Two-photon patterning of a polymer containing Y-shaped azochromophores

APPLIED PHYSICS LETTERS 94, 011115 共2009兲

Two-photon patterning of a polymer containing Y-shaped azochromophores A. Ambrosio,1,a兲 E. Orabona,1 P. Maddalena,1 A. Camposeo,2 M. Polo,2 A. A. R. Neves,2 D. Pisignano,2 A. Carella,3 F. Borbone,3 and A. Roviello3 1

Dipartimento di Scienze Fisiche and CNR-INFM CRS-COHERENTIA, Università degli Studi Federico II, Complesso Universitario di Monte Sant’Angelo, Via Cintia, I-80126 Napoli, Italy 2 CNR-INFM National Nanotechnology Laboratory, Distretto Tecnologico ISUFI, Università del Salento, via Arnesano, I-73100 Lecce, Italy 3 Dipartimento di Chimica, Università degli Studi Federico II, Complesso Universitario di Monte Sant’Angelo, Via Cintia, I-80126 Napoli, Italy

共Received 1 September 2008; accepted 8 December 2008; published online 9 January 2009兲 We report on the patterning of the free surface of azo-based polymer films by means of mass migration driven by one- or two-photon absorption. A symmetric donor-acceptor-donor structured Y-shaped azochromophore is specifically synthesized to enhance two-photon absorption in the polymer. The exposure of the polymer film to a focused laser beam results in light-driven mass migration for both one- and two-photon absorptions. Features with subdiffraction resolution 共250 nm兲 are realized and the patterning dynamics is investigated as a function of the light dose. Furthermore, functional photonic structures, such as diffraction gratings with periods ranging between 0.5 and 2.0 ␮m, have been realized. © 2009 American Institute of Physics. 关DOI: 10.1063/1.3058820兴 In recent years a growing attention has been devoted to the applications based on two-photon absorption 共TPA兲, sustained by the availability of high fluence pulsed laser sources. The quadratic dependence of TPA probability on the light fluence density guarantees a high spatial confinement 共of the order of few hundreds of nanometers兲 in the sample. This phenomenon is largely exploited to overcome the diffraction limit into both high resolution two-photon microscopy1 and lithography.2 In particular, two-photon lithography 共TPL兲 is exploited for high resolution microfabrication of photocurable resins or polymers, with resolution down to few tens of nanometers.3 Among all the materials employed in conventional photolithography, azopolymers have been shown to be suitable for TPL. These compounds contain azo-based moieties in the main polymer chain or as side chains. trans-cis-trans isomerization cycles of the azogroups are activated by UV or visible light and lead to the statistical orientation of the azobenzene molecules perpendicularly to the light polarization direction.4 In recent experiments azopolymers have shown TPA and have been used in lithography experiments aimed at holographic patterning by means of the photoinduced local refractive index change.5–8 Furthermore, trans-cis-trans isomerization cycles induced by photon absorption are known to generate a mass migration 共material displacement兲 on the free surface of the azopolymer films. This phenomenon occurs on azocontaining polymers when illuminated by a light pattern associated with light intensity gradients perpendicular to the light polarization.9 The mechanism of the surface modification following azogroup isomerization is still a matter of discussions. Experimental observations, mainly limited to one-photon absorption and concerning grating fabrication by exposure to laser interference patterns, have ruled out a thermally driven process or the occurrence of ablation in the irradiated regions,10 whereas only few works reported on the a兲

Electronic mail: [email protected].

0003-6951/2009/94共1兲/011115/3/$23.00

possibility of driving the mass migration by TPA,11 although the process is potentially leading to high resolution topographic pattern formation. In this letter, we demonstrate the possibility to drive mass migration by means of a TPA process on the free surface of a specifically synthesized Y-shaped azopolymer film. Furthermore, diffraction gratings with periods of 2 ␮m, 1 ␮m, and 500 nm are realized by TPL, thus accomplishing functional photonic devices by a process based on TPA of the employed polymer. The azopolymer used is a polyurethane obtained by a polycondensation reaction of a chromophore containing, in the main chain, a symmetric azoderivative of 2-共2,6-dimethyl-4H-pyran-4-ylidene兲malononitrile and 2,4tolylendiisocyanate 共PU-AS-Y-DCV兲 共inset in Fig. 1兲. The synthesized azochromophore, characterized by a symmetric donor-acceptor-donor ␲-conjugated structure, is expected to show a high TPA cross section. It is well known that symmetric group distribution into a molecule may favor TPA properties due to symmetric charge transfer from the end of the molecule to the middle.12 The details of synthesis procedure, chemicophysical properties and second order nonlinearity of poled films are described in a previous paper.13 Optical quality thin films, with a thickness of 550 nm, have

FIG. 1. Absorption spectrum of a thin film of PU-AS-Y-DCV spin coated on a glass substrate. The vertical dashed lines point out the wavelengths employed in the experiment, 488 and 800 nm. The inset reports the molecular structure of PU-AS-Y-DCV.

94, 011115-1

© 2009 American Institute of Physics

Downloaded 27 Mar 2009 to 212.189.140.68. Redistribution subject to AIP license or copyright; see http://apl.aip.org/apl/copyright.jsp

011115-2

Ambrosio et al.

FIG. 2. 共Color online兲 共a兲 Topographic AFM image and 共b兲 line profile of a topographical feature obtained in the one-photon process. 共c兲 Topographic AFM image and 共d兲 line profile of the feature obtained in the two-photon process.

been spin coated on glass substrates from pyridine solution 共polymer concentration 5 wt %兲. The typical absorption spectrum of the prepared film is shown in Fig. 1. For one-photon lithography we used a cw argon ion laser emitting at 488 nm wavelength, injected into a 1.3 numerical aperture 共NA兲 oil immersion objective, that focuses the laser beam into a diffraction limited spot on the sample surface. The latter is mounted on a piezoscanner that allows adjusting of the sample position with nominal resolution of the order of few nanometers. In our experiment the sample is first placed in the focal plane, then, the light is injected through the microscope objective. After several seconds of exposure, the laser beam is blocked and the sample is moved into another position on the focal plane. The exposure can be repeated on an area of a few tens of microns wide, depending on the piezoscanner employed. A similar setup is exploited for the two-photon exposures. The main difference being the laser source constituted by a near infrared femtosecond Ti:sapphire laser 共␭ = 800 nm, pulse width ⬍100 fs, and repetition rate of 76 MHz兲 coupled to the high NA 共1.3兲 oil immersion objective of an inverted microscope. Due to the design of our setup, the free surface of the polymer film is in contact with air, whereas the glass substrate is wet by the matching oil of the objective. The same configuration is used for exposures with low NA objectives 共NA= 0.45– 0.5兲, although in this case the matching oil is removed. Topographical features produced on the surface of exposed polymer films are characterized by means of an atomic force microscope 共AFM兲, operating in tapping mode, whereas a free software by Nanotec Electrónica S.L. 共Ref. 14兲 is used for the analysis of the AFM data. Figure 2 shows an example of the topographical features we realized. Figure 2共a兲 reports the topographical image of a structure obtained by irradiating the polymer surface for 60 s at 488 nm wavelength 共one-photon process兲. In this case the light power at the polymer-air interface is no higher than 130 ␮W and a diffraction pattern is obtained by illuminating the interface out of the focal plane of a 0.5 NA microscope objective. The illuminated area has a diameter of about 8.5 ␮m and an average light intensity less than 230 W / cm2 is used. Out-of-focus illumination, permitting to expose a larger sample area than in-focus illumination, turns out to be particularly useful in characterizing the mass-migration pro-

Appl. Phys. Lett. 94, 011115 共2009兲

cess. The resulting diffraction pattern consists of concentric rings, whose alternation of intensity maxima and minima constitutes the light intensity gradient that drives the mass migration, leading to the concentric shape of the polymer topographical features 共Fig. 2兲, which replicates that of the illuminating pattern. Figure 2共b兲 reports the profile of the structure along the direction indicated by the dashed gray line in Fig. 2共a兲. Evidence of the occurrence of mass migration is found, a process that removes material from the brighter intensity regions along the light gradient direction, in accordance with the observations reported in the literature when low optical densities are used.9,15 By comparing the areas of positive and negative parts in the profile of Fig. 2共b兲, the surface changes due to polymer displacement occur with overall volume conservation 共within 10%, i.e., comparable to the experimental error兲. This behavior is a fingerprint of mass migration that distinguishes this phenomenon from others such as ablation or thermally driven embossing.9,16 Figures 2共c兲 and 2共d兲 report topographical image and central profile of a pattern realized by TPA. In this case the polymer-air interface lies in an out-of-focus plane of a 0.45 NA microscope objective that focalizes the femtosecond laser beam at 800 nm wavelength 共estimated peak intensity of the order of few GW/ cm2 obtained by positioning the microscope objective focus about 2 ␮m below the sample surface, resulting into an exposed area diameter of about 15 ␮m兲. The writing wavelength is above the maximum absorbance region 共Fig. 1兲, whereas a TPA process can occur due to the strong absorption of the Y-shaped azopolymer at 400 nm. By referring to the profile reported in Fig. 2共d兲, apparently, for the two-photon structure, the volume conservation typical of mass migration cannot be observed. However a deeper analysis of this profile reveals that it results from the overlap of a hole-shaped baseline and a surface modulation. Indeed, by removing this baseline, a profile similar to that in Fig. 2共b兲, namely, with a balance between the positive and negative parts 共protruding and recessed polymer volumes兲, is found out. The observed volume decrease 共hole-shaped baseline兲 is attributed to structural modifications of the polymer chains, as reported in other works concerning azobenzene containing polymers11,17 and conjugated polymers.18 Another difference between one-photon and two-photon features here shown is that, while the one-photon concentric feature is mostly circular, the two-photon structure appears aligned along a preferential direction. We believe that this difference has its origins both in the optical aberrations occurring during focalization of the femtosecond near-infrared laser and in the influence of the light polarization on the polymer displacement.11 The latter, in particular, is still a matter of discussions even for the well assessed one-photon process. In our experimental work, we also investigated the features obtained by using the high NA objective, which allows the increase in lithography resolution and the investigation of the minimum attainable feature size. These experiments have been performed by using laser peak intensities in the range of 0.3– 3.5 GW/ cm2 in order to prevent other phenomena such as laser ablation, occurring at a peak intensity as high as 50 GW/ cm2. By this way we realized structures down to 250 nm wide employing the diffraction limited spot of the 800 nm wavelength pulsed laser, far below the halfwavelength diffraction limit of the focused laser beam.

Downloaded 27 Mar 2009 to 212.189.140.68. Redistribution subject to AIP license or copyright; see http://apl.aip.org/apl/copyright.jsp

011115-3

Appl. Phys. Lett. 94, 011115 共2009兲

Ambrosio et al.

FIG. 3. 共Color online兲 AFM image of a 10⫻ 10 ␮m2 area of a grating with 2 共a兲 and 1 ␮m 共b兲 periods, respectively. Inset: AFM image of a 500 nm period grating. Vertical scale: 15 nm. The gratings have been realized with a writing speed of 100 ␮m / s.

This method can be used to produce functional devices, namely one-dimensional diffraction gratings. The realized diffraction gratings have periods of 500 nm, 1 ␮m, and 2 ␮m and each pattern covers an area of 1.2⫻ 0.5 mm2. These gratings have been realized by moving the samples with a piezoscanner capable of 300 ␮m X-Y travel in combination with the microscope micrometric stage. Figure 3 reports the AFM images of the realized gratings. The diffraction efficiency 共first-over zero-order ratio兲 of these structures, measured by observing the diffraction of a collimated cw 800 nm wavelength laser beam passing through the gratings, is of the order of 0.03%, comparable to values reported for gratings with similar morphology, realized on azopolymers by interference holography.19 This result evidences that TPL can be used to fabricate functional devices based on azocontaining polymers. Finally, the dynamics of the features growth was investigated by writing several structures at different exposure times and peak intensities and measuring the resulting feature size by AFM scanning. By this way we found out that the trend of feature height versus TPA exposure time 共Fig. 4兲 is well described by a phenomenological model, already used for one-photon driven mass migration on azopolymers.16,20 This model is derived by assuming the mass migration as

FIG. 4. 共Color online兲 Features height vs two-photon exposure time for different values of the laser power injected into the high NA microscope objective. The exposure peak intensities are 3.5 共full circles兲, 1.7 共empty squares兲, 0.9 共full squares兲, and 0.3 GW/ cm2 共empty circles兲. The dashed line represents a fit of the experimental data based on a phenomenological model 共see text兲 for an exposure peak intensity of 3.5 GW/ cm2. Fits of experimental data obtained with lower peak intensities provide similar accordance.

consisting of a classical laminar flow described by Navier– Stokes equations. The rate of variation in the film thickness 共hF兲 predicted by the model is of the form ⳵hF / ⳵t ⬀ hF3 . A fit of our experimental data with the theoretical law is reported in Fig. 4 共black dashed line兲. The accordance with the experimental data constitutes further evidence of the analogy between one- and two-photon lithographies in our experiment. In conclusion, we demonstrated TPA induced mass migration in an azo-containing polymer, allowing to realize topographic features and diffraction gratings with a spatial resolution of few hundreds of nanometers, by low near-infrared laser intensities. This is possible due to the optimized polymer chemical structure. In view of possible applications of our results, the chemical structure of the employed azochromophore can be modified by adding, for instance, photoluminescent chromophores in order to synthesize a light emitting polymer that can be easily patterned with subwavelength resolution by TPL. This possibility opens the way to the exploitation of TPL for the realization of optoelectronic plastic devices. A. Ambrosio is grateful to the European Collaborative Project “S-five” for financial support. The NNL authors are grateful to the support of the Apulia Regional Strategic Project PS_144. D. Winfried, J. H. Strickler, and W. W. Webb, Science 298, 73 共1990兲. S. Kawata, H.-B. Sun, T. Tanaka, and K. Takada, Nature 共London兲 412, 697 共2001兲. 3 D. Tan, Y. Li, F. Qi, H. Yang, Q. Gong, X. Dong, and X. Duan, Appl. Phys. Lett. 90, 071106 共2007兲. 4 G. S. Kumar and D. C. Neckers, Chem. Rev. 共Washington, D.C.兲 89, 1915 共1989兲. 5 X. Xingsheng, M. Hai, W. Pei, L. Zhongcheng, and Z. Qijin, J. Opt. A, Pure Appl. Opt. 4, L5 共2002兲. 6 S. W. Magennis, F. S. Mackay, A. C. Jones, K. M. Tait, and P. J. Sadler, Chem. Mater. 17, 2059 共2005兲. 7 D. Gindre, A. Boeglin, A. Fort, L. Mager, and K. D. Dorkenoo, Opt. Express 14, 9896 共2006兲. 8 H. Ishitobi, Z. Sekkat, and S. Kawata, J. Chem. Phys. 125, 164718 共2006兲. 9 S. Bian, J. M. Williams, D. Y. Kim, L. Li, S. Balasubramanian, J. Kumar, and S. Triphathy, J. Appl. Phys. 86, 4498 共1999兲. 10 P. Rochon, E. Batalla, and A. Natansohn, Appl. Phys. Lett. 66, 136 共1995兲. 11 H. Ishitobi, S. Shoji, T. Hiramatsu, H.-B. Sun, Z. Sekkart, and S. Kawata, Opt. Express 16, 14106 共2008兲. 12 M. Albota, D. Beljonne, J.-L. Brédas, J. E. Ehrlich, J.-Y. Fu, A. A. Heikal, S. E. Hess, T. Kogej, M. D. Levin, S. R. Marder, D. McCord-Maughon, J. W. Perry, H. Röckel, M. Rumi, G. Subramaniam, W. W. Webb, X.-L. Wu, and C. Xu, Science 281, 1653 共1998兲. 13 A. Carella, F. Borbone, U. Caruso, R. Centore, A. Roviello, A. Barsella, and A. Quatela, Macromol. Chem. Phys. 208, 1900 共2007兲. 14 I. Horcas, R. Fernández, J. M. Gómez-Rodríguez, J. Colchero, J. GómezHerrero, and A. M. Baro, Rev. Sci. Instrum. 78, 013705 共2007兲. 15 S. Bian, L. Li, J. Kumar, D. Y. Kim, J. Williams, and S. K. Tripathy, Appl. Phys. Lett. 73, 1817 共1998兲. 16 K. Sumaru, T. Fukuda, T. Kimura, H. Matsuda, and T. Yamanaka, J. Appl. Phys. 91, 3421 共2002兲. 17 H. Ishitobi, Z. Sekkat, and S. Kawata, J. Opt. Soc. Am. B 23, 868 共2006兲. 18 R. Stabile, A. Camposeo, L. Persano, S. Tavazzi, R. Cingolani, and D. Pisignano, Appl. Phys. Lett. 91, 101110 共2007兲. 19 X. L. Jiang, L. Li, J. Kumar, D. Y. Kim, V. Shivshankar, and S. K. Tripathy, Appl. Phys. Lett. 68, 2618 共1996兲. 20 A. Ambrosio, A. Camposeo, P. Maddalena, S. Patanè, and M. Allegrini, J. Microsc. 229, 307 共2008兲. 1 2

Downloaded 27 Mar 2009 to 212.189.140.68. Redistribution subject to AIP license or copyright; see http://apl.aip.org/apl/copyright.jsp