The Chemical Compositions of the Two New HgMn Stars HD 30085 and HD 30963: Comparison to χ Lupi A, ν Cap, and HD 174567

R. Monier; E. Griffin; M. Gebran; T. Kılıçoğlu; T. Merle; F. Royer

doi:10.3847/1538-3881/ab3b59

1. Introduction

The fairly bright stars HD 30085 (HR 1510, A0 IV, V = 6.3), HD 30963 (B9 III, V = 7.2), and HD 174567 (HR 7098, A0V, V = 6.63) have received little attention: only 30, 10, and 47 references, respectively, can be found in SIMBAD for these stars. We have recently undertaken a spectroscopic survey of all apparently slowly rotating, bright early A-stars (A0-A1V) and late B-stars (B8-B9V) observable from the northern hemisphere. This project addresses the fundamental questions of the physics of late B- and early A- stars: (i) can we find new instances of rapid rotators seen pole-on (other than Vega) and study their physical properties (gradient of temperature across the disk, limb, and gravity darkening); and (ii) is our census of chemically peculiar (CP) stars complete up to the magnitude limits we adopted? If not, what are the physical properties of the newly found CP stars? HD 30085 and HD 174567 pertain to the sample of the 47 apparently slowly rotating A0-A1V stars in the northern hemisphere which satisfy v_e sin i ≤ 60 km s⁻¹ and δ ≥ −10° analyzed by Royer et al. (2014). An abundance analysis of selected lines outside the Balmer lines allowed for these stars to be sorted into three groups: 13 CP stars were found (among which 4 are new CPs), 17 superficially normal stars, and 17 spectroscopic binaries (Royer et al. 2014). Monier et al. (2015) reported on a first analysis of the abundances of Fe, Mn, and Hg, and showed that HD 30085, HD 18104, HD 32867, and HD 53588 are four new HgMn stars. Similarly HD 30963 is a slowly rotating B9 star which verifies similar criteria.

In this paper we report on the abundance analysis of 40 chemical elements from high-resolution high-quality échelle spectra of HD 30085, HD 30963, and HD 174567 in the optical range and one spectrum of the HgMn star χ Lupi A (χ Lupi = HD 141556 is a double-lined spectroscopic binary, we are only interested in the abundances of χ Lupi A, the HgMn star). We compare the elemental abundances we find in these four stars to those derived for ν Cap, a bona fide normal late B-type star which we have recently analyzed (Monier et al. 2018) Using model atmospheres and line synthesis, we derive for the first time the abundances of 40 chemical elements in these four stars and find that they depart strongly from solar. The abundances of sodium, magnesium, and calcium have been corrected for non-LTE (NLTE) effects. The overabundances of Mn, Sr, Y, Zr, Pt, and Hg and the underabundances of helium, magnesium, scandium, and nickel lead us to confirm that HD 30085 is indeed a new HgMn star and establish that HD 30963 is a new HgMn star. The superficially normal HD 174567 appears to be a new mild CP star. We have also reanalyzed the cool HgMn star χ Lupi A on the same temperature scale and using the same atomic data and used it as a comparison star.

The paper is divided into five sections. The first section recapitulates what is known of HD 30085, HD 30963, HD 174567, and χ Lupi A, the second section describes new spectroscopic observations of HD 30085, HD 30963, HD 174567, and χ Lupi A. In the third section, we present the derivation of fundamental parameters and the spectral synthesis which we adopted to derive the abundances. In Section 4, we discuss the determination of elemental abundances for each star in light of what is known of other HgMn stars. We also provide identifications of lines absorbing more than 2% of the continuum in HD 30085. In the conclusion, we discuss the chemical peculiarity of the three new CP stars in the light of what is known of other HgMn stars.

2. Recap of Previous Spectroscopic Work on HD 30085, HD 30963, HD 174567, ν Cap, and χ Lupi A

HD 30085 was ascribed a spectral type A0 IV by Cowley et al. (1969) in their survey of 1700 bright northern B9 to A9 stars with a prismatic dispersion of 125 Å mm⁻¹ around H_γ. At that time, Cowley et al. (1969) did not mention any peculiarity of the spectrum. In his study of helium-weak stars, Molnar (1972) classified HD 30085 as a B9 III from 63 Å mm⁻¹ plates centered on H_γ using slightly different MK criteria to assign the temperature and the luminosity class than Cowley et al. (1969) did. He did not comment on any peculiarity of the lines of Si, Sr, and the metals in the spectrum of HD 30085 around H_γ. HD 30085 has been detected as a fairly bright UV source with the S2/68 Ultraviolet Sky Survey Telescope (Jamar et al. 1976), its flux steadily rising toward shorter wavelengths, suggesting that the star is indeed a late B-star rather than an early A-star. Ramella et al. (1989) report on structures in the core of the Mg ii doublet at 4481 Å from 12.4 Å mm⁻¹ dispersion plates. More recently, Monier et al. (2015) have reported on the presence of the Hg ii line at 3984 Å and several strong Mn ii in the high-resolution spectra of HD 30085 and provided overabundances for Mn, Fe, and Hg only based on the spectrum synthesis of a few lines, which clearly established the HgMn nature of this star.

HD 30963 was ascribed a B9 III spectral type by Huang et al. (2010). HD 174567 has been used as a normal comparison star by Smith & Dworetsky (1993) in their abundance study of HgMn and superficially normal stars from International Ultraviolet Explorer (IUE) spectra. Their modeling of coadded IUE spectra of HD 174567 revealed abundances which are nearly solar, confirming the superficially normal status for this star at that time.

The abundances of 22 elements in the atmosphere of χ Lupi A have been derived from optical spectra by Wahlgren et al. (1994) and they are collected in Table 6. The elemental abundances of ν Cap have been recently derived in Monier et al. (2018), which confirms its nearly solar abundances determined by Adelman (1991).

3. Observations

The high-resolution spectra of HD 30085, HD 30963, and HD 174567 have been obtained at the Observatoire de Haute Provence with SOPHIE, the échelle spectrograph in its high-resolution mode (R = 75,000) yielding a full spectral coverage from 3820 to 6930 Å in 39 orders. A detailed technical description of SOPHIE is given in Perruchot et al. (2008). SOPHIE is a cross-dispersed, environmentally stabilized échelle spectrograph dedicated to high -precision radial velocity. The spectra are extracted online from the detector images using a specific pipeline adapted from that of the High Accuracy Radial Velocity Planet Searcher (HARPS).⁷ The sequence or reductions starts with the localization of the 39 orders on the 2D images, the optimal extraction of each order, the wavelength calibration, and finally the correction for flat-field producing a two-dimensional spectrum (e2ds). We have normalized each reduced order separately using a Chebychev polynomial fit with sigma clipping, rejecting points above or below 1σ of the local continuum. Normalized orders were then merged together, corrected by the blaze function and resampled into a constant wavelength step of about 0.02 Å (see Royer et al. 2014, for more details). The observing dates, exposure times, and signal-to-noise (S/N) ratios achieved for each observing run are collected in Table 1. For HD 30085 and HD 30963, the comparison of individual spectra taken at different epochs did not reveal any convincing radial velocity nor residual flux variations in the lines. We therefore coadded the individual orders into merged orders to enhance the S/N up to 350. Complementary mono-order spectra of HD 30085 have been obtained at the Dominion Astrophysical Observatory by Elizabeth Griffin with a lower resolving power of R = 45,000. The three wavelengths intervals observed span from 3874 to 4020 Å, from 4038 to 4183 Å, and from 4430 to 4560 Å. The H and H_δ regions were observed in order to validate the fundamental parameters derived from the Strömgren's photometry (see Section 4.1) and confirm the presence of the Hg ii line at 3983.93 Å.

Table 1. Log of the Observed and Archival Spectra

Star ID	Date	Instrument	R	Barycentric Julian Date	Duration (s)	S/N
HD 30085	2012 Feb 13	SOPHIE	75000	2455971.36664683	800	216
HD 30085	2013 Dec 10	SOPHIE	75000	2456637.61970650	1200	269
HD 30085	2013 Feb 7	DAO	45000	2456331.255	3600	220
HD 30085	2013 Mar 2	DAO	45000	2456354.176	3600	220
HD 30085	2013 Mar 3	DAO	45000	2456355.136	3600	220
HD 30085	2015 Aug 18	DAO	45000	2457253.503	2500	250
HD 30085	2015 Aug 19	DAO	45000	2457254.507	2400	250
HD 30085	2015 Sep 10	DAO	45000	2457276.500	2100	240
HD 30085	2015 Sep 11	DAO	45000	2457277.478	3600	220
HD 30085	2015 Sep 12	DAO	45000	2457278.435	2700	250
HD 30085	2015 Oct 21	DAO	45000	2457317.459	2700	250
HD 30085	2015 Nov 24	DAO	45000	2457351.233	3600	220
HD 30085	2015 Nov 24	DAO	45000	2457351.261	1200	240
HD 30085	2015 Nov 24	DAO	45000	2457351.506	1800	240
HD 30085	2015 Nov 24	DAO	45000	2457351.527	1800	240
HD 30085	2015 Nov 25	DAO	45000	2457352.274	3600	220
HD 30085	2015 Nov 25	DAO	45000	2457352.501	3600	220
HD 30085	2015 Nov 26	DAO	45000	2457353.222	3600	220
HD 30085	2015 Nov 26	DAO	45000	2457353.476	3600	220
HD 30085	2015 Nov 27	DAO	45000	2457354.225	3600	220
HD 30085	2015 Dec 25	DAO	45000	2457382.278	3600	220
HD 30085	2015 Dec 30	DAO	45000	2457387.317	3600	220
HD 30085	2015 Dec 31	DAO	45000	2457388.274	3600	220
HD 30085	2016 Jan 1	DAO	45000	2457389.247	3600	220
HD 30085	2016 Jan 23	DAO	45000	2457411.159	3600	220
HD 30963	2015 Nov 28	SOPHIE	75000	2457354.49106071	2400	118
HD 30963	2015 Nov 28	SOPHIE	75000	2457355.48326524	1800	78
HD 30963	2015 Nov 30	SOPHIE	75000	2457356.51212353	1800	140
HD 30963	2015 Nov 30	SOPHIE	75000	2457357.46945236	1800	101
HD 30963	2015 Dec 1	SOPHIE	75000	2457358.43199975	1680	167
HD 30963	2016 Dec 13	SOPHIE	75000	2457736.46584169	900	129
HD 30963	2016 Dec 13	SOPHIE	75000	2457736.49875775	900	123
HD 30963	2016 Dec 14	SOPHIE	75000	2457737.44071886	1800	165
HD 174567	2009 Aug 5	SOPHIE	75000	2455049.36018803	1200	224
HD 141556	2013 Feb 4	FEROS	48000	2456327.37979	50	325

Download table as: ASCII Typeset image

4. Model Atmosphere Analysis and Synthetic Spectra Computation

The abundances of 40 chemical elements have been derived by iteratively adjusting synthetic spectra to the normalized spectra and looking for the best fit to carefully selected unblended lines. Specifically, synthetic spectra were computed assuming LTE using the SYNSPEC49 code of Hubeny & Lanz (1992) that calculates lines for elements up to Z = 99. For selected lines of Na i, Mg i, and Ca ii, we have provided the NLTE abundances.

4.1. Fundamental Parameters

The fundamental data for HD 30085, HD 30963, ν Cap, HD 174567, and HD 141556 are collected in Table 2. The spectral type retrieved from SIMBAD appears in column 2 and the apparent magnitudes in column 3, the Strömgren indexes b − y, m₁, and c₁ are in columns 4, 5, 6. The photometric data was taken from Hauck & Mermilliod (1998). The radial velocity of HD 30085, HD 174567, and ν Cap are those derived in Royer et al. (2014) using cross-correlation techniques, avoiding the Balmer lines and the atmospheric telluric lines. The normalized spectrum was cross-correlated with a synthetic template extracted from the POLLUX database⁸ (Palacios et al. 2010) corresponding to the parameters T_eff = 11,000 K, log g = 4, and solar abundances. A parabolic fit of the upper part of the resulting cross-correlation function yields the Doppler shift, which is then used to shift spectra to rest wavelengths. The projected rotational velocities were derived from the position of the first zero of the Fourier transform of individual lines; they are taken from Royer et al. (2014). The radial velocity and projected equatorial velocity of HD 30085, HD 30963, ν Cap, HD 174567, and HD 141556 are collected in Table 2.

Table 2. Adopted Fundamental Parameters for HD 30085, HD 30963, ν Cap, HD 174567, and HD 141556

		HD 30085	HD 30963	ν Cap	HD 174567	HD 141556
Sp.T.		A0 IV	B9 III	B9 IV	A0 V	B9 IV
V		6.37	7.20	4.76	6.63	3.95
b − y		−0.038	−0.021	−0.021	0.008	−0.020
m₁		0.135	0.134	0.134	0.123	0.129
c₁		0.849	1.015	1.015	1.083	0.948
T_eff	[K]	11300 ± 200	11476 ± 200	10300 ± 200	10200 ± 200	10608 ± 200
log g	(cgs)	3.95 ± 0.20	3.66 ± 0.20	3.90 ± 0.20	3.55 ± 0.20	3.98 ± 0.20
v_e sin i	[km s⁻¹]	26.0	37.0	24.0	10.5	2.0
v_micr.	[km s⁻¹]	${0.0}_{-0.0}^{+0.4}$	${0.1}_{-0.1}^{+0.3}$	0.50	0.95 ± 0.15	0.10
v_rad	[km s⁻¹]	8.27	3.52	−11.39	−10.84	15.30
M	[M_⊙]	3.1 ± 0.4	3.7 ± 0.4	2.8 ± 0.3	3.4 ± 0.4	2.8 ± 0.3
log t	(t in yr)	8.35 ± 0.05	8.27 ± 0.07	8.50 ± 0.06	8.37 ± 0.15	8.46 ± 0.10
t_MS	(%)	64	85	69	86	63

Download table as: ASCII Typeset image

For the five stars, the effective temperature (T_eff) and surface gravity (log g) were determined using the UVBYBETA code developed by Napiwotzki et al. (1993) and appear in columns 7 and 8. This code is based on the Moon & Dworetsky (1985) grid, which calibrates the uvbyβ photometry in terms of T_eff and log g. The estimated errors on T_eff and log g are ±125 K and ±0.20 dex, respectively (see Section 4.2 in Napiwotzki et al. 1993).

4.1.1. Microturbulent Velocity Determinations

In order to derive the microturbulent velocity of HD 30085, HD 30963, and HD 174567, we have simultaneously derived the iron abundance [Fe/H] for 50 unblended Fe ii lines and a set of microturbulent velocities ranging from 0.0 to 2.0 km s⁻¹. Figure 1 shows the standard deviation of the derived [Fe/H] as a function of the microturbulent velocity. The adopted microturbulent velocities are the values that minimize the standard deviations, i.e., for that value, all Fe ii lines yield the same iron abundance. The microturbulent velocities of the three stars are collected in Table 2.

**Figure 1.** Microturbulent velocity determinations for HD 30085, HD 30963, and HD 174567.
Download figure:
Standard image High-resolution image

4.1.2. Location of the Five Stars in a log g, log T_eff Diagram

The locations of HD 30085, HD 30963, HD 174567, ν Cap, and χ Lupi in a theoretical gravity–temperature (log g, log T_eff) diagram are shown in Figure 2. Evolutionary tracks and isochrones from Bressan et al. (2012)⁹ are displayed for masses from 2.4 to 3.6 M_⊙ with a step of 0.4 M_⊙ and for log t (where t is in years) of 8.25, 8.35, 8.40, 8.45, and 8.50. The evolutionary tracks are computed for a solar initial metallicity of Z = 0.017 including microscopic diffusion. The isocrones are retrieved for the current solar composition, Z = 0.0152. From this diagram we have estimated the masses, ages, and fractional lifetimes on the main sequence, which are collected in the last three rows of Table 2.

4.2. Model Atmospheres

The ATLAS9 code (Kurucz 1992) was used to compute a first model atmosphere for the effective temperature and surface gravity of each star assuming a plane parallel geometry, a gas in hydrostatic and radiative equilibrium, and local thermodynamical equilibrium. The ATLAS9 model atmosphere contains 72 layers with a regular increase in log τ_Ross = 0.125 and was calculated assuming a solar chemical composition (Grevesse & Sauval 1998). It was converged up to log τ = −5.00 in order to attempt to reproduce the cores of the Balmer lines. This ATLAS9 version uses the new opacity distribution function (ODF) of Castelli & Kurucz (2003) computed for that solar chemical composition. Once a first set of elemental abundances was derived using the ATLAS9 model atmosphere, the atmospheric structure was recomputed for these abundances using the opacity sampling ATLAS12 code¹⁰ (Kurucz 2005, 2013). New slightly different abundances were then derived and a new ATLAS12 model recomputed until the abundances of iteration (n-1) differed of those of iteration (n) by less than ±0.10 dex.

4.3. The Line List

Atomic line lists from Kurucz's database have provided the basis to construct our line list.¹¹ These lists collect data mostly from the literature for light and heavy elements (usually critically evaluated transition probabilities from Martin et al. 1988 and Fuhr et al. 1988) and computed data by Kurucz (1992) for the iron group elements. A first line list was built from Kurucz's gfall21oct16.dat,¹² which includes hyperfine splitting levels. We then updated several oscillator strengths and damping parameters with more recent and accurate determinations when necessary. As a rule, we have preferred National Institute of Standards and Technology (NIST)¹³ and Wiese et al. (1996) oscillator strengths for carbon, nitrogen, and oxygen and also Nilsson et al. (2006). The H i lines are calculated using Vidal et al. (1973) tables upgraded by Schoening & Butler up to H 10. The He i lines are computed in SYNSPEC49 using specific tables, either from Shamey (1969) or Dimitrijevic & Sahal-Brechot (1984).

We introduced hyperfine components for a few lines of Mn ii (Holt et al. 1999), isotopic, and hyperfine transitions for a few lines of Ga ii (Nielsen et al. 2000). To model the Hg ii line at 3983.93 Å, we have included nine hyperfine transitions from the various isotopes from Hg¹⁹⁶ up to Hg²⁰⁴ from Dolk et al. (2003).

For the rare earths, we retrieved all relevant transitions from the Database on Rare Earth at Mons University (DREAM) database.¹⁴ We also used specific publications reporting on laboratory measurements: Sm ii (Lawler et al. 2006), Nd ii (Den Hartog et al. 2003), Eu ii (Lawler et al. 2001b), and Tb ii (Lawler et al. 2001a). In Kurucz's line lists, the damping constants are taken from the literature when available. For the iron group elements, they come from the computations of Kurucz (1992). Additional damping constants for a few Si ii lines were taken from Lanz et al. (1988). When they are not available from the line list, damping constants are actually computed in SYNSPEC49 using an approximate formula (Kurucz & Avrett 1981).

4.4. Spectrum Synthesis

We have used only unblended lines to derive the abundances. For a given element, the final abundance is a weighted mean of the abundances derived for each transition (the weights are derived from the NIST grade assigned to that particular transition). For several elements, in particular the heaviest elements, only one unblended line was available and the abundance should be regarded as uncertain. For each modeled transition, the adopted abundance is that which provides the best fit calculated with SYNSPEC49 to the observed normalized profile. A grid of synthetic spectra was computed with SYNSPEC49 (Hubeny & Lanz 1992) to model each selected unblended line of the 40 elements for the four stars. Computations were iterated varying the unknown abundance until minimization of the chi-square between the observed and synthetic spectrum was achieved. For HD 30085, the final synthetic spectrum has allowed the identification of most observed lines.

5. Abundance Determinations

We discuss here the identifications of chemical elements, the lines selected for abundance determinations and the abundance determinations for each element. The abundances derived for each transition and the final abundances¹⁵ are collected with their errors in Table 8. We also provide NLTE abundance corrections for three light elements, namely Na i, Mg i, and Ca ii. We use the Formato code (T. Merle et al. 2019, in preparation) to build simple model atoms, using atomic data from NIST (Kramida et al. 2018) for energy levels, from VALD (Ryabchikova et al. 2015) for radiative bound–bound transitions, and from TOPbase (Cunto & Mendoza 1992) for radiative bound–free transitions. For inelastic collisions with electrons, we use the semi-empirical formula from Seaton (1962) for which a radiative counterpart exists and a collision strength of one for transitions with a forbidden radiative counterpart. For inelastic collisions with hydrogen, we use the formula from Drawin (1969) without a scaling factor, excepted for Mg i where we implemented the mechanical quantum data from Guitou et al. (2015). The statistical equilibrium and the radiative transfer equation for each level and line in each model atom are consistently solved using the NLTE 1D radiative transfer code MULTI (Carlsson 1986, 1992). The references of the atomic data we have used are collected in Table 9. The identications of lines absorbing more than 2% of the continuum are collected in Table 10.

5.1. Helium

The strongest helium lines unambiguously detected are: λ 4026.19, λ 4471.48, and λ 5875.621. The others are either weak or embedded into blends. The least blended, λ 4471.48, has been synthesized to derive the helium LTE abundance. Helium is underabundant in all stars and the deficiency is more important in the three HgMn stars. Figure 3 displays the best fit achieved to model the He i line at 4471.480 Å for HD 30085.

**Figure 3.** Synthesis of the He i line at 4471.480 Å for HD 30085.
Download figure:
Standard image High-resolution image

5.2. The Light Elements (C to Ti)

5.2.1. Carbon

Most of the expected C ii lines are blended with more abundant species. The least blended is the high-excitation C ii triplet at 4267.26 Å which coincides with a 5% line at 4267.18 Å in HD 30085 for instance. Carbon is underabundant in HD 30085, HD 174567, and χ Lupi A, but solar in HD 30963. Sakhibullin (1987) has shown that the C ii line at 4267.26 Å is prone to NLTE effects for B stars hotter than 15,000 K. He suggests the use of C ii lines at 3919 and 3921 Å less influenced by departures from LTE to derive carbon abundances. These lines yield the same carbon abundances.

5.2.2. Oxygen

For O i, the strongest expected allowed lines are those of multiplet 7 (NIST quality B+). The nine transitions of multiplet 7 dominate the opacity from 6155 to 6159 Å. Other lines of O i are blended with lines of iron-peak elements and were therefore discarded to derive the oxygen abundance. Oxygen is underabundant in HD 30085 and χ Lupi A, solar in HD 30963, and slightly overabundant in HD 174567.

5.2.3. Sodium

We have used the two Na i lines at 4497.66 Å (NIST quality B+) and the resonance Na i lines at 5889.95 and 5895.92 Å (quality AA). The LTE overabundances of sodium derived from these lines are −4.97 for HD 30085 and −5.67 (i.e., solar) the upper limit of HD 30963 and HD 174567. The NLTE corrections for the resonance Na i lines at 5889.95 and 5895.92 Å are about −0.52 and −0.60 dex, respectively, which yield NLTE abundances of Na of −5.61 in HD 30085 and −6.27 for HD 30963 and HD 174567 and −5.57 for χ Lupi A, well below the LTE determinations. Sodium is underabundant in HD 30963 and HD 174567 and slightly overabundant in HD 30085 and χ Lupi A.

5.2.4. Magnesium

The unblended Mg ii lines at 4390.514 and 4390.572 Å yield LTE abundances of −4.64, −4.57, −4.42, and −4.46 for HD 30085, HD 30963, HD 174567 and χ Lupi A, respectively. The unblended Mg i line at 5172.68 Å yields similar LTE abundances. We adopt the abundance derived from the unblended Mg i line at 5172.68 Å corrected with the NLTE correction of about −0.33 dex, which yields −4.42, −4.85, −4.59, and −5.02, respectively. Magnesium therefore appears to be underabundant in each star.

5.2.5. Aluminium

The synthesis of the unblended Al ii line at 4663.056 Å shows that aluminium is severely depleted in the three CP stars, but not in the superficially normal HD 174567.

5.2.6. Silicon

The synthesis of strong and unblended lines of Si ii shows that silicon is slightly overabundant in HD 30085, HD 30963, and HD 174567, and slightly depleted in χ Lupi A.

5.2.7. Phosphorus

The lines of P ii expected strongest are λ 6024.18 and 6043.12 of multiplet 2. Phosphorus is overabundant in the four stars.

5.2.8. Sulfur

The lines of S ii at 4162.67 Å, 4142.25 Å, and 4153.06 Å of multiplet 4 are clearly present and unblended in the spectra of all stars. Sulfur is solar in HD 30085, overabundant in HD 30963 and χ Lupi A, and underabundant in HD 30963.

5.2.9. Calcium

The synthesis of the unblended Ca ii lines at 3933.66 and 5019.97 Å yields the calcium LTE abundances, respectively −5.64, −5.46, −5.64, and −5.83 in HD 30085, HD 30963, HD 174567, and χ Lupi A. The NLTE correction for the 3933.66 Å amounts to +0.37 dex, yielding final overabundances of −5.22 for HD 30085 and HD 174567 and −5.37 for HD 30963 and −5.46 for χ Lupi A, respectively. Calcium therefore appears to be overabundant in all four stars.

5.2.10. Scandium

The line of multiplet 1 of Sc ii at 4246.820 Å is the only unblended line (quality A in NIST). Scandium appears to be overabundant in HD 30085, HD 30963, and HD 174567, but underabundant in χ Lupi A

5.2.11. Titanium

We have used 10 unblended lines of Ti ii listed in Table 8 to derive the mean titanium abundances. Titanium is overabundant in all four stars.

5.3. The Iron-peak Elements (V to Zn) and Gallium

5.3.1. Vanadium

The vanadium abundance has been estimated using three lines of V ii at 4005.705, 4023.378, and 4036.78 Å. The adjustment of these lines suggests a significant underabundance of vanadium in HD 30085, overabundances for HD 30963 and HD 174567, and a nearly solar abundance for χ Lupi A.

5.3.2. Chromium

Several strong and unblended lines of Cr ii could easily be found in the spectra of the four stars. Six unblended lines of Cr ii were synthesized and yield chromium overabundances in the four stars. The three resonance lines of Cr i lines at 4245.35, 4274.82, and 4289.73 Å are also present and yield lower overabundances. The next lines of Cr i expected to contribute opacity are those of multiplet 3, but they are all blended. Lines of higher multiplets are expected too faint to be detectable even at an overabundance of −5.43. We have retained the overabundances from Cr ii, which is the dominant ionization stage.

5.3.3. Manganese

The lines of Mn ii are conspicuous in the spectra of HD 30085 and HD 30963. The two lines have hyperfine structures published by Holt et al. (1999). They yield overabundances in all stars, especially large for HD 30085 and HD 30963.

The resonance lines of Mn i at 4030.753 and 4034.483 Å are properly reproduced by a lower manganese overabundance of about −5.13 for HD 30085 for instance. Three other lines at 4055.544 Å, 4526.530 Å, and 4823.524 Å absorb a few percent of the continuum and are properly reproduced by the same abundance. The other Mn i lines are very weak and blended.

5.3.4. Iron

The abundance of iron was derived from the 11 Fe ii lines, all of which are sensitive to small changes in $[\tfrac{\mathrm{Fe}}{{\rm{H}}}]$ . Iron is overabundant in HD 30085, HD 30963, and HD 174567, and slightly underabundant for χ Lupi A.

The 10 lines of Fe i listed in Castelli & Hubrig (2004) are present, the strongest are 4045.812 and 4383.545 Å and are blended in all stars. The others range from 2% to 5% absorption and are often blended. The only unblended Fe i line is 4383.545 Å, and it is adjusted with a solar abundance of iron for all stars.

5.3.5. Nickel

The Ni ii line at λ 4067.031 is weak in the spectra of HD 30085 and χ Lupi A, absorbing about 1% of the continuum and is properly fit with a pronounced underabundance of nickel in HD 30085 and in χ Lupi A. In HD 30963 and HD 174567, the line absorbs about 2% of the continuum and yields a nickel abundance of −6.75 and −5.27. Nickel appears to be underabundant in the three CP stars but slightly overabundant in the superficially normal HD 174567.

5.3.6. Gallium

We have used the high-excitation Ga ii line at λ 6334.069 with hyperfine splitting of the gallium isotopes. It is the strongest expected line of Ga ii in this temperature regime. The eight transitions retrieved and modified from Nielsen et al. (2000) are collected in Table 3.

Table 3. The Ga ii λ 6334.069 Line List

Wavelength	Ion	log gf	Lower Energy Level	EW
(Å)			(cm⁻¹)	(mÅ)
6333.930	Ga ii	0.08	102944.595	14.7
6333.980	Ga ii	0.21	102944.595	20.5
6333.990	Ga ii	0.08	102944.595	14.6
6334.069	Ga ii	0.10	102944.595	15.3
6334.080	Ga ii	0.40	102944.595	26.5
6334.083	Ti ii	−2.08	66521.008	0.0
6334.120	Ga ii	0.02	102944.595	12.8
6334.200	Ga ii	0.09	102944.595	20.9
6334.208	Cr ii	−1.66	89812.422	0.0
6334.290	Ga ii	0.01	102944.595	16.8
6334.368	Ni ii	−2.36	120271.975	0.0

Download table as: ASCII Typeset image

In HD 30963, the line of Ga ii at 6334.069 Å is clearly present. Other lines of Ga ii are present at 4251.108 Å, 4254.032 Å, 4255.64 Å, and 4261.995 Å, and 5360.313 Å, 5363.353 Å, and 5421.122 Å in the red part of the spectrum. The fits to the three lines of Ga ii and their hyperfine structures at 4255.6 and 5360.3 Å are collected in Figures 4 and 5 and consistently yield a large overabundance for HD 30963. In the other three stars, gallium is solar or underabundant

**Figure 4.** Synthesis of the Ga ii 4255.77 Å blend in HD 30963 showing the contribution of the hyperfine structure of gallium.
Download figure:
Standard image High-resolution image

**Figure 5.** Synthesis of the Ga ii 5360 Å blend in HD 30963 showing the hyperfine structure of gallium.
Download figure:
Standard image High-resolution image

5.4. The Sr–Y–Zr Triad

5.4.1. Strontium

The two lines of multiplet 2 have critical evaluation (NIST quality A and AA). The abundance of strontium was derived from the fit to the only unblended line of Sr ii at 4215.52 Å. Strontium is significantly overabundant in HD 30085 and χ Lupi A and overabundant in HD 30963 and HD 174567.

5.4.2. Yttrium

The yttrium lines are conspicuous in the spectra of HD 30085 and HD 30963, most of them absorbing from a few percent up to 10%. The only unblended lines are the low-excitation lines at 3982.592 and 5662.925 Å. Yttrium is very overabundant in each star.

5.4.3. Zirconium

The only two Zr ii unblended lines are λ 4443.008 and λ 4457.431. Zirconium is overabundant in all stars.

5.5. Barium

The two resonance lines of Ba ii (multiplet 1) at 4554.029 Å and 4934.076 Å and the low-excitation line 6141.59 Å are present. The hyperfine structure of the various isotopes of barium we have used is collected in Table 4. Barium is overabundant by a factor of 10 in HD 30085, HD 30963, and HD 174567, and a factor of 5 in χ Lupi A.

Table 4. The Ba ii Line Lists

Wavelength	Ion	log gf	Lower Energy Level	EW
(Å)			(cm⁻¹)	(mÅ)
4553.934	Zr ii	−0.57	19514.840	0.1
4553.995	Ba ii	−1.57	0.000	0.1
4553.997	Ba ii	−1.57	0.000	0.1
4553.998	Ba ii	−1.99	0.000	0.0
4553.999	Ba ii	−1.82	0.000	0.1
4554.001	Ba ii	−1.82	0.000	0.1
4554.001	Ba ii	−2.22	0.000	0.0
4554.011	Cr i	−0.73	33040.093	0.0
4554.029	Ba ii	0.02	0.000	16.6
4554.029	Ba ii	−1.45	0.000	1.3
4554.029	Ba ii	−0.94	0.000	3.8
4554.046	Ba ii	−1.38	0.000	0.3
4554.049	Ba ii	−1.82	0.000	0.1
4554.049	Ba ii	−1.15	0.000	0.6
4554.050	Ba ii	−2.52	0.000	0.0
4554.051	Ba ii	−1.57	0.000	0.1
4554.052	Ba ii	−2.29	0.000	0.0

4934.005	Fe i	−0.61	33507.120	0.4
4934.074	Ba ii	−3.12	0.000	0.0
4934.074	Ba ii	−1.33	0.000	0.8
4934.075	Ba ii	−3.15	0.000	0.0
4934.075	Ba ii	−1.10	0.000	1.3
4934.076	Ba ii	−1.77	0.000	0.3
4934.076	Ba ii	−1.25	0.000	0.9
4934.077	Ba ii	−0.29	0.000	7.0
4934.084	Fe i	−2.30	26627.608	0.1

6141.597	Fe i	−3.12	33765.304	0.0
6141.713	Ba ii	−0.22	5674.824	2.4
6141.713	Ba ii	−1.03	5674.824	0.4
6141.714	Ba ii	−1.18	5674.824	0.3
6141.714	Ba ii	−1.26	5674.824	0.2
6141.715	Ba ii	−1.69	5674.824	0.1
6141.717	Ba ii	−3.07	5674.824	0.0
6141.718	Ba ii	−3.05	5674.824	0.0
6141.731	Fe i	−1.61	29056.322	0.1

Download table as: ASCII Typeset image

5.6. The Rare Earths

We have searched for once-ionized rare earths elements in the spectra of the four stars, namely La ii, Ce ii, Pr ii, Nd ii, Sm ii, Eu ii, Gd ii, Tb ii, Dy ii, Ho ii, and Er ii. As the twice-ionized rare earths often are the dominant stages in these atmospheres of late B stars, we also looked for the twice-ionized species, Pr iii and Nd iii, using the lines listed either in NIST or in DREAM and the line lists published in Ryabchikova et al. (2006) and Ryabchikova et al. (2007).

5.6.1. Lanthanum

The spectrum synthesis of the La ii line at 4042.91 Å feature is consistent with 1 solar abundance upper limit for HD 30085 and an overabundance. Lanthanum is overabundant in HD 174567 and χ Lupi A. The line is blended in HD 30963.

For La iii, the three lines λ 4499.064, λ 5145.718, and λ 5148.391 retrieved from NIST are all blended with more abundant species in the three stars.

5.6.2. Cerium

The synthesis of the unblended Ce ii lines at 4133.80 and 4460.21 Å yields large overabundances of cerium in HD 30085 and χ Lupi A and a 1 solar upper limit for HD 30963 and HD 174567.

For Ce iii, we find coincidences for 7 out of 10 low-excitation lines listed in DREAM, however all of them are heavily blended in the spectra of the four stars.

5.6.3. Praseodymium

The lines of Pr iii at 5284.69 and 5299.99 Å are unblended and provide large overabundances in HD 30085, HD 174567, and χ Lupi A. They are consistent with a solar abundance upper value for HD 30963.

5.6.4. Neodymium

For Nd iii, the strongest expected lines in DREAM are the resonance lines λ 5265.019 and λ 5294.099, which correspond to structures absorbing respectively 3% and 5% of the local continuum inside complex blends. The synthesis of these lines yields moderate to large overabundances of Nd iii in HD 30085, in HD 30963, HD 174567, and in χ Lupi A.

5.6.5. Samarium

The Sm ii lines at 4280.79 and 4424.32 Å are unblended and are consistent with a 1 solar upper limit abundance of samarium in HD 30085, HD 30963, and HD 174567, and an overabundance of 25 in χ Lupi A.

5.6.6. Europium

The Eu ii resonance line at 4129.73 Å is unblended and consistent with a 1 solar upper limit abundance of europium in HD 30085, HD 30963, and HD 174567, and a significant overabundance of about 100 in χ Lupi A.

5.6.7. Gadolynium

The Gd ii line at 4037.32 Å is unblended and properly reproduced with large overabundances of about 100 in HD 30085, HD 174567, and χ Lupi A. It is blended in HD 30963.

5.6.8. Terbium

The Tb ii line at 4005.47 Å is unblended in HD 30085 and HD 30963 and provides a 1 solar upper limit abundance. It is blended in HD 174567 and χ Lupi A.

5.6.9. Dysprosium

The Dy ii line at 4000.45 Å is unblended in HD 30085, HD 174567, and χ Lupi A, and provides a 1 solar upper limit for HD 30085 and an overabundance by a factor of 60, respectively, for HD 174567 and χ Lupi A. This line is blended in HD 30963.

5.6.10. Holmium

The low-excitation line of Ho ii at 4152.62 Å is unblended and provides a 1 solar upper limit in HD 30085, HD 30963, and HD 174567, and an overabundance in χ Lupi A.

5.6.11. Erbium

For Er ii, the line at 4142.91 Å is consistent with a 1 solar upper limit for HD 30085, is blended in HD 30963 and χ Lupi A, and reproduced by a large overabundance of about 173 in HD 174567.

5.6.12. Thulium

The line at 4242.15 Å of Tm ii is blended in HD 30085 and χ Lupi A and is consistent with a 1 solar upper limit for HD 30963 and HD 174567.

5.6.13. Ytterbium

The Yb ii line at 4135.095 Å provides a 1 solar upper limit for HD 30085, HD 174567, and χ Lupi A. It is blended in HD 30963.

5.6.14. Hafnium

The Hf ii line at 3918.08 Å yields a 1 solar upper limit for HD 30085 and HD 174567 and is blended in HD 30963 and χ Lupi A.

5.7. The Very Heavy Elements Osmium, Platinum, Gold, and Mercury

5.7.1. Osmium

The Os ii line at 4158.44 Å is consistent with a 1 solar upper limit in HD 30085, HD 174567, and χ Lupi A, and is blended in HD 30963.

5.7.2. Platinum

The 4514.47 Å Pt ii line is the only unblended line of platinum in the spectra of the four stars. It does not have any hyperfine structure published for the various isotopes of Pt. We therefore derived a crude estimate of the platinum overabundance. There are no other species at this wavelength which could account for the observed absorption as shows the spectrum synthesis without the Pt ii line. The wavelength scale around the Pt ii line was checked using the Fe ii neighboring control lines whose NIST wavelengths are accurate to within ±0.001 Å.

The Pt ii line at 4046.45 Å has hyperfine structure published (five transitions), but it is blended with the nine hyperfine transitions of the Hg i line at 4046.57 Å. Using the mercury abundance obtained in the following paragraph, the synthesis of this blend in the spectra of HD 30085 and HD 30963 also yields an overabundance similar to that derived from the 4514.47 Å line. The final adopted platinum abundance is that derived from the Pt ii line at 4514.47 Å. It ranges from large overabundances of the order of 10⁴ for 30085 and 10³ in HD 30963 to a solar upper limit (−10.65) for HD 174567. This line is blended in χ Lupi A. The synthesis of the Pt ii line at 4514.47 Å in HD 30085 is shown in Figure 6.

**Figure 6.** Synthesis of the Pt ii 4514.17 Å line in HD 30085.
Download figure:
Standard image High-resolution image

5.7.3. Gold

The only Au ii line corresponding with weak features is λ 4052.790, but it is blended with an Fe i line in the spectra of all four stars.

5.7.4. Mercury

The abundance of mercury has been derived from the low-excitation Hg ii line at 3983.93 Å only. This feature absorbs about 12% of the continuum in HD 30085, HD 30963, and χ Lupi A and about 4% in HD 174567. The other Hg ii lines are all high-excitation lines either weak or blended with more abundant species and were not synthesized. In particular Hg ii 6149.4749 Å is blended with Fe ii 6149.258 Å in HD 30085, HD 30963, and HD 174567, which precludes any conclusion on the mercury abundance. The blend is resolved in χ Lupi A and yields an overabundance which agrees from that found with the 3983.93 Å line.

To model the Hg ii line at 3983.93 Å, we have included nine transitions, i.e., all hyperfine structures from the various isotopes from Hg¹⁹⁶ and Hg²⁰⁴ listed in Dolk et al. (2003). These transitions are collected together with blending lines from Ti i, Fe i, Cr is and Cr ii in Table 5.

Table 5. The Hg ii 3983.93 Å Line List

Wavelength	Ion	log gf	Lower Energy Level	EW
(Å)			(cm⁻¹)	(mÅ)
3983.771	Hg ii	−4.50	35514.999	1.2
3983.827	Ti i	−1.46	17075.258	0.0
3983.832	Fe i	−4.81	24338.766	0.0
3983.839	Hg ii	−3.00	35514.000	14.1
3983.844	Hg ii	−3.13	35514.000	12.5
3983.851	Cr ii	−4.31	81962.291	0.0
3983.853	Hg ii	−3.00	35514.000	14.1
3983.874	Fe i	−2.70	34039.515	0.0
3983.899	Cr i	0.35	20520.904	3.2
3983.912	Hg ii	−2.50	35514.000	19.2
3983.932	Hg ii	−3.10	35514.000	12.9
3983.941	Hg ii	−2.90	35514.000	15.3
3983.956	Fe i	−1.02	21999.129	2.5
3983.986	Cr ii	−2.32	90512.561	0.0
3983.991	Cr ii	−2.88	82362.188	0.0
3983.993	Hg ii	−3.00	35514.000	14.2
3984.022	Cr i	−2.47	35397.971	0.0
3984.072	Hg ii	−3.00	35514.000	14.4

Download table as: ASCII Typeset image

The wavelength scale was checked using four control lines on each side of the Hg ii line: shortwards the Zr ii line at 3982.0250 Å, the Y ii line at 3982.59 Å, and longwards the Zr ii lines at 3984.718 and 3991.15 Å. Once corrected for the radial velocities of HD 30085 and HD 30963, the centers of these lines are found at their expected laboratory locations to within ±0.02 Å, which we will adopt as the accuracy of the wavelength scale in this spectral region. After rectification to the red wing of the H line, the core of the Hg ii lines of HF 30085 and HD 30963 are flat and extend from about 3983.90 ± 0.02 Å to 3984.07 ± 0.02 Å, which roughly correspond to the positions of lines of the heaviest isotopes Hg²⁰⁰ and Hg²⁰⁴. The heavy isotopes of Hg thus contribute more to the absorption than the lighter isotopes in HD 30085 and HD 30963 as is the case in the coolest HgMn stars (White et al. 1976; Woolf & Lambert 1999). In χ Lupi A, the Hg ii line center is at 3984.07 Å which corresponds to the line of the heaviest isotope.

First, we checked the influence of possible contaminant species to the Hg ii line from 3983.50 Å up to 3984.50 Å which we estimate to be the maximum extension of the line wings. The only possible contaminants are three weak lines: Fe ii λ 3983.737, Cr i λ 3983.897, and Fe i λ 3983.956. The equivalent widths of these lines were computed for the derived Fe and Cr abundances and the sum of their contributions, which amounts to about 4.9 mÅ, is by far insufficient to reproduced the equivalent width of the observed feature at 3983.93 Å (about 64–70 mÅ) of the Hg ii in HD 30085, HD 30963, and χ Lupi A. Another test consisted in computing the flux without the Hg ii transitions. This consistently resulted into a very weak absorption feature from 3983.50 to 3984.50 Å in the four stars. We can therefore conclude that the observed features at 3983.93 Å are indeed mostly due to the Hg ii line and are almost free of blending.

Including the hyperfine structure of the seven isotopes from Table 2 of Dolk et al. (2003) significantly reduces the mercury abundance, for example to about 120,000 ⊙ in HD 30085 (it would be of the order of 10⁷ ⊙ if we include the single transition retrieved from the Vienna Astrophysical Line Database (VALD). In HD 30085 and HD 30963, the 12% absorption at the core is reproduced but the synthetic profile is significantly offset to the blue by about −0.05 Å from the observed blue wing and + 0.10 Å from the observed red wing. In order to shift redwards the synthetic line core and give it a flat shape comparable to the observed one, we have iteratively altered the oscillator strengths of the individual hyperfine components until the overall observed line profile could be reproduced. We progressively decreased the oscillator strengths of the first four transitions of the three lightest isotopes, Hg¹⁹⁶, Hg¹⁹⁸, and Hg¹⁹⁹, and increased those of five transitions of the heaviest, Hg²⁰⁰, Hg²⁰¹, Hg^202, and Hg²⁰⁴, in order to redistribute the computed line opacity toward longer wavelengths. The final combination of oscillator strengths is recorded in Table 5.

We can give a very rough estimate of the contribution of each isotope in HD 30085 and HD 30963 by dividing the equivalent widths of its hyperfine components to the total equivalent width of the whole line. We thus find increasingly larger contributions as we move to heavier isotopes: Hg¹⁹⁶ 0.4%, Hg¹⁹⁸ 9%, Hg¹⁹⁹ 16%, Hg²⁰⁰ 14%, Hg²⁰¹ 20%, Hg²⁰² 16%, and Hg²⁰⁴ 24%.

The final mercury abundances derived from the synthesis of the 3983.93 Å yield very large overabundances of about −5.83 in HD 30085 (see Figure 7), −5.31 in HD 30963, −5.91 for χ Lupi A, and −7.61 in HD 174567. The observed Hg ii line center in χ Lupi A is at about 3984.072 Å, which corresponds to the heaviest isotope of mercury only. The possible contribution of the various isotopes of mercury in HD 30085, HD 30963, and χ Lupi A is represented as histograms in Figure 8.

**Figure 8.** Possible contribution of the various isotopes of Hg to the 3983.93 Hg ii line in each star.
Download figure:
Standard image High-resolution image

6. Discussion

In Table 6 we compare the abundances we derived for χ Lupi A with those derived from high-resolution spectra in the optical range by Wahlgren et al. (1994).

Table 6. Comparison between the Abundances of χ Lupi A and the Ones Derived in Wahlgren et al. (1994)

Element	$\left\langle \mathrm{log}({\rm{X}}/{\rm{H}})\right\rangle$		Element	$\left\langle \mathrm{log}({\rm{X}}/{\rm{H}})\right\rangle$
	This Work	Wahlgren et al.		This Work	Wahlgren et al.
He i	−1.67 ± 0.23	C ii	−3.76 ± 0.34	−3.85
O i	−3.53 ± 0.10	−3.93	Na i	−4.97	⋯
Mg ii	−4.66 ± 0.15	−4.60 ± 0.06	Al ii	−5.79 ± 0.20	−5.85
Si ii	−4.55 ± 0.05	−4.42 ± 0.28	P ii	−5.47 ± 0.16	⋯
S ii	−4.96 ± 0.08	−4.78 ± 0.04	Ca ii	−5.83 ± 0.16	−5.70
Sc ii	−10.13 ± 0.14	−10.16 ± 0.17	Ti ii	−6.73 ± 0.06	−6.65 ± 0.25
V ii	−8.02 ± 0.12	−7.96 ± 0.23	Cr ii	−6.14 ± 0.17	−5.96 ± 0.27
Mn ii	−6.13 ± 0.04	−6.15 ± 0.13	Fe ii	−4.57 ± 0.16	−4.29 ± 0.21
Ni ii	−6.52 ± 0.14	−5.96 ± 0.10	Ga ii	−9.12	⋯
Sr ii	−7.43 ± 0.09	−7.03 ± 0.14	Y ii	−8.06 ± 0.16	−7.89 ± 0.16
Zr ii	−8.70 ± 0.09	−8.68 ± 0.18	Ba ii	−9.17 ± 0.25	−8.80
La ii	−9.13 ± 0.13	⋯	Ce ii	−8.90 ± 0.25	⋯
Pr iii	−9.09 ± 0.13	⋯	Nd iii	−8.80 ± 0.25	⋯
Sm ii	−8.59 ± 0.13	⋯	Eu ii	−9.61 ± 0.25	⋯
Gd ii	−8.70 ± 0.13	⋯	Dy ii	−9.08 ± 0.25	⋯
Ho ii	−9.50 ± 0.13	⋯	Yb ii	−10.92 ± 0.25	⋯
Os ii	−10.55 ± 0.13	⋯	Au ii	−10.99 ± 0.25	−7.49
Hg ii	−5.91 ± 0.13	−6.34

Download table as: ASCII Typeset image

Our abundance determinations for carbon, magnesium, aluminium, silicon, calcium, scandium, titanium, vanadium, chromium, manganese, and zirconium agree within ±0.20 dex (a representative error on our abundances) with those derived by Wahlgren et al. (1994). For the other nine elements common to both studies, the abundances differ by more than 0.20 dex. This probably comes from using different lines or different more recent atomic data for lines in common, different model atmospheres, and different line synthesis codes.

From the list of identifications provided for the final composition of HD 30085 in Table 10, we can conclude that the following species are present in the spectra of HD 30085 (and HD 30963 as well): C ii, N ii, O i, Na i, Mg ii, Mg i, Mg ii, Si ii, P ii, S ii, Ca ii, Sc ii, Ti ii, V ii, Cr ii, Cr i, Mn ii, Fe ii, Ni ii, Ga ii, Sr ii, Y ii, Zr ii, Ba ii, Pt ii, and Hg ii. Among these species, Si ii, Ti ii, Cr ii, Mn ii, Fe ii, Y ii, and Zr ii, have numerous and strong lines and are important opacity sources. Lines from neutrals, Cr i, Mn i, and Fe i, are also observed and yield lower abundances than Cr ii, Mn ii, and Fe ii, indicating possibly departures from LTE. Gallium is present with a large overabundance in HD 30963.

In HD 30085, we only find evidence for overbundances of four rare earths from unblended lines of Ce ii, Pr iii, Nd iii, and Gd ii. For the other rare earths, we find upper limits: the abundances must be solar or lower. In HD 30963, the blending is higher because of the larger v_e sin i, we can conclude only that Nd iii may be present and overabundant (evidence from one line only), all other lines are blended. In HD 174567, the low v_e sin i favors the detection of weak unblended lines and evidence is found for overabundances of La ii, Pr iii, Nd iii, Gd ii, Dy ii, and Er ii. The presence of once-ionized rare earths is difficult to assess at these temperatures above 10,000 K because they are not the dominant ionization stage. In general, the abundances for the rare earths should be regarded as the least reliable ones in this study because they were usually inferred from the synthesis of one weak line of the once-ionized species only.

The overabundances found for HD 30085 and HD 30963 run from very mild (2.0 ⊙ for Fe) up to quite large (10⁵ ⊙ for Hg). The underabundances run from mild −0.80 ⊙ for carbon to pronounced −0.02 ⊙ for nickel. Helium is quite underabundant in HD 30085, HD 30963, and χ Lupi A.

The abundance patterns of the five stars are compared in Figure 9. The comparison bona fide normal late B-star, ν Cap, has a nearly solar composition for all elements as already found by Monier et al. (2018) and Adelman (1991). For the other four stars, the overall trend is that the light elements (He through Mg) are most often deficient whereas elements heavier than Mn tend to be more and more overabundant as the atomic mass increases (with the exception of nickel). This pattern is most likely caused by radiative diffusion. In particular the overall shapes of the patterns of HD 30085 and HD 30963 compare well with that of χ Lupi A (HD 141556), even though differences for individual abundances exist. There are several similarities for the abundances of the elements lighter than Scandium between χ Lupi A and HD 30963, in particular for helium, magnesium, phosphorus, and calcium. For elements heavier than scandium, the abundances of HD 30963 differ from that of χ Lupi A. For HD 30085, the carbon, magnesium, phosphorus, gallium, strontium, and praseodymium abundances are similar to those of χ Lupi A. Overall, the chemical patterns of HD 30085 and HD 30963 follow that of χ Lupi A, which definitely confirms that these two star are new HgMn stars. Moreover the patterns of HD 30085 and HD 30963 also fall well within the general abundance pattern of confirmed HgMn stars in Figure 1 of the Ghazaryan & Alecian (2016) compilation. HD 30085 and HD 30963 are most likely classical HgMn stars.

The pattern of HD 174567 is intermediate between that of ν Cap and that of the three HgMn stars. We therefore propose that HD 174567 is a new mild CP star with overabundances of elements heavier than strontium less pronounced than in HD 30085 and HD 30963 but definitely larger than in ν Cap. The overabundances for Sr, Y, Zr, several rare earths, and mercury suggest that HD 174567 could be a new cool and mild HgMn star. More observations of HD 174567 are foreseen in order to elucidate the nature of this interesting object.

7. Conclusion

With their characteristic underabundances of most light elements up to calcium and overabundances of manganese and elements heavier than strontium, in particular platinum and mercury, HD 30085 and HD 30963 are definitely two new HgMn stars. Their abundances differ from those of χ Lupi A possibly because their initial abundances differed slightly and because radiative diffusion operated in a slightly different manner in the younger and more massive stars HD 30085 and HD 30963. For HD 174567, the mild deficiencies for light elements and overabundances for strontium, yttrium, zirconium, several rare earths, and mercury suggest that this object should be reclassified as a mild CP star. More observations of this interesting star will help elucidate its nature.

R.M. thanks Pr. Charles Cowley for his insightful comments during the analysis of HD 30085 and HD 174567. We thank the OHP night assistants for their helpful support during the three observing runs. This work has made use of the VALD database, operated at Uppsala University, the Institute of Astronomy RAS in Moscow, and the University of Vienna. We have also used the NIST Atomic Spectra Database (version 5.4) available http://physics.nist.gov/asd. We also acknowledge the use of the ELODIE archive at OHP available at http://atlas.obs-hp.fr/elodie/.

Appendix: Determination of Uncertainties

For a representative line of a given element, six major sources are included in the uncertainty determinations: the uncertainty on the effective temperature ( ${\sigma }_{{T}_{\mathrm{eff}}}$ ), on the surface gravity (σ_{log g}), on the microturbulent velocity ( ${\sigma }_{{\xi }_{t}}$ ), on the apparent rotational velocity ( ${\sigma }_{{v}_{e}\sin i}$ ), the oscillator strength (σ_{log gf}), and the continuum placement (σ_cont). These uncertainties are supposed to be independent, so that the total uncertainty ${\sigma }_{{\mathrm{tot}}_{i}}$ for a given transition (i) verifies

$\begin{eqnarray}&&{\sigma }_{{\mathrm{tot}}_{i}}^{2}={\sigma }_{{T}_{\mathrm{eff}}}^{2}+{\sigma }_{\mathrm{log}g}^{2}+{\sigma }_{{\xi }_{t}}^{2}+{\sigma }_{{v}_{e}\sin i}^{2}+{\sigma }_{\mathrm{log}{gf}}^{2}+{\sigma }_{\mathrm{cont}}^{2}.\end{eqnarray} \tag{ 1 }$

The mean abundance $\left\langle \left[\tfrac{{\rm{X}}}{{\rm{H}}}\right]\right\rangle$ is then computed as a weighted mean of the individual abundances [X/H]_i derived for each transition (i):

$\begin{eqnarray}&&\left\langle \left[\displaystyle \frac{{\rm{X}}}{{\rm{H}}}\right]\right\rangle =\displaystyle \frac{{\displaystyle \sum }_{i}\left({\left[\tfrac{{\rm{X}}}{{\rm{H}}}\right]}_{i}/{\sigma }_{{\mathrm{tot}}_{i}}^{2}\right)}{{\displaystyle \sum }_{i}\left(1/{\sigma }_{{\mathrm{tot}}_{i}}^{2}\right)}\end{eqnarray} \tag{ 2 }$

and the total error, σ, is given by

$\begin{eqnarray}&&\displaystyle \frac{1}{{\sigma }^{2}}=\displaystyle \sum _{i=1}^{N}(1/{\sigma }_{{\mathrm{tot}}_{i}}^{2}),\end{eqnarray} \tag{ 3 }$

where N is the number of lines per element. The uncertainties σ for each element are collected in Table 7.

Table 7. Abundance Uncertainties for the Elements Analyzed in HD 30085, HD 30963, HD 174567, and HD 141556

		He i	C ii	O i	Mg ii	Al ii	Si ii	P ii	S ii	Ca ii
ΔT_eff	+200 K	−0.30	0.00	0.05	0.047	0.075	−0.017	0.00	0.00	0.109
Δ log g	0.15 dex	0.079	0.24	−0.06	−0.052	0.118	−0.017	−0.12	−0.12	−0.032
Δξ_t	+0.20 km s⁻¹	0.00	0.00	0.00	0.00	0.075	−0.035	−0.05	−0.046	0.00
Δ log gf	+0.10	−0.30	−0.12	−0.16	−0.25	−0.028	−0.053	−0.07	−0.071	−0.105
Δcontinuum		−0.12	0.21	0.075	0.13	0.25	0.062	0.06	0.061	0.058

2σ_[X/H]		0.45	0.34	0.19	0.29	0.20	0.092	0.160	0.159	0.165

		Sc ii	Ti ii	V ii	Cr ii	Mn ii	Fe ii	Ni ii	Sr ii	Y ii

ΔT_eff	+200 K	0.22	0.041	0.028	0.057	0.00	0.079	0.176	0.138	0.146
Δ log g	0.15 dex	−0.079	0.015	−0.014	0.03	0.007	0.041	0.114	−0.058
Δξ_t	+0.20 km s⁻¹	−0.038	−0.016	0.00	0.00	0.00	−0.095	0.00	−0.058	0.00
Δ log gf	+0.10	−0.13	−0.23	−0.046	−0.15	−0.015	−0.095	−0.155	−0.058	−0.05
Δcontinuum		0.067	0.03	0.014	0.046	0.014	0.0126	0.079	0.051	0.041

2σ_[X/H]		0.276	0.064	0.23	0.169	0.076	0.162	0.273	0.178	0.165

		Zr ii	Ba ii	La ii	Ce ii	Pr iii	Nd iii	Sm ii	Eu ii	Gd ii

ΔT_eff	+200 K	0.058	0.114	0.11	0.11	0.11	0.11	0.11	0.11	0.11
Δ log g	0.15 dex	0.046	−0.032	−0.125	−0.13	−0.13	−0.13	−0.13	−0.13	−0.13
Δξ_t	+0.20 km s⁻¹	0.00	−0.097	−0.09	−0.09	−0.09	−0.09	−0.09	−0.09	−0.09
Δ log gf	+0.10	−0.066	−0.125	−0.13	−0.13	−0.13	−0.13	−0.13	−0.13	−0.13
Δcontinuum		0.146	−0.097	−0.09	−0.09	−0.09	−0.09	−0.09	−0.09	−0.09

2σ_[X/H]		0.173	0.251	0.25	0.25	0.25	0.25	0.25	0.25	0.25

		Dy ii	Tb ii	Ho ii	Er ii	Tm ii	Yb ii	Hf ii	Os ii	Pt ii

ΔT_eff	+200 K	0.11	0.11	0.11	0.11	−0.155	0.11	0.11	0.11	0.11
Δ log g	0.15 dex	−0.13	−0.13	−0.13	−0.13	0.041	−0.13	−0.13	−0.13	−0.13
Δξ_t	+0.20 km s⁻¹	−0.09	−0.09	−0.09	−0.09	−0.097	−0.09	−0.09	−0.09	−0.09
Δ log gf	+0.10	−0.13	−0.13	−0.13	−0.13	−0.108	−0.13	−0.13	−0.13	−0.13
Δcontinuum		−0.09	−0.09	−0.09	−0.09	0.079	−0.09	−0.09	−0.09	−0.09

2σ_[X/H]		0.25	0.25	0.25	0.25	0.23	0.25	0.25	0.25	0.25

		Au ii	Hg ii

ΔT_eff	+200 K	0.11	−0.155
Δ log g	0.15 dex	−0.13	0.041
Δξ_t	+0.20 km s⁻¹	−0.09	−0.097
Δ log gf	+0.10	−0.13	−0.108
Δcontinuum		−0.09	0.079

2σ_[X/H]		0.25	0.23

Download table as: ASCIITypeset images: 1 2

Table 8. Elemental Abundances from Unblended Lines for HD 30085, HD 30963, HD 174567, and HD 141556

Element	λ (Å)	log gf	Reference	$\left\langle \mathrm{log}({\rm{X}}/{\rm{H}})\right\rangle$
				HD 30085	HD 30963	HD 174567	Chi Lupi A	σ_[X/H]
He i	4471.470	−2.211	NIST/ASD	−2.07	−1.77	−1.37	−1.67
He i	4471.474	−0.287	NIST/ASD	−2.07	−1.77	−1.37	−1.67
He i	4471.474	−1.035	NIST/ASD	−2.07	−1.77	−1.37	−1.67
He i	4471.485	−1.035	NIST/ASD	−2.07	−1.77	−1.37	−1.67
He i	4471.489	−0.558	NIST/ASD	−2.07	−1.77	−1.37	−1.67
He i	4471.683	−0.910	NIST/ASD	−2.07	−1.77	−1.37	−1.67
He i	5875.598	−1.516	NIST/ASD	−2.05	−1.75	−1.35	−1.65
He i	5875.613	−0.339	NIST/ASD	−2.05	−1.75	−1.35	−1.65
He i	5875.615	0.409	NIST/ASD	−2.05	−1.75	−1.35	−1.65
He i	5875.625	−0.339	NIST/ASD	−2.05	−1.75	−1.35	−1.65
He i	5875.640	0.138	NIST/ASD	−2.05	−1.75	−1.35	−1.65
He i	5875.966	−0.214	NIST/ASD	−2.05	−1.75	−1.35	−1.65

$\left\langle \mathrm{log}(\mathrm{He}/{\rm{H}})\right\rangle$				−2.07	−1.76	−1.36	−1.67	±0.23
$\left\langle \left[\tfrac{\mathrm{He}}{{\rm{H}}}\right]\right\rangle$				−1.00	−0.69	−0.29	−0.60	±0.23

C ii	4267.001	0.563	NIST/ASD	−3.63	−3.48	−3.78	−3.78
C ii	4267.261	−0.584	NIST/ASD	−3.63	−3.48	−3.78	−3.78
C ii	4267.261	0.716	NIST/ASD	−3.63	−3.48	−3.78	−3.70

$\left\langle \mathrm{log}({\rm{C}}/{\rm{H}})\right\rangle$				−3.63	−3.48	−3.78	−3.76	±0.17
$\left\langle \left[\tfrac{{\rm{C}}}{{\rm{H}}}\right]\right\rangle$				−0.15	0.00	−0.30	−0.28	±0.17

O i	6155.961	−1.363	NIST/ASD	−3.32	−3.17	−3.06	−3.63
O i	6155.971	−1.011	NIST/ASD	−3.32	−3.17	−3.06	−3.63
O i	6155.989	−1.120	NIST/ASD	−3.32	−3.17	−3.06	−3.63
O i	6156.737	−1.487	NIST/ASD	−3.32	−3.17	−3.06	−3.47
O i	6155.755	−0.898	NIST/ASD	−3.26	−3.11	−3.00	−3.47
O i	6156.778	−0.694	NIST/ASD	−3.32	−3.17	−3.06	−3.47
O i	6158.149	−1.841	NIST/ASD	−3.32	−3.17	−3.06	−3.63
O i	6158.172	−0.995	NIST/ASD	−3.32	−3.17	−3.06	−3.63
O i	6158.187	−0.409	NIST/ASD	−3.32	−3.17	−3.06	−3.63

$\left\langle \mathrm{log}({\rm{O}}/{\rm{H}})\right\rangle$				−3.32	−3.17	−3.06	−3.53	±0.10
$\left\langle \left[\tfrac{{\rm{O}}}{{\rm{H}}}\right]\right\rangle$				−0.15	0.00	0.11	−0.36	±0.10

Na i	4497.657	−1.574	NIST/ASD	−4.97	−5.67 (u.l)	−5.67 (u.l)	−4.97
Na i	4497.657	−2.528	NIST/ASD	−4.97	−5.67 (u.l)	−5.67 (u.l)	−4.97
Na i	5889.950	0.108	NIST/ASD	−5.61	−6.27	−6.27	−5.57
Na i	5895.924	−0.194	NIST/ASD	−5.61	−6.27	−6.27	−5.57

$\left\langle \mathrm{log}(\mathrm{Na}/{\rm{H}})\right\rangle$				−4.97	−5.67	−5.67	−4.97	⋯
$\left\langle \left[\tfrac{\mathrm{Na}}{{\rm{H}}}\right]\right\rangle$				0.70	0.00	0.00	0.7	⋯

Mg ii	4390.572	−0.523	NIST/ASD	−4.64	−4.64	−4.42	−4.72
Mg ii	4427.994	−1.208	NIST/ASD	−4.64	−4.52	−4.42	−4.64
Mg i	5172.684	−0.393	NIST/ASD	−4.97	−4.91	−4.75	−5.01

$\left\langle \mathrm{log}(\mathrm{Mg}/{\rm{H}})\right\rangle$				−4.64	−4.58	−4.42	−4.66	±0.15
$\left\langle \left[\tfrac{\mathrm{Mg}}{{\rm{H}}}\right]\right\rangle$				−0.22	−0.16	0.00	−0.24	±0.15

Al ii	4663.056	−0.244	NIST/ASD	−6.53	−7.53	−5.53	−5.79

$\left\langle \mathrm{log}(\mathrm{Al}/{\rm{H}})\right\rangle$				−6.53	−7.53	−5.53	−5.79	±0.10
$\left\langle \left[\tfrac{\mathrm{Al}}{{\rm{H}}}\right]\right\rangle$				−1.00	−2.00	0.00	−0.26	±0.10

Si ii	4128.054	0.359	NIST/ASD	−4.30	−4.37	−4.45	−4.57
Si ii	4130.894	0.552	NIST/ASD	−4.32	−4.45	−4.45	−4.57
Si ii	5041.024	0.029	NIST/ASD	−4.15	−4.27	−4.15	bl.
Si ii	5055.984	0.523	NIST/ASD	−4.30	−4.35	−4.35	−4.60
Si ii	5056.317	−0.492	NIST/ASD	−4.30	−4.35	−4.30	−4.45
Si ii	6371.37	−0.082	NIST/ASD	−4.15	−4.15	bl.	−4.55

$\left\langle \mathrm{log}(\mathrm{Si}/{\rm{H}})\right\rangle$				−4.25	−4.32	−4.34	−4.55	±0.05
$\left\langle \left[\tfrac{\mathrm{Si}}{{\rm{H}}}\right]\right\rangle$				0.20	0.13	0.11	−0.09	±0.05

P ii	6024.18	0.137	NIST/ASD	−5.55	−5.55	bl.	−5.47
P ii	6043.13	0.384	NIST/ASD	−5.55	−5.55	−5.85	−5.70

$\left\langle \mathrm{log}({\rm{P}}/{\rm{H}})\right\rangle$				−5.55	−5.55	−5.85	−5.47	±0.08
$\left\langle \left[\tfrac{{\rm{P}}}{{\rm{H}}}\right]\right\rangle$				1.00	1.00	0.70	1.08	±0.08

S ii	4153.06	0.395	NIST/ASD	−4.67	−4.19	−4.67	−4.27
S ii	4162.31	0.161	NIST/ASD	−4.67	−4.37	−4.76	⋯
S ii	4162.67	0.830	NIST/ASD	−4.67	−4.37	−4.76	−5.65

$\left\langle \mathrm{log}({\rm{S}}/{\rm{H}})\right\rangle$				−4.67	−4.31	−4.73	−4.96	±0.08
$\left\langle \left[\tfrac{{\rm{S}}}{{\rm{H}}}\right]\right\rangle$				0.00	0.36	−0.06	0.72	±0.08

Ca ii	3933.663	0.135	NIST/ASD	−5.54	−5.46	−5.56	−5.94
Ca ii	5019.971	−0.280	NIST/ASD	−5.64	−6.03	−5.64	−5.83

$\left\langle \mathrm{log}(\mathrm{Ca}/{\rm{H}})\right\rangle$				−5.59	−5.74	−5.60	−5.83	±0.08
$\left\langle \left[\tfrac{\mathrm{Ca}}{{\rm{H}}}\right]\right\rangle$				0.05	−0.10	0.04	−0.19	±0.08

Sc ii	4246.822	0.242	NIST/ASD	−8.53	−8.13	−8.72	−10.13
Sc ii	5031.021	−0.400	NIST/ASD	bl.	bl.	bl.	⋯
Sc ii	5526.813	−0.77	NIST/ASD	bl.	bl.	bl.	no data

$\left\langle \mathrm{log}(\mathrm{Sc}/{\rm{H}})\right\rangle$				−8.53	−8.13	−8.72	−10.13	±0.14
$\left\langle \left[\tfrac{\mathrm{Sc}}{{\rm{H}}}\right]\right\rangle$				0.30	0.70	0.11	−1.30	±0.14

Ti ii	4163.644	−0.128	NIST/ASD	−6.24	−6.20	−6.64	−6.58
Ti ii	4287.873	−2.020	NIST/ASD	−6.28	−6.50	−6.68	−6.68
Ti ii	4290.210	−0.848	NIST/ASD	−6.33	−6.28	⋯	−6.68
Ti ii	4300.042	−0.442	NIST/ASD	−6.03	−6.03	−6.44	⋯
Ti ii	4411.072	−0.6767	NIST/ASD	−6.20	−6.13	−6.50	−6.44
Ti ii	4468.492	−0.620	NIST/ASD	−6.24	−6.24	−6.68	−7.03
Ti ii	4563.758	−0.96	NIST/ASD	−6.20	−6.24	−6.58	−6.74
Ti ii	5129.156	−1.239	NIST/ASD	−6.38	−6.24	−6.75	−6.68
Ti ii	5188.687	−1.220	NIST/ASD	−6.38	−6.24	−6.72	−6.74
Ti ii	5336.780	−1.700	NIST/ASD	−6.44	−6.24	−6.75	−6.74

$\left\langle \mathrm{log}(\mathrm{Ti}/{\rm{H}})\right\rangle$				−6.27	−6.23	−6.63	−6.73	±0.06
$\left\langle \left[\tfrac{\mathrm{Ti}}{{\rm{H}}}\right]\right\rangle$				0.71	0.75	0.35	0.25	±0.06

V ii	4005.71	−0.450	NIST/ASD	bl.	bl.	−7.66	−8.07
V ii	4023.39	−0.610	K10/BGF/WLDS	−8.70	−8.00 (u.l)	−7.50	−8.00
V ii	4036.78	−1.570	NIST/ASD	−8.30	−7.00	−7.60	−8.00

$\left\langle \mathrm{log}({\rm{V}}/{\rm{H}})\right\rangle$				−8.50	−7.50	−7.59	−8.02	±0.12
$\left\langle \left[\tfrac{{\rm{V}}}{{\rm{H}}}\right]\right\rangle$				−0.50	0.50	0.41	−0.02	±0.12

Cr ii	4558.644	−0.660	NIST/ASD	−5.43	−5.55	−5.85	−6.13
Cr ii	4558.787	−2.460	SN14	−5.43	−5.55	−5.85	−6.13
Cr ii	5237.322	−1.160	NIST/ASD	−5.55	−5.85	−6.13	−6.23
Cr ii	5308.421	−1.810	NIST/ASD	−5.63	−5.85	−6.18	−6.15
Cr ii	5313.581	−1.650	NIST/ASD	−5.55	−5.63	−6.10	−6.15
Cr ii	5502.086	−1.990	NIST/ASD	−5.55	−5.68	−6.03	nd

$\left\langle \mathrm{log}(\mathrm{Cr}/{\rm{H}})\right\rangle$				−5.52	−5.68	−6.02	−6.14	±0.09
$\left\langle \left[\tfrac{\mathrm{Cr}}{{\rm{H}}}\right]\right\rangle$				0.81	0.65	0.31	0.19	±0.09

Mn ii	4206.368	hfs	NIST/ASD	−4.87	−4.76	−6.31	−6.13
Mn ii	4259.19	hfs	NIST/ASD	−4.87	−4.71	−6.31	−6.13

$\left\langle \mathrm{log}(\mathrm{Mn}/{\rm{H}})\right\rangle$				−4.87	−4.73	−6.31	−6.13	±0.08
$\left\langle \left[\tfrac{\mathrm{Mn}}{{\rm{H}}}\right]\right\rangle$				1.74	1.88	0.30	0.48	±0.08

Fe ii	4258.148	−3.500	NIST/ASD	bl.	bl.	bl.	−4.40
Fe ii	4273.326	−3.350	NIST/ASD	−4.32	−4.30	−4.38	−4.50
Fe ii	4296.566	−2.900	NIST/ASD	−4.20	−4.20	−4.38	−4.40
Fe ii	4491.397	−2.640	NIST/ASD	−4.20	−4.20	−4.38	−4.60
Fe ii	4508.280	−2.300	NIST/ASD	−4.20	−4.20	−4.40	−4.72
Fe ii	4515.333	−2.360	NIST/ASD	−4.20	−4.20	−4.38	−4.60
Fe ii	4520.218	−2.600	NIST/ASD	−4.16	−4.16	−4.35	−4.57
Fe ii	4923.921	−1.210	NIST/ASD	−4.10	−4.10	−4.38	−4.62
Fe ii	5275.997	−1.900	NIST/ASD	−4.20	−4.20	−4.38	−4.72
Fe ii	5316.609	−1.780	NIST/ASD	−4.15	−4.15	−4.38	−4.65
Fe ii	5506.199	0.860	NIST/ASD	−4.24	−4.20	−4.38	nd

$\left\langle \mathrm{log}(\mathrm{Fe}/{\rm{H}})\right\rangle$				−4.19	−4.19	−4.38	−4.57	±0.08
$\left\langle \left[\tfrac{\mathrm{Fe}}{{\rm{H}}}\right]\right\rangle$				0.31	0.31	0.12	−0.07	±0.08

Ni ii	4067.04	−1.834	K03	−7.45	−6.75	−5.27	−6.52

$\left\langle \mathrm{log}(\mathrm{Ni}/{\rm{H}})\right\rangle$				−7.45	−6.75	−5.27	−6.52	±0.14
$\left\langle \left[\tfrac{\mathrm{Ni}}{{\rm{H}}}\right]\right\rangle$				−1.70	−1.00	0.48	−0.77	±0.14

Ga ii	4251.108	hfs	Dwo98	−9.12 (u.l)	bl.	−9.12 (u.l)	−9.12 (u.l)
Ga ii	4254.032	hfs	Dwo98	−9.12 (u.l)	bl.	−9.12 (u.l)	−9.12 (u.l)
Ga ii	4255.640	hfs	Dwo98	−9.12 (u.l)	−5.88	−9.12 (u.l)	−9.12 (u.l)
Ga ii	4261.995	hfs	Dwo98	−9.12 (u.l)	bl.	−9.12 (u.l)	−9.12 (u.l)
Ga ii	5360.313	hfs	Niel2000	−5.88	−5.88	−9.12 (u.l)	−9.12 (u.l)
Ga ii	5363.353	hfs	Niel2000	bl.	bl.	−9.12 (u.l)	−9.12 (u.l)
Ga ii	5421.122	hfs	Niel2000	bl.	bl.	−9.12 (u.l)	−9.12 (u.l)
Ga ii	6334.07	hfs	Lanz1993	−5.88	−5.88	−9.12 (u.l)	−9.12 (u.l)

$\left\langle \mathrm{log}(\mathrm{Ga}/{\rm{H}})\right\rangle$				−9.12	−5.88	−9.12	−9.12	⋯
$\left\langle \left[\tfrac{\mathrm{Ga}}{{\rm{H}}}\right]\right\rangle$				0.00	3.24	0.00	0.00	⋯

Sr ii	4215.519	−1.610	NIST/ASD	−7.43	−8.08	−8.55	−7.43

$\left\langle \mathrm{log}(\mathrm{Sr}/{\rm{H}})\right\rangle$				−7.43	−8.08	−8.55	−7.43	±0.09
$\left\langle \left[\tfrac{\mathrm{Sr}}{{\rm{H}}}\right]\right\rangle$				1.60	0.95	0.48	1.60	±0.09

Y ii	3982.592	−0.560	NIST/ASD	−6.56	−6.16	−8.56	−8.16
Y ii	5662.922	0.340	Bie11	−6.56	−6.16	−8.56	−7.98

$\left\langle \mathrm{log}({\rm{Y}}/{\rm{H}})\right\rangle$				−6.56	−6.16	−8.56	−8.06	±0.08
$\left\langle \left[\tfrac{{\rm{Y}}}{{\rm{H}}}\right]\right\rangle$				3.20	3.60	1.20	1.70	±0.08

Zr ii	3998.965	−0.520	L06	−7.10	bl.	−8.70	−8.70
Zr ii	4442.992	−0.420	L06	−7.10	bl.	−8.70	−8.70
Zr ii	4457.431	−1.220	L06	−7.40	bl.	−8.40	−8.40
Zr ii	5112.270	−0.850	L06	−6.92	−7.70	−8.40	bl.

$\left\langle \mathrm{log}(\mathrm{Zr}/{\rm{H}})\right\rangle$				−7.13	−7.70	−8.55	−8.70	±0.09
$\left\langle \left[\tfrac{\mathrm{Zr}}{{\rm{H}}}\right]\right\rangle$				2.27	1.70	0.85	0.70	±0.09

Ba ii	4554.04	hfs	VALD??	−8.79	−8.69	−8.87	−9.17
Ba ii	4934.077	hfs	VALD??	−8.79	bl.	−8.87	−9.17
Ba ii	6141.713	hfs	VALD??	−8.79	−8.57	−8.87	−9.17

$\left\langle \mathrm{log}(\mathrm{Ba}/{\rm{H}})\right\rangle$				−8.79	−8.63	−8.87	−9.17	±0.13
$\left\langle \left[\tfrac{\mathrm{Ba}}{{\rm{H}}}\right]\right\rangle$				1.08	1.24	1.00	0.70	±0.13

La ii	4042.91	0.33	Zi99	−10.83 (u.l)	bl.	−9.13	−9.13

$\left\langle \mathrm{log}(\mathrm{La}/{\rm{H}})\right\rangle$				−10.83	−	−9.13	−9.13	±0.13
$\left\langle \left[\tfrac{\mathrm{La}}{{\rm{H}}}\right]\right\rangle$				0.00	−	1.70	1.70	±0.13

Ce ii	4133.80	0.72	LSCI	−8.42	−10.42 (u.l)	−10.42 (u.l)	bl.
Ce ii	4460.21	0.28	PQWB	−8.42	−10.42 (u.l)	−10.42 (u.l)	−8.90

$\left\langle \mathrm{log}(\mathrm{Ce}/{\rm{H}})\right\rangle$				−8.42	−10.42	−10.42	−8.90	±0.13
$\left\langle \left[\tfrac{\mathrm{Ce}}{{\rm{H}}}\right]\right\rangle$				2.00	0.00	0.00	1.52	±0.13

Pr iii	5264.44	−0.66	ISAN	bl.	bl.	bl.	−9.09
Pr iii	5284.69	−0.59	ISAN	−8.99	bl.	−9.99	bl.
Pr iii	5299.99	−0.53	ISAN	−8.89	bl.	−9.59	bl.

$\left\langle \mathrm{log}(\Pr /{\rm{H}})\right\rangle$				−8.94	−	−9.79	−9.09	±0.13
$\left\langle \left[\tfrac{\Pr }{{\rm{H}}}\right]\right\rangle$				2.35	−	1.50	2.20	±0.13

Nd iii	5265.019	−0.66	RRKB	−8.65	bl.	−9.50 (u.l)	−8.80
Nd iii	5294.10	−0.65	RRKB	−8.10	−8.50	−9.50	* bl.

$\left\langle \mathrm{log}(\mathrm{Nd}/{\rm{H}})\right\rangle$				−8.37	−8.50	−9.50	−8.80	±0.13
$\left\langle \left[\tfrac{\mathrm{Nd}}{{\rm{H}}}\right]\right\rangle$				2.13	2.00	1.00	1.70	±0.13

Sm ii	4280.79	−0.33	LA06	−10.99 (u.l)	−10.99 (u.l)	−10.99 (u.l)	bl.
Sm ii	4424.32	0.14	LA06	−10.99 (u.l)	−10.99 (u.l)	noisy	2.40

$\left\langle \mathrm{log}(\mathrm{Sm}/{\rm{H}})\right\rangle$				−10.99	−10.99	−10.99	−8.59	±0.13
$\left\langle \left[\tfrac{\mathrm{Sm}}{{\rm{H}}}\right]\right\rangle$				0.00	0.00	0.00	2.40	±0.13

Eu ii	4129.70	0.19	LWHS	−11.49 (u.l)	−11.49 (u.l)	−11.49 (u.l)	−9.61

$\left\langle \mathrm{log}(\mathrm{Eu}/{\rm{H}})\right\rangle$				−11.49	−11.49	−11.49	−9.61	±0.13
$\left\langle \left[\tfrac{\mathrm{Eu}}{{\rm{H}}}\right]\right\rangle$				0.00	0.00	0.00	1.88	±0.13

Gd ii	4037.32	−0.11	DH06	−8.03	bl.	−8.88	−8.70

$\left\langle \mathrm{log}(\mathrm{Gd}/{\rm{H}})\right\rangle$				−8.03	−	−8.88	−8.70	±0.13
$\left\langle \left[\tfrac{\mathrm{Gd}}{{\rm{H}}}\right]\right\rangle$				2.85	−	2.00	2.18	±0.13

Tb ii	4005.47	−0.02	LA01	−12.10 (ul)	−12.10 (u.l)	bl.	bl.
Tb ii	4033.03	−0.01	LA01	bl.	bl.	bl.	bl.

$\left\langle \mathrm{log}(\mathrm{Tb}/{\rm{H}})\right\rangle$				−12.10	−12.10	bl.	bl.	⋯
$\left\langle \left[\tfrac{\mathrm{Tb}}{{\rm{H}}}\right]\right\rangle$				0.00	0.00	bl.	bl.	⋯

Dy ii	4000.45	0.040	NIST	−10.86 (u.l)	bl.	−9.08	−9.08

$\left\langle \mathrm{log}(\mathrm{Dy}/{\rm{H}})\right\rangle$				−10.86	−	−9.08	−9.08	±0.13
$\left\langle \left[\tfrac{\mathrm{Dy}}{{\rm{H}}}\right]\right\rangle$				0.00	−	1.78	1.78	±0.13

Ho ii	4152.62	−0.93	LA04	−11.74 (u.l)	−11.74 (u.l)	−11.74 (u.l)	−9.50

$\left\langle \mathrm{log}(\mathrm{Ho}/{\rm{H}})\right\rangle$				−11.74	−11.74	−11.74	−9.50	±0.13
$\left\langle \left[\tfrac{\mathrm{Ho}}{{\rm{H}}}\right]\right\rangle$				0.00	0.00	0.00	2.24	±0.13

Er ii	4142.91	−0.72	LA08	−11.07 (u.l)	bl.	−8.77	bl.

$\left\langle \mathrm{log}(\mathrm{Er}/{\rm{H}})\right\rangle$				−11.07	−	−8.77	bl.	⋯
$\left\langle \left[\tfrac{\mathrm{Er}}{{\rm{H}}}\right]\right\rangle$				0.00	−	2.30	bl.	⋯

Tm ii	4242.15	−0.95	NIST	bl.	−12.0 (u.l)	−12.00 (u.l)	bl.

$\left\langle \mathrm{log}(\mathrm{Tm}/{\rm{H}})\right\rangle$				⋯	−12.00	−12.00	bl.	⋯
$\left\langle \left[\tfrac{\mathrm{Tm}}{{\rm{H}}}\right]\right\rangle$				⋯	0.00	0.00	bl.	⋯

Yb ii	4135.095	−0.21	?	−10.92 (u.l)	bl.	−10.92 (u.l)	−10.92 (u.l)

$\left\langle \mathrm{log}(\mathrm{Yb}/{\rm{H}})\right\rangle$				−10.92	−	−10.92	−10.92	±0.13
$\left\langle \left[\tfrac{\mathrm{Yb}}{{\rm{H}}}\right]\right\rangle$				0.00	−	0.00	0.00	±0.13

Hf ii	3918.08	−1.140	?	−11.12 (u.l)	bl.	−11.12 (u.l)	bl.

$\left\langle \mathrm{log}(\mathrm{Hf}/{\rm{H}})\right\rangle$				−11.12	−	−11.12	bl.	⋯
$\left\langle \left[\tfrac{\mathrm{Hf}}{{\rm{H}}}\right]\right\rangle$				0.00	−	0.00	bl.	⋯

Os ii	4158.44	−2.68	?	−10.55 (u.l)	bl.	−10.55 (u.l)	−10.55 (u.l)*

$\left\langle \mathrm{log}(\mathrm{Os}/{\rm{H}})\right\rangle$				−10.55	−	−10.55	−10.55	±0.13
$\left\langle \left[\tfrac{\mathrm{Os}}{{\rm{H}}}\right]\right\rangle$				0.00	−	0.00	0.00	±0.13

Pt ii	4514.17	−1.48	?	−6.50	−7.35	−10.20 (u.l)	bl.

$\left\langle \mathrm{log}(\mathrm{Pt}/{\rm{H}})\right\rangle$				−6.50	−7.35	−10.20	bl.	⋯
$\left\langle \left[\tfrac{\mathrm{Pt}}{{\rm{H}}}\right]\right\rangle$				3.70	2.85	0.00	bl.	⋯

Au ii	4052.79	−1.68	?	−10.99 (u.l)	bl.	−10.99 (u.l)	−10.99 (u.l)

$\left\langle \mathrm{log}(\mathrm{Au}/{\rm{H}})\right\rangle$				−10.99	−	−10.99	−10.99	±0.13
$\left\langle \left[\tfrac{\mathrm{Au}}{{\rm{H}}}\right]\right\rangle$				0.00	−	0.00	0.00	±0.13

Hg ii	3983.931	hfs	Do03	−5.12	−5.31	−7.61	* −5.91

$\left\langle \mathrm{log}(\mathrm{Hg}/{\rm{H}})\right\rangle$				−5.12	−5.31	−7.61	−5.91	±0.12
$\left\langle \left[\tfrac{\mathrm{Hg}}{{\rm{H}}}\right]\right\rangle$				5.75	5.56	3.26	5.00	±0.12

Note. References are defined in Table 5.

Download table as: ASCIITypeset images: 1 2 3 4 5

Table 9. Bibliographical References for the Atomic Data

Element	Bibliographical Source
He i	Ma60 = Martin (1960)
He i	DR2006 = Drake (2006)
C ii	NIST/ASD = Kramida et al. (2018)
O i	KZ91 = Butler & Zeippen (1991)
O i	W96 = Wiese et al. (1996)
O i	NIST/ASD = Kramida et al. (2018)
Mg ii	NIST/ASD = Kramida et al. (2018)
Mg ii	NIST/ASD = Kramida et al. (2018)
Al ii	NIST/ASD = Kramida et al. (2018)
Al i	NIST/ASD = Kramida et al. (2018)
Si ii	NIST/ASD = Kramida et al. (2018)
Si ii	Sh61 = Shenstone (1961)
S ii	GUES = Kurucz (1993)
S ii	KP = Kurucz & Peytremann (1975)
S ii	MWRB = Miller et al. (1974)
Ca ii	NIST = "http://physics.nist.gov/PhysRefData/ASD/"
Ca ii	T89 = Theodosiou (1989)
Sc ii	NIST/ASD = Kramida et al. (2018)
Ti ii	NIST/ASD = Kramida et al. (2018)
Ti ii	NIST/ASD = Kramida et al. (2018)
Ti ii	NIST/ASD = Kramida et al. (2018)
Ti ii	NIST/ASD = Kramida et al. (2018)
V ii	K10 = Kurucz (2010)
V ii	BGF = Biemont et al. (1989)
V ii	WLDSC = Wood et al. (2014)
Cr ii	SN14 = Sansonetti & Nave (2014)
Mn ii	KL01 = Kling et al. (2001)
Fe ii	NIST/ASD = Kramida et al. (2018)
Fe ii	RU98 = Raassen & Uylings (1998)
Ni ii	K03 = Kurucz (2003)
Ga ii	Dwo98 = Dworetsky et al. (1998)
Ga ii	Niel2000 = Nielsen et al. (2000)
Ga ii	Lanz1993 = Lanz et al. (1993)
Sr ii	PBL95 = Pinnington et al. (1995)
Y ii	Bie11 = Biémont et al. (2011)
Y ii	MCS75 = Meggers et al. (1975)
Zr ii	L06 = Ljung et al. (2006)
Ba ii	NBS = Miles & Wiese (1969)
Ba ii	DSVD92 = Davidson et al. (1992)
La ii	Zi99 = Zhiguo et al. (1999)
Ce ii	LSCI = Lawler et al. (2009)
Ce ii	PQWB = Palmeri et al. (2000)
Pr iii	ISAN = A. N. Ryabtsev (2010, private communication)
Nd iii	RRKB = Ryabchikova et al. (2006)
Sm ii	LA06 = Lawler et al. (2006)
Eu ii	LWHS = Lawler et al. (2001b)
Eu ii	ZLLZ = Zhiguo et al. (2000)
Gd ii	DH06 = Den Hartog et al. (2006)
Dy ii	WLN = Wickliffe et al. (2000)
Dy ii	MC = Meggers et al. (1975)
Gd ii	DH06 = Den Hartog et al. (2006)
Tb ii	LA01 = Lawler et al. (2001a)
Ho ii	LA04 = Lawler et al. (2004)
Er ii	LA08 = Lawler et al. (2008)
Hg ii	SR01 = Sansonetti & Reader (2001)
Hg ii	Do03 = Dolk et al. (2003)

Download table as: ASCII Typeset image

Table 10. Identifications of Lines Absorbing More than 2% for HD 30085

λ_obs (Å)	λ_lab (Å)	Species	log gf	E_low	Reference
3920.738	3920.627	Fe ii	−1.330	60628.698	NIST/VALD
3920.801	3920.690	C ii	−0.230	131735.525	NIST/VALD
3922.121	3922.010	Fe ii	−1.110	73606.162	NIST/VALD
3923.571	3923.460	S ii	0.440	130641.115	NIST/VALD
3924.953	3924.842	Fe ii	−1.100	78138.990	NIST/VALD
3926.527	3926.416	Mn ii	−1.600	55757.109	NIST/VALD
3930.407	3930.296	Fe i	−1.490	704.007	NIST/VALD
3930.417	3930.306	Fe ii	−4.300	13671.099	NIST/VALD
3931.063	3930.952	Mn ii	−2.040	52651.876	NIST/VALD
3932.120	3932.009	Ti ii	−1.640	9118.285	NIST/VALD

Only a portion of this table is shown here to demonstrate its form and content. A machine-readable version of the full table is available.

Download table as: Data Typeset image

The Chemical Compositions of the Two New HgMn Stars HD 30085 and HD 30963: Comparison to χ Lupi A, ν Cap, and HD 174567

Article metrics

Permissions

Share this article

Author affiliations

ORCID iDs

Dates

Abstract

1. Introduction

2. Recap of Previous Spectroscopic Work on HD 30085, HD 30963, HD 174567, ν Cap, and χ Lupi A

3. Observations

4. Model Atmosphere Analysis and Synthetic Spectra Computation

4.1. Fundamental Parameters

4.1.1. Microturbulent Velocity Determinations

4.1.2. Location of the Five Stars in a log g, log Teff Diagram

4.2. Model Atmospheres

4.3. The Line List

4.4. Spectrum Synthesis

5. Abundance Determinations

5.1. Helium

5.2. The Light Elements (C to Ti)

5.2.1. Carbon

5.2.2. Oxygen

5.2.3. Sodium

5.2.4. Magnesium

5.2.5. Aluminium

5.2.6. Silicon

5.2.7. Phosphorus

5.2.8. Sulfur

5.2.9. Calcium

5.2.10. Scandium

5.2.11. Titanium

5.3. The Iron-peak Elements (V to Zn) and Gallium

5.3.1. Vanadium

5.3.2. Chromium

5.3.3. Manganese

5.3.4. Iron

5.3.5. Nickel

5.3.6. Gallium

5.4. The Sr–Y–Zr Triad

5.4.1. Strontium

5.4.2. Yttrium

5.4.3. Zirconium

5.5. Barium

5.6. The Rare Earths

5.6.1. Lanthanum

5.6.2. Cerium

5.6.3. Praseodymium

5.6.4. Neodymium

5.6.5. Samarium

5.6.6. Europium

5.6.7. Gadolynium

5.6.8. Terbium

5.6.9. Dysprosium

5.6.10. Holmium

5.6.11. Erbium

5.6.12. Thulium

5.6.13. Ytterbium

5.6.14. Hafnium

5.7. The Very Heavy Elements Osmium, Platinum, Gold, and Mercury

5.7.1. Osmium

5.7.2. Platinum

5.7.3. Gold

5.7.4. Mercury

6. Discussion

7. Conclusion

Appendix: Determination of Uncertainties

Footnotes

4.1.2. Location of the Five Stars in a log g, log T_eff Diagram