**3.3 Retrieval of nanophotonic code**

For a practical demonstration of near-mode retrieval, the pattern of induced optical nearfields generated by irradiating an embedded nanometric structure with light is defined as a *nanophotonic code*. An optical near-field is a non-propagating light field generated in a space extremely close to the surface of a nanometric structure. Because the light distribution depends on several parameters of the structure and the retrieval setup, various types of coding can be considered. Moreover, several novel features of nanophotonics, such as energy transfer (Ohtsu et al., 2008) and hierarchy (Naruse et al., 2005), may be exploited to achieve further functional improvements of nanophotonic codes.

As shown in Fig. 5, we created a sample device to experimentally demonstrate the retrieval of a nanophotonic code within a hologram. The entire device structure, whose size was 15 mm × 20 mm, was fabricated by EB lithography on a Si substrate, followed by sputtering a 50 nm-thick Au layer, as schematically shown in the cross-sectional profile in Fig. 5.

Our prototype device was essentially based on the design of Virtuagram®, developed by Dai Nippon Printing Co., Ltd., Japan, which is a high-definition computer-generated hologram composed of binary-level one-dimensional modulated gratings. As indicated in the lefthand side of Fig. 5, we could observe a three-dimensional image of the earth from the device. Within the device, we formed slightly modified square or rectangular nanometric structures embedded in the original hologram structure, so that near-mode information carried by the nanometric structures was accessible only via optical near-field interactions. The unit size of each embedded structure ranged from 40 nm to 160 nm. For comparison, they were embedded both outside and inside the grid structures, as shown in Fig. 6(a) and (b), respectively.

Note that the original hologram was composed of arrays of one-dimensional grid structures, extending along the vertical direction in Fig. 6 (a). To embed the nanometric structures, the grid structures were partially modified in order to implement the nanophotonic codes. Nevertheless, the grid structures remained topologically continuously connected along the vertical direction. On the other hand, the nanophotonic codes were always isolated from the continuous original grid structures. These geometrical characteristics produce interesting polarization dependence, which is discussed in detail in Sec. 3.3.1.

Fig. 5. Schematic diagram of the sample device for demonstration of nanophotonic hierarchical hologram with a nanophotonic code embedded within the embossed structure of Virtuagram®.

For a practical demonstration of near-mode retrieval, the pattern of induced optical nearfields generated by irradiating an embedded nanometric structure with light is defined as a *nanophotonic code*. An optical near-field is a non-propagating light field generated in a space extremely close to the surface of a nanometric structure. Because the light distribution depends on several parameters of the structure and the retrieval setup, various types of coding can be considered. Moreover, several novel features of nanophotonics, such as energy transfer (Ohtsu et al., 2008) and hierarchy (Naruse et al., 2005), may be exploited to

As shown in Fig. 5, we created a sample device to experimentally demonstrate the retrieval of a nanophotonic code within a hologram. The entire device structure, whose size was 15 mm × 20 mm, was fabricated by EB lithography on a Si substrate, followed by sputtering a 50 nm-thick Au layer, as schematically shown in the cross-sectional profile in

Our prototype device was essentially based on the design of Virtuagram®, developed by Dai Nippon Printing Co., Ltd., Japan, which is a high-definition computer-generated hologram composed of binary-level one-dimensional modulated gratings. As indicated in the lefthand side of Fig. 5, we could observe a three-dimensional image of the earth from the device. Within the device, we formed slightly modified square or rectangular nanometric structures embedded in the original hologram structure, so that near-mode information carried by the nanometric structures was accessible only via optical near-field interactions. The unit size of each embedded structure ranged from 40 nm to 160 nm. For comparison, they were embedded both outside and inside the grid structures, as shown in Fig. 6(a) and

Note that the original hologram was composed of arrays of one-dimensional grid structures, extending along the vertical direction in Fig. 6 (a). To embed the nanometric structures, the grid structures were partially modified in order to implement the nanophotonic codes. Nevertheless, the grid structures remained topologically continuously connected along the vertical direction. On the other hand, the nanophotonic codes were always isolated from the continuous original grid structures. These geometrical characteristics produce interesting

achieve further functional improvements of nanophotonic codes.

polarization dependence, which is discussed in detail in Sec. 3.3.1.

Fig. 5. Schematic diagram of the sample device for demonstration of nanophotonic

hierarchical hologram with a nanophotonic code embedded within the embossed structure

**3.3 Retrieval of nanophotonic code** 

Fig. 5.

(b), respectively.

of Virtuagram®.

Fig. 6. SEM images of embedded nanometric structures (a) outside and (b) inside the grid structures of the original hologram.

#### **3.3.1 Numerical evaluations**

Before conducting the experimental demonstration with fabricated samples, the electric fields at the surfaces of nanometric structures were numerically calculated by the finitedifference time-domain (FDTD) method. As shown in Figs. 7, two types of calculation models were created in order to examine polarization dependency in retrieving the nanophotonic code. The embedded nanometric structure was represented by a square aperture whose side length was 150 nm, shown near the centre of the model. On the other hand, in the model shown in Fig. 7(b), the pitch of the periodic one-dimensional wire-grid structure was 150 nm, and the depth was 100 nm, which models the typical structure of an embossed hologram, and an aperture of the same size as that in Fig. 7(a) was embedded. The material of structures was assumed to be Au, and the structures were assumed to be irradiated with polarized plane waves coming from far above the structures. The wavelength was set to 785 nm. Periodic-conditioned computational boundaries were located 1.5 μm away from the center of the square-shaped aperture. By comparing those two cases, we can predict the effect of the existence of grid structures serving as the environmental structures on nanophotonic code retrieval. Also, we chose the square-shaped structure, which is isotropic in both the *x* and *y* directions in order to clearly evaluate the effects of the environmental structures and ignore the polarization dependency originating in the structure of the nanometric aperture itself.

Nanophotonic Hierarchical Holograms:

and right bars.

Demonstration of Hierarchical Applications Based on Nanophotonics 349

In order to quantitatively investigate how the environmental grid structure affected the electric field in the vicinity of the nanometric structure and the influence of input light polarization, the average electric field intensity was evaluated in the area of the nanometric structure, denoted by code *I* . The average electric field intensity in the area including the surrounding areas is denoted by env *I* . More specifically, code *I* represents the average electric field intensity in the 0.6 μm × 0.6 μm area covering the nanophotonic code, as shown by the dotted square in Fig. 9(a), whereas env *I* indicates that in the 2.5 μm × 2.5 μm area indicated by the dashed square in Fig. 9(a). Figure 9(b) summarizes the calculated values of code *I* and env *I* with each model, respectively shown by the left

First, the polarization dependencies are investigated. As shown in Fig. 8, in the case of the nanometric structure embedded in the environmental grid structure, evident polarization dependency was observed for both code *I* and env *I* in Fig. 9(b), too. Figure 9(c) compares the ratio of code *I* with *x*-polarized input light to that with *y*-polarized input light for the embedded and isolated structures. As shown, code *I* with *x*-polarized input light was about two times larger than code *I* with *y*-polarized input light. On the other hand, the

Second, from the viewpoint of facilitating recognition of the nanophotonic code embedded in the hologram, it is important to obtain a kind of higher *recognizability* for the signals associated with the nanometric structures. In order to evaluate this recognizability, here we

> code num code env

which yields a higher value with higher contrast with respect to code *I* and env *I* (indicated by the term code env *I I* ) and with higher signal intensity (indicated by code *I* ). Figure 9(d) shows the calculated *R*num in the case of *y*-polarized light input to thetwo types of models. The result indicates that the nanometric structure embedded in the environmental grid structure is superior to that of the isolated structure in terms of the

We consider that such polarization dependency and enhanced recognizability are due to the environmental grid structure that extends along the vertical direction. The input light induces oscillating surface charge distributions due to coupling between the light and electrons in the metal. The *y*-polarized input light induces surface charges along the vertical grid; since the grid structure continuously exists along the *y*-direction, there is no chance for the charges to be concentrated. However, in the area of the embedded nanometric structure, there is a structural discontinuity in the grid; this results in higher charge concentration only at the edges of the embedded nanometric structure. On the other hand, the *x*-polarized input light sees structural discontinuity along the horizontal direction due to the vertical grid structure, as well as in the areas of the embedded structures. It turns out that charge concentration occurs not only in the edges of the embedded structures but also at other horizontal edges of the environmental grid structure. Therefore, by comparing this with the *y*-polarized model, enhancement of the electric field intensity and its polarization dependency is evident. In contrast to these nanometric structures embedded in holograms,

= × , (1)

*I R I I*

isolated nanometric structure did not show any polarization dependency.

define a numerical figure-of-merit *R*num as

recognizability defined by eq. (1), as shown in Figure 9.

Fig. 7. Calculation model of embedded nanometric structure (a) without and (b) with environmental grid structure.

Figure 8(a), (c) and (b), (d) show calculated electric field intensity distributions on the surfaces of the structures assuming *x*-polarized and *y*-polarized input light irradiation to each model, respectively. As shown, although the model *without* the grid structure did not reveal any polarization dependency, the model *with* the grid structure revealed evident polarization dependency for the *x*- and *y*-polarized irradiation. Moreover, enhanced electric field intensity was obtained with *x*-polarized irradiation to the model with the grid structure.

Fig. 8. Calculated intensity distribution of electric field produced by (a), (c) *x*-polarized light and (b), (d) *y*-polarized light input to the models without ((a), (b)) and with ((c), (d)) the grid structures.

Fig. 7. Calculation model of embedded nanometric structure (a) without and (b) with

was obtained with *x*-polarized irradiation to the model with the grid structure.

Figure 8(a), (c) and (b), (d) show calculated electric field intensity distributions on the surfaces of the structures assuming *x*-polarized and *y*-polarized input light irradiation to each model, respectively. As shown, although the model *without* the grid structure did not reveal any polarization dependency, the model *with* the grid structure revealed evident polarization dependency for the *x*- and *y*-polarized irradiation. Moreover, enhanced electric field intensity

Fig. 8. Calculated intensity distribution of electric field produced by (a), (c) *x*-polarized light and (b), (d) *y*-polarized light input to the models without ((a), (b)) and with ((c), (d)) the grid

environmental grid structure.

structures.

In order to quantitatively investigate how the environmental grid structure affected the electric field in the vicinity of the nanometric structure and the influence of input light polarization, the average electric field intensity was evaluated in the area of the nanometric structure, denoted by code *I* . The average electric field intensity in the area including the surrounding areas is denoted by env *I* . More specifically, code *I* represents the average electric field intensity in the 0.6 μm × 0.6 μm area covering the nanophotonic code, as shown by the dotted square in Fig. 9(a), whereas env *I* indicates that in the 2.5 μm × 2.5 μm area indicated by the dashed square in Fig. 9(a). Figure 9(b) summarizes the calculated values of code *I* and env *I* with each model, respectively shown by the left and right bars.

First, the polarization dependencies are investigated. As shown in Fig. 8, in the case of the nanometric structure embedded in the environmental grid structure, evident polarization dependency was observed for both code *I* and env *I* in Fig. 9(b), too. Figure 9(c) compares the ratio of code *I* with *x*-polarized input light to that with *y*-polarized input light for the embedded and isolated structures. As shown, code *I* with *x*-polarized input light was about two times larger than code *I* with *y*-polarized input light. On the other hand, the isolated nanometric structure did not show any polarization dependency.

Second, from the viewpoint of facilitating recognition of the nanophotonic code embedded in the hologram, it is important to obtain a kind of higher *recognizability* for the signals associated with the nanometric structures. In order to evaluate this recognizability, here we define a numerical figure-of-merit *R*num as

$$R\_{\rm num} = \frac{\{I\}\_{\rm code}}{\{I\}\_{\rm env}} \times \{I\}\_{\rm code} \tag{1}$$

which yields a higher value with higher contrast with respect to code *I* and env *I* (indicated by the term code env *I I* ) and with higher signal intensity (indicated by code *I* ). Figure 9(d) shows the calculated *R*num in the case of *y*-polarized light input to thetwo types of models. The result indicates that the nanometric structure embedded in the environmental grid structure is superior to that of the isolated structure in terms of the recognizability defined by eq. (1), as shown in Figure 9.

We consider that such polarization dependency and enhanced recognizability are due to the environmental grid structure that extends along the vertical direction. The input light induces oscillating surface charge distributions due to coupling between the light and electrons in the metal. The *y*-polarized input light induces surface charges along the vertical grid; since the grid structure continuously exists along the *y*-direction, there is no chance for the charges to be concentrated. However, in the area of the embedded nanometric structure, there is a structural discontinuity in the grid; this results in higher charge concentration only at the edges of the embedded nanometric structure. On the other hand, the *x*-polarized input light sees structural discontinuity along the horizontal direction due to the vertical grid structure, as well as in the areas of the embedded structures. It turns out that charge concentration occurs not only in the edges of the embedded structures but also at other horizontal edges of the environmental grid structure. Therefore, by comparing this with the *y*-polarized model, enhancement of the electric field intensity and its polarization dependency is evident. In contrast to these nanometric structures embedded in holograms,

Nanophotonic Hierarchical Holograms:

**3.3.2 Experimental demonstration** 

polarization.

Demonstration of Hierarchical Applications Based on Nanophotonics 351

In the experimental demonstration, the optical responses of sample devices, shown in Figs. 6, during near-mode observation were detected using a near-field optical microscope (NOM). A schematic diagram of the detecting setup is shown in Fig. 10. The NOM was operated in an illumination-collection mode with a near-field probe having a tip with a radius of curvature of 5 nm. The fiber probe was connected to a tuning fork. Its position was finely regulated by sensing a shear force with the tuning fork, which was fed back to a piezoelectric actuator of the probe stage. The observation distance between the tip of the probe and the sample device was set at less than 50 nm. The light source used was an LD with an operating wavelength of 785 nm, and scattered light was detected with a photomultiplier tube (PMT). A Glan-Thompson polarizer (extinction ratio 10-6) selected only linearly polarized light as the radiation source, and a half-wave plate (HWP) rotated the

Fig. 10. Schematic diagram of the experimental setup for retrieving a nanophotonic code.

for the isolated square apertures, both the *x*-polarized input light and the *y*-polarized input light have equal effects on the nanostructures.

These mechanisms indicate that such nanophotonic codes embedded in holograms could also exploit these polarization and structural dependences, not only for retrieving nearmode information via optical near-field interactions. For instance, we could facilitate nearmode information retrieval using suitable input light polarization and environmental structures.

Fig. 9. (a) Schematic diagram explaining definition of average electric field intensities code *I* and env *I* , and (b) their graphical representations in each calculation model. Evident polarization dependency was exhibited in the case of the nanometric code embedded in environmental structures. (c) The ratio of code *I* with *x*-polarized input light to that with *y*-polarized input light for the embedded and isolated structures. (d) Numerical visibility *R*num in two types of models with *y*-polarized input light. The result indicates that the visibility of the nanophotonic code was greatly enhanced by embedding it in the environmental structure.

### **3.3.2 Experimental demonstration**

350 Holograms – Recording Materials and Applications

for the isolated square apertures, both the *x*-polarized input light and the *y*-polarized input

These mechanisms indicate that such nanophotonic codes embedded in holograms could also exploit these polarization and structural dependences, not only for retrieving nearmode information via optical near-field interactions. For instance, we could facilitate nearmode information retrieval using suitable input light polarization and environmental

Fig. 9. (a) Schematic diagram explaining definition of average electric field intensities code *I* and env *I* , and (b) their graphical representations in each calculation model. Evident polarization dependency was exhibited in the case of the nanometric code

the visibility of the nanophotonic code was greatly enhanced by embedding it in the

embedded in environmental structures. (c) The ratio of code *I* with *x*-polarized input light to that with *y*-polarized input light for the embedded and isolated structures. (d) Numerical visibility *R*num in two types of models with *y*-polarized input light. The result indicates that

light have equal effects on the nanostructures.

structures.

environmental structure.

In the experimental demonstration, the optical responses of sample devices, shown in Figs. 6, during near-mode observation were detected using a near-field optical microscope (NOM). A schematic diagram of the detecting setup is shown in Fig. 10. The NOM was operated in an illumination-collection mode with a near-field probe having a tip with a radius of curvature of 5 nm. The fiber probe was connected to a tuning fork. Its position was finely regulated by sensing a shear force with the tuning fork, which was fed back to a piezoelectric actuator of the probe stage. The observation distance between the tip of the probe and the sample device was set at less than 50 nm. The light source used was an LD with an operating wavelength of 785 nm, and scattered light was detected with a photomultiplier tube (PMT). A Glan-Thompson polarizer (extinction ratio 10-6) selected only linearly polarized light as the radiation source, and a half-wave plate (HWP) rotated the polarization.

Fig. 10. Schematic diagram of the experimental setup for retrieving a nanophotonic code.

Nanophotonic Hierarchical Holograms:

recognizability *R*exp as

shown in Figure 11.

dependency were exhibited.

Demonstration of Hierarchical Applications Based on Nanophotonics 353

distribution. When a higher intensity is obtained selectively from the area of the nanophotonic code, the difference between *I x*( ) and ( ) env *I x* can be large. On the other hand, if the intensity distribution is uniform along the vertical direction, the difference between *I x*( ) and ( ) env *I x* should be small. Thus, the difference between *I x*( ) and ( ) env *I x* indicates the visibility of the nanophotonic code. We define an experimental

exp () () env *<sup>x</sup>*

Figures 12(b) shows *R*exp as a function of input light polarization based on the NOM results

Fig. 12. (a) Definition of *I x*( ) and ( ) env *I x* for numerical evaluation, and (b) calculated experimental recognizability, *R*exp , of embedded nanometric structure (red curve) and that of isolated nanometric structure (blue curve). Evident recognizability and polarization

*R Ix Ix* = − . (2)

Figures 11(a) and (b) show retrieved results of nanophotonic codes that were outside and inside the environmental grid structures of the hologram, respectively, using a linearly polarized radiation source rotated by 0 degree to 180 degree at 20-degree intervals. As is evident in Fig. 11(a), although small and noisy intensity distributions were obtained, clear polarization dependence was observed in Fig. 11(b); for example, from the area of the nanophotonic code located in the center, a high-contrast signal intensity distribution was obtained with polarizations around 80 degree.

Fig. 11. Observed NOM images of optical intensity distributions of retrieved nanophotonic codes (a) outside and (b) inside the environmental grid structure.

To quantitatively evaluate the characteristics of the embedded nanophotonic code, we investigated two kinds of intensity distribution profiles from the observed NOM images. One is a horizontal intensity profile along the dashed line in Fig. 12(a), which crosses the area of the nanophotonic code, denoted by *I x*( ) , where *x* represents the horizontal position. The other was also an intensity distribution as a function of horizontal position *x*; however, at every position *x*, we evaluated the average intensity along the vertical direction within a range of 2.5 μm, denoted by ( ) env *I x* , which indicates the environmental signal

Figures 11(a) and (b) show retrieved results of nanophotonic codes that were outside and inside the environmental grid structures of the hologram, respectively, using a linearly polarized radiation source rotated by 0 degree to 180 degree at 20-degree intervals. As is evident in Fig. 11(a), although small and noisy intensity distributions were obtained, clear polarization dependence was observed in Fig. 11(b); for example, from the area of the nanophotonic code located in the center, a high-contrast signal intensity distribution was

Fig. 11. Observed NOM images of optical intensity distributions of retrieved nanophotonic

To quantitatively evaluate the characteristics of the embedded nanophotonic code, we investigated two kinds of intensity distribution profiles from the observed NOM images. One is a horizontal intensity profile along the dashed line in Fig. 12(a), which crosses the area of the nanophotonic code, denoted by *I x*( ) , where *x* represents the horizontal position. The other was also an intensity distribution as a function of horizontal position *x*; however, at every position *x*, we evaluated the average intensity along the vertical direction within a range of 2.5 μm, denoted by ( ) env *I x* , which indicates the environmental signal

codes (a) outside and (b) inside the environmental grid structure.

obtained with polarizations around 80 degree.

distribution. When a higher intensity is obtained selectively from the area of the nanophotonic code, the difference between *I x*( ) and ( ) env *I x* can be large. On the other hand, if the intensity distribution is uniform along the vertical direction, the difference between *I x*( ) and ( ) env *I x* should be small. Thus, the difference between *I x*( ) and ( ) env *I x* indicates the visibility of the nanophotonic code. We define an experimental recognizability *R*exp as

$$R\_{\rm exp} = \sum\_{\mathbf{x}} \left| I(\mathbf{x}) - \left\langle I(\mathbf{x}) \right\rangle\_{\rm env} \right| \,. \tag{2}$$

Figures 12(b) shows *R*exp as a function of input light polarization based on the NOM results shown in Figure 11.

Fig. 12. (a) Definition of *I x*( ) and ( ) env *I x* for numerical evaluation, and (b) calculated experimental recognizability, *R*exp , of embedded nanometric structure (red curve) and that of isolated nanometric structure (blue curve). Evident recognizability and polarization dependency were exhibited.

Nanophotonic Hierarchical Holograms:

Science and Technology (MEXT), Japan.

**6. References** 

December

1992

1998

1998

2008

7505, 2010

9271, 2005

*Science*, Vol. 311, pp. 189-193, 2006

*Nanophotonics*, Vol. 1, 011597, 2007

Demonstration of Hierarchical Applications Based on Nanophotonics 355

Excellence (G-COE) Secure-Life Electronics project, and Special Coordination Funds for Promoting Science and Technology sponsored by the Ministry of Education, Culture, Sports,

Lorenzi, R. (2010). Do Mona Lisa's eyes hide a secret code?, *Discovery News*, 13

Javidi, B. & Horner, J. L. (1994). Optical pattern recognition for validation and security

Refregier, P. & Javidi, B. (1995). Optical image encryption based on input plane and Fourier

Rakuljic, G. A., Leyva, V., & Yariv, A. (1992). Optical data storage by using orthogonal

Van Renesse, R. L. (1998). *Optical document scanning*, Altech House Optoelectronics Library,

McGrew, S. P. (1990). Hologram counterfeiting: problems and solutions, *Proceedings of SPIE*, *Optical Security and Anticounterfeiting Systems*, Vol. 1210, pp. 66-76, 1990 Zhdanov, G. S., Libenson, M. N., & Martsinovskii, G. A. (1998). Optics in the diffraction

Ohtsu, M., Kobayashi, K., Kawazoe, T., Yatsui, T., & Naruse, M. (2008). *Principles of* 

Tate, N., Nomura, W., Yatsui, T., Naruse, M., & Ohtsu, M. (2008). Hierarchical hologram

Tate, N., Naruse, M., Yatsui, T., Kawazoe, T., Hoga, M., Ohyagi, Y., Fukuyama, T.,

Ozbay, E. (2006). Plasmonics: Merging Photonics and Electronics at Nanoscale Dimensions,

Nishida, T., Matsumoto, T., Akagi, F., Hieda, H., Kikitsu, A., & Naito, K. (2007). Hybrid

Naruse, M., Yatsui, T., Nomura, W., Hirose, N., & Ohtsu, M. (2005). Hierarchy in optical

Naruse, M., Yatsui, T., Kawazoe, T., Tate, N., Sugiyama, H., & Ohtsu, M. (2008).

Tate, N., Sugiyama, H., Naruse, M., Nomura, W., Yatsui, T., Kawazoe, T., & Ohtsu, M.

Nanostructures, *Applyed Physics Express*, Vol. 1, 112101, 2008

Engineered Nanostructures, *Optics Express*, Vol. 17, 11113, 2009

*Nanophotonics*, Taylor and Francis, Boca Raton, 2008

wavelength-multiplexed volume holograms, *Optics Letters*, Vol. 17, pp. 1471-1473,

limit: principles, results, and problems, *Physics-Uspekhi*, Vol. 41, pp. 719-722,

based on optical near- and far-field responses, *Optics Express*, Vol. 16, pp. 607-612,

Kitamura, M., & Ohtsu, M. (2010). Nanophotonic code embedded in embossed hologram for hierarchical information retrieval, *Optics Express*, Vol. 18, pp. 7497-

recording on bit-patterned media using a near-field optical head, *Journal of* 

near-fields and its application to memory retrieval, *Optics Express*, Vol. 13, pp. 9265-

Nanophotonic Matching by Optical Near-Fields between Shape-Engineered

(2009). Quadrupole-Dipole Transform based on Optical Near-Field Interactions in

verification, *Optical Engineering*, Vol. 33, pp. 1752-1756, 1994

plane random encoding, *Optics Letters*, Vol. 20, pp. 767-769, 1995

The nanophotonic code embedded in the hologram exhibited much greater polarization dependency, as indicated by *embedded* in Fig. 12(b), where the maximum *R*exp was obtained at 80-degree input polarization. On the other hand, only slight polarization dependency was observed with the isolated nanophotonic code, as indicated by *isolated*. These characteristics of retrieving the nanophotonic code in the environmental grid structure agree well with the numerical results in Fig. 9.
