One-step, subwavelength patterning of plasmonic gratings in metal–polymer composites

Abstract

2D and 3D micro/nano fabrication based on two-photon polymerization (TPP) has emerged as a strong contender for additive manufacturing. Here we present direct patterning of gold nanostructures using a femto-second laser inside a commercially available photoresist matrix without incorporating any dye. The nanostructures are written directly by in situ reduction of gold precursor within the photoresist using femto-second laser irradiation. The photo-initiator triggers the reduction of the gold precursor and induces simultaneous polymerization of the photoresist based on two-photon absorption phenomenon. Diffraction gratings with varied loadings of gold precursor in the photoresist have been fabricated and their diffraction efficiencies have been measured in the infrared region. Further 2D and 3D gold loaded mesh structures with subwavelength linewidths have been fabricated with sturdy features which may find metamaterial and plasmonic applications. The effect of gold precursor loading & laser power on line widths have been studied. A minimum line width of 390 nm has been achieved for 5 wt% gold precursor loaded polymers enabling the fabrication of sub-wavelength 2D and 3D plasmonic nanostructures.

---

Introduction

The fabrication of two-dimensional (2D) and three-dimensional (3D) metallic micro/nano structures is a daunting task and has recently being explored for application in fields like MEMS/NEMS, nanophotonics, optoelectronics, lab on chip devices and biomedical applications to name a few.^1–9 1D and 2D subwavelength period diffraction metallic gratings have numerous applications such as pulse compression,¹⁰ enhanced solar cell efficiency,¹¹ plasmonic resonance¹² and polarizers¹³ etc. Different strategies implemented for the fabrication of such gratings are not trivial and require time consuming multistep processes. In this context two-photon lithography (TPL) is of particular interest and allows the direct fabrication of sub-wavelength 2D/3D structures near the polymerization threshold based on photo-polymerizable material.^14–20 Since TPL is restricted to polymeric materials,²¹ one of the current challenge is the direct writing of functional structures inside metal–polymer composite. Metals do not form self-supporting micro/nano structures due to their weak mechanical strength,²² making it difficult to pattern. This acts as a bottleneck in the verification of theoretical concepts in the research area of plasmonics and metamaterials.^23–28 Metal loaded polymer composite structures have been reported in PVA & SU8 matrix previously.^29,30 However PVA–gold composite structures are huge, where as SU8 tends to reduce the gold precursor itself. Thus there is a need to find a suitable material combination in which the gold precursor is stable and large processing window is available for high resolution patterning.

Traditionally TPL utilizes two-photon absorbing dyes for fabricating sub-wavelength structures. However, these dyes typically used to achieve high resolution during lithography are expensive and limits its application.^4,31 Although previous attempts have been made towards dyeless TPL, they have mostly resulted in fabrication of structures limited to a few microns. These low resolutions are a result of using large pulse width of laser, which is constrained by the relatively small two photon absorption cross section of the photoinitiators employed.^15,32–34 Thus the requirement of high TPA coefficient is essential for the fabrication of the sub-wavelength structures.³⁵ Despite its small TPA coefficient, Lucirin TPOL is expected to be a better photoinitiator for TPL because of its high radical quantum yield of 0.99 and better solubility in most of the resins.³⁶ Lucirin TPOL with small loading concentrations (∼3 weight%) has been utilized for the fabrication of larger polymeric micro-cantilevers, optically active microstructures, structures doped with gold nanoparticles, branched hollow fiber microstructures with predefined circular pores.^{32,33,37–39}

Here we report the fabrication of metallic sub-wavelength structures in polymer matrix using 20 wt% Lucirin TPOL loading. We have systematically investigated the effect of laser power, metal precursor loading etc. on the line-widths and fabricated optical grating with variable periodicities. By employing laser parameters such as exposure time, laser power, pulse width and scanning speed, fabrication of gold precursor loaded nanostructures in polymer matrix can be fabricated for photonic & optoelectronics devices.⁴⁰ This advancement is expected to lead to fabrication of complex 2D & 3D plasmonic materials at significantly low cost and with high-throughput.

Experimental procedure

Gold precursor loaded two liquid resins; triacrylicmonomers, tris(2-hydroxy ethyl) isocyanurate triacrylate (SR368) and ethoxylated (6) trimethylolpropane triacrylate (SR499) [Sartomer] were subjected to femtosecond laser irradiation. SR499 reduces structural shrinkage, whereas SR368 confers hardness to the structure during the laser irradiation. The samples were prepared by mixing SR368 and SR499 in the ratio of 48 : 49 by weight percent (wt%). 20 wt% of a photoinitiator ethyl-2,4,6-trimethylbenzoylphenylphosphinate (Lucirin-TPOL from BASF) was added to this mixture. Different wt% of gold(III) chloride trihydrate (HAuCl₄·3H₂O) powder [Alfa Aesar] was loaded in the above solution. Finally a ∼ 3 μm thick film of the final solution was spin coated uniformly on a cover glass. Prior to spin coating, the cover glass was silanized with 3-aminopropyl triethoxysilane to facilitate the adhesion of the mixed resin. Acrylate-coated cover glass was positioned on a XYZ piezo-stage (PI-nanopositioner E-725) of an inverted optical microscope [Olympus-IX81]. The samples were then irradiated with Ti–sapphire laser source (Coherent made, Chameleon Ultra I) at 800 nm wavelength, 140 fs pulse width and a repetition rate of 80 MHz laser pulses via acousto-optic modulator as shown in Fig. S1 (ESI†). A 100× objective lens with numerical aperture of 0.9 was used to focus the beam inside the resin. An EMCCD attached to microscope facilitates the real time monitoring of the fabrication process. The experimental setup was controlled using LabVIEW and various scan rates were employed to fabricate the gratings. The unexposed resin was washed off in dimethylformamide (DMF) after writing. Since the temperature rise in the focal volume is sufficient enough for annealing,⁴¹ no post baking of the sample was performed.

Results and discussion

Several samples of gold–polymer composite lines with 1 wt%, 5 wt% and 10 wt% gold precursor loading were fabricated. Structural analysis was performed by measuring the line width of the grating in the micrographs obtained from scanning electron microscopy (SEM).

Functional grating structures with 1.5 μm, 2.5 μm and 5 μm periodicity were fabricated inside pure polymeric resin as well as in 1 wt%, 5 wt% and 10 wt% gold precursor loaded polymeric matrix. Large area SEM micrograph (560 μm × 280 μm) of the gratings fabricated inside 5 wt% gold loaded polymer are shown in Fig. 1(a). The periodicities of the gratings are 1.5 μm, 2.5 μm and 5 μm from bottom to top respectively. Fig. 1(b) is the zoomed in SEM micrograph of the 1.5 μm period grating showing finely spaced lines. The colour images in Fig. S2† are the optical images of the large area (840 μar× 840 μm) gratings captured in reflection mode using smartphone at different angles, fabricated in 5 wt% gold precursor loaded polymer. Fig. 1(c) is the EDAX spectrum of the 5 wt% gold precursor loaded structure. EDAX for the gold nanoparticle shows strong signal at around 2.30 keV which is in good agreement with the standard data. The presence of the silicon, sodium and potassium peaks in EDAX is from the glass cover slip. Further the presence of gold nanoparticles has been confirmed from the backscattered electron (BSE) image (Fig. 1(d)). One phase in the BSE image corresponds to the polymer and other to the gold nanoparticles. Few large size gold nanoparticles embedded in polymer have been marked by yellow dotted circle in inset of Fig. 1(d) results from the diffusion of gold ions from the surrounding volume to the fabricated structure. Gold ions present at the interface keeps reacting with available Au⁰ and grow making larger nanoparticles which is not possible inside the structure. The full scale backscattered electron images are shown in Fig. S3.†

Fig. 1 (a) Large area (560 μm × 280 μm) SEM image of a grating structure fabricated in 5 wt% gold precursor loaded polymer matrix for grating period of 1.5 μm, 2.5 μm and 5 μm for bottom to top gratings respectively. (b) Zoomed in image of 5 wt% gold precursor loaded grating showing the sturdy structure (c) is the EDAX spectrum showing the presence of gold in the fabricated nanostructure and (d) is the backscattered electron image showing the presence of gold nanoparticles. Inset in (d) is the secondary electron image of the lines with a chunk of gold nanoparticle (marked with yellow circles).

Diffraction patterns from the grating with periodicity 1.5 μm, 2.5 μm and 5 μm fabricated inside 10 wt% gold precursor loaded polymer matrix (SEM images in Fig. 2(a)–(c)) has been recorded using 712 nm laser at normal incidence and are shown in insets respectively. Well-resolved diffraction patterns indicates the high quality of grating structures. Decreasing of line spacing shows an increase in the diffraction angle, hence following the Bragg equation of diffraction. The diffraction patterns obtained from a grating and 2D mesh structure are shown in Fig. S4.†

Fig. 2 Diffraction pattern of red laser (712 nm) from gratings with (a) 1.5 μm, (b) 2.5 μm and (c) 5 μm periodicity fabricated inside 5 wt% gold precursor loaded polymer.

The diffraction efficiency of the grating structures fabricated in pure polymer, 0 wt%, 1 wt%, 5 wt% and 10 wt% gold precursor loaded polymer matrix with constant periodicity of 1.5 μm has been compared in Fig. 3. Tunable laser beam in the wavelength range of 800–1000 nm and constant laser power of 400 mW with fixed distance have been employed for the measurements.

Fig. 3 Measured total diffraction efficiency of the grating structures fabricated in 0 wt%, 1 wt%, 5 wt% and 10 wt% gold precursor loaded polymer matrix. Right side images are the SEM images of respective grating with periodicity of 1.5 μm.

The total diffraction efficiency of gold precursor loaded grating structures has been found to be more than that of pure polymer and is maximum for 10 wt% gold loading. Diffraction efficiency of first order (+1) has also been recorded and shown in the S5.† It can be inferred from Fig. S5† that nearly 40% of the input power is diffracted in the first order diffraction spot for normal incidence. With increase in the gold concentration the line width decreases and the spacing between the two consecutive lines increases (SEM images in the left of Fig. 3) causing more energy distribution in higher order diffraction modes. This phenomenon can be explained on the basis of N slit experiment. A set of N parallel slits illuminated by a monochromatic wave show that the intensity of the light getting passed through the slits will depend upon the angle θ between the direction of the light propagation and the perpendicular to screen as;⁴²

where β = (π/λ)b sin θ, α = (π/λ)d sin θ. I₀ is the intensity of the light at the centre of the diffraction for a single slit of width b, d is the distance between the slits, k = 2π/λ is the wave vector and λ is the wavelength. The first term of the equation in the square brackets describes the Fraunhofer diffraction on single slit and the second term describes the interference from N point sources. The intensity of the light in the centre of diffraction pattern will be proportional to b² causing the reduction in the intensity of 0^th order diffraction pattern. Smaller line-width of the doped polymer grating, where the spacing between the lines is more than undoped polymer, as well as plasmonic inclusion are responsible for enhanced efficiencies of doped grating structures.

The SEM images of large area grating structures with varied spacing have also been fabricated in 0 wt%, 1 wt%, 5 wt% and 10 wt% gold precursor loaded sartomer and are shown in Fig. S5–S8 (ESI†) respectively. Further to show the capability of writing functional 2D and 3D plasmonic devices we have successfully fabricated mesh structures with varied gold precursor loading. We have fabricated 2D mesh structures in 0 wt%, 1 wt%, 5 wt% and 10 wt% gold precursor loaded sartomer (Fig. S9†). Fig. 4(a) is the SEM image of the 2D mesh structure written inside 10 wt% gold precursor loaded polymer. The diffraction pattern of red laser (712 nm wavelength) for this 2D mesh structure showing bright and intense maxima can be seen in the inset of Fig. 4(a). A 3D mesh structure having 3 layers fabricated inside 1 wt% gold precursor loaded polymer is shown in Fig. 4(b). The SEM images of the 3 layer mesh structure at different magnifications are shown in Fig. S10.†

Fig. 4 (a) SEM image of the 2D mesh structure written inside 10 wt% gold precursor loaded polymer. Inset in (a) is the diffraction pattern of red laser (712 nm) from the respective 2D mesh structure. (b) 3D (3 layer) mesh structure fabricated inside 1 wt% gold loaded polymer.

Further to investigate the influence of laser power on feature sizes of 0 wt%, 1 wt%, 5 wt% and 10 wt% gold precursor loaded polymeric lines, a systematic study was performed. Different line widths in the range of 390 nm to 750 nm were achieved by varying the concentration of gold precursor as shown in Fig. 5(a). SEM micrograph of 5% gold loaded polymeric structures in Fig. 5(b) shows feature size of ∼390 nm with gold nanoparticles embedded inside the polymer matrix. Gold nanoparticle loaded polymer lines can only be formed in the laser-exposed regions. The amount of free radicals generated by Lucirin TPOL is responsible for initiation of polymerization process of the sartomer as well as reduction of the gold precursor in the laser irradiated region.

Fig. 5 (a) Plot of laser power vs. line width for pure polymer and gold precursor loaded polymeric structures with different loading. (b) SEM image of 5 wt% gold precursor loaded polymeric lines showing sub-wavelength feature size of 390 nm.

The threshold value of the laser power for writing the structures within polymer with gold precursor up to 5 wt% was found to be roughly the same (∼18 mW) as that for pure polymeric photoresist. The line width was found to increase with increasing laser power and decrease with increase in gold concentration up to nearly 5 wt% loading at ∼18 mW. This decrease in line width with increasing gold concentration can be attributed to the coupling between free radical polymerization and gold reduction. It can be understood as two processes competing against each other giving rise to finer features. It was observed that for 10 wt% gold precursor loaded samples the threshold value of laser power for writing increased to 24 mW; additionally the feature size also increased as compared to 5 wt% gold loaded polymeric lines. A minimum feature size of ∼670 nm was obtained for 10 wt% gold loaded polymeric structures near threshold value, below which writing was not possible. This increase in feature size may be attributed to the increase in the laser threshold value. Initially there is a competition between oligomers and gold ions for free radicals, which leads to reduction in line width up to 5 wt% gold loading. It is well known that small gold nanoparticles are more mobile than the gold ions,²³ which gets evident by the increased line width for 10 wt% gold precursor loading.

Writing process starts with the absorption of 800 nm light by the photo-initiator (Lucirin-TPOL). Free radicals generated by TPA starts to reduce the gold precursor leading to the formation of gold nanoparticles in the polymeric matrix. Unlike SU8–gold precursor film, the sartomer–gold precursor film doesn't show any signs of degradation under normal room light or by just mixing. The gold films were found to be stable even after a week of storage time. The detailed mechanism of simultaneous photo-polymerization and gold reduction is presented below:⁴³

Two photon initiation of Lucirin-TPOL gives rise to two free radicals;

Free radical polymerization initiation of monomer 1:

Propagation of chain polymerization reaction with monomer 2:

Termination by free radical combination:

Reduction of gold from Au³⁺ to Au²⁺ by the free radical:

Fast disproportionation of Au²⁺:

2AuCl₃⁻ → AuCl₂⁻ → AuCl₄⁻

Disproportionation of Au¹⁺ and Au²⁺ to reduce Au³⁺ and Au⁰:

AuCl₂⁻ + AuCl₃⁻ → AuCl₄⁻ + Au⁰ + Cl⁻

Gold nanoparticle formation

nAu⁰ → Au_n

Fig. 6(a) (red curve) shows the variation of line width for various wt% gold loaded polymeric lines written at constant laser power. Blue curve measures the spacing between the two lines for various gold precursor loading, written at constant periodicity of 1.5 μm. Linespan i.e. maximum and minimum line widths obtained at a particular gold loading has also been plotted in Fig. 6(a). The writing threshold is higher for 10 wt% gold loading with less spanning of line width. It is likely that sharing of photoinitiator between gold and oligomer causes this effect. It is apparent that with the increase in the gold loading, more and more hungry gold ions participate in the photoreduction process leading to less spanning in the line width. The upper threshold value for writing the structures with 1%, 5% and 10% loading is ∼186 mW, beyond which the sample starts to ablate. Fig. 6(b) and (c) show the grating structures written inside pure resin and 10 wt% gold precursor doped resin, respectively. The periodicity of both these gratings in the figure was kept constant to show the reduction in line width with increase in the gold loading at constant laser power.

Fig. 6 (a) Measurement of line width for various wt% gold precursor loaded polymeric lines (red cure). Blue curve represents the increase in line spacing with increase in gold precursor loading measured at constant periodicity of 1.5 μm for the gratings. Line spanning at various gold concentrations has also been plotted (green curve). SEM micrographs of grating structures fabricated in (b) 0 wt% and (c) 10 wt% gold loaded polymer matrix respectively at constant laser power just above the threshold value. The grating period is 1.5 μm and constant for both the structures.

Conclusion

In conclusion, we have fabricated large area nanostructures in the form of diffraction gratings of gold precursor loaded polymer composite by using femto-second laser based two-photon lithography without the use of dyes at any stage of fabrication. Free radicals generated by Lucirin-TPOL are found to be responsible for this process. Two photon initiated photoreduction of gold precursor and simultaneous free radical polymerization of sartomer facilitates the fabrication of gold nanoparticle embedded 2D polymeric matrix. This phenomenon can be extended to the fabrication of 3D gold loaded polymeric structures. Features with line widths as small as 390 nm has been observed for 5 wt% gold doped resin. The impact of critical process parameters such as laser power, gold doping etc. on feature sizes has also been studied. This approach is expected to pave way for in situ fabrication of high gold loaded polymer matrix which holds potential application towards novel 3D metamaterials, plasmonics and other optical technologies.

Acknowledgements

Department of Science and Technology, Solar Energy Research Initiative (SERI), Government of India (sanction order no. DST/TM/SERI/2k10/12/(G)) and the Industrial Research and Consultancy Services, Indian Institute of Technology Bombay (grant no. 11IRCCSG025).

References

1. W. U. Huynh, Science, 2002, 295, 2425–2427.
2. C. B. Fox, J. Kim, E. B. Schlesinger, H. D. Chirra and T. A. Desai, Nano Lett., 2015, 15, 1540–1546.
3. D. R. Smith, J. B. Pendry and M. C. K. Wiltshire, Science, 2010, 305, 788–792.
4. S. Shukla, K. T. Kim, A. Baev, Y. K. Yoon, N. M. Litchinitser and P. N. Prasad, ACS Nano, 2010, 4, 2249–2255.
5. S. Balslev, a. M. Jorgensen, B. Bilenberg, K. B. Mogensen, D. Snakenborg, O. Geschke, J. P. Kutter and a. Kristensen, Lab Chip, 2006, 6, 213–217.
6. E. P. Furlani and A. Baev, Phys. Rev. E: Stat., Nonlinear, Soft Matter Phys., 2009, 79, 1–6.
7. H. Chandrahalim, Q. Chen, A. A. Said and X. Fan, Lab Chip, 2015, 15, 2335–2340.
8. D. Wu, L.-G. Niu, S.-Z. Wu, J. Xu, K. Midorikawa and K. Sugioka, Lab Chip, 2015, 15, 1515–1523.
9. T. Wienhold, S. Kraemmer, S. F. Wondimu, T. Siegle, U. Bog, U. Weinzierl, S. Schmidt, H. Becker, H. Kalt, T. Mappes, S. Koeber and C. Koos, Lab Chip, 2015, 15, 3800–3806.
10. D. Strickland and G. Mourou, Opt. Commun., 1985, 56, 219–221.
11. Y. M. Song, J. S. Yu and Y. T. Lee, Opt. Lett., 2010, 35, 276–278.
12. L. Chen, J. J. Wang, F. Walters, X. Deng, M. Buonanno, S. Tai and X. Liu, Appl. Phys. Lett., 2007, 90, 2005–2008.
13. T. Weber, T. Käsebier, E.-B. Kley and A. Tünnermann, Opt. Lett., 2011, 36, 445–447.
14. S. Maruo, O. Nakamura and S. Kawata, Opt. Lett., 1997, 22, 132–134.
15. S. Kawata, H. B. Sun, T. Tanaka and K. Takada, Nature, 2001, 412, 697–698.
16. K. S. Lee, R. H. Kim, D. Y. Yang and S. H. Park, Prog. Polym. Sci., 2008, 33, 631–681.
17. A. Biswas, I. S. Bayer, A. S. Biris, T. Wang, E. Dervishi and F. Faupel, Adv. Colloid Interface Sci., 2012, 170, 2–27.
18. C. N. LaFratta, J. T. Fourkas, T. Baldacchini and R. A. Farrer, Angew. Chem., Int. Ed., 2007, 46, 6238–6258.
19. L. L. Erskine, A. a. Heikal, S. M. Kuebler, M. Rumi, X. Wu, S. R. Marder and J. W. Perry, Solid State Phys., 1999, 398, 51–54.
20. K. D. Belfield, X. Ren, E. W. Van Stryland, D. J. Hagan, V. Dubikovsky, E. J. Miesak and R. V. June, J. Am. Chem. Soc., 2000, 122, 1217–1218.
21. G. Witzgall, R. Vrijen, E. Yablonovitch, V. Doan and B. J. Schwartz, Opt. Lett., 1998, 23, 1745–1747.
22. T. Baldacchini, A.-C. Pons, J. Pons, C. N. LaFratta, J. T. Fourkas, Y. Sun and M. J. Naughton, Opt. Express, 2005, 13, 1275.
23. S. Shukla, X. Vidal, E. P. Furlani, M. T. Swihart, K. T. Kim, Y. K. Yoon, A. Urbas and P. N. Prasad, ACS Nano, 2011, 5, 1947–1957.
24. S. Oldenburg, R. Averitt, S. Westcott and N. Halas, Chem. Phys. Lett., 1998, 288, 243–247.
25. M. S. Rill, C. Plet, M. Thiel, I. Staude, G. von Freymann, S. Linden and M. Wegener, Nat. Mater., 2008, 7, 543–546.
26. S. Shukla, R. Kumar, A. Baev, A. S. L. Gomes and P. N. Prasad, J. Phys. Chem. Lett., 2010, 1, 1437–1441.
27. S. Shukla, A. Baev, H. Jee, R. Hu, R. Burzynski, Y. K. Yoon and P. N. Prasad, ACS Appl. Mater. Interfaces, 2010, 2, 1242–1246.
28. J. Henzie, J. Lee, M. H. Lee, W. Hasan and T. W. Odom, Annu. Rev. Phys. Chem., 2009, 60, 147–165.
29. K. Kaneko, H. B. Sun, X. M. Duan and S. Kawata, Appl. Phys. Lett., 2003, 83, 1426–1428.
30. S. Shukla, E. P. Furlani, X. Vidal, M. T. Swihart and P. N. Prasad, Adv. Mater., 2010, 22, 3695–3699.
31. J. F. Xing, X. Z. Dong, W. Q. Chen, X. M. Duan, N. Takeyasu, T. Tanaka and S. Kawata, Appl. Phys. Lett., 2007, 90, 31–33.
32. P. Tayalia, C. R. Mendonca, T. Baldacchini, D. J. Mooney and E. Mazur, Adv. Mater., 2008, 20, 4494–4498.
33. V. Tribuzi, D. S. Corrêa, W. Avansi, C. Ribeiro, E. Longo and C. R. Mendonça, Opt. Express, 2012, 20, 21107–21113.
34. J.-F. Xing, M.-L. Zheng and X.-M. Duan, Chem. Soc. Rev., 2015, 44, 5031–5039.
35. X. Zhou, Y. Hou and J. Lin, AIP Adv., 2015, 5, 030701.
36. T. Baldacchini, C. N. LaFratta, R. A. Farrer, M. C. Teich, B. E. A. Saleh, M. J. Naughton and J. T. Fourkas, J. Appl. Phys., 2004, 95, 6072–6076.
37. Z. Bayindir, Y. Sun, M. J. Naughton, C. N. Lafratta, T. Baldacchini, J. T. Fourkas, J. Stewart, B. E. A. Saleh and M. C. Teich, Appl. Phys. Lett., 2005, 86, 1–3.
38. C. R. Mendonca, D. S. Correa, F. Marlow, T. Voss, P. Tayalia and E. Mazur, Appl. Phys. Lett., 2009, 95, 113309.
39. M. Bieda, F. Bouchard and A. F. Lasagni, J. Photochem. Photobiol., A, 2016, 319–320, 1–7.
40. J. Henzie, M. I. N. H. Lee and T. W. Odom, Nat. Nanotechnol., 2007, 2, 549–554.
41. K. K. Seet, S. Juodkazis, V. Jarutis and H. Misawa, Appl. Phys. Lett., 2006, 89, 10–13.
42. E. Hecht, Optics, Addison-Wesley, 2002.
43. R. E. Medsker, M. Chumacero, E. R. Santee, A. Sebenik and H. J. Harwood, Acta Chim. Slov., 1998, 45, 371–388.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra22604c

---