We report the first observation of ionization of Rydberg atoms by subpicosecond, circularly polarized THz radiation. The field amplitude in these pulses is non-negligible for only one-quarter of an optical cycle. The experiment is performed in the short-pulse regime, where the duration of the ionizing pulse is shorter than the classical Kepler period of the Rydberg electron. We find that the ionization probability for these atoms is remarkably insensitive to the time-varying polarization of the THz field.Over the years, multiphoton ionization of atoms has been recognized as a process of fundamental importance for understanding the dynamic interaction between atoms and intense radiation fields. Very recently, a number of in vestigations have led to the discovery of novel ionization dynamics during the interaction of highly excited Rydberg atoms with strong, subpicosecond pulses of THz radiation [1][2][3][4][5][6][7]. The unipolar electric field in these pulses resembles one-half of an optical cycle of an oscillating electric field so they are commonly referred to as "half-cycle" pulses (HCPs). To date, experimental and theoretical HCP stud ies have considered only linearly polarized fields. In this configuration, HCPs provide an impulsive unidirectional "kick" that can literally push a Rydberg electron off an atom [1][2][3]8]. Although the ionization dynamics are now relatively well understood for unidirectional fields, previ ous work with multi-cycle laser [9] and microwave pulses [10], as well as classical intuition, suggests that ionization by subpicosecond far-infrared pulses might proceed very differently with circularly polarized radiation.In this Letter, we describe the results of the first experi ment on the ionization of atoms by subpicosecond circu larly polarized THz pulses. The duration of these pulses is so brief that the electric field vector rotates by only 90 ± during the pulse. In the language of classical physics, an electron exposed to this "quarter-cycle" circularly po larized pulse is subjected to a rapidly rotating force in contrast to a unidirectional kick. Therefore, one should expect to observe interesting new ionization dynamics us ing these novel pulses.The interaction of atoms with these pulses is also in teresting from the point of view of collision physics. A quarter-cycle field pulse bears a strong resemblance to the time-dependent field seen by an atom undergoing a colli sion with a charged particle, as shown explicitly in Fig. 1. In each case, the field can be described as a combination of a half-cycle cosine pulse along one axis (the transverse collision field) and a "single-cycle" sine component along an orthogonal axis (the longitudinal collision field). In contrast to true collision experiments, the relative veloc ity, impact parameter, and orientation of the colliding par ticles can be precisely controlled in a "mock collision" by adjusting the duration, field amplitude, and polarization of the quarter-cycle field pulse.It is well known that circularly polarized radiation can...

Unipolar "half-cycle" electric field pulses (HCPs) have been used to recombine free electrons and calcium ions. The field assisted process is very similar to controlled three-body recombination in plasmas. We report on experiments that utilize HCP assisted recombination to probe the probability distribution of continuum electron wave packets and produce bound wave packets that are highly localized in three spatial dimensions.The combination of free electrons and ions to form neutral atoms is a complicated process that can proceed through a variety of different mechanisms including dielectronic, radiative, or three-body recombination [1]. Al though these mechanisms are quite different in detail, in general, recombination is made possible through the trans fer of energy and momentum from the free electron to a third body. The recombination process is inherently time dependent, and the capture of a free electron can result in the formation of a complicated coherent superposition state within each atom. Because of the random nature of electron-ion scattering events in the laboratory, the bound wave packet produced via recombination varies widely from one atom to another due to the different scattering conditions through which each was formed. Consequently, standard theoretical and experimental studies parametrize the process using time-independent recombination rates and branching ratios for capture into various excited states of the atom [1].The experiments described in this Letter are aimed at studying controlled electron-ion recombination in the time domain. The results of these experiments provide new in formation on coherent recombination processes as well as demonstrate the utility of field assisted recombination for probing continuum electron dynamics and producing novel bound wave packets. Specifically, subpicosecond "half cycle" field pulses (HCPs) [2,3] have been used to assist the recombination of a well-characterized continuum wave packet with its parent ion. Although the recombination can be formally classified as radiatively assisted [1], the unique nature of the unipolar field pulse makes the process more closely related to three-body recombination. To make the analogy with three-body recombination the HCP field is equated with the transverse field produced by a passing ion or electron. The impulse provided by the field [4,5] extracts momentum and energy from the free electron, fa cilitating its capture by an ion.In the experiments, a 1.5 psec laser pulse photoionizes a tightly bound electron in calcium, producing a continuum wave packet which travels away from the Ca 1 ion in the form of a thin shell. After the ionizing laser pulse, the ions and free electrons are exposed to a HCP whose duration is so short that the continuum wave packet is essentially frozen during the pulse. The electron receives a R momentum "kick" or impulse, A � � 2 F � HCP �t� dt, from the HCP [4,5], where F � HCP �t� is the HCP field amplitude. All equations are given in atomic units unless otherwise noted. The impulse can hal...

Rydberg atoms are ionized by nearly unipolar, subpicosecond electromagnetic pulses. Deviations from a perfectly unidirectional pulse are found to alter substantially the ionization probability as a func tion of peak field. Quantitative agreement between classical theory and experiment is achieved if the pulse imperfections are significantly attenuated.In recent experiments [1,2], Rydberg atoms have been exposed to ultrashort, electric-field pulses. The electric field in these pulses is nearly unipolar, and has a temporal shape which resembles one-half of a cycle of THz-band radiation [3]. In the experiments [1,2], the 0.5-psec dura tion of these "half-cycle pulses" (HCPs), 1"HCP' is shorter than or comparable to the classical Kepler period of the Rydberg states, 1"K =21Tn 3. The dynamics of the ioniza tion process in these experiments is distinctly different from that in either a long field pulse ('Tpulse> 'TK) [4] or a short laser pulse (1"laser < 1"K ) [5].In a long electric-field pulse, essentially no energy is transferred to the Rydberg electron. Instead, ionization occurs due to the modification of the electronic binding potential [4]. In a HCP, the electron is unable to respond to the rapid changes in the binding potential, and the electron must gain energy from the field in order to es cape the Coulomb attraction of the nucleus. Classically speaking, the electron receives an energy "kick" or im pulse from the rapidly changing field [1],-00where F(t) is the HCP field and v(t) is the velocity of the electron. Therefore the ionization probability is strongly dependent on the velocity or momentum distribution of the initial-state wave function as well as the temporal shape of the electric-field pulse. The time dependence of the electric field is extremely important in the ionization of Rydberg atoms where 1"pulse «1"K· In a Rydberg atom, the electron probability distribution is peaked far from the ion core, and the prob ability for finding the electron near the nucleus during the pulse is 1"pulse l 'TK. Therefore, in order to have an ion ization probability greater than 1"pulse l 1"K' energy transfer between the electron and the time-dependent field must occur far from the ion core where the electron is essen tially free. Equation (1) clearly shows that a free electron can gain energy from a unipolar field pulse. However, this electron cannot gain energy from a pulse whose time-integrated electric field is zero. Indeed, Rydberg atoms may be ionized with 100% efficiency by a HCP, but not by a short laser pulse [1,5].The results of the ionization experiments [1] are in qualitative agreement with classical simulations [1,6].The agreement is quantitative if the experimental field values are rescaled by a multiplicative factor of 2.5. A more recent field calibration suggests that the field discrepancy is actually only a factor of 1.6. A description of the differences between the two calibration techniques is given later in the paper. A possible source for the per sistent disagreement between experiment and theory...

An approach to experimentally measuring the speed of a moving object by direct application of the Doppler effect for sound is discussed. The method presented here uses a Windows computer and sound card to record Doppler shifted sound from a moving source. This sound card approach allows for direct acquisition of Doppler shifted sound intensity as a function of time, affording much analytical and pedagogical freedom in undergraduate lab instruction. In addition, the acquisition of such data allows for the experimental study of not only constant velocity sound sources, but of accelerated sound sources as well.

The oscillation between bound-state configurations in a rapidly autoionizing three-body Coulomb system has been directly observed. Using a 500-fsec laser pulse, calcium atoms are excited to the pure 4p 3/2 15d twoelectron configuration at an energy greater than 3 eV above the ionization limit. As a result of configuration interaction, the electrons scatter coherently into multiple bound and continuum configurations. The oscillation between the degenerate 4p 1/2 n�d and 4p 3/2 nd modes as well as autoionization into 4s 1/2 �l , 3d 3/2 �l , and 3d 5/2 �l continua are observed explicitly using bound-state interferometry. The measured time dependence of the 4p 3/2 15d character is in excellent agreement with the Fourier transform of the frequency domain excitation cross section. To our knowledge, this is the first experimental demonstration of the equivalence of time and frequency domain spectra in a multiconfigurational system involving bound and continuum channels.

[1] Nighttime water-leaving radiance is a function of the depth-dependent distribution of both the in situ bioluminescence emissions and the absorption and scattering properties of the water. The vertical distributions of these parameters were used as inputs for a modified one-dimensional radiative transfer model to solve for spectral bioluminescence waterleaving radiance from prescribed depths of the water column. Variation in the waterleaving radiance was consistent with local episodic physical forcing events, with tidal forcing, terrestrial runoff, particulate accumulation, and biological responses influencing the shorter timescale dynamics. There was a >90 nm shift in the peak water-leaving radiance from blue ($474 nm) to green as light propagated to the surface. In addition to clues in ecosystem responses to physical forcing, the temporal dynamics in intensity and spectral quality of water-leaving radiance provide suitable ranges for assessing detection. This may provide the information needed to estimate the depth of internal light sources in the ocean, which is discussed in part 2 of this paper.

Picosecond laser pulses have been used to produce Rydberg wave packets in calcium atoms in the presence of a strong static electric field. The dynamics of the Stark wave packets have been observed by measuring the momentum-space probability distribution as a function of time. The full precession of the electronic orbital angular momentum, the appearance of a large-amplitude, linear oscillation of the electronic dipole moment, and a pronounced, periodic up-down asymmetry in the momentum distribution are all observed directly.A large amount of theoretical and experimental work on Rydberg electron dynamics in a variety of different circum stances has been performed �1�. However, until very re cently, the lack of good experimental techniques to monitor the full time-dependent probability distribution of wave packets has severely limited the insight gained from experi mental data alone, without the aid of theoretical simulations. Nevertheless, comparisons between the results of experi ments and theory have shown that it is possible to produce well-controlled wave packets under a variety of different conditions �1�, and that with alternative methods it is pos sible to experimentally recover their full time-dependent probability distribution with high fidelity �2-4�. In fact, one can now perform dynamics spectroscopy on uncharacterized wave packets. The full electronic motion as viewed in an experiment can be used to interpret the physics behind the motion directly, without relying on theoretical simulations. The refinement of this approach is a necessary prerequisite to controlling wave-packet motion in complex systems where complete quantum mechanical calculations are not readily available.The results presented in this Rapid Communication pro vide a complete experimental view of the time-dependent dynamics of a Rydberg wave packet in combined Coulomb and uniform static electric fields. Although numerous experi mental studies of Stark wave packets have been performed over the last decade �5-7�, this paper describes experiments where the complicated multidimensional evolution of the wave packet can be seen directly. Specifically, the precession of the electronic orbital angular momentum, strong oscilla tions of the electronic dipole moment along the static field direction, and a periodic asymmetry in the momentum distri bution along the static field axis are all linked in the motion of the wave packet and are clearly identified in the measured probability distributions.In the experiment, ground-state 4s4s 1 S 0 Ca atoms in a thermal beam are promoted to an intermediate 4s4 p 1 P 1 level using a 5-nsec dye laser pulse. A 1.5-psec laser pulse then drives a fraction of the excited-state atoms into a 4snd 1 D 2 , 26�n�30 radial Rydberg wave packet �3�. The wave packet is initially localized near the Ca � ion core but immediately propagates radially outward, reflects from the Coulomb potential, and returns to the ion core after one Ke pler period, � K �2�N 3 �3.0 psec �8�. In the absence of the static field, the wave packet...

Seawater has been irradiated using a train of 70 ns flashes from a 440 nm laser source. This wavelength is on resonance with the blue absorption peak of Chlorophyll pigment associated with the photosystem of in vitro phytoplankton. The resulting fluorescence at 685 nm is instantaneously recorded during each laser pulse using a streak camera. Delayed fluorescence is observed, yielding clues about initiation of the photosynthetic process on a nanosecond time scale. Further data processing allows for determination of the functional absorption cross section, found to be 0:0095 Å 2 , which is the first reporting of this num ber for in vitro phytoplankton. Unlike other flash-pump studies of Chlorophyll, using a LED or flashlamp based sources, the short laser pulse used here does not reveal any pulse-to-pulse hysteresis (i.e., variable fluorescence), indicating that the laser pulses used here are not able to drive the photosynthetic process to completion. This is attributed to competition from a back reaction between the photoexcited photo system II and the intermediate electron acceptor. The significance of this work as a new type of deploy able ocean fluorimeter is discussed, and it is believed the apparatus will have applications in thin-layer phytoplankton research.