2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
C O M M U N I C A T I O N
and the surace processing step. The resolution and structureintegrity can be very good; however, as the density o the metalbinding sites on the structure cannot be controlled, the met-allization quality can vary. In addition, along with the suraceo the structures, the substrate is also activated; the metalliza-tion is thereore not selective, oten requiring an extra step toremove the structures rom the metalized substrate.
This isaddressed by an alternative approach, which employs a photo-polymer doped with the metal binding sites. In this case, themetallization is selective, the density and distribution o thebinding sites can be controlled, and the resolution, structuralintegrity, and metallization quality limitations derive only romthe photopolymer used.
To date, the DLW approach with sub-sequent metallization has been applied up to now with greatsuccess in the realization o inrared metamaterials.
Here, we report the nanoabrication o 3D metallic woodpilestructures with eatures below 100 nm, obtained by DLW andselective silver coating. The structures, which have cc geom-etry
and in-layer periodicity 600 nm, are abricated usingan organic inorganic, zirconium-silicon hybrid material dopedwith a metal-binding monomer, 2-(dimethylamino) ethyl meth-acrylate. They are subsequently metalized using electrolessplating. The metalized structures are shown to exhibit ohmicconductivity, comparable to pure silver. We also present hereexperimental and theoretical electromagnetic characterizationresults o the abricated structures, showing a cut-o at around300 THz (1
m) and a second band gap inside the optical parto the spectrum, centered at 450 THz. To the best o our knowl-edge this is the rst time that metalized structures with suchresolution, ohmic response, and band gaps in the visible areabricated and characterized.In what ollows we discuss rstly the structure abrica-tion and metallization steps, we continue with the resistancemeasurements, which demonstrate the ohmic response o thestructure, and with the structural characterization, which dem-onstrates the long range-order characterizing our structuresand provides the geometrical eatures used in the simulations.Finally we discuss the experimental and theoretical electromag-netic characterization o the structures, through transmissionmeasurements and simulations, which reveal a bandgap in thevisible range.The photosensitive material used or the abrication o thethree dimensional photonic nanostructures is a zirconium-silicon organic-inorganic material doped with tertiary aminemetal-binding moieties.
Hybrid (organic-inorganic) mate-rials are a very popular class o photosensitive materials, asthey are easy to prepare, modiy and photopolymerize and,ater polymerization, they are optically, mechanically andchemically stable. As a result, the have ound many applica-tions in 3D photonic and biomedical devices.
In addition,hybrid materials chemistry provides the possibility o the inclu-sion o unctional groups, such as nonlinear optical moleculesand quantum dots.
Building on our previous work on thesynthesis and DLW processing o a silicon-zirconium hybridmaterial,
we have added methacrylate moieties by addinga 2-(dimethylamino) ethyl methacrylate (DMAEMA) monomer,capable o participating in the photopolymerization step toincorporate covalently bound metal-binding groups in the 3Dstructures.
The main materials used here or the synthesis o photopolymer were methacryloxy-propyltrimethoxysilane(MAPTMS, 99%), dimethyl- aminoethyl methacrylate (DMAEMA,
99%) and zirconium
-propoxide (ZPO) 70% solution in 1-propanol. The molar ratios were 7:3 or MAPTMS/ZPO and 3:7or DMAEMA/MAPTMS. MAPTMS was rstly hydrolyzed byadding HCl (0.1 M) at a 1:0.1 ratio and the mixture was stirred or20 minutes. Next, ZPO was mixed with DMAEMA and the mix-ture was stirred or 15 min. The MAPTMS sol was gently addedto the stirred ZPO sol and the mixture was stirred or 30 min.Finally, the photoinitiator 4,4-bis(diethylamino) benzophenonewas added at a 1% w/w concentration to the nal solution. Thesolution was ltered using 0.2
m syringe lters. Beore struc-turing, a small droplet o the composite solution was placedon a on a 100 micron thick glass slide and heated at 50 ºC or30 min. Ater DLW processing, the sample was developed and thematerial not exposed to the laser radiation removed by immer-sion in a 7:3 isopropanol/1-propanol solution. The metallizationprocedure has been described in detail previously.
For the 3D structuring a Ti:Sapphire emtosecond laser(800 nm, 75 MHz,
20 s) was ocused into the photopoly-merisable composite using a high numerical aperture ocusingmicroscope objective lens (100
1.4, Zeiss, PlanApochromat). Sample movement was achieved using piezoelec-tric and linear stages, or accurate and step movement, respec-tively (PI). The whole DLW setup, which is described in detailin Re. , was computer-controlled using the 3DPoli sotware.Here, the average laser power used or the abrication o thehigh-resolution structures was 18 mW, measured beore theobjective, while the average transmission to the sample was20%. To avoid contact with the lens immersion oil, all struc-tures were abricated upside down with the glass substratein contact with the oil. They were built layer-by-layer startingrom the top with the last layer adhering to the substrate. Thisway, the laser beam did not cross an already polymerized layer,causing second polymerization or beam distortion.In order to measure the conductivity o the metalized struc-tures, solid bulk dielectric boxes were created and metalized viaelectroless plating, keeping the same conditions as in the wood-pile structures, and the resistance at the aces o those boxes wasmeasured. The equipment used or the measurement was a Tek-tronix 370 curve tracer, which monitors the current,
, or eachspecic voltage,
, applied between the two metallic needles o thetracer which are placed on the samples. Modiying the voltage weobserved the linear
response o the structures, and thus theirOhmic behavior, with average resistance R
. Cal-culation o the resistivity,
, taking into account the needle sizeand the thickness o the metal gave
·m,and, thus, conductivity
. Thisresult is close to the conductivity o pure silver, which is6.3·10
(DC value). The thickness o the metal wasestimated rom SEM images o structures beore and atermetallization and was ound to be in the range
1 shows the Scanning Electron Microscope (SEM)pictures o some typical metalized woodpile structures with twodierent periodicities, showing the fexibility and the high reso-lution that can be achieved using this abrication technology. Inorder to achieve maximum resolution, the lowest intensity thatwas enough to abricate ully 3D structures was used.