Work under the heading of Laboratory Plasma Spectroscopy may be conveniently separated into three classes depending on the extent to which the interaction of the emitting atoms with their plasma environment is central to the investigation. Zero order, the longest established use of laboratory plasmas in connection with astrophysics, concerns the use of hot plasmas for the excitation, measurement, and identification of the spectra of highly-stripped ions. In such work the properties of the plasma itself are usually of secondary importance. In first-order, plasma spectroscopy is used to determine fundamental atomic data concerned with the interaction of an atom with a single particle, usually either a photon or an electron, i.e.: the determination of oscillator strengths and collision cross-sections. Finally, higher-order processes in which the plasma nature of the surrounding medium is most relevant concern the study of line-shapes, and related topics such as the excitation of satellite spectral features by plasma oscillations. Developments in plasma diagnostic techniques in the last five years have greatly extended the scope of the second and third categories and have yielded much astrophysically important information from laboratory studies. Recent advances in these areas are reviewed.

In a recent T-tube experiment, Greig et al. (1970) reported the observation of a blue shift of ~0.05 Å for the first member of the resonance series of ionized helium. This was explained as due to some polarization effect in the plasma (Berg et al., 1962; Griem, 1966) and a classical estimate allowing for quantum mechanical effects was presented.

Absorption spectra of light elements were observed in the vacuum ultraviolet with an original technique described in an earlier paper (Mehlman-Balloffet and Esteva, 1969). The method utilizes a two-vacuum spark mounting: one of the sparks is emitting the continuous background, the other one generates the absorbing plasma. Several light element have been successsively introduced in the spark anode. For all of them new autoionizing levels have been observed in Rydberg series of resonances exhibiting the asymmetric ‘Beutler-Fano’ profile.

In the beryllium and magnesium spectra three new series corresponding to two-electron excitation process have been identified while for boron, carbon, nitrogen, aluminium and silicon the resonances observed correspond to single subshell electron excitation such as: 2s22p2P0 → 2s2p(3P0)np2De for the case of boron.

All these series lie in the photoionization continuum of the absorbing atomic species and usually between the first and second ionization limit. This means that they were observed with a normal incidence grating spectrograph in the spectral range 500–1500 Å. In the extreme ultraviolet some other transitions involving inner-shell electron excitation were observed. In the beryllium spectra a series lying between 100 and 110 Å was identified while, the magnesium spectra exhibited only isolated resonances in the 220-265 Å range together with an inner-shell 2p electron photoioniation continuum.

A complete description of experimental results with numerical data is being submitted for publication (Esteva and Mehlman-Balloffet, 1972).

An essential prescription for a review paper prepared for a conference is that it treat an area of currently flourishing research – a requirement directly conflicting with limitations of time and space. Under the present title, a large output of diverse and enterprising experimental endeavour has indeed appeared in the three years since the Maryland meeting, so that the reviewer is forced to be very selective. Fortunately, the situation is alleviated by the inclusion in Dr Burgess’s review of ‘Plasma Spectroscopy’ of much material describable as ‘fundamental data’ – such as f-value measurements obtained from plasma sources, shock-tubes and arcs. Again little reference is needed here to the important and extensive work on line identifications and classifications in spectra of medium to high ionization stages, since these will be covered by Dr C. Jordan’s review and several contributed papers.

An experiment for the determination of the absorption spectrum of S i was performed using a flash-pyrolysis technique.

The material (pure Sulphur) was introduced by vacuum deposition into the absorption cell which was surrounded by a helical flash-lamp. A condensor bank of 120 μF charged to 10 kV was discharged through the flash-lamp. The absorption of the strong flux of light results in the evaporation of the material and in its dissociation. With delays of 100–500 μsec a Garton source was fired providing the background continuum.

Widths of Stark broadened Ni I Lines in the spectral range of 3000 Å–4000 Å have been measured using a gas driven shock tube as a spectroscopic source. Electron densities were determined using measured pressure and temperature data. Within the density range 6 × 1016 cm-3 to 12 × 1016 cm-3 the observed widths were linear functions of the electron density. There was no apparent correlation of observed widths with neutral density. Van der Walls broadening was presumed to be negligible because we observed narrow lines which should have had the same Van der Walls widths as computed with a classical theory as some of the wide lines we observed. In this way we were able to attribute almost all of the observed broadening to the Stark effect. The experimental results were compared with results of calculations of Stark broadening parameters using the program developed by Cooper and Oertel. The measured widths were about a factor three larger than the theoretical predictions. A part of this discrepancy is probably due to the fact that only LS allowed interactions were considered in the broadening calculations and possible incomplete classifications.

New Spectra of highly ionized atoms of the fifth period are to be published soon. Selected energy levels obtained with 104 A peak current vacuum sparks are presented in Table I.

A discussion is presented of some theoretical methods for predicting atomic transition wavelengths and transition probabilities that are especially useful for highly stripped ions. Particular attention is given to procedures involving model potentials and pseudo-potentials and to procedures involving expansions in inverse powers of the nuclear charge. The importance of relativistic effects is assessed. The results of some recent calculations are compared to experimental data. Theoretical studies of molecular absorption and emission are briefly summarized.

In the spectra of highly ionized atoms, lines corresponding to different degrees of ionization appear simultaneously on the same region of the photographic plate. Many difficulties are encountered in experimental assignment of the lines to the ionization they correspond to. As a result of this there is a special significance in exact calculations of energy levels. An accurate determination by some theoretical procedure can solve the problem of ionization differentiation.

Throughout this work we present theoretical calculations of energy levels of d and p electrons for ions of the fourth row in the isoelectronic sequences of Li i, Be i. We have also calculated the splitting of the various levels.

The technique for absolute calibration using the branching ratio of two transitions which have a common upper level but widely separated energies has been employed and described by Griffin and McWhirter (1961).

Using the plasma produced by irradiation of a carbon target with a 1 GW, 6 nsec duration, neodymium laser pulse, we have been able to calibrate a grazing incidence spectrograph at the wavelength (25.83 Å) of the 7–1 transition in C VI by direct comparison with the intensity of the 7–6 (3434 Å) transition emitted from the same ions in the plasma. The spectrum from a similar carbon plasma has been described by Boland et al. (1968) and in several ways is ideal for this type of measurement.

A normal pressure arc having local thermal equilibrium, is a useful source for measurements of atomic parameters in the vacuum ultraviolet region (Boldt, 1970). However, it imposes severe difficulties if the spectral lines ending on low-lying levels should be observed. The reason is the absorption effective in cold and inhomogeneous gas layers intervening in the optical path between the homogeneous hot plasma column and the vacuum monochromator. Here a version of a cascade arc with a differential pumping system is described which makes possible the observation of argon resonance lines at 1067 and 1048 Å, considerably free of absorption. For this purpose it was found necessary to use a bored anode as a pumping port.

Line radiation emitted by highly ionized atoms embedded in hot laboratory plasmas can be utilized to obtain collisional rate coefficients for excitation and ionization. After a discussion of the principles underlying these measurements, the plasma device mostly used is explained briefly as are the various experimental techniques. All experimental results obtained so far are finally discussed and compared with theoretical calculations where possible.

Laser produced plasmas are a useful source of highly stripped ions for XUV spectroscopy and it is therefore important to understand the relation between the incident laser flux, plasma temperature and ionisation equilibrium.

Ionization equilibrium calculations, which have a number of astrophysical applications, have been computed by Jordan for the ions of C, N, O, Ne, Mg, Si, S, Fe, and Ni.

The aim of this paper is to extend the evaluation to other elements relevant to the investigation of the solar atmosphere. Ions of Na, Al, P, Cl, A, K, Ca, Cr, and Mn have been considered and the ionization balance has been computed for temperature from 104 to 107 K.

The computation is performed including collisional ionization, autoionization, radiative recombination and dielectronic recombination.

The collisional ionization of atomic systems by structureless charged particles of arbitrary mass and charge is considered. For the theoretical study, the classical impulse approximation is used. In this case, analytical expressions Q for the ionization cross-sections are given by Garcia et al. (1968). From these expressions, a reduced cross-section QR is obtained as a function of a scaled energy X. It is found that, for a particle of mass M, charge Z, and energy E, incident on a target having ξ equivalent electrons with a binding energy I, QR is given by:

where

IH is the hydrogen ionization potential and me is the electron mass.

X-ray radiation is obtained from a point plasma produced by pinch effect Cohen et al. (1968) in a vacuum spark similar to that used by Lie and Elton (1971) with a 220 kA peak current. The extremely dense and hot minute plasma appears at a distance of 0.5 to 1 mm from the anode tip and has maximum dimensions of a few microns.

The spectroscopic observations are performed by means of a focusing Cauchois spectrometer so as to avoid effects of fluctuations in the source position.

In this brief review of solar line identifications the progress made since the previous conference in this series, which was held in Maryland in 1968, will be discussed.

Only lines which appear in emission when seen in the solar disk will be included, thus excluding lines from low stages of ionization at wavelengths of above about 2000Å which can appear in emission at the limb.

It is convenient to divide the solar spectrum between about 1.5 Å and 2300 Å into wavelength regions, since the various types of transitions which will be discussed tend to appear over restricted wavelength ranges.

The spectra of Feix to xvi which dominate the XUV spectrum include the 3pn–3p-1 3d transitions studied earlier (Fawcett et al., 1966, 1967, 1968) (Gabriel et al., 1966) and the 3s23pn–3s3pn transitions. A comprehensive classification of spectra from both these configurations is now completed at the Astrophysics Research Unit. The experimental method used to extend the classification of the 3s23pn–3s3pn+1 lines in the period between potassium and vanadium was to compare the spectra of laser produced plasmas which possessed lines belonging to different ionization stages and is described by Fawcett (1970). Too many lines due to low stages of ionization obscured the 3s23pn–3s3pn+1 lines in chromium, manganese, iron, cobalt and nickel in these laser produced spectra for successful analysis. Volatile compounds of these elements were therefore added to the filling gas of a specially designed theta-pinch (Fawcett et al., 1971). Separation of lines of different stages of ionization was accomplished through the application of a microsecond shutter and adjusting the theta-pinch parameters. Radiation from the cooler ends of the plasma was avoided through viewing the theta-pinch from the side.

In this work we decided to concentrate on the region 50–380 Å for which the grazing incidence grating spectrograph is the most suitable instrument. This portion of the spectrum contains many rather intense lines lying very close together which arise primarily from very highly ionized atoms in the solar corona. Wavelength measurements and identifications have been made for a portion of the lines, but for many of them, especially the weaker ones, the situation was not completely clear and many doubts existed. Therefore we decided to build a rocket spectrograph which would significantly improve the resolution and wavelength accuracy for lines in the solar spectrum. Thus, we attempted to achieve wavelength accuracy of 0.003 Å and to limit the instrumental broadening of the lines to 0.03 Å.