Laser interference lithography serves as an established manufacturing technique for generating precise and periodic microstructures across functional surfaces. Among various optical configurations, the non-orthogonal Lloyd’s mirror interferometer functions as a wavefront-division interferometer characterized by a simple architecture and short non-common optical paths, providing mechanical and environmental stability even within standard laboratory and machine-shop settings. The technique allows maskless exposure over large surface areas and facilitates straightforward tuning of interference fringe periods by adjusting mirror angles. Despite these operational merits, conventional setups remain constrained by predefined interference intensity fields, where the resulting geometric shapes are fundamentally locked to fixed optical arrangements. To fabricate non-standard or alternative patterns, researchers traditionally have had to redesign the entire optical layout, alter beam numbers, or adjust physical polarization elements. Addressing this fundamental limitation, a research team developed a dynamic approach integrating a spatial light modulator into a non-orthogonal two-axis Lloyd’s mirror interferometer to achieve active wavefront control and arbitrary translation of interference fringes during exposure.
The underlying operational principle relies on superposing three coherent laser sub-beams to produce a two-dimensional grid of interference spots, commonly referred to as dot array fringes. In the optical framework, a single collimated laser beam is divided into three distinct paths: one sub-beam impinges directly onto a target substrate, while the second and third sub-beams reflect off an X-mirror and a Y-mirror mounted at specific opening angles relative to the substrate surface. The interference among these three sub-beams creates periodic constructive intensity peaks whose pitch along orthogonal axes is defined mathematically by the laser wavelength and the mirror inclination angles. By routing the incident laser light through a reflective spatial light modulator driven by complementary metal-oxide semiconductor electronics prior to entering the interferometer, distinct phase modulations are applied across the sub-beams. Rather than altering the underlying spatial period of the interference field, shifting the phase of a given sub-beam introduces a controlled constant offset that displaces the constructive interference nodes along the corresponding Cartesian coordinate.
Theoretical and mathematical derivations established that the displacement of the interference fringe array in the horizontal and vertical directions is directly proportional to the dynamic phase shifts imparted to the respective sub-beams. When the phase of the sub-beam reflecting from the X-mirror is shifted, the fringe array translates along the X-axis, and independent phase shifts applied to the Y-mirror sub-beam translate the pattern along the Y-axis. Because the wave-vector differences between the interfering beams remain fixed throughout modulation, the spatial pitch and intrinsic grid arrangement of the dot array remain completely constant during translation. Consequently, applying coordinated, continuous phase modulation over time allows the light spots to follow designed continuous paths across the substrate. This dynamic movement effectively converts stationary interference lithography into a continuous drawing-type exposure process, enabling the synthesis of complex exposure dose distributions without modifying the physical positions of lenses, mirrors, or holders.
To evaluate the validity of the operational model, the investigators performed numerical calculations simulating the motion of the interference spot array along predetermined trajectories. A standard reference parameter set, encompassing optical wavelength, electric field amplitudes, polarization states, and mirror angles, was utilized to compute the accumulated exposure dose distributions. Initial simulations confirmed that setting specific phase combinations produced independent orthogonal shifts matching theoretical predictions. The numerical calculations were subsequently expanded to simulate multi-directional motion, modeling stationary exposure alongside L-shaped, square, and circular paths. By accumulating the interference intensity distributions along these designated paths, the simulations demonstrated that the original elliptical spot profile could be transformed into diverse non-elliptical geometries, including square profiles, loops, and expanded boundaries, thereby expanding pattern flexibility while retaining the stability advantages of Lloyd’s mirror systems.
The numerical calculations also revealed an intrinsic physical trade-off associated with moving-spot interference lithography: continuous translation inherently blurs the accumulated dose distribution, which progressively diminishes the optical contrast. In a stationary exposure regime, the normalized ratio of maximum to minimum intensity represents an ideal contrast of 1:0. However, as the spot translates across L-shaped, square, and circular trajectories, the minimum baseline intensity accumulates within non-targeted regions, shifting the calculated intensity ratios to 1:0.233, 1:0.412, 1:0.464, and 1:0.799, respectively. When mapped onto a 0.7-micrometer-thick layer of photoresist, this gradual loss of contrast corresponded to a simulated decrease in the maximum developable relief depth from an initial 0.70 micrometers down to 0.54, 0.41, 0.38, and 0.14 micrometers under the respective exposure paths. The researchers observed that while dynamic translation limits achievable depth under fixed conditions, this reduction can potentially be compensated by tailoring photoresist thickness and optimizing chemical exposure parameters to meet specific target profile depths.
To substantiate the theoretical and simulated framework, the researchers constructed a prototype experimental apparatus. The light source comprised a single-frequency-mode laser operating at an emission wavelength of 405.1 nanometers with an initial collimated beam diameter of 0.4 millimeters. The beam passed through a spatial filter to suppress spatial noise before a beam expander broadened its diameter to approximately 9.0 millimeters. The central, uniform region of the expanded beam was directed onto a reflective liquid-crystal spatial light modulator featuring an active area of 15.36 millimeters by 9.6 millimeters, an array pitch of 8.0 micrometers, a pixel size of 7.8 micrometers, and 1024 phase-modulation steps with a reported phase stability of approximately 0.001 pi radians. The modulated beam was directed into the interferometer, which incorporated two high-flatness right-angle prism mirrors with surface qualities of one-tenth of the wavelength mounted on four-axis precision manual stages. Behind the substrate, an integrated optical observation unit composed of an objective lens, a tube imaging lens, and a complementary metal-oxide semiconductor sensor monitored the fringe pattern in real time, with the entire apparatus supported on a vibration-isolated optical table in a darkroom.
Prior to executing microfabrication runs, the physical stability and positional precision of the constructed optical system were quantitatively assessed. In-plane positioning stability was monitored over a duration exceeding 1200 seconds by tracking the intensity centroid of individual interference spots from real-time video captured at one frame per second. The measured standard deviations of positional drift remained below 53 nanometers along both the X- and Y-directions, representing approximately 1.4 percent of the four-micrometer fringe pitch despite operation in an ordinary laboratory setting. Regional consistency was subsequently tested across four corner zones of the exposure field. The calibration curves exhibited consistent linear displacement responses, showing maximum regional variations of 240.0 nanometers in the X-direction and 246.0 nanometers in the Y-direction, corresponding to 0.60 percent and 0.615 percent of the fringe pitch. Although these variations corresponded to an optical phase deviation of 0.12 pi radians—exceeding the native 0.001 pi stability of the spatial light modulator—the deviation was attributed to optical alignment tolerances and system-level mounting errors rather than instability in the modulator.
The evaluation of orthogonal motion independence revealed minimal cross-axis coupling during phase modulation. By adjusting the grayscale input driving the spatial light modulator from 0 to 612 in increments of 36, the fringe spots were translated across cumulative distances of 6951 nanometers along the X-axis and 6745 nanometers along the Y-axis. During these sweeps, the maximum crosstalk displacement observed along the X-direction reached 58.51 nanometers, yielding a crosstalk ratio of 0.842 percent, while the maximum crosstalk displacement along the Y-direction reached 111.1 nanometers, representing a crosstalk ratio of 1.65 percent. Because these minute deviations exhibited no systematic correlation with the applied phase levels, the investigators concluded that parasitic displacements stemmed from residual optical misalignments and minor beam overlap imperfections rather than electro-optic modulation flaws, indicating that further mechanical and optical refinement could suppress crosstalk even further.
Experimental lithography was conducted on glass substrates spin-coated with OFPR-800 photoresist optimized for 405.1-nanometer illumination. The incident laser power density was calibrated to 677.2 microwatts per square centimeter, and substrates were exposed for 80 seconds, delivering an accumulated exposure dose of 487.6 millijoules per square centimeter at the central interference peak. Following exposure, substrates were developed for four seconds in a diluted developer mixture comprising two parts OFPR Developer 3 to one part deionized water, followed by a twenty-second rinse in deionized water. Although the theoretical writing speed limit of the system was estimated at approximately 20 micrometers per second, experimental writing speeds were maintained between 202 and 370 nanometers per second to protect profile resolution and contrast. During exposure, the phase patterns displayed on the spatial light modulator were updated every 200 milliseconds to match the device response time, fabricating five distinct microstructures that corresponded to the planned stationary, linear, and multi-axis trajectory designs.
Quantitative metrology performed using scanning electron microscopy and atomic force microscopy verified the fidelity and uniformity of the fabricated microstructures. Statistical analysis of the scanning electron microscopy images revealed consistent structural dimensions, with measured parameter variations across individual shapes remaining within 8 percent. Comparing physical metrics—including area, circularity, and orthogonal dimensions—against the simulated dose profiles yielded relative errors within 8.51 percent across all five pattern designs. Topographic characterization via atomic force microscopy confirmed three-dimensional profile formation, showing mean absolute depth deviations ranging from 0.13 to 0.20 when compared to normalized simulation data, aligning with the expected reduction in fringe contrast from 1:0 to 1:0.799. The authors noted that the operational exposure area in the current prototype was constrained by the Gaussian profile of the laser beam, and indicated that integrating beam-shaping elements to produce a uniform flat-top intensity profile, alongside expanding the modulated area, represents a clear pathway to scale up fabrication across larger surfaces.
Subject of Research: Two-dimensional micropattern fabrication using active wavefront control in laser interference lithography
Article Title: Fabrication of Two-Dimensional Micropattern Structures through Active Wavefront Control in Laser Interference Lithography
Article References: Takahiro, N., Shin, D. W., Banno, R., & Shimizu, Y. (2026). Fabrication of Two-Dimensional Micropattern Structures through Active Wavefront Control in Laser Interference Lithography. Nanomanufacturing and Metrology, 9(1), Article 36. https://doi.org/10.1007/s41871-026-00318-8
Image Credits: AI Generated
DOI: 10.1007/s41871-026-00318-8
Keywords: laser interference lithography, spatial light modulator, Lloyd mirror interferometer, wavefront division, dynamic phase modulation, micropatterning, photoresist, scanning electron microscopy, atomic force microscopy, microfabrication, optical lithography, fringe displacement, crosstalk displacement, surface patterning, photolithography, phase modulation
Cite Scienmag News
Denise Maddox. (November 16, 2026). Active Wavefront Control Enables Versatile Micropattern Fabrication in Laser Interference Lithography. Scienmag. https://scienmag.com/active-wavefront-control-enables-versatile-micropattern-fabrication-in-laser-interference-lithography/
Denise Maddox. "Active Wavefront Control Enables Versatile Micropattern Fabrication in Laser Interference Lithography." Scienmag, 16 November 2026, https://scienmag.com/active-wavefront-control-enables-versatile-micropattern-fabrication-in-laser-interference-lithography/. Accessed 18 November 2026.
Denise Maddox. "Active Wavefront Control Enables Versatile Micropattern Fabrication in Laser Interference Lithography." Scienmag. November 16, 2026. https://scienmag.com/active-wavefront-control-enables-versatile-micropattern-fabrication-in-laser-interference-lithography/








