Open Access
Add to BibSonomy
Abstract
3D printing of optical components can broaden access to optical fabrication. However, consumer options for 3D printing have been limited due to the form and roughness requirements for optics. Previous efforts have established a protocol for the fabrication of singlet lenses using a stereolithographic printer and simple post-processing techniques. Here we further elevate this research by building a consumer-grade 3D printed spectrometer utilizing achromatic doublet printed lenses. These lenses are fabricated using stereolithographic printers with a filled cavity and reduce chromatic focal shift by a factor of 6 over singlet lenses. The proof-of-concept spectrometer system incorporates a pinhole, two doublet lenses, and a dispersing prism. Opto-mechanics for the system were fabricated using an FDM 3D printer. Results from the fabricated system closely matched results obtained with a commercially available spectrometer device.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Additive manufacturing has expanded greatly in several fields in recent years [1–4]. 3D printing is an attractive technology due to the relatively low fabrication costs and high degree of design freedom that is available when compared to traditional methods of fabrication. However, efforts to 3D print optical components on consumer-grade printers have generally been restricted to the terahertz regime [5,6], due to the high requirements for surface roughness and form accuracy at shorter wavelengths. In response, efforts to 3D print optical components have shifted towards customized inkjet printing and two-photon printing. While inkjet printing using specialized machines can provide functional components [7,8], they may require an iterative printing process [9]. Two-photon printing has demonstrated the ability to fabricate a wide array of optical components including waveguide and fiber-based components [10,11], lenses and collimation optics [12,13], diffractive structures [14,15], phase modulation elements [16,17] and spectrometer systems [18]. However, both custom inkjet and two-photon printing systems have high costs, rendering accessibility of optical fabrication to specialized laboratories. By developing protocols for low-cost printing of optical systems, accessibility to optical fabrication can be expanded to a broader community. Past efforts to adapt consumer-grade 3D printing for whole spectrometer construction has focused on printing of casings rather than the printing of optical elements [19,20]. Here we will show a new application of consumer-grade printing for the creation of an optical system in which the optical elements were printed in a more affordable manner.
One subset of consumer-grade technologies is stereolithography. This method utilizes optics and UV curable resin to fabricate parts in a layer-by-layer manner [21]. For optical fabrication, stereolithography is an attractive process, with significantly larger printing volumes, and far lower costs and printing times when compared to two-photon printing. Additionally, stereolithography allows for a variety of clear printing materials, and can achieve higher resolutions (around 25 µm commercially) than those achievable with an extrusion-based printing method. Being able to fabricate optical systems using stereolithographic printing would greatly increase optical fabrication access by reducing both costs and infrastructure.
Previously, we have shown that stereolithographic printing materials were suitable for optical fabrication and described a methodology for the fabrication of singlet optical lenses made through the stereolithography printing process and simple post-processing steps [22]. Here, we advance our previous efforts with regard to complete system implementation and complexity of optical components. We present new processes for fabricating doublet lenses and optical dispersing prisms. These components are combined with a pinhole (fabricated through simple micrometer pin punching), FDM printed mounts and system opto-mechanics. All these components are assembled to construct a consumer-grade 3D printed spectrometer system. Spectral characteristics obtained from this spectrometer closely matched to results from a commercial spectrometer system. The spectrometer works as a proof-of-concept system to demonstrate the functionality of consumer-grade printers for the fabrication of complete optical instruments.
2. Spectrometer design
The spectrometer was designed to include two optical lenses and a dispersing prism. The general layout of the system is shown below in Fig. 1.
Before, we have fabricated singlet lenses using stereolithographic printing. However, due to expected high chromatic aberration of singlet lenses, the spectrometer was designed to incorporate doublet lenses fabricated using the previously characterized clear printing resin and NOA 61, a common UV curable optical epoxy. The NOA 61 was filled into a cavity post-printing, as had been previously demonstrated using two-photon printing [23]. The dispersing prism was designed to be composed solely of the clear resin material. Refractive index data for the resin printing material was input into Zemax and optical designs were generated for both the doublet lenses and the spectrometer system. Curvatures of doublet surfaces were selected from those that inversely matched the curvatures of commercially available concave lenses. These glass lenses acted as reusable surface finishing molds. The chosen doublet design had two convex clear resin surfaces with curvatures of 34.5 mm and 11.5 mm, with the NOA 61 portion having a concave surface of radius -11.5 mm and a planar final surface. The resulting focal length of the doublet was 88 mm. The system was designed to disperse wavelengths between 500 and 700 nm onto the sensor surface. Since curvatures of doublets were chosen based on commercially available concave lenses, the system was not fully optimized to work at diffraction limited performance. However, future molds could be made to be used with any designed curvature. Table 1 below lists the system design parameters and Fig. 2 below shows expected rms spot size of the doublet imaging system (excluding prism) at 500, 600, and 700 nm.
Fig. 2. Expected spot diagrams of doublet imaging system at 500, 600, and 700 nm. Airy disk shown in center of each plot.
Table 1. System prescription table.
3. Component fabrication
3.1 Doublet lenses
Optical elements were fabricated using a Form 2 stereolithographic printer and the corresponding clear resin as described previously [22]. Due to the high chromatic aberration expected from singlet lenses over broadband wavelengths, the fabrication method previously applied to create singlet lenses was adapted for the fabrication of doublet lenses. Doublet lenses were composed of the clear resin printing material and NOA 61, a common UV curable optical epoxy. The doublet was designed to have a diameter of 12.7 mm, but to enable easier fabrication an outer brim was designed of diameter 25 mm. The brim was needed to include a cavity in the design that would be filled with NOA 61 during the fabrication process. A schematic of the 3D printed portion of the doublet is shown below in Fig. 3.
Fig. 3. Schematic of the printed doublet design from top, side, and bottom views. An open brim is left in the top of the design to allow for filling of NOA 61 after finishing of the 3D printed surfaces. Brim on bottom surface is a closed, planar cylinder with lens surface protruding out.
Doublets were printed using the 25 µm resolution setting of the printer in such a way that printed supports did not touch optical surfaces. This resolution is sufficient for the project due to the internal isotropy of printed material and post-processing of exterior surfaces. After printing, standard washing and post-curing steps were followed. Clear resin surfaces were then post-processed by placing drops of resin (approximately 175 µL in total) on the surface of glass concave lenses with matching curvature (purchased from Thorlabs). This methodology had previously been shown to produce singlet lenses with average RMS surface roughness of 6 nm and RMS wavefront error of 0.048 wave [22]. The doublet structure was then out-gassed in a vacuum chamber (Zeny VP125+, approximately $100 on amazon.com) and cured in a Formlabs curing chamber without heat for six hours. Glass lenses could be removed following the curing process by placing the structure in the freezer for approximately five minutes due to differences in thermal expansion between the materials. After removal of the glass surfaces, NOA 61 was poured into the cavity to form a matching concave surface, out-gassed, and cured. Succeeding the curing process, the NOA 61 surface was placed on approximately 175 µL of additional NOA 61 on top of a planar optical glass window to create the final planar surface. The doublet was then out-gassed and cured a final time and the glass surface was removed by placing in the freezer an additional time. A graphical representation of the fabrication process is shown in Fig. 4. An image of a completed doublet can be seen below in Fig. 5.
Fig. 5. Completed doublet lens shown after completion of fabrication process. After processing steps, doublet surfaces are smooth and transparent. Text shown under the lens to illustrate clarity of doublet after finishing.
3.2 Dispersing prism
A 45-45-90 prism of thickness 20 mm was chosen to act as the dispersing element for the spectrometer system. Prisms were printed, washed, and post-cured according to standard printing operation. After printing, the hypotenuse of the prisms was placed on approximately 300 µL of resin on an optical flat glass surface, out-gassed, and cured. The glass was removed by placing the structure in the freezer and the glass surface was cleaned off. One leg of the prism was then placed on approximately 250 µL of resin on the glass surface, out-gassed, and cured to serve as the input surface of the prism. A completed prism can be seen below in Fig. 6.
Fig. 6. Image of completed prism surface after processing of prism surfaces. Prism is placed on Rice University logo to illustrate surface clarity. Note that for illustration purposes 3 sides of the prism were finished. Prisms used in experiments had only the hypotenuse and one leg post-processed.
3.3 Pinhole
The pinhole for the structure was fabricated by puncturing black foil with a 100 µm drill bit commercially purchased (Mudder, 57 Piece Hand Drill Bit Set – cost < $15, amazon.com). Microscope analysis of the pinhole post-fabrication indicated a diameter of approximately 70 µm.
4. Doublet characterization
Chromatic correction performance of the doublets was demonstrated by comparing the focal shift of the doublets at various wavelengths in comparison to the shift seen in singlet lenses with similar focal length (doublet focal length of 88 mm, singlet focal length of 98 mm). This was accomplished by imaging a resolution target through the printed lenses onto a camera that was positioned on a micrometer stage. 10 nm bandpass filters were placed in the illumination setup, and the amount the micrometer stage needed to be adjusted to focus the image on the camera between wavelengths was measured and recorded as the shift needed in reference to 500 nm focus. The results were also compared with expected Zemax model. The imaging was performed in a 1x setup using a doublet lens with a 7 mm iris placed before the lens to ensure only the designed optical surfaces of the doublet lens were illuminated. The singlet imaging is slightly worse at all wavelengths (see Fig. 7), likely due to the 10 nm full width at half maximum (FWHM) bandpass of the illumination filters causing chromatic aberration. Results comparing the singlet and doublet performance are shown in Tables 2 and 3 with corresponding images in Figs. 7 and 8. Figure 9 plots the comparison between focal shifts seen in the singlet and doublet lens.
Fig. 7. Images of resolution target at measured wavelengths for singlet lens. Images are digitally altered in contrast and brightness for consistency between wavelengths
Fig. 8. Recorded doublet images at measured wavelengths. Images are digitally altered in contrast and brightness for consistency between wavelengths.
Fig. 9. Comparison between focal shift for singlet and doublet. Trend line shows first order polynomial fit between measured wavelengths
Table 2. Comparison of expected to measured focal shift in singlet printed lenses.
Table 3. Comparison of expected to measured focal shift in doublet printed lenses.
Results showed a greater than six times reduction in focal shift using the doublet lenses in place of the singlet lenses. Differences between the expected focal shift as plotted in Zemax and the measured focal shift likely stem from slight variations of refractive index values that occur over time in the Clear Resin material. Studies on the refractive index of the material over a period of a week indicated fluctuations in abbe number between 51.1 and 56.9. Even with these variations, the chromatic correction of the doublets significantly improves chromatic focal shift seen in the singlets.
5. Opto-mechanic design of the spectrometer system
An image of the constructed system is shown below in Fig. 10. Component housing and system opto-mechanics were constructed using an Ultimaker S5 3D printer and black PLA. The spectrometer system consists of the entrance pinhole, a 7 mm system stop, collimating doublet, 45-45-90 dispersing prism, focusing doublet, and Flea3 monochromatic camera.
Fig. 10. Image of the printed system. System illumination consisting of fiber light source and 3D printed singlet lens not shown. Housing for individual components designed to slide in grooves in whole system casing to allow for focusing control. From left to right in image: Entrance pinhole, system stop, collimating lens, dispersing prism, focusing lens, and Flea3 monochromatic camera. Insert in top left shows image of pinhole as seen through a light microscope. Pinhole diameter was approximately 70 µm and was located at dashed box in image.
Housing for individual components included a post on the bottom that fit into grooves in the system housing to allow for adjustment of the position of individual elements. The housing for the prism included posts that allowed for rotational freedom. The camera used in the system was a Flea3 monochromatic camera that was attached to a metallic post. The post was then screwed into the 3D printed housing for the camera. The pinhole was placed over a cavity designed in the entrance wall of the system. Light was provided from a Leeds Ace fiber optic illuminator, and this light was focused using a 3D printed singlet lens of focal length 25 mm that was fabricated using the glass-curing method previously described [22].
6. System performance validation
The angle of the prism in the system was adjusted such that the sensor captured a wavelength range of 500-700 nm and the placement of individual components was aligned to focus light onto the camera. Calibration of pixel to wavelength conversion was performed by placing 10-nm bandpass filters of various center wavelength into the illumination setup of the system. The pixel column with peak intensity values was calibrated to be at the center wavelength of each filter. Calibration was performed using a polynomial fit between center wavelengths of filters used and the pixels of peak intensity. In result, a look-up table of pixel coordinates for specific wavelengths was generated. After calibration was performed, four different foil filters with different spectral distributions (parry sky blue, light green, pale rose pink, and gypsy red) from Roscolux were placed in the system and the measured spectrum was captured on the camera. A sample of captured images for the unfiltered light source and the captured image with the pale rose pink foil in place are shown in Fig. 11.
Fig. 11. Raw images captured on camera. Panel A shows a white light source. Panel B shows recorded image with the pale rose pink foil filter in the illumination setup. Sensor captured a range of approximately 500-700 nm across the length of the sensor. Image processing was done only on 500-700 nm range.
Measured spectrum was flatfield corrected by dividing the intensities captured for a white light illumination source. Only the ten central pixel rows, corresponding to the ten rows with highest light intensity, were used for image analysis. Using the look-up table previously described, recorded pixel irradiance was converted to normalized spectral distribution. Spectrums obtained using the 3D printed system were compared to those obtained using an Ocean Optics USB 4000 visible light spectrometer. Results obtained for each foil closely match the results acquired with the reference spectrometer. A comparison between the 3D printed spectrometer and the Ocean Optics results is provided in Fig. 12.
Fig. 12. Comparison of 3D printed spectrometer to commercial Ocean Optics spectrometer. For all plots, blue curves indicate 3D printed spectrum and orange indicates the commercial spectrometer results. Upper left: Parry Sky Blue foil filter, Upper right: Light Green foil filter, Lower left: Pale Rose Pink foil filter, Lower right: Gypsy Red foil filter
While the 3D printed spectrometer closely matches results taken from the commercial spectrometer, there are some observable differences, particularly when looking at data from the blue and red foils. This is most likely due to the larger point spread function of the 3D printed spectrometer impacting neighboring wavelengths. Both the parry sky blue foil and the gypsy red foil provide steeper curves than those seen in the pale rose pink and light green foils, necessitating the need for higher resolution. This hypothesis on the impact of the larger point spread function is further supported by measuring the FWHM of a filter placed in the illumination system. While the nominal FWHM of the filter given by the supplier was 10 nm, the 3D spectrometer gives a FWHM of 18 nm. The measured FWHM closely matches the Zemax simulated spectral resolution at 610 nm of 15 nm. Measured FWHM can be seen in Fig. 13 in comparison to the FWHM measurement of the Ocean Optics spectrometer. There is an observable difference in peak location between the two spectrometers of 2 nm. This differential likely stems from the asymmetry of dispersion in the 3D system, as well as the commercial spectrometer measuring the filter peak to be at a slightly shorter wavelength than the expected 610 nm center wavelength.
Fig. 13. Measurement of FWHM of 610 nm filter taken from 3D printed spectrometer in comparison to a measurement using the Ocean Optics spectrometer. Nominal FWHM is 10 nm but measured FWHM for printed spectrometer is approximately 18 nm. Difference in peaks of 2 nm is likely due to asymmetrical dispersion
7. Discussion
We have shown a method to use consumer-grade printers to develop a spectrometer system. To enable this, we developed a protocol for the fabrication of doublet lenses incorporating NOA 61. The spectrometer system closely matched results from a commercial ocean optics system. A design challenge when using the clear resin material is the fluctuating changes in abbe number observed over time. More long-term studies are needed to characterize the trends in refractive indices and abbe number to assess conditions for the material to stabilize. Development of additional printing materials for optical applications would also be beneficial for the design of systems using stereolithographic printers.
Consumer-grade printing of optical elements allows for a variety of customizable designs by changing of materials and curvatures. All glass surfaces used for fabrication can be cleaned and repurposed as many times as needed to perform small batch fabrication of optical elements and systems. A limitation to the use of glass surfaces to finish printed surfaces is the loss of geometrical freedom that makes 3D printing an attractive option for optical fabrication. While the work presented here uses glass surfaces to fabricate optical elements, other finishing methods, such as spin-coating, are potentially suitable for fabrication depending on the quality of elements needed for a given project. Further improvement in pinhole fabrication would likely lead to better performance of the system. Here, we showed a simple cost-effective fabrication methodology of punching a hole using a commercially purchased drill bit set. However, this methodology provides a limitation on achievable pinhole size. Developing better methodology for pinhole fabrication would improve system performance.
In conclusion, we have demonstrated a method for fabricating doublet lenses and prisms using a consumer-grade stereolithographic printer. The doublets were validated to provide chromatic correction in comparison to singlet lenses fabricated in a similar manner. A 3D printed spectrometer system using these doublets and prisms was housed in entirely 3D printed opto-mechanics and used to measure spectrum from various color foil filters. Results show that the 3D printed system gives comparable performance to that of a commercial spectrometer system. This work serves as a proof-of-concept for the application of consumer-grade printing to construct whole optical systems.
Funding
National Science Foundation Precise Advanced Technologies and Health Systems for Underserved Populations (PATHS-UP) Engineering Research Center (#1648451).
Disclosures
Dr. Tomasz Tkaczyk has financial interests in Attoris LLC focusing on applications and commercialization of hyperspectral imaging technologies.
Data Availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. S. V. Murphy and A. Atala, “3D bioprinting of tissues and organs,” Nat. Biotechnol. 32(8), 773–785 (2014). [CrossRef]
2. M. Vukicevic, B. Mosadegh, J. K. Min, and S. H. Little, “Cardiac 3D Printing and its Future Directions,” JACC: Cardiovascular Imaging 10(2), 171–184 (2017). [CrossRef]
3. S. J. Trenfield, A. Awad, A. Goyanes, S. Gaisford, and A. W. Basit, “3D Printing Pharmaceuticals: Drug Development to Frontline Care,” Trends Pharmacol. Sci. 39(5), 440–451 (2018). [CrossRef]
4. C. M. B. Ho, S. H. Ng, K. H. H. Li, and Y. J. Yoon, “3D printed microfluidics for biological applications,” Lab Chip 15(18), 3627 (2015). [CrossRef]
5. A. I. Hernandez-Serrano, M. Weidenbach, S. F. Busch, M. Koch, and E. Castro-Camus, “Fabrication of gradient-refractive-index lenses for terahertz applications by three-dimensional printing,” J. Opt. Soc. Am. B 33(5), 928–931 (2016). [CrossRef]
6. H. Acikgoz, R. K. Arya, and R. Mittra, “Statistical analysis of 3D-printed flat GRIN lenses,” in 2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016 - Proceedings (2016).
7. B. G. Assefa, M. Pekkarinen, T. Saastamoinen, J. Biskop, M. Kuittinen, J. Turunen, and J. Saarinen, “Design and characterization of 3D-printed freeform lenses for random illuminations,” in (2018).
8. J. Gawedzinski, M. E. Pawlowski, and T. S. Tkaczyk, “Quantitative evaluation of performance of three-dimensional printed lenses,” Opt. Eng. 56(08), 084110 (2017). [CrossRef]
9. B. G. Assefa, M. Pekkarinen, H. Partanen, J. Biskop, J. Turunen, and J. Saarinen, “Imaging-quality 3D-printed centimeter-scale lens,” Opt. Express 27(9), 12630–12637 (2019). [CrossRef]
10. M. Schröder, M. Bülters, C. von Kopylow, and R. B. Bergmann, “Novel concept for three-dimensional polymer waveguides for optical on-chip interconnects,” J. Eur. Opt. Soc. 7, 12027 (2012). [CrossRef]
11. M. Schumann, T. Bückmann, N. Gruhler, M. Wegener, and W. Pernice, “Hybrid 2D-3D optical devices for integrated optics by direct laser writing,” Light Sci. Appl. 3, e175 (2014). [CrossRef]
12. S. Thiele, T. Gissibl, H. Giessen, and A. M. Herkommer, “Ultra-compact on-chip LED collimation optics by 3D femtosecond direct laser writing,” Opt. Lett. 41(13), 3029–3032 (2016). [CrossRef]
13. T. Gissibl, S. Thiele, A. Herkommer, and H. Giessen, “Sub-micrometre accurate free-form optics by three-dimensional printing on single-mode fibres,” Nat. Commun. 7, 11763 (2016). [CrossRef]
14. J. Mu, Z. Liu, J. Li, T. Hao, Y. Wang, S. Sun, Z.-Y. Li, J. Li, W. Li, and C. Gu, “Direct laser writing of pyramidal plasmonic structures with apertures and asymmetric gratings towards efficient subwavelength light focusing,” Opt. Express 23(17), 22564–22571 (2015). [CrossRef]
15. B. K. Hsiung, R. H. Siddique, D. G. Stavenga, J. C. Otto, M. C. Allen, Y. Liu, Y. F. Lu, D. D. Deheyn, M. D. Shawkey, and T. A. Blackledge, “Rainbow peacock spiders inspire miniature super-iridescent optics,” Nat. Commun. 8(1), 2278 (2017). [CrossRef]
16. T. Gissibl, M. Schmid, and H. Giessen, “Spatial beam intensity shaping using phase masks on single-mode optical fibers fabricated by femtosecond direct laser writing,” Optica 3(4), 448–451 (2016). [CrossRef]
17. S. Lightman, R. Gvishi, G. Hurvitz, and A. Arie, “Shaping of light beams by 3D direct laser writing on facets of nonlinear crystals,” Opt. Lett. 40(19), 4460–4463 (2015). [CrossRef]
18. A. Toulouse, J. Drozella, S. Thiele, H. Giessen, and A. Herkommer, “3D-printed miniature spectrometer for the visible range with a 100 × 100 µm2 footprint,” Light: Advanced Manufacturing 2(1), 20–30 (2021). [CrossRef]
19. T. C. Wilkes, A. J. S. McGonigle, J. R. Willmott, T. D. Pering, and J. M. Cook, “Low-cost 3D printed 1 nm resolution smartphone sensor-based spectrometer: instrument design and application in ultraviolet spectroscopy,” Opt. Lett. 42(21), 4323–4326 (2017). [CrossRef]
20. T. C. Wilkes, T. D. Pering, A. J. S. McGonigle, J. R. Willmott, R. Bryant, A. L. Smalley, F. M. Mims, A. V. Parisi, and R. A. England, “The pispec: A low-cost, 3D-printed spectrometer for measuring volcanic SO2 emission rates,” Front. Earth Sci. 7, 1 (2019). [CrossRef]
21. F. P. W. Melchels, J. Feijen, and D. W. Grijpma, “A review on stereolithography and its applications in biomedical engineering,” Biomaterials 31(24), 6121 (2010). [CrossRef]
22. G. D. Berglund and T. S. Tkaczyk, “Fabrication of optical components using a consumer-grade lithographic printer,” Opt. Express 27(21), 4323–4326 (2019). [CrossRef]
23. M. A. Tadayon, S. Chaitanya, K. M. Martyniuk, J. C. McGowan, S. P. Roberts, C. A. Denny, and M. Lipson, “3D microphotonic probe for high resolution deep tissue imaging,” Opt. Express 27(16), 22352–22362 (2019). [CrossRef]
