Optical coherence tomography for process control
of laser micromachining
Markus Wiesner,1Jürgen Ihlemann,1Heike H. Müller,2Eva Lankenau,2
and Gereon Hüttmann2
1Laser-Laboratorium Göttingen e.V., Hans-Adolf-Krebs-Weg 1, D-37077 Göttingen, Germany
2Institute of Biomedical Optics, University of Lübeck, Peter-Monnik-Weg 4, D-23562 Lübeck, Germany
?Received 2 November 2009; accepted 15 February 2010; published online 24 March 2010?
In situ surface imaging for nondestructive evaluation ?NDE? by optical coherence tomography
?OCT? before, during, and after ablative laser processing is presented. Furthermore, it is shown that
the ability of in situ characterization is beneficial for samples such as optical fibers, which are
difficult to handle in the standard analysis. Surface images taken by the OCT are compared with
these common analysis tools such as scanning electron microscopy ?SEM?, reflected-light, and
confocal microscopy. An axial resolution of ?126 nm for surface detection and a lateral resolution
?2.5 ?m are obtained and the potential of the setup to imaging structures with high aspect ratio is
demonstrated. © 2010 American Institute of Physics. ?doi:10.1063/1.3356080?
High resolution laser micromachining is a demanding
application for positioning and alignment systems. It requires
careful adjustment of the sample and online controlling of
parameters such as ablation rate and focus position. Nonde-
structive testing and evaluation ?NDT/NDE? is an important
part of the laser process and of the quality management in
general. Ex situ characterization of the machined work piece
with the need of subsequent readjustment into the processing
system are time consuming and hold the risk of contaminat-
ing and damaging the microstructure. In this paper, we will
present a concept to solve these problems by the integration
of an optical coherence tomography ?OCT? module for in
situ control into an existing F2-laser microprocessing system.
OCT is a recently introduced method for nondestructive,
noncontact, and multidimensional visualization, with many
applications in the investigation of biomedical specimens,
especially in the area of ophthalmology.1The method has
been under constant development since its introduction in the
year 1991.2It achieves high resolution imaging on the mi-
cron level with a high sensitivity and a large dynamic range
by the interferometric detection of backscattered and re-
flected light from structures and interfaces in the sample.3
Based on the principle of white-light interferometry, the
most common OCT setup is a Michelson interferometer illu-
minated by a temporally low-coherence light source. Mainly
near-infrared light sources are used to penetrate the surface
of opaque or turbid materials by some millimeters to yield
depth information. A depth scan ?A-scan? of the reference
mirror provides interference between the sample and refer-
ence beams and a depth-resolved reflectivity profile is ob-
tained. By deflecting the sample beam with a xy-scanner,
cross-sectional ?B-scans?, or three-dimensional ?3D? imaging
Besides the traditional time-domain OCT4new tech-
niques have been developed and established to enhance per-
formance and imaging quality. Spectral-domain ?SD-?OCT5,6
and swept-source ?SS-?OCT7lead to improved sensitivity8–10
and imaging speed. Scan rates up to 568 600 A-scans/s for
SS-OCT11and 312 500 A-scans/s for SD-OCT12have been
realized, which enables real-time monitoring. Supported by
improved light sources ultrahigh resolution ?UHR-?OCT
with depth resolutions down to 1 ?m and superior imaging
quality becomes available.13,14High resolution imaging com-
bined with new contrast mechanisms is often required for
many applications in materials research because in many
cases structures with sizes of a few microns are relevant.
In recent years, applications for nonmedical purposes
have become interesting and now find their way into indus-
trial use.15,16Implementations can be found in the field of
optical metrology17and noncontact material characterization.
The decoupling of lateral and axial resolution allows the
measurement of structures with a high aspect ratio like bore
holes.18The determination of the refractive index19is also
possible as well as measurements of thickness20
distances21or the survey of roughness of a specimen’s
A more sophisticated utilization of OCT is the combina-
tion with established measurement and manufacturing pro-
cesses such as laser-induced breakdown spectrometry to pro-
vide additional ablation depth information for spectrometric
analysis24or high speed in situ cross-sectional depth profil-
ing of ultrafast micromachining by using one light source for
both imaging and machining.25Process control with OCT in
general has been reported to supplement manufacturing26or
to replace sensors in harsh conditions to enhance measure-
ment precision.21Recently, imaging rates were presented,
which allow investigating the temporal dynamics of the laser
ablation process itself.27The introduction of so called func-
tional OCT has added more aspects to this field of applica-
tions. Spectroscopic analysis,28flow characterization29for
microfluidics, and evaluation of strain fields30are just a few
Hence, one of the largest fields of industrial utilization
REVIEW OF SCIENTIFIC INSTRUMENTS 81, 033705 ?2010?
0034-6748/2010/81?3?/033705/7/$30.00© 2010 American Institute of Physics
so far can be found in NDT/NDE. Imaging and characteriza-
tion of laser-induced damage sites in optical components31or
the investigation of different polymers, ceramics, and com-
posite materials32,33have been reported.
However, for material investigations many applications
demand the visualization of the whole sample surface, which
cannot obtained by point ?A-scan? or cross-sectional ?B-
scan? measurements. Thus transversal ?en-face? scanning is
needed to yield the essential information.34,35It has been
shown that this technique is capable to outperform imaging
methods that rely solely on the confocal principle.36In gen-
eral, noncontact process control for high resolution laser mi-
cromachining can be fulfilled by many optical techniques,
e.g., fringe projection, triangulation, confocal microscopy, or
interferometry. Of all optical methods, the white light inter-
ferometry is distinguished by the fact that the accuracy with
which a surface can be determined does not depend on the
aperture, i.e., the focusing angle of the incident radiation to
the surface.37In addition, the use of a SD-OCT makes me-
chanical scanning of the sample in z direction unnecessary
and keeps the mechanical complexity moderate. Therefore,
fast and high precision surface detection, even with large
working distance, limited space, and within structures with
high aspect ratio can be achieved. Especially the compara-
tively simple integration into established measurement and
manufacturing processes seems a promising way to establish
new techniques and shows the versatile capabilities of OCT
for industrial applications. In particular, high precision laser
processing methods such as ablation, pattering, or 3D micro-
machining can benefit from this kind of process control. This
paper will present the integration of a UHR-SD-OCT with
the ability of surface imaging into an existing F2-laser pro-
cessing system. Possible applications such as NDT of micro-
machined surfaces or alignment purposes will be shown.
II. OPTICAL SYSTEM AND SETUP
The F2-laser processing system consists of a F2-laser
?Lambda Physik LPF 220i?, a 157 nm beam-shaping and
-delivery system ?MicroLas Lasersystem?, high precision tar-
get positioning drives, and beam and sample-alignment diag-
nostics. It has been described in detail previously.38Ablative
processing of materials is performed in a mask projection
configuration. The laser delivers up to 25 mJ single-pulse
energy with 15 ns pulse duration at 1–200 Hz repetition rate.
Homogenization and imaging optics are assembled in a 3 m
long aluminum chamber, which is flushed with nitrogen gas
to provide transparency at 157 nm. The mask is illuminated
with the homogenized laser beam and imaged onto the work
piece at 25? demagnification using a Schwarzschild objec-
tive of 0.4 numerical aperture ?NA?. The optical system
provides a uniformly illuminated target field of size
240?240 ?m2. The target sample is positioned outside the
chamber on a high precision translation stage. A gas flow
nozzle separates the imaging optics from the sample to pro-
vide a protective and transparent stream of nitrogen gas to
the working area on the target surface ?Fig. 1?. With this
setup, ablative machining even of transparent materials such
as glass and quartz is possible, as the short wavelength of
157 nm serves for efficient absorption. Thus, high precision
manufacturing of optical components such as submicron
gratings39or microlenses on the end faces of optical fibers40
has been demonstrated. Due to the short wavelength ? being
used in combination with the high NA of the Schwarzschild
objective the depth of field ?DOF? is very limited. According
to the well-known equation41
DOF= ? 0.5
the DOF is about 1.0 ?m. This must be taken into account
during the alignment process to secure precise processing. In
a basic version, the alignment of the image plane ?z-
alignment? is accomplished using the camera of the sample
alignment system or by applying a test series of ablation
spots with varying z to find the exact image plane. Both
methods are time consuming and not very reliable, the latter
is even damaging part of the work piece. Furthermore, online
control of the sample position for longer manufacturing pro-
cesses is not possible and the detection of tilted sample sur-
faces a major problem. For these reasons, a more sophisti-
cated alignment and focus control method is needed. The
adoption of the OCT ability for surface detection is imple-
mented here to solve these problems. The OCT system con-
sists of a commercial SD-OCT imaging system ?Thorlabs
FIG. 1. ?Color online? High resolution Schwarzschild configuration of the F2-laser optical processing system with OCT add on.
033705-2 Wiesner et al. Rev. Sci. Instrum. 81, 033705 ?2010?
Inc.?, including the data acquisition and the illumination,
combined with a customized Michelson interferometer and a
galvanometric scanner system. The OCT imaging system has
an A-scan sample rate of 1.4 kHz and uses a superlumines-
cent diode for illumination with an output power of 8.4 mW.
The central wavelength is ?0=926.3 nm with a full width at
half maximum bandwidth of ??=95.1 nm. The axial reso-
lution ?z of the OCT systems can be described by42
2=2 ln 2
and depends on the coherence length lcof the light source.
Since SD-OCT acquires a depth-resolved reflectivity profile
with a single A-scan, the maximum scanning depth zmaxis
where n is the sample’s refractive index and N is the number
of spectrometer detector elements. According to this equa-
tion, the maximum scanning depth scales linearly with the
number of detector elements. With N=1024 the theoretical
measurement depth of our OCT system is zmax=2.3 mm
with an axial resolution of ?z=3.9 ?m.
An obvious advantage of OCT is that the axial and lat-
eral resolutions are decoupled, so that it is possible to opti-
mize the system design for lateral scanning without affecting
the axial resolution. The lateral resolution itself depends on
the optical constraint of the sample arm optics and is de-
scribed by the resolution limit for image formation in
?x = 0.61?
In our case, the OCT uses the same Schwarzschild type re-
flective objective that is used for the laser processing beam.
This is possible due to the broadband reflective coating and
the achromatic nature of this optics device. The 0.4 NA of
the Schwarzschild objective provides a lateral resolution of
?x=1.4 ?m at the OCT wavelength and, according to Eq.
?1? a DOF of 5.8 ?m. For defining the position of a surface,
the measured OCT peak was fitted with a parabolic function
and the exact position of the maximum was determined.
With optical smooth surfaces, the measurement precision
was limited by mechanical vibration and with optically scat-
tering surfaces the accuracy was reduced due to speckle for-
The setup and beam path of the Michelson interferom-
eter is shown in Fig. 2. The emitted light is guided from the
imaging system by a single mode fiber to the Michelson
interferometer and split into the reference and sample arms.
The reference arm is folded twice to keep the setup as com-
pact as possible. The integration of the OCT through the
existing sample alignment system makes in situ monitoring
of the target field and the mask possible. For coupling the
OCT beam path into the sample alignment beam path, a cube
beam splitter was integrated under the illumination port of
the sample alignment. After that the OCT beam path is iden-
tical to the sample alignment beam path, passes the field lens
of the F2-laser optical processing system, and is focused into
the mask plane. From there, the focus spot is imaged by the
Schwarzschild objective onto the target field. The sample
arm optics were customized so that it meets the necessary
optical conditions for monitoring the target field of both the
Schwarzschild objective and the mask. For monitoring the
mask plane, it is required to match the shorter sample arm by
swing a mirror tub into the reference arm.
For surface scans, the sample arm can be deflected by a
galvo scanner system ?6210H, Cambridge Technology,
USA?. The scanner system allows us to scan the whole target
field ?240?240 ?m2? of the F2-laser processing system
with a maximum number of 542?542 A-scans for one im-
age. In principle, it is possible to scan the whole Schwarzs-
child objective field of view of 720 ?m, but the empty
mask holder restricts the maximum scanning field to
III. SYSTEM PERFORMANCE
Main purpose of the installed OCT is the in situ surface
detection and imaging for alignment purposes and NDE be-
fore, during, and after laser processing. We demonstrate here
the potential of the setup to imaging structures with high
aspect ratio, which are not accessible for most common ana-
lyze tools such as confocal microscopy or scanning electron
microscopy ?SEM?. Furthermore, it is shown that the ability
of in situ characterization is beneficial for samples which are
difficult to handle in the standard analysis such as optical
fibers, which may be too long and bulky for investigation
under a microscope or SEM. In particular, SEM is restricted
to spot test or the inspection of prototypes because of the
contamination by the required metal coating.
A. Surface inspection
Figure 3 shows a SEM image of a line grating on the end
face of a 400 ?m multimode optical fiber which was manu-
factured by the F2-laser processing system. Due to the large
size of the grating, ablation of each line by a slit mask was
FIG. 2. ?Color online? ?a? Configuration and ?b? photographic views of the
customized Michelson interferometer at the backside of the sample align-
033705-3Wiesner et al. Rev. Sci. Instrum. 81, 033705 ?2010?
not possible, so each line was written by imaging a spot on
the surface and then moving the fiber relative to the laser
with a velocity of 14 ?m/s. The laser was triggered with
50 Hz and the fluence was about 6.8 J/cm2. The slightly
damaged area of the line grating in the upper right was in-
vestigated by the OCT to demonstrate the ability of detecting
small damage sites within NDT. The scan area is
200?200 ?m2and consists of 400?400 A-scans ?Fig.
4?a??. Figure 4?b? provides a closer look of the structure. The
scan area is 80?80 ?m2consisting of 540?540 A-scans.
The different ablation depth for the coating, cladding, and
the fused silica core, as a result of the different ablation rates
are recognizable ?Fig. 4?c??. The grating period is 20 ?m
with a gap width of 7 ?m and covers the whole core of the
fiber. The depth of the structure is about 13 ?m at the core
area. The aforementioned cracks are also clearly visible
?Figs. 4?a? and 4?b??.
B. Work piece alignment
OCT is not limited to inspection purposes, also work
piece alignment is feasible. It is possible to even remove the
specimen from its holder and to realign it within the optical
system with the help of the OCT.
Figures 5?a?–5?c? shows a simple “stepped pyramid”
structure, which was ablated in a three step process, by re-
ducing the lateral length of a square mask ?2.5, 1.25, and 0.5
mm? inserted in the mask plane for every ablation step. First,
the surface was detected by the OCT and moved to the image
plane of the Schwarzschild objective.Asquare aperture mask
with a lateral length of 2.5 mm was used for the ablation.
Sixteen pulses with a fluence of 4 J/cm2were applied, so
that a square was ablated in the material. An A-scan by the
OCT at the center of the structure delivered a depth of
?2.1 ?m. The sample was released from the holder and
investigated under a reflected-light microscope ?Fig. 5?a??.
The sample was then replaced to the holder and realigned by
the OCT. For the next step, the lateral size of the square
aperture mask was reduced to 1.25 mm and the abovemen-
tioned ablation was applied again. The structure depth mea-
sured by the OCT at the center was now 4.8 ?m. A third
step with a lateral mask size of 0.5 mm was performed and
resulted in a final depth of 8 ?m at the center of the struc-
ture. Figure 5?a?–5?c? shows that realignment of the sample
was successful and every ablated structure has sharp bound-
aries. A comparison with depth measurements taken by a
confocal microscope ?Sensofar Plu 2300? after the end of
manufacturing are shown in Table I.
FIG. 3. SEM image of 20 ?m line grating at the end face of a 400 ?m
multimode optical fiber.
FIG. 4. ?Color online? ?a? Surface image of the fiber end face ?cf. Fig. 3?
showing an area of 200?200 ?m2. Thin white line indicates where the
B-scan was obtained. ?b? In this image, the scan area was reduced to
80?80 ?m2consisting of 540?540 A-scans. In both images, the damaged
area is clearly visible. ?c? B-scans show the different ablation depth for the
coating ?left?, cladding ?middle? and the fused silica core ?right?, separated
by the dotted lines.
FIG. 5. Three step manufacturing of a simple “stepped pyramid.” Between
each step the sample was released from its holder and the microscopic
image was taken.After realignment, the next step was applied. This example
shows the reliability of the OCT alignment capabilities. After the ablation,
the measured depth via A-scan in the middle of the structure was ?a?
2.1 ?m, ?b? 4.8 ?m, ?c? 8 ?m.
033705-4Wiesner et al.Rev. Sci. Instrum. 81, 033705 ?2010?
A surface image of a more suitable pyramid structure
was taken to demonstrate the capability to detect debris
which is produced during the manufacturing process. Com-
parison between the reflected-light microscopic and the OCT
images shows that the debris deposited around the topmost
square could be clearly resolved by the OCT ?Figs. 6?a? and
6?b??. Furthermore the roughness induced by the F2-laser ab-
lation on the surface of the micromachined structure43can be
detected by the OCT ?Figs. 6?c? and 6?d??.
To evaluate the performance of our measurement sys-
tem, we used a lithographically manufactured phase mask.
This four-step phase mask was well characterized44and used
for beam shaping of partial coherent UV laser beams. Every
step has a height of 126?3 nm with a lateral pixel size
of 2.5 ?m. Figures 7?a? and 7?b? shows surface images
taken with the OCT. The picture ?a? shows an area of
150?150 ?m2consisting of 300?300 A-scans. The differ-
ent heights of the steps are well-defined and gray-scale re-
solved. By reducing the scan area to 50?50 ?m2, also with
300?300 A-scans, not only the individual pixels can be
clearly distinguished, it also makes defects smaller than the
pixel size of 2.5 ?m visible ?Fig. 7?b??. For comparison pur-
poses, pictures taken by SEM and a reflected-light micro-
scope are shown in Fig. 7?c? and 7?d?. For defining the po-
sition zSof a surface, the measured OCT peak was fitted with
a parabolic function by
?Ip−1+ Ip+1− 2 · Ip?·1
where zpand Ipare the coordinates of the maximum value
and the proximate values Ip−1and Ip+1. Figure 8 shows ex-
emplary the parabolic fitting to the OCT data for two surface
points at nearby pixels with different step heights. The dif-
ference of these two points is the step height. In this ex-
ample, we found a height difference of ?zS=0.125 ?m.
We have presented an implementation of an UHR-SD-
OCT into an existing F2-laser processing system for in situ
NDE of micromachined surfaces and alignment purposes.
We compared the surface images taken by the OCT with
common analysis tools such as SEM, reflected-light, and
confocal microscopy. The results show that it is possible to
use OCT for surface detection and imaging in the field of
laser micromachining. Furthermore reliable focusing and tilt
TABLE I. Comparison of depth measurements by OCT and a commercial
FIG. 6. Comparison between reflected-light microscope and surface OCT
images. ?a? 500? magnification of the microstructure and ?b? matchable
OCT presentation with 400?400A-scans. The debris around the structure is
resolved in both images. Furthermore the OCT image contains grayscale
encoded height information. By increasing the magnification to 1000? ?c?
and reducing the scan area from 160?160 to 80?80 ?m2?d? the rough-
ness of the surface becomes visible.
FIG. 7. Surface OCT images of the phase mask with the size of
150?150 ?m2?a? and 50?50 ?m2?b?. Defect ?circle? smaller than
2.5 ?m can be also seen in the upper right of ?b?. Both images consist of
300?300 A-scans. For comparison images were also taken with a reflected-
light microscope ?c? and a SEM ?d?.
033705-5Wiesner et al. Rev. Sci. Instrum. 81, 033705 ?2010?
detection of the sample within the alignment process is pos-
sible. Multistep ablation with intermediate removal of the
sample becomes possible. An axial resolution of ?126 nm
for surface detection and a lateral resolution ?2.5 ?m were
obtained. Unfortunately, the low scanning rate of 1.4 kHz for
A-scans makes high resolution imaging time consuming.
Here, further improvement of the current setup is needed.
Scanning rates up to 200 kHz are desirable. With proper
synchronization to the F2-laser processing system online
measurement of the ablation rate or investigation of the tem-
poral dynamics of the laser ablation process itself should be
possible. Especially for industrial application a more com-
pact and rugged setup without fiber guidance is needed. On
balance, the OCT has proven to be versatile for both surface
imaging and sample alignment. The integration into the
F2-laser processing system has already reduced processing
time, rejection rate, and enhanced the whole manufacturing
process itself by the capability of in situ NDE.
We would like to thank Jörg Meinertz, Tim Bonin and
the Thorlabs HL AG for useful discussions and Markus
Sailer for skilled SEM and confocal microscopy measure-
ments. Financial support by the German Federal Ministry of
Economics and Technology is gratefully acknowledged
?Grant No. 16IN0351?.
1M. Wojtkowski, R. Leitgeb, A. Kowalczyk, T. Bajraszewski, and A. F.
Fercher, J. Biomed. Opt. 7, 457 ?2002?.
2D. Huang, E. A. Swanson, C. P. Lin, J. S. Schuman, W. G. Stinson, W.
Chang, M. R. Hee, T. Flotte, K. Gregory, C. A. Puliafito, and J. G.
Fujimoto, Science 254, 1178 ?1991?.
3A. F. Fercher, W. Drexler, C. K. Hitzenberger, and T. Lasser, Rep. Prog.
Phys. 66, 239 ?2003?.
4E. A. Swanson, D. Huang, M. R. Hee, J. G. Fujimoto, C. P. Lin, and C. A.
Puliafito, Opt. Lett. 17, 151 ?1992?.
5A. Fercher, C. Hitzenberger, G. Kamp, and S. El-Zaiat, Opt. Commun.
117, 43 ?1995?.
6G. Häusler and M. W. Lindner, J. Biomed. Opt. 3, 21 ?1998?.
7M. A. Choma, K. Hsu, and J. A. Izatt, J. Biomed. Opt. 10, 044009 ?2005?.
8J. F. de Boer, B. Cense, B. H. Park, M. C. Pierce, G. J. Tearney, and B. E.
Bouma, Opt. Lett. 28, 2067 ?2003?.
9R. A. Leitgeb, C. K. Hitzenberger, and A. F. Fercher, Opt. Express 11, 889
10M. A. Choma, M. V. Sarunic, C. Yang, and J. A. Izatt, Opt. Express 11,
11C. M. Eigenwillig, W. Wieser, B. R. Biedermann, and R. Huber, Opt. Lett.
34, 725 ?2009?.
12B. Potsaid, I. Gorczynska, V. J. Srinivasan, Y. Chen, J. Jiang, A. Cable,
and J. G. Fujimoto, Opt. Express 16, 15149 ?2008?.
13M. Wojtkowski, V. J. Srinivasan, T. H. Ko, J. G. Fujimoto, A. Kowalczyk,
and J. S. Duker, Opt. Express 12, 2404 ?2004?.
14W. Drexler, U. Morgner, F. X. Kärtner, C. Pitris, S. A. Boppart, X. D. Li,
E. P. Ippen, and J. G. Fujimoto, Opt. Lett. 24, 1221 ?1999?.
15D. Stifter, Appl. Phys. B: Lasers Opt. 88, 337 ?2007?.
16P. Targowski, M. Gora, and M. Wojtkowski, Laser Chem. 2006, 35373
17M. L. Dufour, G. Lamouche, V. Detalle, B. Gauthier, and P. Sammut,
Insight 47, 216 ?2005?.
18T. Dresel, G. Häusler, and H. Venzke, Appl. Opt. 31, 919 ?1992?.
19G. J. Tearney, M. E. Brezinski, J. F. Southern, M. R. Hee, B. E. Bouma,
and J. G. Fujimoto, Opt. Lett. 20, 2258 ?1995?.
20M. Haruna, M. Ohmi, T. Mitsuyama, H. Tajiri, H. Maruyama, and M.
Hashimoto, Opt. Lett. 23, 966 ?1998?.
21L. Giniunas, R. Karkockas, and R. Danielius, Appl. Opt. 37, 6729 ?1998?.
22P. Ettl, B. Schmidt, M. Schenk, I. Laszlo, and G. Häusler, Proc. SPIE
3407, 133 ?1998?.
23M. M. Amarala, M. P. Raelea, J. P. Calyb, R. E. Samada, and A. Z.
Freitasa, Proc. SPIE, 7390, 73900Z ?2009?.
24E. A. Kwiatkowska, J. Marczak, R. Ostrowski, W. Skrzeczanowski, M.
Sylwestrzak, M. Iwanicka, and P. Targowskia, Proc. SPIE 7391, 73910F
25P. J. Webster, M. S. Muller, and J. M. Fraser, Opt. Express 15, 14967
26G. M. Guss, I. L. Bass, R. P. Hackel, C. Mailhiot, and S. G. Demos, Appl.
Opt. 47, 4569 ?2008?.
27W. Y. Oh, S. H. Yun, B. J. Vakoc, G. J. Tearney, and B. E. Bouma, Appl.
Phys. Lett. 88, 103902 ?2006?.
28C. Joo, K. H. Kim, and J. F. de Boer, Opt. Lett. 32, 623 ?2007?.
29Y.-C. Ahn, W. Jung, and Z. Chen, Appl. Phys. Lett. 89, 064109 ?2006?.
30K. Wiesauer, M. Pircher, E. Götzinger, C. K. Hitzenberger, R. Engelke, G.
Grützner, G. Ahrens, R. Oster, and D. Stifter, Insight 49, 275 ?2007?.
31S. G. Demos, M. Staggs, K. Minoshima, and J. Fujimoto, Opt. Express 10,
32D. Stifter, K. Wiesauer, M. Wurm, E. Leiss, M. Pircher, E. Götzinger, B.
Baumann, and C. K. Hitzenberger, in 17th World Conference on Nonde-
structive Testing, 2008, p. 17.
33D. Stifter, K. Wiesauer, M. Wurm, E. Schlotthauer, J. Kastner, M. Pircher,
E. Götzinger, and C. K. Hitzenberger, Meas. Sci. Technol. 19, 074011
34B. R. Biedermann, W. Wieser, C. M. Eigenwillig, G. Palte, D. C. Adler, V.
J. Srinivasan, J. G. Fujimoto, and R. Huber, Opt. Lett. 33, 2556 ?2008?.
35K. Wiesauer, M. Pircher, E. Götzinger, S. Bauer, R. Engelke, G. Ahrens,
G. Grützner, C. K. Hitzenberger, and D. Stifter, Opt. Express 13, 1015
36J. A. Izatt, M. R. Hee, and G. M. Owen, Opt. Lett. 19, 590 ?1994?.
37M. C. Knauer, C. Richter, and G. Häusler, Laser J. 1, 33 ?2006?.
38P. R. Herman, K. P. Chen, M. Wei, J. Zhang, J. Ihlemann, D. Schäfer, G.
420421 422423 424425426427428429430431
Depth z [µm]
Intensity I [dB]
Depth z [µm]
Intensity I [dB]
FIG. 8. ?Color online? Parabolic fit to the OCT data for two different surface
points of the phase mask ?red dots and blue diamonds?. ?a? Overview of the
parabolic fit to the found OCT peak. ?b? Enlargement of the peak to show
the different position of the two vertexes. The difference of these two points
is step height of two pixel structures.
033705-6Wiesner et al.Rev. Sci. Instrum. 81, 033705 ?2010?
Marowsky, P. Oesterlin, and B. Burghardt, Proc. SPIE 4274, 149 ?2001?.
39J. Ihlemann, S. Müller, S. Puschmann, D. Schäfer, P. R. Herman, J. Li, and
M. Wei, Proc. SPIE 4941, 94 ?2003?.
40T. Fricke-Begemann, J. Li, J. Dou, J. Ihlemann, P. Herman, and G. Ma-
rowsky, Proceedings of the Third International WLT-Conference on La-
sers in Manufacturing, 2005, p. 733.
41M. Born and E. Wolf, Principles of Optics, 6th ed. ?University Press,
Cambridge, 1997?, pp. 370–458.
42P. H. Tomlins and R. K. Wang, J. Phys. D: Appl. Phys. 38, 2519 ?2005?.
43P. E. Dyer, C. D. Walton, and K. A. Akeel, Opt. Lett. 30, 1336 ?2005?.
44D. Schäfer, “Diffractive phase elements for partial coherent UV laser
beams,” Ph.D. thesis, Georg-August-Universität Göttingen, 2001.
033705-7Wiesner et al. Rev. Sci. Instrum. 81, 033705 ?2010?