Novel Fabrication Method of Minute Cylindrical Structures Such as

Stents using Lithography, Etching, and Chemical Polishing

Toshiyuki Horiuchi

, Kaiki Ito

, Jun-ya Iwasaki

and Hiroshi Kobayashi

Tokyo Denki University, 5 Senju-Asahi-cho, Adachi-ku, Tokyo, Japan

Keywords: Fabrication Method, Cylindrical Structure, Stent, Lithography, Chemical Etching, Chemical Polishing,

Rotary Scan-projection Exposure, Stainless-steel Pipe.

Abstract: Applicability of characteristic subtractive processes of stainless-steel pipes to fabrication of minute structures

such as stents was demonstrated. Pipes with an outer diameter of 2 mm, a thickness of 50 µm, and a length of

50 mm were coated with a resist PMER N-CA3000 PM in approximately 10 µm thick in 20 mm area at the

tip parts of pipes. Next, stent-like mesh patterns composed of 30 rhombuses on an ordinary flat film reticle

were replicated on a pipe using a rotary scan-projection exposure system, in which patterns were precisely

and homogeneously replicated by synchronously scanning the reticle linearly in perpendicular to the pipe axis

and rotating the pipe at a constant speed. All the patterns on the reticle were continuously replicated during

the pipe was rotated 360º. After printing the stent-like mesh patterns, the pipes were processed in two steps.

In the first step, they were wetly etched in FeCl

aqueous solution, and in the second step, they were

chemically polished in a chemical compound on the market. As a result, a stent-like meshed pipe with mesh

widths of 83±6 µm was precisely fabricated.

1 INTRODUCTION

Diseases of blood vessels are one of the typical

illnesses in the present age. Choked, damaged, or

adhered vessels prevent the supply of fresh blood to

organs and tissues, and cause the necrosis and

degradation of organs. For this reason, effective

medical treatments have to be given in the early

stages of diseases.

Insertions of stents are considered as prospective

treatments (Varghese, 2015) (Rahal, 2014) (Caiazzo,

2015). Stents are cylindrical components with net-

like structures, and have appropriate elasticity and

rigidity. When the stents are inserted in blood vessels,

they should be folded or shrunk in diameter. On the

other hand, once they are inserted, they have to

support the blood vessels from inside and secure the

blood paths (Tammareddi, 2016).

For this reason, various methods have been

researched and developed for fabricating stents

superior in usability, functionality, and mechanical

properties. Knitting of wires or fibres (Rebelo, 2015)

and cutting of metal pipes using laser beams

(Kesavan, 2013) (Ando, 2017) (Nishi, 2013) are

https://www.dendai.ac.jp/en/graduate/

typical methods.

On the other hand, the authors have researched on

lithography systems for patterning on fine cylindrical

specimens such as pipes, wires, shafts, and others. For

this reason, as a typical application of the system and

related lithography and etching technologies,

fabrication of stent-like structures is investigated

here.

Generally speaking, patterning accuracies of

lithography and etching are higher than laser beam

cutting. In addition, it is expected that roughness of

part-edges obtained by lithography and etching is

improved by serially adding chemical polishing after

the etching. Based on these considerations, this

research has been performed.

As a lithography tool, a handmade rotary scan-

projection exposure system (Horiuchi, 2015)

(Horiuchi, 2016) is used. In this system, patterns on a

flat reticle in an oblong slit area is projected on a

cylindrical ridge of specimen pipe. When the reticle

is linearly scanned across the slit and the pipe is

rotated around its axis synchronously, patterns on the

reticle are continuously replicated on the cylindrical

surface of the pipe.

Horiuchi, T., Ito, K., Iwasaki, J. and Kobayashi, H.

Novel Fabrication Method of Minute Cylindrical Structures Such as Stents using Lithography, Etching, and Chemical Polishing.

DOI: 10.5220/0008997201690175

In Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2020) - Volume 1: BIODEVICES, pages 169-175

ISBN: 978-989-758-398-8; ISSN: 2184-4305

169

Lithography tools for printing on cylindrical pipes

have not been commercially available. Although

various ideas have been reported (Lee, 2011) (Haoa,

2011) (Lim, 2014), performances comparable with or

superior to those of our system have never been

shown for printing aimed complicated patterns.

After the pipes with stent-like patterns of resist are

etched in an aqueous solution of ferric chloride

(FeCl

), original shapes of stent-like meshed pipes

were fabricated. Fabrication methods up to this stage

have been already developed in the past researches of

authors (Ito, 2017) (Horiuchi, 2019).

However, in the past researches, the minimum

mean mesh width was 109 µm, and it was difficult to

reduce the widths without making locally broken

parts. For this reason, as a new idea, chemical

polishing process is added here. By polishing the

meshed pipes slowly and carefully, the mesh widths

are reduced down to 83 µm. The roughness of the

meshes is also improved.

2 FABRICATION PROCESS OF

STENT-LIKE MESHED PIPES

Meshed patterns were designed by connecting

rhombuses at their corners sequentially. The

rhombuses were arrayed 5 in the axial direction and 6

in the circumferential direction. Figure 1 shows the

reticle patterns designed by extending the aimed

meshed pipe shape. After 5×6=30 rhombuses were

regularly arrayed, connections at 12 corners were

intentionally separated for giving flexibility or

deformability to the meshed pipe. The rhombus

pattern width was 130 µm.

It was planned to fabricate this structure according

to the following processes, as shown in Figure 2. At

first, the meshed patterns of resist were printed on

Figure 1: Reticle patterns designed for printing stent-like

rhombus mesh.

(a) Resist coating

b) Rotary scan-projection exposure

(d) Etching in ferric chloride

(e) Chemical Polishing

Figure 2: Processes for fabricating the aimed structure.

Stainless-steel pipe

Resist

Exposure light

Slit

Reticle

Patterns

Linear scan

Stainless-steel pipe

Rotation

Resist

Projection lens

Etchant

Resist Developer

Stainless-steel pipe

Meshed pipe with resist

Polishing chemicals

Polished pipe

2 mm

Intentionally separated parts

BIODEVICES 2020 - 13th International Conference on Biomedical Electronics and Devices

170

SUS304 stainless-steel pipes using the rotary scan-

projection lithography. The contents of SUS304 are

shown in Table 1. The other part of the alloy is iron

(Fe). Main components are Fe, Cr, and Ni.

Table 1: Contents of SUS304 except iron.

Element Content

Cr 18.00-20.00

Ni 8.00-10.50

Mn ≤ 2.00

Si ≤ 1.00

C ≤ 0.08

P ≤ 0.045

S ≤ 0.030

In this lithography system, patterns on the reticle

were replicated on a pipe coated with a resist film

during the flat reticle was linearly scanned in the

horizontal direction, and rotating a pipe 360º

synchronously around its axis.

Next, the pipe was etched in an aqueous solution

of FeCl

using the mesh patterns of resist as etching

masks. After the etching, meshed pipe with the still

remained resist patterns on the outer surface was

polished in a chemical liquid compound on the

market.

Although the resist was almost removed during

the polishing, it remained partially. For this reason,

the polished pipe was cleaned finally by dipping in

acetone and adding ultra-sound wave vibration.

By these processes, the meshed pipes were

smoothed and homogeneously thinned. As a result,

precisely meshed pipes were obtained.

3 RESIST PATTERN PRINTING

Using the reticle, meshed patterns were printed on

SUS304 stainless-steel pipes with outer and inner

diameters of 2 and 1.9 mm and a length of 50 mm.

The thickness of the pipe wall was 50 µm. As a resist,

the negative PMER N-CA3000 PM (Tokyo Ohka

Kogyo) was used, and coated in approximately 10 µm

thick in 20 mm area from the tip of a pipe one by one.

Although positive resists had high resolution, it

was worried that the exposure doses between

sensitized or not would change critically, and the

exposure dose margin for the patterning became

small. In addition, it was feared that the vague

exposure influenced by the slit width and the pipe

curvature led up to the contrast degradation of pattern

images, and notable fluctuation of pattern widths and

thicknesses occurred. In contrast, widths and

thicknesses of the negative resist patterns tended to

saturate if the exposure dose was sufficiently given.

For this reason, it was thought that good patterning

homogeneity would be obtained. Besides, negative

resists with high transmittance were commercially

available, and thick resist patterning durable for long-

time etching with high speed stirring was applicable.

The reticle patterns were printed using a

handmade exposure system, as shown in Figure 3.

The typical exposure time for rotating the pipe 360º

was 30 s.

An example of resist patterns on a pipe is shown

in Figure 4. Mesh patterns are finely printed.

Measurement results of pattern widths are shown in

Figures 5 and 6. They show the width distribution in

the axial and circumferential directions, respectively.

It is known that pattern widths are very homogeneous.

The 3σ deviation was 12.0 µm.

However, the mean width was 208 nm, and very

wide comparing with the reticle pattern width of 130

µm. It was considered that this width increase was

caused by the following reasons. One reason is that

the actual projection ratio was not 1 but 1.13.

Figure 3: Handmade rotary scan-projection exposure

system.

Figure 4: An example of resist patterns on a pipe.

Rod lens for

illumination

Slit

Reticle

Projection lens

Stainless-

steel pipe

100 mm

1 mm

Gaps

Resist

Novel Fabrication Method of Minute Cylindrical Structures Such as Stents using Lithography, Etching, and Chemical Polishing

171

Figure 5: Width fluctuation of mesh resist patterns in the

axial direction.

Figure 6: Width fluctuation of mesh resist patterns in the

circumferential direction.

Accordingly, the projected width of mesh pattern

becomes 130×1.13=147 µm. In addition, because the

slit width was 800 µm and the mesh patterns were

projected on the curved surface of the pipe, the

pattern widths were slightly widened. Besides,

projected positions were also shifted at the both

outside ends of the slit. Since the slit width of 800 µm

was also magnified to 800×1.13=904 µm, the angle θ

between the ends A and B of the projected area from

the centre line are calculated referring to Figure 6,

θ=sin

-1

(0.904/2) =26.9º. (1)

Therefore, the pattern width printed at the both

outside ends of the slit becomes 147 µm /cos

=165

µm.

On the other hand, the arc length L of the

circumference MB of the pipe in the angle of

calculated to be

L=(26.9×π/180) ×1000 µm = 469 µm. (2)

Accordingly, the pattern at B is printed shifting

469-(904/2) =17 µm to the right outside. In contrast,

the pattern at A is printed shifting 17 µm to the left

outside. Accordingly, pattern width including the

vaguely exposed parts become 165+17+17=199 µm.

For this reason, if the resist is exposed sufficiently to

stabilize the pattern widths, the widths become more

than this width, and the widths shown in Figures 4 and

5 are within the reasonable range.

Figure 7: Figure for considering the pattern position shift

and size change on the curved pipe surface.

4 ETCHING AND CHEMICAL

POLISHING

Patterned pipes were etched in an aqueous solution of

FeCl

（ Sanhayato, H-1000A ） for 25 min at 40-

45ºC. The etchant was stirred by a propeller stirrer

rotating at 250 rpm. After the etching, the pipes were

rinsed by pure water. The patterns on the pipe surface

were remained as they were.

Next, etched pipes with the remained resist

patterns were chemically polished in a chemical

compound on the market (Sasaki Chemical, S-250)

heated on a hotplate at 93-95 ºC for 1.5 min. During

the polishing, the pipe was held by an L-shape tool

handmade with a hard plastic material, and moved in

Image of slit

904

Pipe

100

150

200

250

300

0 5 10 15

Series1

Series2

Series3

Resist pattern width (µm)

Position from the tip end (mm)

130

250

Rotation angle of the pipe (deg)

100

150

200

250

300

0 60 120 180 240 300 360

Series1

Series2

Series3

Resist pattern width (µm)

Rotation angle of the pipe (deg)

0.75

6.75

14.25

Position from the tip end (mm)

BIODEVICES 2020 - 13th International Conference on Biomedical Electronics and Devices

172

the chemical compound by swinging it using the tool.

The polishing was advanced at a speed of

approximately 7 µm/min in width directions.

On the other hand, the resist began to be peeled

off from approximately 40 s after beginning the

polishing. Therefore, it was supposed that the

polishing was advanced in the thickness direction also

after that.

After the polishing, the pipes were rinsed by pure

water. However, because resist fragments remained

on the polished pipes were observed, they were

removed by washing the pipes in acetone with ultra-

sonic vibration. Figure 8 compares stent-like meshed

pipes with and without the chemical polishing. In the

case of a pipe etched only by FeCl

, obtainable mean

mesh width was 109μm, as shown in Figure 8(a). In

contrast, when the chemical polishing was added, the

mesh-part width was reduced to 83 μm, as shown in

Figure 8(b).

(a) Pipe meshed by only etching in FeCl

(b) Pipe finished by polishing in chemicals.

Figure 8: Comparison of meshed-pipe preciseness between

with and without finishing by chemical polishing.

Besides, side wall roughness of the meshed parts

was decreased from around 10 µm to 5 µm, as

compared in Figure 9. It is known that the chemically

polished pipes became smooth and sheeny.

In Figures 8 and 9, photographs of meshed pipes

were taken by inserting a black hexagonal wrench

into the meshed pipes. For this reason, only the front

sides of the structures are shown. The vague

horizontal lines in the photographs are the ridges of

wrenches.

Next, to investigate the width fluctuation of stent-

like meshed parts, mesh widths were measured at 4

sides of all 30 rhombuses. Measured width

distributions are shown in Figures 10 and 11. It was

clarified that the widths were almost homogeneous in

both axial and circumferential directions. The 3σ

deviation of width was 6.2 µm. This fluctuation

(a) Pipe meshed by only etching in FeCl

(b)

Pipe finished by polishing in chemicals.

Figure 9: Comparison of meshed-pipe roughness between

with and without finishing by chemical polishing.

was far smaller than the 3σ deviation value of 13.5

µm obtained in the pipe shown in Figure 8(a) finished

without adding the chemical polishing in the

compound. In addition to the fact that the mean width

was reduced, width fluctuation was noticeably

decreased.

Components of S-250 are not made public. In the

instruction manual, it is written that the chemical is

composed of strong acid and gloss finisher. Judging

from the description of “nitrous acid gas is

generated”, it is supposed that the main strong acid is

nitric acid. The gloss polisher is a dangerous goods

class 4/class 2 petroleum. Pipe surfaces etched only

by FeCl

without the resist was rough and not sheeny.

The surface smoothness was inferior to that obtained

by the polishing.

The authors thought that the distortions of pattern

sizes and the shifts of printed positions could be

reduced by using a slit with a narrower width.

However, the illumination optics had to be improved

for collecting the light flux in the slit efficiently,

2 mm

1 mm

Novel Fabrication Method of Minute Cylindrical Structures Such as Stents using Lithography, Etching, and Chemical Polishing

173

Figure 10: Width fluctuation of meshes in the axial

direction after the chemical polishing.

Figure 11: Width fluctuation of meshes in the

circumferential direction after the chemical polishing.

because the exposure time was lengthened. In

addition, since the meshed part widths were largely

thinned by the undercut during the etching and

polishing, printed wide mesh patterns were rather

preferable.

Mechanical performance of the polished stent-like

parts have not been investigated yet, and should be

clarified hereafter. However, it was checked in the

past research that a similar meshed pipe without

polishing kept deformation linearity and elasticity

when the pressing force was less than 0.6 N at least.

Because the axial cross section of the stent-like part

is 2×15=30 mm

, pressing force of 0.6 N corresponds

to the differential blood vessel pressure of 0.6/30

N/mm

=0.02 MPa=200 hPa=150 mmHg, which is

approximately the maximum differential blood vessel

pressure.

5 CONCLUSIONS

Stent-like meshed pipes were successfully fabricated

by a novel method using lithography, wet etching and

chemical polishing. As a raw material, stainless-steel

SUS304 pipes with outer and inner diameters of 2 and

1.9 mm, respectively, were used.

At first, resist patterns were printed on a pipe

surface using a rotary scan-projection exposure

system. In the system, stent-like mesh patterns on a

reticle were continuously replicated on a resist film

coated on a pipe by synchronously scanning the

reticle linearly and rotating the pipe at a constant

speed. After the patterning, pipes with stent-like resist

patterns were etched in an aqueous solution of FeCl

Next, the meshed pipes were chemically polished

using a chemical compound on the market.

As a result, very fine meshed parts with a mean

width of 83 µm were obtained. Jogged edges of the

etched pipes were smoothed, and the 3σ deviation of

mesh widths was largely reduced from 13.5 µm to 6.2

µm. It was demonstrated that the subtractive process

of stainless-steel pipes would be applicable to

fabrication of minute complicated structures such as

stents.

ACKNOWLEDGEMENTS

This work was partially supported by JAPS

KAKENHI Grant Number 17K05021, the grants

from JI Engineering, and the grants from Shonan

Instruments.

REFERENCES

Ando, K., Ishii, K., Tada, E., Kataoka, K., Hirohata, A.,

Goto, K., Kobayashi, K., Tsutsui, H., Nakahama, M.,

Nakashima, H., Uchikawa, S., Kanda, J., Yasuda, S.,

Yajima, J., Kitabayashi, H., 2017. Prospective multi-

center registry to evaluate efficacy and safty of the

newly developed diamond-like carbon-coated cobalt-

chromium coronary stent system, Cardiovascular

Intervention and Therapeutics, Vol. 32, Issue 3, 225–

232.

Rotation angle of the pipe (deg)

100

150

200

250

300

0 5 10 15

Series1

Series2

Series3

130

250

Mesh width (µm)

Position from the tip end (mm)

100

150

200

250

300

0 60 120 180 240 300 360

Series1

Series2

Series3

0.75

6.75

14.25

Position from the tip end (mm)

Mesh width (µm)

Rotation angle of the pipe (deg)

BIODEVICES 2020 - 13th International Conference on Biomedical Electronics and Devices

174

Caiazzo, G., Fabris, K. E., Serdoz, O., Mattesini, A., Foin,

N., Rosa S. D., Indolfi. C., Mario, C. D., 2015. Absorb

bioresorbable vascular scaffold: What have we learned

after 5 years of clinical experience? International

Journal of Cardiology Vol. 201, 129–136.

Haoa, X., Wang, L., Wanga, O., Guoa, F., Tanga, Y., Dinga,

Y., Lua, B., 2011. Surface micro-texturing of metallic

cylindrical surface with proximity rolling-exposure

lithography and electrochemical micromachining.

Applied Surface Science. Vol. 257, 8906-8911.

Horiuchi, T., Fujii, H., Yasunaga, K., 2015, Lithography

onto Surfaces of Fine-Diameter Pipes Using Rotary

Scan-Projection Exposure. Journal of Photopolymer

Science and Technology, Vol. 28, No. 2, 273-278.

Horiuchi, T., Furuhata, T., Muro., 2016. Synchronous scan-

projection lithography on overall circumference of fine

pipes with a diameter of 2 mm. Japanese Journal of

Applied Physics, Vol. 55, 06GP13, 1-6.

Horiuchi, T., Ito, K., Kobayashi, H., Yanagida, A., 2019.

Investigation on stainless-steel stents fabricated using

projection lithography and wet etching, 2

EUROPEAN BIOSENSOR SYMPOSIUM (EBS

2019), FLORENCE, ITALY, 240.

Ito, K., Suzuki, Y., Horiuchi, T., 2017. Fabrication of

Cylindrical Micro-Parts Using Synchronous Rotary

Scan-Projection Lithography and Chemical Etching.

Proceedings of SPIE, Vol. 10454, 1045410, 1-6.

Kesavan, S., Strange, J. W., Johnson, T. W., Flohr-Roese,

S., Baumbach, A., 2013. First-in-man evaluation of the

MOMO cobalt-chromium, carbon-coated stent,

EuroIntervention, Vol. 8, No. 9, 1012-1018.

Lee, D., Hiroshima, H., Zhang, Y., Itoh, T., Maeda, R.,

2011. Cylindrical projection lithography for microcoil

structures. Microelectronic Engineering. Vol. 88,

2625–2628,

Lim, H., Choi, K., Kim, G., Lee, S., Park, H., Ryu, J., Jung,

S., Lee, J., 2014. Roll-to-roll nanoimprint lithography

for patterning on a large-area substrate roll.

Microelectronic Engineering. Vol. 123, pp. 18–22,

2014.

Nishi, S., Nakayama, Y., Ishibashi-Ueda, H., Yoshida, M.,

Yonetani, H., 2013. Treatment of rabbit carotid

aneurysms by hybrid stents (microporous thin

polyurethane-covered stents). Preservation of side-

branches, Vol. 28, issue: 7, 1097-1104.

Rahal, J. P., Dandamudi, V. S., Heller, R. S., Safain, M. G.,

Malek. A. l., 2014. Use of concentric Solitaire stent to

anchor Pipeline flow diverter constructs in treatment of

shallow cervical carotid dissecting pseudoaneurysms,

Journal of Clinical Neuroscience Vol. 21, 1024–1028.

Rebelo, R., Vila, N., Fangueiro, R., Carvalho, S., Rana, S.,

2015. Influence of design parameters on the mechanical

behavior and porosity of braided fibrous stents,

Materials and Design. Vol. 86, 237–247.

Tammareddi, S., Sun, G., Li, Q., 2016. Multiobjective

robust optimization of coronary stents, Materials and

Design. Vol. 90, 682–692.

Varghese, T. G., Thomas, S. C., Revanka, V. R., 2015.

Emergency stenting of unprotected left main coronary

artery occlusion using bare metal stent with good long-

term results, -A case report-, Journal of Indian college

of cardiology, Vol. 5, 345-348.

Novel Fabrication Method of Minute Cylindrical Structures Such as Stents using Lithography, Etching, and Chemical Polishing

175