Department of Physics, Imperial College, London SW7 2BZ, UK

Astrophysics Research Centre, School of Mathematics and Physics, Queen’s University, Belfast, BT7 1NN, Northern .K

Harvard-Smithsonian Center for Astrophysics, MS-3, 60 Garden Street, Cambridge, MA 02138, USA

Department of Physics, Laboratory, University of Oxford, Parks Road, Oxford, OX1 3PU, UK Received date / Accepted date ABSTRACT Context. The I(1 5.01 Å)/I(16.78 Å) emission line intensity ratio in Fe XVII has been reported to deviate from its theoretical value in solar and stellar X-ray spectra. This is attributed to opacity in the 15.01 Å line, leading to a reduction in its intensity, and was interpreted in terms of a geometry in which the emitters and absorbers are spatially distinct. Aims. We study the I(15.01 Å)/I(16.78 Å) intensity ratio for the active cool dwarf EV Lac, in both flare and quiescent spectra. Methods. The observations were obtained with the Reflection Grating Spectrometer on the XMMNewton satellite. The emission measure distribution versus temperature reconstruction technique is used for our Analysis Results. We find that the 15.01 Å line exhibits a significant enhancement in intensity over the optically thin value. To our knowledge, this is the first time that such an enhancement has been detected on such a sound statistical basis. We interpret this enhancement in terms of a geometry in which the emitters and absorbers are not spatially distinct, and where the geometry is such that resonant pumping of the upper level has a greater effect on the observed line intensity than resonant absorption in the line-of-sight. Key words. radiative transfer - stars activity - stars coronae - stars individual (EV Lacertae) - techniques spectroscopic - Xrays stars 1. Introduction Although opacity effects have been frequently detected in the chromospheric and transition region emission lines of lowly ionized species in solar and stellar sources (see, for example, 200 6 ), this is not the case for coronal transitions in more highly ionized systems. In some respects, this is not surprising, as the optical depth at line centre scales directly with wavelength, Send offprint requests to

e-mail s.rose@imperial.ac.uk and is inversely proportional to the square root of the plasma temperature (Mitchell & Zemansky 1961). Coronal lines in highly charged ions will generally lie at short wavelengths (in the X-ray spectral region), and be formed at high values of plasma temperature

( 1997 ) and Phillips et al. (1996 ) found some evidence of opacity in the Fe XVII 15.01 Å line from an Analysis of solar active regions. However, no evidence of coronal opacity was found in the RS CVn binaries Capella

2001 ) and II Peg

2001 ), nor in a sample of 26 late-type stars

2003 ), although

(2004 ) claim the detection of opacity in the

18.9 7 Å emission line of II Peg. More recently

(2005 ) have analysed a 50 ksec XMM-Newton observation of the late-type star AB Dor. They found that the observed emission line intensities were generally in very good agreement with theoretical values from an optically thin model. However, the Fe XVII intensity ratio I(15.01 Å)/I(16.78 Å) was significantly smaller during the flare period of observation than the optically thin prediction. This was attributed to opacity in the 15.01 Å transition, which has a much larger oscillator strength than the 16.78 Å feature. There was no evidence of opacity in the quiescent X-ray Spectrum of AB Dor. EV Lac (GJ 873 ) is a very active dM4

5e flare star at a distance of 5 pc. It exhibits spot-related quasi-periodic variations, and has been reported to have a solar-like activity cycle (Mavridis & Avgoloupis 1986). EV Lac has one of the highest surface magnetic flux in the recent survey of low mass stars by Reiners & Basri (2007). In the present paper we study the X-ray Spectrum of EV Lac, with emphasis on the effects of resonant scattering, via the Analysis of Fe XVII line intensity ratios..5e flare star at a distance of 5 pc. It exhibits spot-related quasi-periodic variations, and has been reported to have a solar-like activity cycle (Mavridis & Avgoloupis 1986). EV Lac has one of the highest surface magnetic flux in the recent survey of low mass stars by Reiners & Basri (2007). In the present paper we study the X-ray Spectrum of EV Lac, with emphasis on the effects of resonant scattering, via the Analysis of Fe XVII line intensity ratios. 2. Observations & Analysis The observations of AB Dor reported by (200 5 ) were part of a larger dataset that also included the star EV Lac. Observations of these stars were made during both flare and quiescent periods and were obtained with the XMM-Newton Reflection Grating Spectrometers RGS1 and RGS2, which provide spectra with a resolution of 70 -500 in the 0.2 -2.5 keV range. Data were reduced with the XMM Science Analysis System (SAS) software v5.3.3. The RGS1 and RGS2 spectra are divided into 3400 channels with widths ranging from 0.007 to 0.014 Å, and could not be summed directly as their wavelength grids do not coincide, and were therefore rebinned into a new grid with a width of 0.02 Å for further Analysis A total line spread function was defined for the summed Spectrum, which is the sum of the individual line spread functions weighted by the effective area (see

2004 ). The Analysis of the RGS spectra is based on the plasma emission measure distribution versus temperature reconstruction technique, and was performed using the IDL-based software PINTofALE (Kashyap & Drake 2000 ). In our Analysis, the first step was the identification of the strongest lines in each Spectrum, using the latest version (Version 5.2 ) of the CHIANTI atomic database

1997; Landi et al. 2006 ). The line spread function of the RGS instrument is characterized by extended wings, and multicomponent fitting involving the adjacent transitions was therefore necessary for most of the lines. In this case the total emissivity is the sum of the emissivities of the single lines, each multiplied by the corresponding element abundance. Lines rejected from our Analysis include those whose inverse emissivity curves did not agree with those of other lines of the same ion. Weak lines, which are affected by large uncertainties and strong blending, were also excluded from the emission measure reconstruction. Further details of our Analysis procedures may be found in

(2005 ). 3. Results and Discussion The observed intensity ratios of the Fe XVII 1 5.01 Å (3 d → 2 p) to 16.78 Å (3 s → 2 p) transitions are consistently smaller that those calculated by collisional-radiative models. In an effort to explain the discrepancy

(2000 ) measured the ratio using the NIST EBIT and found it in agreement with the ratios predicted using a distorted wave collisional model. Similar experiments in the LLNL EBIT-II showed that resonant excitation can make a significant contribution to the 3 s → 2 p line intensities

2002 ). Additional processes therefore have to be taken into account. Doron & Behar (2002 ) have shown that the inclusion of radiative recombination, dielectronic recombination, resonant excitation and inner-shell collisional ionization into a collisional-radiative model, can rectify some of these discrepancies. The Doron & Behar (2002 ) calculations and are used in our Analysis We note that the most recent calculations of Landi & Gu (2006 ) are in agreement with Doron & Behar (2002 ). In Table 1 we present the observed I(15.01 Å)/I(16.78 Å) line intensity ratios for EV Lac during 4 different periods of observation, namely 1 quiescent and 3 flare, where the quoted errors in Robserved correspond to 2 σ. The predicted ratios for an optically thin plasma from the calculations of Doron & Behar (2002 ), Rtheory, and the ratios generated from the differential emission measure (DEM) reconstruction, RDEM, are also given for each case. Values quoted for Rtheory are for 2 different estimates of the plasma temperature (Te), with Rtheory = 1.75 calculated for Te = 7. 2 ×106 K, determined using the measured I(Fe XVIII 16.07 Å)/I(Fe XVII 16.78 Å) intensity ratio in conjunction with the calculations of

(1997 ). By contrast, Rtheory = 1.85 is the theoretical ratio for Te = 107 K, the temperature corresponding to the peak of the Fe XVII 15.01 Å line contribution function