3D-printed miniature spectrometer for the visible range with a ......the visible range from 490 nm...
Transcript of 3D-printed miniature spectrometer for the visible range with a ......the visible range from 490 nm...
-
Article Open Access
3D-printed miniature spectrometer for thevisible range with a 100 × 100 μm2 footprintAndrea Toulouse1,2, Johannes Drozella1,2, Simon Thiele1,2, Harald Giessen2,3 and Alois Herkommer1,2
Abstract
The miniaturisation of spectroscopic measurement devices opens novel information channels for size critical ap-plications such as endoscopy or consumer electronics. Computational spectrometers in the micrometre size rangehave been demonstrated, however, these are calibration sensitive and based on complex reconstruction al-gorithms. Herein we present an angle-insensitive 3D-printed miniature spectrometer with a direct separated spa-tial-spectral response. The spectrometer was fabricated via two-photon direct laser writing combined with a super-fine inkjet process. It has a volume of less than 100 × 100 × 300 μm3. Its tailored and chirped high-frequency grat-ing enables strongly dispersive behaviour. The miniature spectrometer features a wavelength range of 200 nm inthe visible range from 490 nm to 690 nm. It has a spectral resolution of 9.2 ± 1.1 nm at 532 nm and 17.8 ± 1.7 nm ata wavelength of 633 nm. Printing this spectrometer directly onto camera sensors is feasible and can be replicatedfor use as a macro-pixel of a snapshot hyperspectral camera.
IntroductionThe field of micro-optics has been transformed by the
use of femtosecond direct laser writing as a 3D printingtechnology since the early 2000s. Both the complexity andsurface quality have advanced from simple microlenses1,2
and achromats3 to multi-lens and multi-aperture objectives4,5.A similar development has occurred with diffractive optics,which have evolved from simple gratings6 to stacked dif-fractive microlens systems7. Along with imaging optics,photonic crystals8, waveguides9,10, and collimators11, com-plex beam shapers12–14 have been demonstrated. This rapiddevelopment reflects the significant potential of 3D print-ing technology in micro-optics. Advancements have en-abled access to the millimetre scale and larger with 3Dprinting technology15,16. Meanwhile, the almost unlimited
optical design freedom that it offers makes further mini-aturisation possible, thereby increasing the given function-ality in decreasing volumes on the micrometre scale. Theaim of this study was to enhance both the complexity andminiaturisation of 3D-printed micro-optics and to create anentire integrated measurement system-a spectrometer-in a100 × 100 × 300 μm3 volume.The miniaturisation of spectroscopic measurement
devices opens novel information channels for size-criticalapplications. For instance, medical engineering, consumerelectronics, and downscaled chemical engineering couldbenefit from a cost-effective, efficient, and readily integ-rated spectroscopic micro-device. Spectra could be re-trieved from the tip of a distal-chip endoscope with a bend-ing radius smaller than that of an optical fibre to exploreregions that are otherwise inaccessible. With the size ofone or two orders of magnitude smaller than a smartphonecamera objective, integration into consumer electronics isapparently realisable for applications such as skin diseasediagnosis17 and counterfeit bank note detection18. Further-
© The Author(s) 2020Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, dis-tribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a
link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article′sCreative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article′s Creative Commons license andyour intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
Correspondence: Andrea Toulouse ([email protected])1Institute of Applied Optics (ITO), University of Stuttgart, Pfaffenwaldring9, 70569 Stuttgart, Germany2Research Center SCoPE, University of Stuttgart, Pfaffenwaldring 57,70569 Stuttgart, GermanyFull list of author information is available at the end of the article.
Toulouse et al. Light: Advanced Manufacturing (2020)1:5 Official journal of the JHL 2689-9620https://doi.org/10.37188/lam.2020.005 www.light-am.com
http://creativecommons.org/licenses/by/4.0/mailto:[email protected]
-
more, chemical micro-reactors (such as that proposed byPotyrailo et al.19) could be further miniaturised by preciseabsorption and emission line observation via 3D-printedminiature spectrometers integrated into a microreactor plat-form.Various concepts for miniaturised spectrometers have
been demonstrated and commercialised. There are threemain categories for these spectrometers: direct, computa-tional and filter-based. Spectrometers in the first categorycreate a spatial-spectral response by redistributing incom-ing light, which can be measured directly by a line or im-age sensor. Spectrometers in the second category create amixed but spectrum-dependent unique signal from whichthe original spectrum can be computationally reconstruc-ted (wavelength multiplexing). The third category com-bines all concepts with narrowband wavelength filteringper filter patch. The 3D-printed spectrometer presented inthis paper is assigned to the first class of direct spectromet-ers.To the best of our knowledge, direct spectrometers
presented in the literature possess a footprint area that is atleast two orders of magnitude larger than our 3D-printedspectrometer (see Fig. S1). These range from commercialgrating spectrometers20,21 and scanning micro-opto-electro-mechanical system spectrometers22 to integrated ap-proaches with digital planar holograms23 or ring-resonatorenhancement24. An exceptionally high ratio of bandwidthper resolution has been reached with a grating multiplex-ing approach in a centimetre-size range25. Meanwhile, thefootprint is reduced by two to six orders of magnitude inour 3D-printed spectrometer, whereas the spectral band-width per resolution ratio of these direct spectrometers canbe one to three orders of magnitude higher.In the computational spectrometer category, several new
technologies have been proposed, such as colloidalquantum dot26 or photonic crystal slab27 patches forwavelength multiplexing. The single patches are in the100-micrometre size range; however, an entire spectromet-er consists of multiple patches. Thus, the filter size must bemultiplied accordingly. It is our understanding that onlytwo spectrometers have been demonstrated in which thesystem sizes are of the same order of magnitude as ourfootprint of 100 × 100 μm2. They are based on nanowires28,29
or disordered photonic structures30, and they have a band-width per resolution ratio that is similar to the one in ourapproach. However, owing to wavelength multiplexing, thesystem must be calibrated with an iterative reconstructionalgorithm to deduce the original spectrum. In contrast, wepresent an angularly insensitive spectrometer with a dir-ectly separated spatial-spectral response.The third category-filter-based spectrometers-is based on
traditional concepts, such as dyes or Fabry-Perot etalons,along with recent concepts, such as nanowire grids, plas-monic nanohole arrays, quantum dots31, colloidal quantumdot sensors combined with Bragg mirror filter arrays32, andultrafast pyroelectric metasurfaces33. However, owing totheir narrow bandwidth filtering, a considerable portion ofthe signal is lost. In contrast, our 3D-printed spectrometercollects light with a numerical aperture (NA) that is largerthan 0.4-a light collection area of 50 × 50 μm2 at the firstlens-and redistributes the wavelengths with minimal losses.Overall, the 3D-printed miniature spectrometer presen-
ted herein marks an innovation for 3D-printed micro-op-tics in terms of complexity, while exploiting for the firsttime the micrometre size range for direct spectrometers.
Results
y
We assessed the performance of the spectrometer in aminiature volume of 100 × 100 × 300 μm3 for a wavelengthrange of 490 nm to 690 nm. In the design concept Fig. 1,the strengths of 3D printing are considered: freeform sur-faces and inherent near-perfect alignment owing to simul-taneous mount and lens fabrication, which enable heavilytilted and asymmetrical surfaces. Incoming light is collec-ted by a lens with a quadratic aperture of 50 × 50 μm2,which corresponds to an NA of 0.42 in the -direction atthe slit. The slit is realised not in air but in a photopolymerto serve simultaneously as an ink barrier. The divergingrays exiting the slit are collimated by the subsequent sur-face. This collimation lens is terminated by a tilted surfaceto compensate in advance the grating deflection angle,thereby maintaining the spectrometer within its narrowfootprint. The grating surface is then illuminated by thecollimated rays. The last surface serves as an imaging lensto separate the wavelengths at an output plane nearby, suchas for an image sensor in future applications. The full spec-trometer thus has a length below 300 μm, which is insidethe high-precision piezo z-stage range of the 3D printer.This last surface is also tilted with respect to the gratingsurface and thus restricts the chief ray angles (CRAs) at theoutput image plane to a maximum angle of 16° at a 690 nmwavelength. Therefore, each field could be coupled effi-ciently without crosstalk into an image sensor pixel withCRA limitations.The implementation of the described design principle is
shown as a sequential ray-tracing model in Fig. 1b. In ex-tending the general design concept, slight curvatures andaspherical extensions were permitted for the tilt and grat-ing surfaces to minimise aberrations. Detailed surface typedescriptions are given in the Materials and Methods sec-tion. The most important functional surfaces of the spectro-meter are highlighted on the right side of the figure. Be-
Toulouse et al. Light: Advanced Manufacturing (2020)1:5 Page 2 of 11
-
cause the spectrometer size is in the micrometre range, dif-fraction must be considered for the slit design. The slitwidth was designed to be 1.3 μm to create a single slit dif-fraction pattern with an opening angle that is suitable forthe collection NA of 0.42 (with regard to its first minimapositions at 550 nm).The grating is designed to consist of as many lines as
can be manufactured to maximise the spectrometer resolu-tion. The fabrication process generally accommodates fea-ture sizes down to 100 nm34. However, if the photoresist isprinted as a bulk material, as in the case of the grating, theresolution decreases owing to the accumulated laser power,and thus polymerisation increases in the laser focus vicin-ity. Consequently, the grating period was restricted to aminimum period of 650 nm to be sufficiently resolved (seeFig. S3).The first diffraction order of the grating is used as the
spectrometer measurement signal. The grating topographywas thus established for maximum diffraction into the firstdiffraction order. Both the zeroth and second order are spa-tially separated such that the measurement signal is not af-fected. The period is chirped and spans from 650 nm to860 nm across the grating width of 38 μm. This is a consid-erably smaller portion of the 100 μm footprint for two reas-ons. First, the total length of the spectrometer is restrictedto
-
from the first order of interest without exceeding the foot-print. Accordingly, multiple spectrometers could be prin-ted in close vicinity without crosstalk. The first-order dif-fraction is visible as being coloured with separated foci,while the non-dispersed zeroth order is initially reflected atthe right lens edge and redirected to the right part of the slitimage plane, which is well separated from the first-orderfoci.
z
Furthermore, the focal length of the last lens was optim-ised such that the first diffraction order is focused withinthe working distance of the printing objective in the -dir-ection (
-
Dip39 combined with the spectral distribution of the mono-chromator that was used for the measurement (see Fig. S2).The average slit width at full width half maximum(FWHM) is 1.3 μm, which is in good agreement with thedesign. To quantify the signal-to-noise ratio, a sinc2 func-tion with a FWHM of 1.3 μm is subtracted from the meas-urement data, and the remaining positive signal is regardedas noise. The maximum noise level is 15%, while the aver-age noise level is 5% ± 3%.
f = 2
The miniature spectrometer was experimentally charac-terised using the setup depicted in Fig. 3. A fibre-coupledmonochromator or laser (only for the linewidth measure-ment) is used as illumination source. The output facet ofthe fiber is imaged with a 100x microscope objective (fo-cal length mm) to the front of the 3D-printed spec-trometer. This configuration is used for multiple reasons.First, the illumination is concentrated on the region of in-terest, namely the spectrometer. Second, stray light fromoutside the 100 × 100 μm2 footprint is suppressed. In areal-world application, stray light could be shielded by achromium aperture without tight alignment tolerances atthe spectrometer front. Third, the angular extent of thefibre facet emission is increased owing to the magnifica-tion of −0.13 and the high NA of 0.9 of the described ima-ging setup. Accordingly, the angular insensitivity of thespectrometer, that is, the successful spatial filtering, can bedemonstrated.On the microscope side of the measurement setup
Fig. 3b, the dispersed slit image plane is recorded with a50x microscope objective with an NA of 0.55 combined
y
with a tube lens and a monochromatic image sensor. In therecorded images, the first diffraction order is visible as atight line that moves in the -direction when shifting thewavelength. In contrast, the zeroth order is stationary at theleft end of the recorded images. To measure the spatial-spectral response, an intensity profile across the centre ofthe first-order movement range is recorded, as indicated bythe white dashed line.In Fig. 4a, the normalised spatial-spectral response
measured at the slit image plane of the spectrometer isshown for a wavelength range from 490 nm to 690 nm in10 nm steps. The peaks of the single profiles are well sep-arated, and the medium noise level is on the order of 5% to20%. Each profile can be fitted with a sinc2 function toeliminate the noise, as depicted in Fig. 4b. The width of thesingle wavelength profiles underestimates the resolutioncapacity of the miniature spectrometer because the mono-chromator-line spectral width has a similar magnitude asthe resolution capacity to be measured. Fig. 4e depicts theresolution capacity.
yThe wavelength-dependent centre positions of the sinc2
fits are plotted as -positions or wavelength shifts per mi-crometre, respectively, in Fig. 4c and d. These fitted meas-urement data are consistent with the respective ray-tracingsimulation data; however, the wavelength shift per micro-metre is larger in the measurement than in the simulation.This is presumably due to the shrinkage of the photoresist,which leads to a focus shift towards the last lens block andthus to closer spacing of the wavelengths.The spectral resolution of the miniature spectrometer
a bMonochromator
or laser
Multimode fiber
λi
Image of fiber facet Dispersed slit image
Monochromatic
image sensor
Illumination
100× objective 50× objective
Miniature
spectrometer
Counts
Pix
els
Tube lens
Microscope
λn
λ1 λnλ1
λi
Fig. 3 Measurement setup. a Illumination side: A fibre-coupled monochromator or laser is imaged to the front of the miniature spectrometer.b Microscope side: The image plane of the spectrometer is magnified with a microscope. The dispersed image of the slit (first diffraction order, right) shifts correspond-ing to the illumination wavelength while the zeroth diffraction order (left) is stationary and spatially separated from the region of interest. The profile position for fur-ther evaluations is indicated as a white dashed line. The monochromatic measurements of the fibre facet and slit images are coloured for visualisation. Scale bars 20 μm
Toulouse et al. Light: Advanced Manufacturing (2020)1:5 Page 5 of 11
-
λ = 532λ = 633
532633
was assessed with narrow bandwidth illumination sources(green laser at nm and helium neon laser at
nm) (Fig. 4e). These two laser wavelengths samplethe resolution in the shorter wavelength range ( nm)and the longer wavelength range ( nm) of the totalspectral range of our spectrometer. For speckle reduction, avibrating diffuser was integrated before coupling into themultimode fibre, and multiple measurements were ob-tained for temporal averaging. At a 532 nm illuminationwavelength, the measured profile has a centre peak with acomparable and even slightly narrower FWHM linewidththan the profile simulated with the WPM. In exchange, theside lobes are more pronounced. This behaviour may bedue the non-uniform illumination of the slit or grating,comparable to the intensity distribution of a Bessel func-tion as the diffraction pattern of a ring pupil. In such a case,the FWHM width is smaller than that of the correspondingairy disc, while the side lobes carry more energy40. The res-olution at 532 nm is determined from the FWHM of1.01 μm indicated in the graph.From Fig. 4d, the measurement data of the wavelength
range from 500 nm to 560 nm are averaged to calculate the
medium wavelength shift per micrometre around the laserwavelength of 532 nm. Multiplication of this value by theFWHM yields a spectral resolution of 9.2 ± 1.1 nm, whichis close to the diffraction limit of 10.4 nm and the wave-op-tical linewidth simulation of 12.5 ± 1.5 nm. The linewidthmeasurement at the 633 nm wavelength is evaluated ac-cordingly. Here, the medium wavelength shift per micro-metre in the range from 600 nm to 660 nm is considered. Ityields a spectral resolution of 17.8 ± 1.7 nm at awavelength of 633 nm. The wave-optical linewidth simula-tion suggests a slightly better resolution of 14.5 ± 1.4 nm.This deviation could be due to imperfections in the refract-ive optical surfaces printed with the high-resolutionphotoresist IP-Dip, which has some deficiencies in terms ofsmoothness.Finally, the spectrum of an unknown light source (broad-
band multiple LED light) was recorded and compared tothat of a commercial spectrometer (see Fig. 5). Because ofthe relatively high noise level, an averaged, weighted noisesignal was subtracted from all the measured spectra (seethe Materials and Methods section and Supplementary Ma-terial for further information). A wavelength-dependent
a
b
c e
d
1.01 μm
2.48 μm
690 nm490 nm1.0
0.8
0.6
0.4
0.2
0
−10 0 10 20y (μm)
y (μ
m)
ΔλΔy
−1/(n
m μ
m−1
)
−10 0 10y (μm)
I (a.
u.)
1
0
I (a.
u.)
Raytracing simulationMeasurement
Wave-optical simulationMeasurement
λ = 633 nm
λ = 532 nm
500 550 600 650 700
λ (nm)
1.0
0.8
0.6
0.4
0.2
0
−10 0 10 20y (μm)
−10 0 10y (μm)
I (a.
u.)
1
0
I (a.
u.)
500 550 600 650 700
λ (nm)
40
20
20
25
10
15
0
5
0
−20
Fig. 4 Spatial-spectral response of the miniature spectrometer. a Measured normalised intensity profiles at the image plane of the spectrometer for illu-mination wavelengths ranging from 490 nm to 690 nm in 10 nm steps (monochromator, profile position is indicated in Fig. 3b). b Sinc2 fits of the intensity profiles froma. c Centre positions of the sinc2 fits per wavelength. d Wavelength shift per micrometre deduced from c. e Linewidth simulation and measurement with a red or greenlaser, respectively. The measured full width at half maximum is indicated with a pair of arrows. The combination of measurements d and e yield a spectral resolution of9.2 ± 1.1 nm at 532 nm and 17.8 ± 1.7 nm at 633 nm wavelength
Toulouse et al. Light: Advanced Manufacturing (2020)1:5 Page 6 of 11
-
calibration factor was determined by the quotient of aknown spectrum (halogen lamp) that was measured withboth our spectrometer and a commercial spectrometer. Themeasurement of the unknown light source was multipliedby this calibration factor. The unknown spectrum measure-ment shows a slightly lower signal in the range of 500-570 nm and a slightly higher signal in the range of 620-670 nm, with overall satisfactory accordance, although ithas a footprint eight orders of magnitude smaller. Espe-cially in the lower wavelength range, small features at dis-tinct wavelengths that are visible on the spectrum recordedusing the commercial spectrometer, are distinctly percept-ible in the measurement obtained using our spectrometer.
DiscussionThe study presented herein proves the feasibility of a
fully functional spectrometer fitted in a volume of only100 × 100 × 300 μm3. The size comparison with a standardsmartphone camera lens Fig. 6 highlights the potential ofthis technology. At a size of more than one order of mag-nitude smaller than the objective, the spectrometer can beintegrated almost non-invasively into a smartphone cam-era, i.e. printed directly onto the camera sensor, such as forsingle spectral measurements or as individual pixels of asnapshot hyperspectral camera.The feasibility of two-photon 3D printing directly onto
image sensors has been demonstrated in previous studies4,5.It should be noted that the wavelength shift per micro-metre has a value of 9.1 ± 1.1 nm/μm at 532 nm and 7.2 ±0.7 nm/μm at 633 nm. Therefore, a monochromatic imagesensor with a pixel pitch below 1 μm must be used to re-solve the spectral lines. Current trends in this direction areevident. For example, Samsung has announced a 47.3 MPsensor with a 0.7 μm pixel pitch (ISOCELL Slim GH1).Moreover, high-resolution monochromatic image sensors
have entered the mass market through Huawei Mate 9smartphone.
12·200 nm ·
(1∆λ532
+1∆λ633
)= 16.5
Furthermore, the implementation of a spectrometer fab-ricated strictly by 3D printing can be highlighted as a nov-elty. The measured ratio of bandwidth per resolution of
is in a margin similar
to those of other spectrometers that have been demon-strated in this size range28,30. In the category of direct spec-trometers, the microspectrometer presented here is the firstof its kind in this size range (see Fig. S1).
z
Nevertheless, our spectrometer has a limitation in that arelatively high noise level can be identified. A possiblesolution could be the design of an ink basin with improvedsuppression enabled by an increase in the ink layer that isclose to the slit. However, a slit elongated in the direc-tion for improved absorptive ink layer thickness wouldlikely decrease the spectrometer light efficiency. Anotherpart of the noise could originate in the utilisation of thehigh-resolution photoresist IP-Dip, which enables smallvoxel sizes and is thus well suited for the creation of high-frequency grating structures. However, a deficiency of thismaterial is the roughness of the fabricated surfaces. Kirch-ner et al.41 reported the surface roughness of IP-Dip on theorder of 100 nm, which makes this photoresist type disad-vantageous for refractive lenses. Since diffractive and re-fractive structures are both part of the spectrometer, the re-fractive surfaces are compromised. The use of a smootherphotoresist, for example, IP-S, which can exhibit an optic-al-quality root mean square surface roughness that is be-low 10 nm14, could be examined in future work. Thisphotoresist may be beneficial for spectrometers with a lar-
1.0
0.8
0.6
0.4
0.2
0
500 550 600 650 700
λ (nm)
3D-printed spectrometer
Commercial spectrometer
I (
a.u.)
Fig. 5 Comparison of a spectrum measured with the 3D-prin-ted spectrometer and with a commercial spectrometer
1 mm
100 μm
Fig. 6 Size comparison of the 3D-printed spectrometer (whitebox) with an iPhone 5s camera lens. The inset shows a micro-scope image of the fabricated spectrometer (left) and its opticaldesign principle (right)
Toulouse et al. Light: Advanced Manufacturing (2020)1:5 Page 7 of 11
-
ger footprint and lower grating frequencies. Multi-materialprinting, as demonstrated by Schmid et al.3 and Mayer etal.42, could resolve these conflicting requirements. Furtheroptimisation of 3D-printing parameters, as proposed byChidambaram et al.43, could lead to smoother surfaces atthe expense of printing time. The printing parameters havethus far been optimised with a focus on the slit and grating,which both require high resolution instead of smoothness.As presented in the introduction, our spectrometer is
classified in the direct spectrometer category because of itsdirect spatial-spectral response. However, in combinationwith a computational approach, the (static) noise can be in-terpreted as a multiplexed wavelength signal. An iterativereconstruction algorithm, such as that used by Bao et al.26,could be employed to sharpen the spectral lines of ourspectrometer. In its current configuration, the spectrometercan readily be used as a wavemeter because the centres ofthe sinc2 fits yield distinct results in the presence of noise.As a miniature wavemeter, it could be employed as ahighly integratable laser wavelength stabiliser.Additionally, the wavelength range of the spectrometer
presented here is extendable. 3D printing comes with ahigh degree of individuality at a comparably low cost.Therefore, an array of slightly different spectrometerscould be fabricated, with each being optimised for anotherpart of the spectrum. This is a major advantage with regardto spectrometers fabricated using non-classical approaches,such as colloidal quantum dots26, photonic crystal slabs27,nanowires28, and disordered photonic structures30. Thesemethods have either a rather small bandwidth or cannot beintrinsically extended to the infrared region. In our classic-al grating spectrometer approach, the only limit forwavelengths in the infrared region is either the transmit-tance of the photoresist or the absorption of the sensor thatis used to record the signal. The photoresist IP-Dip has anextinction coefficient well below 0.1 mm−1 in the infraredregion up to a 1 600 nm wavelength39. Thus, our approachcould readily be employed in this region.Moreover, the design can be based on the same prin-
ciple, as presented in Fig. 1. By leveraging minor adap-tions of surface tilts, the grating phase profile, and re-op-timisation of the aspherical surface coefficients, the spec-tral cut-out could be shifted towards longer wavelengths.For instance, a grid of four individual spectrometers couldenable a spectrometer with a wavelength range of 800 nmon a footprint of 200 × 200 μm2.
ConclusionIn this study, the feasibility of a 3D-printed spectromet-
er in a miniature volume of 100 × 100 × 300 μm3 was the-oretically assessed and experimentally proven. The fabric-
ated spectrometer has a wavelength range of 200 nm in thevisible range, and a spectral resolution of 9.2 ± 1.1 nm at532 nm and 17.8 ± 1.7 nm at a 633 nm wavelength. To thebest of the authors' knowledge, the presented spectrometeris the first spectrometer with a direct spatial-spectral re-sponse that has been demonstrated in this size range. Com-pared to computational and filter-based approaches, its an-gular insensitivity and relatively large aperture of 50 ×50 μm2 are emphasised. A hybrid direct computational ap-proach could significantly improve the average noise levelof 5 to 20%. Wavelength range extensions of 200 nm per a100 × 100 μm2 footprint were proposed in a multi-apertureapproach by implementing the same design principle foradjacent parts of the spectrum. Possible applications rangefrom endoscopy devices and consumer electronics tochemical microreactors.
Materials and methods
x ztoroidal (y)
The proposed spectrometer was designed using ZEMAXOpticStudio optical design software in the sequential modefor a wavelength range of 490 nm to 690 nm. The glassmodel that was utilised is the photoresist IP-Dip presentedby Gissibl et al.44. For ray tracing, the spectrometer was di-vided into two parts: the light collector (part 1, top lens toslit) and the dispersing imager (part 2, slit to the dispersedslit image plane). All surfaces except the grating surfaceand the photoresist/air interface behind the slit are de-scribed by a toroidal surface with a rotation radius of infin-ity. The surfaces have a cylinder-like shape with a 50 μmextension in the -direction. The surface sags aredefined by
ztoroidal (y)=cy2
1+√
1−c2y2+a1y2+a2y4+a3y6+a4y8+a5y10 (1)
c ai
x = 0
where curvature and coefficients are optimisation vari-ables. Part 1 consists of a toroidal surface with coefficientsup to the second order followed by a propagation distanceof 90 μm inside the photoresist to the slit. In part 2, ray-tra-cing for all fields starts at the centre of the slit with thesame NA as the rays focused by the collector. Accordingly,spatial filtering at the slit is considered, and part 2 can beoptimised for the best imaging (minimum spot width perwavelength at ) of the slit plane. After a distance of4.5 μm, the photoresist/air interface is simulated by a planeto complete the ink basin described later in this section.
zgrating (y)
The following optical surfaces, except the grating sur-face, are described by Eq. 1 with coefficients up to thetenth order. The grating surface is described by the sequen-tial surface-type elliptical grating 1, and the surface sags
are given by
Toulouse et al. Light: Advanced Manufacturing (2020)1:5 Page 8 of 11
-
zgrating (y) =cb2y2
1+√
1−b2y2+a1y+a2y2+a3y3+a4y4+a5y5
(2)
b, c ai
de f f
where , and coefficients are optimisation variables.This polynomial surface is superimposed by a phase pro-file resembling the variable grating period with the effect-ive grating period defined as
de f f (y) =1T0+αy+βy2+γy3+δy4 (3)
T0, α, β, γ δwhere the optimisation variables are , and .This phase profile was translated into a topography usingMATLAB by applying the first-order blazed condition fora wavelength of 550 nm at a fixed incident angle of thechief ray to the field-dependent (chirped) grating deflec-tion angle.
dy = dz = 20 x = 0
E = 0
Subsequent to the optical design, lens mounts and an inkbasin were added to the optimised lens surfaces in theSolidWorks CAD programme. A two-dimensional scalarwave-optical simulation using the wave propagation meth-od35,36 (WPM) was performed with discretisation of thevolume model with nm at for a planewave normal input that covers the full width of the spectro-meter. To approximate the absorbing behaviour of the inkbasin, a perfectly absorbing ( ) layer was introducedto the model at the inside edges of the basin. For the res-ults presented in Fig. 1d, a spectrum simulation of 490 to690 nm in steps of 40 nm was performed. Each resultingintensity distribution was normalised to its own maximumvalue in the observation plane and converted to an RGBrepresentation according to Bruton45 for display purposes.
y
In the next step, the CAD model was sliced at a distanceof 100 nm, hatched strictly in the direction of the gratinglines ( -direction) at a distance of 250 nm, and fabricatedby means of 3D dip-in two-photon DLW using PhotonicProfessional GT2 (Nanoscribe GmbH, Karlsruhe, Ger-many) and the proprietary negative-tone photoresist IP-Dip44 on a glass substrate with an ITO coating. The lasersource of the 3D printer is a pulsed femtosecond fibre laserwith a centre wavelength of 780 nm, a specified averagelaser power output of 120 mW, a pulse length of approxim-ately 100 fs at the source, and an approximate repetitionrate of 80 MHz. The spectrometer was printed with 7.5 mWlaser power and a scan speed of 15 mm/s. The total print-ing time of the single spectrometer was just below 2 h. Thedispersed slit image plane coincided with the glass surface.The polymerised samples were developed in propyleneglycol methyl ether acetate for 8 min to wash out the unex-posed photoresist. Subsequently, the sample was rinsed inan isopropanol bath for 2 min and dried with a nitrogenblower.
The slit was fabricated with the SIJ-S030 super-fineinkjet printer (SIJ Technology, Inc., Tsukuba, Japan). Theink utilised for the creation of the non-transparent struc-tures was NPSJ (Nanopaste Series, Harimatec, Inc., Geor-gia, USA). It is a conductive ink that comprises a silvernanoparticle content of 65 mass % with a particle size of12 nm. When high voltage is applied, the printer can dis-pense droplet volumes of 0.1 fl to 10 pl from a needle tipwith a diameter below 10 μm, which fits into the gapbetween the collector lens and the ink basin walls. (Detailson this fabrication method are provided in Toulouse et al.37,38.)This process was applied to fill the ink basin that wasdefined in the CAD design.The spatial-spectral response measurements were per-
formed with a fibre-coupled home-built monochromatorusing the spectrum of a dimmable 150 W quartz halogenlamp (model I-150, CUDA Products Corp., FiberopticLightsource, Florida, USA), which was used as the illumin-ation source. The linewidth measurements were performedwith a green laser source (532 nm, 5 mW) or red heliumneon laser (632.8 nm, 5 mW, model 05-LLP-831, MellesGriot, Darmstadt, Germany), respectively, in combinationwith a vibrating diffuser coupled into the same fibre as themonochromator (300 μm, NA 0.39, model M69L02, Thor-labs, New Jersey, USA). The output fibre facet was im-aged with a 100x objective with NA 0.9 (M Plan Apo HR100X, Mitutoyo, Kawasaki, Japan) to the front of the 3D-printed spectrometer. The dispersed slit image plane of thespectrometer was recorded with a monochromatic videomicroscope.
f = 200
y
x
The video microscope consisted of a 50x objective withNA 0.55 (50X Mitutoyo Plan Apo, Edmund Optics, NewJersey, USA) in combination with a mm tube lens(MT-4, Edmund Optics, New Jersey, USA) and a mono-chromatic camera (UI-3180CP-M-GL R2, IDS, Obersulm,Germany). For the linewidth measurement, the video mi-croscope was focused on the minimum detectable linewidthper wavelength, and the simulated linewidth was offset inthe -direction for the best fit. Before each measurementand for each wavelength, the camera integration time wasadjusted to record a high signal without saturation. Foreach measurement, 20 images and 20 profiles in the -dir-ection per image were recorded and averaged after subtrac-tion of a dark image. The slit width measurement was sim-ilarly performed; however, instead of the entire spectromet-er, only part 1 (collector lens, slit, and ink basin) was fab-ricated and the video microscope was focused on the slit.Here, the integration time of the camera was adjusted to theoverall maximum signal and was the same for all wavelengths.For all spectrum measurements, the noise was evaluated
and subtracted (see Supplementary Material for further in-
Toulouse et al. Light: Advanced Manufacturing (2020)1:5 Page 9 of 11
-
Im (y)Im (λ)
Ith,halogen (λ)
k (λ) = Ith,halogen (λ)/Im,halogen (λ)
formation). The wavelengths outside the spectrum of ourspectrometer were filtered with a long-pass filter (pass >500 nm) and a short-pass filter (pass < 700 nm) (modelsFEL0500 and FES0700, Thorlabs, New Jersey, USA). Thesetup shown in Fig. 3 was used. The measured profile
was translated into a wavelength-dependent signal by evaluating the centres of the sinc2 fits (see Fig. 4b-
e). The relative intensity calibration of the spectrometerwas conducted with a 150 W quartz halogen lamp (modelI-150, CUDA Products Corp., Fiberoptic Lightsource, Flor-ida, USA). A reference spectrum was recordedwith a commercial spectrometer (AvaSpec-ULS2048CL-EVO-FCPC, Avantes, Mountain Photonics GmbH, Land-berg am Lech, Germany) and was normalised to its maxim-um intensity. The calibration factor was calculated as
.
Im,LED,calib (λ) =k (λ) · Im,LED (λ)
As a light source for the unknown spectrum measure-ment, a white light LED (model MCWHLP1 and collimat-or SM2F32-A, Thorlabs, New Jersey, USA) was coupledinto the multimode fibre of the measurement setup. The finalspectrum measurement was calibrated as
.
AcknowledgementsThis work was funded by Bundesministerium für Bildung undForschung (Printoptics, Printfunction), Baden-Württemberg StiftunggGmbH (Opterial), and the European Research Council (ComplexPlas,3DPrintedOptics). The authors thank Simon Amann for building, provid-ing, and supporting the ITOM-operated monochromator used for themeasurements. We also thank Mario Hentschel for performing the SEMmeasurements of the grating surface.
Author details1Institute of Applied Optics (ITO), University of Stuttgart, Pfaffenwaldring9, 70569 Stuttgart, Germany. 2Research Center SCoPE, University of Stut-tgart, Pfaffenwaldring 57, 70569 Stuttgart, Germany. 34th Physics Insti-tute (PI4), Pfaffenwaldring 57, 70569 Stuttgart, Germany
Author contributionsAndrea Toulouse performed the optical and mechanical design, fabrica-tion steps, measurements, evaluations, interpretations, and article writ-ing. Johannes Drozella implemented and executed the wave-opticalsimulations. Simon Thiele, Harald Giessen, and Alois Herkommer de-vised the general concept together with Andrea Toulouse and providedadvice throughout the work, including the conception and writing ofthe article.
Conflict of interestThe authors declare that they have no conflict of interest.
Supplementary information is available for this paper at https://doi.org/10.37188/lam.2020.005.
Received: 09 April 2020 Revised: 12 August 2020 Accepted: 31 August 2020
Published online: 30 November 2019
References Guo, R. et al. Log-pile photonic crystal fabricated by two-photon pho-topolymerization. Journal of Optics A: Pure and Applied Optics 7, 396-399 (2005).
1.
Malinauskas, M. et al. A femtosecond laser-induced two-photon photo-polymerization technique for structuring microlenses. Journal of Op-tics 12, 35204 (2010).
2.
Schmid, M. et al. Three-dimensional direct laser written achromaticaxicons and multi-component microlenses. Optics Letters 43, 5837-5840 (2018).
3.
Gissibl, T. et al. Two-photon direct laser writing of ultracompact multi-lens objectives. Nature Photonics 10, 554-560 (2016).
4.
Thiele, S. et al. 3D-printed eagle eye: compound microlens system forfoveated imaging. Science Advances 3, e1602655 (2017).
5.
Winfield, R. J. et al. Fabrication of grating structures by simultaneousmulti-spot fs laser writing. Applied Surface Science 253, 8086-8090(2007).
6.
Thiele, S. et al. 3D printed stacked diffractive microlenses. Optics Ex-press 27, 35621-35630 (2019).
7.
von Freymann, G. et al. Three-dimensional nanostructures for photon-ics. Advanced Functional Materials 20, 1038-1052 (2010).
8.
Schumann, M. et al. Hybrid 2D-3D optical devices for integrated opticsby direct laser writing. Light: Science & Applications 3, e175 (2014).
9.
Lindenmann, N. et al. Photonic wire bonding: a novel concept for chip-scale interconnects. Optics Express 20, 17667-17677 (2012).
10.
Thiele, S. et al. Ultra-compact on-chip LED collimation optics by 3Dfemtosecond direct laser writing. Optics Letters 41, 3029-3032 (2016).
11.
Schmidt, S. et al. Tailored micro-optical freeform holograms for integ-rated complex beam shaping. Optica 7, 1279-1286 (2020).
12.
Dietrich, P. I. et al. In situ 3D nanoprinting of free-form coupling ele-ments for hybrid photonic integration. Nature Photonics 12, 241-247(2018).
13.
Gissibl, T. et al. Sub-micrometre accurate free-form optics by three-di-mensional printing on single-mode fibres. Nature Communications 7,11763 (2016).
14.
Jonušauskas, L. et al. Mesoscale laser 3D printing. Optics Express 27,15205-15221 (2019).
15.
Heinrich, A. & Rank, M. 3D Printing of Optics. (SPIE, 2018).16. Kim, S. et al. Smartphone-based multispectral imaging: system devel-opment and potential for mobile skin diagnosis. Biomedical Optics Ex-press 7, 5294-5307 (2016).
17.
Rissanen, A. et al. MEMS FPI-based smartphone hyperspectral imager.Proceedings of SPIE 9855, Next-Generation Spectroscopic Technolo-gies IX. Baltimore, Maryland, United States: SPIE, 2016, 985507.
18.
Potyrailo, R. A. et al. Fluorescence spectroscopy and multivariate spec-tral descriptor analysis for high-throughput multiparameter optimiza-tion of polymerization conditions of combinatorial 96-microreactor ar-rays. Journal of Combinatorial Chemistry 5, 8-17 (2003).
19.
Hamamatsu Datasheet. Mini-Spectrometer. SMD Series C14384MA-01(2019).
20.
Ibsen Photonics. PEBBLE VIS. 380-850 nm OEM Spectrometer (2019).21. Huang, J. et al. Miniaturized NIR spectrometer based on novel MOEMSscanning tilted grating. Micromachines 9, 478 (2018).
22.
Calafiore, G. et al. Holographic planar lightwave circuit for on-chipspectroscopy. Light: Science & Applications 3, e203 (2014).
23.
Kyotoku, B. B. C., Chen, L. & Lipson, M. Sub-nm resolution cavity en-hanced microspectrometer. Optics Express 18, 102-107 (2010).
24.
Pang, Y. J., Yao, M. L. & Liu, S. Grating multiplexing structure basedhigh-resolution infrared spectrometer. Infrared Physics & Technology 104, 103148 (2020).
25.
Bao, J. & Bawendi, M. G. A colloidal quantum dot spectrometer. Nature 523, 67-70 (2015).
26.
Toulouse et al. Light: Advanced Manufacturing (2020)1:5 Page 10 of 11
https://doi.org/10.37188/lam.2020.005https://doi.org/10.37188/lam.2020.005https://doi.org/10.37188/lam.2020.005https://doi.org/10.37188/lam.2020.005ds高亮Editorial office will modify it before online
ds高亮Please kindly revise the reference according to the following example:
Complete book: Atkinson, K. et al. (eds) Clinical Bone Marrow and Blood Stem Cell Transplantation (Cambridge Univ. Press, 2004).
Chapter in book: Harley, N. H. & Vivian, L. in Mechanisms of Disease 4th edn, Vol. 2 (eds Sodeman, W. A. & Smith, A.) Ch. 3 (Saunders, 1974).
ds高亮Please kindly revise the reference according to the following example:
Complete book: Atkinson, K. et al. (eds) Clinical Bone Marrow and Blood Stem Cell Transplantation (Cambridge Univ. Press, 2004).
Chapter in book: Harley, N. H. & Vivian, L. in Mechanisms of Disease 4th edn, Vol. 2 (eds Sodeman, W. A. & Smith, A.) Ch. 3 (Saunders, 1974).
-
Wang, Z. et al. Single-shot on-chip spectral sensors based on photoniccrystal slabs. Nature Communications 10, 1020 (2019).
27.
Yang, Z. Y. et al. Single-nanowire spectrometers. Science 365, 1017-1020 (2019).
28.
Zheng, B. J. et al. On‐chip measurement of photoluminescence withhigh sensitivity monolithic spectrometer. Advanced Optical Materials 8,2000191 (2020).
29.
Redding, B. et al. Compact spectrometer based on a disorderedphotonic chip. Nature Photonics 7, 746-751 (2013).
30.
Gildas, F. & Dan, Y. P. Review of nanostructure color filters. Journal ofNanophotonics 13, 020901 (2019).
31.
Tang, X., Ackerman, M. M. & Guyot‐Sionnest, P. Acquisition of Hyperspec-tral data with colloidal quantum dots. Laser & Photonics Reviews 13,1900165 (2019).
32.
Stewart, J. W. et al. Ultrafast pyroelectric photodetection with on-chipspectral filters. Nature Materials 19, 158-162 (2019).
33.
Thiel, M. & Hermatschweiler, M. Three-dimensional laser lithography.Optik & Photonik 6, 36-39 (2011).
34.
Schmidt, S. et al. Wave-optical modeling beyond the thin-element-ap-proximation. Optics Express 24, 30188-30200 (2016).
35.
Drozella, J. et al. Fast and comfortable GPU-accelerated wave-opticalsimulation for imaging properties and design of highly aspheric 3D-printed freeform microlens systems. Proceedings of SPIE 11105, NovelOptical Systems, Methods, and Applications XXⅡ. San Diego, California,United States: SPIE, 2019, 1110506.
36.
Toulouse, A. et al. Alignment-free integration of apertures and non-37.
transparent hulls into 3D-printed micro-optics. Optics Letters 43, 5283-5286 (2018). Toulouse, A. et al. Super-fine inkjet process for alignment-free integra-tion of non-transparent structures into 3D-printed micro-optics. Pro-ceedings of SPIE 10930, Advanced Fabrication Technologies forMicro/Nano Optics and Photonics XⅡ. San Francisco, California, UnitedStates: SPIE, 2019, 109300W.
38.
Schmid, M., Ludescher, D. & Giessen, H. Optical properties of photores-ists for femtosecond 3D printing: refractive index, extinction, lumines-cence-dose dependence, aging, heat treatment and comparisonbetween 1-photon and 2-photon exposure. Optical Materials Express 9, 4564-4577 (2019).
39.
Gross, H. Handbook of Optical Systems. Volume 2: Physical ImageFormation. (Weinheim: Wiley-VCH, 2005).
40.
Kirchner, R. et al. Reducing the roughness of 3D micro-optics. SPIENewsroom. http://dx.doi.org/10.1117/2.1201611.006788 (2017).
41.
Mayer, F. et al. Multimaterial 3D laser microprinting using an integ-rated microfluidic system. Science Advances 5, eaau9160 (2019).
42.
Chidambaram, N. et al. Selective surface smoothening of polymer Mi-crolenses by depth confined softening. Advanced Materials Technolo-gies 2, 1700018 (2017).
43.
Gissibl, T. et al. Refractive index measurements of photo-resists forthree-dimensional direct laser writing. Optical Materials Express 7,2293-2298 (2017).
44.
Bruton, D. RGB VALUES FOR VISIBLE WAVELENGTHS. at http://www.physics.sfasu.edu/astro/color/spectra.html (1996).
45.
Toulouse et al. Light: Advanced Manufacturing (2020)1:5 Page 11 of 11
http://dx.doi.org/10.1117/2.1201611.006788http://www.physics.sfasu.edu/astro/color/spectra.htmlhttp://www.physics.sfasu.edu/astro/color/spectra.htmlhttp://dx.doi.org/10.1117/2.1201611.006788http://www.physics.sfasu.edu/astro/color/spectra.htmlhttp://www.physics.sfasu.edu/astro/color/spectra.htmlhttp://dx.doi.org/10.1117/2.1201611.006788http://www.physics.sfasu.edu/astro/color/spectra.htmlhttp://www.physics.sfasu.edu/astro/color/spectra.html