A Study on Double-sided Optical Focusing Alignment of Transparent

Substrate

Chia-Lien Ma

, Chih-Chung Yang, Yu-Hsuan Lin and Kuo-Cheng Huang

Taiwan Instrument Research Institute, National Applied Research Laboratories, 20 R&D RD. VI, Hsinchu, Taiwan

Keywords: Auto Focus, Optical Inspection, Assembly Error Detection.

Abstract: The non-contact optical focus positioning technology proposed in this study uses the principle of optical

imaging to develop a system that is simple, fast, and has a microscopic image inspection function. This study

uses a dual optical path design for the upper and lower systems. The near-semi-reflective substrates in the

system have different reflectivity coatings. The energy density of the transmitted and reflected light is

captured through image processing. The position shift or tilt of the components is then used to calculate the

energy density difference of the upper and lower systems. It can be quickly converted into focus position,

system uniformity, surface height, component assembly offset and tilt. The technology of this research can

overcome the problems of manual focus, additional device focus module and human eye misjudgement, and

provide a measurement method and tool with simple operation and high accuracy. It can effectively adjust the

Z-axis offset range of the lower system can be effectively adjusted by ±1 mm, the lower system tilts range

±1 ̊, and the substrate tilts range ±1 by the difference of the upper and lower energy densities. The Y-axis

offset range is ±0.1 mm by the maximum energy density of the upper system.

1 INTRODUCTION

As semiconductor technology continues to evolve,

line widths continue to decrease and manufacturing

processes become more complex. At present, the

industry has entered the 7 nm process stage and

requires nearly a thousand processes. Advanced

processes and complex processes are becoming more

and more demanding for measuring speed and

accuracy, and the demand for automated optical

measuring equipment for precision components is

increasing.

Automated optical inspection (AOI) is a high-

speed and high-precision optical image detection

system. It uses machine vision as a detection standard

technology to improve the shortcomings of human-

based detection (Wang et al, 2019). The application

level includes high-tech industry research and

development, manufacturing quality control, national

defense, life, medical, environmental protection,

electricity, etc. AOI is a common detection technique

in industrial processes. Optical instruments are used

to obtain the surface state of the sample, and image

processing technology is used to detect flaws.

https://orcid.org/0000-0003-4681-0351

Because it is non-contact testing, semi-finished

products can be inspected during the process (Chon et

al, 2001). High-precision optical image detection

system, including measurement lens technology,

optical lighting technology, focus positioning

measurement technology, electronic circuit testing

technology, image processing technology and

automation technology application, etc., its

development and application not only meet the

development needs of high-tech industry, but also can

be extended to the defence military industry (Zureik

et al, 2010). The analysis and research of military

weapons manufacturing, night vision combat

systems, and strategic topography are closely related

to this imaging technology.

Among them, the focus positioning measurement

technology is often applied to accurately adjust the

image position, and there are many setups and

technologies that can apply this technology. The

focus positioning measurement system is roughly

divided into two kinds of systems. Most of the

systems need to be set up with another microscope.

When the system focus is confocal with the focal

plane of the microscope, the surface image is

obtained, and the distance is calculated by geometric

Ma, C., Yang, C., Lin, Y. and Huang, K.

A Study on Double-sided Optical Focusing Alignment of Transparent Substrate.

DOI: 10.5220/0008915500800085

In Proceedings of the 8th International Conference on Photonics, Optics and Laser Technology (PHOTOPTICS 2020), pages 80-85

ISBN: 978-989-758-401-5; ISSN: 2184-4364

principle (Browne et al, 1992)(Maly et al, 1994).

However, if the sample is a penetrating material, the

reflection signal of the lower interface will be so weak

that it is difficult to judge the location of the interface.

Therefore, only surface focusing can be performed,

and the focused optical path and the microscope

imaging optical path are not coaxially designed, so

the two focal points cannot overlap. In another

system, a chromatic confocal detection module or a

laser rangefinder is combined with an external

module to perform positioning and single point

position measurement. First move the external

module to the detection position, define the detection

line, and finally detect it (Tiziani et al, 1994)(Kim et

al, 2013). In addition, the distance between the two

modules needs to be corrected in advance, which

requires an additional calibration time, resulting in a

slow measurement speed.

In view of the fact that these measurement

systems require a human eye, manual focus or an

external focus module to obtain a clear position

image, it will result in reduced measurement

efficiency and accuracy. Therefore, this research

develops a measurement module with microscopic

image inspection and autofocus function, and has the

characteristics of fast, easy operation and high

longitudinal measurement range. In this study, a self-

developed focusing module was designed, which is

designed for dual light paths in the upper and lower

systems. Combined with a microscope objective, it

features microscopic image detection for quick

alignment of measurement locations. Using the

principle of optical imaging, the anti-reflective

coating of the near-half-reflective substrate allows the

energy density of penetration and reflection to be

approximately the same at the focal position. By

calculating the difference in energy density between

the upper and lower systems, which is caused by the

shift or tilt of the lower system and the near-half-

reflective substrate, the focus position information,

system uniformity and surface height can be

instantaneously obtained, and fast autofocus

operation can be performed. At the same time, after

rapid focusing, the assembly error of the system or

component can be known by the change in energy

density of the upper and lower systems.

2 EXPERIMENTAL PRINCIPLE

AND SETUP

A lens is a combination of multiple lenses that correct

aberrations such as astigmatism, coma, field

curvature, distortion, and spherical aberration by the

shape, thickness, and curvature of the lens and the

spacing between the lenses. The lens can be viewed

as a combination of two spherical surfaces, so the

imaging results can be obtained using a single

spherical imaging formula. The lens imaging path is

shown in Figure 1. It is assumed that the lens is placed

in a uniform medium, and the medium is assumed to

be air. The light source is irradiated onto the object

and refracted through the lens to focus the image.

Figure 1: Focus image.

The imaging formula, also known as the lens-

maker’s formula, is as follows (Nayar and Nakagawa,

1994):

𝑆



𝑆



𝑓

(1)

where 𝑆



is the object distance, which is the distance

between the object and the lens; 𝑆



is the image

distance, which is the distance between the image and

the lens; and f is the focal length of the lens. It can be

seen from the lens-maker’s that a relative image

distance can be obtained from one object distance,

and vice versa. When the sample is in the object plane,

the outline of the sample is focused by the lens and

forms a sharp image. The image at this time is called

a focus image. When the sample is not in the object

plane, a circle is formed in the image plane according

to the trigonometric principle. The blurred image was

out of focus at the time.

The optical design is shown in Figure 2. The

design concept is the upper and lower double light

path. The upper system includes a CCD for capturing

images, a focusing microscope objective, and a near-

half reflecting substrate. The lower system consists of

a light source, a mirror, a CCD and a focusing

microscope objective. This model is created by using

a single lens to represent the focusing microscope

objective lens, and it is selected as a standard product.

The purpose is to establish the model in the simplest

way. If a good focusing effect can be achieved with

this structure, a more accurate result can be obtained

Object

Image

Lens

𝑆



𝑆



A Study on Double-sided Optical Focusing Alignment of Transparent Substrate

by replacing the focusing microscope objective during

actual measurement. Since only surface reflection is

considered, the focus position is designed on the lower

surface of the near-half reflecting substrate.

Figure 2: Schematic diagram.

The Figure 2 shows, the light source transmits

through the half mirror to the focus lens, focusing on

the lower surface of the near semi-reflective substrate,

which is the focus position. Due to the dual optical path

design, the light will have two paths. A part of the light

is reflected by the lower surface and returned to the

CCD of the lower system. The other part of the light

continues to propagate through the near-half reflecting

substrate. After passing through the focusing lens,

focus on the CCD of the upper system.

The wavelength of the simulated light source is 550

nm representing the visible light band. The mirror is

50% transmitted and 50% reflective. The lens has a

curvature of 10 mm, the thickness is 5 mm, the material

is BK7, and no surface coating. The near-half

reflecting substrate has a length of 10 mm and a width

of 10 Mm, thickness 2 mm, material BK7. In this paper,

non-sequential optical tracking does not consider

optical wave effects. Because the diffraction and near-

field effects caused by the objects in the experimental

structure are negligible. In the process of optical

simulation, the material and interface have been

correctly set, so the calculation of Fresnel equations

will be considered and calculated by the FRED optical

simulation software.

3 EXPERIMENTAL RESULTS

AND DISCUSSION

In order to obtain a clear image when the CCD of the

upper and lower systems is in the focus position, the

upper surface and the lower surface of the near-half-

reflective substrate are coated, which is determined by

simulation. The clarity of the image is positively

correlated with the optical energy density, so the

energy density can be used as a basis for judging the

clarity of the image.

The simulation results are shown in Figure 3. First,

the reflection and transmittance of pure glass materials

are analysed. At the focus position, the energy density

received by the upper system is much larger than that

received by the lower system, which is about 33.5

times. In other words, in the case where the upper and

lower surfaces of the near-half-reflective substrate are

not coated, only the upper system can capture a clear

image, and the lower system can only receive a very

weak and blurred image. In order to improve the

excessive difference in energy density between the

upper and lower systems, the reflective coatings on the

lower surface of the near-half-reflective substrate were

simulated at 30%, 60%, and 80%, respectively. Among

them, in the case of 30% and 80% reflective coating on

the lower surface coating, the difference in energy

density between the upper and lower systems is 3.46

and 0.37 times, respectively. Only one of the systems

can receive clear images at the focus position, not both

systems simultaneously. For both systems to capture a

clear image, the lower surface needs to be plated with

a 60% reflective coating to receive a clear image at the

same time, as shown in Figure 3. This coating

specification is ideally feasible, but in fact, the coating

quality and mechanism error must also be considered,

so it needs to be measured to determine.

In order to calibrate the assembly error of the

system and components, the CCD and lens of the

above system are used as the reference to simulate the

four assembly conditions, which are the lower system

is displaced in the Z-axis direction, shifted in the Y-

axis direction, the angle θ is tilted with the light source

as a centre, and the near-half-reflective substrate is

tilted at an angle ψ from the centre of the substrate, as

shown in Figure 4.

PHOTOPTICS 2020 - 8th International Conference on Photonics, Optics and Laser Technology

Figure 3: Energy density change plots for different coating

systems.

In the case of different reflective coatings, the

lower system is offset in the Z-axis direction, and the

Z-direction offset can be adjusted by the difference in

energy density between the upper and lower systems,

and the range is ±1 mm, as shown in Fig. 5 (a). (b),

difficult to adjust compared to other cases where there

is a coating. Wherein, as observed in Figure 5, the left

and right curves are asymmetrical because the focus

position will be in the near semi-reflective substrate

when the lower system is closer to the upper system;

when the lower system is far away from the upper

system, the focus position will be in the air, and due

to the refractive index of the substrate and air are not

the same, so the energy density near and away from

the system is asymmetric.

Figure 4: System and component assembly error types.

(a)

(b)

(c)

(d)

Figure 5: Substrate (a) uncoated, (b) 30%, (c) 60%, and (d)

80% reflective coating, lower system energy density

variation.

Since the energy density of the upper and lower

systems can be made the same when the lower surface

of the substrate is 60% reflective coating, it is used as

0,00

50,00

100,00

150,00

200,00

250,00

300,00

350,00

0 1020304050607080

Energy Density (W/mm2)

Lower surface AR coating (%)

Lower system

Upper system

0,00

50,00

100,00

150,00

200,00

250,00

300,00

350,00

-1 -0,75 -0,5 -0,25 0 0,25 0,5 0,75 1

Energy Density (W/mm2)

Relative position of the lower system and the substrate (mm)

Lower system

Upper system

0,00

50,00

100,00

150,00

200,00

250,00

-1 -0,75 -0,5 -0,25 0 0,25 0,5 0,75 1

Energy Density (W/mm2)

Relative position of the lower system and the substrate (mm)

Lower system

Upper system

0,00

20,00

40,00

60,00

80,00

100,00

120,00

140,00

-1 -0,75 -0,5 -0,25 0 0,25 0,5 0,75 1

Energy Density (W/mm2)

Relative position of the lower system and the substrate (mm)

Lower system

Upper system

0,00

20,00

40,00

60,00

80,00

100,00

120,00

140,00

160,00

180,00

200,00

-1 -0,75 -0,5 -0,25 0 0,25 0,5 0,75 1

Energy Density (W/mm2)

Relative position of the lower system and the substrate (mm)

Lower system

Upper system

A Study on Double-sided Optical Focusing Alignment of Transparent Substrate

a standard sheet. When the system is offset in the Y

direction, the Y-direction offset of the lower system

can be adjusted by the maximum energy density of

the upper system, and the range is ±0.1 mm, as shown

in Figure 6.

Figure 6: System energy density change diagram under the

Y-axis offset of the lower system.

When the lower system tilts the θ angle with the

light source as the centre, the tilt of the lower system

can be adjusted by the difference in energy density

between the upper and lower systems, and the range

is ±1 ̊, as shown in Figure 7.

Figure 7: System energy density change diagram at lower

system tilt.

When the near-half-reflective substrate is tilted at

an angle ψ from the centre of the substrate, the tilt

angle can be determined by the change of the energy

density of the lower system, and the range is ±1 ̊, and

the substrate uniformity can be detected, as shown in

Figure 8.

Figure 8: System energy density change diagram of

substrate tilt.

Based on the above simulation results, in an actual

setup, a power meter and profiler is used to measure

the energy density difference between the upper and

lower systems. If the energy density of the lower

system is greater than the upper system, the tilt angle

of the substrate must be adjusted. If the energy

density difference is greater than 90, the Y-axis

displacement of the lower system must be adjusted; if

the energy density difference is greater than 20, the

Z-axis displacement of the lower system must be

adjusted first; if the energy density difference is less

than 20, the tilt of the lower system must be adjusted

angle.

4 CONCLUSIONS

The substrate coating simulation results show that the

60% coated substrate is used as the calibration

standard film, so that the upper and lower systems can

capture clear images. In the assembly error simulation

results, the Z-axis offset range of the lower system

can be effectively adjusted by ±1 mm, the lower

system tilts range ±1 ̊, and the substrate tilts range ±1

by the difference of the upper and lower energy

densities. The Y-axis offset range is ±0.1 mm by the

maximum energy density of the upper system.

Further, the uniformity of the substrate can be

detected as a result of the tilt of the substrate.

The focus positioning technology proposed in this

study is a self-developed non-contact optical

measurement module that overcomes the problems of

the traditional autofocus measurement module and

can be widely used in autofocus. It is suitable for

single point height, uniform measurement and system

assembly error detection of microstructure.

0,00

20,00

40,00

60,00

80,00

100,00

120,00

140,00

-0,1 -0,075-0,05-0,025 0 0,025 0,05 0,075 0,1

Energy Density (W/mm2)

Y-axis offset position of the lower system (mm)

Lower system

Upper system

0,00

20,00

40,00

60,00

80,00

100,00

120,00

140,00

-1 -0,75 -0,5 -0,25 0 0,25 0,5 0,75 1

Energy Density (W/mm2)

Tilt angle of the lower system (∘)

Lower system

Upper system

0,00

20,00

40,00

60,00

80,00

100,00

120,00

140,00

-1 -0,75 -0,5 -0,25 0 0,25 0,5 0,75 1

Energy Density (W/mm2)

Tilt angle of the substrate (∘)

Lower system

Upper system

PHOTOPTICS 2020 - 8th International Conference on Photonics, Optics and Laser Technology

ACKNOWLEDGEMENTS

The authors would like to express their appreciation for

financial aid from the Ministry of Science and

Technology, R.O.C under grant numbers MOST 108-

2221-E-492-019, MOST 108-2218-E-492-010 and

MOST 108-2622-E-492-009-CC3. The authors would

also like to express their gratitude to the Taiwan

Instrument Research Institute of National Applied

Research Laboratories for the support.

REFERENCES

Wang, X., Zheng, Z., Fan, Y., Lai, M., Wei, J., Wu, X., “A

focus automatic positioning system of high-power laser

beam based on plasma ultraviolet radiation,” High

Power Laser and Partical Beams, Vol. 31, 2019.

Chon, S. M., Choi, S. B., Kim, Y. W., Kim, K. W., Lim, K.

H., Choi, S. Y., and Jun, C. S., 2001. “Development of

automated contact inspection system using in-line CD

SEM,” IEEE, 399-401.

Zureik, E., Lyon, D., Abu-Laban, Y., Surveillance and

Control in Israel/Palestine: Population, Territory and

Power, Routledge, 2010.

Browne, M. A., Akinyemi, O., and Boyde, A., 1992.

“Confocal surface profiling using chromatic

aberration,” Scanning, Vol. 14, 145-153.

Maly, M. and Boyde, A., 1994. “Real-time stereoscopic

confocal reflection microscopy using objective lens

with linear longitudinal chromatic dispersion,”

Scanning, Vol. 16, 187-192.

Tiziani, H. J., and Uhde, H. M., 1994. “3-Dimensional

image sensing by chromatic confocal microscopy,”

Appl. Opt., Vol. 33, 1838–1843.

Kim, T., Kim, S. H., Do, D., Yoo, H., and Gweon, D., 2013.

“Chromatic confocal microscopy with a novel

wavelength detection method using transmittance,”

OPTICS EXPRESS, Vol. 21, 6286-6294.

Nayar, S.K., and Nakagawa, Y., “Shape from focus

system,” 1994. IEEE, Vol. 16, 824-831.

A Study on Double-sided Optical Focusing Alignment of Transparent Substrate