Strain engineering allows researchers to alter a material's electronic and optical behavior by changing its physical structure rather than its chemical composition. The approach is well established in crystalline materials, whose ordered atomic lattices respond predictably to deformation. Applying the same principle to amorphous materials is more difficult. Their atoms lack long-range order, and conventional thermal or mechanical treatments can cause crystallization, cracking, elemental segregation, or damage at material interfaces.
As reported in Advanced Photonics , researchers led by Heping Zeng at East China Normal University recently developed a laser-based approach for modifying an amorphous semiconductor while preserving its predominantly disordered structure. The method uses programmable sequences of femtosecond laser pulses to generate localized mechanical stress. The researchers call the approach "digital cold strain engineering," or DCSE.
The method is "digital" because the processing condition is controlled by changing the integer number of sub-pulses within each laser burst. It is described as "cold" because the femtosecond pulses deposit energy faster than it can initially diffuse through the material. This limits early thermal spreading, although it does not mean that the process is entirely free of heat.
The researchers tested DCSE on thin films of amorphous gallium oxide, or a-Ga₂O₃. This wide-bandgap semiconductor is of particular interest for solar-blind ultraviolet photodetectors, which detect UV wavelengths that are largely absorbed by Earth's atmosphere and therefore encounter little background from sunlight at ground level. Zeng remarks, "This is important because it could enable detection of faint UV signals — for example, from flames or electrical discharges — even in daylight."
When the focused laser strikes the gallium oxide surface, it produces a dense plasma. The plasma's rapid expansion generates a transient pressure wave in the material. The laser system delivers bursts at a kilohertz-scale external repetition rate, but the individual sub-pulses within each burst are separated at a frequency of approximately 1.08 GHz. At this spacing, the pressure response from one sub-pulse has not fully relaxed before the next arrives. The responses can therefore overlap and accumulate.
At constant average laser power, increasing the number of sub-pulses reduces the energy carried by each one. Dividing the burst into lower-energy interactions helps limit the severity of each event and suppress pronounced crater-forming ablation. At the same time, the closely spaced sequence retains a cumulative pressure effect. This balance allows the laser to act not only as an energy source but also as a localized mechanical processing tool.
The team used several complementary techniques to evaluate the treated films. Grazing-incidence X-ray diffraction found a broad diffuse pattern without detectable sharp diffraction peaks, consistent with preservation of the predominantly amorphous structure. Atomic force microscopy indicated that the treated surface was about 0.17 nm smoother, with no pronounced crater-like features in the examined regions. Ellipsometry indicated a film-thickness decrease of about 13.6 nm, with cross-sectional electron microscopy independently supporting the reduction trend.
The treatment also changed the film's optical and electronic characteristics. Its refractive index at 587.3 nm increased from 1.853 to 1.859. X-ray photoelectron spectroscopy showed shifts in the gallium and oxygen binding energies, indicating changes in the local chemical and electronic environment. The ultraviolet absorption peak shifted from 263 to 256 nm. Together, the measurements suggest that the laser treatment modified the amorphous network and its near-surface electronic environment without producing crystallization detectable by the methods used.
To determine whether these changes affected device performance, the researchers fabricated solar-blind UV photodetectors from treated and untreated films. Under 254 nm illumination at 90 μW/cm² and a bias of 1 V, the treated device produced a photocurrent of approximately 12.9 nA, compared with 8.2 nA for the reference device, an increase of about 57 percent.
Responsivity increased by about 58 percent, from approximately 0.080 to 0.126 mA/W. Specific detectivity increased by about 25 percent, from approximately 4.4 × 10⁹ to 5.5 × 10⁹ Jones. These improvements were accompanied by a moderate increase in dark current and slower response kinetics. The treated device's approximate rise and decay times were 0.5 and 3.2 seconds, compared with 0.2 and 1.5 seconds for the reference device.
The team also varied the number of sub-pulses from 16 to 96 while keeping average laser power constant. Film roughness, thickness, refractive index, binding energy, optical absorption, and photoresponse all changed with the selected sub-pulse condition. The trends were not always monotonic, suggesting competition between the decreasing energy of each sub-pulse and the increasing number of closely spaced interactions.
Zeng notes that important questions remain: "The pressure simulations describe cumulative behavior qualitatively rather than determining the film's absolute stress. The thickness reduction may result from more than one process, including structural modification and limited surface-material removal." Moreover, "Before the method's practical value for manufacturing can be assessed, further research is needed to clarify the coupled effects of laser-processing parameters, plasma dynamics, stress and thermal processes, and material–substrate interactions; develop predictive models for material responses; and evaluate scalability and long-term stability."
Overall, the central advance is a discrete programming variable that links laser-burst structure to measurable changes in the structural, optical, and electronic properties of an amorphous material. The improved photodetector response provides a device-level demonstration of this approach. With further study, the strategy could be explored in other amorphous oxides, phase-change materials, flexible electronics, and photonic coatings.
For details, see the original Gold Open Access article by T. Yuan et al., " Digital cold strain engineering: photon forging via GHz sub-pulse programming ," Adv. Photon. 8(5), 056015 (2026), doi: 10.1117/1.AP.8.5.056015
