Implementation of a Large-Area Diffractive Lens Using Multiple Sub-Aperture Diffractive Lenses and Computational Reconstruction

Round 1

Reviewer 1 Report

The authors have employed Golay-like synthetic aperture technique to record higher resolutions using diffractive lenses (multiple) with smaller bandwidth (diffractive lenses with higher bandwidth exceeds lithography capabilities). Then the reconstructions were performed numerically using synthesised PSF using LRR algorithm. The authors have found that Golay-like recording when combined with LRRA reconstruction provides better resolution than Golay-like recording combined with NLR reconstruction. The reported simulation and experimental results verify the findings. The following are my comments.

 

1. Golay-like configuration is compared with Gloay-like configuration (Fig.4 to Fig.7). Other bandlimited diffractive lens configurations are compared separately (Fig.1). Will it not be more reasonable to do a cross comparison between using Golay-like and the bandlimited lens configurations. If it is difficult to fabricate the other bandlimited lenses then atleast adding simulation reconstructions of bandlimited lenses (shown in fig.1(b,c)) to Fig.5 will provide better understanding of the benefits of 'Golay-like + LRRA' systems. The above addition will also help in validating the authors claim of 'overcoming lithography based limitations' in diffractive optics.

 

2. It will be good to include details such as size (radius), pitch, grayscale-levels etc of the fabricated LADISA to section-4.

 

3. The improvement in resolution has been qualititaively verified (Fig.9). Is it not possible to quantify the results? Since authors are using USAF test object, I guess this should be an easy addition (using the binary line patterns).

 

4. It makes me think that an object of size 0.12 x 0.11 mm (shown in Fig.9) will be resolved the same (or even better) when looking high-res microscope (ie. direct imaging). I may be wrong since the group number used in the USAF target is not mentioned by the authors. Will it not be be a better idea to first quantitatively set a targeted object resolution (using USAF), that is beyond 'direct imaging' and 'lithography' capabilities then do the experiments to verify them. This will also put the paper in a strong foothold (since Golay systems are known for super-resolution). 

 

I see the paper is well written, easy to understand and proposes a novel concept supported by experimental results.  But I see there is space for further improvements (1~2), and quantitative verification of 'higher-resolution beyond lithography and direct-imaging limitations' (3~4), will add significant impact, which I encourage the authors to do.  

 

Author Response

Please check the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The article is devoted to the important problem of increasing the resolution of imaging systems. The main idea of the article is to use small sub-aperture diffractive lenses instead of one large one. To avoid the problem of precision adjustment of the mutual position of diffractive lenses, it is proposed to make each lens off-axis so that the images formed by them do not intersect on the camera sensor. The images are then processed by different methods in such a way as to restore the original image with high resolution. The article has a well-developed mathematical part, but the experimental part is based on a very mediocre experimental technique with low resolution (about 20 µm). Micrographs of the diffractive zones show large sampling errors due to the fact that the annular diffractive zones are formed by a device operating in Cartesian coordinates. One can say that the purpose of the experiment is to prove the feasibility of the idea. Therefore, the spatial resolution of the technology is not so important. BUT, the choice for approbation of the idea of a technology with a spatial resolution that is 1-2 orders of magnitude worse than widely available technologies significantly reduces the impression of the experimental part of the article. It would be better to explain that so low fabrication resolution is enough for the task.

As a justification for the use of sub-aperture lenses, the thesis is put forward that it is difficult to manufacture large diffractive lenses. Apparently, the author's experience is based on electron-beam lithography. In the case of using laser lithography [1] and diamond turning [2], it is quite feasible to manufacture rotationally symmetric diffractive lenses up to 800 millimeters in size [3]. Therefore, the authors should more thoroughly substantiate the areas of application of sub-aperture diffractive lenses.

1. Veiko, V.P., Korolkov, V.P., Poleshchuk, A.G. et al. Laser technologies in micro-optics. Part 1. Fabrication of diffractive optical elements and photomasks with amplitude transmission. Optoelectron. Instrument.Proc. 53, 474–483 (2017). https://doi.org/10.3103/S8756699017050077

2. https://www.lightpath.com/capabilities/diamond-turning/

3. Paul D. Atcheson, Chris Stewart, Jeanette Domber, Kevin Whiteaker, Jerold Cole, Peter Spuhler, Aaron Seltzer, Jerald A. Britten, Shamasundar N. Dixit, Brandon Farmer, Lensey Smith, "MOIRE: initial demonstration of a transmissive diffractive membrane optic for large lightweight optical telescopes," Proc. SPIE 8442, Space Telescopes and Instrumentation 2012: Optical, Infrared, and Millimeter Wave, 844221 (21 September 2012); https://doi.org/10.1117/12.925413

In the abstract there is an error “One such methods…”

 "One such methods is the widely used indirect imaging 20 method with Golay configuration telescopes."

Figure 7. “…(b) optical microscope and (c) SEM image…” It is necessary to change “…(b) optical microscope image and (c) SEM image…” The picture doesn’t show OPTICAL MICROSCOPE.

 

Author Response

Please check the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

I find the authors have convincingly addressed all the reviewers concerns in the revised version. I recommend the paper to be accepted for publication in the current format.