In the absence of strong external electric or magnetic fields, atomic systems of different magnetic sublevels (but otherwise having the same principal and angular momentum quantum numbers) are degenerate in energy. Because of this degeneracy the measurement of the energies of atomic transitions is not enough to get a complete characterization of the quantum state of the system--the magnetic quantum numbers are also needed. The magnetic quantum numbers are related to the spatial orientation of the atom or ion. In the case of pronounced external symmetry, spatial effects can be important. For example excitation in cylindrically symmetric situations can lead to oriented or aligned systems where the magnetic substates with different angular momentum projections have the same energy but different populations. If such a system undergoes a transition which results in a photon or electron emission the emitted radiation will show anisotropic and polarized behavior.
The above situation is relevant in several natural and laboratory environments. Generally cylindrical symmetry applies for cases when atomic excitation takes place in the interaction by a directed beam of charged particles. An astrophysical example is the atomic excitation occurring in solar flares where a plasma made of charged particles (ions and electrons) moves along strongly directed magnetic field lines. Excitation due to directed flow of particles can be observed also in supernova shock waves. The same situation occurs in many of the laboratory experiments where electron or ion beams excite or ionize atoms or ions.
The EBIT capabilities for measuring electron impact ionization, excitation and recombination cross sections have already been demonstrated in several cases. Transitions in highly charged ions usually involve the emission of photons in the X-ray region. This fact offers an obvious choice of the use of X-ray analyzers (usually solid state detectors and crystal spectrometers) for measuring the above mentioned cross sections. In the EBIT device the ions interact with a narrow (about 0.06 mm diameter) beam of electrons. This very well collimated electron beam acts as a quantization axis making the cylindrical symmetry a natural case for electron-ion interactions inside an EBIT machine. Because of this fact care has to be taken in interpreting X-ray line intensities when they are used for obtaining electron-ion interaction cross sections. They can be strongly affected by anisotropic and polarized emission. The latter is most important in the case of crystal spectrometers where the energy dispersion is polarization selective. On the other hand the measurement of the polarization or the angular distribution of the X-ray emission can give information about the magnetic sublevels involved in the electron-ion collision . This information remains hidden in a simple energy dispersive measurement because of the degeneracy of the magnetic sublevels.
We have successfully extended EBIT spectroscopy of highly charged ions into the range of the spectrum by observing forbidden (M1) transitions within the ground term of titanium-like barium and xenon using a scanning grating monochromator . These transitions are of particular interest for their use in the remote monitoring of ion temperatures by Doppler broadening in large future tokamak fusion machines. Because light detected in these experiments arises from transitions between very close lying levels (fine structure), the wavelength is sensitive to small Zeeman shifts which are usually negligible in the X-ray spectra. The availability of advanced reflective and refractive optics for visible light allows one to achieve both high efficiency and high resolution simultaneously. We have recently applied Fabry-Perot interferometry to achieve resolution sufficient to see Doppler-blurred Zeeman broadening from ions in the trap. The present accuracy of our measurement of the wavelength of the visible light is sufficient to reveal large disagreements (~4%) with ab initio Dirac-Fock calculations.
Our UV spectroscopy capabilities are currently being extended into the VUV in order to join up smoothly with our x-ray capabilities and provide continuous coverage from the visible to the x-ray regime.
When an electron collides with an atom or ion, there is a small probability that the electron kicks out another electron, leaving the ion in the next highest charge state (charge q increased by +1). This is called electron-impact ionization and is the dominant process by which atoms and ions become more highly charged. The rate equation is given by:
e- + A(q) → e- + e- + A(q+1).
From energy conservation, it is clear that the initial energy of the incident electron must be larger than the ionization potential of the electron being removed.
Theorists have found electron impact ionization cross sections difficult to calculate from first principles, even for relatively simple systems like hydrogen-like (one electron) and helium-like (two electron) ions. There has been recent progress in developing a phenomenological theory by Dr. Yong-Ki Kim here at NIST.
A simple empirical formula for calculating electron impact ionization cross sections was developed by W. Lotz over 25 years ago. It is not very accurate, but it does give experimentalists a useful qualitative picture.
Radiative recombination is a process which takes place when a positively charged ion captures an electron to one of its bound orbits with a simultaneous emission of a photon:
e- + A(q+) → A(q-1) + photon
In the dielectronic recombination process the energy which becomes available during the capture process is carried away by the promotion of a bound electron to another bound orbit:
e- + A(q+) → A(q-1)** → A(q-1) + photon
In the second step of the dielectronic recombination process a photon is emitted characteristic to the doubly excited state (**) of the q-1 times ionized ion. The dielectronic recombination is a resonant process, because of the discrete energy nature of the bound electron orbits.
Both radiative and the dielectronic recombination are important capture processes which play a dominant role in determining the charge state balance of highly ionized astrophysical and laboratory plasmas.
In a recent investigation carried out with our EBIT , scandium-like and titanium-like barium ions were created, trapped, and excited. X-ray peaks arising from both radiative recombination and dielectronic recombination were studied simultaneously. In the DR process a 2p electron was promoted to the 3d orbital. One of the M-shell electrons of the recombined ion subsequently decayed radiatively to the 2p vacancy, and emitted an x-ray of energy almost twice the incident kinetic energy of the projectile electron. Comparison with theoretical estimates showed a favorable agreement with the data. The theoretical calculations were carried out by the theory group of the University of Connecticut (McLaughlin and Hahn).
In order to have a better understanding of the motion of the highly charged ions inside the trap a computer simulation was developed by Eric Meyer. The program accounts for the presence of the magnetic field of the superconducting coils, the electron beam, and the neutral background gas atoms. It can also simulate the effect of changing electron beam currents which has importance in time resolved (e.g., lifetime) measurements. Here are a few examples of different orbits.
The charge state balance (e.g., the population of different charge states) inside the EBIT is determined by the balance between the different ionization and recombination processes. In order to predict the charge states of the ions inside the trap and to help to determine the EBIT parameters for the optimalization for the production of a certain charge state these processes should be taken into account. To this end a computer program which was originally developed by Penetrante in Livermore and modified by Margolis at Oxford was installed and further modified. The predictions of the simulation can be checked by analyzing the x-rays emitted from the trap. An alternative way to look into the same problem is the extraction and charge to mass ratio analysis of the highly charged ions. Experimental efforts in this direction are under way.
The measurement of excited-state lifetimes is complementary to measuring transition wavelengths as a way of studying atomic structure. Although the lifetimes are determined by the same wavefunctions as the energy levels the measurements of the atomic decays carry different information since they are sensitive to the long-range behavior of the wavefunctions. The knowledge of the lifetimes also has important practical applications. They are critical in the density diagnostics of laboratory and astrophysical plasmas.
The principle of measuring lifetimes with an EBIT lies in the periodic fast switching of different voltages in the machine. Since the ions are created and excited with the same beam of electrons, by changing the electron beam energy one can selectively exclude certain levels from being excited. This can simply be done by setting the electron beam energy below the excitation threshold of the level to be excluded. Without further excitation the time dependence of the emitted photon signal carries the information about the lifetime of the level. After a certain period of time (determined by the lifetime of the level) the electron beam energy is set to be above the excitation threshold to repopulate the level and repeat the sequence. An alternative method for measuring lifetimes with an EBIT is to switch off the electron beam completely, take data, and turn the beam back again to re-excite the ions in the trap. While the electron beam is off, the ions remain trapped by the magnetic field. The lifetime range that can be measured with an EBIT is determined by the capabilities for the fast switching of voltages. In principle the 10 ns to 10 ms lifetime range can be addressed by this method. Since this lifetime range is only partially covered by other methods the EBIT is a unique tool for measuring the lifetime of long living metastable levels.
In a recent experiment we have measured the lifetime of a visible light emitting metastable level . The transition takes places within the ground state configuration of titanium-like ions. The measured lifetimes fall into the millisecond range.
The EBIT was originally developed for in situ spectroscopic measurements of highly charged ions. It soon turned out, however, that the machine is easily capable of producing the highest charge states not accessible even for sophisticated ion sources. To use the EBIT as a source for ions an extraction system has to be built which removes the ions from the trap region and transports them outside the machine. The construction of the extraction system on the NIST EBIT was completed in the first half of 1995 [12,20]. When the ions are extracted from the EBIT they pass several ion beam steering and shaping ion-optical elements. A bending magnet is available as a part of the system to charge-to-mass select the ions and direct them into experimental chambers. Recent experiments with the extracted ions involve the study of nanometer scale damages of insulator surfaces caused by the impact of highly charged ions. X-ray  and Electron spectroscopic investigations  of metal-surface -- highly-charged-ion interactions are under way.
Both the atomic spectroscopy and ion-surface interactions component of our work is proceeding in connection with International SEMATECH.