“Generating ever-shorter and brighter light pulses is a central goal of ultrafast science, enabling coherent control and observation of electron dynamics on their natural timescale,” writes Meng Han, assistant Professor at Kansas State University’s Department of Physics, in his recent publication1. This vision has now been realized in a remarkable way: the generation of 18-attosecond light pulses, representing a major milestone in attosecond science.
This achievement advances beyond the 43-attosecond pulses demonstrated in 2017 by Hans Jakob Wörner’s group at ETH Zurich, which were generated using a Ti:Sapphire laser system featuring Light Conversion’s TOPAS-PRIME-HE optical parametric amplifier2. This time, the breakthrough was achieved with a system powered by Light Conversion’s Yb-based PHAROS femtosecond laser and the CASCADE (Cascaded focus and compression) setup, introduced by Prof. Ming-Chang Chen and collaborators at National Tsing Hua University (Hsinchu, Taiwan)3. In essence, CASCADE employs multiple noble gas-filled cells, where the laser pulses are focused to nonlinearly generate a broader bandwidth and subsequently compressed using chirped mirrors. The resulting 3.7 fs pulses from this setup are then directed into the attosecond streaking beamline built by Prof. Han.
Prof. Han focuses his research on experimental atomic, molecular, and optical physics, including CEP-stabilized near-single-cycle laser technology, short isolated attosecond pulse generation, characterization, and applications. In this interview, he reflects on the collaboration that made the recent achievement possible, the challenges of working on the edge of what lasers can deliver, and why the field of attosecond science continues to be a place where researchers are quite literally pushing physics to its limits.
How did your collaboration with Prof. Ming-Chang Chen begin?
I first met Prof. Ming-Chang Chen at the ATTO conference in Orlando in 2022, when I was a postdoc presenting my poster on circularly polarized attosecond metrology. He was very interested in my work, and we had a long conversation about the future of attosecond science. Ming-Chang also showed me his impressive pulse compression results, and we both shared the belief that Yb-based lasers would become the main workhorse in the field. Later, when I started my own lab at Kansas State University, I decided to invest my limited resources into Yb amplifiers – a bit of a gamble for a new assistant professor 😉. Ming-Chang and his student, Ming-Shian, were incredibly supportive and helped us a lot during the early days of setting up the lab.
The previous record for the shortest laser pulse stood for nine years. Why did it take so long to surpass it?
In short, generating such an ultrashort attosecond pulse requires three key ingredients: a very broad spectral bandwidth, precise control of the spectral phase, and a bright, stable flux suitable for characterization experiments such as attosecond streaking. Back in 2017, two pioneering groups – Zenghu Chang’s group at UCF and Hans Jakob Wörner’s group at ETH – demonstrated attosecond streaking using OPA or OPCPA lasers centered around 1.8 μm. Since then, progress has been limited by the technical challenges of simultaneously achieving all three requirements at the necessary level of performance, which explains why it took nearly nine years to break the record.
Your pulses reached the “atomic unit” timescale, defined as the time required for a 1S electron to travel one Bohr radius. How does this classical definition of time and distance match the fact that orbitals are density of probability? Could you explain its relevance?
Strictly speaking, a 1S electron in a stationary state has no motion in the quantum sense – it’s described by a probability density rather than a classical trajectory. However, once the 1S electron is partially excited to another state, we create a superposition of two states. This superposition gives rise to electron dynamics, where the electron density evolves in time. The larger the energy gap between the states, the faster these dynamics occur. One atomic unit of time – about 24.2 attoseconds – represents the natural timescale of the fastest possible valence-electron motion across all atoms, which is why it is such a meaningful benchmark.
In your article, you describe the various laser sources used to generate the shortest possible pulses. Ti:Sapphire lasers still produce shorter and more energetic pulses than Yb lasers, so one would think that they are more suitable for generating these short pulses. What is the advantage of a Yb laser? Is it mostly the gain in cut-off because of the longer pump wavelength compared with a Ti:Sapphire laser amplifier?
Compared to Ti:Sapphire lasers, the main advantage of Yb lasers for the attosecond community is their exceptional stability. With post-compression techniques, Yb systems can also deliver very short pulses. For pulse durations in the 20–50 fs range, I would say Ti:Sapphire lasers remain highly competitive, since they can also be made stable in that regime and, more importantly, the pulse contrast is very clean. But when it comes to generating few-cycle pulses, Yb lasers have shown clear advantages over Ti:Sapphire, combining stability with scalability. That’s why they are becoming increasingly attractive for pushing attosecond science forward.
Is 320 eV the limit for 1 µm drivers? Would increasing pulse energy help surpass this? Or achieving even shorter pulses from an MPC?
I wouldn’t say 320 eV is necessarily the limit for 1-µm drivers. In our experiment, we only used about 0.8 mJ to generate the attosecond pulses, so there is certainly room to explore higher photon energies by increasing the driver pulse energy. It would be very interesting to repeat these experiments with higher-energy, shorter-pulse Yb drivers to see how far the cutoff can be pushed, and whether even shorter pulses can be achieved.
How robust is your attosecond setup – both in maintaining short pulses during extended experiments and when starting from a cold start?
Our attosecond streaking setup is very robust – it can run continuously for several weeks, even months, without interruption. From a cold start, it typically takes just a few hours to warm up the laser and bring all the instruments into optimal condition for data collection.
What would be the most compelling next experiment you or other groups should pursue using your setup with these pulses?
There are many exciting directions to explore with these pulses. One example would be angle-resolved attosecond streaking, which could allow us to measure partial-wave–resolved photoionization time delays. Another example would be tabletop attosecond-resolved inner-shell electron spectroscopy4. This would provide unprecedented insight into electron dynamics on their natural timescale.
What further developments would you like to see in Yb amplifiers?
Personally, I would like to see Yb amplifiers that can deliver higher single-pulse energies while maintaining the current pulse duration, for example, on the order of 10 mJ at 200 fs. Such systems would open exciting possibilities for scaling attosecond pulse generation.
The result achieved by Meng Han and his collaborators is not an endpoint but another step in a continuous journey. Extreme light generation has become one of the most dynamic frontiers of modern physics, with each advance opening new ways to probe and control the fastest processes in nature. Here, Yb-based femtosecond lasers play an increasingly central role. Thanks to their stability, scalability, and versatility, these lasers now power many of the most ambitious experiments in the field.
References
1. J. Gao, M. Hasan, H. Liang, Z. Guo, C. Zhang, M. Tsai, Y. Yuan, Z. Eisenhutt, C. H. Keitel, C. Lin, C. Jin, Y. Liu, M. Chen, M. Han, Bright Isolated Light Pulses Reaching 18 Attoseconds from High-Order Harmonic Generation by Post-Compressed Yb Lasers, Nat Commun 17 (2026).
2. T. Gaumnitz, A. Jain, Y. Pertot, M. Huppert, I. Jordan, F. Ardana-Lamas, H. J. Wörner, Streaking of 43-Attosecond Soft-X-Ray Pulses Generated by a Passively CEP-Stable Mid-Infrared Driver, Opt. Express 25, 27506-27518 (2017).
3. M. Tsai, A. Liang, C. Tsai, P. Lai, M. Lin, M. Chen, Nonlinear Compression Towards High-Energy Single-Cycle Pulses by Cascaded Focus and Compression, Sci. Adv. 8, eabo 1945 (2022).
4. M. Hasan, J. Gao, H. Liang, Y. Yuan, Z. Eisenhutt, M. Tsai, M. Chen, H. J. Wörner, A. Rudenko, M. Han, Time-Domain Measurement of Auger Electron Dynamics in Xenon and Krypton Atoms After Giant Resonance Photoionization, Phys. Rev. Lett. 136, 163201 (2026).
For further information, please contact