Open Access
Add to BibSonomy
Abstract
In this study, a mold of a line generator consisting of a cylindrical lens array with a triangular diffractive grating was ultra-precision diamond machined with Moore 350FG 5-axis ultra-precision machine. The design and the simulation of the hybrid structure were performed using ray tracing and exact electromagnetic theory. The machined hybrid lens structure was replicated to a UV-curable polymer and was then characterized and tested. The optical measurements show that the hybrid element works as designed.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Multi-axis ultraprecision diamond machining can be used to fabricate optical components such as microlenses, mirrors, light guides, lens arrays, Fresnel lenses and refractors [1, 2]. Freeform diamond turning [3] allows making of freeform surfaces that are used in lighting applications, medical devices, bio-implants and energy applications [4, 5]. The freeform multi-axis diamond machine has several cutting modes including turning, milling, fly-cutting, grinding and ruling, and is capable of machine multi-axis tool paths. Also machining a diffractive grating on top of a freeform surface is possible, although the part shape and feature sizes are limiting factors [6, 7]. Patterning size for diffractive elements starts from 500 nm and for other optical elements from 10 µm [8]. Freeform machining holds many advantages over lithographical processes: an ability to pattern 3D-surfaces [9–11], good shape of the pattern, high patterning speed, a capacity to process different materials, manufacturing process observed in real time, and on-fly adjustments are possible [12, 13].
Lithographical processes are conventionally used in making diffractive gratings and even hybrid diffractive phase element for line generator [14] but they lack the ability to pattern three-dimensional surfaces. In this work, we attempt to overcome this limitation by generating diffractive gratings on a cylindrical lens array with ultra-precision diamond machining. We chose ruling as a machining strategy because it is well suited for machining contour type tool paths with high accuracy. The replication of the machined structure can be done with either injection molding or UV-curing processes.
In this work hybrid structure is diamond machined on a Ni-coated copper piece and then copied into a UV-curable polymer. The intensity distributions of the UV-casts are measured with laser and CCD-camera setup. Additionally, the machined slab is studied with an optical profiler.
The idea is to establish an industrial-like manufacturing process for making and replicating hybrid structures. The process from design to manufacturing revealed valuable information from the process steps and testing the optical functionality of the final structure. The handling of a large amount of optical and manufacturing data also creates challenges [15, 16].
2. Design
The hybrid structure consists of a row of three cylindrical lenses and a linear triangular shaped diffraction grating on the surface of the cylinders. The height and the radius of curvature of each cylinder are 92 µm and 5.475 mm, respectively. The period and the height of the grating are d = 4.3 µm and h = 2.2 µm, respectively, which lead to an apex angle of 90 degrees. The length of the cylinder is 10 mm and the diameter is 2 mm, which gives the total area of 10 mm × 6 mm. The illustration of the structure is shown in Fig. 1 where the grating period and the axes of the element are not in scale for clarity.
We assumed that the refractive index of the material of the hybrid object is 1.52 and the grating is optimized to operate at the wavelength of 633 nm. The cylindrical lenses focus the incident light into three lines which are located at the distance of 10.7 mm from the back surface of the object. When we add the grating on the surface of the cylindrical lenses, the combined effect is that the grating stretches the lines generated by the lenses. We designed the diffraction grating to split the light at normal incidence angle into −3 and 3 diffraction orders with a total efficiency of 78% in TE-polarization and 63% in TM-polarization. The diffraction efficiencies of the grating are shown in Fig. 2. The efficiencies are calculated using the Fourier modal method [17].
Fig. 2 Theoretical diffraction efficiencies of the grating splitting light into −3 and 3 diffraction orders in (a) TE-polarization and (b) TM-polarization.
The design and the simulation of the hybrid structure were carried out by using Zemax [18]. Since Zemax cannot calculate exact diffraction efficiencies, we created a user-defined function written in C + + to compute the amplitude and phase information of the diffracted rays. The user-defined function makes use of pre-calculated table to speed up the computations. The cylindrical lenses were also modeled using the user-defined function. The simulation of the lines generated by the hybrid element is shown in Fig. 3. The element is oriented so that the cylindrical lenses are located at the side of the illumination. The light source in the simulations is a collimated rectangular beam with the size of the element. The detector is located at the distance of 10.7 mm from the backplane of the object.
Fig. 3 Simulated detector image of the lines generated by the hybrid object. The image is calculated at the distance of 10.7 mm behind the object’s backplane.
3. Manufacturing
To generate a CAD surface file from the optical design, the cylindrical shape containing X, Y, and Z data points within 1 µm interval was imported into Creo Parametric CAD program. Two identical NURBS (non-uniform rational basis spline, a model for representing a surface) curves were generated from the points and a cylindrical surface of 10 mm × 6 mm with 0.1 µm surface accuracy was then generated. To make a comparison study of machining optical patterns three distinctive working areas of 10 mm × 6 mm were altogether generated: planar area with planar diffractive grating, period 4.3 µm and the height 2.2 µm; cylindrical area without diffractive grating; and cylindrical area with diffractive grating, period 4.3 µm and the height 2.2 µm. This allows comparison with three different areas and the use of this data for troubleshooting the manufacturing chain. Cutting paths were then generated in the Creo software and three optical areas were high speed milled with a Mikron XSM 400 milling machine on a round aluminum slab. Then 100 µm of high phosphorous (12 w-%) electroless nickel was deposited on the pre-machined slab. The coated part was then transferred to Moore 350 FG ultraprecision 5-axis diamond machine tool for finish cutting of three optical areas. Ruling based cutting toolpaths with 0.1 µm surface following accuracy were generated over three surfaces in Creo program.
The diamond machining was done in three phases: First, a controlled waviness diamond lathing tool was used to cut three optical areas, secondly cylindrical lenses were cut in one of the areas, and finally, the diffractive pattern was cut with a sharp diamond tool in one planar area and on top of the cylindrical lenses. The cutting data is given in Table 1.
Table 1. Cutting data.
Mist coolant nozzles were also set to blow a fine mixture of petroleum-based lubricant and air to the tooltip. The ruling cutting of cylindrical areas consists of moving the tool along a linear trajectory over the surfaces to be cut. In this way, the tool makes downward and upward cutting motion depending on the slope of the pattern. In ruling cutting moving along the downhill in a plunging cutting strategy is rather difficult. In the manufactured cylindrical patterns, the slope of the pattern is only 5 degrees so having 10 degrees of front clearance was sufficient for the plunging style ruling cut. Figure 4 shows the CAD design of the part with all the three optical areas that were manufactured. The ruling direction and an image of the sharp diamond tool used for ruling the diffractive gratings are shown in the picture.
To monitor the cutting process, six temperature probes were set inside the machine enclosure to observe the temperature changes. For thermal stabilization of the machining process, all the cutting programs were let to dry run for 6 hours before making the actual cut. Temperature changes recorded were within ± 0.1 °C during the successful six hour cutting time for one optical area. After the machining, the diamond tools were inspected with a 1000 × magnification confocal microscope and some 100 nm rounding off the sharp diamond tool tip was detected.
4. Characterization & Optical testing
4.1 Surface roughness measurements
The machined cylindrical area without the diffractive grating was characterized with the Wyko NT9300 optical profiler. Each of three inverse cylinders was measured in several points using the magnification of 20 in the area of approximately 230 µm × 310 µm. The measured values for the surface roughnesses Ra and Rq, as well as the radius of curvature, are presented in Table 2. Surface curvatures are approximately 0.2 mm smaller than the simulated ones.
Table 2. Surface roughness values and radius of curvature measured from the cylindrical area. Different inverse cylinders are marked with a, b, and c and 1, 2 and 3 stand for different measurement positions inside the cylinder.
4.2 Imaging of the hybrid structure
UV-casts for the characterization and optical testing were taken from the machined hybrid structure using Ormocomp UV curable polymer (US-S4, Ormocer). The cross-section of the grating lines in the hybrid structure is presented in Fig. 5. The UV-casts were imaged using LEO 1550 Gemini scanning electron microscope (SEM).
Fig. 5 SEM image from the UV-cast showing the cross-section of the grating lines in the hybrid structure.
4.3 Measurement of the intensity distribution
The intensity distribution of the fabricated element was measured by illuminating the hybrid element using a collimated HeNe laser beam and capturing the image by a CCD camera. The light of the HeNe laser was first purified by a spatial filter consisting of a microscope objective and a pinhole and then the light was collimated to the hybrid element using a lens. The image was captured by a CCD camera from a partially transparent paper screen, see Fig. 6.
Fig. 6 Cross-section of the measured intensity distribution of the hybrid element (left) and CCD camera picture of the hybrid element (right).
Figure 6 shows the measured intensity distribution at the distance of 11 mm from the hybrid element. The lengths of the lines are 20 mm as predicted by the theoretical calculations. The measured diffraction efficiencies of the hybrid element are presented in Fig. 7.
Fig. 7 Measured diffraction efficiencies of the hybrid element in (a) TE polarization, (b) TM polarization.
5. Discussion & Conclusion
The Plastics Industry Association (PLASTICS) has standards for the finish of the molds for plastic products. The smoothest standard, A-1, is the only one suitable for optical applications. The measured surface roughnesses from the lenses (Table 2) are well below the requested surface roughness values for SPI A-1 standard (Ra: 12 – 25 nm). Furthermore, the measured roughness values prove that the surface is smooth enough for optical applications. The surface roughness of the final hybrid structure on UV-cast should have surface roughness of the same order of magnitude as the measured cylindrical surface, and since the roughness is significantly smaller than used wavelength, the surface roughness has little effect on the efficiency of the beam splitter.
The fabricated hybrid structure works well compared to theoretical calculations. The measured and theoretical diffraction efficiencies, in Figs. 2 & 7, are similar; the measured total diffraction efficiencies of diffraction orders −3 and 3 for TE and TM polarizations were 48% and 43%, respectively. The light diffracting to diffraction orders −2 and 2 explains the difference between theoretical calculations and measurements. This is most likely due to some disturbances in the fabrication processes, according to our calculations, a 10% shrinkage of the UV-cast explains light going into diffraction orders −2 and 2. The measured intensity distribution correlates well with the theoretical calculations.
In this study, a pre-designed cylindrical lens array with a triangular diffractive grating, designed to work as a line generator, was ultra-precision diamond machined and replicated to UV curable polymer. The structure was then characterized and tested and it was shown that the hybrid structure works as predicted by the theoretical calculations. This technique allows the manufacturing of different types of hybrid line generators for applications in e.g. machine vision and beam alignment. By optimizing the grating on top of the cylindrical lenses, we can manipulate the intensity of the lines and achieve, for example, dashed line pattern.
Funding
TEKES/European Union-European Regional Development Fund (project no. 525/31/2010).
Acknowledgments
The work of A. Eronen was funded by the Ministry of Education, Finland, through the Graduate School of Modern Optics and Photonics.
References and links
1. E. Brinksmeier, Y. Mutlugünes, F. Klocke, J. C. Aurich, P. Shore, and H. Ohmori, “Ultra-precision grinding,” CIRP Ann. Manuf. Technol. 59(2), 652–671 (2010). [CrossRef]
2. Y. Takeuchi, S. Maeda, T. Kawai, and K. Sawada, “Manufacture of Multiple-focus Micro Fresnel Lenses by Means of Nonrotational Diamond Grooving,” CIRP Ann. Manuf. Technol. 51(1), 343–346 (2002). [CrossRef]
3. F. Z. Fang, X. D. Zhang, A. Weckenmann, G. X. Zhang, and C. Evans, “Manufacturing and measurement of freeform optics,” CIRP Ann. Manuf. Technol. 62(2), 823–846 (2013). [CrossRef]
4. E. Nova, “Structured Surfaces for Improved LED and Solar Efficiency,” EUSPEN Special Interest Group, Meeting, IPT Aachen, February, (2010).
5. E. Brinksmeier and L. Schönemann, “Ultraprecision Machining - Mechanical Processes for the Manufacturing of Functional Surfaces,” in Nanotech Europe (2009).
6. X. Jiang, P. Scott, and D. Whitehouse, “Freeform surface characterisation - a fresh strategy,” CIRP Ann. Manuf. Technol. 56(1), 553–556 (2007). [CrossRef]
7. C. Buß and D. Lindemann, “Production of Freeform Optics,” in EUSPEN Special Interest Group Meeting IPT (2010).
8. F. Van Hulst, P. Geelen, A. Gebhartdt, and R. Steinkopf, “Diamond tools for producing micro-optic elements,” Ind. Diamond Rev. 3, 58–62 (2005).
9. B. McCall, G. Birch, M. Descour, and T. Tkaczyk, “Fabrication of plastic microlens arrays for array microscopy by diamond milling techniques,” Proc. SPIE 7590, 75900A (2010). [CrossRef]
10. M. Rahman, H. Lim, K. Neo, A. Senthil Kumar, Y. Wong, and X. Li, “Tool-based nanofinishing and micromachining,” J. Mater. Process. Technol. 185(1-3), 2–16 (2007). [CrossRef]
11. L. B. Kong, C. F. Cheung, S. To, and W. B. Lee, “An investigation into surface generation in ultra-precision raster milling,” J. Mater. Process. Technol. 209(8), 4178–4185 (2009). [CrossRef]
12. T. Blümel, M. Bosse, and M. Kurz, “Application Report: μPhase® 2 HR on an Ultra-precision Lathe,” Fisba Optik GmbH, Berlin and IWF (Technical University Berlin, 2007).
13. W.-L. Zhu, S. Yang, B.-F. Ju, J. Jiang, and A. Sun, “On-machine measurement of a slow slide servo diamond-machined 3D microstructure with a curved substrate,” Meas. Sci. Technol. 26(7), 075003 (2015). [CrossRef]
14. L. C. Neto, L. B. Roberto, P. Verdonck, R. D. Mansano, G. A. Cirino, and M. A. Stefani, “Design and fabrication of a hybrid diffractive optical device for multiple-line generation over a wide angle,” Appl. Opt. 40(2), 211–218 (2001). [CrossRef] [PubMed]
15. J. Väyrynen, T. Saastamoinen, J. Mutanen, P. Pääkkönen, K. Mönkkönen, and M. Kuittinen, “Manufacturing of cylindrical diffractive lens by ruling,” Proc. SPIE 7927, 79270M (2011). [CrossRef]
16. R. Gläbe, “Distributed manufacturing of optics – challenges in data handling and quality management,” in International Molded Optics Conference (2009)
17. J. Turunen, “Diffraction theory of microrelief gratings” in Micro-optics: Elements, Systems and Applications, H. Herzig, ed. (Taylor & Francis, 1997).
18. Radiant Zemax, http://www.radiantzemax.com
