Novel technique for high-quality microstructuring with excimer lasers

Stephan Roth*, Prof. Manfred Geiger

Chair of Manufacturing Technology, Univ. of Erlangen-Nuremberg, 91058 Erlangen, Germany

ABSTRACT

Laser micromachining has become increasingly established in many microsystems applications during the past years. These
new fields occasion higher demands on the quality of micromachined devices combined with high resolution and working
velocity. Due to the disadvantages of conventional excimer laser processing, a novel technique is required to meet these
demands. The main problems of conventional excimer laser machining are the redeposition of ablated material on the
irradiated work piece and the formation of a strong melting phase especially for metals. These difficulties greatly reduce the
applicability of excimer laser material processing for manufacturing microsystems technology components. By applying a
thin water film to the substrate surface, the redeposition of ablated material can be completely avoided, which results in a
better quality of the microstructures. Usage of a water film, however, has proved to lead to a marked reduction of the
ablation rate for the examined materials - ceramics and stainless steel. Therefore, one of the objectives of future research
will be to raise the ablation rate in order to render excimer laser processing more interesting economically. Adding alcoholic
additives, among others, has improved the wetting of the liquid films on the surface. The effect of the modified chemical
composition of the liquid on ablation rate and structure quality for various materials is presented here.

Keywords: Production process, excimer laser, microstructuring, structure quality

1. INTRODUCTION
The miniaturization of parts for diverse areas of technical applications increasingly determines the design of new products.
This trend also makes new demands on the respective manufacturing methods. A consequence of the increasing demand for
miniaturized work pieces is that the required processing quality has to be improved further. In many cases, laser irradiation
is a suitable tool for material processing. It can be used to machine a great variety of materials.1 Its importance keeps
increasingly especially in the area of microsystems technology. Here, ceramic materials are used more and more for parts
under stress from heat. Because of their special properties, ceramic materials are machined mainly with excimer lasers.
Since the absorption length of UV light is very short for most materials, the energy is deposited in a very thin surface layer.
The high pulse energy and the extremely short pulse duration of excimer lasers, typically a few 10 ns, lead to a sudden
melting, evaporation and plasma formation of the irradiated material. A large portion of the laser energy is immediately
removed with the ablated material so that the total heat transfer into the bulk material is relatively small. Therefore, this
method is very well suited for machining materials with high thermal shock and stress sensitivity, such as ceramics and
glass. In the case of metals, the material is mainly removed as molten metal. Sub-micron particles or melt droplets deposit
on or close to the already produced structures. Avoiding this redeposition is a requirement for improving the processing
quality of excimer laser microstructuring. The subsequent deposition removal —if at all possible — is very costly and rarely
economical.
Several approaches to avoid these material redepositions have been examined.2'3 A main focus of this research is the
application of an additional layer onto the surface to be treated. This can be a thin water film, which is sprayed onto the
surface, or a coat of lacquer, which has to be removed after processing. In the past, a thin water film was used, for example,
to remove impurities from waver surfaces with a laser beam.4 This method was then transferred to the processing of mostly
ceramic materials: a redeposition of material is to be avoided through the creation of a thin water film at the point of
processing.5 Because of the successes made so far in improving the structural quality, this approach is the subject of further
recent research.

*
Correspondence: Email: [email protected]; WWW: http:llwww.lft.uni-erlangen.de; Telephone: +49 9131 852-
7953; Fax: +49 9131 36403

In Laser Applications in Microelectronic and Optoelectronic Manufacturing V,

338 Henry Helvajian, Koji Sugioka, Malcolm C. Gower, Jan J. Dubowski, Editors,
Proceedings of SPIE Vol. 3933 (2000) • 0277-786X100/$15.OO

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 07/02/2015 Terms of Use: http://spiedl.org/terms

 2. EXCIMER LASER MICROSTRUCTURING
There are some significant differences between excimcr laser irradiation and the systems used mainly in the automobile
industry and mechanical engineering, such as CO? and solid state lasers. The short pulse duration, high pulse energy and
extremely small penetration depth of the radiation into the irradiated surfaces facilitate the machining ol ceramics and glass.
Another special characteristic is the large beam profile of this kind of laser. This makes it possible to create many
differently shaped structures at the same time. The so-called mask projection technique projects a mask containing the
geometrical information onto the surface of the work piece; the mask is either made of cut stainless sheet steel or chrome-
coated and structured silica glass substrate. In order to achieve on the work piece an energy density sufficient for the
processing. the area of the laser beam must be decreased with optical elements. Only high-quality mirror and lens systems
are used to avoid optical errors that could negatively influence the processing quality.
The length of the process is mainly determined by the number of pulses needed for structuring. Here, the ablation rates of
the different materials are important. For ceramics and steels they are between 0.1 and 0.2 urn per pulse. For polymers.
however, they can be up to I jim per pulse. Because of the low ablation rates, deep structures take a lot of time and arc
therefore often uneconomic. This is especially true for ceramic and metallic materials.
Figure 1 shows the different processing areas that are distinguished in the excimer laser machining of ceramics. Here, the
ablation threshold is of fundamental importance. It marks the energy density value where ablation of the material begins.
Below this threshold, the substrate material remains virtually unaffected, only organic residues (grease, oil) left in surface
indentations are removed. Additionally, the irradiation is reflected from the walls of existing cavities, which can lead to a
sufficient energy density at the bottom of these indentations. Both phenomena evidently increase the roughness of the
substrate. When the ablation threshold is exceeded slightly. the roughness peaks (if the work piece surface inch. This
material flows into the existing indentations, which decreases the roughness. Thus. excimer laser irradiation can be used to
selectively modify the roughness.
When the ablation threshold of the processed material is exceeded distinctly, extensive ablation results. Because of reduced
projection achieved through mask projection, excimer laser machining can realize structures measuring only a few
micrometers. Especially high-quality lenses make it possible to create even structures measuring within the dimensions of
the utilized laser wavelength. This technique is used above all in the production of semiconductors.

energy density below ablation threshold

surface modification
//'
:¼. '/ /1/,

/ ////, ///z / roughening

-: energy density above ablation threshold

'" z.'! vJYtN

/
initial surface
energy density distinctly above
ablation threshold

microstructuring 2
I'.—

microstructure

Fig. I: Processing of ceramics by excimer laser radiation1

339

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 07/02/2015 Terms of Use: http://spiedl.org/terms

 3. MACHINING UNDER THIN WATER FILM
The findings on the removal of impurities on silicium wafers were used as a basis for experiments on the improved ablation
of ceramic materials. Here, though, the water film is continuously sprayed onto the vertical work piece; moreover, spraying
on the water film and machining coincide. This makes sure that drained and evaporated fluid is continually replaced. A
disadvantage of this method is that atomizing of the water causes a thick fog to develop at the machining point. This fog
weakens and disperses the excimer laser beam, which in turn reduces the quality of the machined structure.

3.1. Materials

The experiments regarding the qualification of excimer laser ablation under a sprayed-on liquid film on one the hand use
materials that can be structured very well with excimer laser radiation, such as different ceramics. On the other hand, they
concentrate on materials whose structurization with an excimer laser has so far been problematic, but which play an
important part in many industrial applications. This is especially true for steels: a distinct melting phase often occurs during
their processing.
The following ceramics were chosen for the experiments: aluminum oxide (Al203, FRIALIT-DEGUSSIT F 99,7,
FRIATEC AG), silicon carbide (SiC, Ekasic D) and silicon nitride (Si3N4, Ekasin S, both Elektroschmelzwerk Kempten
GmbH). As so-called high-performance ceramics, they are used in many areas of vehicle construction, mechanical
engineering and machine-making. A high-speed steel 1.3343 (S 6-5-2) was used as steel material. It is utilized mainly for
cutting and forming tools.

3.2. Experimental setup

The laser treatment was carried out with KrF (X 248 nm) and XeCl (X = 308 nm) excimer laser radiation. Here, the KrF
laser stands out due to its higher pulse frequency of 150 Hz at a pulse energy of 800 mJ. The XeC1 laser in contrast has a
pulse energy of 2 J at a maximum pulse frequency of 20 Hz. High quality beam controls with beam homogenizers are
available for both laser systems. They work according to the mask projection method. This makes it possible to create
structures measuring only micrometers.
A working chamber with a closed liquid cycle was constructed for the experiments. From a reservoir, the liquid mixture, in
a pre-defined chemical composition, is pumped into the working chamber and there atomized using an atomizer pistol. The
lance of the atomizer pistol can be exchanged to produce differently sized droplets. The liquid is sprayed onto the vertical
work piece, collected in the chamber and piped back to the reservoir through a filter unit. The strong atomization creates a
fog in the working chamber, which fouls the optical element such as the imaging lens. Therefore, the chamber is closed up
with a UV permeable silica glass top at the point of entry of the light beam. This avoids liquid droplets on the lens. The
silica glass top can be cleaned easily after the processing. In this way, a negative influence on the beam properties is
reduced. In order to keep the spray from overly dispersing the laser light over the work piece, the focus of the spray jet is
placed over the actual point of processing. Thus, the laser beam hits the surface of the work piece within the draining water
jet. The thickness of the film is determined by the amount of water being sprayed on.

Fig. 2: Working chamber for applying liquid film during laser irradiation

340

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 07/02/2015 Terms of Use: http://spiedl.org/terms

 3.3. Ablation behavior

The experiments with a water film sprayed on during excimer laser ablation showed that the reposition of ablated work
piece particles can be avoided nearly completely. There are no melt particles or warps either around or in the structures.
Figure 3 shows SiC structures produced both in the conventional, dry way (left) and with a sprayed-on liquid film (right).
The dry-processed structure shows strong impurities. The original surface of the work piece is completely covered with
material deposits around the structures. After processing under the water film, however, the surface remains practically
unchanged.

sic

[KrF excimer laser

wavelength: A = 248 nm
number of pulses: N = 500
pulse frequency: I,, = 20 Hz

Fig. 3: Conventional dry processing of SiC (left): machining of SiC under thin tap water film (right)

The ablation rate is an important variable in excimer laser processing. Figure 4 shows a comparison of the ablation rates of
different ceramics for dry processing and when using a water film (tap water). As the energy density grows, the ablation
rates steadily increases once the ablation threshold has been exceeded. Then, a saturation value is reached at about 0.2 to 0.3
m per pulse. Further raising the energy density increases the ablation rate only negligibly. For most ceramics, this behavior
does not change much when a water film is used, though the ablation rate level is pushed toward lower values. Al2O is an
exception to this rule, here, the ablation rate increases considerably. The quality ol' the structure, however, decreases
distinctly for this material.5 There is no projection of the structures on the mask. The structures created arc characterized by
a cratered bed. Evidently, larger chunks of the material broke away during lasering.

0,3

pulse

0,2

0,15

0,1

0,05

0 0
0 5 10 15 20 i/cm'
energy density H —4' energydensityH —0
KrF excimer laser:
—ci— AI,O, —0— SiC
wavelength: A = 248 nm
number of pulses: N = 500 —h-— SiN, —0.— Zr02
pulse frequency f. = 20Hz

Fig. 4: Ablation rates for structuring under thin tap water titm

341

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 07/02/2015 Terms of Use: http://spiedl.org/terms

 During observation of the ablation process it is observed that both acoustic and light emissions decrease drastically during
lmcrostructuring under a water fIlm in contrast to conventional processing. Around the irradiated surface area. the water is
vaporized explosively, and accelerated away from the surface of the work piece. The sequence of events can he made
visible by short-time recordings (Figure 5). For the experiments, the samples were irradiated with an excimer laser Ironi
straight above after a thin water film had been deposited on the horizontal samples. The recordings show that fog and
ablated material do not break the water film until 5 to 10 ps after the laser pulse has ended. At a high speed. they arc
catapulted out of the processing area vertically. This process is finished after about 500 ps. If you compare this temporal
sequence with that of conventional excimer laser processing (start of ablation already during the excinicr laser pulse. after
about 5 to 10 ns, and fading of the plasma light after about 2.5 ps) . a considerable delay of ablation phenomena becomes
clear. An explanation f'or this behavior could he that super heated water with a temperature (it noticeably more than l(XYC
can exist in an instable state for a certain time before it finally vaporizes explosively ("defervesccncc"),
At the water-work piece interface, the water is heated and supcrhcated suddenly. Defervescence does not occur until after a
temporal delay of a few ps — long after the laser pulse has finished -, and the superheated water vaporizes explosively. The
higher the energy density of the laser pulse, the higher the superheating of the water, and the higher the blast pressure that
has to be expected. Because of the water film's cooling, the steam bubbles implode after a certain period of time. The
implosion shock waves can cause cavity erosions.
Absorption, reflection and dispersion losses caused by the water film and water droplets as well as changed ablation
mechanisms are possible explanations for the usually lower ablation rates. Additionally, the cooling of the irradiated surface
by the water film could cause a slightly lower ablation rate. On the other hand, the absence of processing plasma allows
more laser energy to reach the work piece surface than conventional irradiation. Up 30% of laser energy are absorbed in the
plasma in excimer laser processing.7

r=1Ojs r=2OpS r4Ops

XeCl excimer laser

waveIength 2. = 308 nm
energy density H, = 25 J/cm

tap water film

Al,O,
rl5Ops t=500ps

Fig. 5: Behavior of the water film during laser irradiation (short-time shutter camera; exposure timeS ns)

While during conventional excimer laser machining, melting and vaporization of the material and finally appearance of a
plasma occur immediately after the beginning of the laser pulse. contact with a water film evidently suppress these events.
Enclosing the interaction zone in a water film, or the latter's cooling effect possibly make the vaporization and sublimation
of the material much more difficult. This could explain the absence of processing plasma. But since sonic laser processed
work piece surfaces show structures that suggest a solidified melting phase. is it safi to assume that at least some of the
irradiated material melts and finally evaporates.

342

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 07/02/2015 Terms of Use: http://spiedl.org/terms

 3.4. Mechanical Stability

During excimer laser processing with a water film the explosive evaporation of the water causes strong blast waves —at
least according to today's understanding of ablation mechanisms. This can possibly damage the marginal zone of the
processing area, which might negatively influence the stability of ceramic work pieces. Therefore, transverse rupture tests
are conducted with A1203 and Si3N4 ceramics. Both materials where chosen because of their totally different ablation
behaviors which suggest different ablation mechanisms. With A1203 ablation rate increases greatly when processed under
water. At the same time, many break outs can be seen at the edges of the machined microstructures, which suggest a greater
mechanical stress caused by the processing. This makes it safe to assume that the marginal zone is damaged as well. Si3N4,
however, does not show these effects which gives reason to hope that the stability of the work piece is not influenced
negatively. The results show that excimer laser processing decreases the stability of the materials when compared to their
unprocessed initial state. With Si3N4, however, processing under water results in a slightly higher stability than conventional
excimer laser processing. The blast waves caused by the water' s evaporation therefore do not necessarily lead to a
weakening of the work piece. The Al203 ceramic shows contrary results. Here, processing under water drastically decreases
the stability.

4. LIQUID MIXTURES WITH ALCOHOLIC ADDITIVES

A even liquid film is also necessary for a good processing result. In order to achieve this, the surface of the work piece must
be wetted evenly. Optimizing the wetting behavior of the utilized liquid mixtures is another important goal of the
experiments. The angle of contact between the liquid and the substrate is to be reduced by adding appropriate additives.
This is possible, for example, with alcoholic (methanol or ethanol) or non-alcoholic additives.

4.1. Wetting behavior

In order to attain a thinner and more even liquid film, additives are to improve the wettability of the liquid on the substrate
surface. The formation of droplets can locally change the thickness of the film, which can directly influence the processing
result by leading to a decreased quality of the structure. In order to avoid this formation of liquid droplets on the substrate
surface, the surface tension of the water is reduced with alcoholic additives. It is well-known that low amounts of methanol
or ethanol can significantly reduce the surface tension of water. The tests were done with a contact angle measuring system,
which optically analyses the angle between the lateral face of a liquid droplet and the substrate surface. Figure 6 shows that
as the wetting angle decreases significantly the concentrations of both the ethanol (C2H60) and the methanol (CFIO)
additive increase. Aluminum oxide and silicon carbide were chosen as substrate materials, since both materials show very
different surface properties. Reservations that the irradiation could lead to heavy combustion reactions as the share of
alcohol in the liquid was too high led to the decision that the water-alcohol mixture should contain a maximum of 5 %
alcohol. For both investigated materials (Al203 and SiC), the wetting angle is a maximum of 75° with this percentage of
alcohol.
I00

AI2OjdistiIIed water and methanol

) 60
—fr-.- SiC/distilled water and methanol

° Al2Ojdistilled water and ethanol

I 40
c _____—, .-4-— SiC/distilled water and ethanol

20

0
0 10 20 30 40 % 60
additives concentration

Fig. 6: Influence of methanol and ethanol on wetting angle

343

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 07/02/2015 Terms of Use: http://spiedl.org/terms

 What must not be forgotten, however, is that is not only the wetting of the liquid that is change but possibly the transmission
behavior as well. The addition of additives may not impair the transmission for the wavelengths in use. Otherwise, it would
lead to a partly irregular absorption of the irradiation both in the water fog and in the liquid film on the surface. In order to
avoid this unwanted influence, the mixtures were examined with regard to their transmission. It turns out that the
transmission does not decrease continually as the share of alcohol in the mixture increases. It is also established, however,
that there are only minimal differences in transmission between the different mixtures. At the KrF excimer laser's
wavelength (X = 248 nm) transmission is between 80 % and 85 %. At this wavelength, the differences in transmissions for
methanol are even smaller than for ethanol.

4.2. Ablation rate

As has been discussed extensively in the previous chapter, the spraying-on of water can significantly increase the quality of
the structure in excimer laser microstructuring. Problems that have so far occurred in ablation processing (material
redeposition; formation of melting) are avoided for almost all examined materials. However, the ablation rates of most
ceramic and metal materials are greatly reduced. Therefore, the essential goals of further tests are to increase the ablation
rates and the quality of the structure for the processing of aluminum oxide.
The results presented so far were attained with simple tap water. Therefore, distilled water is used in the following
experiment. In order to improve wetting, 5 % of methanol and ethanol, respectively, are added. The low amount of additives
can be explained by the fact the pre-tests showed that a higher amount of alcohol can lead to heavy combustion reactions
during irradiation. The resulting flame has a similar effect as the plasma that is produced during conventional processing.
The irradiation is greatly absorbed and dispersed, the processing results are distinctly worse and inhomogeneous.
The results for distilled water are similar to those attained with tap water. The ablation threshold is pushed towards higher
energy density values. The reposition of material is avoided completely with the use both of methanol and ethanol. Figure 7
shows, however, that the ablation rates are similar to those of tap water. This is true for Al203 as well. The structure
qualities of the examined materials are comparable as well.

tte__
I
0,14

_ • S
J/m2 10 0 2 3 4 5 6 7 8 J/cm2lO
energy density HE —k
1
energydensityHE —
—4-- AI,03, distilledwater KrF excimer laser: .. SiC, disthled water
wavelength: ? = 248 nm
—,-- AI,O,, distilled water + 5 % methanol sc, datld water + 5 % methanol
number of pulses: N = 500
—0— 203, distilled water + 5 % ethanol pulse frequency: f, = 20 Hz —U— SiC, distilled water + 5 % ethanol

Fig. 7: Ablation rates for structuring under distilled water with additions of ethanol and methanol

4.3. Roughness
When structuring under a liquid film, the structure ground is roughened strongly, as Figure 8 shows on silicon nitride. Some
of the resulting micro-cavities show a distinct orientation. Depending on the application, this increased roughness of the
structures can lead to an unwanted limitation of work piece properties. On the other hand, the formation of oriented cavities
offers the possibility of using this orientation to optimize bearing work pieces, for example. Therefore, it is a possible
approach to avoid or purposely create the phenomena for the respective application. This is not only true for the roughness

344

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 07/02/2015 Terms of Use: http://spiedl.org/terms

 within the structures hut also and especially for modification of the roughness of the entire work piece surface. The
roughness values R, measured within the structures and as presented in Figure 9 do not show a significant dependency on
the used liquid. Moreover, the results are similar for all examined materials. All roughness values are slightly higher than
those of conventionally machined structures.

SLN,

KrF excimer laser

wavelength ,. = 248 nm
number of pulses N = 500
pulse frequency f, = 20 Hz

Fig. : Surface formation at the structure ground of a SiN4 sample

8,0

urn

4.0

2.0

0,0
0,0 2,0 4,0 6,0 J/cm' 10,0 0,0 2.0 4,0 6,0 J/cm 10,0

energy density H1 —0' energy density H —÷

KrF excimer laser:

—4-— Si,N. —-'- AlO, —4-— SI.N, —U-- A1,O,
wavelength: , — 248 nm
—v— steel 1.3343 —A— SiC number of pulses: N 500 —,-— steel 1.3343 —A—— SiC
pulse frequency I, = 20 Hz

Fig. 9: Roughness value R for different materials

5. CONCLUSION

The influences of the energy density of the utilized laser irradiation and the composition of the liquid on the processing
under a sprayed-on water film were examined for different ceramics (SiC, SiN4, A1203) and steel. Besides ablation rate and
structure quality, the focus was on the attainable degree of surface roughness. Similar to "dry" cxcimer laser processing. the
energy density of the laser irradiation considerably influences the processing result. It is decisive for the ablation rates of the
different materials. It is, however, reduced distinctly when a liquid film is used during irradiation. This was proven l'or both
tap water and the examined ethanol and methanol mixtures, whose goal was to improve wetting. It became clear that the
original approach to achieve an increased processing quality and higher ablation rates by improving the wetting of' water on
the substrate surface did evidently not have the desired effect. However, so far only alcoholic additives (ethanol and
methanol) were used, which can cause heavy combustion reactions during irradiation when their concentration is too high.
In first tests, the use of soapy additives, which can also decrease the surface tension, showed that a strong loam builds

345

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 07/02/2015 Terms of Use: http://spiedl.org/terms

 during atomization, which disqualifies this possibility. Because of the inadequate results for an improved wetting, another
approach is going to be the use of liquids that show a reduced wetting on the substrate. This prevents a premature draining
of the liquid film before the next laser pulse, which can occur at high spray pressures.

ACKNOWLEDGEMENTS
Financial support for this research was given by the Deutsche Forschungsgemeinschaft (DFG) within the
Schwerpunktprogramm ,,Mikromechanische Produktionstechnik 1998" under reference number Ge 530/29-3. The authors
wish to thank Mrs. A. Topperwien for her extensive and helpful work within this research project.

REFERENCES

1. W. Becker, Oberflachenausbildung und tribologische Eigenschaften excimerlaserstrahlbearbeiteter

Hochleistungskeramiken, Reihe Fertigungstechnik, Band 95, Hanser, München, 1999.
2. P. Burck, W. Reitzenstein, H.K. Tönshoff, and H. Kappel, "Mikrostrukturierung von Keramiken in reaktiven
Gasatmosphären mit UV-Lasern", ECLAT'96, F. Dausinger, H.W. Bergmann, and J. Sigel, pp. 699-706, AWT e.V.,
Wiesbaden, 1996.
3. H. Kappel, Oberflachenmodzfikation technischer Keramik mit Excimer-Laserstrahlung, Fortschr.-Ber. VDI Reihe 2
Nr. 469, Düsseldorf, 1998.
4. W. Zapka, W. Ziemlich, and A.C. Tam, "Efficient pulsed laser removal of 0.2 sm sized particles from a surface",
App!. Phys. Lett. 58, pp. 2217-2219, 1991.
5. M. Geiger, S. Roth, and W. Becker, "Microstructuring and surface modification by excimer laser machining under
thin liquid films", Laser Surface Processing, V.1. Pustovoy, SPIE Vol. 3404, pp. 200-208, Limoges, 1997.
6. N. Lutz, Oberflachenfeinbearbeitung keramischer Werkstoffe mit XeC1-Excimerlaserstrahlung, Reihe
Fertigungstechnik, Band 39, Hanser, MUnchen, 1994.
7. 0. Gedrat, Strukturierung technischer Keramik mit Excimer-Laserstrahlung, Fortschr.-Ber. VDI Reihe 2, Nr. 276,
Düsseldorf, 1992.

346

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 07/02/2015 Terms of Use: http://spiedl.org/terms