场的参数为: 光束束腰半径w = 186 mm; 超高斯阶数N = 6; 中心角频率ω0 = 1.79 × 1015 Hz; 光束波长λ0 = 1053 nm; CPP的PV值为7.3λ0. 由于CPP面形分布的随机性, 光束通过CPP后在靶面形成散斑. 对该散斑的光强和位相的统计特性进行分析, 典型结果如图4所示. 图 4 激光束经过CPP整形后靶面光强和位相统计特征 (a) CPP整形后的靶面光强分布; (b) CPP整形后的靶面振幅分布; (c) CPP位相与远场位相统计分布 Figure4. The statistical characteristics of the laser beam's intensity and phase of the target plane after CPP's shaping: (a) Intensity distribution of the target plane after CPP's reshaping; (b) amplitude distribution of the target plane after CPP's shaping; (c) statistical distribution of CPP's phase and far field phase.
从图4(a)和(b)可知, 激光束经过CPP整形之后, 在靶面形成散斑光强的统计分布近似为负指数分布, 振幅的统计分布近似为瑞利分布, 与圆型复数高斯随机变量满足相同的统计特征. 在图4(c)中, 拟合曲线服从正态分布, 可见CPP位相的统计分布大致为正态分布, 且CPP的位相分布的统计特性与靶面散斑的位相分布的统计特性大致一致. 因此, 圆型复数高斯随机变量中的位相分布?应该与CPP位相的统计分布一致, 即满足正态分布. 我们进一步利用衍射积分模型对激光束通过CPP后的靶面光强分布进行数值模拟, 并对由圆型复数高斯随机变量描述的靶面光强分布进行统计分析. 将连续位相板视为由512 × 512位相单元组成, 进而采用常规评价参数对两种方法得到的瞬时焦斑的光强和位相两个方面进行比较, 并运用FOPAI曲线比较两者靶面瞬时光强的不同峰值功率占总功率的份额, 采用位相统计分布规律来比较数值求解的远场位相分布与光场特性的统计表征方法得到的远场位相分布. 典型结果如图5所示. 图 5 数值求解与的瞬时远场光强特性比较 (a)瞬时焦斑光强FOPAI对比; (b)数值求解远场位相与解析求解远场位相统计特性 Figure5. Comparison of characteristics of instantaneous far-field intensity solved by numerical analysis and that Solved by analytical solution: (a) FOPAI's comparison instantaneous focal spot intensity; (b) statistical characteristics of numerical solution far-field phase and analytical solution far-field phase.