New website getting online, testing
Metasurface beamsplitter with large field of view and equal diffraction angle interval
  • Abstract

    LiDAR currently mainly uses a Dammann grating as the laser beamsplitter. However, as a periodic diffraction optical device, the Dammann grating satisfies the grating equation requiring each diffraction angle's sine value to form an arithmetic progression, which cannot achieve uniform angular beam-splitting. The theoretical diffraction efficiency is also limited. This paper uses the angular spectrum and random search optimization algorithm to design a more flexible non-periodic beamsplitter. Simulations show that the metasurface beamsplitter can achieve a 70-degree field angle of 41 beams with an equal diffraction angle interval. The simulated diffraction efficiency reaches 84% which is higher than the diffraction limit of a binary phase device. In experiments, the metasurface beamsplitter has good beam-splitting uniformity and can promote the development of LiDAR.

    Keywords

  • 但是,达曼光栅作为周期性二元衍射光学器件,主要存在两个方面的问题。第一方面,达曼光栅各分束光束的衍射角满足光栅方程,在正入射条件下,对于给定的周期和波长,其各衍射级次角度的正弦值构成正比。当衍射角满足小角度近似时,我们可认为出射线束的衍射角度等间隔分布;当衍射角较大时,出射线束的衍射角度就不再均匀。在实际应用场景中,我们希望衍射角度构成等差数列或其它任意分布。而达曼光栅无法实现在指定的角度序列出射的功能,如在等差数列角度上实现均匀的光束分束。另一方面,达曼光栅是二元相位衍射光学器件。二元相位衍射光学器件的理论最大衍射效率为81%。由于设计差异,不同分束个数的二元达曼光栅的衍射效率不同。以转折点为0.73526的3分束达曼光栅和转折点为0.22057和0.44563的4分束达曼光栅为例,两者的衍射效率分别为66.42%及70.63%[],都低于理论最大衍射效率。

    随着物联网技术的快速发展,自动驾驶在过去几年里取得了重大进展,即将实现工业化[]。为保障自动驾驶的安全,我们需要实现更精确的汽车导航感知系统。目前,自动驾驶汽车导航的感知系统包括摄像头、毫米波雷达和激光雷达等[]。其中,摄像头的准确度与周围环境的明暗相关。而激光雷达通过主动发射激光照亮周围环境,从测量激光反射回波得到实际距离,可以获得比摄像头这类被动探测方案更为精确的距离信息。因此,激光雷达是无人驾驶所需的重要传感器[],主要由发射器、扫描器、接收器、处理器等部件组成。目前商用的激光雷达有单线至128不等的线束数,其中高线数的激光雷达可以实现更高的分辨率和帧率。为了实现高线束数,需利用分束器对激光进行分束。在激光分束器设计中,学界及工业界主要采用以达曼光栅[]为代表的二相位台阶衍射光学器件。达曼光栅通过调控二台阶相位转折点的位置可在远场实现等强度的光斑阵列。目前,2至32束及64束[]的达曼光栅数值解均已求出,圆环达曼光栅[-],三维达曼阵列[-]等广义达曼光栅也得到了研究和发展。

    为解决达曼光栅无法实现任意角度调制和低衍射效率的问题,我们需要使用具有任意波前调控能力的器件。超构表面是具有亚波长特征尺寸的衍射光学器件,可以调控光的振幅[]、相位[-]、波长[]及偏振[-]等参数。为了突破周期性光栅分束器衍射角的局限性,需要非周期性的波前调制。超构表面具有灵活的二维空间结构分布,完美契合了分束器非周期性波前相位调控的需要。同时,超构表面有相位连续调控能力,其理论最大衍射效率为100%,可突破二元相位器件的衍射效率极限。除此之外,超构表面亚波长周期的纳米结构单元可以消除高级次衍射,具有最大可达±90°衍射角,可实现超大角度的光束分束。近年来已有超构表面分束的文献报道。Song等人提出利用复振幅调控的电介质超构表面实现二维选择性衍射[]。 Ni等人利用纯相位调控的电介质超构表面实现了大视场角的高均匀性一维和二维分束[]。然而,现有超构表面分束器多为周期性器件,受制于光栅方程对衍射角的限制,尚未有实现等角度间隔分束的文献报道。在本文中,我们提出了一种适用于任意角度分束分布的分束器设计优化算法,充分发挥了超构表面灵活的相位调控能力。我们利用该算法设计了一款在70°全角内等角度间隔分束41束的超构表面分束器。仿真和实验测试表明其显著地扩展光束分束器的功能。

    Figure 1. Schematic diagram of the designed metasurface beamsplitter and Dammann grating
    Full-Size Img PowerPoint

    Schematic diagram of the designed metasurface beamsplitter and Dammann grating

    图1所示,达曼光栅和超构表面分束器均可将入射激光转化为多束光束。超构表面分束器由于其更小的特征尺寸可以实现比达曼光栅更大的衍射角。对于激光雷达应用而言,由于达曼光栅的衍射角无法覆盖整个视场角,以其为分束器的激光雷达依赖外部机械部件实现大视场角的环境扫描。超构表面分束器更大的衍射角可以增大探测范围,简化外部扫描机械结构,提高扫描帧率。

    Figure 2. (a) Schematic diagram of the metasurface cell, with a length of 480 nm, a width of 240 nm, a height of 760 nm, and a period of 650 nm. The substrate is 500-μm thick; (b) Efficiency of cross circularly polarized light at 650 nm period. The black dot in (b) is the selected metasurface cell; the length and width of the nanopillar are 480 nm and 240 nm, respectively; the conversion efficiency is 99.2%
    Full-Size Img PowerPoint

    (a) Schematic diagram of the metasurface cell, with a length of 480 nm, a width of 240 nm, a height of 760 nm, and a period of 650 nm. The substrate is 500-μm thick; (b) Efficiency of cross circularly polarized light at 650 nm period. The black dot in (b) is the selected metasurface cell; the length and width of the nanopillar are 480 nm and 240 nm, respectively; the conversion efficiency is 99.2%

    为了设计高效率的超构表面单元,我们使用时域有限差分法对超构表面单元进行了仿真。分束器的设计波长为1550 nm,该波长属于红外通信窗口,具有成熟的配套激光器、探测器等器件。由于液态水对该波长具有较大的吸收,1550 nm激光对人眼相对安全。作为激光雷达的重要部件,该分束器设计在1550 nm波长,可以承受较高的激光功率,提高信噪比。如图2(a)所示,超构表面单元由500 μm厚的二氧化硅基底和760 nm高的非晶硅柱组成。该器件采用几何相位原理调控光的相位,当入射圆偏振光通过超构表面时,出射光将变为正交圆偏振光,并附带2θ的相位,其中θ是硅柱的面内旋转角。由于硅柱的平面投影为长方形,满足C2对称的硅柱携带的相位为转角的两倍[-]。相较于传播相位调控,几何相位调控可以提供更准确的相位响应。在实际加工过程中,当硅柱小幅偏离设计尺寸时,其携带的几何相位由于转角不变而不受影响,从而对加工误差具有更大宽容度。通过全波仿真,我们将周期确定为650 nm,超构表面的周期应小于半波长,可消除高级次衍射。如图2(b)所示,我们计算了此周期下,不同长度和宽度的超构表面结构单元的圆偏振光转化效率。我们选取几何尺寸为480 nm,宽240 nm的矩形硅柱为超构表面的基本组成单元(图2(b)中黑点)。该尺寸的硅柱具有高达99.2%理论转化效率,同时其对角线长度(536 nm)较小。因此,相邻单元的硅柱不易粘连,便于器件加工。

    Figure 3. (a, b) One-dimensional simulation of the beamsplitter intensity angle distribution. The blue line is the simulation result, and the orange dashed line is the optimization target; (c) The unwrapped phase of the beamsplitter; (d) The full width at half maximums (FWHMs) of spots in (a). The blue line is the simulation result, the yellow dashed line is the diffraction limit, and the red dotted line is the optimization target; (e) 41 rays on the 40 cm distancing plane obtained by the numerical Rayleigh-Somerphy diffraction integral. Each light spot is cut into 3 mm squares and spliced together
    Full-Size Img PowerPoint

    (a, b) One-dimensional simulation of the beamsplitter intensity angle distribution. The blue line is the simulation result, and the orange dashed line is the optimization target; (c) The unwrapped phase of the beamsplitter; (d) The full width at half maximums (FWHMs) of spots in (a). The blue line is the simulation result, the yellow dashed line is the diffraction limit, and the red dotted line is the optimization target; (e) 41 rays on the 40 cm distancing plane obtained by the numerical Rayleigh-Somerphy diffraction integral. Each light spot is cut into 3 mm squares and spliced together

    其中:$ {f}_{x} $为空间频率,$ \lambda $为波长,L为超构表面直径,$ {\theta }_{n} $为各分束衍射角,$ \beta $表示衍射极限的远场发散角与目标的远场发散角的比值。为避免出现高于衍射极限的频率成分,我们将$ \beta $设为0.9。如图3(a, b)所示,橙色虚线展示了远场强度优化目标。

    E=(IItarget)2Nα0thorderIN0thorder,

    为设计分束器的相位分布,我们采用快速傅里叶变换和随机搜索相结合的优化算法。直径1 mm的分束器需在远场水平方向形成41个点状光斑,0级光斑居中,全角为70°。各级次衍射角构成等差数列,因此衍射级次为m的衍射角$ {\varphi }_{m}=m\Delta \varphi $,其中,$ m= 0, \pm 1,\pm 2,\cdots ,\pm 20 $,$\Delta \varphi$为相邻衍射级次设定的角度间隔,在此为1.75°。由于器件光斑沿水平方向展开,竖直方向无需叠加相位调制,可以将其相位设计简化为一维相位优化问题。即已知器件表面及远场的强度分布,求器件表面及远场的相位。基于角谱算法,我们可以根据器件表面的复振幅求得远场的复振幅分布。长度为N的一维离散复振幅$ U\left(n\right) $的角谱为$ A\left(m\right)= \displaystyle{\sum} _{n=0}^{N-1}U\left(n\right){W}_{N}^{mn} $ ,其中${W}_{N}={{\rm{e}}}^{-2\pi /N}$,为 N 次单位根之一。分束器将近场超构表面的复振幅傅里叶变换为远场指定角度出射的光束。

    为实现大视场角的等角度间隔分束,我们首先需要确定远场强度优化目标。远场的强度分布可用衍射角度的函数描述。由离散傅里叶变换公式可知,空间频率对应角度的正弦值构成等差数列$ {\theta }_{n}={{\arcsin}} \left(\dfrac{\lambda }{L}(n-N/2)\right) $, 其中$ \lambda $为波长,L为计算窗口的长度,N为向量长度,n为向量索引,$ 0\le n < N-1 $。由于等间隔的衍射角度$ {\varphi }_{m} $不与格点$ {\theta }_{n} $重合,其无法直接用$ {\theta }_{n} $上的强度数列描述。为获得更高的计算精度,我们将一个超构表面单元周期分成四个具有相同相位的像素单元,此时计算周期为162.5 nm。我们增大计算窗口长度以缩小频率间隔,用更多离散点近似表示远场不同衍射级次的光束。通过将实际尺寸为1 mm的器件外填充零至20 mm的计算窗口,我们得到20倍的频率格点,进而可较为准确地描述远场各衍射级次光束的细节。得益于快速傅里叶变换$ O\left(N\log\right(N\left)\right) $的较低计算复杂度[],增加傅里叶变换向量长度的时间代价较小。由衍射光学可知,不同衍射角出射的分束光束的远场发散角与其衍射角有关。以$ \theta $角从长度为L 的狭缝出射的平面波的复振幅可写为$ {U}_{{\mathrm{i}}}\left(x\right)={{\mathrm{e}}}^{-{\mathrm{i}}kx\mathrm{s}\mathrm{i}\mathrm{n}\theta }\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{t}\left({x}/{L}\right) $。其傅里叶变换为$ FT\left\{{U}_{{\mathrm{i}}}\right\}=L*\mathrm{s}\mathrm{i}\mathrm{n}\mathrm{c}\left(L\left({f}_{x}-{\mathrm{sin}\theta }/{\lambda }\right)\right) $。因此,其0级斑的远场发散角为$\mathrm{\Delta }\theta ={\mathrm{arcs}\mathrm{i}\mathrm{n}}\left({\pi \lambda }/{L}+\mathrm{sin}\theta \right)+ {\mathrm{arcsin}}\left({\pi \lambda }/{L}-\mathrm{sin}\theta \right)$。当$ \mathrm{sin}\theta \gg {\pi \lambda }/{L} $时,其0级斑的远场发散角为正入射时远场发散角的$ {1}/{\mathrm{cos}\theta } $倍。我们将各衍射级次分束的单缝衍射图样强度进行叠加,可得远场光强的理论优化目标为

    其中:I为当前光强,${I}_{{\rm{target}}}$为目标光强,N为像素个数,$ \alpha $为非负的经验系数,取为0.25。超构表面的入射光为平面波,其振幅强度设为1。为避免陷入局部最优解,我们将全部1538个超构表面单元的优化顺序随机打乱,然后依次确定各个单元的相位。保持其他单元的相位不变,此单元的相位依次选取0–2π范围内的16个等间隔相位台阶,同时计算分别取不同相位台阶时的评价函数值,从中挑选出评价函数最小的相位更新该单元的相位。我们将随机遍历全部1538个超构表面单元视为1次迭代。在经过50次迭代后,图3(c)为优化得到的相位分布。由于分束个数41为奇数,分束器的相位中同时包含了奇函数和偶函数成分。该器件的远场光强仿真结果如图3(a, b)中蓝色实线所示。可以看出,归一化远场光强仿真结果和光强优化目标相吻合。全部41个光斑强度利用强度峰值的平均值进行归一化后的标准差为0.011,极差(最大值与最小值之差)为0.052。数值仿真表明,在±35°衍射范围内的41个光斑总光强占入射光的84.2%,即器件的理论衍射效率为84.2%。图3(d)的蓝线为仿真光斑的半高全宽(full width at half maximum,FWHM),其与优化目标(红色点划线)吻合。为了进一步验证仿真结果,我们利用瑞利-索莫非衍射积分数值计算[]分束器在40 cm处平面的光强分布。如图3(e)所示,高级次光斑较低级次光斑在水平方向有所展宽,且峰值亮度下降。

    Itarget (fx)=n=2020sinc2(Lβ(fxsinθnλ)),

    在确定远场强度优化目标后,我们采用随机搜索算法求解该一维相位优化问题。为了兼顾远场衍射光强度与强度优化目标的一致性及器件的整体衍射效率,我们将评价函数设置为均方根误差(RMSE)和各级次0级斑衍射效率(以各衍射级次0级斑光强计)的线性叠加

    本文设计的超构表面分束器采用了标准的微纳加工工艺制备。首先,在石英衬底沉积760 nm厚的非晶硅薄膜。然后,在硅膜上旋涂厚度为100 nm的正电子束光刻胶并进行烘烤。再使用加速电压为100 kV的电子束光刻(JEOL,JBX 6300FS)对电子束胶进行曝光。显影后,在电子束胶上沉积10 nm厚的铬膜作为掩模。通过剥离工艺将矩形硅柱图案转移到铬膜上。再使用电感耦合等离子体蚀刻机(Oxford,Plasma Pro System100 ICP380)蚀刻760 nm厚的硅膜。最后,使用铬刻蚀液去除残留的铬掩模,得到几何相位超构表面。

    本文中超构表面分束器还有很多优化及改进的空间。从设计方法上看,本文基于标量衍射理论来优化衍射图案,忽略了光的偏振和矢量性质,没有考虑到近场耦合效应的影响。各纳米柱采用FDTD仿真时,我们在四周采用了周期性边界条件。实际纳米柱周围的纳米柱的几何相位转角不一定与其相同,采用周期性边界条件仿真忽略了相邻纳米柱的影响。近场耦合效应可使光在纳米结构之间的传输和重新分配,这会导致一部分能量被束缚在近场中。从而以倏逝波或反射波形式损耗光能,将对衍射效率造成负面影响。近场耦合还可改变纳米柱周围的电磁场分布,使超构表面的波前相位调制偏离原有设计,从而影响到衍射图案的形状和强度分布。在后续研究中,考虑到一维分束器在与相位梯度正交的方向上的重复性,计算区域可简化为狭长条状区域。在计算资源充足时,进行大尺寸全波仿真从而充分考虑纳米柱之间的近场耦合是可能的,进而使得仿真结果更加精确。

    从拓扑优化的角度考虑,可以优化超构表面单元的形状及晶格排列方式。超构表面单元的形状需要和入射光的偏振态相匹配。如沿用当前的圆偏振光入射,则考虑三角形、长方形、五边形及具有各向异性的无定形结构等可以产生几何相位的拓扑结构[]。如改用线偏振光入射,则考虑圆形、方形、长条状及无定形结构等。晶格排列方式方面可以考虑四边形和六边形网格等。六边形网格可能较四边形网格有更高的衍射效率及更弱的采样锯齿效应[]

    Figure 4. (a, b) Metasurface morphology imaged by 5x and 100x objective microscopes. Scalebars: 200 μm and 2 μm; (c) Design drawing of the same area in Fig. 4(b), with a scale of 2 μm; (d) Schematic diagram of the experimental optical path. P: polarizer; QWP: quarter-wave plate; MS: metasurface; (e) Captured 41 rays. Each beam is cut into 3 mm squares and spliced together; (f) Experimentally measured power of beams. The blue line represents the beams from the 1st to 20th orders and the orange dotted line is from the −1st to −20th orders; (g)Experimentally measured FWHMs. The solid blue line with asterisks and the orange dotted line with circles are the horizontal FWHMs from the 1st to the 20th orders and from the −1st to the −20th orders, respectively. The dashed yellow line with regular triangles and the dashed-dotted purple line with inverted triangles are the vertical FWHMs from the 1st to the 20th orders and from the −1st to the −20th orders, respectively. The solid green line is the horizontal diffraction limit.
    Full-Size Img PowerPoint

    (a, b) Metasurface morphology imaged by 5x and 100x objective microscopes. Scalebars: 200 μm and 2 μm; (c) Design drawing of the same area in Fig. 4(b), with a scale of 2 μm; (d) Schematic diagram of the experimental optical path. P: polarizer; QWP: quarter-wave plate; MS: metasurface; (e) Captured 41 rays. Each beam is cut into 3 mm squares and spliced together; (f) Experimentally measured power of beams. The blue line represents the beams from the 1st to 20th orders and the orange dotted line is from the −1st to −20th orders; (g)Experimentally measured FWHMs. The solid blue line with asterisks and the orange dotted line with circles are the horizontal FWHMs from the 1st to the 20th orders and from the −1st to the −20th orders, respectively. The dashed yellow line with regular triangles and the dashed-dotted purple line with inverted triangles are the vertical FWHMs from the 1st to the 20th orders and from the −1st to the −20th orders, respectively. The solid green line is the horizontal diffraction limit.

    在样品加工完后,我们使用WITEC显微镜的白光反射照明模式观察器件的表面形貌。5倍物镜下的显微图像如图4(a)所示,该样品直径为1 mm。100倍物镜下的显微图像(图4(b))与超构表面相同区域的设计图纸(图4(c))吻合,表明样品具有较好的加工效果。从显微图像中可以看出纳米硅柱的竖直方向仅具有重复的超构表面单元结构,相位调制只存在于水平方向。

    从纳米结构单元角度考虑,本文中的分束器工作时光束是正入射的。当超构表面单元周期为亚波长时,由光栅方程可知,最大衍射角为90°。当前超构表面的70°分束全角是根据具体需求设计的,其分束全角可以根据设计需求变得更大。我们对超构表面单元的厚度、周期和长宽等参数进行过扫描。就最大衍射角而言,只要保证超构表面的周期小于波长,其最大衍射角均为±90°。但从提高衍射效率角度出发,周期应较小,从而保证相位分布的连续性。本器件相位的最大衍射角为±35°,对1550 nm的激光等效周期为2.7 μm,相当于4.2个超构表面周期。如果等效的超构表面周期数进一步减少,由于相位的离散性增加,器件效率会迅速降低。

    图4(d)展示了我们搭建的实验表征光路。1550 nm激光从激光器光纤端面出射,利用短焦透镜准直为平行光,出射后经过线偏振片和四分之一波片转化为圆偏振光。圆偏振光在超构表面纳米柱面正入射并在远场形成41个水平分布光斑。仿真中,该分束器的入射光是均匀平面波。实验上,为保证入射光场的均匀性,实验中光束直径被扩束至约5 mm,远大于分束器直径(1 mm),减少了实验与仿真的偏差。为在远场条件下进行大角度范围拍摄,我们将铟镓砷相机(HAMAMATSU,C14041-10U)固定在金属杆上。该金属杆可绕超构表面正下方的支点在水平面内旋转,相机距离转轴40 cm。由于相机距离超构表面较远,单个相机像素对应的远场发散全角仅0.0029°,可以精细分辨光斑的细节并测量其半高全宽。我们旋转相机在不同衍射级次上依次拍摄41个光斑,其光强分布如图4(e)所示,大部分光斑具有较好的高斯分布。随着衍射级次的增加,光斑会产生形变,在水平方向上有所拉伸,亮度也有所下降。为使照明在超构表面上的光束接近设计的平面波,入射光束尺寸会大于器件尺寸,进而导致0级光斑存在较强的背景光,出现过曝现象。图4(f)为实验测得的$ \pm 1,\pm 2,\cdots ,\pm 20 $级光斑的相对强度。随着衍射级次增加,衍射角度也逐渐变大,光强呈下降趋势。各光斑总光强的标准差为0.179。如图4(g)所示,为分析各级次光斑的均匀度,我们统计了实验中$ \pm 1,\pm 2,\cdots ,\pm 20 $级光斑的水平及纵向半高全宽。随着级次增大,横向半高全宽缓慢增加,而纵向半高全宽基本不变。由于衍射光斑在水平方向上存在一定展宽,光斑的横向半高全宽平均值为0.143°,略大于纵向半高全宽平均值0.1234°,与设计结果基本吻合。

    理论上,该一维分束器还可被扩展为二维分束器。二维M×N分束的分束器的相位由一维M束分束相位和一维N束分束相位叠加得到,即$\varphi \left(x,y\right)={\varphi }_{1{D}_{x}}\left(x\right)+ {\varphi }_{1{D}_{y}}\left(y\right)$,其中${\varphi }_{1{D}_{x}}\left(x\right)$和${\varphi }_{1{D}_{y}}\left(y\right)$为一维的分束器相位。该1维41分束相位可用于构造41×41束分束器相位。如果每个方向上的分束数改变,上述的${\varphi }_{1{D}_{x}}$或${\varphi }_{1{D}_{y}}$需按照本文讨论的方法分别设计。在本方法中,只需要设计一维相位分布即可,优化速度快于二维相位分布的设计速度,从而提升了器件的设计效率。

    本文设计并实验测试了一种红外超构表面激光分束器。将角谱法和随机搜索算法结合,可优化得到该分束器的相位分布。该分束器在实验测试中实现了等角度间隔、相同半高全宽和具有均匀强度分布的41分束效果。相较于现有的二元相位达曼光栅,该分束器打破了分束衍射角正弦值成正比的制约,同时具有70°的分束全角,高达84%的理论衍射效率。实验测得的光束功率标准差为0.179,光束的横向平均半高全宽为0.143°,这表明超构表面分束器具有较好的分束均匀性,将促进激光雷达的发展,拓展超构表面的应用场景。

    所有作者声明无利益冲突

  • References

    [1]

    Li Y, Ibanez-Guzman J. Lidar for autonomous driving: the principles, challenges, and trends for automotive lidar and perception systems[J]. IEEE Signal Process Mag, 2020, 37(4): 50−61.

    DOI: 10.1109/MSP.2020.2973615

    CrossRef Google Scholar

    [2]

    Leonard J, How J, Teller S, et al. A perception‐driven autonomous urban vehicle[J]. J Field Robot, 2008, 25(10): 727−774.

    DOI: 10.1002/rob.20262

    CrossRef Google Scholar

    [3]

    Dammann H, Görtler K. High-efficiency in-line multiple imaging by means of multiple phase holograms[J]. Opt Commun, 1971, 3(5): 312−315.

    DOI: 10.1016/0030-4018(71)90095-2

    CrossRef Google Scholar

    [4]

    Zhou C H, Liu L R. Numerical study of Dammann array illuminators[J]. Appl Opt, 1995, 34(26): 5961−5969.

    DOI: 10.1364/AO.34.005961

    CrossRef Google Scholar

    [5]

    Zhao S, Chung P S. Design of a circular Dammann grating[J]. Opt Lett, 2006, 31(16): 2387−2389.

    DOI: 10.1364/OL.31.002387

    CrossRef Google Scholar

    [6]

    Wen F J, Chung P S. Use of the circular Dammann grating in angle measurement[J]. Appl Opt, 2008, 47(28): 5197−5200.

    DOI: 10.1364/AO.47.005197

    CrossRef Google Scholar

    View full references list
  • Author Information

  • Copyright

    The copyright belongs to the Institute of Optics and Electronics, Chinese Academy of Sciences, but the article content can be freely downloaded from this website and used for free in academic and research work.
  • About this Article

    DOI: 10.12086/oee.2024.240141
    Cite this Article
    Xia Ruixing, Zhao Dong, Li Ziqin, Huang Kun. Metasurface beamsplitter with large field of view and equal diffraction angle interval. Opto-Electronic Engineering 51, 240141 (2024). DOI: 10.12086/oee.2024.240141
    Download Citation
    Article History
    • Received Date June 16, 2024
    • Revised Date July 14, 2024
    • Accepted Date July 15, 2024
    • Published Date August 24, 2024
    Article Metrics
    Article Views(650) PDF Downloads(200)
    Share:
  • Related Articles

[1]

Li Y, Ibanez-Guzman J. Lidar for autonomous driving: the principles, challenges, and trends for automotive lidar and perception systems[J]. IEEE Signal Process Mag, 2020, 37(4): 50−61.

DOI: 10.1109/MSP.2020.2973615

CrossRef Google Scholar

[2]

Leonard J, How J, Teller S, et al. A perception‐driven autonomous urban vehicle[J]. J Field Robot, 2008, 25(10): 727−774.

DOI: 10.1002/rob.20262

CrossRef Google Scholar

[3]

Dammann H, Görtler K. High-efficiency in-line multiple imaging by means of multiple phase holograms[J]. Opt Commun, 1971, 3(5): 312−315.

DOI: 10.1016/0030-4018(71)90095-2

CrossRef Google Scholar

[4]

Zhou C H, Liu L R. Numerical study of Dammann array illuminators[J]. Appl Opt, 1995, 34(26): 5961−5969.

DOI: 10.1364/AO.34.005961

CrossRef Google Scholar

[5]

Zhao S, Chung P S. Design of a circular Dammann grating[J]. Opt Lett, 2006, 31(16): 2387−2389.

DOI: 10.1364/OL.31.002387

CrossRef Google Scholar

[6]

Wen F J, Chung P S. Use of the circular Dammann grating in angle measurement[J]. Appl Opt, 2008, 47(28): 5197−5200.

DOI: 10.1364/AO.47.005197

CrossRef Google Scholar

[7]

Yu J J, Zhou C H, Jia W, et al. Three-dimensional Dammann array[J]. Appl Opt, 2012, 51(10): 1619−1630.

DOI: 10.1364/AO.51.001619

CrossRef Google Scholar

[8]

Yu J J, Zhou C H, Jia W, et al. Three-dimensional Dammann vortex array with tunable topological charge[J]. Appl Opt, 2012, 51(13): 2485−2490.

DOI: 10.1364/AO.51.002485

CrossRef Google Scholar

[9]

Fu W W, Zhao D, Li Z Q, et al. Ultracompact meta-imagers for arbitrary all-optical convolution[J]. Light Sci Appl, 2022, 11(1): 62.

DOI: 10.1038/s41377-022-00752-5

CrossRef Google Scholar

[10]

Zhao D, Dong Z G, Huang K. High-efficiency holographic metacoder for optical masquerade[J]. Opt Lett, 2021, 46(6): 1462−1465.

DOI: 10.1364/OL.419542

CrossRef Google Scholar

[11]

李瑞琛, 邹毅军, 陈天航, 等. 宽频消色散超表面全息成像[J]. 光电工程, 2023, 50(8): 230118.

DOI: 10.12086/oee.2023.230118

Li R C, Zou Y J, Chen T H, et al. Broadband achromatic metasurface holography[J]. Opto-Electron Eng, 2023, 50(8): 230118.

DOI: 10.12086/oee.2023.230118

CrossRef Google Scholar

[12]

杨睿, 于千茜, 潘一苇, 等. 基于片上超表面的多路方向复用全息术[J]. 光电工程, 2022, 49(10): 220177.

DOI: 10.12086/oee.2022.220177

Yang R, Yu Q Q, Pan Y W, et al. Directional-multiplexing holography by on-chip metasurface[J]. Opto-Electron Eng, 2022, 49(10): 220177.

DOI: 10.12086/oee.2022.220177

CrossRef Google Scholar

[13]

Li G X, Zhang S, Zentgraf T. Nonlinear photonic metasurfaces[J]. Nat Rev Mater, 2017, 2(5): 17010.

DOI: 10.1038/natrevmats.2017.10

CrossRef Google Scholar

[14]

冯睿, 田耀恺, 刘亚龙, 等. 拓扑优化超表面的偏振复用光学微分运算[J]. 光电工程, 2023, 50(9): 230172.

DOI: 10.12086/oee.2023.230172

Feng R, Tian Y K, Liu Y L, et al. Polarization-multiplexed optical differentiation using topological metasurfaces[J]. Opto-Electron Eng, 2023, 50(9): 230172.

DOI: 10.12086/oee.2023.230172

CrossRef Google Scholar

[15]

Zhao H, Wang X K, Liu S T, et al. Highly efficient vectorial field manipulation using a transmitted tri-layer metasurface in the terahertz band[J]. Opto-Electron Adv, 2023, 6(2): 220012.

DOI: 10.29026/oea.2023.220012

CrossRef Google Scholar

[16]

Song X, Huang L L, Tang C C, et al. Selective diffraction with complex amplitude modulation by dielectric metasurfaces[J]. Adv Opt Mater, 2018, 6(4): 1701181.

DOI: 10.1002/adom.201701181

CrossRef Google Scholar

[17]

Ni Y B, Chen S, Wang Y J, et al. Metasurface for structured light projection over 120° field of view[J]. Nano Lett, 2020, 20(9): 6719−6724.

DOI: 10.1021/acs.nanolett.0c02586

CrossRef Google Scholar

[18]

Xie X, Pu M B, Jin J J, et al. Generalized pancharatnam-berry phase in rotationally symmetric meta-atoms[J]. Phys Rev Lett, 2021, 126(18): 183902.

DOI: 10.1103/PhysRevLett.126.183902

CrossRef Google Scholar

[19]

Lin D M, Fan P Y, Hasman E, et al. Dielectric gradient metasurface optical elements[J]. Science, 2014, 345(6194): 298−302.

DOI: 10.1126/science.1253213

CrossRef Google Scholar

[20]

Johnson S G, Frigo M. A modified split-radix FFT with fewer arithmetic operations[J]. IEEE Trans Signal Process, 2007, 55(1): 111−119.

DOI: 10.1109/TSP.2006.882087

CrossRef Google Scholar

[21]

Shen F B, Wang A B. Fast-Fourier-transform based numerical integration method for the Rayleigh-Sommerfeld diffraction formula[J]. Appl Opt, 2006, 45(6): 1102−1110.

DOI: 10.1364/AO.45.001102

CrossRef Google Scholar

[22]

Kim S, Kim J, Kim K, et al. Anti-aliased metasurfaces beyond the Nyquist limit[Z]. arXiv: 2406.11261, 2024. https://doi.org/10.48550/arXiv.2406.11261.

Google Scholar

Related Articles
Show full outline

Catalog

    Corresponding author: Huang Kun, huangk17@ustc.edu.cn

    1. On this Site
    2. On Google Scholar
    3. On PubMed
    Metasurface beamsplitter with large field of view and equal diffraction angle interval
    • Figure  1

      Schematic diagram of the designed metasurface beamsplitter and Dammann grating

    • Figure  2

      (a) Schematic diagram of the metasurface cell, with a length of 480 nm, a width of 240 nm, a height of 760 nm, and a period of 650 nm. The substrate is 500-μm thick; (b) Efficiency of cross circularly polarized light at 650 nm period. The black dot in (b) is the selected metasurface cell; the length and width of the nanopillar are 480 nm and 240 nm, respectively; the conversion efficiency is 99.2%

    • Figure  3

      (a, b) One-dimensional simulation of the beamsplitter intensity angle distribution. The blue line is the simulation result, and the orange dashed line is the optimization target; (c) The unwrapped phase of the beamsplitter; (d) The full width at half maximums (FWHMs) of spots in (a). The blue line is the simulation result, the yellow dashed line is the diffraction limit, and the red dotted line is the optimization target; (e) 41 rays on the 40 cm distancing plane obtained by the numerical Rayleigh-Somerphy diffraction integral. Each light spot is cut into 3 mm squares and spliced together

    • Figure  4

      (a, b) Metasurface morphology imaged by 5x and 100x objective microscopes. Scalebars: 200 μm and 2 μm; (c) Design drawing of the same area in Fig. 4(b), with a scale of 2 μm; (d) Schematic diagram of the experimental optical path. P: polarizer; QWP: quarter-wave plate; MS: metasurface; (e) Captured 41 rays. Each beam is cut into 3 mm squares and spliced together; (f) Experimentally measured power of beams. The blue line represents the beams from the 1st to 20th orders and the orange dotted line is from the −1st to −20th orders; (g)Experimentally measured FWHMs. The solid blue line with asterisks and the orange dotted line with circles are the horizontal FWHMs from the 1st to the 20th orders and from the −1st to the −20th orders, respectively. The dashed yellow line with regular triangles and the dashed-dotted purple line with inverted triangles are the vertical FWHMs from the 1st to the 20th orders and from the −1st to the −20th orders, respectively. The solid green line is the horizontal diffraction limit.

    • Figure  1
    • Figure  2
    • Figure  3
    • Figure  4