Beam shaping imaging system for laser microprocessing with scanning optics Alexander Laskin a, Nerijus Šiaulys b, Gintas Šlekys b, Vadim Laskin a a AdlOptica GmbH, Rudower Chaussee 29, 12489 Berlin, Germany b Altechna/WOP, Konstitucijos ave. 23C-604, LT-08105, Vilnius, Lithuania ABSTRACT Laser beam shaping systems converting Gaussian to flattop or other irradiance profiles are used in various solar cell manufacturing laser technologies to enhance their performance. Scanning over whole working field with using popular 2- and 3-axis galvo mirror scanners is very often important part of microprocessing systems. Therefore, combining of beam shaping optics with scanning heads is an important technical task in field of solar cells manufacturing. As the beam shaping optics it is suggested to apply field mapping refractive beam shapers Shaper having some important features: low output divergence, high transmittance, extended depth of field, capability to work with TEM 00 and multimode lasers, as result providing a freedom in building various optical systems. Demagnifying of flattop laser beam can be realized with using imaging technique; the imaging optical system to be composed from F- lens of scanning head and additional collimating system to be used right after a Shaper. One of the problems in this approach is implementation of compact design of the collimating part. As a solution it is suggested to apply a specially designed Beam Shaping Unit being based on Shaper and locating between a laser and a scanning head; the functions of that combined system are: conversion from Gaussian to flattop laser beam irradiance profile, compact collimator design, alignment features, easy adaptation to a laser and a scanning head used in particular equipment. There will be considered design features of refractive beam shapers Shaper and Beam Shaping Unit, examples of optical layouts to generate flattop laser spots, which sizes span from several tens of microns to millimetres. Examples of real implementations and results of material processing will be presented as well. Keywords: beam shaping, flattop, tophat, micromachining, scribing, drilling, fired contacts, galvo scanner. 1. INTRODUCTION Uniform irradiance distribution of laser spot, realized by beam shaping optics, is highly advisable in various micromachining techniques like scribing, drilling, laser doping and laser-fired contact processes used in production of solar cells, PCB and Through-Silicon Via (TSV) drilling, display repair. At the same time scanning over working field using 2- and 3-axis galvo mirror scanners is another important part of industrial microprocessing systems. Therefore combining of beam shapers, converting Gaussian to flat-top (uniform) laser beam profile, with scanning optical heads is an insistent technical task. To meet the requirements of modern laser techniques the beam shaping optics should be compatible with other optical components applied in technological installations, should allow adaptation to real production conditions. Depending on an application either round or square laser spots are required, therefore optics of beam shaping systems should provide possibility to realize not only variable intensity distributions but also various spot shapes. When a laser technology requires a laser spot of less than 100 µm size, uniform profile, steep edges and a particular shape (round, square, etc.), one of fruitful optical approaches is to apply a refractive beam shaper like Shaper and imaging optical system. The F- lens of a scanning system can be a part of that imaging layout, and then a workable optical solution can be built on the base of usual off-the-shelf industrial components.
2. THEORETICAL CONSIDERATIONS 2.1 Motivation increasing efficiency of using laser energy All laser applications have specific features of interaction of material and laser radiation. There is, however, something common for all single mode (TEM 00 ) lasers the Gaussian function of intensity distribution. Therefore approximate evaluation of the efficiency of using laser energy can be done by considering just the geometrical features of the Gaussian function, without taking into account effects accompanying laser treatment of materials like burning, melting, etc. The three-dimension intensity profile can be interpreted as a geometrical figure bounded by a horizontal plane and a surface of the Gaussian function I(r) I (r) = I 0 exp(-2hr 2 / 2 ) (1) where r is variable beam radius in polar coordinates, is waist radius of the Gaussian laser beam, I 0 is constant. Fig.1 shows a section of such a figure, its volume has physical sense of energy of the laser beam. Let s denominate by variables E 1, E 2, E 3 different parts of that figure: E 1 - an apex of Gauss function is an excess of intensity over the working level I h, very often this is a loss of energy or a source of overheating the central portion of a zone under treatment, E 2 - tails of Gaussian distribution, almost always this is a loss of energy or a source of unwished effects like heat affected zone (HAZ), and E 3 - effective cylinder of energy. Figure 1. To evaluation of efficiency of using the laser energy By mathematical transformation one can get following formulas to calculate the energy parts E 1, E 2 and E 3 : E 1 = 1 h + lnh h E 2 = h (2) E 3 = lnh h E 1 +E 2 = 1 + lnh h where h = I h / I max. The results of calculations are presented on right diagram in Fig.1. The unconditional energy loss E 2, tails, can reach a very high level - if a working energy level I h is half of maximum (h = 0.5, very often just this level is considered as a working one) the energy loss is 50% of full laser beam energy! In the case of laser treatment of thin films the energy part E 1, apex of Gauss, is also considered as a loss of energy because this part exceeds the working energy level I h. Thus both energy parts E 1 and E 2 are losses, the sum E 1 + E 2 has sense of combined losses and minimum of this function is 0.63! In other words, when treating thin films, in the best case, only 63% of energy is lost and 37% is effective! No doubt, transformation of the original Gaussian shape to an effective cylinder, called as a flattop or top hat profile would help to save laser energy, increase productivity and improve those technologies where uniform intensity is most desirable.
2.2 Refractive Beam Shaper Among popular types of beam shaping optics the field mapping refractive beam shapers 1,2,3 like Shaper 5,6,10,11, Fig. 1, demonstrate high flexibility in building various optical systems due to their principle of operation implying saving of a beam consistency and providing a flat wave front of output beam. Figure 2 Principle of the Shaper operation Their main optical features are: - refractive optical systems to transform laser intensity distribution from Gaussian to flattop (tophat, uniform); - almost 100% efficiency; - high transmittance; - the transformation is realized through the phase profile manipulation in a control manner, without deterioration of the beam consistency and increasing its divergence; - the output phase profile is maintained flat, hence output beam has low divergence; - adaptability to real lasers through variation of distance between the Shaper components; - TEM 00 or multimode beams applied; - Output beam is collimated and resulting beam profile is kept stable over large distance; - Galilean design, no internal focusing; - achromatic optical design - several lasers can be used simultaneously. Fig.3 presents experimental data with TEM 00 laser. Important feature of the refractive field mapping beam shapers is that the transformation is realized in control manner by accurate introducing and further compensation of wave aberration, therefore the resulting collimated output beam has low divergence and there is no deterioration of the beam consistency. On the other hand this allows adapting the beam shapers to create final laser spots of required shape and intensity profile. Figure 3 Experimental and theoretical intensity profiles: Left Input TEM 00 beam, Right - after the Shaper (Courtesy of Laser-Laboratorium Göttingen e.v.)
2.3 Propagation of Flat-top beams Behavior of light is very good investigated, for example the diffraction theory 4,7 is successfully used to analyze the irradiance distribution transformation while the light beam propagation. When a TEM 00 laser beam with Gaussian intensity distribution propagates in space its size varies due to inherent beam divergence but the intensity distribution stays stable, this is a famous and widely used feature of TEM 00 beams. But this brilliant feature is valid for Gaussian beams only! When propagation of coherent light beams with non-gaussian intensity distributions, for example flat-top beams, they get simultaneously variation of both size and intensity profile, Fig. 4: at certain distance from initial plane with uniform intensity distribution there appears a bright rim that is then transformed to more complicated circular fringe pattern, finally at long distance (far field) the profile is featured with relatively bright central spot and weak diffraction rings this is the well-known Airy disk intensity distribution described mathematically by formula I ( ) = I 0 [J 1 (2 )/(2 )] 2 (3) where J 1 is the Bessel function of 1 st kind, 1 st order, is polar radius, I 0 is a constant. The Airy disk function is result of Fourier-Bessel transform for a circular beam of uniform initial intensity 7. Figure 4. Intensity profile variation by a flat-top beam propagation Evidently, even a pure theoretical flat-top beam is transformed to a beam with essentially non-uniform intensity profile. There exists, however, certain propagation length where the profile is relatively stable, - this length is in reverse proportion to a wavelength and in square proportion to beam size. For visible light and flat-top beam of 6 mm diameter the length, where deviation from uniformity doesn t exceed +10%, is about 200-300 mm, for the 12 mm beam it is about 1 meter. 2.4 Imaging of Flat-top Beams A proved and reliable way to overcome the above considered unwished diffraction effects is imaging of the output aperture of a beam shaper. Then a flat-top profile generated at the Shaper output is restored in the image plane with a transverse magnification defined by the imaging system applied. This approach was considered in details in literature 8,9, we point out here several important for practice issues: - in geometrical optics each image point is created by a beam of rays from a corresponding object point, - Object and Image planes are optically conjugated, - the real image is always behind the lens focus, - the product of object size h and aperture angle u (exactly sinu) is constant over whole optical system: h. u = h. u = const, (4) it is implied here that an optical system is free of aberrations, for example, is aplanatic. The imaging system can be implemented as an aplanatic lens. In practice of building industrial microprocessing systems with scanning optics it is very convenient to apply two-lens imaging system, Fig. 5, where the F- lens can be one of lenses: - output of Shaper, the Object, is located in front focal plane of the lens 1, thereby this lens works as a collimator producing a collimated beam from each Object point, - the lens 2 creates the Image in its back focal plane, this can be just F- lens, - the transverse magnification is defined as a ratio of focal lengths of those lenses: = - h / h = - f 2 / f 1, (5)
- the common focus F 1+2 of this two-lens system is located between the lens 2 and its back focus F 2, so the image is again located behind the common focus of the entire system, - beams from object points are collimated in space between the lenses 1 and 2 (marked by dashed line), hence the distance between lenses isn t critical, a galvo scanner can be located there. Figure 5. Two-lens imaging layout The layout at Fig. 5 b) demonstrates the behavior of intensity profile of a low divergent laser beam: - the Object is uniformly illuminated, - a beam from each point of the object plane has low divergence, near the same like a laser beam of the similar size, 2u = 2, these features are typical for output beam of a refractive field mapping beam shapers like Shaper. Uniform intensity distribution at the Object is transformed to Airy disk (1) in common focus F 1+2 and is then restored to uniform one in Image plane. It is implied that the Image plane is just working plane, therefore, there is no need to take care for the intensity profile variation due to diffraction in other parts of optical path. To avoid any unwished diffraction effects it is necessary to take care to transmitting of full light energy through an optical system and prevent any clipping of a beam. 3. BEAM SHAPING UNIT FOR SCANNING SYSTEMS 3.1 Compact Imaging System One of implementations of two-lens imaging system is shown in Fig. 6, top: the laser beam is expanded by the beamexpander to provide a proper input beam size at the entrance of Shaper; output aperture of the Shaper is imaged into working plane by an imaging system composed from a Collimator and F- lens; scanning of image spot over working field is realized by two-mirror scanner installed between the collimator and the F- lens. Typical Shaper output beam sizes are about 6 mm, while final spot sizes in microprocessing technologies to be less than 100 µm, therefore strong de-magnifying, = 1/100 x 1/200 x, is required. On the other hand the focal length of the F- lens is usually determined by a linear working field in a particular laser equipment, typically f 2 = 100 250 mm. Hence, according to (5), the collimator focal length f 1 reaches relatively big values, up to tens of meters, that makes realization of this approach very difficult.
Figure 6. Simple (top) and compact (centre) layouts of complete Beam Shaping systems, example of creating square spots: On left input of Shaper, Centre - output of Shaper, On right final square spot 50 x 50 m 2 To realize a compact imaging system it is necessary either to apply mirrors bending the optical path, or to implement a compact design of the collimator, for example in form of telephoto lens 4 with negative and positive lenses. Example of telephoto design of collimator is presented in Fig.6 centre, usually it is realized with using off-the-shelf lenses and allows up to 5-6 times reduction of total length of the collimating part. To provide proper operation and reasonable beam size on the collimator lenses it is suggested to apply right after the Shaper a field lens that reduces beam size on collimator lenses but has no influence on the system magnification. The Object at the exit of the Shaper can be implemented as a physical aperture or iris diaphragm, this means a real object, and then the Image will have very sharp edges and repeat the shape of that aperture. Evidently, the iris diaphragm will provide a simple way to vary the resulting spot size. Applying of a square aperture is a simple and reliable way to create square shaped spot with flat-top profile and steep edges. This approach is demonstrated in Fig.6 bottom, where beam profiles at the Shaper input and output as well as final spot of 50 x 50 m 2 size are presented. If no apertures applied and output collimated beam simply propagates towards the imaging lens the Object has no a definite plane and whole space after the Shaper, where the intensity profile is flat-top, will be mapped to a corresponding space on image side. Depending on the laser specifications like wavelength, M 2, beam diameter as well as Shaper model, the flat-top profile of output beam is stable over hundreds of mm or even meters. Hence, the beam profile is stable over relatively long distance in the image space as well, in other words the extended depth of field (DOF) is provided. The DOF length can be approximately evaluated with taking into account that: - longitudinal magnification of imaging system is equal to square of the transverse magnification 4, - a virtual part of the output beam, on left from the Shaper exit in Fig. 6, to be considered as well, i.e. the length to be doubled.
3.2 Evaluation of Possible Magnification Let s evaluate the transverse magnification that can be achieved in imaging systems based on widely used in industry optical components. Assume the working wavelength = 532 nm and the imaging optical system is composed from a collimator and an F- lens as shown in Fig. 6. Let the focal length of F- lens f 2 = 100 mm and the entrance pupil diameter D = 10 mm, optical designs of modern F- lenses with such specifications allow to provide diffraction limited image quality over whole working angular field. Evidently, the aperture angle u can be found as ratio of the pupil diameter and the focal length: u = D / 2f 2. (6) On the other hand the double aperture angle 2u is defined by specifications of the beam shaping optical system providing beam profile in object space. In case of the refractive field mapping beam shaper Shaper it is the same like natural divergence of a laser beam: 2u = 2. The angle of divergence of a TEM 00 laser beam is defined as = HM 2 /( H ) (7) where - wavelength, M 2 - laser beam quality factor, - a waist radius of the Gaussian beam. Transforming the formulas (4), (5), (6) and (7) and taking the = h, that is valid for refractive beam shapers like Shaper 6_6, one can get a common expression for the achievable transverse magnification: = -2 HM 2 Hf 2 /( H HD) (8) By substituting values of the considered example: = 532 nm, = 3 mm (for Shaper 6_6), M 2 = 1, f 2 = 100 mm, D = 10 mm the calculations give the magnification down to 1/1000 x! In other words, theoretically with ordinary modern off-the-shelf industrial optical components and lasers it is possible to reduce drastically the output beam of a beam shaping system and provide resulting spot sizes of several tens of microns. In practice usually the demagnifying is down to 1/200 x. The imaging approach with refractive beam shaper Shaper is typically recommended to be applied in such applications as microwelding, patterning on polymer layers, laser marking, microprocessing applications like drilling blind vias in PCB, flat panel display repair where working laser spots of 30-50 µm size with both high flatness of intensity profile and high steepness of edges are required. 3.3 Compact Beam Shaping Unit Typical set of optical components in laser equipment for microprocessing includes: a laser, a beam-expander, galvo scanner, F- lens and some mirrors to arrange the optical path. Applying of beam shaping optics presumes using additional optical components, for example, realization of the layout in Fig. 6 implies installing the Shaper, lenses of collimating system, auxiliary mirrors to arrange optical path and simplify the system alignment. Any type of beam shaping optics requires careful alignment since behavior of flat-top beams differs from Gaussian ones. In spite of basic simplicity of using those additional optical components very often integration of beam shaping optics is braked because of additional demands to components alignment, unusual view of beam profiles, etc. To overcome these difficulties it is suggested to combine all optical components between the laser and the scanner in a separate Beam Shaping Unit (BSU) as shown in Fig. 6, centre.this BSU can be assembled and carefully adjusted out of equipment and then easily integrated in a particular industrial machine, example of BSU is presented in Fig. 7. Thus the BSU is a device to provide conversion from Gaussian to flat-top laser beam irradiance profile, has compact collimator design, and functions of laser beam expanding, adjustment and adaptation to laser and scanning Figure 7. Beam shaping unit head applied in a particular equipment. One of particular BSU designs includes Shaper 6_6 and internal collimator with focal length 6 m, thus providing 1/100 x de-magnifying when applied with F- lens of 60 mm focal length, at the same time the total length of that BSU is about 600 mm only! Easy integration in already existing equipment is a serious advantage of the BSU.
4. EXPERIMENTAL RESULTS Fig. 8 presents the experimental data of thin layer ablation on a typical photovoltaic s material. When a TEM 00 laser with Gaussian profile is used the resulting hole has smooth edge, HAZ, variable depth, elliptic shape due to astigmatism of the laser beam, Fig. 8, left. All these processing imperfections are completely removed when using the above described BSU, Fig. 8, centre: steep edges, no HAZ, constant engraving depth, round shape, exact hole size controlled by iris diaphragm installed inside BSU after the Shaper. Scanning of the final laser spot using standard galvo mirrors and F- lens allows repeatable reproduction of hole size and shape over whole working field of the system, Fig. 8, right. Figure 8. Results of thin-film ablation using fs laser, spot size about 30 µm: on left TEM 00 laser without Shaper, centre - with Shaper-based BSU, on right over scanner working field. 5. CONCLUSIONS Beam shaping optics improves those micromachining laser technologies in which laser spot homogeneity is advisable or required. Using imaging optical systems in combination with refractive beam shapers allows creating round and square flat-top laser spots of less than 100µm size with simultaneous providing the functions of scanning using usual galvo mirrors and F- lenses. To simplify integration of beam shaping optics in industrial equipment it is suggested to combine the beam expander, beam shaper and collimating part of imaging system in a separate Beam Shaping Unit that can be assembled and adjusted out of equipment and installed later in space between the laser and the scanner. 6. REFERENCES [1] Dickey, F. M., Holswade, S. C., [Laser Beam Shaping: Theory and Techniques], Marcel Dekker, New York, (2000). [2] Hoffnagle, J. A., Jefferson, C. M., Design and performance of a refractive optical system that converts a Gaussian to a flattop beam Appl. Opt. vol. 39, 5488-5499 (2000). [3] Kreuzer, J., Coherent light optical system yielding an output beam of desired intensity distribution at a desired equiphase surface US Patent 3476463, (1969). [4] Smith, W.J. [Modern Optical Engineering], McGraw-Hill, New York, (2000). [5] Laskin, A. Achromatic refractive beam shaping optics for broad spectrum laser applications Proc. SPIE 7430, Paper 7430-03 (2009). [6] Laskin, A., Achromatic Optical System for Beam Shaping US Patent 8023206, (2011). [7] Goodman, J.W. [Introduction to Fourier Optics], McGraw-Hill, New York, (1996). [8] Laskin, A.V., Laskin, V.V. Imaging techniques with refractive beam shaping optics Proc. SPIE 8490, Paper 8490-19 (2012). [9] Laskin A.V., Laskin V.V. Applying of refractive spatial beam shapers with scanning optics Proc. ICALEO 2011, Paper M604 (2011). [10] Laskin A., Laskin V. Variable beam shaping with using the same field mapping refractive beam shaper Proc. SPIE 8236, Paper 82360D (2012). [11] Laskin, A.V. [http://www.pishaper.com]. 6. ACKNOWLEDGEMENTS The authors are grateful to users of Shaper in Laser-Laboratorium Göttingen e.v. and Altechna/WOP for their active and patient work with optics discussed in this paper and kind permission to publish some experimental results.