FULL PAPER

www.aem-journal.com

Fabrication of High Aspect-Ratio Surface Micro Patterns

on Stainless Steel using High-Speed Direct Laser
Interference Patterning
Valentin Lang,* Bogdan Voisiat, Tim Kunze, and Andres Fabián Lasagni

where tailored topographies with micro- and

This work addresses the fabrication of periodic micro patterns on stainless nano-features lead to super-hydrophobic
steel by means of Direct Laser Interference Patterning. A modular optics properties.[9–11] In medical technology, the
assembly is introduced which produces two-beam interference patterns biocompatibility of implants in the human
body has been improved by well-defined
within a 5
100 μm rectangular shaped laser spot. The optical assembly
surface textures with surface roughness’s
considers optimal utilization of the available laser power in order to enable from some tens of nanometers[12–14] up to
high processing rates. Using different optical elements, the spatial period of several micrometers.[15,16] In addition,
the periodic structures was set to 8 μm and 19 μm. The effect of the microstructures are known to reduce the
processing parameters on the morphology of the produced patterns is bacterial adhesion,[17] which permits pre-
vention of serious infections (for example
investigated by means of confocal microscopy, scanning electron microscopy,
with Staphylococcus,[18–20] Escherichia
and energy-dispersive X-ray spectroscopy. It is found that the pulse repetition coli,[19,20] Pseudomonas aeruginosa,[19] Pseu-
rate shows a strong influence on the structure height accompanied by domonas fluorescens,[21] and diatom attach-
considerable heat accumulation effects, which is reported for the first time ment[22]). In the field of large-area optical
for this method. In addition, aspect ratios slightly higher than one could be applications, such as organic light emitting
realized, which is also reported for the first time when using Direct Laser diodes (OLED)[23–25] or photovoltaic devi-
ces,[26–28] well-controlled diffraction effects
Interference Patterning with ns-pulsed lasers. Particularly for the smaller
introduced by micro-gratings are used to
periods, the melt flow impairs pattern formation due to the overmelting of improve their efficiencies. The height of the
the material (humping). Moreover, variations of the surface chemistry are periodic topography features (respectively
observed, especially for higher accumulated laser fluences. the aspect ratio) is often of key importance
for the surface function.[29–31]
Nowadays, various methods are avail-
able for the fabrication of high-resolution
micro-patterns down to the nanometer
1. Introduction scale such as lithography or electron beam technologies. They
typically require high invest and operation costs accompanied by
It is well known that periodic micro-patterns on surfaces can
restrictions on component dimensions and geometry as well as
influence and improve different properties. A prominent example
low processing speeds. Consequently, there is a lack on
is the shark skin, which has shown that repetitive surface patterns
industrially-relevant one-step methods permitting high process-
with feature sizes on the micrometer scale can significantly reduce
ing speeds.
drag (friction between the fluid and the surface).[1–3] These surface Direct Laser Interference Patterning (DLIP) technology has
features have been successfully reproduced in different materials, turned out as a cost-efficient tool for the mass production of
including steels,[4,5] polymer,[6] acrylic,[7] and textiles.[8] Other bio- surface structures in the micrometer and submicrometer
inspired examples are self-cleaning surfaces such as the lotus leaf, range.[32] The DLIP technology takes advantage of a periodic
intensity distribution, obtained from two or more superimposed
V. Lang, Dr. B. Voisiat, Prof. A.F. Lasagni
coherent beams. In the case of two-beam interference, a line-like
Technische Universität Dresden intensity pattern is obtained where the spatial period Λ is
George-Bähr-Strasse 3c typically defined by the used laser wavelength λ and angle of
01069 Dresden, Germany incidence α between impinging laser beams, as shown in
E-mail: valentin.lang@tu-dresden.de Equation (1).
V. Lang, Dr. T. Kunze, Prof. A.F. Lasagni
Fraunhofer-Institut für Werkstoff- und Strahltechnik (IWS) λ
Winterbergstraße 28, 01277 Dresden, Germany Λ¼ : ð1Þ
2  sinα =2
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/adem.201900151.
This allows resolutions below those of conventional methods
DOI: 10.1002/adem.201900151 such as micro milling[33,34] or direct laser writing.[35]

Adv. Eng. Mater. 2019, 1900151 1900151 (1 of 9) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.aem-journal.com

The use of short pulse lasers (pulse duration >1 ns) results 2. Experimental Section
in a considerable amount of melting of the substrate material
2.1. Materials
due to heat diffusion. As a result, the topographies achievable
with DLIP using ns-pulsed lasers are generally limited in
their aspect ratios (ratio between structure depth and spatial Stainless steel substrates (1.4031, X5CrNi18-10) with a thickness
period) to up to 0.5.[36] While high-aspect-ratio structures are of 0.9 mm were electro-polished obtaining a surface roughness
already possible using ps pulsed lasers in combination with (Sa) of 53 nm. Before the laser experiments, the metallic
the DLIP technology,[37] ns laser sources are still very substrates were cleaned using a lint-free cellulose paper with
interesting for industrial high-speed structuring applications isopropyl alcohol.
due to the lower invest costs and higher available laser
powers. Thus, there is a demand for structuring solutions to
structure metals with higher aspect ratios using DLIP with 2.2. Direct Laser Interference Patterning
ns-pulsed lasers.
This work focuses on the high-speed structuring of stainless The laser experiments were carried out using a nanosecond
steel employing ns-pulsed laser radiation using DLIP. In pulsed slab-shaped solid-state laser (Edgewave IS400) operating
addition, for the first time, heat accumulation during the DLIP at its fundamental wavelength λ of 1064 nm. This system
process is investigated when utilizing ns laser pulses. permits to change the repetition rate from single pulse to
Rectangular shaped laser spots permit high surface processing 10 kHz, with a maximum average power of 150 W (@ 10 kHz
rates by enabling the treatment of larger surface areas per laser repetition rate). The pulse duration τ was 10 ns. In the performed
pulse.[38,39] Therefore, an optical DLIP setup is designed, experiments, the laser fluence was kept constant at 3.2 J cm2.
permitting to form the interfering laser beams into a This can be done by setting the pulse energy to 16 mJ (at the
rectangular shape containing a well-defined interference materials’ surface) considering that the area which is irradiated
pattern. A similar approach has been already successfully by each laser pulse is 0.1
5 mm2. In consequence, depending
applied to achieve exceptionally high structuring speeds close to on the used repetition rate, the laser power level was adjusted in
1 m2 min1.[40] By using different optical elements, the spatial order to keep this value constant, considering optical losses in
period of the micro-structures was set to 8 and 19 μm. The the beam path of 33% (0.1 kHz: 2.4 W, 1 kHz: 24.2 W, 2 kHz:
influence of the processing parameters (laser fluence, repeti- 48.4 W, 5 kHz: 121 W).
tion rate, feed speed) on the morphology of the produced The interference patterns were generated using an optical
patterns is investigated. Additionally, the melt flow character- assembly developed by Fraunhofer IWS (see Figure 1). By using
istics as a result of heat accumulation as well as the amount of two bi-prisms, the laser beam is split into two sub-beams which
molten material are taken in account. The obtained surfaces are are overlapped on the sample surface. Additionally, the laser
examined by Confocal Microscopy (CFM) and Scanning beams are formed into a rectangular shape of 5 mm x 100 μm on
Electron Microscopy (SEM). Moreover, energy dispersive X- the sample surface by first expanding the laser beam in one
ray spectroscopy (EDX) measurements are also performed in spatial direction with a cylindrical telescope, and secondly
order to determine possible changes in the chemistry of the focusing the sub-beams one-dimensionally with an additional
treated surface. cylindrical lens (see Figure 1). The intersection angle between

Figure 1. Experimental setup (center) illustrating the optical assembly (right side) that forms the line-like shaped irradiation area containing a two beam
interference pattern and the areal scanning strategy (left side).

Adv. Eng. Mater. 2019, 1900151 1900151 (2 of 9) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.aem-journal.com

the overlapping sub-beams is defined by the wedge angles of the of material composition in near-surface layers, it can provide
bi-prisms. In this work, the spatial periods Λ of 8 μm and 19 μm useful information for the determination of qualitative changes
were used, which corresponds to intersection angles α of 7.6 and across periodic patterns as has been previously shown.[42] For
3.1, respectively. performing the EDX analysis, the acceleration voltage was set to
In order to structure larger areas, the relative movement 15 kV. The line-scans were realized with a resolution of 284
between the laser irradiated area (which contains the interfer- points/100 μm, corresponding to a separation distance of
ence pattern) and the sample surface was performed using a 352 nm between two data points.
linear stage. The surface processing rate A0 is determined by the
length l of the laser irradiated area per pulse and the feed rate v,
as shown in Equation (2). 3. Results and Discussion
The DLIP approach was used to produce periodic surface
A0 ¼ v  l ð2Þ
structures on stainless steel with spatial periods of 8 μm and
19 μm. The influence of the pulse-to-pulse (PTP) overlap on the
resulting structure height was investigated. Figure 2 shows SEM
images of structured samples spatial periods of 19 μm
2.2.1. Surface Characterization (Figure 2a,b) and 8 μm (Figure 2c,d) employing a low PTP
overlap (Figure 2a,c) and high PTP overlap (Figure 2b,d),
The topographies of the patterned surfaces were characterized respectively. Figure 2a,c show the topography when using a
using a scanning electron microscope (Philips ESEM). The 40 μm PTP feed (60% overlap) while Figure 2b and d show the
structure heights were investigated by measuring the structure sample surface when utilizing a PTP feed (99% overlap) of only
height hst at different positions on the substrates using confocal 1 μm. As expected, the DLIP micro-structures become more
microscopy (Sensofar S-Neox).[41]The structure height of the pronounced with increasing the number of pulses.
patterns was evaluated using the step height definition, The individual pulses are still visible for large PTP feeds (see
representing the standard type A1 in the norm ISO5436-1.[41] Figure 2a and c), while they cannot be distinguished in the
Areas of 200
200 μm2 and 100
100 μm2 for the spatial surface profile contour for very small feeds (high overlaps, see
periods of Λ ¼ 19 μm and Λ ¼ 8 μm, respectively, were divided Figure 2b and d). Particularly, it can be seen in Figure 2a, that the
into a series of 18 profiles (cross-sections of the interference molten material at the interference maxima positions forms two
pattern). These profiles were used to calculate the average values independent melt fronts which follow the thermal gradient
as well as the standard errors for each measurement. In order to perpendicular to the line-like laser intensity distribution. These
determine the effect of the laser treatment on the surface two fronts of molten material merge with the material release
chemistry, the treated surfaces were examined with Energy- from the neighboring interference maxima at the interference
dispersive X-ray spectroscopy (EDX, Philips ESEM). Although minima regions when increasing the pulse overlap (Figure 2b),
EDX is of limited use for the precise quantitative determination especially for shorter spatial periods due to the reduced

Figure 2. SEM images of treated steel surfaces (X5CrNi18-10/1.4301) with spatial periods of a) Λ ¼19 μm and 80 % PTP overlap, b) Λ ¼ 19 μm and 99 %
PTP overlap, b) Λ ¼ 8 μm and 80% overlap and c) Λ ¼ 8 μm and 99% PTP overlap. The used laser fluence was 3.2 J cm2 in all cases.

Adv. Eng. Mater. 2019, 1900151 1900151 (3 of 9) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.aem-journal.com

geometrical distance (Figure 2c). At high PTP overlaps (short the structure height evolution for patterns with spatial periods of
PTP feeds), the melt fronts convergence in combination with the Λ ¼ 19 μm, while Figure 4c and d show the structure height
large amount of molten material resulting in a significant evolutions for spatial periods of Λ ¼ 8 μm. In addition,
increase in the structure height, as shown in Figure 2b and d. Figure 4a and c show the influence of different pulse-to-pulse
Figure 3 shows the confocal microscope images of the same feeds between 1 μm and 20 μm (corresponding to pulse overlaps
line-like structures as in Figure 2. With shorter PTP feed (higher between 99 % and 80 %, respectively) using the same pulse
PTP overlap), considerably deeper structure heights of 8.7 μm repetition rate of 5 kHz. Figure 4b and d also show the influence
and 18.5 μm were achieved for the 8 and 19 μm spatial periods, of different pulse repetition rates based on the results obtained
respectively (see Figure 3b,d). In comparison, the heights of the with pulse repetition rates from 0.1 kHz to 5 kHz at a fixed pulse-
periodic structures formed with larger PTP feed (lower overlap) to-pulse feed of 2 μm.
reached only structure heights of 3.4 μm and 1.5 μm for the From Figure 4a is visible, that with increasing PTP overlap the
spatial periods of 8 μm and 19 μm, respectively (see Figure 3a,c). structure height increases as well. For high pulse overlaps higher
The aspect ratios for both periods are very similar when using than 98%, maximum structure heights of more than 20 μm
high overlap conditions (AR of 1.09 and 0.97 for 8 μm and 19 μm (corresponding to aspect ratios of more than 1) are achieved. In
period, respectively), at 99% overlap and 3.2 J cm2 laser fluence all cases, the structure height is always lower at the beginning of
per pulse (see Figure 3c,d). For low overlap conditions the aspect the process and gradually increases with increasing distance
ratios for both periods differ (AR of 0.42 and 0.08 for 8 μm and relative to the process starting point (infeed) until either a steady
19 μm period, respectively) at 80% overlap at same laser fluence state condition is reached or even a reduction of the structure
(see Figure 3a,c). height occurs. Different mechanisms can be responsible of the
Based on the topographic analysis, the steadiness of structure changes in the structure height observed during the process.
height of the produced patterns was investigated by evaluating One possible reason is the heat accumulation which is produced
the structure height during the course of the laser process. Since during the laser treatment, meaning that the irradiated steel
the substrate is moved relative to the DLIP head during the substrates are heated up.[43,44] Additionally, the produced surface
process, the structure height development was determined by structures on the substrate can have a significant influence on
measurements at different positions on the processed area on the optical absorption of the laser radiation due to multi-
the substrates. The structure heights were measured at the reflections in the generated trenches, which means that a higher
beginning of the process (infeed) and at subsequent positions amount of energy is coupled in the material as the structure is
(dx ¼ 0.5, 1, 2, 3, 4, 5, 10, 20, and 50 mm). formed.[45,46] The effect of heat accumulation is further
Figure 4 shows the development of the position-dependent promoted by incubation effects during multi-shot laser
structure height during the DLIP process. Figure 4a and b show ablation.[47] It has been also reported that rippled surface

Figure 3. Confocal Microscope images of treated steel substrates (X5CrNi18-10/1.4301) with interference periods of a) Λ ¼ 19 μm at pulse overlap of
80%, b) Λ ¼ 19 μm at pulse overlap of 99%, c) Λ ¼ 8 μm at pulse overlap of 80% and d) Λ ¼ 8 μm at pulse overlap of 99%. The used laser fluence per
pulse was 3.2 J cm2.

Adv. Eng. Mater. 2019, 1900151 1900151 (4 of 9) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.aem-journal.com

patterns can change the thermal diffusion into the substrate,

thus influencing the resulting surface patterns.[48,49]
In general, an increase of the structure height with increasing
pulse overlap was observed for both spatial periods: Λ ¼ 19 μm
(Figure 4a) and Λ ¼ 8 μm (Figure 4c). However, for the smaller
spatial period (Λ ¼ 8 μm) a maximum in the structure height was
reached at a PTP feed of 2 μm followed by a decrease in structure
height for further increasing pulse overlap. For this condition (PTP
feed of 2 μm) the evolution of the structure height shows a
maximum at the 2 mm position and then subsequently decreases
until 10 mm and saturates from 20 mm. Consequently, at the
spatial period of Λ ¼ 8 μm an optimum condition in terms of
structure height can be expected in between the PTP feed values
applied. A pulse overlap exceeding this optimum value leads to
over-melting, resulting in reduced structure heights.
Regarding the influence of the pulse repetition rate on the
structure height for the short spatial period (Λ ¼ 8 μm, Figure 4d), a
similar behavior is observed as for the large periods (Figure 4b),
which means that for higher pulse repetition rates higher
structures are obtained. However, for the smaller period (Λ ¼ 8
μm) this dependence shows a minor extent, meaning that the
average structure heights achieved with 0.1 kHz is close to that of
achieved with 5 kHz (average structure heights 7.4 μm with
0.1 kHz and 8.7 μm with 5 kHz, respectively). A possible
explanation for this phenomenon is that for the shorter periods
the merging of the melt fronts occurs earlier in the course of the
process since melt has to travel a shorter distance.
The results shown in Figure 4 indicate a clear influence of the
applied pulse repetition rate on the resulting structure heights
since the applied average fluence per pulse is kept constant at FP,
2
av ¼ 3.2 J cm during all experiments. The change of the
structure height during the laser process indicates that the heat
is accumulated at the material if a large number of successive
laser pulses is used. Heat accumulation in laser material
processing has already been reported, but usually at compara-
tively higher pulse repetition rates (starting at several hundred
kHz).[50–52] Regardless of the applied pulse overlap or pulse
repetition rate, the structure heights saturate after a certain
distance from the start point. At higher pulse repetition rates, the
equilibrium of the structure height occurs later and with higher
equilibrium structure heights. Figure 4. Evolution of structure heights of trench patterns produced with
For interpretation of the obtained results, calculations of the DLIP: a,b) interference period Λ ¼ 19 μm; c,d) interference period
thermal evolution due to heat accumulation in the laser-affected Λ ¼ 8 μm; a,c) variation of the PTP feed at a fixed pulse repetition rate
zone were carried out according to the model described by Weber frep ¼ 5 kHz; b,d) variation of the pulse repetition rate at a fixed PTP feed
et al.[53] (see Equation (3)): of 0.002 mm; Material: stainless steel X5CrNi18-10/1.4301. The used laser
fluence per pulse was 3.2 J cm2.
   1 r2nD
Θ t  N1
4κ
Q nD X
NP
fL tN1
T Sum;nD ðtÞ ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi e fL
ð3Þ coordinate r2nD to zero the following simplified equation is
ρ  cp  ð4  π  κÞ N¼1  nD
t  N1 f obtained:
L

 
where QnD is the residual heat introduced into the material, ρ is 2  Q Heat =A NP Θ t  N1
X fL
the mass density of the material, cp its specific heat capacity, T Sum;nD ðtÞ ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð4Þ
ρ  cp  ð4  π  κÞ N¼1 t  N1
κ ¼ λth  (ρ  cp)1 is the diffusivity, λth the heat conductivity, t is fL
time, r the coordinate of space, fL the laser repetition rate and NP
the number of pulses. where QHeat is the residual heat and A is the irradiated area.
Due to the elongated shape of the laser irradiated area, the Three different pulse repetition rates of 5 kHz, 1 kHz, and
solution of the 1D heat flow can be assumed, which gives nD ¼ 1 0.1 kHz were used in the calculations with a constant laser
for the indicator of the dimension. By further setting the volume fluence of 3.2 J cm2. Figure 5 shows the calculated temporal

Adv. Eng. Mater. 2019, 1900151 1900151 (5 of 9) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.aem-journal.com

evolution of the surface temperature for each of the investigated schema shows that the molten material at the interference
pulse repetition rates. It can be seen that even at repetition rates maxima positions moves from the hot to the colder peripheral
of 1 kHz, an important accumulation of heat is observed, areas due to surface tension gradients induced by the
reaching a temperature increase of 600 K after 100 pulses. A temperature differences. Later, the fronts of molten metal start
further increase of the rep. rate to 5 kHz produces an increase of to merge at the interference minima locations. The application
1500 K, also after 100 pulses. The calculated temperature of subsequent laser pulses (with the interference intensity
evolutions demonstrate that heat accumulation is possible also at distribution) which is in particular more relevant when using
relative low repetition rates (KHz range), what in the case of ps high PTP overlaps results in additional melt leading to patterns
laser sources is observed at significantly higher frequencies (e.g., with higher aspect ratios. While the structure height can be
of some MHz).[54,55] This simple calculation clearly demon- significantly boosted especially for the larger spatial periods due
strates that the heat accumulation effects can take place in the to stacking of molten material, a stronger increase of the PTP
process used in the experiments and can also explain the overlap result in a reduction of the structure height for smaller
variation of the height during the DLIP process. periods due to material over melting (Figure 6c). Overmelting
Figure 6a shows the development of the average aspect ratios leads to a local collapse of the structures and thus to the
of the structure heights for the larger (Λ ¼ 19 μm) and smaller emergence of overhang structures (see for example the SEM
(Λ ¼ 8 μm) interference periods as a function of the number of image of Figure 2d). This would be in accordance with known
accumulated laser pulses per area. The aspect ratios are phenomena in the wake of Marangoni convection, such as the
calculated as the quotient of the average structure height (from humping phenomenon in laser welding.[58–60] This effect was
Figure 4) and the corresponding interference period. The not observed for larger periods, as in the latter case a larger
observed differences in the structure heights depending on the amount of melt is required due to the longer distances that have
spatial period and PTP overlap can be interpreted as a result of to be covered. In fact, overhang structures can be seen as an
the melt displacement which is increased due to the accumula- advantage compared to simple structures. For instance, micro-
tion of heat. A generally accepted theory for near-surface melt textures with overhang structures can enable to obtain super-
dynamics during laser processes is provided by Marangoni hydrophobic surfaces, even in materials which are intrinsically
convection.[56] Furthermore, D’Alessandria et al.[57] have already hydrophilic.[61,62]
reported on the different regimes responsible for the pattern These considerations explain that heat accumulation has
formation in DLIP when using ns pulses depending on the similar effect on smaller and larger interference periods, but that
applied pulse fluence, which are either Marangoni convection or for smaller periods the results obtained exceed the state of
the recoil pressure. In both cases, the obtained molten metal is maximum structure height, whereas this is not the case for the
driven towards the minima positions. However, the here larger periods.
presented results demonstrate for the first time the possibility In addition to the topography analysis, EDX measurements
of reaching structure heights in the same order of the spatial were performed for verifying the above described structuring
period (or aspect ratios slightly higher than one).[57] A scheme of mechanism. Figure 7 shows the results of an EDX line-scan
the possible structuring mechanism applying the Marangoni measured across a DLIP-structure (perpendicular to the
convection model for the DLIP using multiple pulses is shown in interference lines) for iron, oxygen, chromium, and carbon
Figure 6b and c, for long and short periods, respectively. The elements. The absolute response of the material during EDX

Figure 5. Temporal evolution of the temperature increase ΔTSum,1D on the surface of a semi-infinite body of stainless steel at the location of a one-
dimensional source for different repetition rates (5 kHz, 1 kHz, and 0.1 kHz) with equal heat source energy of 16 mJ per laser pulse with pulse durations
of τ ¼ 10 ns.

Adv. Eng. Mater. 2019, 1900151 1900151 (6 of 9) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.aem-journal.com

Figure 6. a) Average Structure Height Aspect Ratio as function of Pulse Overlap PO for Λ ¼ 19 μm and Λ ¼ 8 μm patterns; Schematic representation of
the structuring mechanism showing the melt flow in the DLIP processes for b) large and c) short interference periods.

measurements varies very strongly depending on whether the concentration at the interference minima positions. This
sensing spot strikes in the trenches or on the ridges of the periodic modulation of the material composition underlines
topography. For this reason, for each measurement point of the the movement of the melt flow from the interference maxima to
line scan, the individual percentages of the measured elements minima positions, since the molten metal is longer exposed to
were calculated. The gray bars in the background of Figure 7 the air atmosphere and thus allowing the diffusion of oxygen and
represent the interference maxima positions which correspond finally forming oxides. Both composition distributions for iron
to the locations of the trenches in the produced surface pattern. and oxygen also show slight double periodic peaks, which is in
The image shows that both iron and chromium elements agreement with the previous explanation and represent the two
(corresponding to stainless steel) have a higher concentration at fronts of molten metal which are finally merged at the
the interference maxima positions. Differently, the oxygen interference minima. These effects have been also investigated
content shows a opposite distribution, having a maximal in the past on steel surfaces which were irradiated using ns pulse

Adv. Eng. Mater. 2019, 1900151 1900151 (7 of 9) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.aem-journal.com

Figure 7. Ratio of the proportions of the most important elements determined as a line diagonally over the cross section of the DLIP trench structures,
based on EDX measurement.

DLIP by Rosenkranz et. Al.[63] Using XPS in combination with explained due to the accumulation of the heat, in particular for
Ar ion etching as well as atom probe tomography, a sharp repetitions rates of few kHz. Thus, this research work permitted
interface for transition between the oxidic and metallic state was to show for the first time heat accumulation processes using
found for the unpatterned sample, whereas for the laser- Direct Laser Interference Patterning with ns laser pulses.
irradiated specimens, not only a mixed oxidic state was observed Depending on the process parameters, aspect ratios slightly
but also an increase of the oxide layer thickness. In addition, this higher than one could be realized, which has not been previously
layer consisted on an iron oxide layer on top followed by achieved using DLIP with ns-pulsed lasers. Particularly for the
chromium oxide.[63] smaller periods, the melt flow impairs pattern formation due to
the overmelting of the material (humping). Finally, a mechanism
for the structure formation was presented, which was verified by
4. Conclusions EDX analyses.

In this work, high-speed surface structuring of stainless steel

was performed, by developing a DLIP optical configuration
producing elongated laser spots containing the interference
Acknowledgements
patterns. This was realized using a modular optics assembly This work was supported by the German Research Foundation (DFG)
combined the functions of beam splitting (two sub-beams for the (Programm SPP 1676 “Trockenumformen  Nachhaltige Produktion
production of trench patterns) and beam shaping in order to durch Trockenbearbeitung in der Umformtechnik”).
achieve a rectangular laser spot of 5 mm x 100 μm for
interference patterning. Patterns with two different spatial
periods (8 μm and 19 μm) were fabricated on stainless steel. The Conflict of Interest
impact of multi-pulsing was investigated by varying the feed The authors declare no conflict of interest.
rates. It was found that higher overlaps (or a larger amount of
pulses) lead to increase the structure heights. Furthermore, it
was found that during the DLIP process, a rapid increase in the
structure height occurs, particularly at the beginning of the Keywords
process (2 mm after the infeed position). This effect could be laser interference, manufacturing, micromachining, surface engineering

Adv. Eng. Mater. 2019, 1900151 1900151 (8 of 9) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.aem-journal.com

[33] D. J. Guckenberger, T. E. de Groot, A. M. D. Wan, D. J. Beebe,

E. W. K. Young, Lab Chip 2015, 15, 2364.
Received: February 8, 2019
[34] B. Hyun Kim, S. Hyoung Ryu, D. Ki Choi, C. Nam Chu, J. Micromech.
Revised: March 28, 2019
Microeng. 2005, 15, 124.
Published online:
[35] H. J. Booth, Thin Solid Films 2004, 453–454, 450.
[36] A. Lasagni, F. Mücklich, M. R. Nejati, R. Clasen, Adv. Eng. Mater.
2006, 8, 580.
[1] M. J. Walsh, AIAA J. 1983, 21, 485 [37] A. I. Aguilar-Morales, S. Alamri, A. Fabián Lasagni, J. Mater. Process.
[2] D. W. Bechert, W.-E. Reif, G. Hoppe, 23rd Aerospace sciences meeting Tech. 2018, 252, 313.
1985, 546. [38] P. Shi, D. Li, H. Zhang, Y. Wang, K. Du, Optics Commun. 2004, 229,
[3] D. M. Bushnell, K. J. Moore, Annu. Rev. Fluid Mech. 1991, 23, 65. 349.
[4] D. W. Bechert, M. Bruse, W. Hage, Exp. Fluids 2000, 28, 403. [39] B. Burghardt, H.-J. Kahlert, Optics for forming a sharp illuminating line
[5] D. W. Bechert, M. Bruse, W. Hage, R. Meyer, 28th Fluid Dynamics of a laser beam. Veröffentlichungsnr. US Patent 5,721,416.
Conf. 1997, 1960. [40] V. Lang, T. Roch, A. Fabián Lasagni, Adv. Eng. Mater. 2016, 18,
[6] T. Wan Kim, J. Nanosci. Nanotechnol. 2014, 14, 7562. 1342
[7] G. D. Bixler, B. Bhushan, J. Colloid Interface Sci. 2013, 393, 384. [41] ISO 5436–1:2000. 2000-03-15. ISO 5436–1:2000(E), Geometrical
[8] J. N. A. Matthews, Phys. Today 2008, 61, 32. Product Specifications (GPS) — Surface texture: Profile method;
[9] N. A. Patankar, Langmuir 2004, 20, 8209. Measurement standards.
[10] A. Marmur, Langmuir 2004, 20, 3517. [42] A. Lasagni, F. Mücklich, Practical Metallogr. 2006, 43, 1.
[11] W. Barthlott, C. Neinhuis, Planta 1997, 202, 1. [43] G. Raciukaitis, M. Brikas, P. Gecys, M. Gedvilas, in Proc. of SPIE,
[12] J. H. Fitton, B. A. Dalton, G. Beumer, G. Johnson, H. J. Griesser, 7005: SPIE Digital Library, 2008, 70052L.
J. G. Steele, J Biomed Mater Res. 1998, 42, 245. [44] R. Weber, C. Freitag, T. V. Kononenko, M. Hafner, V. Onuseit,
[13] L. Zhao, S. Mei, P. K. Chu, Y. Zhang, Z. Wu, Biomaterials 2010, 31, P. Berger, T. Graf, Phys. Procedia 2012, 39, 137.
5072. [45] L. K. Ang, Y. Y. Lau, R. M. Gilgenbach, H. L. Spindler, Appl. Phys. Lett.
[14] C. Liang, H. Wang, J. Yang, Y. Cai, X. Hu, Y. Yang, B. Li, H. Li, H. Li, 1997, 70, 696.
C. Li, X. Yang, ACS Appl. Mater. Interfaces 2013, 5, 8179. [46] D. Bergström, J. Powell, A. F. H. Kaplan, J. Appl. Phys. 2007, 101,
[15] A. F. von Recum, T. G. van Kooten, J. Biomater. Sci. Polym. Ed. 1996, 113504.
7, 181 [47] F. Di Niso, C. Gaudiuso, T. Sibillano, F. Paolo Mezzapesa, A. Ancona,
[16] F. Yu, F. Mücklich, P. Li, H. Shen, S. Mathur, C.-M. Lehr, U. Bakowsky, P. Mario Lugarà, Optics Exp. 2014, 22, 12200.
Biomacromolecules 2005, 6, 1160. [48] T. D. Bennet, D. J. Krajnovic, C. P. Grigoropoulus, P. Baumgart,
[17] A. Jaggessar, H. Shahali, A. Mathew, P. K. D. V. Yarlagadda, J. A. C. Tam, J. Heat Transfer 1997, 119, 589.
Nanobiotechnol. 2017, 15, 64 [49] J. M. Prusa, G. Venkitachalam, P. A. Molian, Int. J. Mach. Tool
[18] L.-C. Xu, C. A. Siedlecki, Acta Biomaterialia 2012, 8, 72. Manufac. 1999, 39, 431.
[19] N. Lu, W. Zhang, Y. Weng, X. Chen, Y. Cheng, P. Zhou, Food Control [50] F. Brygo, C. Dutouquet, F. Le Guern, R. Oltra, A. Semerok,
2016, 68, 344. J. M. Weulersse, Appl. Surface Sci. 2006, 252, 2131.
[20] A. Vikram Singh, V. Vyas, T. S. Salve, D. Cortelli, D. Dellasega, [51] H. Wook Kang, H. Lee, A. J. Welch, J. Appl. Phys. 2008, 103,
A. Podestà, P. Milani, W. N. Gade, Biofabrication 2012, 4, 25001. 83101.
[21] C. Díaz, P. Schilardi, M. F. L. de Mele, Artificial Organs 2008, 32, 292. [52] A. Ancona, S. Döring, C. Jauregui, F. Röser, J. Limpert, S. Nolte,
[22] A. J. Scardino, E. Harvey, R. de Nys, Biofouling 2006, 22, 55. A. Tünnermann, Optics Lett. 2009, 34, 3304.
[23] S. Eckhardt, M. Siebold, A. Fabián Lasagni, Optics Exp. 2016, 24, [53] R. Weber, T. Graf, P. Berger, V. Onuseit, M. Wiedenmann, C. Freitag,
A553. A. Feuer, Optics Exp. 2014, 22, 11312.
[24] C. Fuchs, T. Schwab, T. Roch, S. Eckardt, A. Lasagni, S. Hofmann, [54] B. Luther-Davies, Optical Eng. 2005, 44, 51102.
B. Lüssem, L. Müller-Meskamp, K. Leo, M. C. Gather, R. Scholz, [55] S. M. Eaton, H. Zhang, M. Li Ng, J. Li, W.-J. Chen, S. Ho,
Optics Exp. 2013, 21, 16319. P. R. Herman, Optics Exp. 2008, 16, 9443.
[25] L. Zhou, X. Dong, Y. Zhou, W. Su, X. Chen, Y. Zhu, S. Shen, ACS Appl. [56] J. Mazumder, Optical Eng. 1991, 30, 1208.
Mater. Interfaces 2015, 7, 26989. [57] M. D’Alessandria, A. Lasagni, F. Mücklich, Appl. Surface Sci. 2008,
[26] M. L. Brongersma, Y. Cui, S. Fan, Nature Mater. 2014, 13, 451. 255, 3210.
unas, S. Indrisiunas, B. Voisiat, R. Mazeikienė, J. Laser Micro.
[27] A. Vinci [58] U. Gratzke, P. D. Kapadia, J. Dowden, J. Kroos, G. Simon, J. Phys. D
Nanoeng. 2013, 8, 244. 1992, 25, 1640.
[28] L. Müller-Meskamp, Y. Hyun Kim, T. Roch, S. Hofmann, R. Scholz, [59] R. Fabbro, J. Phys. D: Appl. Phys. 2010, 43, 445501.
S. Eckardt, K. Leo, A. Fabián Lasagni, Adv. Mater. 2012, 24, 906. [60] Y. Kawahito, M. Mizutani, S. Katayama, J. Phys. D: Appl. Phys. 2007,
[29] M. Bieda, C. Schmädicke, T. Roch, A. Lasagni, Adv. Eng. Mater. 2015, 40, 5854.
17, 102. [61] L. Cao, T. P. Price, M. Weiss, D. Gao, Langmuir 2008, 24, 1640.
[30] C. Greiner, A. del Campo, E. Arzt, Langmuir 2007, 23, 3495. [62] S. Shin, J. Seo, H. Han, S. Kang, H. Kim, T. Lee, Materials 2016, 9,
[31] N. J. Shirtcliffe, S. Aqil, C. Evans, G. McHale, M. I. Newton, 116.
C. C. Perry, P. Roach, J. Micromech. Microeng. 2004, 14, 1384. [63] A. Rosenkranz, L. Reinert, C. Gachot, H. Aboufadl, S. Grandthyll,
[32] A. Fabian Lasagni, D. Benke, T. Kunze, M. Bieda, S. Eckhardt, K. Jacobs, F. Müller, F. Mücklich, Adv. Eng. Mater. 2015, 17,
T. Roch, D. Langheinrich, J. Berger, Proc. LPM2014 2014, 1. 1234.

Adv. Eng. Mater. 2019, 1900151 1900151 (9 of 9) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim