1. Introduction

The reflection and transmission of light at a solid surface can be controlled by a periodic structure formed on the surface with a period size less than the wavelength of the incident light. Changing these optical properties in a visible region, the surface is colorizing, the so-called structural coloration [1]. Since this color is not due to the optical absorption by the bonding structure of molecules, the discoloration does not occur by the destruction of the molecular structure, which is induced by, for example, an irradiation of ultraviolet light. This significant advantage has attracted much interest in the method to produce a surface nanostructure inducing vivid and bright coloration, suggesting potential applications such as color marking and painting of an automotive body [2].

So far, several methods have been proposed to produce the nanostructures on solid surfaces [3–16]. Using the chemical oxidation [3,4] or the deposition of colloidal nanoparticles [5,6], the periodic structure with a uniform size can be formed homogeneously in a large area. The surface is colorizing homogeneously, where the color depends on the thickness of the oxidation layer or the size of nanoparticles. To draw figures and/or characters with different colors, the periodic structure is formed by using electron-beam lithography and etching process [7], or ablation with focused ion beam [8]. However, these methods for the color marking need much time to produce the nanostructure on the surface in a large area, leading to the large costs.

Laser pulses can also produce the structure in air to be able to apply the color marking [9–16]. Recently, high power laser systems with a high repetition rate and high-speed beam scanning systems have been developed and commercially available. This situation has been expected to reduce the process time and costs. However, it is difficult for the laser pulses to form the nanostructure to make vivid and bright colors, because the spatial resolution is usually limited to the order of the laser wavelength by the diffraction limit of light. To avoid this theoretical restriction, it has been reported the metal nanoparticle formation for plasmonic absorption [9], the oxide film formation for optical interference [10], and the periodic nanostructure (PNS) formation for optical diffraction [11–16]. In these processes, the PNS formed with femtosecond (fs) laser pulses has been observed on various kinds of materials such as dielectrics [17–23], semiconductors [24–27], and metals [28–32], which size is much below than the laser wavelength. This surface phenomena, therefore, has been focused on the application to the laser color marking. However, because the PNS has a non-uniform period size, the surface has poor diffraction efficiency and scatters the incident light strongly, resulting in the dull and dark coloration.

Miyaji and Miyazaki have proposed the dominant physical mechanism of the PNS formation on solid surfaces irradiated with the fs laser pulses [22,23,26,33,34], and developed new methods to fabricate the homogeneous PNS with a uniform period, the so-called nanograting [32,35]. While tight-focused fs laser pulses can form the homogeneous PNS having few grooves in the focal spot of few µm in diameter by using a microscope objective lens [36,37], our methods can offer an unique advantage that the nanograting can be formed in the whole area of few 100 µm in diameter with a low-numerical-aperture lens. This paper reports the vivid and bright coloration of the stainless steel (SUS) surface with the nanograting formed by the fs laser ablation. In the experiment, we evaluated the surface color with a photograph taken as functions of an incident angle of white light and an observation angle. Calculation using a simple model of diffraction reproduced well the characteristic changes of the colors observed as functions of these angles. The experimental and theoretical results clearly show that the nanograting can spatially disperse well the incident light into individual wavelength owing to the diffraction with high efficiency, leading to the vivid and bright coloration.

2. Experimental

2.1 Nanograting formation with femtosecond laser pulses

As a target, we used a polished SUS 304 plate of 10-mm square and 2-mm thickness. The root-mean-square value of the surface roughness was measured to be ∼ 2 nm with a scanning probe microscope (SPM). The target was held on an automatically xy translation stage. In the preliminary experiment, we measured the single-shot ablation threshold as 600 ± 10 mJ/cm² for the SUS target.

The nanograting was formed on the SUS surface by using the two-step ablation process with the fs laser pulses [35]. We used linearly-polarized, 800-nm, 100-fs laser pulses from a Ti:sapphire laser system operated at a repetition rate of 10 Hz. The optical setup for the ablation process was almost the same as that in the previous experiment [35]. Briefly, the fs laser output was split into two beams (1 and 2) with a half mirror, and the focused beams were superimposed at the target surface. The horizontally-polarized beam 1 and 2 were normally and obliquely incident on the target, respectively. The incident angle θ of beam 2 was set to be θ = 59° to make an interference pattern with the fringe period Λ = λ/sinθ = 933 nm. In the optical line of the beam 2, we set a retroreflector on an x stage to align the delay time between the beam 1 and 2. The beam 1 (beam 2) was focused with a 250-mm (200-mm) focal-length lens to the focal spot of 120 µm (75 µm) in diameter at the 1/e² intensity. To form the distinct fringe pattern, the peak fluences F⁽¹⁾ and F⁽²⁾ of beams 1 and 2 were independently controlled with a pair of half-wave plate and polarizer, the center of the focal spots of beam 1 and 2 was superimposed, and the delay time was optimized. In the first step, the beam 1 and 2 were focused on the target to overlap in space and time, and the fringe pattern was formed on the surface through ablation. Irradiating the beam 1 and 2, the stage was moved in a 1 mm × 1 mm area at a constant velocity v₁. In the second step, the only beam 1 was irradiated on the surface with the fringe pattern. The stage was moved at a constant velocity v₂ to form the nanograting in the 1 mm × 1 mm area on the SUS surface.

We observed morphological change of the target surface as functions of F⁽¹⁾, F⁽²⁾, and v₁ in the first step, and F⁽¹⁾ and v₂ in the second step, using a scanning electron microscope (SEM) and the SPM. The spatial-frequency distribution of the structured surface was evaluated with the two-dimensional Fourier transform of the SEM image. The period d was determined from the distribution in the direction parallel to the polarization direction.

2.2 Evaluation of surface color

The color of the target surface was evaluated with a photograph taken with a digital camera. Figure 1(a) shows a schematic illustration of the optical configuration for the color reading procedure. We use a light-emitted diode (LED) as a white light source. The light diffused by a paper set in front of the LED is illuminated uniformly on the whole of the target surface located 250 mm away from the LED at an incident angle (or so-called illumination angle) of α = 0–45°. The photograph of the target surface is taken by means of the camera (Canon, PowerShot SX60HS) located 250 mm away from the target at an angle (or so-called observation angle) of β = 15–75°. To avoid the incidence of the reflected light at the target surface into the camera directly, we observe the surface in the same direction as the LED. The groove direction of the structured SUS target is set perpendicular to the incident (zx) plane of the light, and the camera is set in the zx plane. The spectrum of the light from the LED was measured with a spectrometer. As shown in Fig. 1(b), the light has a broad spectrum in a visible region of λ = 420–700 nm. To evaluate the color of the target surface, the white balance and the exposure of the camera were adjusted with a following procedure. Firstly, illuminating with the light from the LED, we took the photograph of a white card in the color card 158a of the practical color coordinate system determined by Japan Color Research Institute. The white balance of the camera was adjusted so that the red (R), green (G), and blue (B) values in the acquired image data had the same value. Next, we set the exposure, the shutter speed, and the ISO (International Organization for Standardization) sensitivity to be F8, 0.3, and 100, respectively, so as not to saturate the RGB values in the image data of the structured surface observed at all of α and β.

 

Fig. 1. (a) Schematic drawing of color reading procedure of the SUS surface with the periodic structure, where α, β, and d is the incident angle of the light, the observation angle of the camera, and the period of the structure, respectively. (b) The spectrum of the light from the light-emitted diode (LED). The color bar in (b) represents the visible spectrum.

Download Full Size | PDF

To evaluate the color brightness given human intuitively, or relative sense to light and dark, we used the HSV (hue, saturation, value) color space, which can be transformed from standard RGB color space [38]. The brightness V of the color can be expressed numerically as V (%) = Max (R, G, B) / 255 × 100, where the RGB values (each 0 to 255) were obtained from the image data of the structured surface, and Max (R, G, B) denotes a maximum among the R, G, and B values.

3. Results and discussion

3.1 Surface morphology of the SUS surface irradiated with fs laser pulses

In the preliminary experiment, we fabricated the PNS on the SUS surface to compare with the coloration of the SUS surface with the nanograting formed by the two-step ablation. The single fs laser pulses (beam 1) were irradiated onto the surface at the fluence F for the scanning speed v. Figure 2(a) shows the SEM and SPM images of the PNS formed on the SUS surface at F = 600 mJ/cm² for v = 40 µm/s, together with the spatial-frequency distribution transformed from the SEM image in the direction parallel to the polarization direction. The scanning direction was vertical. It clearly shows that line-like nanostructures extending in the direction perpendicular to the polarization direction are formed on the target and have a broad spatial-frequency distribution, corresponding to non-uniform periods in a range of d = 280–700 nm, which represents the peaks at d = 345 nm and d = 656 nm.

 

Fig. 2. SEM (upper) and SPM (middle) images of SUS surface irradiated with fs laser pulses, together with the spatial-frequency spectrum of the SEM image (bottom). (a) PNS at F = 600 mJ/cm² for v = 40 µm/s, (b) interference pattern at F⁽¹⁾ = 400 mJ/cm², F⁽²⁾ = 300 mJ/cm², and θ = 59° for v₁ = 400 µm/s, and (c) nanograting at F⁽¹⁾ = 350 mJ/cm² for v₂ = 17 µm/s. The number at each peak in the spectrum denotes the period d in nm of the structure. The scanning direction is vertical on the images.

Download Full Size | PDF

Using the two-step ablation process, the nanograting was formed on the SUS surface. In the first step, we produced the interference pattern on the surface irradiated with the fs pulses (beam 1 and 2) at F⁽¹⁾ = 400 mJ/cm² and F⁽²⁾ = 300 mJ/cm² for v₁ = 400 µm/s, as shown in Fig. 2(b). The fringe pattern represents the depth of ∼ 60 nm and the frequency peak at Λ = 937 nm with the peaks of its harmonic components. In the second step, the interference pattern was downsized with the single fs pulses (beam 1) at F⁽¹⁾ = 350 mJ/cm² for v₂ = 17 µm/s. The scanning direction in the two-step process was vertical. As shown in Fig. 2(c), the surface structure is a homogeneous nanograting with a uniform period of d = 467 nm ∼ Λ/2, which is also seen with the isolated peak in the spectrum.

3.2 Color of the SUS surface irradiated with fs laser pulses

We took the photographs of the SUS surfaces with the PNS and the nanograting as functions of the illumination angle α and the observation angle β every 5°. The results are shown in Fig. 3. The blank area represents no observation, because the camera could not be set owing to interfere with the LED. The colors of the SUS with the PNS include blue, green, brown, and purple, as seen in Fig. 3(a). On the other hand, those with the nanograting include purple, blue, yellow, green, and red, which have more variation than those with the PNS, as seen in Fig. 3(b). In addition, we confirmed that these colors did not change five months later. To discuss the origin of the coloration, we consider the optical diffraction at the periodically structured surface. The incident angle α of light and the diffraction angle, here the observation angle β, should satisfy the grating equation described as

 

Fig. 3. Photograph of the SUS surface with (a) PNS and (b) nanograting taken at α and β. The dotted, dashed, and solid lines denote A = sin α + sin β = 1.05, 1.20, and 1.35, respectively. The size of the color pallet is 1-mm square. The blank area denotes no observation.

Download Full Size | PDF

To see the color difference between the SUS surfaces with the PNS and the nanograting, we calculated λ_D for these periodic structures. The results are shown in Table 1, where we set m = 1 in Eq. (1). Since the PNS has a broad frequency spectrum at peaks of d = 345, 656 nm, the incident light with λ_D = 362, 689 nm and λ_D = 414, 787 nm are strongly diffracted at A = 1.05 and 1.20, respectively. In a visible region, the intense red and weak green, and blue, components diffracted at the surface are mixed to look brown. Because λ_D shifts to be larger in larger A, the mixture color looks to change from light brown to dark brown. At A = 1.35, λ_D is 467, 886 nm, resulting in the coloration in blue. For the nanograting with a uniform period of d = 467 nm, λ_D at A = 1.05, 1.20, and 1.35 are 490, 560, and 630 nm, respectively. These wavelengths of the diffracted light at the structured surface are in good agreement with the colors shown in Fig. 3. The results show that the first-order diffracted light at the surface should strongly determine the surface color, and that the homogeneous nanograting with a uniform period can clearly separate the incident light into individual wavelength to look vivid colors.

Table 1. Wavelength λ_D in nm (color) of the first-order diffracted light at the SUS surface with the PNS and the nanograting as functions of A and d.

View Table

Finally, to evaluate the brightness V of the color of the SUS surface, we calculated V from the acquired image data of the surface. The result is shown in Fig. 4. Comparing Fig. 4(a) with Fig. 3(a), the surface with the PNS has the high brightness of V > 80% in brown and green. As seen in Fig. 3(b) and 4(b), the surface with the nanograting has the high brightness of V > 80% in the colors continuously distributing from purple to red, and the area of V > 80% is larger than that with the PNS. These results indicate that the nanograting has higher diffraction efficiency in the visible region than the PNS.

 

Fig. 4. Brightness V of the color of the SUS surface with (a) the PNS and (b) the nanograting plotted as functions of α and β. The blank area denotes no observation.

Download Full Size | PDF

4. Conclusion

We have investigated the coloration of the SUS surface having the nanograting structure formed by the two-step ablation process with fs laser pulses. The results demonstrate that the surface has more color variation and brightness than that with a periodic nanostructure produced with the single-beam fs laser pulses, leading to a conclusion that the nanograting can spatially disperse well the incident light into individual wavelength owing to diffraction with high efficiency. Calculation using the diffraction equation reproduced the characteristic change of the color observed as functions of these angles.