1. Introduction
In recent decades, the pursuit of miniaturization has been crucial in nanofabrication, fostering innovation, and enabling novel applications in chip manufacturing, nanophotonics, and quantum devices
[1],
[2]. Advancements in nanofabrication technology are driven by the demand for higher component density and performance, necessitating precise material processing in atmospheric environments. Traditional techniques such as electron beam lithography
[3] and nanoimprint lithography
[4] have achieved manufacturing resolution at a scale of tens of nanometers, offering high precision, scalability, and the ability to create complex nanoscale structures. Currently, the 3 nm process, pioneered by Samsung Electronics (Republic of Korea) and the Taiwan Semiconductor Manufacturing Company (TSMC; China) using extreme ultraviolet nanolithography, is the most advanced mass production technology in the semiconductor industry. Competition for the development of the 2 nm advanced process has significantly expanded to several major chip manufacturing companies, including TSMC, Intel, and Samsung (
Table 1), with production targeted for approximately 2025.
Since their introduction in 1960, lasers have been considered a viable technology for maskless nanolithography and direct three-dimensional (3D) writing. However, owing to the optical diffraction limit, nanopatterning beyond the diffraction limit is challenging when using laser irradiation in far-field and atmospheric environments. Comparing optical near- and far-field laser manufacturing, the near-field approach offers the advantages of robust subwavelength precision and high spatial resolution, enabling localized energy deposition and minimal heat-affected zones. However, it requires complex equipment, flattens the sample surface, and has a limited processing area, making it unsuitable for large-scale applications. Conversely, far-field laser processing features simple optical setups, broad material applicability, and high processing speeds, making it suitable for industrial-scale production. However, it is constrained by the diffraction limit, resulting in a lower resolution. In recent decades, considerable efforts have focused on improving the spatial resolution of far-field laser manufacturing. Various mechanisms, including stimulated emission depletion (STED)
[5], multiphoton absorption
[6], and optical far-field-induced near-field enhancement (O-FIB)
[7], have led to feature sizes as small as tens of nanometers, promoting laser-based nanomanufacturing towards finer resolutions. This study provides a brief overview of recent developments in nanomanufacturing using optical near- and far-field laser irradiation, as depicted in
Fig. 1, and discusses the challenges and limitations in this field.
2. Optical near-field laser nanomanufacturing
The basic mechanism of near-field laser processing involves manipulating the optical field via discontinuous interfaces with nanoscale dimensions. This is typically achieved by irradiating a small aperture or tip with a far-field-focused spot to excite the evanescent waves that propagate along the irradiated surface. For nanostructures with feature sizes much smaller than the incident laser wavelength, generating electromagnetic field components with stronger transverse wave vectors is feasible, leading to a spatially localized field distribution that exceeds the diffraction limit
[8]. Localized near-field enhancement was first demonstrated by combining an 800 nm femtosecond laser with a scanning probe
[9], enabling the fabrication of nanostructures on a gold film as small as approximately 11 nm (
). Since then, several studies have reported processing sizes of 10–30 nm
[10],
[11]. The high resolution of near-field laser processing comes at the expense of its evanescent characteristics. Once the optical near-field moves away from the discontinuous boundary that excites it, the light intensity decays exponentially. Consequently, the working distance for near-field laser processing is significantly short, relying primarily on near-contact exposure for nanofabrication.
In optical near-field processing, feature size primarily depends on the physical dimensions of the near-field probe and is less sensitive to the wavelength and pulse width of the excitation light in the far field. For an unflattened sample surface, any external disturbance can cause a nanoscale change (
; distance
from the probe to the substrate surface in near field), leading to exponential energy decay and reduced scanning speed. To address the short working distance of the evanescent field during dynamic processing, precision controls, such as optical tweezers, atomic force microscopy (AFM), and microspheres, are essential. Optical tweezers are used to manipulate microspheres in a water environment, utilizing the near-field focusing effect of the microspheres to fabricate arbitrary patterns with a resolution of approximately 100 nm (
)
[12]. A quantum cascade laser combined with AFM-enhanced near-field nanolithography has achieved a maximum
in situ resolution of approximately 35 nm on an organic sample surface
[13], which was applied for biomolecular information storage. Moreover, the microsphere can be elevated to create a gap between it and the substrate surface, extending its working distance to several micrometers and enabling non-contact operation. By precisely controlling the microsphere and adjusting the thickness of the phase-change film, a minimum feature size of approximately 30 nm (
) can be achieved
[14]. The fitting results of the ablation width versus thickness indicate that the ablated feature size may be further reduced to approximately 15 nm at film thicknesses as low as approximately 10 nm.
3. Optical far-field laser nanomanufacturing
Near-field enhancement primarily focuses on controlling the optical field, without fully exploring the specific mechanisms of laser–material interactions during processing. This technique typically requires a smooth sample surface owing to the narrow working distance, limiting its applicability. Therefore, research into ultrafast laser nanolithography, from near-field to far-field nonlinear effects, is an inevitable trend. Laser processing involves complex physical effects, and nonlinear laser–material interactions have been shown to surpass the optical diffraction limit, enabling super-resolution processing.
Table 2 [5],
[6],
[7],
[15],
[16],
[17],
[18],
[19],
[20] summarizes the nanomanufacturing results obtained with feature sizes beyond the diffraction limit using far-field laser approaches reported in the past five years. These approaches include STED, multiphoton absorption, O-FIB, longitudinal field superposition, and dual-beam overlapping. Multiphoton absorption using a femtosecond laser induces an optical chemistry effect for nanostructure creation. For instance, 3D nanoengraving has been established using STED, applicable in optical data storage, with a capacity reaching the petabyte scale by dual-laser direct writing and reading of multilayer data points at a minimum size of approximately 55 nm
[5]. A 785 nm femtosecond laser successfully generated 3D nanostructures of hydrosilicates, achieving a feature size of 26 nm that corresponds to a processing precision of
[6].
To ensure uniform longitudinal energy deposition in the far field and confinement of the lateral subwavelength light beam in the near field during laser–matter interaction, a new strategy called optical far-field-induced near-field breakdown was proposed for surface nanofabrication
[7],
[15]. The seed-induced backscattering diffusion mechanism for ultrafast laser-induced nanostructuring forms a positive feedback loop that extends and homogenizes longitudinal energy deposition. By dynamically modulating the polarization of an 800 nm femtosecond laser, nanopatterns with a spatial resolution of < 20 nm (
λ/40) were created on the surface of titanium dioxide thin film
[15]. Modulation of a high-quality longitudinal optical field also facilitates the creation of nanostructures with feature sizes beyond the optical limit. Nanoholes with diameters of 10–30 nm were fabricated on crystalline aluminum oxide using longitudinal field irradiation
[16], achieving an aspect ratio exceeding 16. Hybrid laser irradiation schemes have been employed for precise engineering of diverse materials to address the diffraction limit in the far field. Additionally, dual-beam irradiation is a strategy used to achieve finer nanofabrication in the far field. Dual 405 nm nanosecond laser overlapping can achieve a minimum linewidth of 5 nm
[17]. In using dual orthogonal polarization overlapping of 800 nm femtosecond laser irradiation, direct 12 nm (
λ/66)
[18] nanostructures were fabricated on a silicon surface by coupling the two beams at orthogonal polarization and optimizing laser processing parameters at high repetition rates. By adjusting the dual pulses of a 1064 nm nanosecond laser, direct writing of approximately 30 nm on a single-crystal silicon surface was also achieved
[19]. These studies indicate that pulsed laser precision engineering is a viable approach for producing sub-10 nm features.
4. Challenges and outlooks
Ultrafast laser manufacturing has advanced significantly in precision engineering, enabling feature sizes from microns to nanoscale. However, several challenges remain unresolved.
First, the rapid development of nanofabrication still struggles with achieving nanometric high-aspect-ratio processing. This challenge arises from the diffraction-limited propagation of light, which contradicts the requirement for uniform axial and subwavelength lateral energy deposition for nanoscale processing at high aspect ratios. Additionally, complex light–material interactions exacerbate the inhomogeneity of nanocreation, including linear and nonlinear absorption, scattering, and thermal and stress effects resulting from intense energy deposition. To address these issues, two primary solutions have been proposed. One involves axially stretched or multifocus beam irradiation, which can be achieved by modulating the phase and amplitude of the incident pulses to balance the depth of focus with the laser spot size
[21]. The other is a dicing technology based on a Bessel–Gaussian beam. Study has shown that the Bessel-like beam can create holes in transparent hard materials at a diameter of approximately 35 μm and an aspect ratio of over 1000
[22]. However, the aspect ratio of nanofabrication using far-field femtosecond laser irradiation is mostly limited to 10, which does not satisfy industrial requirements. Recently, a new strategy based on backscattering interference crawling and chemical etching has been shown to achieve spatial widths in the tens of nanometers range, with aspect ratios ranging from 1000 to 10 000 on transparent functional solids. This approach also reduces the roughness of the cutting surface to 9.4 nm
[7], offering a promising method for realizing nanofabrication at high aspect ratios.
Second, owing to the instability of the nonlinear threshold phenomenon, the processing size achievable by the ultrafast laser was constrained to approximately 10 nm. Threshold tracing and locking have been reported to enhance the spatial resolution of femtosecond laser processing up to the quantum limit of two-dimensional material surfaces
[20]. At laser energies near the atomic damage threshold, atomic removal may not occur precisely at the center of the laser irradiation area. In this case, the laser energy gradient becomes less pronounced, leading to an ineffective breakdown region, which is commonly determined by the laser energy gradient. Instead, localized atomic removal occurs sporadically within the ablation area, significantly influenced by the position and energy fluctuations of the local electrons rather than by the laser-induced thermal gradient. This effect has enabled feature sizes below 5 nm
[20], surpassing the optical diffraction limit and the secondary gap between atomic point-defect complexes. These studies indicate that sub-10 nm laser nano-manufacturing should consider the effects of the fundamental lattice structure and elementary electron vibrations of materials, providing a new perspective for super-resolution nanoscale and atomic-level processing.
Both ultrafast laser fabrication systems and laser sources are crucial for efficient, high-quality laser nanostructuring. High-repetition-rate femtosecond lasers are mostly preferred for nanostructuring owing to their high stability. Over the past decades, pulse-shaping technology has become a potential method for achieving better performance in laser material processing. Recently, gigahertz burst-mode femtosecond lasers
[23], which emit femtosecond pulse trains with extremely short pulse intervals of several hundred picoseconds, have garnered significant attention for their ability to achieve high-quality, high-efficiency, and precise material processing. Additionally, achieving both high resolution and efficiency in laser nanomanufacturing presents a significant challenge, particularly for large-scale nanofabrication requiring diverse industrial needs. To address this issue, parallel laser beam manufacturing, including microlens arrays
[24] and spatial light modulators
[25], offers scalable solutions that balance the demands of resolution and throughput in large-scale nanofabrication. Consequently, ultrafast laser processing, which holds promise for nanofabrication, has been proposed. However, many new phenomena remain unexplained owing to complex nonlinear interactions between femtosecond lasers and materials, necessitating further investigation.
CRediT authorship contribution statement
Zhenyuan Lin: Writing – original draft. Lingfei Ji: Writing – original draft. Minghui Hong: Writing – original draft, Conceptualization.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
This work was supported by the National Natural Science Foundation of China (51975017 and 52405448), the Human Resource Training Project (HRTP–[2022]–53) of Innovation Laboratory for Sciences and Technologies of Energy Materials of Fujian Province (IKKEM). Zhenyuan Lin is grateful for the support by the China Postdoctoral Science Foundation (2024M750149 and GZC20240087).
PDF
(438KB)
