• Home
  • Sitemap
  • Contact us
Article View

Brief Report

Applied Science and Convergence Technology 2024; 33(1): 32-35

Published online January 30, 2024


Copyright © The Korean Vacuum Society.

Characteristics of a Plasmonic Color Filter Using Simple Focused Ion Beam Process

Taehun Kima , Byungwoo Leeb , and Ha Sul Kimb , ∗

aNational Forensic Service, Wonju 26460, Republic of Korea
bDepartment of Physics, Chonnam National University, Kwangju 61748, Republic of Korea

Correspondence to:hydenkim@jnu.ac.kr

Received: November 27, 2023; Revised: December 19, 2023; Accepted: December 20, 2023

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc-nd/4.0/) which permits non-commercial use, distribution and reproduction in any medium without alteration, provided that the original work is properly cited.

This study is aimed to develop fabrication methods for a simple and inexpensive surface plasmonic device working as a color filter in the visible wavelength range. A two-dimensional hexagonal plasmonic filter with pitch size of 360 nm, hole size of 120 nm, and thickness of 125 nm was made by simple focused ion beam (FIB) process. We used a commercial light-emitting diode (LED) chip as a light source for surface plasmon excitation. The international commission on illumination (CIE) chromaticity coordinates of the LED light source were located at x = 0.30 and y = 0.34. After the incident LED light passed through the plasmonic color filter, the peak wavelength of the transmitted light was observed at around 530 nm. The CIE chromaticity coordinates of the transmitted light were calculated as x = 0.29 and y = 0.66. This study demonstrates a simple plasmonic device fabrication method using the FIB process. The results can be applicable for filtering for a specific wavelength range in the visible spectrum.

Keywords: Plasmon, Plasmonic color filter, International commission on illumination chromaticity

Surface plasmon (SP) has become a field of interest to many scientists working in biology, chemistry, and applied physics. Furthermore, development of semiconductor-processing technology also induced the manufacture of nanoscale plasmonic chip in industries. SP is a wave that propagates along the surface of a conductor. When light shines on a conductor, SPs are formed on the surface due to interaction between free electrons and incident light [13]. During such interaction, the free electrons react collectively by resonating and oscillating with the incident light. The resonant interaction between surface charge oscillation and electromagnetic field of the incident light wave is called surface plasmon resonance (SPR) [46].

SPR can be used in the excitation and perturbation of SPs to investigate biological events or improve device performance occurring on metal surfaces. For example, the absorption of solution phase molecules on the metal surface alters the plasmon vibration and the vibration resonance frequency. This frequency change is related to the refractive index of the adsorbent material. Therefore, SPR can be applied to study bioprotein adsorption and biomolecular interactions [7]. Another example is local SPR between focused electromagnetic waves and free electrons on a metal surface. When electrons on a metal surface are excited by an intensified light, the chemical activation energy of plasmon-driven reaction is reduced, resulting in a more effective chemical reaction. As a result, chemical catalytic reactions can be significantly promoted [8]. Yet another example, SP can also be applied to energy harvesting research. An enhancement in energy conversion efficiency and short-circuit current density in amorphous silicon p-i-n solar cells could be realized with around 8 % improvement due to forward scattering by SP in Au nanoparticles deposition on the device. This result was confirmed by finite-element electromagnetic simulation, which showed the expected increase in transmission of electromagnetic radiation at visible wavelengths [9].

To fabricate plasmonic device with hundred-nanometer-scale dimension, we have to use electron beam exposure or the semiconductor photolithography process [10,11]. Therefore, it is both inconvenient and complex to fabricate a SP device that has a large area due to multiple alignment or thin photoresist exposure issues, thereby leading to a high cost SP device. In the present study, we aimed to develop fabrication methods for a simple and inexpensive SP device working as a color filter in visible wavelength. Also, we want to show that a single commercial light-emitting diode (LED) light instead of expensive, bulky xenon light can be used as light source for plasmonic experiment. Therefore, research on the implementation of plasmonic color filters can be applied in various ways through a very simple method.

When electromagnetic waves are incident on a metal surface, interaction occurs between the free charges and electromagnetic field on the metal surface. Depending on the type and structure of the thin film metal used, a specific wavelength can be transmitted or reflected due to the surface plasmonic effects. In other words, it is possible to induce a change in the resonance conditions according to the specific incident direction and the geometrical structure of the thin metal film. Therefore, we can design a plasmonic color filter structure that can effectively transmit a specific wavelength of the incident electromagnetic wave according to the changes in resonance conditions. Applying the boundary conditions of electromagnetic waves to dielectric and hexagonal plasmonic devices, we can solve the Maxwell equation [12]. The maximum peak wavelength of transmitted light can be expressed as


Where ϵm is the dielectric constant of the silver metal, ϵd is the dielectric constant of the dielectric, P is the distance between the one hole and the next hole in the nanoscale hole array. i and j are the integers representing the scattering order in a hexagonal array. Therefore, it is possible to adjust the specific transmitted wavelength arbitrarily where the maximum transmittance occurs in proportion to the changes in the hole spacing of the metal surface.

The peak wavelength of the transmitted spectrum was predicted according to Eq. (1). Then, the transmitted spectral characteristics were estimated using a commercial finite-difference time-domain (FDTD) simulator (Lumerical’s multiphysics simulation suite). To begin with, we have set up the structure of the device. Then, we virtually deposited a gold film on top of the glass. Subsequently, we designed the hexagonal structure of an etched hole array in the Au film. During the simulation setup process, we determined that the necessary materials for the simulation were available in the material database. Aiming to obtain final high-accuracy results, we increased the mesh accuracy setting during the convergence testing stage. Next, we set up a planar wave light source at normal incidence in air above the device structure. The start and stop range of the source had been set to the range of interest which is from 400 to 700 nm. A virtual device was simulated by changing the kind of metal, size, and shape of the device for a plasmonic color filter. By performing these computer simulations, it was possible to predict any change in the electric field of the transmitted electromagnetic waves according to positions, shape, and size of the device.

A glass substrate with favorable transmittance in visible wavelength range was used as a basic substrate of the device. Thin gold, silver, and aluminum films are currently used in the manufacture of plasmonic devices. For aluminum and silver, aerial oxidation causes formation of aluminum oxide and silver oxide on the surface of the metal, respectively. Such oxidation causes less SP effect due to reduction of free charge on the surface. In addition, the refractive index of the metal surface may change to other values. This may lead to change in the maximum peak value of the transmitted wavelength.

Due to the above-mentioned issues with aluminum and silver, gold with high chemical stability was used as a plasmonic metal in the current work. However, when gold is deposited directly on a glass substrate, it does not adhere to the surface due to the inability to form an oxide-metal interface for adhesion. Therefore, a gold film easily peels off the glass substrate. To overcome this problematic phenomenon, a titanium film of thickness 3 nm was first deposited on the glass substrate because it can create an oxide interface easily with favorable bonding strength on the glass. Then, a gold film of thickness 125 nm was deposited on the titanium layer.

Figure 1(a) shows the results of computer simulation of the transmission spectrum. Pitch values used were from 360−380 nm and a diameter was set to 120 nm. As seen from Eq. (1), the maximum transmission wavelength shifted toward longer wavelengths as the pitch of the hole increased. Also, Fig. 1(a) shows the predicted transmission spectrum when the pitch was set to 360 nm and the hole diameter was increased from 80 to 120 nm. As the diameter of the hole increased, the overall transmittance also increased since the amount of light transmitted increased. The inset in Fig. 1(a) provides a threedimensional (3D) representation of the overall model of the plasmonic device. The gold film has a thickness of 125 nm, and the holes are structured in a hexagonal pattern. Figure 1(b) shows the fabrication process. First, the glass substrate was cleaned thoroughly by washing sequentially with acetone, methanol, and isopropyl alcohol to remove all contaminants. Next, a titanium layer of thickness 3 nm was deposited at a rate of 0.7 Å/sec on the glass substrate using an electron beam evaporator. To note, deposition at any higher rate than the above value increased the surface roughness which in turn led to lower transmittance. Therefore, a gold film of thickness 125 nm was also deposited at 0.7 Å/sec above the titanium layer.

Figure 1. The results of computer simulation represent the transmission spectrum using a commercial FDTD simulator. The top three graphs in (a) depict the transmission for hole diameter of 120 nm with pitch values of 360, 370, and 380 nm, respectively. The black, red, and blue graphs illustrate the predicted transmission spectra when the pitch is fixed at 360 nm, and the hole diameter is increased to 80, 100, and 120 nm, respectively in (a). The inset in (a) provides a 3D representation of the overall model of the plasmonic device. (b) Schematic diagram of the fabrication process.

For the FIB-milling process, nanoscale hole arrays were designed using commercial auto computer-aided design software. The diameter of hole was set to 120 nm and the pitch between the holes was designed to be 360 nm. Six holes were arranged to make up two-dimensional (2D) hexagonal lattice. During the FIB-milling process, the maximum area that can be exposed at one processing time was 10 × 10 µm2. For the convenience of measurement, FIB-milling was repeated until the size of the total pattern was about 100 × 100 µm2. Furthermore, the ion beam current was set to 30 pA. Figure 2 is a scanning electron microscope (SEM) image of the plasmonic color filter and the inset image is a magnified image of the plasmonic device surface.

Figure 2. This shows the full SEM image of the fabricated device by FIB process using an ion beam current of 30 pA. The inset image shows a magnified view of the plasmonic device.

Figure 3 is a schematic diagram showing the method of measuring the transmission spectrum of the manufactured plasmonic color filter. Commercial white LED light was lighted at the back of the plasmonic device. Next, the light transmitted through the periodic hole array pattern was measured using a high sensitivity mini-spectrometer (C10082CA, Hamamatsu) with high resolution of ~ 1 nm.

Figure 3. Schematic diagram showing the method of measuring the transmission spectrum of the manufactured plasmonic color filter.

Figure 4(a) shows the spectrum of input light from the white LED, with radiating range from 425 to 700 nm. This LED light consists of two major peaks. The strongest intensity was detected at 459 nm, showing a blue color. The second peak was observed at around 540 nm having a broad and asymmetric shape. Figure 4(b) represents the spectrum of transmitted light through a plasmonic color filter. The peak wavelength of the transmitted light was observed at around 530 nm with full width at half maximum of 57 nm. The expected peak wavelengths obtained from computer simulation was in the range 515−525 nm depending on the hole size. The difference in peak wavelengths between the experiment and the computer simulation was possibly due to uneven hole shape, hole size, and roughness of the thin gold film.

Figure 4. (a) Spectrum of a white LED light. The strongest intensity was detected at 459 nm, showing a blue color. (b) Spectrum of the transmitted light through a plasmonic color filter. The center wavelength of the transmitted light was observed at around 530 nm.

The strongest light at 459 nm was measured at a very low intensity due to the skin effect of thin gold film. We think that the thin gold film with thickness 125 nm did not block the transmission of strong LED light completely at 459 nm. However, it can be expected to improve the skin effect by depositing a thicker gold film at around 150 nm.

Figure 5 shows the chromaticity coordinates of the LED light in the international commission on illumination (CIE) 1931 chromaticity coordinates. The chromaticity coordinate values of the LED light were located at x = 0.30 and y = 0.34, which are close to ideal white light. On the contrary, the chromaticity coordinates of transmitted light through the plasmonic color filter moved to x = 0.29 and y = 0.66, representing a yellowish green color.

Figure 5. The chromaticity coordinate values of the LED light are located at x = 0.30 and y = 0.34. The chromaticity coordinates of transmitted light through the plasmonic optical filter moved to x = 0.29 and y = 0.66, respectively.

In conclusion, a plasmonic color filter using gold hole array was fabricated using FIB process. 2D hexagonal arrays showed that the pitch between one hole to the next hole was around 360 nm and the size of the hole was around 120 nm. The incident LED light consisted of two major peaks of which the stronger intensity one was at 459 nm, showing blue color, while the other one was observed around 540 nm with broad and asymmetric shape. Also, the chromaticity coordinate value of CIE 1931 was at x = 0.30 and y = 0.34, which are close to ideal white color. Under the surface normal incident light, the spectrum of the transmitted light through plasmonic color filter revealed that the peak wavelength of the transmitted light was around 530 nm, showing a minor wavelength shift of 20 nm to the longer wavelength compared to the FDTD simulation. Also, the chromaticity coordinate value of transmitted light through the plasmonic color filter moved to x = 0.29 and y = 0.66, representing a yellowish green color. These results indicated that the plasmonic optical filter can be easily implemented using a metal hole array made through the FIB process and a single LED chip. Therefore, we can achieve a more inexpensive plasmonic color filter and plasmonic light source using a small LED chip.

This study was financially supported by Chonnam National University (Grand number: 2022-0196) and Basic Science Research Program through the National Research Foundation of Korea (NRF) granted by the Korea government (2022R1F1A1072506).

  1. T. W. Ebbesen, H. J. Lezec, H. F. Ghaemi, and P. A. Wolff, Nature 391, 667 (1998).
  2. V. E. Ferry, M. A. Verschuuren, H. B. T. Li, E. Verhagen, R. J. Walters, R. E. I. Schropp, H. A. Atwater, and A. Polman, Opt. Express 18, 237 (2010).
    Pubmed CrossRef
  3. Y. Wang, E. W. Plummer, and K. Kempa, Adv. Phys. 60, 799 (2011).
  4. M. Nam, N.-K. Chung, S. Shim, J.-Y. Yun, J.-T. Kim, and S. G. Pyo, J. Korean Phys. Soc. 71, 299 (2017).
  5. A. Kim, C. Choi, Y. H. Yu, and W. K Jung, J. Korean Phys. Soc. 45, 67 (2004).
  6. H. H. Nguyen, J. Park, S. Kang, and M. Kim, Sensors 15, 10481 (2015).
    Pubmed KoreaMed CrossRef
  7. Y. Tang, X. Zeng, and J. Liang, J. Chem. Educ. 87, 742 (2010).
    Pubmed KoreaMed CrossRef
  8. J. Wang, W. Lin, E. Cao, X. Xu, W. Liang, and X. Zhang, Sensors 17, 2719 (2017).
    Pubmed KoreaMed CrossRef
  9. D. Derkacs, S. H. Lim, P. Matheu, W. Mara, and E. T. Yu, Appl. Phys. Lett. 89, 093103 (2006).
  10. L. Petti, R. Capasso, M. Rippa, M. Pannico, P. La Manna, G. Peluso, A. Calarco, E. Bobeico, and P. Musto, Vib. Spectrosc. 82, 22 (2016).
  11. S. Baek, G. Kang, C.-W. Lee, and K. Kim, Sci. Rep. 6, 30476 (2016).
    Pubmed KoreaMed CrossRef
  12. Y.-T. Chang, T.-H. Chuang, M.-W. Tsai, L.-C. Chen, and S.-C. Lee, J. Appl. Phys. 101, 054305 (2007).

Share this article on :

Stats or metrics