Pulsed Rabi oscillations describe how a resonant optical pulse drives a quantum two-level system between its ground and excited states. In the ideal short-pulse limit, the system’s final excitation is determined by the pulse area. But when spontaneous emission occurs during the interaction, an emitted photon can reset the system and allow the remaining pulse to excite it again. This re-excitation changes photon-count statistics and can undermine single-photon generation.
Understanding this distinction matters when evaluating quantum emitters and designing pulsed excitation schemes. The area theorem remains a useful starting point, but it does not by itself describe every emission event or the multiphoton errors that can arise in a real system.
Pulsed Rabi oscillations and the area theorem
A quantum two-level system is modeled with a ground state and an excited state connected by an optical transition. When a resonant pulse drives that transition, the system coherently oscillates between the two states. The amount of evolution is set by the pulse area: the time-integrated strength of the interaction between the pulse’s electric field and the system’s dipole.
For a system that begins in the ground state, the ideal excited-state probability follows a sinusoidal relationship with pulse area. A pulse with area π ideally leaves the system excited, while other pulse areas produce different final populations. This dependence is known as the area theorem.
The theorem assumes the pulse acts before spontaneous emission significantly affects the system. In practical terms, the pulse width must be short compared with the spontaneous-emission lifetime. If that condition is not met, the system may emit while the pulse is still present, and its subsequent evolution no longer follows the simple, uninterrupted Rabi rotation.
How spontaneous emission creates re-excitation
Photon emission changes the system’s state: after emitting, it returns to the ground state. If the driving pulse has not finished, the remaining portion can excite the system again. That creates the possibility of more than one photon being emitted in response to a single pulse.
A useful way to describe this process is through quantum trajectories. Each trajectory represents one possible sequence of events, including the times at which photons are emitted. Between emissions, the system evolves under the pulse. An emission resets it to the ground state, effectively restarting its response to the remaining pulse area.
This picture connects the timing of an emission to the probability of another. An emission early enough in the pulse leaves more driving area available for re-excitation. The resulting photon-count distribution therefore depends not only on total pulse area, but also on pulse duration and shape.
Pulse duration affects single-photon performance
For a short π pulse, the intended outcome is one emitted photon after the system has been excited. However, an emission during the pulse can be followed by another excitation and a second emission. The probability of these multiphoton events increases as the pulse becomes less short relative to the system’s decay time.
The study’s analytical trajectory model estimates emission probabilities in the short-pulse regime and compares them with numerical calculations and measurements from a semiconductor quantum dot. For square pulses, its approximations agreed well with numerical results up to pulse widths on the order of the spontaneous-emission lifetime. The approximation becomes less reliable as pulse duration grows, in part because it neglects cases where multiple emissions occur during the pulse and the influence of spontaneous emission on the system’s driven evolution.
For increasingly long pulses, multiple emissions become possible and the photon statistics tend toward a Poisson distribution. In that regime, the behavior is no longer well described by a simple short-pulse picture.
Why even multiples of π can favor photon pairs
A particularly striking result occurs when the total pulse area is an even multiple of π. Under ideal area-theorem dynamics, the system returns to the ground state at the end of such a pulse. That might suggest little or no emission. But spontaneous emission during the pulse changes the outcome: if a photon is emitted after roughly π of area has interacted, the remaining area can drive the system back toward its excited state, making a second photon likely.
For even-π pulses, the single- and two-photon emission probabilities can both be proportional to pulse duration in the short-pulse limit. Their relative sizes depend on pulse shape; the reported model gives a two-photon probability greater than the one-photon probability for the cases discussed. This illustrates why the area theorem alone cannot predict photon counts when emission and continued driving overlap.
The second-order coherence, commonly written as g2(0), provides a measure of photon-number correlations. For a good single-photon source, it approaches zero as the pulse becomes very short. Around even multiples of π, two-photon emission can instead produce photon bunching and strongly super-Poissonian statistics.
Practical implications for quantum emitters
For engineers and researchers working with quantum-optical devices, the main lesson is to treat pulse area and pulse duration as linked design considerations, not interchangeable controls. A target area may set the intended coherent rotation, while the pulse width relative to the emission lifetime helps determine whether re-excitation creates unwanted additional photons.
- For single-photon operation: short pulses reduce the opportunity for emission followed by re-excitation, improving the single-photon character of the output.
- For multiphoton operation: even-π pulse areas can favor two-photon emission in the short-pulse regime, although the balance depends on pulse shape and duration.
- For model selection: analytical short-pulse estimates are useful within their regime, while longer pulses call for methods that capture multiple emissions and their effect on driven dynamics.
These findings concern quantum two-level systems and photon sources rather than conventional industrial automation. Their broader technical value is a clear demonstration of how an idealized control rule can break down when system dynamics and irreversible events occur on comparable timescales.
Conclusion
Pulsed Rabi oscillations are governed by pulse area only under conditions where the interaction is effectively completed before spontaneous emission intervenes. When the pulse is longer, photon emission can reset the system and enable re-excitation, changing both photon counts and source statistics. Accounting for pulse duration, emission lifetime, and pulse shape is therefore essential when assessing single-photon performance or exploring two-photon generation.