In 1954, a physicist named R.H. Dicke predicted something that sounded almost too theatrical to be real: if you excite a collection of atoms and let them radiate together, their spontaneous emission doesn't just add up — it synchronizes, amplifying the light far beyond what any individual atom could produce. The effect scales with the number of atoms squared, a signature of quantum cooperation that physicists had been chasing in gases and isolated systems for seven decades.
Now a team at DGIST, the Daegu Gyeongbuk Institute of Science and Technology in South Korea, says it has confirmed Dicke's prediction in a solid-state material for the first time, using a layered semiconductor called MoS₂ and a computational framework borrowed from open quantum systems theory. The result, published in Advanced Science in March 2026, identifies the microscopic mechanism behind an ultrafast decoherence problem that had puzzled HHG researchers for more than a decade.
The core finding is a form of destructive interference. When intense laser light drives electrons in the MoS₂ lattice, two radiation processes compete simultaneously. One is the Dicke superradiance, where excited electrons radiate in lockstep. The other is a broadband emission from individual electron scattering events that mimics blackbody radiation. The DGIST team, led by Gimin Bae, Youngjae Kim, and Jae Dong Lee, showed that these two processes cancel each other out, shortening the effective electron scattering time to a few femtoseconds and explaining why coherent radiation in solid-state HHG decays so fast.
The team used the Lindblad master equation, a standard tool in quantum optics for modeling systems that leak energy to their environment, combined with a one-dimensional Hubbard model describing electron interactions in the solid. This approach goes beyond traditional quantum master equations, which typically assume Markovian dynamics, meaning the environment has no memory. The Lindblad formulation lets the model capture non-Markovian effects, where the environment's state evolves alongside the system it is disturbing.
"We find a strong destructive interference between the Dicke superradiance and the broadband emission, which makes a scale down of the effective electron scattering time and leads to just few-femtosecond dephasing time T2," the authors write in the paper. T2, the decoherence time that sets how long quantum information survives in the material, is among the most operationally important numbers in any solid-state quantum system.
The decoherence problem in HHG is not abstract. When a femtosecond laser pulse drives electrons in a crystal lattice, those electrons should emit coherent radiation at integer multiples of the driving frequency, the harmonic generation that makes HHG useful for generating extreme ultraviolet light and probing material properties. But the coherence collapses in roughly one to two femtoseconds, too fast for many applications and too fast to explain with conventional scattering theory. The DGIST result does not solve that limitation. It explains why it exists.
The solid-state Dicke superradiance finding is the more striking claim. Dicke's original 1954 paper, published in Physical Review, described cooperative emission in atomic gases. Observing it in a crystalline solid required conditions the authors had to engineer: a small interatomic spacing that enhances the cooperative coupling, and a pumping frequency in the mid-infrared at 0.26 electronvolts that puts the electron dynamics on the same timescale as electron-electron scattering. Under those conditions, the superradiance intensity scaled as N², where N is the number of atoms per unit length, exactly as Dicke predicted.
For quantum technologists, the result sits somewhere between fundamental and applicable. Dephasing mechanisms in solid-state systems matter for any hardware that tries to store or manipulate quantum information in a material: superconducting qubits, spin qubits in silicon, valley qubits in 2D materials. If the T2 bottleneck in HHG is driven by a general interference mechanism between cooperative and incoherent emission channels, the same physics likely applies wherever electrons in a solid interact with their environment. That makes the result a candidate explanation for decoherence across a category of systems, though the paper does not demonstrate control over the mechanism.
The press release accompanying the paper is less restrained. "This study opens a pathway to connect ideal quantum theory to practical and reliable quantum engineering," it reads, framing the result as a bridge from textbook physics to real technology. That is a longer leap than the paper makes. DGIST has identified and modeled a mechanism. Whether it leads to engineered solutions for quantum information systems is a different question the paper does not answer. Longer coherence times, better material designs, smarter control pulses — those are the targets the press release implies, not what the paper delivers. The distinction between explaining a phenomenon and engineering around it matters, and the paper is careful about it even if the press release is not.
One thing the paper does leave open is the practical regime. The effective blackbody temperature fitted to the broadband emission was around 3,800 Kelvin, roughly the surface temperature of a small star, an artifact of the modeling not a real condition in the experiment. The Lindblad approach and 1D Hubbard model are computational approximations, and the authors acknowledge their simplifications. Their code is available on GitHub under a Creative Commons license, which means other groups can test whether the mechanism holds in different materials or under different excitation conditions.
The result also raises a question the paper does not fully address: whether the Dicke superradiance signature is unique to MoS₂ or appears broadly in correlated electron materials. The parameter regime the team identified, weakly to moderately correlated electrons with U/t = 1 to 2 in Hubbard notation, includes a range of transition metal dichalcogenides and other layered materials. If the effect is general, it would be a feature of ultrafast electron dynamics across an entire class of quantum materials, not a curiosity of one semiconductor.
What is true at the end of this is narrower than what the headline promised. DGIST has produced a computational model that identifies a plausible microscopic driver of ultrafast decoherence in solid-state HHG, and it has made the first claim of solid-state Dicke superradiance. Whether that changes how anyone builds a quantum computer or designs a material is a question the paper does not resolve, but it is the right question to start asking.
