Mask-Shifting-Based Projection Lithography for Microlens Array Fabrication
Round 1
Reviewer 1 Report
This manuscript proposed a new approach for fabrication of microlens array. By employing a projection-based mask-shift filtering technique, the authors show the ability to achieve lower linewidths and higher surface figure accuracy compared to conventional methods. It is well illustrated by the data in the manuscript. However, there are still some comments that need to be considered.
1. “Direct-write lithography techniques comprise electron beam direct-write lithography, ion beam direct-write lithography, and two-photon laser direct-write lithography.” The author's manuscript is incomplete in describing the fabrication methods for microlens arrays. For example, in addition to the three direct writing techniques mentioned in the text, the most widely used technique in hard and brittle materials is the etching-assisted femtosecond laser maskless direct writing technique (Laser & Photonics Reviews，11，1600115 (2017), Laser & Photonics Reviews 13，1800272 (2019)).
2. The authors choose the 0.2x lithography objective lens. Can higher fabrication accuracy be achieved by changing the objective lens?
3. The innovation of the method proposed in this manuscript should be fully described in the text.
4. there are some format issues. For example, the format of the figure notes and references is not harmonized. “Figure 1 a) … Figure 3 (a)….” ”Ref. 16 19 23 24.”
Author Response
- “Direct-write lithography techniques comprise electron beam direct-write lithography, ion beam direct-write lithography, and two-photon laser direct-write lithography.” The author's manuscript is incomplete in describing the fabrication methods for microlens arrays. For example, in addition to the three direct writing techniques mentioned in the text, the most widely used technique in hard and brittle materials is the etching-assisted femtosecond laser maskless direct writing technique (Laser & Photonics Reviews，11，1600115 (2017), Laser & Photonics Reviews 13，1800272 (2019)).
Re: Thanks for your advice. We have added the related references in the resubmitted manuscript. These references are respectively listed as [14], [15], and [16] in References.
- The authors choose the 0.2x lithography objective lens. Can higher fabrication accuracy be achieved by changing the objective lens?
Re: The critical dimension (CD) of feature pattern in the mask determines the CD of final exposure pattern in the wafer for the proximity contact lithography, because it has constant exposure characteristics. And the 0.2x lithography objective lens can accomplish the smaller CD than the proximity contact lithography method in this study. In addition, the manufacturing precision can be addressed by the high-precision motion control in the projection lithography equipment. This approach not only improves the processing accuracy and CD, but also reduces the cost of mask processing.
- The innovation of the method proposed in this manuscript should be fully described in the text.
Re: Thanks for your advice. The innovation of the method proposed in this manuscript have been described in the last paragraph of chapter Introduction. Meanwhile, the superiority of the proposed method can be confirmed by the experiment.
- there are some format issues. For example, the format of the figure notes and references is not harmonized. “Figure 1 a) … Figure 3 (a)….” ”Ref. 16 19 23 24.”
Re: We apologize for our negligence. We have carefully checked and revised these format in this resubmitted manuscript.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Review comments for “Mask Shifting-Based Projection Lithography for Microlens Array Fabrication”
To editor and authors:
The focus of this paper is a novel approach for microlens array fabrication using a projection lithography process with mask shifting. The research method used in this paper is the experimental method. The results of this research study represents the achieved high-fidelity surface figure accuracy of the microlens array. But this study cannot provide the complete the results of the experiments. (Ex. The 3D profile measurement and surface roughness of the microlens array.)
I would like to recommend its publication upon a major revision. Some comments could be found as follows.
1. Is the microlens array a spherical lens or the aspheric lens?
2. In figure 7, the unit is presented in micron?
3. The micro-lens has cross-sectional profile measurement, but there is no actual measurement with optical microscope or SEM or 3D profile measurement.
4. What application of the thickness of the lens produced in this paper is only 2.5um?
Author Response
- Is the microlens array a spherical lens or the aspheric lens?
Re: Thank you very much for your question. The method proposed in this manuscript can enable processing of the spherical lens and aspheric lens.
- In figure 7, the unit is presented in micron?
Re: The unit is presented in micron in Figure. Thanks for your advice.
- The micro-lens has cross-sectional profile measurement, but there is no actual measurement with optical microscope or SEM or 3D profile measurement.
Re: Thanks for your advice. We have added a series of measurement data by 3D profile measurement in Figure 9 and Figure 10 of this resubmitted manuscript.
- What application of the thickness of the lens produced in this paper is only 2.5um?
Re: The vector height of the lens produced by the proposed approach in this paper is 2.5 um, and the microarray lens can be processed on a glass substrate with a thickness of 1 mm. It can be employed in the optical imaging field, such as Hartmann sensors.
Author Response File: Author Response.docx
Reviewer 3 Report
This work proposed a fabrication method for microlens array by combining projection lithography with mask shifting to achieve better surface figure accuracy and surface roughness compared with contact lithography approach. The work is well organized and presented. However, I do have a couple of questions to be addressed before any further consideration.
1. First of all, the authors claimed "surface roughness at the nanometer level" in the Abstract. But no experiment data is shown. The profiles in Figure 7/8 can't show the nanometer level roughness. Is the nanometer level roughness really possible after the development of photoresist? Please clarify.
2. The phrase "surface figure" is a bit confusing. By comparing Figure 7 and 8, the authors claim that the proposed method is better in "surface figure"? Can the authors quantify that? What does it mean? Sharper profile? In Line 289, what's "higher quantization order"?
These two conclusions seem to be key of the work. They should be clearly addressed.
3. The introduction part about microlens (array) fabrication should be improved. More recent literatures should be included, instead of just works before 2010 or even 2000.
For example:
Inkjet printing
Fabrication of Microlens Arrays with High Quality and High Fill Factor by Inkjet Printing. https://doi.org/10.1002/adom.202200677.
Electrohydrodynamic jet printing
Drop-on-Demand Electrohydrodynamic Jet Printing of Microlens Array on Flexible Substrates. 2023. https://doi.org/10.1021/acsapm.3c00054.
UV-assisted nanoimprinting
Curvature-Adjustable Polymeric Nanolens Fabrication Using UV-Controlled Nanoimprint Lithography. Micromachines 2022, 13 (12). https://doi.org/10.3390/mi13122183.
Thermal reshape
Wavelength-Scale Lens Microscopy via Thermal Reshaping of Colloidal Particles. Nanotechnology 2012, 23 (28), 285708. https://doi.org/10.1088/0957-4484/23/28/285708.
4. The description of section 2.1.
* Equation (1), should be right hand side be m^2
* Equation (6), f(x, y)?
5. I still don't understand how the authors solve the orthogonal errors. Can the authors show some experiment data?
6. Quite some typos. Please do the proof more carefully.
* Line 55, binary overlayl?
* Line 172, which figure?
* Line 226/229, missing of )? It makes the sentences difficult to understand.
* Figure 6, there are two (a)-(d)
* Figure 7, there are extra ")"?
* Line 255, mthod
* Line 257, proposed method
* Line 299, propose a with
* Could be a few more...
Many typos. Should be improved.
Author Response
- First of all, the authors claimed "surface roughness at the nanometer level" in the Abstract. But no experiment data is shown. The profiles in Figure 7/8 can't show the nanometer level roughness. Is the nanometer level roughness really possible after the development of photoresist? Please clarify.
Re: Thanks for your advice. The roughness (RMS) of the microarray lens can reach the nanometer level. And the surface data of these microarray lens measured by the 3D profile instrument have been shown in Figure 9 and Figure 10 of this resubmitted manuscript.
- The phrase "surface figure" is a bit confusing. By comparing Figure 7 and 8, the authors claim that the proposed method is better in "surface figure"? Can the authors quantify that? What does it mean? Sharper profile? In Line 289, what's "higher quantization order"?
These two conclusions seem to be key of the work. They should be clearly addressed.
Re: The surface shape refers to the contour shape of the microstructure surface. The quantization order refers to the number of strip-shaped division areas of the target microstructure during mask design. Since the proposed method based on projection lithography utilizes a 0.2x lithography objective lens for exposure imaging, under the same circumstances, the microstructure outline size of the mask surface is 5 times that of ordinary close contact lithography equipment. Therefore, this article with the proposed method, the number of target microstructure strip segmentation regions can be more.
- The introduction part about microlens (array) fabrication should be improved. More recent literatures should be included, instead of just works before 2010 or even 2000.
Re: Thanks for your advice. We have added the related references in the resubmitted manuscript. These references are respectively listed as [19], [28], [29], and [31] in References.
- The description of section 2.1.
* Equation (1), should be right hand side be m^2
* Equation (6), f(x, y)?
Re: Thanks for your advice, and we apologize for our negligence. We have carefully checked and revised these problems in this resubmitted manuscript.
- I still don't understand how the authors solve the orthogonal errors. Can the authors show some experiment data?
Re: For traditional mobile exposure devices, the two moving directions of X and Y cannot be absolutely 90°, and there will always be a certain degree of orthogonality error ∆θ, and the deviation of the moving direction will cause two problems in the micro-pattern: 1) causing the pattern changes from a square clear aperture to a diamond-shaped clear aperture; 2) leading to serious misalignment of graphics units.
- Quite some typos. Please do the proof more carefully.
Re: We apologize for our negligence. We have carefully checked and revised these problems in this resubmitted manuscript.
Round 2
Reviewer 2 Report
Review comments for “Mask Shifting-Based Projection Lithography for Microlens Array Fabrication”
To authors:
This study was provided the complete the results of the experiments. (Ex. The 3D profile measurement and surface roughness of the microlens array.) But no responses some questions.
Some comments could be found as follows.
How to confirm this method proposed in this manuscript can enable processing of the spherical lens and aspheric lens?
No
Author Response
How to confirm this method proposed in this manuscript can enable processing of the spherical lens and aspheric lens?
Re: Thanks for your question. The difference between spherical lenses and aspheric lenses is mainly due to the different changes in the radius of curvature. For the two types of lenses, the proposed calculation model of feature pattern in the manuscript can calculate the required feature pattern distribution. And the processing of these two types of lenses can be manufactured using the mask-shifting method. To verify the two types of lenses, the 3D profilometer scanning method can be used to conduct high-precision detection of the lens surface shape.
Reviewer 3 Report
Thanks the authors for addressing my questions carefully.
One final question on the surface roughness in Figure 9 and Figure 10. It seems that the surface roughness becomes higher when lens aperture is small. Can the authors provide some explanation? Is it because sharper spatial change of light intensity when the lens aperture is small?
Other than that, all my questions have been well addressed. The manuscript can by published in its current format.
Author Response
One final question on the surface roughness in Figure 9 and Figure 10. It seems that the surface roughness becomes higher when lens aperture is small. Can the authors provide some explanation? Is it because sharper spatial change of light intensity when the lens aperture is small?
Re: Thanks for your question. As the aperture of the lens continues to decrease, its processing requirements become more and more stringent. According to processing requirements, the balance between aperture and surface roughness cannot currently be balanced. This causes in that the smaller the aperture, the more difficult it is to control the surface roughness. In addition, surface roughness is also related to the imaging quality of the imaging system.
