Design and fabrication of complicated diffractive optical elements on multiple curved surfaces

Ran Tian; Juan Liu; Xin Li; Xugang Wang; Yongtian Wang

doi:10.1364/OE.23.032917

1. Introduction

There are various applications of the diffractive optical elements (DOEs) on curved surfaces, such as thin-film transistors [1], artificial compound eye [2] and so on. It has been a challenge to fabricate arbitrary DOEs on multiple planar or curved surfaces simultaneously. In recent years, many methods, such as: the laser direct writing [3–6 ], the ion beam proximity lithography [7], the soft lithography [8,9 ], the nano-imprint lithography [10], and the interference lithography [11–16 ], are employed to fabricate micro/nano-structures on planar or curved surface. Most methods are time-consuming and expensive while interference method is simple and fast. Shi et al. have designed and fabricated arbitrary DOEs on planar surface by interference [17], and Zhao et al. improved his method and further fabricated arbitrary DOEs on one curved surface [18], where the precise alignment of two laser beam is required. To solve this problem, Wang et al. proposed a convenient method to fabricate DOEs on curved surface using the computer generated hologram which also adopted for three-dimensional display in [19,20 ], where only one laser beam modulated by spatial light modulator (SLM) is employed [21]. However, all the methods above can only fabricate one DOE at a time and the depth of field is also small. If multiple arbitrary DOEs on multiple curved surfaces are required to fabricate, the common method need multiple exposures on different surfaces, which will lead to the difficulties of the collimation of different curved surfaces, and also it’s time-consuming. Stimulated by those challenges, in this paper, we propose a method to fabricate multiple arbitrary DOEs on multiple curved surfaces simultaneously.

In this paper, we present an effective method to design and fabricate multiple arbitrary intensity distribution of DOEs on multiple curved surfaces. The proposed design method is combined the Fidoc method [22] (improved GS algorithm) and the Huygens diffraction [23] in scalar diffraction domain. 3D binary patterns and 3D gray level patterns are simulated and fabricated on multiple curved surfaces, which verifies the feasibility of this method. The efficiency of the fabrication is improved and the multiple collimation of different curved surfaces is avoided since it requires only one optimized phase (loaded in SLM) to produce multiple desired profiles of DOEs. In addition, the depth of field is larger. It is believed that it is an effective method to fabricate multiple arbitrary complicated DOEs on curved surfaces simultaneously, and it could be useful in various optical fields.

2. Basic principles

The basic principle is shown in Fig. 1 . $P_{1}$ , $P_{2}$ denote the curved surfaces on which the desired DOEs will be fabricated. $H$ denotes the plane that we place a phase distribution. $P_{1} (Χ_{1})$ , $P_{2} (Χ_{2})$ and $H (Ξ)$ denote the complex amplitude distributions the on surface $P_{1}$ , $P_{2}$ and $H$ respectively, where $X_{1}$ , $X_{2}$ and $Ξ$ represent the Cartesian coordinate systems $x_{1} y_{1} z_{1}$ , $x_{2} y_{2} z_{2}$ and $ξ η z$ respectively. Since the DOEs are fabricated by exposure, the profiles of DOEs can transfer into the 3D intensity distributions $I (X_{i}) = {| P (X_{i}) |}^{2}$ on corresponding surfaces. Then, based on the desired intensity distributions $I (X_{i})$ , the phase distribution $H (Ξ)$ on plane $H$ will be optimized by the following method.

Fig. 1 Schematic view of the light propagation between the plane $H$ and the curved surfaces $P_{1}$ , $P_{2}$ .

Download Full Size | PDF

At first, it is necessary to clarify that the proposed method to fabricate multiple DOEs on multiple curved surfaces simultaneously is based on the principle of independently propagating and superposition of light. Then, the Huygens diffraction and the inverse Huygens diffraction [24,25 ] in scalar diffraction domain (considering the paraxial approximation) is adopted to denote the process of light propagation. And the positive propagation direction is from $H$ to curved surfaces. So, the complex amplitude distributions on curved surfaces $P_{1}$ , $P_{2}$ propagated from plane H can be expressed as Eq. (1)

P_{i} (Χ_{i}) = \frac{1}{j λ} \iint_{Γ} H (Ξ) \frac{\exp (j 2 π r_{i} / λ)}{r_{i}} d σ (i = 1, 2)

respectively, and the distribution

H (Ξ)

on plane H propagated from curved surfaces

P_{1}

and

P_{2}

can be expressed as Eq. (2)

H (Ξ) = \sum_{i = 1}^{2} \frac{j}{λ} \iint_{Γ'_{i}} P_{i} (X_{i}) \frac{\exp (- j 2 π r'_{i} / λ)}{r'_{i}} d σ_{i}'

where the subscript

i = 1, 2

means the two curved surfaces

P_{1}

,

P_{2}

respectively.

λ

is the wavelength of light.

Γ

denotes the plane

H

and

Γ'_{1 (2)}

is the curved surface

P_{1}

(

P_{2}

). The distance between surface

H

and

P_{1}

is d₀, and d is the distance between surface

P_{1}

and

P_{2}

. The diffraction distance between two points on

H

and curved surface

P_{1} (P_{2})

is

r_{1, 2} = \sqrt{{(x_{1, 2} - ξ)}^{2} + {(y_{1, 2} - η)}^{2} + l {(z_{1, 2})}^{2}}

, where

l (z_{1, 2})

is the z-component of distance which is variable for different points on curved surfaces.

Then, combined the Huygens diffraction we adopted above with the Fidoc method [22], Fig. 2 shows the whole process of the method we propose to design the pure-phase distribution to fabricate multiple arbitrary DOEs on multiple curved surfaces. The intensity distributions of DOEs we fabricate combined with random phases are adopted as the initial complex amplitude distributions $P_{10} (X_{1})$ and $\times$ on the curved surfaces, respectively. When the $j$ th iteration is performed, at first, the complex amplitude distributions $P_{1 j} (X_{1})$ and $P_{2 j} (X_{2})$ on the curved surfaces conduct the inverse Huygens diffraction (denoted by HuF^- ¹ {…}). And according to Eq. (2), the distribution $H_{j} (Ξ)$ on plane $H$ is obtained. After that, the phase of $H_{j} (Ξ)$ is kept and the amplitude is imposed to unity, and the distribution ${\bar{H}}_{j} (Ξ)$ is obtained. Then, according to Eq. (1), the Huygens diffraction (denoted by HuF {…}) of ${\bar{H}}_{j} (Ξ)$ from plane $H$ to curved surfaces is computed separately to acquire the complex amplitude distributions ${\tilde{P}}_{1 j} (X_{1})$ on $P_{1}$ and ${\tilde{P}}_{2 j} (X_{2})$ on $P_{2}$ . Then, the phase is kept, while the amplitude is both replaced with $A_{i, j + 1}$ , _where the feedback parameter k adds negative feedback to each pixel value to force it to converge faster, M is a function of zero at the positions of the zero padding and unity at the image’s pixels, and the noise suppression parameter $γ$ ,combined with M, affect the noise of the reconstructed patterns [22]. And the obtained complex amplitude distributions $P_{i, j + 1} (X_{i})$ is adopted to update $P_{i, j} (X_{i})$ for next iteration. The iteration will continue until the number of iteration reaches the maximum or the reconstructed intensity distributions from ${\bar{H}}_{j} (Ξ)$ have acceptable and convergent relative errors from the ideal intensity distributions. If feedback parameter k and the noise suppression parameter $γ$ are reasonable, the algorithm will converge within tens of iterations and the errors are stably reduced so that the noise is stable and acceptable. Finally, the optimized pure-phase distribution $φ_{j}$ can be obtained which will be loaded in the phase-only SLM, and 3D intensity distributions of DOEs on multiple curved surfaces can be acquired. After exposure once, the multiple desired DOEs can be fabricated.

Fig. 2 Flowchart of the proposed optimization method.

Download Full Size | PDF

3. Numerical simulations

In this section, we will conduct numerical simulations to evaluate our proposed approach. At first, we should establish the mathematical model of the curved surfaces. Considering the practicality of the subsequent fabrication, we choose two cylindrical surfaces, one of which is concave and the other one is convex, as shown in Fig. 3 . Figure 3(a) shows the top view, and the pink part is the cylindrical surfaces we established which sampled into a series of grids with equal interval as shown in Fig. 3(b). The subscript 1 and 2 means the two cylindrical surfaces respectively, L is the side length of cylindrical surface, R is the radius of the curvature, 2α is the field angle, and d is the distance of two surfaces. Then, we simulate the process of the proposed method to calculate the pure-phase distribution on a plane to modulate the 3D intensity distributions on the curved surfaces simultaneously.

Fig. 3 (a) Top view and (b) sampling schematic diagram of cylindrical surfaces.

Download Full Size | PDF

We apply a 3D grey level pattern and a 3D binary pattern on the two cylindrical surfaces respectively to conduct the numerical simulation. The initial intensity distributions on the two cylindrical surface is shown in Fig. 4(a) : the left one on concave cylindrical surface $P_{1}$ is a 3D binary pattern, while on the right convex surface $P_{2}$ , a 3D grey level badge pattern is adopted. Both two patterns with pixel of 256 $\times$ 256 are expended to the patterns with pixels of 400 $\times$ 400 by zero padding. The parameters for the simulation are as follows: L _1,2 = 12 mm, R _1,2 = 51.852 mm, d = 2 mm, $λ$ = 532 nm. The hologram is comprised of 800 $\times$ 800 pixels, each having square size of $8 μ m$ . And the distance between the plane $H$ and the cylindrical surface $P_{1}$ is d ₀ = 361 mm which needs to satisfy the sampling theorem. We choose the noise suppression parameter $γ = 1$ , the feedback parameter k = 0.9. The relative error (RE), defined as $R E = R_{e} / R_{i} \times 100 %$ , is selected to evaluate the quality of the reconstruction, where $R_{e} = \sum_{m = 1}^{M} \sum_{n = 1}^{N} {(| I' (m, n) | - | I (m, n) |)}^{2}$ is the error item, $R_{i} = \sum_{m = 1}^{M} \sum_{n = 1}^{N} {| I (m, n) |}^{2}$ represents the ideal distribution item, $I'$ denotes the reconstructed intensity distribution, $I$ denotes the desired intensity distribution, M and N represents the number of row and column of the sampling points. So the smaller value of RE implies the higher quality of reconstruction. After 30 iterations, the reconstructed intensity distributions on curved surfaces is shown in Fig. 4(b). Figure 4(c) shows the relationship between the values of RE and the number of iteration. The values of RE both decrease with the increase of iterations, and after 35 iterations, the RE reduce to 3.28% and 2.67% respectively which represent the high quality of the reconstructed patterns.

Fig. 4 Numerical simulation of 3D patterns on cylindrical surfaces: (a) the intensity distributions of the initial patterns, (b) the intensity distribution of the reconstructed patterns after 30 iterations, (c) the relationship between relative errors and the number of iteration.

Download Full Size | PDF

The mathematical model above is inerratic. So in order to better demonstrate the availability of the method we proposed, we design a more complicated mathematical model to conduct the simulation, which can be described by the formula: $z = a \sin (b x) \times \sin (c y)$ , where a = 0.2, b = c = 0.5. The ideal patterns we selected is the same as the preceding patterns in first case with 256 $\times$ 256 pixels, and they are expended to 400 $\times$ 400 pixels by zero padding. Other parameters are also the same as those used in the first case. The results of the numerical simulation are shown in Fig. 5 . After 30 iterations, the reconstructed patterns is shown in Fig. 5(b) and the value of RE decrease to only 3.15% and 2.99% as shown in Fig. 5(c). So it is proved furthermore that this method is effective to implement the intensity modulation on various sorts of curved surfaces.

Fig. 5 Numerical simulation of 3D patterns on complicated surfaces: (a) the intensity distributions of the original patterns, (b) the intensity distribution of the reconstructed patterns after 30 iterations, (c) the relationship between relative errors and the number of iteration.

Download Full Size | PDF

For the purpose to better demonstrate the feasibility of the method we proposed to modulate 3D intensity distributions on multiple curved surface, we construct a more complicated mathematical models which contains three curved surface ( $P_{1}$ and $P_{3}$ are cylindrical surfaces, $P_{2}$ is the same as the second case where a = 0.3, b = c = 0.5) to conduct the numerical simulation. The ideal patterns on surfaces $P_{1}$ and $P_{3}$ are the same as the preceding patterns, and the ideal pattern on surface $P_{2}$ is a new gray level pattern as show in Fig. 6(a) . All the patterns with 128 $\times$ 128 pixels are expended to 256 $\times$ 256 pixels by zero padding. The distance between the plane $H$ and the surface $P_{1}$ is d₀ = 1015 mm, the distance between these three curved surfaces are both 2 mm, the pixels of hologram is 768 $\times$ 768, the noise suppression parameter $γ = 1$ , the feedback parameter $k = 1$ and other parameters are also the same as those used in the first case. After 20 iterations, the reconstructed patterns is shown in Fig. 6(b) and the value of RE decrease to only 1.39%, 1.58% and 1.36% respectively as shown in Fig. 6(c). According to the results, the method is proved its feasibility to realize the 3D intensity modulation on multiple arbitrary curved surfaces simultaneously.

Fig. 6 Numerical simulation of 3D patterns with 128 $\times$ 128 pixels on multiple surfaces: (a) the intensity distribution of the original patterns, (b) the intensity distribution of the reconstructed patterns after 20 iterations, (c) the relationship between relative error and the number of iteration.

Download Full Size | PDF

4. Experiment and discussion

The method we adopt to conduct the experiment is the holographic projection technique. Figure 7 shows the schematic view of optical experimental setup. The laser with $λ$ = 532 nm (Oxxius 532-300-COL-PP-LAS-01462) is adopted as light source. A laser beam is spatial-filtered and collimated and then illuminates the SLM (HOLOEYE PLUTO), in which the pure-phase hologram is loaded to modulate the incident collimating laser beam. The 4-f system and a high-pass filter are adopted to remove the strong zero-order noise in the modulated light which reflected from the SLM. Through these setups, the desired intensity distributions of DOEs are generated and then expose the photopolymer on the surfaces of two cylindrical lens located at their corresponding position. The parameters of the photopolymer we use is as follows: the refractive index modulation is $Δ n$ = 0.04, the thickness is $15 μ m$ and the exposure is 15 mJ/cm². In the experiment, according to the power of exposing light, the exposing time we chose is 3 seconds. Finally, the desired DOEs will be fabricated on two pieces of photopolymer.

Fig. 7 Schematic view of the optical experimental setup.

Download Full Size | PDF

The two initial patterns we selected is the same as the patterns of the first case in section 3 with 256 $\times$ 256 pixels and expended to 400 $\times$ 400 pixels by zero padding. And the parameters in this experiment are as follows: L _1,2 = 12 mm, R _1,2 = 51.852 mm, $λ$ = 532 nm, the distance between two cylindrical surfaces is d = 2 mm, and the diffraction distances between cylindrical surface $P_{1}$ and the modulation plane is d₀ = 361 mm. The pixels of the SLM we adopt is 800 $\times$ 800 in the middle. The focal length of Fourier lens in the 4-f system is 500 mm, so the total distance between SLM and cylindrical surface $P_{1}$ is 2361 mm.

The results of the modulation on cylindrical surface $P_{1}$ and $P_{2}$ are shown in Fig. 8 & Fig. 9 . They are both captured by Canon EOS 5D Mark II. Figure 8(b) and Fig. 9(b) shows the partial enlarged images. The proportional dimension is also shown at the bottom of the images. According to these images of experimental results, the two patterns are effectively fabricated on two different curved surfaces simultaneously. So the feasibility of the method we proposed is demonstrated.

Fig. 8 (a) Picture of the binary pattern fabricated on cylindrical surface P ₁ and (b) its partial enlarged picture.

Download Full Size | PDF

Fig. 9 (a) Picture of the gray level pattern fabricated on cylindrical surface P ₂ and (b) its partial enlarged picture.

Download Full Size | PDF

However, some details of the produced profiles are not very perfect. The reasons might be as follows. At first, the dust in air cannot avoid during the experiments, which would reduce the quality of the pattern projected on photopolymer. Then, the non-uniformity and the unevenness of the photopolymer will influence the fabrication results. If the environment of experiment is improved, the results could be better.

5. Conclusion

In brief, we propose an effective method based on phase optimization to design the pure-phase distribution for fabricating multiple arbitrary DOEs on multiple curved surfaces. Since the proposed method can be applied to fabricate complex 3D patterns on multiple curved surfaces simultaneously, the multiple collimation of different curved surfaces in the system is avoided and the efficiency of DOEs’ fabrication is significantly improved. Moreover, the depth of field is larger in the method we propose. Both the numerical simulations and the optical fabrications are performed and they are in good agreement. And the curved surfaces can be more complicated according to the simulating results. The proposed method is easy and efficient. It is believed that it is an effective method to fabricate multiple arbitrary complicated DOEs on curved surfaces simultaneously, and it could be useful in various optical fields such as optical communication, optical interconnects, optical design, and so on.

Acknowledgments

This work was supported by the National Natural Science Founding of China (NSFC) (61575024, 61235002 and 61420106014), the National High Technology Research and Development Program of China (863 Program Grant No. 2015AA015905), and the National Basic Research Program of China (973 Program Grant No. 2013CB328801 and No. 2013CB328806).

References and links

1. G. Yoo, H. Lee, D. Radtke, M. Stumpf, U. Zeitner, and J. Kanicki, “A maskless laser-write lithography processing of thin-film transistors on a hemispherical surface,” Microelectron. Eng. 87(1), 83–87 (2010). [CrossRef]

2. D. Radtke, J. Duparré, U. D. Zeitner, and A. Tünnermann, “Laser lithographic fabrication and characterization of a spherical artificial compound eye,” Opt. Express 15(6), 3067–3077 (2007). [CrossRef] [PubMed]

3. Y. Xie, Z. Lu, and F. Li, “Fabrication of large diffractive optical elements in thick film on a concave lens surface,” Opt. Express 11(9), 992–995 (2003). [CrossRef] [PubMed]

4. D. Radtke and U. D. Zeitner, “Laser-lithography on non-planar surfaces,” Opt. Express 15(3), 1167–1174 (2007). [CrossRef] [PubMed]

5. H. Nishiyama, M. Mizoshiri, T. Kawahara, J. Nishii, and Y. Hirata, “SiO₂-based nonplanar structures fabricated using femtosecond laser lithography,” Opt. Express 16(22), 17288–17294 (2008). [CrossRef] [PubMed]

6. T. Wang, W. Yu, D. Zhang, C. Li, H. Zhang, W. Xu, Z. Xu, H. Liu, Q. Sun, and Z. Lu, “Lithographic fabrication of diffractive optical elements in hybrid sol-gel glass on 3-D curved surfaces,” Opt. Express 18(24), 25102–25107 (2010). [CrossRef] [PubMed]

7. P. Ruchhoeft, M. Colburn, B. Choi, H. Nounu, S. Johnson, T. Bailey, S. Damle, M. Stewart, J. Ekerdt, S. V. Sreenivasan, J. C. Wolfe, and C. G. Willson, “Patterning curved surfaces: template generation by ion beam proximity lithography and relief transfer by step and flash imprint lithography,” J. Vac. Sci. Technol. B 17(6), 2965–2969 (1999). [CrossRef]

8. R. J. Jackman, S. T. Brittain, A. Adams, M. G. Prentiss, and G. M. Whitesides, “Design and fabrication of topologically complex, three-dimensional microstructures,” Science 280(5372), 2089–2091 (1998). [CrossRef] [PubMed]

9. J. G. Kim, N. Takama, B. J. Kim, and H. Fujita, “Optical-softlithographic technology for patterning on curved surfaces,” J. Micromech. Microeng. 19(5), 055017 (2009). [CrossRef]

10. T. Senn, J. P. Esquivel, N. Sabate, and B. Lochel, “Fabrication of high aspect ratio nanostructures on 3D surfaces,” Microelectron. Eng. 88(9), 3043–3048 (2011). [CrossRef]

11. T. A. Savas, S. N. Shah, M. L. Schattenburg, J. M. Carter, and H. I. Smith, “Achromatic interferometric lithography for 100-nm-period gratings and grids,” J. Vac. Sci. Technol. B 13(6), 2732–2735 (1995). [CrossRef]

12. K. M. Baker, “Highly corrected close-packed microlens arrays and moth-eye structuring on curved surfaces,” Appl. Opt. 38(2), 352–356 (1999). [CrossRef] [PubMed]

13. A. J. Turberfield, M. Campbell, D. N. Sharp, M. T. Harrison, and R. G. Denning, “Fabrication of photonic crystals for the visible spectrum by holographic lithography,” Nature 404(6773), 53–56 (2000). [CrossRef] [PubMed]

14. K. Kintaka, J. Nishii, and N. Tohge, “Diffraction gratings of photosensitive ZrO₂ gel films fabricated with the two-ultraviolet-beam interference method,” Appl. Opt. 39(4), 489–493 (2000). [CrossRef] [PubMed]

15. J. Nishii, “Glass-imprinting for optical device fabrication,” in Advances in Optical Materials, OSA Technical Digest (CD) (Optical Society of America, 2009), paper AThC1.

16. A. Mizutani, S. Takahira, and H. Kikuta, “Two-spherical-wave ultraviolet interferometer for making an antireflective subwavelength periodic pattern on a curved surface,” Appl. Opt. 49(32), 6268–6275 (2010). [CrossRef] [PubMed]

17. R. Shi, J. Liu, J. Xu, D. Liu, Y. Pan, J. Xie, and Y. Wang, “Designing and fabricating diffractive optical elements with a complex profile by interference,” Opt. Lett. 36(20), 4053–4055 (2011). [CrossRef] [PubMed]

18. H. Zhao, J. Liu, R. Xiao, X. Li, R. Shi, P. Liu, H. Zhong, B. Zou, and Y. Wang, “Modulation of optical intensity on curved surfaces and its application to fabricate DOEs with arbitrary profile by interference,” Opt. Express 21(4), 5140–5148 (2013). [CrossRef] [PubMed]

19. T. Shimobaba, H. Nakayama, N. Masuda, and T. Ito, “Rapid calculation algorithm of Fresnel computer-generated-hologram using look-up table and wavefront-recording plane methods for three-dimensional display,” Opt. Express 18(19), 19504–19509 (2010). [CrossRef] [PubMed]

20. P. Tsang, W.-K. Cheung, T.-C. Poon, and C. Zhou, “Holographic video at 40 frames per second for 4-million object points,” Opt. Express 19(16), 15205–15211 (2011). [CrossRef] [PubMed]

21. X. Wang, J. Liu, J. Han, N. Zhang, X. Li, B. Hu, and Y. Wang, “3D optical intensity modulation on curved surfaces by optimization method and its application to fabricate arbitrary patterns,” Opt. Express 22(17), 20387–20395 (2014). [CrossRef] [PubMed]

22. A. Georgiou, J. Christmas, N. Collings, J. Moore, and W. A. Crossland, “Aspects of hologram calculation for video frames,” J. Opt. A, Pure Appl. Opt. 10(3), 035302 (2008). [CrossRef]

23. J. W. Goodman, Introduction to Fourier Optics (McGraw-Hill, 1996).

24. J. R. Shewell and E. Wolf, “Inverse diffraction and a new reciprocity theorem,” J. Opt. Soc. Am. 58(12), 1596–1603 (1968). [CrossRef]

25. S. Yan, Design of Diffractive Micro-optics (National Defense Industry Press, 2011), pp. 25, 70, 284 (in Chinese).

Design and fabrication of complicated diffractive optical elements on multiple curved surfaces

Abstract

1. Introduction

2. Basic principles

3. Numerical simulations

4. Experiment and discussion

5. Conclusion

Acknowledgments

References and links

Cited By

Figures (9)

Equations (2)

Optics Express