An experimental investigation of the atomic europium ...

Home

Search

Collections

Journals

About

Contact us

My IOPscience

An experimental investigation of the atomic europium anion

This article has been downloaded from IOPscience. Please scroll down to see the full text article. 2004 J. Phys. B: At. Mol. Opt. Phys. 37 1961 (http://iopscience.iop.org/0953-4075/37/9/015) View the table of contents for this issue, or go to the journal homepage for more

Download details: IP Address: 134.197.14.10 The article was downloaded on 03/05/2013 at 17:19

Please note that terms and conditions apply.

INSTITUTE OF PHYSICS PUBLISHING

JOURNAL OF PHYSICS B: ATOMIC, MOLECULAR AND OPTICAL PHYSICS

J. Phys. B: At. Mol. Opt. Phys. 37 (2004) 1961–1965

PII: S0953-4075(04)75985-4

An experimental investigation of the atomic europium anion V T Davis1 and J S Thompson2 1

Photonics Research Centre, United States Military Academy, West Point, NY 10996, USA Department of Physics and Chemical Physics Programme, University of Nevada, Reno, Nevada 89557-0058, USA 2

Received 11 February 2004 Published 23 April 2004 Online at stacks.iop.org/JPhysB/37/1961 (DOI: 10.1088/0953-4075/37/9/015) Abstract The electron affinity (EA) of europium has been measured using laser photodetachment electron spectroscopy. The EA of Eu(8S7/2) was determined to be 1.053 ± 0.025 eV. The data also show that Eu− has at least one bound excited state with binding energy of 0.864 ± 0.024 eV relative to the ground state of the europium atom. The experimental results are consistent with recent EA measurements of lanthanide atoms whose anions are predicted to have [Xe](4f n 6p) (where n is odd) configurations.

The study of negative ions has been a laboratory for characterizing electron correlation in complex atomic and molecular systems for many years. The formation of negative ions is dependent on the subtleties of the dynamics of the electron–electron interactions to provide a potential sufficient to bind an extra electron to a neutral system. Sophisticated calculations, that account in detail for electron correlation, are employed for investigating even simple atomic negative ions. Approximation schemes are typically used to reduce the number of terms included in the calculations so that numerical techniques may be successfully employed. Furthermore, as atomic Z becomes large, the contribution to the total energy of the system by relativistic effects becomes significant and increases the complexity of the calculation. Lanthanide negative ions are particularly interesting because of their unique properties, which result from the interaction of their 4f, 5d and 6s orbital electrons. Although the small radii of the 4f orbitals limit their interactions with outer valence electrons, their binding energies are nevertheless comparable to those of their outer neighbours. Since the spread of energies within a particular configuration is much larger than the spread in those binding energies, the various configurations overlap one another to a considerable degree, making theoretical calculations based on the mixing of the relevant basis functions extremely difficult [1]. Several reviews of negative ion research [2–4] have pointed out the computational complexity encountered by theoretical investigations of lanthanide atomic anions and the limited number of experimentally derived results for these ions. Experimental verification of the existence of the predicted negative ion structure is important for validating approaches used in theoretical calculations. In particular, since knowledge of the electron affinities (EAs) of 0953-4075/04/091961+05$30.00 © 2004 IOP Publishing Ltd Printed in the UK

1961

1962

V T Davis and J S Thompson

Figure 1. Schematic diagram of the LPES experimental apparatus. See text for further information.

rare earth atoms is limited, there is keen interest in experimental data concerning the electron affinities of the lanthanides [2]. Semi-empirical estimates of the electron affinities of certain lanthanides have been reported [5–7]. Recent calculations, reported by a number of authors using several different computational techniques, predict the formation of stable negative lanthanide ions by the attachment of a 6p electron [8–16] rather than a 4f electron, due to stronger correlation effects between the 4f electrons [1]. Lanthanides are best described by jj-coupling schemes, which leads to the expectation that Eu− exists as a (7/2, 1/2)J =3,4 doublet. Previous experimental investigators have reported production of stable atomic europium anions using accelerator mass spectrometry (AMS) techniques [17, 18]. The relative yields of sputtered negative ions were used to compare binding energies, but only as an indication of relative values. Using this technique, Nadeau et al have reported a lower limit value of 0.05 eV for the electron affinity of Eu [18]. This experimental study of Eu− was performed using the laser-photodetachment electronspectroscopy (LPES) technique. A detailed description of the experimental apparatus has been given elsewhere [19], so only a brief description is presented. The experimental apparatus consisted of a commercial caesium-sputter negative-ion source, accelerator, and an interaction chamber in which photoelectrons were produced in a crossed laser-ion beams geometry (see figure 1). The source of the negative ions was a target pellet consisting of a mixture of copper powder, europium powder and sodium carbonate. The negative ions extracted from the ion source were accelerated by a 10 kV potential, mass-selected by a 90◦ bending magnet, then focused and steered into the interaction chamber. Once inside the chamber, the ion beam crossed a photon beam at an intersection angle of 90◦ . The photon beam was produced by a continuous Nd:YAG laser operating in single-line mode at 1064 nm and typically delivering between 6–8 W to the interaction chamber. Copper dimer anions (A Cu− 2 , A = 126) produced from sputtering of the copper powder were used as mass markers to identify the 153Eu− and 151 Eu− beams. Electrons photodetached in the interaction region were energy analysed using a spherical sector, 160◦ electrostatic kinetic energy analyser which operated in a fixed pass-energy mode. The electron spectrometer was positioned below the plane which contained the laser and ion beams at a 45◦ declination angle. Electrons with the correct energy for transmission through the spherical-sector analyser were detected with a channel electron multiplier (CEM). Analogue outputs from the ion-beam current and the laser power meters were converted to frequencies by a voltage-to-frequency converter, and logged with counters for normalization of electron counts.

An experimental investigation of the atomic europium anion

1963

1400

Normalized Electron Counts

1200 1000 800 600 400 200 0 0.0

0.1

0.2

0.3

0.4

0.5

Electron Energy (eV) in the Centre of Mass Frame

Figure 2. Typical photoelectron kinetic energy spectrum for photodetaching Eu− using a Nd– YAG laser (1064 nm). The laser output power was 6.2 W. The kinetic energy of the Eu− ions in the beam was 10 keV, and the ion current, measured in the interaction chamber, was 150 pA. The data accumulation time for each data point was 120 s and the spectrum took approximately 2.67 h to complete. The two Gaussian-shaped photoelectron peaks are superimposed on an linear background of low-energy electrons created by collisional detachment of Eu− ions. This spectrum is for 151Eu−. Scans of 153Eu− give identical spectra within experimental uncertainties.

A typical photoelectron kinetic energy spectrum for Eu− is shown in figure 2. Fourteen Eu photoelectron spectra were recorded. Since Eu occurs naturally in two stable isotopes of almost equal amounts; 153Eu (52.2%) and 151Eu (48.8%), scans were taken for both isotopes with identical results within experimental uncertainties. The energy scale for all the Eu− photoelectron kinetic energy spectra taken was determined using the photoelectron energy spectra of Na− produced by the sputtering of the sodium carbonate, and the known electron affinity of Na [2]. Electron energy spectra were taken for the photodetachment of Na− either before or after each Eu− photoelectron spectrum was accumulated. The energy scale for the Eu− photoelectron spectra in the laboratory frame was then transformed into the ion rest frame using the Na− photoelectron spectra as a reference. The energy separation of the photoelectron peaks corresponds to the initial and final states for the process hν + Eu− → Eu + e− , where Eu and Eu− can be in excited states. Conservation of energy requires that the kinetic energy of the photoelectron, Ec , is given by −

Ec = Eγ − Eea − Ea + Een , Eea

(1)

the excitation energy of the final state of the atom, Ea where Eγ is the photon energy, the electron affinity, and Een is the excitation energy of the initial negative ion state. For this experiment, Eea = 0 because the first excited state in europium lies 1.603 eV above the europium ground state and is not accessible at the photon energy employed in this experiment. The photoelectron peaks in figure 2 were fit to Gaussian functions superimposed over a linear background using a weighted least-squares technique to determine the energy centroid of each peak. The width of each Gaussian peak was fixed to match the width of each fine structure resolved Na− reference scan. Analysis of the photoelectron peaks in the electron spectra revealed two transitions of energies 1.053 ± 0.025 eV and 0.884 ± 0.024 eV relative to the 8S7/2 ground state of the europium atom. The data indicated that the electron affinity of europium is 1.053 ± 0.025 and that Eu− has at least one bound, long-lived, excited state. This bound excited state of Eu− must be long lived since the flight time to the interaction region

1964

V T Davis and J S Thompson

for an ion in the beam was approximately 56 µs. The observed increase in electron counts near 0 eV was due to low-energy electrons created by collisional detachment of Eu− ions in the beam by background gas and ion-aperture scattering. The measured relative intensities of the two photoelectron peaks in the electron energy spectra were the same averaged over all of the data sets, within statistical uncertainties. The variation in the relative intensities of the two photoelectron peaks was probably due to changes in the overlap of the ion and laser beams during the course of collecting each electron energy spectrum. The change in overlap of the two beams is most likely a result of the variation in the areal cross section and ion density of the ion beam over the collection times (∼3 h) for each electron energy spectrum. The reported uncertainty in the measurements represents one standard deviation of the mean. The uncertainty includes statistical and systematic contributions due to the photoelectron count rates for Eu− and the fitting of the data to Gaussian functions for the Eu− and Na− photoelectron energy spectra, the uncertainty in the electron affinity of Na, and the determination of the ion-beam kinetic energy. The reported uncertainty was dominated by the variance in the measured energy centroids resulting from fitting the data to Gaussian function for peaks in the Eu− photoelectron spectra. This variance was due to the relatively low photoelectron count rates in the Eu− photoelectron spectra resulting from the low-ion-beam intensities obtained during this set of experiments. In summary, the electron affinity of europium has been measured using laser photoelectron energy spectroscopy. The electron affinity of Eu(8S7/2) was determined to be 1.053 ± 0.025 eV. These results are consistent with the estimate by Nadeau et al of a lower limit of 0.05 eV for the EA of Eu [18]. The results also support the prediction of the formation of Eu− by the attachment of a 6p electron as evidenced by the ensuing (7/2, 1/2)J =3,4 doublet, for which the present measurement indicates a splitting of 0.169 eV. These results are consistent with recent experimental measurements for the electron affinities of praseodymium [20] and thulium [21], whose negative ions are predicted to have an active electron configuration of 4f n 6p, where n is odd. The electron affinities of Pr, Tm and Eu are all near 1 eV and all three negative ions support at least one long-lived, bound excited state. These three atomic lanthanide negative ions exhibit strong similarities in the observed photoelectron energy spectra and the measured atomic electron affinities. This indicates that the dominant contributions to the total energy for these lanthanide negative ions are similar in origin as might be expected for heavy ions with similar electron configurations. Further theoretical studies are needed to investigate the observed similarities in the structure of these negative ions. References [1] Cowan R 1981 The Theory of Atomic Structure and Spectra (Berkeley and Los Angeles: University of California Press) [2] Andersen T, Haugen H K and Hotop H 1999 J. Chem. Phys. Ref. Data 28 1511 [3] Blondel C 1995 Phys. Scr. T 58 31 [4] Andersen T 1991 Phys. Scr. T 34 23 [5] Zollweg R J 1969 J. Chem. Phys. 50 4251 [6] Angelov B M 1976 Chem. Phys. Lett. 43 368 [7] Bratsch S 1983 Chem. Phys. Lett. 98 113 [8] Vosko S H, Chevary J A and Mayer I L 1991 J. Phys. B: At. Mol. Opt. Phys. 24 L225 [9] Vosko S H, Lagowski J B, Mayer I L and Chevary J A 1991 Phys. Rev. A 43 6389 [10] Gribakina A A, Gribakin G F and Ivanov V K 1992 Phys. Lett. A 168 280 [11] Vosko S H and Chevary J A 1993 J. Phys. B: At. Mol. Opt. Phys. 26 873 [12] Datta D and Beck D R 1993 Phys. Rev. A 47 5198 [13] Dinov K, Beck D R and Datta D 1994 Phys. Rev. A 50 1114 [14] Chevary J A and Vosko S H 1994 J. Phys. B: At. Mol. Opt. Phys. 27 657

An experimental investigation of the atomic europium anion [15] [16] [17] [18] [19] [20] [21]

Eliav E and Kaldor U 1995 Phys. Rev. A 52 291 O’Malley S M and Beck D R 2000 Phys. Rev. A 61 034501 Garwan M A, Litherland A E, Nadeau M-J and Zhao X-L 1993 Nucl. Instrum. Methods B 79 631 Nadeau M-J, Garwan M A, Zhao X-L and Litherland A E 1997 Nucl. Instrum. Methods B 123 521 Covington A M, Calabrese D, Williams W W, Thompson J S and Kvale T J 1997 Phys. Rev. A 56 4746 Davis V T and Thompson J S 2002 J. Phys. B: At. Mol. Opt. Phys. 35 L11 Davis V T and Thompson J S 2002 Phys. Rev. A 65 010501R

1965