We demonstrate that black silicon (b-Si) made by dry plasma etching is a promising substrate for laser three-dimensional (3D) micro/nano-polymerization. High aspect ratio Si-needles, working as sacrificial support structures, have flexibility required to relax interface stresses between substrate and the polymerized micro-/nano- objects. Surface of b-Si can be made electrically conductive by metal deposition and, at the same time, can preserve low optical reflectivity beneficial for polymerization by direct laser writing. 3D laser polymerization usually performed at the irradiation conditions close to the dielectric breakdown is possible on non-reflective and not metallic surfaces. Here we show that low reflectivity and high metallic conductivity are not counter- exclusive properties for laser polymerization. Electrical conductivity of substrate and its permeability in liquids are promising for bio- and electroplating applications.
© 2013 Optical Society of America
Three-dimensional (3D) laser structuring on and in optically transparent substrates, usually glasses and photo-polymers, has number of applications ranging from micro-optics to imaging of cell cultures and micro-organisms in controlled and well-defined cages [1–4]. Such 3D enclosures are important for tracing several generations of cells and their division . Reflection microscopy as well as micro- optical elements performing in reflection can find all very same applications as compared with those functional in transmission. However, conditions to make polymerized structures on reflective surfaces are problematic since reflection alters 3D micro/nano-structuring of polymers. Interference is important at the pre-surface region and even short femtosecond laser pulses (100 fs takes up 30 μm in length) experience self-interference with its reflected replica [6, 7]. In case of 3D direct laser writing (DLW) and holographic recording  this causes change in structural morphology of the polymerized structures.
DLW on thin absorbing or reflective films e.g., indium tin oxide (ITO) or thin coated metal layers (Au, Ag, Cr), usually results in ablation and peeling off at the irradiation location even at the fluence/irradiance of polymerization threshold. This is due to strong absorption and the high fluence conditions required for 3D laser writing in photo-polymer. How reflectivity of substrate affects formation of extended 3D micro and nano-structures has an important engineering aspect to be better understood. Capability to form 3D structures on conductive substrate can be used for electrochemical coating and metalization and is popular for fabrication of plasmonic meta-materials.
Here, we compare DLW by fs-laser pulses on glass, reflective Si (covered with a 2-nm-thick native SiO2), and black-Si  (b-Si) which is made by dry plasma etching and has reflectivity below 2% at the wavelength of structuring, λ = 515 nm. For comparison, the same substrates were coated by 100 nm of Au and polymerization was carried out. Low reflectivity useful in recovery of 3D polymerized structures at pre-breakdown irradiance can be combined with high electrical conductivity important for electro-plating of micro-optical elements. Sheet resistance of b-Si well below 1 Ω/sq can be obtained after sputtering few hundreds of nanometers of Au.
2.1. Black silicon
Substrates used for 3D polymerization of woodpile structures were: (i) cover glass, (ii) Si, (iii) b-Si, (iv) 100 nm Au coated over a cover glass and (v) 100 nm Au coted on b-Si. The scanning electron microscopy (SEM) image and the reflection spectrum of b-Si used in this study is shown in Fig. 1. We used b-Si after 25 min of dry plasma etching in SF6 and O2 mixture. The obtained b-Si had comparatively uniform height of the spikes of 3 μm and the aspect ratio approaching 10. An SEM image and reflection spectrum of b-Si is shown in Fig. 1. The reflectivity measurement was performed with integrating sphere setup. A p-type boron doped 100 mm commercial silicon wafer was used as b-Si  substrate. It has a specific resistitivy of 10–20 Ωcm, 〈100〉 orientation and 525 ± 25 μm thickness (Atecom Ltd, Taiwan). The dry etching process was carried out on an Oxford PlasmaLab 100 ICP380 system: SF6 gas flow rate of 65 sccm, O2 gas flow rate 44 sccm, process pressure of 35 mTorr, 100 W RIE power, 20°C electrode temperature and 10 Torr He backside cooling pressure.
2.2. Direct laser writing
The logpile photonic crystal structures with 2 μm lateral period were chosen as a test objects, as such woodpiles would have a stop gap at approximately 4–5 μm. After experimentally obtaining optimal DLW parameters for structuring, a gradient pore cage structures for cell culturing/isolation were produced on a Au coated b-Si. Such templates could localize cells and supply the needed growth medium from beneath in order to control their expansion and proliferation. Additionally, electric signals could be induced to trigger specific cell behavior.
The resist SZ2080  with Irgacure 369 (Irg) photo-initiator was used for the DLW. Photo-polymer was drop-casted and dried over night (12 h) at room ambience and then used for laser structuring. More, details on resist composition and properties can be found elsewhere . DLW was performed using second harmonic wavelengths λ = 515 nm of Yb:KGW ”Pharos” (Light Conversion, Ltd.) in the window of laser powers from 70 to 250 μW at 200 kHz repetition rate . All the samples were scanned at 1 mm/s using Aerotech linear stages (assembled by Altechna R&D, Ltd.) which resulted to an overlap of approximately ∼100 pulses over the focal spot which was d = 1.22λ/NA = 450 nm in diameter under focusing with an objective lens of numerical aperture NA = 1.4. Laser structuring of SZ2080 with Irg at 515 nm corresponds to the conditions of maximized two- photon absorption (TPA) and smallest nonlinear refractive index n2. Indeed, the DLW wavelength of 515 nm factored by x = 0.7 (where the maximum of TPA is expected) is close to the maximum of photoluminescence which corresponds to the direct (one- photon) absorption process: 0.7 × 515 = 360 nm . After DLW, the development of the samples was carried out in 4-methyl-2-pentanone for 1 h to reveal the produced microstructures.
It is instructive to estimate irradiance at the focal spot for typical conditions when high quality 3D structures were made on b-Si. For the typical Ep = 0.5 nJ the irradiance per pulse assuming Gaussian temporal profile the irradadiance is Ip = 2(Ep/tp)/(π(d/2)2) = 2 TW/cm2, where pulse duration tp = 300 fs. This corresponds to a pre-breakdown condition in transparent dielectics .
2.3. Sheet resistance and characterization
Woodpile structures of the face centered cubic (fcc) symmetry  with lateral period of 2 μm were fabricated on different substrates and irradiances; their morphology was inspected by SEM (Hitachi TM-1000 and CamScan Apollo 300 for a high resolution). Electical conductivity of the Au films sputtered (with Quorum Q1507 ES) on the substrates was measured by van der Pauw method using Hall mobility measurement setup HMS3000. The sheet resistance Rs [Ω/sq] was measured on the simultaneously co-sputtered b-Si and glass substrates. Four corner contacts required for the van der Pauw method were applied to the samples at 200 °C temperature using Sn(95%)-In(5%) alloy. Resistance measurements were carried out at two different currents of 0.1 and 1 mA in order to check the ohmic character of the contacts (if contacts are not ohmic strong departure between the determined values of Rs would appear). The average of 4 measurements was taken for Rs.
3. Results and discussion
3.1. Laser polymerization on (non)reflective substrates
Figure 2 shows set of SEM images of woodpile structures recorded at different laser powers (at 200 kHz repetition rate) on substrates with different reflectivity. Missing parts of the structures are caused by optical breakdown which was observed during DLW and caused extended damage region later washed out during development. As one can see, the b-Si substrate with lowest reflectivity was the most suitable to recover high quality 3D woodpile structures within the widest range of irradiances. It is also obvious, that high reflectivity of substrates was detrimental for recovery of woodpiles with good integrity and quality. This is understandable since a narrow window of irradiances/fluences suitable for polymerization by DLW is already very close to the dielectric breakdown threshold as we demonstrated earlier . Hence, minor intensity augmentation due to interference or inhomogeneity of absorption in the resist drives catastrophic explosion due to breakdown. As a result, quality of polymerized structures suffers or they are destroyed and not recovered after development.
As shown in Fig. 3 with higher magnification, flexibility of black-Si spikes helped to relax the interface stresses between the 3D woodpile structure and substrate. Close-up view shows perfect integration of the tip-tops of spikes into the 3D polymerized structure. Also, after development of unexposed resist, the surface of such complex and high aspect ratio pillars was recovered to apparently initial conditions. This is a remarkable feature not expected for fragile and high aspect ratio spikes. Also, the stress relaxation within couple of the first layers allowed to obtain good quality woodpiles (see, (b)). As photonic crystals (PhCs) have a reflection band at the designed wavelength, the surface which has no reflection is a welcome feature for optical performance of PhCs in reflection. The b-Si substrate would suite PhCs functional in visible (note, the structures fabricated in this work have the stop-gap in IR around 4–5 μm where b-Si has strong reflectivity).
At high pulse energy the optical breakdown is possible as shown in Fig. 4. The very first layer is shown failing due to micro-explosions triggered by dielectric breakdown Fig. 4(a). Reflection as well as light field enhancement and thermal heating at the tips of spikes all contributed to the enhanced polymerization on the Au-coated b-Si. Due to the optical breakdown, woodpile structures formed on reflective substrates and metals are not recovered after development. As the structures start to be formed from the substrate, this is the location where the failure, detachment, surface ablation of metal, etc., takes place. Though a narrow, a window of 3D structuring exists on Au coated b-Si (Fig. 2).
3.2. Laser polymerization on electrically conductive black-Si
How the reflectivity of b-Si is affected by Au deposition was examined next. Having electrically conductive substrates with sheet resistance smaller than 1 Ω/sq is required for electrochemical coating. It would be beneficial to have electrically conductive but not reflective substrate. This was tested on b-Si. Figure 5 shows how sheet resistance evolves with thickness of Au coating on b -Si and test sample of glass co-sputtered simultaneously. Black appearance of gold coated b-Si was not changed suggesting strong scattering and absorption. With 100 nm Au coating made by magnetron sputtering the 1 Ω/sq sheet resistance required for reliable electrical contact is obtained on glass. The b-Si has its p-type doping-defined resistivity for a thin, up to 10-nm-thick Au films. Such layer does not make continuous film over b-Si and forms nano- islands. As the islands grow and percolate for thicker Au films, a conductive film develops as recognized by a steep drop of the sheet resistance (Fig. 5(a)). The following plateau region in resistance of b-Si ≈ 50 Ω/sq with further sputtering of Au corresponds to formation of the bottom layer of Au and its advancement onto the ”trunks” of the b-Si spikes. when percolation of the Au islands of the vertical trunks is finished a new lower Rs value ≈ 1 – 2 Ω/sq is reached. From approximately t = 200 nm, addition of Au layers by coating makes a sheet resistance increasingly smaller; this can be rationalized as an increasing number of resistors (thin films) connected in parallel. It is instructive to compare resistivity of a t = 100 nm Au film on glass (Rs = 1Ω/sq) with that of bulk Au ρAu = 2.44 × 10−8 Ωm. One would find that ρ = Rs × t = 1 × 10−7 Ωm, which is close to 10% resistivity of pure Au.
We coated 100 nm of Au over b-Si, such layer was used for laser polymerization (see Fig. 2), and measured reflectivity for the normal incidence (not shown here). Such thickness of Au and normal incidence corresponds to the DLW conditions (see Fig. 2). Reflectivity was close to 0.1% and had a slight modulation over the employed 600 – 900 nm spectral window. The interference corresponded to the 3.7-μm-thick layer of air (refractive index n = 1). It is close to the actual height of b-Si spikes which was 3 ± 0.3μm. The difference can be caused by the effective refractive index of air and Si-spikes. Surprisingly, the measured reflectivity was by 0.1% smaller than that of uncoated b-Si. The same trend of b-Si becoming more “black” (non-reflective) was confirmed for thicker Au coatings as well. Figure 5(b) shows a photo image where b-Si with 500 nm Au coating appears slightly darker.
Experimental measurements of the sheet resistance over a broad change of more than three orders of magnitude show that surface conductivity can be tuned to the values required for electro-plating and electrochemical deposition from electrolyte solutions. At the same time, reflectivity of b-Si is small and DLW can be performed with the same quality and fidelity as on the popular glass substrates. Having liquid permeable substrate below polymerized woodpiles can be useful for their coating by electroplating. Such metal coated PhC structures perform as edge filters at IR wavelengths .
One strategy to enlarge the fabrication window would be to write an interfacing part of the woodpile structure at a lower pulse energy, and reach the required dose by repeated scans. This approach is working on other conductive and rough substrates . In this study, we compared exactly the same conditions of polymerization on various substrates, hence, optimization was not carried out. Localization of heat at the tips of needles, plasmonic field enhancement effects, and temporal chirp influence of ionization can all influence polymerization and have to be investigated separately. A diffusional recovery of polymerization from quenching  is expected to be similar to cooling time at the irradiation spot; the temperature recovery on nano-objects such as tips of nano-needles takes longer . Such optimization should show pathways to broaden the window of high-quality fabrication.
Micro/nano-structured Si surfaces are used to control cell adhesion . Cages of different size and wall geometries can be formed by DLW on black-Si. The idea is illustrated in Fig. 6; the height of the walls can be arbitrary and are only constrained by mechanical properties of polymer [19, 20]. The height of the spikes can be controlled by duration of dry etching and was linear for the first 25 min with the growth rate of 90 nm/min. The cells in the cages would be interconnected at the sub-substrate level which is liquid permeable but not transparent for the cell migration. Au-coated spikes of Si can be used as a mechanical perturbation of cell membranes and forms hot-spots on Au-coated tips under illumination. Substrates with mechanical and optical functionality at the nano-scale level interacting with cell membrane are strongly sought after in bio-medical and bio-physics fields and could help to reveal the mechanisms governing membrane’s receptors and drug delivery. Low reflectance of b-Si can prove useful in cell microscopy and imaging applications as well.
Novel substrate which combines a low back reflectance with a high electrical conductivity is demonstrated on the basis of the plasma etched black-Si. Mechanical flexibility of Si needles facilitates stress release at the interface, a highly required feature for the large area 3D poly-merization by DLW. Low reflectivity of the substrate provides a wider parameter space when high-quality 3D structures can be fabricated. Permeability of b-Si in liquid solutions opens applications in electrochemical coating and bio-/medical microscopy and imaging as well as cell studies. These directions are currently at the front edge of fabrication of metamaterials for plasmonics and cell bio-imaging. As b-Si can be made inexpensively by plasma and laser assisted processes , wide range of applications in light harvesting, photo- electrodes, bolometers, catalysts, surface enhanced Raman scattering, fuel cells and super- capacitors [10, 22–25] is expected to benefit from useful mechanical and optical properties of b-Si.
This work was suported in part by a VP1-3.1-SMM-01-V-02-001 visit project grant (for SJ) from the Reseach Council of Lithuania. AZ, MM and AK acknowledges the research support by a research grant No. VP1-3.1-SMM-10-V-02-007 ( Development and Utilization of a New Generation Industrial Laser Material Processing Using Ultrashort Pulse Lasers for Industrial Applications) from the European Social Fund Agency. We are grateful to Dr. Tadas Malinauskas for setup and measurements of back-reflection from black silicon and to Colin Welch and Robert Gunn from Oxford Instruments Plasma Technology, UK for initial dry etching recipe set-up and discussions.
References and links
1. Y. Hanada, K. Sugioka, H. Kawano, I. S. Ishikawa, A. Miyawaki, and K. Midorikawa, “Nano-aquarium for dynamic observation of living cells fabricated by femtosecond laser direct writing of photostructurable glass,” Biomed. Microdev. 10, 403–410 (2008) [CrossRef] .
2. Y. Hanada, K. Sugioka, I. Shihira-Ishikawa, H. Kawano, A. Miyawaki, and K. Midorikawa, “3D microfluidic chips with integrated functional microelements fabricated by a femtosecond laser for studying the gliding mechanism of cyanobacteria,” Lab Chip. 11, 2109–2115 (2011) [CrossRef] [PubMed] .
3. D. Day, K. Pham, M. J. Ludford-Menting, J. Oliaro, D. Izon, S. Russell, and M. Gu, “A method for prolonged imaging of motile lymphocytes,” Immunol. Cell. Biol. 87, 154–158 (2009) [CrossRef] .
4. M. Malinauskas, H. Gilbergs, A. Žukauskas, K. Belazaras, V. Purlys, M. Rutkauskas, G. Bičkauskaitė, D. Paipulas, R. Gadonas, A. Piskarskas, M. Farsari, and S. Juodkazis, “Femtosecond laser polymerization of hybrid/integrated micro-optical elements and their characterization,” J. Opt. 12, 124010 (2010) [CrossRef] .
5. K. Pham, R. Shimoni, M. J. Ludford-Menting, C. J. Nowell, P. Lobachevsky, Z. Bomzon, M. Gu, T. P. Speed, C. J. McGlade, and S. M. Russell, “Divergent lymphocyte signalling revealed by a powerful new tool for analysis of time-lapse microscopy,” Immunol. Cell Biol. [CrossRef] (2012 (in press)).
6. C. Reinhardt, S. Passinger, B. Chichkov, C. Marquart, I. Radko, and S. Bozhevolnyi, “Laser-fabricated dielectric optical components for surface plasmon polaritons,” Opt. Lett. 31, 1307–1309 (2006) [CrossRef] [PubMed] .
7. S. Rekstyte, A. Zukauskas, V. Purlys, Y. Gordienko, and M. Malinauskas, “Direct laser writing of 3D micro/nanostructures on opaque surfaces,” Proc. SPIE 8431, 843123 (2012) [CrossRef] .
8. T. Kondo, S. Matsuo, S. Juodkazis, and H. Misawa, “A novel femtosecond laser interference technique with diffractive beam splitter for fabrication of three-dimensional photonic crystals,” Appl. Phys. Lett. 79, 725–727 (2001) [CrossRef] .
9. H. Jansen, M. de Boer, R. Legtenberg, and M. Elwenspoek, “The black silicon method: a universal method for determining the parameter setting of a fluorine-based reactive ion etcher in deep silicon trench etching with profile control,” J. Micromech. Microeng. 5, 115–120 (1995) [CrossRef] .
10. J. Pezoldt, T. Kups, M. Stubenrauch, and M. Fischer, “Black luminescent silicon,” Physica Status Solidi (c) 8, 1021–1026 (2011) [CrossRef] .
11. A. Ovsianikov, J. Viertl, B. Chichkov, M. Oubaha, B. MacCraith, I. Sakellari, A. Giakoumaki, D. Gray, M. Vamvakaki, M. Farsari, and C. Fotakis, “Ultra-low shrinkage hybrid photosensitive material for two-photon polymerization microfabrication,” ACS Nano 2, 2257–2262 (2008) [CrossRef] .
12. M. Malinauskas, A. Žukauskas, G. Bičkauskaitė, R. Gadonas, and S. Juodkazis, “Mechanisms of three-dimensional structuring of photo-polymers by tightly focussed femtosecond laser pulses,” Opt. Express 18, 10209–10221 (2010) [CrossRef] [PubMed] .
13. M. Malinauskas, V. Purlys, M. Rutkauskas, A. Gaidukevičiutė, and R. Gadonas, “Femtosecond visible light induced two-photon photopolymerization for 3D micro/nanostructuring in photoresists and photopolymers,” Lith. J. Phys. 50, 201–208 (2010) [CrossRef] .
14. K. Ho, C. Chan, C. Soukoulis, R. Biswas, and M. Sigalas, “Photonic band gaps in three dimensions: new layer-by-layer periodic structures,” Sol. Stat. Comm. 89, 413–416 (1994) [CrossRef] .
15. V. Mizeikis, S. Juodkazis, R. Tarozaitė, J. Juodkazytė, K. Juodkazis, and H. Misawa, “Fabrication and properties of metalo-dielectric photonic crystal structures for infrared spectral region,” Opt. Express 15, 8454–8464 (2007) [CrossRef] [PubMed] .
16. I. Sakellari, E. Kabouraki, D. Gray, V. Purlys, C. Fotakis, A. Pikulin, N. Bityurin, M. Vamvakaki, and M. Farsari, “Diffusion-assisted high resolution direct femtosecond laser writing,” ACS Nano 27, 2302–2311 (2012) [CrossRef] .
17. K. Ueno, S. Juodkazis, T. Shibuya, V. Mizeikis, Y. Yokota, and H. Misawa, “Nano-particle-enhanced photo-polymerization,” J. Phys. Chem. C 113, 11720–11724 (2009) [CrossRef] .
18. A. Ranella, M. Barberoglou, S. Bakogianni, C. Fotakis, and E. Stratakis, “Tuning cell adhesion by controlling the roughness and wettability of 3D micro/nano silicon structures,” Acta Biomaterialia 6, 2711–2720 (2010) [CrossRef] [PubMed] .
19. S. Juodkazis, V. Mizeikis, K. K. Seet, H. Misawa, and U. G. K. Wegst, “Mechanical properties and tuning of three-dimensional polymeric photonic crystals,” Appl. Phys. Lett. 91, 241904 (2007) [CrossRef] .
20. T. Kondo, S. Juodkazis, and H. Misawa, “Reduction of capillary force for high-aspect ratio nanofabrication,” Appl. Phys. A 81, 1583–1586 (2005) [CrossRef] .
21. T.-H. Her, R. J. Finlay, C. Wu, S. Deliwala, and E. Mazur, “Microstructuring of silicon with femtosecond laser pulses,” Appl. Phys. Lett. 73, 1673–1675 (1998) [CrossRef] .
24. L. Taylor, T. Kirchner, N. Lavrik, and M. Sepaniak, “Surface enhanced raman spectroscopy for microfluidic pillar arrayed separation chips,” Analyst 137, 1005–1012 (2012) [CrossRef] .
25. K. Juodkazis, J. Juodkazytė, P. Kalinauskas, E. Jelmakas, and S. Juodkazis, “Photoelectrolysis of water: Solar hydrogen - achievements and perspectives,” Opt. Express 18, A147–A160 (2010) [CrossRef] [PubMed] .