Open Access
Add to BibSonomyAbstract
The design of a new type of plasmonic ultra-high extinction ratio micropolarizing transmission filter is presented along with an experimental demonstration. A pair of dielectric coated metal gratings couple incident TM polarized light into surface plasmons, which are fed into a central metal-insulator-metal (MIM) waveguide, followed by transmission through a sub-wavelength aperture. Extinction ratios exceeding 1011 are predicted by finite element simulation. Good absolute agreement for both the spectral and polarization response is obtained between measurement and simulations using measured geometric parameters. The filters can be easily fabricated and sized to match the pixel pitch of current focal plane arrays.
©2011 Optical Society of America
1. Introduction
Polarization resolved imaging of light plays an important role in a number of applications [1,2]. It can yield information about the orientation, material type, and roughness of a surface [3,4]. Early embodiments of this concept involved time-sequenced serial or amplitude-division parallel polarization filtering in the aperture plane of a conventional imaging system [5]. More recent designs involve micropolarizing elements that are directly integrated into the structure of individual pixels of detector arrays. These have been proposed and demonstrated using liquid crystal micropolarizing elements [6], metal nanowire and nanoparticle polarizers [7–15], and polymer micropolarizers [16]. To enable very sensitive measurements, high extinction ratios along with high transmission for the desired polarization state are needed.
Here we present simulation and experimental results for a new plasmonic linear micropolarizer with extinction ratios which can be made arbitrarily high constrained only by processing considerations, where the undesired linear polarization state is limited by direct transmission through a thick metal film. Example structures are demonstrated with near-field extinction ratios exceeding 109.
2. Overall device structure
The plasmonic micropolarizer consists of a metal film fabricated with a linear aperture in a central cavity that is surrounded by linear gratings parallel to the aperture. This is shown in cross section in Fig. 1(a) . The aperture and gratings are covered with a transparent dielectric layer to form a surface plasmon slab waveguide that supports only TM polarized modes. This part of the device is based on a structure that we have previously described, which exhibits enhanced transmission [17]. TE transmission is removed by the addition of a metal cap that covers the aperture and forms a metal-insulator-metal (MIM) waveguide. The gratings couple light into surface plasmons, and the cavity width (wc) is adjusted to obtain constructive interference between the surface plasmons and incident light at the edge of the cap. TE modes are cut off due to the sub-wavelength dielectric thickness in the MIM waveguide, while the TM modes have no cutoff [18].
Fig. 1 Plasmonic micropolarizer structure. (a) Central structure of linear micropolarizer showing aperture, cavity and first grating period. The taper angles of the aperture and gratings along with the dielectric profile were chosen to match the fabricated structures. (b) Simulated time average power flow through the aperture of a structure with 635 nm period input gratings having 19 grooves per side.
The micropolarizer behaves like a plasmonic lens, concentrating incident TM light over a large collection area into a sub-wavelength aperture where it is transmitted (see Fig. 1(b)). The characteristics of the concentration and the effective spectral range are adjustable by varying the extent of, and feature size within, the device. Large collection area structures transmit high absolute power over a narrow bandwidth, while smaller structures exhibit high transmission efficiencies over a wider bandwidth. By pixel-pitch matching to a photodiode or CCD focal plane array, measurement of the first three components of the Stokes vector (corresponding to linear and unpolarized states) can be obtained with a four pixel “super-pixel.” An illustration of this with an inset of the instantaneous magnetic field is shown in Fig. 2 . One pixel may be replaced by a micropolarizer sensitive to circular polarization, which is currently being developed and will be discussed in a future publication. This would yield the complete Stokes vector.
Fig. 2 Polarization super-pixel: illustration of how to integrate the structure as a super-pixel, which can be laid on a detector array. The inset shows a snapshot of the magnetic field perpendicular to the plane for the structure of Fig. 4 (with 20 grooves). The arrows under the aperture are the time average Poynting vector of the light exiting the aperture.
3. Experimental verification
The devices used for the experimental demonstration were fabricated on polished microscope slides coated with 200-250 nm of gold on top of a 2.5 nm titanium adhesion layer. All patterning was done by electron beam lithography in a JEOL 840 scanning electron microscope with NPGS software. In the first patterning level, linear apertures and alignment marks were defined in the positive resist PMGI (polydimethylglutarimide) and cut completely through the metal using broad beam Ar ion milling. In the second patterning level, gratings with 19 grooves per side and the central cavity were defined in the positive resist PMMA (polymethylmethacrylate) and cut partially into the gold surface using Ar ion milling. For the samples reported here, the groove and cavity depth were approximately 35 nm with grating period and cavity width as variable parameters. This was followed by blanket deposition of SiO2 using plasma enhanced chemical vapor deposition. In the final patterning level, a metal cap was fabricated over the aperture using a lift-off process. The alignment marks were also covered at this processing level to minimize the transmission of unfiltered light. Geometrical parameters (grating period and depth, cavity width, aperture width and position, cap width and position) of the fabricated structures were measured using a combination of scanning electron microscopy and atomic force microscopy.
Experimental far-field spectra were collected using an optical microscope configured for transmission measurements coupled to an Acton 300i spectrometer with a Princeton Instruments Spec-10:100BR liquid nitrogen cooled silicon CCD array detector. A fiber coupled tungsten halogen lamp provided white light for an input arm which included divergence control lenses and a rotatable high extinction ratio polarizer (Thorlabs LPVIS050). Linearly polarized white light was incident on the air side of the structure and collected with a long working distance objective (N.A. = 0.5) after transmission through the polarizing element and 1mm thick glass substrate. The ~24µm collection spot size was smaller than the 50µm pattern length and comparable to the pattern width. The collected light was focused onto the end of an optical fiber positioned in a conjugate relationship to the sample plane. The opposite end of the fiber was optically coupled to the entrance slit of the spectrometer. Raw spectra were background subtracted and normalized to a white light measurement made with no sample in place to give absolute transmission.
Two-dimensional finite element simulations were compared to experimental results. The grooves were assumed to be infinite in length in the models, which is an appropriate approximation as the lengths of the grooves are much longer than pertinent lateral dimensions. Simulations were performed with the commercial finite element analysis package, COMSOL Multiphysics, which solves the frequency domain Helmholtz equation for the electric and magnetic fields. Spectroscopic ellipsometry of fabricated structures was used to determine the optical constants and thickness of the deposited SiO2, along with optical constants of the deposited gold for use in simulation. The base gold thickness was determined to be 242 nm by fitting to the measured transmission through the smooth gold near the pattern, which is in reasonable agreement with a measurement of 230 nm performed using a stylus profilometer on a separate witness slide. The models were truncated using perfectly matched layers [19]. The far-field electric field was calculated using the Stratton-Chu formula [20]. The far-field was then integrated over an angle appropriate to the collection angle of our measurement objective (NA = 0.5) for comparison to experimental data.
Structures were fabricated with 19 grooves in each of the two gratings with 500 nm period and variable cavity width. The far-field transmission spectrum of a representative structure is shown in Fig. 3a for TM polarization (example SEM images of structure are shown in inset). The simulated and measured far-field TM transmission show good agreement in terms of peak location and absolute transmission efficiency. For this structure, 1.07% and 1.53% of the light incident on the entire structure was measured and simulated, respectively, to be collected in the far-field. Further optimization yields much higher efficiencies (see Fig. 4(a) below). This demonstrates the ability of the structure to focus light from a large collection area into the sub-wavelength aperture.
Fig. 3 Measured and simulated spectral response: model validation and ultra-high extinction ratio prediction. (a) Measured and simulated absolute TM transmission through a high-selectivity structure with 500 nm period input gratings having 19 grooves per side. The inset shows a representative SEM image of a complete structure and an off-angle detail of the central region. (b) TE transmission with the curve labeled 'TE simulation' estimating power collected by the microscope objective, and the 'TE near field' curve simulating all of the power exiting the aperture. (c) TM/TE extinction ratios. The measurement is background limited for wavelengths >600 nm.
Fig. 4 Tunable bandwidth and transmission efficiency. (a) Simulated transmission efficiency for structures with 635 nm grating period and variable number of grooves. (b) Enhancement factor as defined in the text for the same structures. All presented structures were modeled using an 800 nm cap width, a 2000 nm cavity, and a 635 nm periodicity. The 2 and 4 groove models use an h value of 45 nm, while the 20 groove model uses an h value of 20.
Two different simulated TE transmission spectra along with the measured TE transmission for the same structure are shown in Fig. 3(b). The primary contribution to the far field TE signal is, in fact, direct transmission through the gold film with the value increasing toward the green due to an interband transmission at about 470 nm wavelength [21]. The “TE near field” shown in Fig. 3(b) is the simulated power exiting the aperture. The near field TE transmission through the aperture is many orders of magnitude lower than the far-field transmission because the cap increases the effective metal thickness reducing direct transmission near the aperture. While the TE measurement agrees well with the far field simulation at short wavelengths, at longer wavelengths there is a deviation between the two, because the measurement was background limited above about 600 nm. The flat background indicates white light contamination.
The TE measurement in Fig. 3(b) was performed with an illumination spot roughly twice the size of the grating structure. Increasing the illumination spot size with a constant collection spot size (roughly the size of the structure) causes the flat background to increase uniformly while the short wavelength transmission remains unchanged. This leads to the conclusion that the flat background is due to transmission away from the structure through pinholes in the metal film. The light undergoes multiple reflections inside the glass substrate until it is scattered into the collection optics at the aperture (the major scatterer on the output side of the gold film). The measured far-field, simulated far-field, and simulated near-field extinction ratios (ER) are shown in Fig. 3(c). The simulated near-field extinction ratio peaks at 1.1x109 and remains near that value at longer wavelengths, and is limited at low wavelengths by the high direct transmission through the gold film. The good agreement between TM simulation and measurement, and TE simulation and measurement at low wavelengths confirms the near-field simulated behavior of these structures.
4. Structure optimization
Because measured TE transmission at higher wavelengths is limited by pinholes and direct transmission through the metal film itself, improving film quality, thickness or using a metal with lower skin depth will allow for increased extinction ratios. Near the peak wavelength at 700nm, in the MIM waveguide the propagation length of the lowest TE and TM mode are 33nm and 2.5um, respectively. The skin depth of gold is 14.4nm. Therefore, if the transmission through the MIM waveguide dominates, an order of magnitude in extinction ratio can be gained by adding 77nm of length on the cap while decreasing TM transmission by only 3% of its value. If direct transmission through the metal is the limiting factor, for roughly every 33nm of additional gold thickness an order of magnitude in ER will be gained.
Aluminum has a small skin depth compared to gold (and silver) over the entire visible and IR wavelength range, but also higher loss. Simulations were performed replacing the gold substrate with a 200nm aluminum/50nm gold stack, and replacing the gold cap with a 50nm gold/100nm aluminum stack. The cap was also widened by 400nm. The peak near field and far field extinction ratios increased to 2.39x1011 and 7.41x1010, respectively. The peak TM transmission maintained a high value, decreasing only from 14.0% to 9.5%. This process can be carried further allowing essentially arbitrarily large theoretical extinction ratios, which in practice will be limited only by fabrication imperfections such as surface roughness, or the finite length of the device (long MIM structures will be necessary for the highest extinction ratios). For improved performance into the blue, the gold in this structure could be replaced with silver.
The structures can be further optimized to realize very high transmission efficiencies, up to 29.9% for the desired polarization state. Figure 4(a) contains transmission spectra from simulated structures having different numbers of grooves in the gratings demonstrating how the band width and transmission efficiency can be tailored. Increasing the number of grating grooves reduces the bandwidth, but also reduces the transmission efficiency; absolute power increases as the collection area of the structures increase, but due to the increased area of the device, the efficiencies decrease. Figure 4(b) replots the groove number simulations as an enhancement factor, which is defined as the transmitted power divided by the power transmitted by a bare aperture. Enhancement factors of more than 12 were achieved for 20 grooves. This demonstrates how the structure collects more power as the collection area is increased.
All gratings are sensitive to the incoming light’s angle of incidence. The spectral performance of a high efficiency, 4 groove design was examined for incident angles from 0 to 20 degrees. Transmission efficiency decreases from about 32% to 15% and the FWHM increases slightly as the incident angle is increased to 4 degrees. For incident angles of 6-20 degrees, the transmitted signal splits into two distinct peaks with maximum efficiency remaining around 15%. However, because of the MIM cutoff based polarization filtering, there is no significant change in extinction ratio over all incident angles.
The wavelength of the peak transmission is also tunable by scaling lateral dimensions. Experimental results are shown in Fig. 5 for structures fabricated by simple geometric scaling of the grating period, cavity width and cap width. The inset shows the peak wavelength vs. grating period. Because only lateral dimensions were varied, structures for filtering different spectral bands may be fabricated using the same lithography steps, making this structure amenable to large scale integrated fabrication for simultaneous color and polarization micro-filters.
Fig. 5 Tunable peak wavelength. measured TM transmission curves for three structures whose linear dimensions (cavity width and grating periodicity) have been scaled to shift the transmission peak. The blue curve was fabricated with a period of 450 nm and a cavity width of 2070 nm. The black curve has a period of 500 nm and a cavity width of 2300 nm, and the red curve has a grating periodicity of 575 nm and a cavity width of 2650 nm. (inset) plot of peak wavelength vs grating period showing approximate linear scalability.
5. Applications of design
The overall dimensions of these micropolarizers can be matched to the pixel size of CCD or CMOS focal plane arrays. The design presented here has distinct advantages over alternate approaches. The demonstrated extinction ratio is already a factor of ten higher than alternatives operating at similar wavelengths [9,14], and potential extinction ratios are far higher than easier to fabricate infrared polarizers [13]. While alternate approaches provide only polarization filtering, this design combines spectral and polarization filtering into a single structure, which has distinct advantages in applications such as biomedical imaging [22] or advanced polarimetric image reconstruction [23]. Micropolarizers are commonly integrated with a buffer layer between the micropolarizers and detector elements, resulting in optical cross-talk that can greatly decrease the precision of polarization selectivity [24]. The design presented here can be directly fabricated into the Ohmic contact metallization of a detector, completely eliminating optical cross-talk produced by the filter. Alternatively, the basic design of Fig. 1 can be modified to include a grating structure on the output surface of the metal, which can cause beaming from sub-wavelength apertures maintaining a small spot size many wavelengths from the output face [25]. Because the micropolarizer focuses the light to a small spot size, sensor sizes may be reduced with a corresponding decrease in noise and increase in response speed compared to alternate approaches. Also, because of the plasmonic lensing effect, the microlens array commonly used to increase collection fill factor in commercial focal plane detector arrays may be eliminated.
6. Conclusion
In summary, this paper presents simulation and experimental results demonstrating a spectral and polarization micro-filter, which can be integrated onto focal plane arrays of detectors. These micro-filters have ultra-high extinction ratios, controllable spectral behavior, and can be easily manufactured using standard lithography techniques.
Acknowledgments
This work was sponsored by the Air Force Office of Scientific Research, contract number FA9550-10-C-0024. This work was supported in part by a grant of computer time from the DoD High Performance Computing Modernization Program at the Air Force Research Laboratory DoD Supercomputing Resource Center. We acknowledge fruitful discussion with Charles Durfee, help with illustrations from Kate Bachman, and fabrication assistance from Greg Nuebel and Tim Rea.
References and links
1. J. S. Tyo, D. L. Goldstein, D. B. Chenault, and J. A. Shaw, “Review of passive imaging polarimetry for remote sensing applications,” Appl. Opt. 45(22), 5453–5469 (2006). [CrossRef] [PubMed]
2. M.-R. Antonelli, A. Pierangelo, T. Novikova, P. Validire, A. Benali, B. Gayet, and A. De Martino, “Mueller matrix imaging of human colon tissue for cancer diagnostics: how Monte Carlo modeling can help in the interpretation of experimental data,” Opt. Express 18(10), 10200–10208 (2010). [CrossRef] [PubMed]
3. L. B. Wolff, “Applications of polarization camera technology,” IEEE Intell. Syst. 10(5), 30–38 (1995).
4. L. B. Wolff, “Surface orientation from polarization images,” Proc. SPIE 850, 110–121 (1995).
5. J. L. Pezzaniti and R. A. Chipman, “Mueller matrix imaging polarimetry,” Opt. Eng. 34(6), 1558–1568 (1995). [CrossRef]
6. X. J. Zhao, F. Boussaid, A. Bermak, and V. G. Chigrinov, “High-resolution thin “guest-host” micropolarizer arrays for visible imaging polarimetry,” Opt. Express 19(6), 5565–5573 (2011). [CrossRef] [PubMed]
7. G. R. Bird and M. Parrish Jr., “The wire grid as a near-infrared polarizer,” J. Opt. Soc. Am. B 50(9), 886–891 (1960). [CrossRef]
8. G. P. Nordin, J. T. Meier, P. C. Deguzman, and M. W. Jones, “Micropolarizer array for infrared imaging polarimetry,” J. Opt. Soc. Am. A 16(5), 1168–1174 (1999). [CrossRef]
9. J. Guo and D. J. Brady, “Fabrication of thin-film micropolarizer arrays for visible imaging polarimetry,” Appl. Opt. 39(10), 1486–1492 (2000). [CrossRef] [PubMed]
10. J. Zhang, Y. Yan, X. Cao, and L. Zhang, “Microarrays of silver nanowires embedded in anodic alumina membrane templates: size dependence of polarization characteristics,” Appl. Opt. 45(2), 297–304 (2006). [CrossRef] [PubMed]
11. V. Gruev, R. Perkins, and T. York, “CCD polarization imaging sensor with aluminum nanowire optical filters,” Opt. Express 18(18), 19087–19094 (2010). [CrossRef] [PubMed]
12. A. Stalmashonak, G. Seifert, A. A. Unal, U. Skrzypczak, A. Podlipensky, A. Abdolvand, and H. Graener, “Toward the production of micropolarizers by irradiation of composite glasses with silver nanoparticles,” Appl. Opt. 48(25), F37–F44 (2009). [CrossRef] [PubMed]
13. Z. Wu, P. E. Powers, A. M. Sarangan, and Q. Zhan, “Optical characterization of wiregrid micropolarizers designed for infrared imaging polarimetry,” Opt. Lett. 33(15), 1653–1655 (2008). [CrossRef] [PubMed]
14. M. Guillaumée, L. A. Dunbar, Ch. Santschi, E. Grenet, R. Eckert, O. J. F. Martin, and R. P. Stanley, “Polarization sensitive silicon photodiodes using nanostructured metallic grids,” Appl. Phys. Lett. 94(19), 193503 (2009). [CrossRef]
15. Y. Zhou and D. J. Klotzkin, “Design and parallel fabrication of wire-grid polarization arrays for polarization-resolved imaging at 1.55 microm,” Appl. Opt. 47(20), 3555–3560 (2008). [CrossRef] [PubMed]
16. V. Gruev, J. Van der Spiegel, and N. Engheta, “Dual-tier thin film polymer polarization imaging sensor,” Opt. Express 18(18), 19292–19303 (2010). [CrossRef] [PubMed]
17. P. D. Flammer, I. C. Schick, R. T. Collins, and R. E. Hollingsworth, “Interference and resonant cavity effects explain enhanced transmission through subwavelength apertures in thin metal films,” Opt. Express 15(13), 7984–7993 (2007). [CrossRef] [PubMed]
18. Q. Wang and S.-T. Ho, “Ultracompact TM-pass silicon nanophotonic waveguide polarizer and design,” IEEE Photonics J. 2(1), 49–56 (2010). [CrossRef]
19. J. Jin, The Finite Element Method in Electromagnetics, 2nd ed. (Wiley, New York, 2002).
20. J. A. Stratton, Electromagnetic Theory (McGraw-Hill, New York, 1941).
21. P. G. Etchegoin, E. C. Le Ru, and M. Meyer, “An analytic model for the optical properties of gold,” J. Chem. Phys. 125(16), 164705 (2006). [CrossRef] [PubMed]
22. A. Safrani, O. Aharon, S. Mor, O. Arnon, L. Rosenberg, and I. Abdulhalim, “Skin biomedical optical imaging system using dual-wavelength polarimetric control with liquid crystals,” J. Biomed. Opt. 15(2), 026024 (2010). [CrossRef] [PubMed]
23. D. A. LeMaster, “Stokes image reconstruction for two-color microgrid polarization imaging systems,” Opt. Express 19(15), 14604–14616 (2011). [CrossRef] [PubMed]
24. A. A. Cruz-Cabrera, S. A. Kemme, J. R. Wendt, R. R. Boye, T. R. Carter, and S. Samora, “Polarimetric imaging cross talk effects from glue separation between FPA and micropolarizer arrays at the MWIR,” Proc. SPIE 6478, 64780Q, 64780Q-13 (2007). [CrossRef]
25. H. J. Lezec, A. S. Degiron, E. Devaux, R. A. Linke, L. Martin-Moreno, F. J. Garcia-Vidal, and T. W. Ebbesen, “Beaming light from a subwavelength aperture,” Science 297(5582), 820–822 (2002). [CrossRef] [PubMed]
