In the standard "textbook" description of a Bose-Einstein Condensate (BEC), we imagine a cloud of atoms acting as a single, unified wave. This is the mean-field approximation, often calculated using the Gross-Pitaevskii (GP) equation. It assumes that every particle in the system occupies the exact same quantum state, or "orbital."
While this simplification works for static systems, it often breaks down when the system starts to move, tunnel, or interact strongly. As Jörg Schmiedmayer demonstrated in Conserving symmetries in Bose-Einstein condensate dynamics requires many-body theory, mean-field theories can even fail to satisfy basic conservation laws like angular momentum during dynamics. To capture the true physics, we need a way to track how particles correlate and "fragment" into multiple different states.
The challenge is computational: the number of possible configurations in a many-body system grows exponentially with the number of particles. The solution developed by researchers like Alexej I. Streltsov, Ofir E. Alon, and Lorenz S. Cederbaum is to use time-dependent variational orbitals.
The Intuition: An Adaptive Spotlight
Imagine trying to film a high-speed car chase at night. You could set up thousands of fixed floodlights to illuminate the entire city (a static basis set), but this is incredibly expensive and wasteful. Alternatively, you could use a few high-powered spotlights mounted on gimbals that automatically track the cars as they move (an adaptive basis).
In quantum dynamics, the "cars" are the regions of configuration space where the wave function is actually present. Instead of using a massive, fixed set of mathematical functions to describe the system, the Multiconfigurational Time-Dependent Hartree for Bosons (MCTDHB) method uses a small number of orbitals that change shape over time.
As explained in Exact Quantum Dynamics of a Bosonic Josephson Junction, these orbitals are not chosen by hand. Instead, they follow a variational principle: at every tiny step in time, the equations of motion adjust the shape of the orbitals to ensure they provide the best possible representation of the many-body wave function. This allows the simulation to maintain "numerically exact" accuracy with a fraction of the computational cost.
Capturing Fragmentation and Loss of Coherence
Why does this matter? When a BEC is pushed—for example, by splitting it into two wells of a potential—it doesn't always stay in one piece. In Reduced density matrices and coherence of trapped interacting bosons, the authors show that the system can undergo fragmentation, where the atoms begin to occupy two or more different spatial orbitals.
A single-orbital theory like Gross-Pitaevskii is blind to this. It sees a single density cloud, but it cannot see that the atoms inside that cloud are no longer coherent with one another. By using multiple time-dependent orbitals, MCTDHB can track how coherence is lost. For instance, Axel U. J. Lode and colleagues used this approach in How an interacting many-body system tunnels through a potential barrier to open space to show that tunneling isn't a simple flow, but a complex interference of multiple single-particle processes that lead to fragmentation.
From Averages to Single-Shot Reality
One of the most striking applications of this method involves simulating what an experimentalist actually sees. Standard theories predict the average density of atoms. But in a real lab, as Mark A. Kasevich points out in Single-shot simulations of dynamic quantum many-body systems, a single image of a gas might show random vortices or interference fringes that disappear when you average many images together.
Because the time-dependent orbital approach provides a full many-body wave function, researchers can sample it to produce "single-shot" simulations. This bridges the gap between abstract quantum theory and the noisy, granular reality of experimental data. As Kaspar Sakmann argued in his thesis, these simulations reveal that even the venerable Bose-Hubbard model—the "gold standard" for lattice physics—fails to capture the structural evolution of the orbitals themselves, which he addressed by introducing "time-dependent Wannier functions."
Scaling Beyond Bosons
The power of adaptive basis sets isn't limited to cold atoms. In molecular chemistry, Ove Christiansen and his team have applied similar logic to molecular vibrations. In Efficient time-dependent vibrational coupled cluster computations..., they demonstrate that using time-dependent basis sets allows them to simulate molecules with hundreds of vibrational modes. By letting the basis "breathe" and move with the nuclei, they reduce the computational scaling from exponential to cubic, making it possible to model complex molecular dynamics that were previously unreachable.
Whether modeling a cloud of atoms tunneling through a barrier or a large molecule vibrating, the core insight remains the same: don't force the physics into a static box. By letting the mathematical basis evolve alongside the system, we can capture the intricate correlations that define the quantum many-body world.