- Email: [email protected]
Resonant excitation of Fe14+ observed with a compact electron beam ion trap
Nuclear Instruments and Methods in Physics Research B 408 (2017) 191–193
Contents lists available at ScienceDirect
Nuclear Instruments and Methods in Physics Research B journal homepage: www.elsevier.com/locate/nimb
Resonant excitation of Fe14+ observed with a compact electron beam ion trap Takashi Tsuda a, Erina Shimizu a, Safdar Ali a, Hiroyuki A. Sakaue b, Daiji Kato b,c, Izumi Murakami b,c, Hirohisa Hara d,e, Tetsuya Watanabe d,e, Nobuyuki Nakamura a,⇑ a
Institute for Laser Science, The University of Electro-Communications, Tokyo 182-8585, Japan National Institute for Fusion Science, Gifu 509-5292, Japan Department of Fusion Science, SOKENDAI, Gifu 509-5292, Japan d National Astronomical Observatory of Japan, Tokyo 181-8588, Japan e Department of Astronomical Science, SOKENDAI, Tokyo 181-8588, Japan b c
a r t i c l e
i n f o
Article history: Received 15 November 2016 Received in revised form 7 March 2017 Accepted 17 March 2017 Available online 30 March 2017 Keywords: Extreme ultraviolet spectra Electron beam ion trap Resonant excitation Iron ions
a b s t r a c t We present extreme ultraviolet spectra of highly charged Fe ions observed with a compact electron beam ion trap as a function of monoenergetic electron beam energy. For 3s3p–3s3d lines in Fe XV, strong intensity enhancement at a specific electron energy is confirmed. The enhancement is assigned as the resonant excitation via dielectronic capture followed by autoionization. The resonance contribution to line intensity ratio, which is important for the diagnostics of astrophysical plasmas, is discussed. Ó 2017 Elsevier B.V. All rights reserved.
1. Introduction Spectra of highly charged Fe ions in the extreme ultraviolet (EUV) range are important for the spectroscopic diagnostics of astrophysical hot plasmas such as solar corona [1]. In such diagnostics, plasma parameters, such as electron temperature and density, are determined through the comparison between the observed spectra and theoretical spectra calculated with plasma models. For accurate diagnostics, the model spectra should thus be examined by laboratory benchmark spectra obtained under well-defined conditions. We have been studying EUV spectra of highly charged Fe ions with an electron beam ion trap (EBIT), which can realize well-defined plasma consisting of a quasimonoenergetic electron beam and trapped ions with a narrow charge state distribution [2–4]. In our previous study [3], the electron density dependence of the intensity ratio was measured for several density-sensitive emission lines in Fe XIII, XIV, and XV. Although good quantitative agreement was found between the experiment and our model calculation for Fe XIII and XIV, a significant discrepancy was found for the intensity ratio between the 3
1
3s3p3P2 –3s3d D3 (233.9 Å) and 3s3p1P1 –3s3d D2 (243.8 Å) transi⇑ Corresponding author. E-mail address: [email protected] (N. Nakamura). http://dx.doi.org/10.1016/j.nimb.2017.03.099 0168-583X/Ó 2017 Elsevier B.V. All rights reserved.
tions in Fe XV. The discrepancy for this line ratio has also been reported so far from solar observations (for example, see Ref. [5]), and line blending has often been pointed out as the origin of this discrepancy [5–7]. However, high resolution measurements in our previous study [4] have ruled out the possibility of line blending. In this paper, we study the electron energy dependence of the line ratio, whereas the density dependence was the main subject of interest in previous studies. Resonant excitation contribution to the line ratio is observed and discussed. 2. Experiment The present experiment was performed with a compact EBIT, called CoBIT [8]. Highly charged Fe ions were produced by successive ionization of iron injected as a vapor of ferrocene (Fe(C5H5)2) by a 500 eV electron beam. After a ‘‘cooking” time of 1600 ms, the electron energy was swept between 400 and 500 eV for 10 ms (probing time), and kept at 500 eV for 10 ms (keeping time) for preserving the charge distribution. After the probing and keeping periods were repeated 100 times, the ions were eliminated and the cycle was started again from the cooking time. The electron beam current was 10 mA throughout the measurement. The EUV emission from the trapped Fe ions was observed with a grazing incidence flat field grating spectrometer [9] employing a
192
T. Tsuda et al. / Nuclear Instruments and Methods in Physics Research B 408 (2017) 191–193
1200 gr/mm concave grating with 13,450 mm radius of curvature (Hitachi 001-0660). In the present setup, no entrance slit was used because the EBIT represents a line source which can be regarded as a slit. The diffracted EUV photon was detected by a position sensitive detector (PSD) consisting of five micro channel plates and a resistive anode (Quantar Technology Inc., model 3391). The front of the micro channel plate was coated by CsI for enhancing the sensitivity. When a photon was detected, the position on the PSD, pulse height, and the electron energy were recorded on a PC in list mode. 3. Results and discussion A result of about 80 h data acquisition is shown in Fig. 1. In this wavelength range, lines from Fe XIII to XV were observed with a few lines from impurity ions. Fe XV lines of present interest are 3
1
3s3p3P2 –3s3d D3 at 234 Å and 3s3p1P1 –3s3d D2 at 244 Å. As seen in the figure, intensity enhancement was observed at an electron energy of 420 eV for both the lines whereas no energy dependence was confirmed for almost all the other lines in the present electron energy and wavelength ranges. This enhancement is considered to be due to excitation realized by dielectronic capture followed by autoionization:
e þ Fe14þ ð3s2 Þ ! Fe13þ ð2p1 3s2 3p3dÞ ! Fe14þ ð3s3dÞ þ e:
ð1Þ
This is a resonant process that has a sharp dependence on electron energy. Intensity enhancements can be recognized not only at 420 eV but also around 450–460 eV though the amount of the enhancement is not so large as that at 420 eV. This enhancement 2
is considered to be the resonant excitation via Fe13+ (2p1 3s2 3d ). It is noted that the electron energy scale was calibrated with the theoretical resonance energy calculated with HULLAC [10]. The electron beam energy in an EBIT is primarily determined by the potential difference between the electron gun and the middle electrode of the ion trap. However, the actual interaction energy between the electron and the trapped ions is affected by the space charge potential, which is generally difficult to distinguish experimentally. In the present study, the difference between the energy estimated from the potential difference and the calibrated energy was about 15 eV, which may be due to the space charge potential.
Fig. 2. Similar to Fig. 1 but for the wavelength and energy ranges of present interest. The yellow lines represent the on-resonance spectrum at 420 eV and the off resonance spectrum at 412 eV. (For interpretation of the references to colour in this figure caption, the reader is referred to the web version of this article.)
Fig. 2 shows a closeup view for the lines of present interest at the electron energy range where the resonance was observed. Spectra at off- and on-resonance energies are shown. The observed intensity ratio of 234 Å to 244 Å lines obtained by fitting Gaussian functions to the spectra is 0:54 0:03 and 0:87 0:04 at off- and on-resonance, respectively (the uncertainty was simply estimated from the fitting error in the unweighted least squares fitting). Gaussian functions are used because the peak function is considered to be determined by the Gaussian spatial distribution of the electron beam, which acted as an entrance slit of the spectrometer. As confirmed in this example, not only intensity but also the intensity ratio between the two discussed lines is strongly affected by the resonance. It is thus important to take the resonance contribution into account for estimating the ratio by CR model calculation. However, the resonance contribution cannot account for the discrepancy between the experiment and the model found in the previous study [3]. The observation in the previous study was done at the electron energy corresponding to the present cooking energy, which is well above the resonance. In the present study, it is confirmed that the line ratio for the off-resonance region is 0.5–0.6 and almost independent of electron energy for the 400–500 eV range. This value is thus higher than the CR model ratio, which is 0.3 [3], obtained for the typical electron density of CoBIT (1010 cm3). As can be seen in Fig. 2, the line at 227 Å also shows intensity enhancement at an electron energy of 420 eV. It is clearly seen in the on-resonance spectrum whereas it is negligibly weak in the off-resonance spectrum. This line corresponds to the 3s3p3P1 – 3
3s3d D2 transition, and the enhancement is considered to be due to the resonance via the 2p1 3s2 3p3d resonance state, similarly to the 234 and 244 Å lines. The amount of intensity enhancement at the resonance should be proportional to the resonant excitation cross section. Although it is rather difficult to obtain absolute cross section, relative cross section can be deduced from the present observation. The analysis for the cross section is in progress, and will be published elsewhere. Acknowledgment This work was performed under the Research Cooperation Program in the National Institutes of Natural Sciences (NINS). We thank Dr. Robinson Kenneth for English proofreading. References Fig. 1. EUV spectra of highly charged Fe ions observed with a compact EBIT as a function of electron energy. The intensity is represented by color (blue to red). The wavelength scale was calibrated with well-known Fe lines, and the electron energy was calibrated with the theoretical resonance energy calculated with HULLAC [10]. The spectrum integrated for whole electron energy range is shown in the upper panel. (For interpretation of the references to colour in this figure caption, the reader is referred to the web version of this article.)
[1] T. Watanabe, Spectroscopy of highly charged ions in solar and astrophysical plasmas, J. Phys.: Conf. Ser. 163 (2009) 012002. [2] H.A. Sakaue, N. Yamamoto, S. Morita, N. Nakamura, C. Chen, D. Kato, H. Kikuchi, I. Murakami, S. Ohtani, H. Tanuma, T. Watanabe, H. Tawara, Electron density dependence of intensity ratio for Fe XXII extreme ultraviolet emission lines arising from different ground levels in electron beam ion trap and large helical device, J. Appl. Phys. 109 (2011) 073304.
T. Tsuda et al. / Nuclear Instruments and Methods in Physics Research B 408 (2017) 191–193 [3] N. Nakamura, E. Watanabe, H.A. Sakaue, D. Kato, I. Murakami, N. Yamamoto, H. Hara, T. Watanabe, Intensity ratio of density-sensitive lines in Fe ions observed with a well-defined laboratory plasma, Astrophys. J. 739 (2011) 17. [4] E. Shimizu, H.A. Sakaue, D. Kato, I. Murakami, N. Yamamoto, H. Hara, T. Watanabe, N. Nakamura, Intensity ratio of EUV emission lines in Fe XV studied with electron beam ion traps, J. Phys.: Conf. Ser. 583 (2015) 012019. [5] P.L. Dufton, A.E. Kingston, K.G. Widing, Fe XV line ratios in the sun: discrepancies between theory and observations, Astrophys. J. 353 (1990) 323–328. [6] S.O. Kastner, A.K. Bhatia, Optically thin and thick Fe XV spectrum: effect of self-absorption on the 284.16 Å resonance line, Astrophys. J. 553 (2001) 421–428.
193
[7] F.P. Keenan, K.M. Aggarwal, D.S. Bloomfield, A.Z. Msezane, K.G. Widing, An investigation of Fe XV emission lines in solar flare spectra, Astron. Astrophys. 449 (2006) 1203–1208. [8] N. Nakamura, H. Kikuchi, H.A. Sakaue, T. Watanabe, Compact electron beam ion trap for spectroscopy of moderate charge state ions, Rev. Sci. Instrum. 79 (2008) 063104. [9] H. Ohashi, J. Yatsurugi, H.A. Sakaue, N. Nakamura, High resolution extreme ultraviolet spectrometer for an electron beam ion trap, Rev. Sci. Instrum. 82 (2011) 083103. [10] A. Bar-Shalom, M. Klapisch, J. Oreg, HULLAC, an integrated computer package for atomic processes in plasmas, J. Quant. Spectrosc. Radiat. Trans. 71 (2001) 169.
Contents lists available at ScienceDirect
Nuclear Instruments and Methods in Physics Research B journal homepage: www.elsevier.com/locate/nimb
Resonant excitation of Fe14+ observed with a compact electron beam ion trap Takashi Tsuda a, Erina Shimizu a, Safdar Ali a, Hiroyuki A. Sakaue b, Daiji Kato b,c, Izumi Murakami b,c, Hirohisa Hara d,e, Tetsuya Watanabe d,e, Nobuyuki Nakamura a,⇑ a
Institute for Laser Science, The University of Electro-Communications, Tokyo 182-8585, Japan National Institute for Fusion Science, Gifu 509-5292, Japan Department of Fusion Science, SOKENDAI, Gifu 509-5292, Japan d National Astronomical Observatory of Japan, Tokyo 181-8588, Japan e Department of Astronomical Science, SOKENDAI, Tokyo 181-8588, Japan b c
a r t i c l e
i n f o
Article history: Received 15 November 2016 Received in revised form 7 March 2017 Accepted 17 March 2017 Available online 30 March 2017 Keywords: Extreme ultraviolet spectra Electron beam ion trap Resonant excitation Iron ions
a b s t r a c t We present extreme ultraviolet spectra of highly charged Fe ions observed with a compact electron beam ion trap as a function of monoenergetic electron beam energy. For 3s3p–3s3d lines in Fe XV, strong intensity enhancement at a specific electron energy is confirmed. The enhancement is assigned as the resonant excitation via dielectronic capture followed by autoionization. The resonance contribution to line intensity ratio, which is important for the diagnostics of astrophysical plasmas, is discussed. Ó 2017 Elsevier B.V. All rights reserved.
1. Introduction Spectra of highly charged Fe ions in the extreme ultraviolet (EUV) range are important for the spectroscopic diagnostics of astrophysical hot plasmas such as solar corona [1]. In such diagnostics, plasma parameters, such as electron temperature and density, are determined through the comparison between the observed spectra and theoretical spectra calculated with plasma models. For accurate diagnostics, the model spectra should thus be examined by laboratory benchmark spectra obtained under well-defined conditions. We have been studying EUV spectra of highly charged Fe ions with an electron beam ion trap (EBIT), which can realize well-defined plasma consisting of a quasimonoenergetic electron beam and trapped ions with a narrow charge state distribution [2–4]. In our previous study [3], the electron density dependence of the intensity ratio was measured for several density-sensitive emission lines in Fe XIII, XIV, and XV. Although good quantitative agreement was found between the experiment and our model calculation for Fe XIII and XIV, a significant discrepancy was found for the intensity ratio between the 3
1
3s3p3P2 –3s3d D3 (233.9 Å) and 3s3p1P1 –3s3d D2 (243.8 Å) transi⇑ Corresponding author. E-mail address: [email protected] (N. Nakamura). http://dx.doi.org/10.1016/j.nimb.2017.03.099 0168-583X/Ó 2017 Elsevier B.V. All rights reserved.
tions in Fe XV. The discrepancy for this line ratio has also been reported so far from solar observations (for example, see Ref. [5]), and line blending has often been pointed out as the origin of this discrepancy [5–7]. However, high resolution measurements in our previous study [4] have ruled out the possibility of line blending. In this paper, we study the electron energy dependence of the line ratio, whereas the density dependence was the main subject of interest in previous studies. Resonant excitation contribution to the line ratio is observed and discussed. 2. Experiment The present experiment was performed with a compact EBIT, called CoBIT [8]. Highly charged Fe ions were produced by successive ionization of iron injected as a vapor of ferrocene (Fe(C5H5)2) by a 500 eV electron beam. After a ‘‘cooking” time of 1600 ms, the electron energy was swept between 400 and 500 eV for 10 ms (probing time), and kept at 500 eV for 10 ms (keeping time) for preserving the charge distribution. After the probing and keeping periods were repeated 100 times, the ions were eliminated and the cycle was started again from the cooking time. The electron beam current was 10 mA throughout the measurement. The EUV emission from the trapped Fe ions was observed with a grazing incidence flat field grating spectrometer [9] employing a
192
T. Tsuda et al. / Nuclear Instruments and Methods in Physics Research B 408 (2017) 191–193
1200 gr/mm concave grating with 13,450 mm radius of curvature (Hitachi 001-0660). In the present setup, no entrance slit was used because the EBIT represents a line source which can be regarded as a slit. The diffracted EUV photon was detected by a position sensitive detector (PSD) consisting of five micro channel plates and a resistive anode (Quantar Technology Inc., model 3391). The front of the micro channel plate was coated by CsI for enhancing the sensitivity. When a photon was detected, the position on the PSD, pulse height, and the electron energy were recorded on a PC in list mode. 3. Results and discussion A result of about 80 h data acquisition is shown in Fig. 1. In this wavelength range, lines from Fe XIII to XV were observed with a few lines from impurity ions. Fe XV lines of present interest are 3
1
3s3p3P2 –3s3d D3 at 234 Å and 3s3p1P1 –3s3d D2 at 244 Å. As seen in the figure, intensity enhancement was observed at an electron energy of 420 eV for both the lines whereas no energy dependence was confirmed for almost all the other lines in the present electron energy and wavelength ranges. This enhancement is considered to be due to excitation realized by dielectronic capture followed by autoionization:
e þ Fe14þ ð3s2 Þ ! Fe13þ ð2p1 3s2 3p3dÞ ! Fe14þ ð3s3dÞ þ e:
ð1Þ
This is a resonant process that has a sharp dependence on electron energy. Intensity enhancements can be recognized not only at 420 eV but also around 450–460 eV though the amount of the enhancement is not so large as that at 420 eV. This enhancement 2
is considered to be the resonant excitation via Fe13+ (2p1 3s2 3d ). It is noted that the electron energy scale was calibrated with the theoretical resonance energy calculated with HULLAC [10]. The electron beam energy in an EBIT is primarily determined by the potential difference between the electron gun and the middle electrode of the ion trap. However, the actual interaction energy between the electron and the trapped ions is affected by the space charge potential, which is generally difficult to distinguish experimentally. In the present study, the difference between the energy estimated from the potential difference and the calibrated energy was about 15 eV, which may be due to the space charge potential.
Fig. 2. Similar to Fig. 1 but for the wavelength and energy ranges of present interest. The yellow lines represent the on-resonance spectrum at 420 eV and the off resonance spectrum at 412 eV. (For interpretation of the references to colour in this figure caption, the reader is referred to the web version of this article.)
Fig. 2 shows a closeup view for the lines of present interest at the electron energy range where the resonance was observed. Spectra at off- and on-resonance energies are shown. The observed intensity ratio of 234 Å to 244 Å lines obtained by fitting Gaussian functions to the spectra is 0:54 0:03 and 0:87 0:04 at off- and on-resonance, respectively (the uncertainty was simply estimated from the fitting error in the unweighted least squares fitting). Gaussian functions are used because the peak function is considered to be determined by the Gaussian spatial distribution of the electron beam, which acted as an entrance slit of the spectrometer. As confirmed in this example, not only intensity but also the intensity ratio between the two discussed lines is strongly affected by the resonance. It is thus important to take the resonance contribution into account for estimating the ratio by CR model calculation. However, the resonance contribution cannot account for the discrepancy between the experiment and the model found in the previous study [3]. The observation in the previous study was done at the electron energy corresponding to the present cooking energy, which is well above the resonance. In the present study, it is confirmed that the line ratio for the off-resonance region is 0.5–0.6 and almost independent of electron energy for the 400–500 eV range. This value is thus higher than the CR model ratio, which is 0.3 [3], obtained for the typical electron density of CoBIT (1010 cm3). As can be seen in Fig. 2, the line at 227 Å also shows intensity enhancement at an electron energy of 420 eV. It is clearly seen in the on-resonance spectrum whereas it is negligibly weak in the off-resonance spectrum. This line corresponds to the 3s3p3P1 – 3
3s3d D2 transition, and the enhancement is considered to be due to the resonance via the 2p1 3s2 3p3d resonance state, similarly to the 234 and 244 Å lines. The amount of intensity enhancement at the resonance should be proportional to the resonant excitation cross section. Although it is rather difficult to obtain absolute cross section, relative cross section can be deduced from the present observation. The analysis for the cross section is in progress, and will be published elsewhere. Acknowledgment This work was performed under the Research Cooperation Program in the National Institutes of Natural Sciences (NINS). We thank Dr. Robinson Kenneth for English proofreading. References Fig. 1. EUV spectra of highly charged Fe ions observed with a compact EBIT as a function of electron energy. The intensity is represented by color (blue to red). The wavelength scale was calibrated with well-known Fe lines, and the electron energy was calibrated with the theoretical resonance energy calculated with HULLAC [10]. The spectrum integrated for whole electron energy range is shown in the upper panel. (For interpretation of the references to colour in this figure caption, the reader is referred to the web version of this article.)
[1] T. Watanabe, Spectroscopy of highly charged ions in solar and astrophysical plasmas, J. Phys.: Conf. Ser. 163 (2009) 012002. [2] H.A. Sakaue, N. Yamamoto, S. Morita, N. Nakamura, C. Chen, D. Kato, H. Kikuchi, I. Murakami, S. Ohtani, H. Tanuma, T. Watanabe, H. Tawara, Electron density dependence of intensity ratio for Fe XXII extreme ultraviolet emission lines arising from different ground levels in electron beam ion trap and large helical device, J. Appl. Phys. 109 (2011) 073304.
T. Tsuda et al. / Nuclear Instruments and Methods in Physics Research B 408 (2017) 191–193 [3] N. Nakamura, E. Watanabe, H.A. Sakaue, D. Kato, I. Murakami, N. Yamamoto, H. Hara, T. Watanabe, Intensity ratio of density-sensitive lines in Fe ions observed with a well-defined laboratory plasma, Astrophys. J. 739 (2011) 17. [4] E. Shimizu, H.A. Sakaue, D. Kato, I. Murakami, N. Yamamoto, H. Hara, T. Watanabe, N. Nakamura, Intensity ratio of EUV emission lines in Fe XV studied with electron beam ion traps, J. Phys.: Conf. Ser. 583 (2015) 012019. [5] P.L. Dufton, A.E. Kingston, K.G. Widing, Fe XV line ratios in the sun: discrepancies between theory and observations, Astrophys. J. 353 (1990) 323–328. [6] S.O. Kastner, A.K. Bhatia, Optically thin and thick Fe XV spectrum: effect of self-absorption on the 284.16 Å resonance line, Astrophys. J. 553 (2001) 421–428.
193
[7] F.P. Keenan, K.M. Aggarwal, D.S. Bloomfield, A.Z. Msezane, K.G. Widing, An investigation of Fe XV emission lines in solar flare spectra, Astron. Astrophys. 449 (2006) 1203–1208. [8] N. Nakamura, H. Kikuchi, H.A. Sakaue, T. Watanabe, Compact electron beam ion trap for spectroscopy of moderate charge state ions, Rev. Sci. Instrum. 79 (2008) 063104. [9] H. Ohashi, J. Yatsurugi, H.A. Sakaue, N. Nakamura, High resolution extreme ultraviolet spectrometer for an electron beam ion trap, Rev. Sci. Instrum. 82 (2011) 083103. [10] A. Bar-Shalom, M. Klapisch, J. Oreg, HULLAC, an integrated computer package for atomic processes in plasmas, J. Quant. Spectrosc. Radiat. Trans. 71 (2001) 169.