基于选择性光谱响应结构的横向色散多焦点超透镜

Transversely Dispersive Multi⁃Foci Metalens Based on Selective Spectral Response Structure

光谱分离成像是一种可同时获取光谱信息、空间位置信息的探测技术,被广泛应用于地理、环境、生物、光学等领域,集成化是该技术的重要发展方向之一。提出了一种基于超透镜的光谱分离技术,采用选择性光谱响应单元结构对红、绿、蓝(RGB)三个波段的聚焦相位进行编码,基于单片器件即可收集不同位置的光谱信息,简化了获取光学信息的过程,推动了小型化光谱分离成像器件的进一步发展。

Objective Spectral information is among the most essential characteristics of a substance and can be detected and analyzed to obtain multidimensional information,such as composition,density,and shape.Spectral separation imaging technology based on this technology can obtain the spectral and spatial position information of the detected object simultaneously,for which reason it is widely used in earth science,environmental monitoring,intelligent agriculture,biomedicine,and other fields.Traditional methods are usually based on a mechanical turntable that rotates multiple filters to achieve spectral separation.This requires the use of multiple serial imaging methods to collect the desired spectral information,which entails the disadvantages of high system complexity and inconvenient operation.Although considerable research has been conducted on simplifying device complexity,integration remains a problem to be solved.Methods In this study,a transversely dispersive multi-focus metalens based on a selective spectral response structure is proposed.Three sets of unit structures are designed to independently respond to circularly polarized light with wavelengths of 473,532,and 633 nm.The focusing phase maps corresponding to the red,green,and blue(RGB)wavebands are calculated using a transversely dispersive mechanism.Furthermore,the three groups of unit structures are combined with three focusing phase maps using a binary matrix coding method.Spectral information from different locations can then be collected using a single device,which simplifies the process of obtaining optical information and avoids the shortcomings of traditional spectral imaging devices that require switching bulky optical splitting devices.In addition,owing to the shared aperture design characteristics of transversely dispersive multi-foci metalenses,it has a larger numerical aperture(NA)than the metalens array,showing better imaging performance.Results and Discussions Figure 1 shows the design concept of a transversely dispersive multi-foci metalens that linearly superimposes three sub-metalenses to achieve multi-focus focusing on the three wave bands of the RGB without sacrificing NA.The basic unit structure is shown in Fig.2(a).For the unit structure with a high polarization conversion efficiency at the designed wavelength and a low polarization conversion efficiency at the other two wavelengths,as shown in Fig.2(b),three groups of unit structures,RectB,RectG,and RectR,are obtained.They have the same period(P=400 nm)and height(H=700 nm).Figure 2(c)shows the relationship between the polarization conversion efficiencies of RectB,RectG,and RectR in independent modulation and the wavelength of the incident light.Figure 2(d)shows the phase modulation function of the designed cell structure using RectB as an example.Figures 2(e)‒(g)show that the Hadamard product is calculated by multiplying the binary matrix corresponding to the three wavebands of the RGB and the focusing phase maps,and the final phase map of the transversely dispersive multi-foci metalens can be obtained by linear superposition.The axial light field distributions of the metalens are shown in Figs.3(a),(d),and(g).The focal lengths are 49.75(blue),49.71(green)and 49.73μm(red),respectively,which are very close to the design values,demonstrating that the designed metalens can achieve the function of multi wave band focusing at the target position.Figures 3(b),(e),and(h)show the cross-sectional light field distributions shown by the white dotted lines in Figs.3(a),(d),and(g),respectively,showing the excellent symmetry of the focus.Figures 3(c),(f),and(i)show the one-dimensional horizontal transversals at the peak of the focus shown in Figs.3(b),(e),and(h),respectively.The results show that the full width at half maximum(FWHM)at each focus position is 502(blue),573(green),and 670 nm(red),and the theoretical value is very close to the simulation value,indicating good focusing performance.Furthermore,crosstalk analysis between each waveband of the metalens is performed.The geometric phase-unit structure exhibits wideband response characteristics.Although the unit structure is avoided as much as possible in the previous design to modulate light beyond the working waveband,there is still some modulation that causes the metalens to focus outside the target position,as shown in Fig.4.The analysis shows that although crosstalk is inevitable,the target focus intensity is significantly stronger than that of the crosstalk focus.Therefore,this crosstalk only slightly reduces the focusing efficiency and does not affect the spectral separation function during actual use.Conclusions In this paper,a spectral separation device is implemented.When the incident light is blue(473 nm),green(532 nm),and red(633 nm),the FWHM of the planarization device is 502,573,and 670 nm,respectively,which can realize multi-wavelength focusing close to the diffraction limit.Although crosstalk exists in the transversely dispersive multi-foci metalens,it does not affect the actual spectral separation function.The spectral response characteristics can be further optimized by changing the geometric size,material,arrangement,and other parameters of the unit structure to reduce crosstalk and improve spectral separation performance.On the basis of the realization of the basic functions,metalenses have the characteristics of planarization,miniaturization,and integration,which provides a new solution for reducing the complexity of spectral separation imaging equipment and further promotes the practical application of metasurfaces.Meanwhile,the spectral separation method proposed in this paper is not limited to the three wavebands of RGB,but can be extended to more bands through the design of the unit structure and the calculation of the phase map to achieve multispectral imaging and even hyperspectral imaging.