Introduction

Planetary nebulae play a crucial role as significant representatives of the late-stage evolutionary pathway for low- to intermediate-mass stars, ultimately leading to the formation of white dwarfs as their remnants. The elemental abundances derived from PNe not only reflect the outcomes of internal nucleosynthesis processes within their parent stars but also mirror the interstellar elemental compositions present during the formation of the precursor stars. The materials expelled by PNe mix with the existing interstellar matter, initiating mechanisms of chemical enrichment. Although this enrichment is not as dramatic as that caused by supernova events, the materials produced by intermediate-mass stars are noteworthy due to the large number of objects involved, especially when compared to the relatively rare occurrences of supernovae. This contribution is particularly significant for elements such as He, C, N, Ne, and s-process elements. Therefore, accurate determinations of the chemical composition of PNe are critical for tracing the overall chemical evolution of our Milky Way galaxy.

One of the contemporary techniques for acquiring the spectra of PNe is the Integrated Field Unit (IFU) spectroscopy. The IFU spectra of PNe offer numerous advantages over traditional long-slit spectra. Researchers can analyze emission lines, velocity patterns, and the spatial distribution of various elements within PNe by applying IFU data, which comprises a three-dimensional data cube containing two spatial dimensions and one spectral dimension. This framework enables the extraction of detailed information regarding the kinematics, ionization structure, and abundance gradients present in these objects. In this paper, we present an IFU analysis of the planetary nebula IC 4642 (Shapley 2, PN G334.3-09.3).

IC 4642 has received limited attention in the literature. No detailed spectral analysis has been conducted except for the work of Kingsburgh and Barlow1. Their analysis was constrained by low spectral dispersion and a narrow optical wavelength range ( 3400Å  to 5200Å), which allowed them to identify only a few emission lines. The fluxes of these lines were adjusted for interstellar extinction using a coefficient value that is uncertain (see Table 31). Consequently, the findings regarding the physical conditions and chemical composition of IC 4642 are questionable.

Based on H\(\alpha\) and [O III] images2, the morphology of IC 4642 has been categorized as elliptical with multiple shells (EM class)3. Additionally, it has been proposed as a PN characterized by embedded knots within a bipolar structure (BE class)4. Take advantage of three-dimensional hydrodynamical numerical simulations5, reported that IC 4642 exhibits a morphology defined by bipolarity and point-symmetry.

The central star of IC 4642 does not have a definitive spectral classification; it is referred to in the literature as an absorption line star due to the presence of specific absorption features in its observed spectrum6,7,8. The temperature of the CS was derived through the Zanstra method, yielding \(T_z(\text {He II}) = 1.16 \times 10^5\) K and \(T_z(\text {H I}) = 0.62 \times 10^5\) K9. Additionally, it was determined using the energy-balance method, resulting in \(T = 0.996 \times 10^5\) K10.

The primary objective of this study is to conduct a detailed investigation of the planetary nebula IC 4642, which has received limited attention in the literature, particularly regarding its physical and chemical properties. Furthermore, it highlights the recently observed variability of its central star, as documented by the Gaia mission. However, a comprehensive analysis of this variability requires follow-up studies and is therefore not addressed within the scope of this work.

The article is organized as follows:  Sect. “Observations & data reduction” describes the observations and data reduction. Section “The morphology of IC 4642” illustrates the morphology and ionization structures observed in the object.  Section “Physical and chemical analysis” discusses the plasma diagnostics, physical conditions, and chemical composition of the PN.  Section “The variable CS of IC 4642” highlights the variability of the CS. Lastly, the conclusions drawn from the study are summarized in the final section.

Observations & data reduction

On 15th May 2018, the observations of IC 4642 were carried out using the Wide Field Spectrograph (WiFeS) instrument11 mounted on the 2.3-m telescope at the Australian National University (ANU) located at Siding Spring Observatory. WiFeS is a powerful spectrograph known for its impressive capabilities, including integral field, dual-beam, concentric, and image-slicing functionalities. It offers a substantial field of view measuring 25 arcsec \(\times\) 38 arcsec, accompanied by a remarkable spatial resolution of (1.0 arcsec\(\times\) 1.0 arcsec). The data were acquired in low-resolution mode (R\(\sim 3000\)), providing a full width at half-maximum resolution resolution of \(\sim\)100 km \(\hbox {s}^{-1}\) (\(\sim 1.5\) Å). During these observations, the CCD was configured in double binning mode, resulting in a data format of 4096\(\times\)2048 with 1.0 arcsec pixels in the spatial direction. Simultaneous observations were performed using two gratings, namely B3000 and R3000, with the RT560 dichroic that cuts at 5600Å  to ensure substantial wavelength coverage overlap. This setup enabled continuous wavelength coverage from \(\sim 3400\) to \(\sim 8950\,\)Å. The wavelength scale was calibrated using Ne-Ar arc lamp observations taken during nighttime, while the nebular flux was calibrated using the standard stars HD 111980 and HD 031128. The PyWiFeS pipeline12 was used for the overall steps of data reduction. Table 1 provides a summary of the WiFeS observations. The three integrations with different exposure times (10s, 100s, and 1000s) were taken to ensure that we captured both the bright and faint features of the nebula without saturating the detector. The shorter exposures (10s and 100s) were used to avoid saturation in the brightest regions, while the longest exposure (1000s) was necessary to detect the faintest emission lines and structures. The combination of these exposures allowed us to achieve a balanced dataset that covers the full dynamic range of the nebula’s emission.

Table 1 The observing log of IC 4642.
Full size table

The morphology of IC 4642

Figure 1
figure 1

A deep composite RGB image of IC 4642, captured using H\(\alpha\) and [O III] filters by the chart32.de team in June 2017. This image reveals a very faint, spherical outer halo, as well as several outer [O III] features that extend beyond the field of view of the WiFeS spectrograph. The overall size of the nebula, including its faint outer halo, is \(\sim 65\) arcsec. The plate scale has been overlaid on the image (private communication). In this and subsequent figures, the orientation maintains that the northern direction is upwards, and the eastern direction is to the left.

Full size image
Figure 2
figure 2

The emission-line maps of IC 4642 presented in the left, middle, and right panels focus on different ionization states of argon: [Ar III] at 7134 Å, [Ar IV] at 4710 Å, and [Ar V] at 7005 Å. These maps effectively illustrate the variations in inner nebular emissions and ionization gradients. As one moves from the lower ionization states shown in the [Ar III] map to the higher ionization states represented in the [Ar V] map, a slight reduction in the size of the nebula is observed. Throughout the maps, an elliptical shape is consistently observed.

Full size image
Figure 3
figure 3

The figure displays three images of IC 4642, highlighting emission lines of H\(\alpha\) (left panel), He I at 3888 Å (middle panel), and He II at 4685 Å (right panel). They all show roughly the same structures.

Full size image
Figure 4
figure 4

The emission-line maps of IC 4642 in three different ions: [Ne III] at 3868 Å, [O III] at 5007 Å, and [Cl IV] at 8045 Å. All maps consistently display the same structures seen in Figures 2 and 3. However, there are four illuminated corners that align with the boundaries of the two inner elongated lobes.

Full size image

Figure 1 illustrates a deep and high-resolution composite color image of IC 4642, obtained using H\(\alpha\) and [O III] filters. This image was captured by the Chilean Advanced Robotic Telescope (CHART32) team [Johannes Schedler, Philipp Keller, Konstantin Buchhold, Volker Wendel, and Bernd Flach-Wilken; (https://www.chart32.de/]) in June 2017. In addition to the multiple ionized shells of the PN, the image unveils a faint spherical outer halo for the first time. This halo is associated with enhanced emission towards the north and patchy emission towards the southwest, both of which are more pronounced in the [O III] filter. This faint halo represents the remnants from the asymptotic giant branch (AGB) phase, which are expelled during the later stages of this evolutionary phase. According to the stellar-wind model13, stars lose their outer envelopes through a slow but intense outflow known as the AGB wind. This mass loss is driven by substantial thermal pulsations occurring in AGB stars.

The faint outer halo and its associated features lie outside the field of view of the WiFeS instrument, making it impossible to directly measure the kinematical properties of the halo, including its velocity. Based on the plate scale of the image, we estimate that the total size of the nebula, including its faint outer halo, is \(\sim 65\) arcsec. Future observations with a wider field of view and higher sensitivity will be crucial to explore the properties of the halo in greater detail.

At first glance, the overall morphology of IC 4642 exhibits notable central symmetry, characterized by a bright, nearly elliptical shape. However, a closer examination reveals a more intricate structure. The observed features suggest that IC 4642 may correspond to a bipolar nebula viewed face-on, akin to NGC 2392. Concentric layers are evident surrounding the core, and asymmetric emissions in opposing directions may indicate the presence of knots or lobes of ejected material, potentially linked to stellar winds or magnetic fields. Further high-dispersion spectroscopic observations are necessary to confirm this interpretation and investigate the underlying mechanisms driving these features.

In Fig. 2, we present three images in the emission lines of [Ar III] (left panel), [Ar IV] (middle panel), and [Ar V] (right panel). Overall, the shape exhibits a distinctive elliptical morphology, characterized by two elongated lobes extending from the northeast to southwest, which roughly encircle the central zone of the PN. The three emission-line maps show minimal variation in the internal structure and size of the PN. While the two elongated inner structures are connected in the quadruple ionized Argon line, they appear disconnected in the double and triple ionized Argon lines. Moreover, as the ionization levels increase from the [Ar III] and [Ar IV] maps to the [Ar V] map, there is a discernible decrease in the nebula’s size, indicating the presence of ionization stratification. The measurements of the PN’s sizes were found to be 24.7, 22.4, and 19.3 arcsec for the [Ar III], [Ar IV], and [Ar V] emission lines, respectively. The inner structures of the nebula do not resemble the low-ionization features commonly observed in many PNe, as they exhibit similar brightness in both low- and high-ionization lines.

Figure 3 displays three images of the PN, showcasing the emission lines of H\(\alpha\) (left panel), He I (middle panel), and He II (right panel). Notably, all the images manifest a consistent elliptical morphology featuring two inner elongated lobes. It is worth mentioning that the brightness of the eastern lobe appears reduced in the middle section.

The emission-line maps of neon [Ne III], oxygen [O III], and chlorine [Cl IV], presented in Fig. 4, illustrate a similar oval shape of the nebula, with four illuminated corners that reveal the boundaries of the two inner elongated lobes.

Although Figures 2, 3, and 4 display similar morphological features, each figure emphasizes different ionization states and elements that are essential for understanding the ionization structure and chemical composition of the nebula.

The data cubes hold significant potential for creating line ratio maps, which provide valuable information on dust distribution and low-ionization structures (LIS), as well as allowing us to visualize temperature and density distributions across the entire nebula. However, the low signal-to-noise ratio (SNR) in our data, particularly in the fainter regions of the nebula, makes it difficult to construct reliable maps.

Physical and chemical analysis

Global spectra and line fluxes

To extract the integrated spectrum of the PN from the data cubes, the software QFitsView v3.3 [a FITS file viewer utilizing the QT widget library developed at the Max Planck Institute for Extraterrestrial Physics by Thomas Ott] was used. An annular aperture with a radius of 11 arcseconds was chosen to match the apparent size of the nebula in our observations, ensuring that the extracted spectrum was representative of the entire PN. The spectra obtained from the B and R gratings were combined applying the scombine task within the IRAF software.

Line identification and flux measurements, along with their associated uncertainties, were derived using the ALFA code14. The line fluxes were checked once more using the splot task in the IRAF software. To derive the interstellar extinction coefficient and nebular plasma diagnostics, we applied the Nebular Empirical Abundance Tool (NEAT)15.

Applying the Galactic reddening law16 to the observed Balmer decrements (H\(\alpha\)/H\(\beta\), H\(\gamma\)/H\(\beta\), and H\(\delta\)/H\(\beta\)), assuming case B condition at \(T_{eff} = 10,000\) K, yields an average extinction coefficient of \(c({\text{ H }}\beta ) = 0.191^{+0.049}_{-0.051}\). To tackle statistical uncertainties in the extinction coefficient and subsequent steps of plasma diagnostics, such as determining physical conditions and chemical abundances-NEAT integrated the Monte Carlo technique, allowing for a precise assessment of these uncertainties.

Table 2 presents the values of both the observed, \(F(\lambda )\), and dereddened, \(I(\lambda )\), line fluxes relative to H\(\beta =100\) for \(\sim 150\) collisional excitation lines (CELs) and optical recombination lines (ORLs). The dereddened total fluxes of H\(\beta\) and H\(\alpha\), determined to be I(H\(\beta\)) \(= 4.57 \times 10^{-12}\) erg \(\hbox {cm}^{-2}\) \(\hbox {s}^{-1}\) and I(H\(\alpha\)) = \(1.26\times 10^{-12}\) erg \(\hbox {cm}^{-2}\) \(\hbox {s}^{-1}\), respectively, are in excellent agreement with the values reported in the literature17,18.

Spectral analysis reveals that the emission lines of low-ionization species, such as [O I], [N I], [O II], and [N II], exhibit very weak intensities or are absent. Conversely, there is a pronounced intensity in the He II line at \(\lambda\)4686, measuring \(\sim 1.2\) times the intensity of the H\(\beta\) line, which in consistent with the value derived by Tylenda et al.19. This observation suggests that IC 4642 is a PN with very high excitation class. As a result, it indicates that the effective temperature of the central star exceeds 85 kK, aligning with the temperature derived from Zanstra \(T_z(\text {He II})\) and energy-balance methods (see “Observations & data reduction” section). Following the methodology outlined in the literature20, a significantly high EC value of 11.0 was obtained. This finding is consistent with the presence of the highly excited line [Ne IV], which possesses an atomic ionization energy of 97.1 eV.

The analysis of the emission lines shows that the flux ratio of [N II] \(\lambda\)6583 and [S II] \(\lambda\)6731 relative to H\(\alpha\) being below 0.1 implies that the nebula is optically thin21.

To explore the physical conditions and chemical composition of the interior structures of the PN and to compare these with the characteristics of the entire nebula, we extracted the integrated spectra from five positions (P1–P5) within the nebula (Fig. 4, right panel). Positions 1–4 correspond to the four corners of the two elongated inner lobes, while position 5 corresponds to the central region of the nebula, which includes the invisible CS in the IFU emission-line maps. The interstellar extinction coefficients for regions P1 through P5 are shown in Table 3. The small discrepancies between these values and the value for the complete PN can be attributed to variations in internal dust distribution.

Table 2 The fluxes of the emission lines, \(F(\lambda )\), and de-reddened intensities, \(I(\lambda )\), relative to H\(\beta = 100\), are presented for IC 4642. The rest and observed wavelengths are provided in units of angstroms (Å). The line fluxes were derived from the spectrum of the entire PN. The complete table is available as supplementary material.
Full size table

Density and temperature of IC 4642

The prominent emission lines (CELs and ORLs) detected within the integrated nebular spectrum were used to derive the density and temperature of IC 4642 across various ionization states. The nebular density was determined through the analysis of specific line ratios: [O II] (\(\lambda\)3727/\(\lambda\)3729), [S II] (\(\lambda\)6716/\(\lambda\)6731), [Cl III] (\(\lambda\)5517/\(\lambda\)5537), and [Ar IV] (\(\lambda\)4711/\(\lambda\)4740). In parallel, the nebular temperature was calculated by examining the line ratios: [O III] (\(\lambda\)4959 + \(\lambda\)5007)/\(\lambda\)4363, [Ar III] (\(\lambda\)7135 + \(\lambda\)7751)/\(\lambda\)5192, [N II] (\(\lambda\)6548 + \(\lambda\)6584)/\(\lambda\)5754, [O II] (\(\lambda\)7319 +\(\lambda\)7330)/(\(\lambda\)3726+\(\lambda\)3729), [S II] (\(\lambda\)6717 + \(\lambda\)6731)/(\(\lambda\)4068+\(\lambda\)4076), He I (\(\lambda\)7281/\(\lambda\)6678), as well as, Balmer and Paschen jumps.

Table 3 presents the electron densities and temperatures of IC 4642, along with those found in the literature. The results indicate that the density computed from [O II] diagnostic lines exceeds that reported by1, while the density inferred from [S II] lines is lower. Additionally, the temperature determined from the [O III] lines is lower than ours. Additionally, the electron densities and temperatures derived from five interior locations (P1-P5) are listed in Table 3, revealing consistency with the overall nebula within the margin of error. However, this conclusion should be interpreted with caution, as the spectra of these inner nebular positions are contaminated by neighboring gases. Generally, the average temperature deduced from the high ionization zone exceeds that of the medium zone, which in turn is higher than the value derived from the low ionization zone. The mean nebular density calculated from the low and medium ionization zones shows rough consistency.

While the spectral lines of many low-ionization species, such as [O I], [N I], [O II], [N II], and [S II], exhibit very weak intensities, they remain above the noise threshold in our full spectrum (see Table 2). However, the physical conditions derived from these lines should be interpreted with caution due to their low signal-to-noise ratios. In particular, the electron temperatures estimated from the [O I] and [S II] lines are considered unreliable and are marked with a colon (:) in Table 3 to indicate their uncertain nature.

Although the data cubes hold significant potential for creating diagnostic maps of PNe, which would enable visualization of temperature, density, and interstellar extinction distributions across the entire nebula, the poor signal-to-noise ratios inherent in the data prevented us from presenting such maps.

Table 3 The physical parameters of IC 4642, which are derived from its global spectrum. A colon (:) is used to mark values that are of uncertain values.
Full size table

Ionic and elemental abundances

Table 4 provides the calculated ionic and elemental abundances of IC 4642 using the NEAT code. The abundances of nitrogen, oxygen, neon, argon, sulfur, and chlorine were determined through the analysis of CELs, while the abundances of helium and carbon were derived from the ORLs. The ionic abundances were computed by considering the appropriate temperature and density values for their specific ionization zones. In cases where multiple lines were observed for the same ion, the average abundance was calculated. The elemental abundances were obtained by applying the ionization correction factors (ICFs)26 to the calculated ionic abundances. The results reveal a deficiency in heavy elements, characterized by a low helium abundance (He/H < 0.125), a nitrogen abundance (log(N/H) + 12) < 8.0, and a N/O ratio < 0.25, consistent with the chemical signature of Type III PNe27.

To validate the chemical classification of IC 4642 through kinematical analysis of the PN, we evaluated its radial peculiar velocity and its height relative to the Galactic plane. We determined the distance to the PN using its observed parallax of \(0.243 \pm 0.061\) mas, as provided by Gaia DR3. Additionally, we corrected the heliocentric radial velocity of the PN (\(44.0 \pm 3.0\) km \(\hbox {s}^{-1}\))7 to the Local Standard of Rest (LSR) radial velocity by applying the solar motion parameters28 \((u, v, w) = (10, 5.5, 7.17)\)km \(\hbox {s}^{-1}\). The results reveal that IC 4642 is located 669 pc below the Galactic plane and possesses a radial peculiar velocity of \(123 \pm 19\) km \(\hbox {s}^{-1}\), which exceeds the upper limit of \(\sim 60\) km \(\hbox {s}^{-1}\) for objects associated with the Galactic thin-disk. These kinematic parameters support the classification of the PN as a Type III, suggesting that the precursor star of the PN is probably an old, low-mass star with a mass between 1.0 and 1.2 \(\hbox {M}_{\odot }\)27.

A comparison of the total abundances of IC 4642 with those of the five inner positions (P1–P5), Type I PNe, non-Type I PNe, solar abundances, and literature values is presented in Table 5. The results reveal good agreement between the abundances derived from the five inner positions and the overall nebula, within the margin of error. Therefore, the primary distinction between the emission line structures of these two lobes and the rest of the nebula is their increased brightness rather than their distinct physical or chemical properties.

In general, the elemental abundances of IC 4642 are more consistent with those of non-Type I PNe and solar abundances than with those of Type I PNe. The high uncertainty associated with the nitrogen and sulfur abundances can be attributed to the very weak intensity of the corresponding lines, and the total abundances of both elements are estimated based on ionic abundances derived from only one ionization state (\(\hbox {N}^+\) and \(\hbox {S}^+\)). Furthermore, the results from Table 5 also indicate an underabundance of sulfur, a phenomenon commonly observed in many PNe and known as the sulfur anomaly.

Although we have detected numerous ORLs of nitrogen and oxygen (as listed in Table 2), their extremely weak intensities hindered our ability to determine the total abundances from these lines with acceptable accuracy. Consequently, this limitation also precludes the calculation of the abundance discrepancy factor (ADF) for both elements.

Table 4 Ionic and total abundances of IC 4642.
Full size table
Table 5 Comparing elemental abundances of IC 4642 with previous studies, Type I, non-Type I PNe, and the Sun.
Full size table

The variable CS of IC 4642

The axisymmetric shapes of PNe, including bipolar and elliptical morphologies, arise from the axisymmetric mass loss experienced by progenitor stars during their interactions with binary companions30. These companions can be stellar or sub-stellar objects. Based on this hypothesis, PNe are categorized into four classes: (1) No interaction with companions; (2) Interaction with stellar companions, without a common envelope (CE) phase; (3) Interaction with stellar companions, with a CE phase; and (4) Interaction with sub-stellar companions with a CE phase, which accounts for 56% of PNe sample. In this context, IC 4642 is tentatively classified (with low confidence) as class 4, indicating that its morphology likely arises from its progenitor star being part of a close binary system, in CE phase, with a sub-stellar companion. This classification further supports the presence of spherical halos surrounding elliptical planetary nebulae, a feature that is evident in IC 4642 (see Fig. 1).

The CS of IC 4642 has been identified as a variable star based on data from Gaia DR331. To investigate the nature of this variability, we conducted an extensive search for time-series data from both ground-based and space-based observations, including the Zwicky Transient Facility (ZTF), Optical Gravitational Lensing Experiment (OGLE), Transiting Exoplanet Survey Satellite (TESS), and Kepler missions. Unfortunately, our search did not yield any positive results.

We then used the available Gaia data to explore the cause of the CS variability. A periodic search was performed to identify prominent signals through periodograms (Fig. 5). The periodogram reveals a dense array of peaks , with two significant concentrations around periods of approximately 0.18 days and 1.9 days. The analysis identified several periodicities within the error margins; however, none could be confirmed as definitive or statistically significant. The light curves associated with these ambiguous periods exhibited characteristics similar to those of a binary system, indicating variability potentially caused by irradiation effects. Nevertheless, the absence of clear and consistent patterns, along with the presence of multiple aliases and spurious peaks in the periodogram, hindered our ability to establish a robust conclusion regarding the variability of the CS or to identify a reliable period. A follow-up observations-specifically photometric and radial velocity measurements are essential for further clarification.

Figure 5
figure 5

Periodogram of the central star of IC 4642.

Full size image

Conclusions

In this study, we have presented the first integral field spectroscopic observations of the Galactic planetary nebula IC 4642, covering the optical range of 3400–8950 Å. These observations have yielded important information into the physical, chemical, and morphological properties of the nebula.

The integrated spectrum reveals numerous CELs and ORLs, which have enabled us to determine the nebular density and temperature through various diagnostic line ratios. The consistency of these physical parameters with those derived from five inner positions within the PN-four at the corners of the two elongated lobe structures and one at the central region-suggests relatively uniform physical conditions throughout the nebula.

Our analysis indicates that IC 4642 belongs to a high-excitation class, characterized by very weak intensities or the absence of low-ionization lines. The prominent intensity of the He II line at \(\lambda 4686\) confirms the high excitation of the PN. The optically thin nature of the nebula is further supported by the very weak intensities of low-ionization lines, such as [N II] and [S II], relative to the H\(\alpha\) line.

The ionic and elemental abundances of IC 4642, determined from the entire nebula using available CELs and ORLs, show good agreement with that computed from the five interior positions. The elemental abundances align also with those of non-Type I PNe and solar abundances. According to Peimbert’s classification scheme, IC 4642 is categorized as a type III PN, indicating it originated from an old, low-mass progenitor star with a mass ranging from 1.0 to 1.2 \(\hbox {M}_{\odot }\).

The analysis of emission-line maps, which include low, intermediate, and high ionization lines, consistently reveals an elliptical shape with two inner illuminated lobes surrounding the nebular core, suggesting a likely bipolar structure with an equatorial torus.

The analysis of Gaia data identified several potential periodicities in the central star’s variability, but none could be confirmed as definitive due to inconclusive patterns and aliases. Follow-up observations, including photometry and radial velocity measurements, are necessary to verify the variability of the central star.