High‐Brightness, Wide‐Gamut, and High‐Resolution Structural Colors via Ultrafast Laser Oxidation of Ti/TiO2 Films

TL;DR

A new laser method creates bright, vivid structural colors on titanium films with high resolution and durability for use in art and anti-counterfeiting.

Contribution

A scalable ultrafast laser technique for generating high-resolution structural colors using Ti–TiO2–Ti films with controllable oxidation.

Findings

Structural colors achieve over 80% sRGB gamut and 60% peak reflectance.

Resolution of up to 30,000 DPI is achieved with precise color difference control (ΔE < 3).

Laser-induced colors show excellent durability in thermal and corrosion tests.

Abstract

The scalable fabrication of high‐brightness, wide‐gamut, and high‐resolution structural colors remains challenging due to the reliance on complex nanofabrication or coating techniques. Here, we demonstrate a facile and flexible approach to generating vivid structural colors via ultrafast laser processing of Ti–TiO2–Ti sandwich‐structured thin films. The multilayer design enables controlled oxidation with oxide thicknesses exceeding 200 nm under laser irradiation. Optical characterization reveals dual absorption enhancement peaks and a reflection enhancement peak in the visible spectrum. By tuning the accumulated laser fluence, we modulate the oxidation state of Ti and the optical properties of TiO2, achieving vibrant structural colors with high reflectivity and a color gamut exceeding 80% sRGB. Peak reflectance reaches 60%, and the highest color difference (ΔE) control precision is less than 3. Moreover, a resolution up to 30 000 DPI is attained. The laser‐induced structural colors exhibit excellent durability, passing standard tests for thermal stability and corrosion. This method enables direct patterning of high‐luminance colors and holds promise for applications in art, anti‐counterfeiting, signage, and flexible photonic devices.

Ultrafast laser writing on titanium (Ti)–titanium dioxide (TiO2)–Ti sandwich films triggers controllable oxidation (〉200 nm) to create vivid, durable structural colors. The colors reach 〉 80% of the standard RGB (sRGB) gamut, up to 60% peak reflectance, and 30,000 dots per inch (DPI) resolution for art and anti‐counterfeiting.

Figures

Schematic and demonstration of laser‐induced coloration in a Ti‐TiO2‐Ti sandwich structure. (a) High‐brightness, full‐gamut structural colors are directly written using a picosecond laser. By introducing a thin TiO2 layer at an optimal position within the Ti film, the trilayer structure enables simultaneous enhancement of brightness and color gamut under identical laser processing conditions. (b) Correlation between laser parameters and the resulting TiO2 thickness. The multilayer structure (40 nm Ti/70 nm TiO2/120 nm Ti) was irradiated with laser pulses ranging from 6 to 10.4 µJ at a scanning speed of 16 mm/s, resulting in controllable oxidation of both the top and bottom Ti layers and corresponding shifts in the reflectance spectra. (c) Optical microscope images of representative laser‐printed structural colors. The laser fluence is indicated in the figure. Sample I was produced at a scanning speed of 60 mm/s, while samples II–IX were processed at 1 mm/s. A broad range of colors—such as yellow, orange, purple, red, blue, cyan, and green—was achieved.

Laser‐printed color images, reflectance, and gamut of laser‐modified Ti–TiO2–Ti films. (a,b) Laser‐printed structural colors by tuning single‐pulse energy and scanning speed, based on a single 200 nm Ti layer and Ti–TiO2–Ti sandwich structure with layer thicknesses of 40 nm (top Ti), 20–140 nm (TiO2), and 120 nm (bottom Ti). (c) Color gamut coverage in the CIE 1931 chromaticity diagram. The crosses indicate the chromaticity coordinates of the colors shown in (a,b). By simultaneously adjusting the laser energy and initial film structure, a color gamut covering approximately 80% of the sRGB space is achieved. (d) Reflectance spectra corresponding to the structure in (b), demonstrating fine spectral tuning via modulation of laser power density: by increasing single‐pulse energy (6.8–9.2 µJ) at 1 mm/s or adjusting scanning speed (5–19 mm/s) at 6.8 µJ. The reflectance spectra showed gradual shifts with a step size of approximately 10 nm. (e) Representative reflectance spectra after laser oxidation for samples with varying initial thin‐film structures, where the pre‐sputtered TiO2 middle layer ranges from 70 nm to 140 nm. All spectra exhibit peak reflectance exceeding 60%.

Cross‐sectional microstructure and compositional evolution under different laser irradiation doses. (a–c) FIB‐SEM cross‐sectional images of the Ti–TiO2–Ti multilayer before laser processing, after low fluence, and after high fluence. Morphological changes and oxide growth are observed with increasing laser energy; the scale bar is 200 nm. (d–f) XPS depth profiling results showing the elemental distribution of Ti, O, and Si as a function of depth for the initial, low fluence, and high fluence films.

High‐resolution laser‐printed structural colors and patterns. (a) Structural color dots fabricated using a high‐NA (0.8) objective lens, the thin film consists of 40 nm (top Ti)/140 nm (middle TiO2)/120 nm (bottom Ti). The single‐pulse energy was varied from 1.6 to 2.4 nJ, and the dwell time from 50 to 600 ms to tune the deposited fluence (i‐ii). SEM characterization of the blue color dot (iii), revealing diameters smaller than 800 nm. (b) Structural color lines (i) and the logo of Westlake University (ii). (c) Laser‐printed image of Westlake University Yungu Campus on a 2‐inch polished silicon wafer. (d) Laser‐printed image of Westlake Institute for Optoelectronics on a 2‐inch rough silicon wafer substrate.

Full text

Introduction

1

Structural colors, originating from the interaction between visible light and micro/nanostructures, differ from pigment‐based colors by relying on interference rather than absorption [1, 2, 3, 4, 5, 6, 7, 8, 9]. They are abundant in nature (e.g., butterfly wings, peacock feathers, seashells) and have inspired artificial platforms such as plasmonic nanostructures [10, 11, 12, 13, 14], Mie resonators [15], Fabry–Pérot cavities [16, 17, 18, 19], dielectric multilayers [20, 21, 22, 23, 24], and metasurfaces [25, 26, 27]. These systems offer ultrahigh resolution, vivid brightness, wide color gamut, and long‐term stability. However, conventional fabrication often relies on costly, intricate techniques—such as e‐beam lithography [28, 29], nanoimprinting [30, 31], and multilayer deposition [20, 21]—which limit scalability and practical adoption. Laser‐based nanostructuring presents a promising alternative, offering non‐contact operation, high spatial resolution, and mask‐free patterning. Approaches such as laser‐induced coloring of noble metals [32], femtosecond laser‐induced periodic surface structures (LIPSS) [33, 34, 35, 36], and laser‐induced oxidation [37, 38, 39, 40, 41] have been explored. Among these, plasmonic colors from noble metals suffer from a limited color gamut and strong dependence on the substrate [42]. LIPSS offer iridescent colors that vary with viewing angle and typically require low‐throughput processing [43]. In contrast, laser‐induced oxidation allows for nanoscale precision in controlling oxide thickness, yields angle‐independent structural colors, and ensures high‐throughput fabrication [44]. This mechanism—primarily governed by thermal, plasmonic, and photochemical effects—has been widely used in materials like titanium alloys and stainless steel, where surface oxidation occurs as molten metal captures atmospheric oxygen under localized heating [45, 46]. Titanium dioxide (TiO_2_), with its high refractive index and low absorption, is an ideal material for structural color formation [47]. Nonetheless, for bulk titanium, nanosecond laser‐induced oxidation typically yields oxide layers below 80 nm; further increasing energy leads to ablation and blackening, limiting the achievable color gamut to ∼15% sRGB [48, 49, 50]. Recently, researchers have proposed various multilayer thin‐film architectures to overcome substrate dependence while significantly expanding the achievable color gamut. For instance, in our previous work [41, 44], we developed TiAlN–TiN hybrid films that enabled wide‐gamut, highly stable, and high‐speed structural color printing. However, due to the intrinsic absorption characteristics of these films, the peak reflectance of the structural colors did not exceed 20%. Lapidas et al., [39]. Further demonstrated Ti–TiO_2_ hybrid films, where a combination of laser‐induced periodic surface structures (LIPSS), surface ripples, and oxidation produced a diverse color palette with a maximum resolution of 20 000 DPI and peak reflectance of 30%. Despite these advances, simultaneously achieving wide color gamut, high brightness, and high spatial resolution remains a major challenge.

In this study, we propose a Ti‐based precursor thin‐film architecture incorporating a thickness‐tunable TiO_2_ interlayer. Upon picosecond laser irradiation, controlled oxidation of the residual Ti layer occurs, enabling both the growth and precise modulation of the TiO_2_ thickness. TiO_2_ forms a Fabry–Pérot‐like optical cavity with the adjacent metallic Ti layers, generating multiple resonance modes across the visible spectrum. This strategy enables precise color control (ΔE < 3), ultrahigh resolution up to 30 000 DPI, and a peak reflectance of 60%, while maintaining a color gamut exceeding 80% of the sRGB space, representing a significant advance toward practical, high‐performance laser‐fabricated structural color technologies.

Results and Discussion

2

As illustrated in Figure 1a, the sandwich‐structured film consists of a TiO_2_ nanolayer pre‐deposited beneath the surface of a Ti metal film. The top Ti layer has a thickness of 30–50 nm, the TiO_2_ interlayer ranges from 0 to 150 nm, and a typical bottom Ti layer is 120 nm. Upon laser irradiation, the top Ti layer reaches high temperatures first and undergoes thermal oxidation. As the laser energy input increases, the oxidation front gradually propagates inward. Meanwhile, the pre‐deposited TiO_2_ interlayer, which has a relatively high melting point, becomes thermally softened or even molten under high fluence but does not chemically interact or compete for oxygen due to its already oxidized state. This allows the oxygen species to diffuse further into the structure, enabling oxidation of the bottom Ti layer. This sequential oxidation process results in a total TiO_2_ thickness composed of both the pre‐deposited oxide and the laser‐induced conversion of Ti, enabling an improved reflectance and a broadened color gamut compared to conventional single‐layer Ti oxidation. Figure 1b illustrates the correlation between the laser parameters and oxide thickness, clearly showing that the TiO_2_ thickness exhibits an approximately linear dependence on the deposited laser fluence. When the single‐pulse energy ranges from 6 to 6.8 µJ, oxidation occurs only in the upper Ti layer, resulting in TiO_2_ thicknesses of approximately 22–64 nm. As the pulse energy exceeds 7 µJ, oxidation also occurs in the bottom Ti layer, and the total laser‐induced TiO_2_ thickness reaches up to ∼152 nm. Combined with the pre‐sputtered 70 nm TiO_2_ interlayer, the maximum total TiO_2_ thickness achieves approximately 222 nm. Figure 1c shows, under optical microscopy, continuous and finely resolved color transitions accompanied by smooth surface morphology. Details on the coating procedures, laser setup, and processing parameters can be found in the Materials and Methods section. The observed color changes originate from a triple‐resonance reflection spectrum generated by the TiO_2_/Ti bilayer within the visible range. Changes in the TiO_2_ layer thickness on top then modulate the optical path length, resulting in systematic resonance peak shifts. As shown in Figure S1, increasing TiO_2_ from 0 to 80 nm, the interference‐induced reflection spectrum undergoes a blue shift, resulting in a limited gamut. Further increasing the TiO_2_ thickness to 210 nm induces a pronounced red shift in the spectral peaks, thereby significantly broadening the achievable color gamut. Moreover, at low laser fluence, the top Ti partially oxidizes, transforming the Ti–TiO_2_–Ti stack into a transitional four‐layer state (TiO_2_/Ti/TiO_2_/Ti). Simulations confirm that this configuration further broadens the accessible color gamut and identify an optimal thickness window for the top Ti layer (see Figure S2). We also assess the influence of the back reflector (Ti vs. Cu/Al/SUS) on the reflectance spectra (see Figure S3).

Schematic and demonstration of laser‐induced coloration in a Ti‐TiO2‐Ti sandwich structure. (a) High‐brightness, full‐gamut structural colors are directly written using a picosecond laser. By introducing a thin TiO2 layer at an optimal position within the Ti film, the trilayer structure enables simultaneous enhancement of brightness and color gamut under identical laser processing conditions. (b) Correlation between laser parameters and the resulting TiO2 thickness. The multilayer structure (40 nm Ti/70 nm TiO2/120 nm Ti) was irradiated with laser pulses ranging from 6 to 10.4 µJ at a scanning speed of 16 mm/s, resulting in controllable oxidation of both the top and bottom Ti layers and corresponding shifts in the reflectance spectra. (c) Optical microscope images of representative laser‐printed structural colors. The laser fluence is indicated in the figure. Sample I was produced at a scanning speed of 60 mm/s, while samples II–IX were processed at 1 mm/s. A broad range of colors—such as yellow, orange, purple, red, blue, cyan, and green—was achieved.

A 200 nm Ti thin film was deposited on a silicon substrate and subjected to laser‐induced oxidation under varying energy inputs. The resulting colors under different conditions are shown in Figure 2a–i. To investigate the limits of the color gamut achievable with the Ti–TiO_2_–Ti sandwich structure, we prepared a series of precursor films with varying thicknesses of the intermediate TiO_2_ layer—20, 40, 60, 70, 80, 100, 120, and 140 nm—while keeping the top and bottom Ti layers fixed at 40 nm and 120 nm, respectively. The resulting laser‐printed structural colors are presented in Figure 2a(ii–viii) and Figure 2b. As the pre‐deposited TiO_2_ thickness increases, the tunable color range gradually narrows. This reduction is attributed to the diminished participation of the bottom Ti layer in the laser‐induced oxidation process. Our results show that a pre‐deposited TiO_2_ thickness of approximately 70 nm achieves an optimal balance, offering both a broad tunable range and an extended color gamut. This configuration enables the generation of vivid structural colors with high reflectance and broad chromatic coverage. By simultaneously adjusting the pre‐deposited TiO_2_ thickness and the accumulated laser fluence, we achieved a color gamut coverage exceeding 80% of the sRGB space, as illustrated in Figure 2c. We measured the reflectance spectra of selected colors shown in Figure 2b. As the accumulated laser fluence increased, a noticeable redshift in the spectral peaks was observed, accompanied by a decrease in peak reflectance from 53% to 40%, as illustrated in Figure 2d. In addition, we evaluated the reflectance spectra for other samples with different pre‐deposited TiO_2_ thicknesses (80, 100, 120, and 140 nm). The reflectance curves obtained under their respective optimal oxidation conditions are plotted in Figure 2e. All of these samples exhibit peak reflectance values exceeding 60%. Compared with related publications, our design attains a high peak reflectance while preserving a wide color gamut (see Figure S4); moreover, the reflectance of our structural colors surpasses that of a commercial reflective film (see Figure S5).

Laser‐printed color images, reflectance, and gamut of laser‐modified Ti–TiO2–Ti films. (a,b) Laser‐printed structural colors by tuning single‐pulse energy and scanning speed, based on a single 200 nm Ti layer and Ti–TiO2–Ti sandwich structure with layer thicknesses of 40 nm (top Ti), 20–140 nm (TiO2), and 120 nm (bottom Ti). (c) Color gamut coverage in the CIE 1931 chromaticity diagram. The crosses indicate the chromaticity coordinates of the colors shown in (a,b). By simultaneously adjusting the laser energy and initial film structure, a color gamut covering approximately 80% of the sRGB space is achieved. (d) Reflectance spectra corresponding to the structure in (b), demonstrating fine spectral tuning via modulation of laser power density: by increasing single‐pulse energy (6.8–9.2 µJ) at 1 mm/s or adjusting scanning speed (5–19 mm/s) at 6.8 µJ. The reflectance spectra showed gradual shifts with a step size of approximately 10 nm. (e) Representative reflectance spectra after laser oxidation for samples with varying initial thin‐film structures, where the pre‐sputtered TiO2 middle layer ranges from 70 nm to 140 nm. All spectra exhibit peak reflectance exceeding 60%.

Precise control over the optical thickness of the TiO_2_ film is key to achieving these results. Owing to the continuous modulation of laser deposition energy, the optical thickness of the TiO_2_ layer can be finely tuned in a continuous manner. Measured reflectance spectra in Figure 2b demonstrate an average spectral shift resolution of ∼10 nm, corresponding to an energy increment of 0.2 µJ per pulse. This enables highly refined color gradations with an average color difference (ΔE) below 8 and a minimum ΔE of less than 3. Our design achieves exceptional color saturation in the violet, blue, and green regions—colors that are typically challenging to obtain using conventional structural coloration approaches.

The laser had a pulse duration of 10 ps, a central wavelength of 1030 nm, and a repetition rate of 1000 kHz. The laser system and optical setup are detailed in the Supporting Information. After galvanometric scanning and focusing, the beam spot size was approximately 70 µm, with a line scanning pitch of 25 µm. The oxidation of Ti is strongly dependent on laser power density, which can be approximated by:P  =  (E · f · d/v)/S  =  K · E/v. Where K  =  (f · d)/S, E is the single‐pulse energy, v is the scanning speed, f is the repetition rate, d is the line pitch, and S is the laser spot area. Higher pulse energy or lower scanning speed leads to increased energy accumulation, resulting in deeper oxidation. However, the final oxidation state depends not only on the input power density. As illustrated in Figure 2b, both yellow A (25 mm/s, 7 µJ) and yellow B (250 mm/s, 9.4 µJ) exhibit similar hues. However, the input power density for yellow B is significantly lower than that for yellow A. This discrepancy is due to concurrent heat dissipation through conduction and radiation, both of which are functions of E and v. Thus, the net energy accumulation can be described as:P_net=K·E/v−Loss(E,v). The mode of energy deposition plays a crucial role in determining the structural evolution of the multilayer film. To achieve the same oxidation depth, significantly higher pulse energies are required at high scan speeds, whereas lower energies are sufficient at slow scan speeds. This behavior is attributed to the reduced thermal accumulation at faster scanning, which limits the temperature rise and hence the oxidation efficiency. Beyond pulse energy, pulse width critically shapes TiO_2_ formation. At equal average power density, ps pulses have much higher peak power than ns pulses, producing distinct thermal histories: electrons absorb energy rapidly, transiently overheating relative to the lattice; subsequent electron–phonon coupling governs heat flow and oxidation dynamics. Although a full quantitative model is pending, our data show that the 10 ps processing yields a uniform oxide with few microcracks and consistent color, whereas 100 ns produces pronounced microcracking and reduced color uniformity (see Figure S6).

To understand the structural evolution of the thin films under laser irradiation, we selected three representative samples corresponding to the initial structure, low laser fluence, and high laser fluence, and a comprehensive set of characterizations was performed. The surface morphology and roughness was analyzed using optical microscope and atomic force microscopy (AFM), revealing that the processed films maintained good surface flatness (see Figure S7, S8). To evaluate the multilayer structure, including its elemental distribution and microstructure, we employed focused ion beam scanning electron microscopy (FIB‐SEM) and X‐ray photoelectron spectroscopy (XPS). Cross‐sectional SEM images are shown in Figure 3a–c. As seen in Figure 3a, the initial structure consists of a well‐defined three‐layer stack: 40 nm top Ti, 65 nm TiO_2_, and 114 nm bottom Ti. After laser‐induced oxidation at a low fluence, the top Ti layer is fully oxidized and merges with the pre‐deposited TiO_2_ layer, forming a single oxide layer. Meanwhile, the bottom Ti layer remains largely unchanged with a thickness of over 100 nm, as shown in Figure 3b. Upon increasing the laser fluence, the bottom Ti layer is partially consumed, reducing its thickness to 68 nm. Simultaneously, the oxide layer grows to approximately 221 nm, and microcracks appear within the film structure, as illustrated in Figure 3c. FIB‐SEM analysis revealed that the Ti layer gradually oxidized into TiO_2_ under laser irradiation, with increasing oxide thickness at higher fluences. The optical thickness and uniformity of the oxide layer varied with laser energy, accounting for the observed spectral shifts and color changes. At higher energy levels, microcracks and pores appeared due to accumulated thermal stress, which compromised the structural integrity of the film and increased optical absorption. This degradation in film quality explains the reduced reflectance observed at high laser energy densities, consistent with the experimental results shown in Figure 2d. Depth‐resolved XPS was performed to further confirm TiO_2_ formation as shown in Figure 3 d–f. Argon ion sputtering was used to remove successive layers, and high‐resolution spectra were acquired at 20 to 200 s etch time intervals. The depth profiles of Ti 2p, O 1 s, and Si 2p signals show that the surface region is dominated by TiO_2_ after laser irradiation, evidenced by the high oxygen and titanium signals at the surface. As the etching proceeds, the oxygen signal gradually decreases while the titanium signal shifts from oxidized Ti (Ti⁴⁺) to metallic Ti (Ti⁰), indicating a transition from TiO_2_ to underlying Ti layers. The Si signal emerges at later stages, corresponding to the silicon substrate, confirming the multilayer structure and the film thickness. Compared with the untreated region, the laser‐irradiated area exhibits a thicker TiO_2_ layer, as reflected by the prolonged presence of oxygen and oxidized titanium signals, suggesting enhanced oxidation induced by laser energy. Cross‐sectional EDS analysis further confirmed that the TiO_2_ content increased with rising laser fluence (see Figure S9). Indeed, laser processing can produce different TiO_2_ polymorphs (anatase and rutile) with distinct refractive indices that, in principle, affect the optical thickness. To assess this, we measured Raman spectra for three representative colors (i.e., different oxide thicknesses; see Figure S10). All spectra show mixed anatase–rutile signatures with broad, low‐intensity peaks, indicating modest crystallinity; at the highest fluence, the rutile bands (447, 612 cm^−1^) strengthen slightly, but the films remain mixed phase. These results indicate that phase‐induced changes in n eff_are secondary, and that the observed structural colors are governed primarily by the physical thickness increase of laser‐grown TiO_2.

Cross‐sectional microstructure and compositional evolution under different laser irradiation doses. (a–c) FIB‐SEM cross‐sectional images of the Ti–TiO2–Ti multilayer before laser processing, after low fluence, and after high fluence. Morphological changes and oxide growth are observed with increasing laser energy; the scale bar is 200 nm. (d–f) XPS depth profiling results showing the elemental distribution of Ti, O, and Si as a function of depth for the initial, low fluence, and high fluence films.

The resolution of laser printing is primarily determined by the size of the focused laser spot. To investigate the resolution limit, we employed a high‐numerical‐aperture (NA  =  0.8) objective lens to significantly reduce the spot size. The multilayer stack was Ti (120 nm) / TiO_2_ (140 nm) / Ti (40 nm). We patterned 3 × 8 dot arrays with a 3 µm pitch as shown in Figure 4a(i). For each row, we varied the single‐pulse energy (1.6, 2.0, 2.4 nJ) and the dwell time (50–600 ms) to tune the deposited fluence. The resulting checkerboard exhibits adjacent dots with clearly different hues. Optical microscopy images reveal that the structural color dots have diameters below 1 µm, with the smallest features measuring near 500 nm. Thus, different color outputs were achieved by varying the thickness of the TiO_2_ interlayer, demonstrating the tunability of structural color at the microscale. SEM measurements in Figure 4a(ii‐iii) further confirms that even at the highest deposited energy the dot diameter remains below 800 nm, corresponding to a printing capability > 30 000 DPI (25.4 mm/0.8 µm ≈ 31,750 DPI), and the structural color regions exhibit no structural damage, with only minor contrast changes, supporting that the color modulation originates from controlled oxidation rather than surface ablation. We also fabricated 1 µm‐wide lines and the logo of Westlake University, as shown in Figure 4c. These structures further demonstrate the capability of our method for high‐resolution and precise patterning. In addition, we demonstrated the capability for large‐area and complex patterning; we laser‐printed detailed images of the Westlake University Yungu Campus and the Westlake Institute for Optoelectronics, as shown in Figure 4c,d. The parameters used for fabricating different colors can be found in Figure 2b. The resulting patterns exhibit vibrant colors with high brightness and saturation. A multicolor structural color plate was observed under different viewing angles ranging from 0° to 80° (step = 10°), as shown in Figure S11. In addition, a representative green region was selected, and its chromaticity coordinates were measured under the same viewing angles (see Figure S12). All measurements exhibit excellent consistency, confirming the angle‐independent optical performance of the laser‐fabricated structural colors. Scalability of the process was demonstrated by batch‐fabricating 40 identical samples using the same laser writing pattern. As shown in Figure S13, the samples exhibit excellent uniformity and reproducibility. To assess the long‐term durability of the printed colors, we subjected the samples to salt spray, damp heat, and condensation tests. Post‐exposure analysis revealed negligible changes in color shift, confirming the strong environmental stability of the structural colors (see Figure S14). In addition, we extended the coating and laser patterning process to stainless steel substrates (see Figure S15). The structural coloration remained consistent, demonstrating the broad substrate compatibility of our approach.

High‐resolution laser‐printed structural colors and patterns. (a) Structural color dots fabricated using a high‐NA (0.8) objective lens, the thin film consists of 40 nm (top Ti)/140 nm (middle TiO2)/120 nm (bottom Ti). The single‐pulse energy was varied from 1.6 to 2.4 nJ, and the dwell time from 50 to 600 ms to tune the deposited fluence (i‐ii). SEM characterization of the blue color dot (iii), revealing diameters smaller than 800 nm. (b) Structural color lines (i) and the logo of Westlake University (ii). (c) Laser‐printed image of Westlake University Yungu Campus on a 2‐inch polished silicon wafer. (d) Laser‐printed image of Westlake Institute for Optoelectronics on a 2‐inch rough silicon wafer substrate.

Conclusion

3

In summary, compared to conventional pigment‐based colors, structural colors derived from nanometric thin films represent a promising next‐generation coloring technology. We present a Ti–TiO_2_–Ti sandwich structure fabricated via magnetron sputtering, in combination with picosecond laser‐induced oxidation, to achieve nanometer‐precision control over TiO_2_ layer thickness directly on metallic Ti films. This approach enables vivid structural coloration with high saturation and precisely tunable hue differences, covering over 80% of the sRGB gamut, achieving reflectivity exceeding 60%, and offering exceptional resolution up to 30,000 DPI. The resulting films exhibit outstanding durability, including resistance to corrosion and abrasion. Importantly, our deposition and laser direct‐writing strategy is substrate‐independent, allowing for versatile integration across different surfaces. The method combines high fabrication throughput with low cost, offering a scalable and customizable platform for high‐quality structural color production. Potential applications span a wide range of fields, including information security, artistic design, advertising, and environmental technologies.

Materials and Methods

4

Multilayer Films Coating

4.1

The Ti and TiO_2_ multilayer films were deposited using a magnetron sputtering system under high vacuum conditions, with a base pressure maintained below 5 × 10^−4^ Pa. During the deposition of Ti layers, a DC power supply of 500 W was applied, with argon (Ar) and oxygen (O_2_) flow rates of 100 and 3 sccm, respectively. For the TiO_2_ layers, an RF power supply of 500 W was used, with Ar and O_2_ flow rates of 197 and 3 sccm, respectively. The working pressure was kept at 0.4 Pa throughout the process, and the deposition was performed at room temperature with a target‐to‐substrate distance of 10 cm. The deposition rates were approximately 0.21 nm/s for Ti and 0.04 nm/s for TiO_2_. The layer thicknesses were accurately controlled by adjusting the sputtering time.

Laser Processing Setup and Optical Path

4.2

In our experiments, surface coloring was performed using a custom‐built laser marking system as shown in Figure S16. The laser source was a 1030 nm picosecond laser (Amplitude, Tangerine) with a pulse duration of 10 ps. Both the repetition rate and pulse energy were tunable to precisely control the laser‐material interaction. The initial laser beam was linearly polarized in the S‐polarization state and had a diameter of approximately 3 mm. The laser beam first passed through an energy attenuator composed of a half‐wave plate and a polarizing beam splitter, allowing fine adjustment of the incident pulse energy. Subsequently, the beam passed through a Glan‐Taylor prism with an extinction ratio of 1 00 000:1 to further purify the polarization. After that, a quarter‐wave plate was used to convert the linearly polarized beam into circular polarization, which effectively suppresses the formation of laser‐induced periodic surface structures. The circularly polarized beam was then directed into a galvanometric scanner for dynamic beam steering and finally focused onto the sample surface using a lens with a focal length of 20 cm. The focal spot diameter was measured to be approximately 120 µm using a beam profiler, and the line spacing is 25 µm. All laser processing was conducted under ambient environmental conditions, with the temperature maintained at room temperature (∼25°C) and the pressure at standard atmospheric pressure (∼101.3 kPa).

Optical and Micro/Nano Characterization

4.3

Normal‐incidence reflectance spectra were measured using a CRAIC 20/30PV UV–vis microspectrophotometer (CRAIC Technologies Inc.). The measurement area was 20 µm × 20 µm with a spectral resolution of less than 2 nm. The reflectance data were processed using Python and converted into RGB values and CIE 1931 chromaticity coordinates (x, y). Numerical calculations and spectral plotting were performed using the numpy and matplotlib packages. Angle‐dependent reflectance spectra were obtained using a Lambda 1050 spectrophotometer (PerkinElmer). Bright‐field optical images of the sample surfaces were acquired using an LSM 900 optical microscope (Carl Zeiss) under various objective lenses, with a maximum magnification of 100× (NA = 0.90). Surface morphology and elemental composition were characterized using a ZEISS GeminiSEM 360 scanning electron microscope (SEM) equipped with an in‐lens detector and energy‐dispersive X‐ray spectroscopy (EDS). Cross‐sectional images were obtained using a Helios 5 UX focused ion beam–scanning electron microscope (FIB–SEM, Thermo Fisher Scientific), where a gallium ion beam was used to mill the sample surface. Depth‐profiling X‐ray photoelectron spectroscopy (XPS) analysis was conducted using a Nexsa G2 system (Thermo Scientific). Surface roughness was characterized using a Dimension ICON atomic force microscope (Bruker).

Color Difference (ΔE) Calculation

4.4

The reflectance spectra were measured under a D65 standard illuminant and converted into CIE 1931 XYZ tristimulus values using the CIE standard color matching functions:

$$[eqn]$$

where R(λ) is the measured reflectance, S(λ) is the D65 illuminant spectrum, and k is a normalization constant:

$$[eqn]$$

The XYZ values were then transformed into CIE Lab* coordinates relative to the D65 reference white point(*X_n_ *,*Y_n_ *,*Z_n_ *):

$$[eqn]$$

with

$$[eqn]$$

Finally, the color difference between pristine and tested samples was quantified using the CIE76 formula:

$$[eqn]$$

Color Fastness Tests

4.5

The color fastness of the laser‐processed samples was evaluated by conducting a salt fog test and a damp heat (double‐0) test. The salt fog test was performed in accordance with the GB/T 10125–2021 standard for a duration of 96 h. A 5 wt.% NaCl solution with a pH of 6.8 was used, and the test was conducted at a constant temperature of 35 °C. The salt spray sedimentation rate was maintained at 1.5 mL/(80 cm^2^·h). The double‐80 test (damp heat test) was conducted following the GB/T 1740–2007 standard, also for 96 h. The test chamber was maintained at a temperature of 80 °C and a relative humidity of 80% to simulate high‐temperature and high‐humidity environmental conditions. To quantitatively assess the color stability after these durability tests, the color difference (ΔE) between the pristine and tested samples was measured using an integrating sphere spectrophotometer under a D65 standard illuminant. The results showed that the average color difference remained below 7, indicating good color fastness under both corrosive and humid environments.

Author Contributions

M.Q. conceived the concept and supervised this research. R.C. and W.L. performed the experimental work on coating, laser processing and characterization. Y.S. assisted in setting up the experimental apparatus. R.C., H.Z. and Y.C. designed the patterns and conducted related experiments. R.C. performed data analysis and figure preparation. R.C. wrote the manuscript with M.Q. All authors contributed to revisions and comments of the manuscript and discussed the results.

Conflicts of Interest

The authors declare no conflict of interest.

Supporting information

Supporting File: advs73515‐sup‐0001‐SuppMat.docx.