Abstract:Direct X-ray imaging by a Fresnel zone plate (FZP) has achieved a spatial resolution of 10 nm on a synchrotron beamline. It may be used to obtain submicron-resolution X-ray images of laser-plasma sources or fusion targets. However, none of previous imaging experiments with laser-plasma kilo-elelctron-volt X-ray sources shows such a high resolution. In comparison with the FZP imaging on a synchrotron, we consider a case of imaging an extended object with a laser-plasma X-ray source that the illumination monochromaticity is lower and the field of view larger. Our simulations show that the spatial resolution is affected by both the object size and the spectral bandwidth of the source, which can explain the previous experiments. We conclude that by using a 100-zone FZP to image an object with up to 700 μm in size, a spatial resolution better than 1 μm can be realized by using X-rays of several kilo-electron volts and a spectral bandwidth just less than 3%. In this paper, we report a proof-of-principle study in simulation and experiment in an optical range centered at 632.8 nm. The simulation is performed with the same method as that previously used for X-ray imaging but with a 100-zone FZP working in the optical range. Simulations show that with the increase of the object size, the field-of-view contrast is degraded, but the spatial resolution is nearly unchanged. With the increase of the spectral bandwidth for the illumination, both the contrast and the resolution are degraded. In the experiments, different spectral bandwidths are realized by band-pass filters and different object sizes by an adjustable aperture. The experimental results are confirmed to be in agreement with the simulations. These results reveal that given a satisfied spectral bandwidth of laser-plasma X rays, the FZP imaging will be a promising approach to 1 μm or higher resolution X-ray imaging of a 1-mm-size object. Keywords:imaging of zone plate/ extended source/ laser plasma
在考察扩展光源光谱带宽对成像的影响中, 固定可变光阑尺寸使扩展光源直径为3 mm, 利用滤波片来选择入射光的光谱带宽. 图8的左列给出了光谱带宽参数w分别为0.5%, 1.5%, 8%, 12%时成像实验结果, 右列给出了相应像扣除像探测器背景后经过其图案中心沿x2方向(参见图8(a))的强度分布. 可见随着入射光的光谱带宽增加, 像变得越来越模糊, 对比度下降. 图 8 扩展光源直径为3 mm时不同光谱带宽的成像结果及其沿x2轴方向强度分布 (a) w = 0.5%; (b) w = 1.5%; (c) w = 8%; (d) w = 12% Figure8. Images of 3 mm-diameter source for different spectral bandwidth, and the corresponding intensity distribution along x2 axis: (a) w = 0.5%; (b) w = 1.5%; (c) w = 8%; (d) w = 12%.
参考图8(a)右图的虚线所示, 按4.1节所述方法处理像图案的中心圆环区强度分布的上升沿, 可获得像对比度以及对应的视场中央分辨能力. 图9给出了像对比度与视场中央分辨能力随入射光光谱带宽的变化, 图中空心符号点及连线是图6的模拟结果. 结果表明, 随着扩展光源光谱带宽w从0.5%增加至12%, 视场中央分辨能力从26.4 ${\text{μ}}{\rm m}$下降至76.0 ${\text{μ}}{\rm m} $, 像对比度从0.74下降至0.41. 这一变化趋势与数值模拟结果相同(参见图9中虚线). 定量比较, 实验与数值模拟结果也符合得较好. 例如, 当光谱带宽参数$w = 8$%时, 实验给出视场中央的分辨能力为70.0 ${\text{μ}}{\rm m}$, 像对比度为0.50, 而数值模拟给出视场中央的分辨能力为75.1 ${\text{μ}}{\rm m}$, 像对比度为0.56. 实验和数值模拟结果的相对偏差约10%, 这可能来自实验数据的处理: 像探测器像素有一定大小以及接收的光强信号有涨落. 图 9 视场中央分辨能力与像对比度随光谱带宽变化的实验结果 Figure9. Experimental results for spatial resolution and image contrast in the field-of-view center versus spectral bandwidth.
25.2.扩展光源尺寸对FZP成像的影响 -->
5.2.扩展光源尺寸对FZP成像的影响
实验中使用w = 0.5%的准单色光成像, 来验证扩展光源尺寸对成像影响的模拟结果. 4.2节的数值模拟以及5.1节的实验结果表明, 对于当前所使用的FZP, 入射光谱带宽分别为0和0.5%所成的像, 视场分辨能力与像对比度基本一样. 因此可以把0.5%的准单色光近似看成单色光. 控制可变光阑大小, 扩展光源的直径分别设为3和12 mm, 记录成像结果. 采用与图8相同的数据处理方法, 得到像对比度与视场中央分辨能力随扩展光源尺寸的变化, 结果如图10所示, 图中空心符号点及连线是图4的数值模拟结果. 对于3和12 mm扩展光源, 像对比度分别为0.74和0.30, 视场中央的分辨能力分别为26.4和28.7 ${\text{μ}}{\rm m} $. 可见, 随着光源尺寸的增加, 视场中央分辨能力略微下降. 考虑到分辨能力的改变量(2.3 ${\text{μ}}{\rm m} $)小于像探测器的像素尺寸(5.2 ${\text{μ}}{\rm m}$), 可以认为分辨能力基本不变. 与此不同的是, 像对比度显著降低. 这些变化趋势与图4的数值模拟结论一致. 图4的数值模拟结果给出了当扩展光源尺寸分别为3和12 mm时, 视场中央分辨能力分别为22.6和23.2 ${\text{μ}}{\rm m}$, 像对比度分别为0.80和0.26. 因此, 定量比较数值模拟与实验结果也符合得较好. 图 10 视场中央分辨能力与像对比度随扩展光源尺寸变化的实验结果 Figure10. Experimental results for spatial resolution and image contrast in the field-of-view center versus extended source size.