Tracing PAH Size in Prominent Nearby Mid-Infrared Environments

C. Knight; E. Peeters; D. J. Stock; W. D. Vacca; A. G. G. M. Tielens

doi:10.3847/1538-4357/ac02c6

1. Introduction

The mid-infrared (MIR) spectra of many astronomical sources are dominated by emission features at 3.3, 6.2, 7.7, 8.6, 11.2, and 12.7 μm, attributed to the IR fluorescence of polycyclic aromatic hydrocarbon molecules (PAHs; e.g., Léger & Puget 1984; Allamandola et al. 1985). It has been well established that PAH molecules with sizes from 50 to 100 carbon atoms are the carriers of these bands (e.g., Allamandola et al. 1989; Puget & Léger 1989). In recent works, different authors have adopted various monikers to refer to these features, such as the aromatic emission features (AEFs; e.g., Werner et al. 2004a), the aromatic infrared bands (AIBs; e.g., Croiset et al. 2016), or simply the PAH emission features (e.g., Boersma et al. 2012; Peeters et al. 2012, 2017). Other related species are also considered to be carriers, such as polycyclic aromatic nitrogen heterocycles (PANHs; Hudgins et al. 2005; Bauschlicher et al. 2008) or PAHs with functional groups attached (e.g., Joblin et al. 1996; Sloan et al. 1997; Pilleri et al. 2015; Maltseva et al. 2016; Shannon & Boersma 2019). These PAH emission features have been found in a wide variety of Galactic and extragalactic sources, such as H ii regions, reflection nebulae (RNe), post-AGB stars, planetary nebulae (PNe), the diffuse interstellar medium, and galaxies (e.g., Sellgren et al. 1996; Hony et al. 2001; Verstraete et al. 2001; Peeters et al. 2002).

The PAH emission features show variations in intensities, peak positions, and band profiles (e.g., Bregman et al. 1989; Hony et al. 2001; Peeters et al. 2002; Galliano et al. 2008). Relative intensity variations reveal how these bands relate to each other and to the intrinsic properties of these species. The 6.2, 7.7, and 8.6 μm bands are well correlated over a wide range of environments (Joblin et al. 1996; Galliano et al. 2008). Likewise, the 3.3 and 11.2 μm bands are well correlated (Russell et al. 1977; Hony et al. 2001) but do not strongly correlate with the 6–9 μm PAH bands. Laboratory and quantum chemical studies show that the most significant driver of these intensity variations is the PAH charge state: 6.2, 7.7, and 8.6 μm bands are prominent in ionized PAHs, whereas the 3.3 and 11.2 μm bands are strongest in neutral PAHs (e.g., Allamandola et al. 1989; Bauschlicher et al. 2008). This in turn is strongly dependent on the UV radiation field of nearby stellar sources incident upon the environment in which these PAHs reside (as well as local gas density and temperature; e.g., Tielens 2005; Galliano et al. 2008; Boersma et al. 2013; Stock et al. 2016; Peeters et al. 2017; Stock & Peeters 2017).

The MIR spectra can be strongly influenced not only by ionization but also by changes in the chemical structure of these species. In particular, changes in PAH size can be traced by relative intensity variations within an extended region (e.g., Mori et al. 2012; Croiset et al. 2016). We investigate the PAH size distribution through the measurement of the 11.2/3.3 PAH intensity ratio in the photodissociation regions (PDRs) adjacent to three astronomical sources: two RNe, NGC 2023 and NGC 7023, and the region southeast from the Orion Bar. RNe are ideal test beds for the study of the PAH emission bands, as they have strong MIR emission and have a well-defined structure without contamination from atomic lines (i.e., Sellgren 1983; Rapacioli et al. 2005; Berné et al. 2007; Boersma et al. 2013, 2016), whereas the Orion Bar has long been considered the prototypical PDR (Tielens & Hollenbach 1985), i.e., regions where far-ultraviolet (FUV) photons with energies between 6 and 13.6 eV control the physics and chemistry of the gas.

In this paper, we relate the PAH sizes in each source considered (NGC 2023, NGC 7023, and the region southeast of the Orion Bar) to other properties of these species, as well as the physical characteristics of their PDRs. Section 2 details the properties of each of the sources studied here. In Section 3, the observations used in this study are detailed, while in Section 4, the data reduction performed is discussed. Section 5 presents the results of our analysis in the form of spatial maps of emission feature ratios and line projections of these ratios extending from the illuminating sources. These results are discussed in Section 6, and conclusions are given in Section 7.

2. Astronomical Sources

For our analysis, we selected three well-known sources that show prominent PAH emission features. We briefly summarize the general morphology and properties of these sources relevant to this study below.

NGC 2023.—NGC 2023 is a well-known RN illuminated by the B1.5V type star HD 37903 and located at a distance of 403 ± 4 pc (Kounkel et al. 2018). Its MIR emission has been the focus of many studies over the past 40 yr (e.g., Sellgren 1984; Gatley et al. 1987; Abergel et al. 2002; Peeters et al. 2012, 2017). It shows the same morphology at a wide range of wavelengths: a limb-darkened shell or a "bowl"-shaped PDR surrounding the interior cavity (Figure 1). We can see into the various layers of the PDR, along with some diffuse filamentary structure throughout the RN. PDR models of the cavity have shown an FUV radiation field strength of ∼4000 times that of the average interstellar field (position H5; Steiman-Cameron et al. 1997), as well as gas densities ranging from 10⁴ to 10⁵ cm⁻³ (e.g., Burton et al. 1998; Wyrowski et al. 2000; Sandell et al. 2015). We focus our study of this RN to the south of HD 37903, where the presence of multiple MIR emission peaks is suggestive of significant PAH emission along the PDR boundaries and is associated with the densest gas structures in the region. These peaks are labeled as in Peeters et al. (2017) as the S ridge, the SSE ridge, the SE ridge, and the S' ridge.

**Figure 1.** FLITECAM 3.3 μm (top left), IRAC 8.0 μm (top right), and the 11.2 μm PAH band extracted from IRS SH (bottom) observations of NGC 2023. The illuminating star, HD 37903, is indicated by a cyan circle, and the young stellar objects, C and D, are indicated by black circles, the south IRS SH FOV by a white dashed rectangle, and the ISO-SWS FOV by a black rectangle. The cross-cut used is shown as a white line, at the same position of cross-cut used by Peeters et al. (2017). The south ridge and south–southeast ridge as referred to by Peeters et al. (2017) are labeled the S ridge and SSE ridge, respectively. A 2MASS point source embedded within the S ridge is indicated by a black circle (see Section 6). The units are surface brightness (MJy sr^–1) for FLITECAM 3.3 μm and IRAC 8.0 μm images and integrated fluxes (×10⁻⁶ W m⁻² sr⁻¹) for the 11.2 μm image.
Download figure:
Standard image High-resolution image

NGC 7023.—NGC 7023 is another RN that is well known for its strong MIR emission (e.g., Cesarsky et al. 1996; Boissel et al. 2001; An & Sellgren 2003; Werner et al. 2004a; Rapacioli et al. 2005; Berné & Tielens 2012; Boersma et al. 2013, 2014). It is illuminated by the Herbig Be star HD 200775 and is located at a distance of 361 ± 6 pc (Gaia Collaboration et al. 2016, 2018). There are three main regions of study located around this star: 40'' northwest, 70'' south, and 170'' east (Berné et al. 2007). Each of these regions coincides with a PDR front. In Figure 2, only the NW and S PDRs are visible. The NW PDR is the most prominent of the three with an edge-on structure in which clear stratification of emission is visible (e.g., Pilleri et al. 2012). Estimates for the UV radiation field strength at the NW PDR front are 2600 times that of the average interstellar value with gas densities on the order of 4 × 10³ cm⁻³ (Chokshi et al. 1988). Due to its prominence, the NW PDR of NGC 7023 is the focus of our study of this RN.

**Figure 2.** FLITECAM 3.3 μm (top left), IRAC 8.0 μm (top right), and the 11.2 μm PAH band extracted from IRS SH (bottom) observations of NGC 7023. The illuminating star, HD 200775, is indicated by a cyan circle, the northwest IRS SH FOV by a white dashed rectangle, the ISO-SWS FOV by a black rectangle, and the cross-cut used is shown as a white line. We use the same units as given in Figure 1.
Download figure:
Standard image High-resolution image

Orion Bar.—The Orion Bar or Bright Bar refers to the ionization ridge in the Orion Nebula or the interface between the H ii region and PDR located at a distance of 414 ± 7 pc (Menten et al. 2007). The MIR emission within the Orion Bar has been well studied, showing a stratified plane-parallel structure in which different species peak at a range of distances from the source of UV radiation, θ¹ Ori C (e.g., Aitken et al. 1979; Sellgren 1981; Bregman et al. 1989; Geballe et al. 1989; Sellgren et al. 1990; Tielens et al. 1993; Giard et al. 1994; Cesarsky et al. 2000; Rubin et al. 2011; Boersma et al. 2012; Haraguchi et al. 2012). In Figure 3, the bar structure is quite prominent, as it separates the extremely bright Trapezium Cluster responsible for the production of the strong UV radiation field in the northwest from the relatively colder, lower-density region extending southeast from the Bright Bar. PDR models of the Orion Bar suggest a gas density of 5 × 10⁴ cm⁻³ and FUV radiation field strength of 4 × 10⁴ times that of the average interstellar value (e.g., Tauber et al. 1994). The UV field of the H ii region surrounding θ¹ Ori is expected to be too harsh to allow a significant PAH population to prosper. Indeed, most of the MIR emission to the north of the Orion Bar is associated with warm dust within the ionized gas (Salgado et al. 2016). Thus, our study of the Orion Nebula extends beyond the PDR front within the Bright Bar southward toward the outer Veil.

**Figure 3.** FLITECAM 3.3 μm (left) and IRAC 8.0 μm (right) observations of the Orion Nebula. The position of the illuminating source, θ¹ Ori C, is indicated by a cyan circle, the IRS SH maps for each position specified in Rubin et al. (2011) by white dashed rectangles, and the ISO-SWS FOVs by black rectangles. We note that the FLITECAM 3.3 μm and IRAC 8.0 μm images use a square root scale to show the entire range in emission structure. We use the same units as given in Figure 1.
Download figure:
Standard image High-resolution image

3. Observations

We used multiple MIR observations to measure the dominant PAH emission bands. First, we obtained photometric observations from the First Light Infrared TEst CAMera (FLITECAM; Temi et al. 2012) on board the Stratospheric Observatory for Infrared Astronomy (SOFIA; Young et al. 2012) to probe the 3.3 μm PAH emission. Second, PAH emission in the 7–9 μm range was investigated using observations from the Infrared Array Camera (IRAC; Fazio et al. 2004) on board the Spitzer Space Telescope (Werner et al. 2004b). Finally, the 11.2 μm PAH emission was extracted from observations in the short-high mode (SH) of the Infrared Spectrograph (IRS; Houck et al. 2004) on board the Spitzer Space Telescope. These observations are detailed below and are summarized in Table 1.

Table 1. Log of Observations

Data	S/N ^a	R.A. ^b	Decl. ^b	Nod	Exp. ^l
				(arcsec, deg) ^c	(s)
NGC 2023
FC^d	12	5 41 37.89	−2 15 52.2	1000, 245	460
IRS SH^e		5 41 37.63	−2 16 42.5
IRAC 8.0^f ^, ^g	46	5 41 37.89	−2 15 52.2

NGC 7023
FC^d ^, ^h	76	21 01 36.92	68 09 47.7	600, 0	50
IRS SHⁱ		21 01 32.89	68 09 52.5
IRAC 8.0^f ^, ^h	80	21 01 36.92	68 09 47.7

Orion
FC^d	40	5 35 26.00	−5 26 12.0	1500, 50	360
IRS SH^j
I4		5 35 23.3	−5 25 20.7
I3		5 35 24.4	−5 26 39.1
I2		5 35 26.3	−5 26 12.1
I1		5 35 28.1	−5 27 43.8
M1		5 35 29.9	−5 27 14.4
IRAC 8.0^k	39	5 35 22.33	−5 24 36.0

Notes.

^aThis value corresponds to the peak S/N in each frame.^b α, δ (J2000) refer to the center of the map; α has units of hours, minutes, and seconds, and δ has units of degrees, arcminutes, and arcseconds.^cNod parameters (throw, angle).^dFLITECAM.^eSee, e.g., Peeters et al. (2012, 2017); Shannon et al. (2016).^fSpitzer Heritage Archive.^gFleming et al. (2010).^hCroiset et al. (2016).ⁱSee, e.g., Berné et al. (2007); Boersma et al. (2013); Stock et al. (2016).^jRubin et al. (2011); Boersma et al. (2012).^kMegeath et al. (2012); α, δ in this case reference the center of a subimage extracted from the original mosaic.^lExposure time.

Download table as: ASCII Typeset image

3.1. SOFIA Observations

We obtained SOFIA-FLITECAM observations in the fourth to sixth cycles both for NGC 2023 (Figure 1) and toward the southeast of the Orion Bar (Figure 3) in the "PAH" filter, which has an effective wavelength of 3.302 μm and a bandwidth of 0.115 μm (PID: 04_0058, PI: A. Tielens). The FLITECAM instrument has a 1024 pixel × 1024 pixel InSb detector that covers an 8' × 8' area on the sky with 0 farcs 475 × 0 farcs 475 pixels.

The NGC 2023 field of view (FOV) is centered on the central star, HD 37903, while the Orion FOV was chosen such that it is centered on the Spitzer-IRS SH pointing I2 as defined in Rubin et al. (2011), which probes the region southeast of the Orion Bar, in order to cover the five Spitzer-IRS SH pointings closest to the Trapezium Cluster observed by Rubin et al. (2011). To account for background emission, the instrument was used in nod mode, where the nod parameters were set so that the telescope was moved sufficiently off source between each observation. In the case of NGC 7023 (Figure 2), we have used SOFIA-FLITECAM archival data previously published by Croiset et al. (2016). The FLITECAM 3.3 μm images were found to have a point-source FWHM of ∼2 farcs 1.

3.2. Spitzer Observations

We obtained Spitzer-IRAC 8.0 μm observations (channel 4) of each source (see Figures 1, 2, and 3). The IRAC 8.0 μm band is a broadband filter with a bandwidth of 2.9 μm and a nominal wavelength of 7.87 μm and is found to be an excellent tracer for the 7–9 μm PAH emission (e.g., Smith et al. 2007; Stock et al. 2014, see Appendix). The IRAC 8.0 μm observations were found to have a point-source FWHM of 2''.

For each source, we measured the 11.2 μm PAH emission using IRS SH data, which cover the wavelength range of 10–20 μm at a spectral resolution of 600. IRS SH spectral maps were obtained over a portion of the nebula for both NGC 2023 and NGC 7023 (see Figures 1 and 2, respectively). The spectral map of NGC 2023 covers the S and SSE emission ridges, which correspond to the PDR front as traced by H₂ emission (following the nomenclature of Peeters et al. 2017), as well as YSO C defined by Sellgren (1983). The spectral map of NGC 7023 is composed of the region surrounding the NW PDR. For Orion, we use IRS SH pointings taken along a line from 2.1 to 5 farcm 1 extending radially away from the illuminating source θ¹ Ori C near the Bright Bar (Rubin et al. 2011; Boersma et al. 2012). The spectra obtained for each pointing correspond to a 2 × 10 aperture grid pattern or "chex" of SH spectra (for details, see Rubin et al. 2011). The resulting FOV of each of these pointings (25 farcs 4 × 16 farcs 3) and the position of each pointing are shown in Figure 3. These pointings are given the following designations by Rubin et al. (2011) with increasing distance from the Bar: "inner" (I4, I3, I2, I1) and "middle" (M1).

4. Data Processing

The same general reduction procedure was used for all sources to generate emission ratio maps.

4.1. Flitecam

To extract the 3.3 μm emission feature, the raw FLITECAM frames were co-added using a noise-weighted mean into a single image and flux-calibrated using standard stars observed during the same night as the observations. These images were slightly misaligned with respect to the IRAC 8.0 μm images. We correct for this misalignment by shifting our FLITECAM images to align with the IRAC 8.0 μm images; this is done by aligning the prominent stellar sources or the extended emission peaks in the FLITECAM frames with the corresponding sources in the IRAC 8.0 μm frames. The FLITECAM images are converted from units of surface brightness (Jy pixel⁻¹) to flux (W m⁻² sr⁻¹) by multiplying by the bandwidth of the FLITECAM filter, assuming a nominal flat spectrum. This is done in order to have comparable units with the 11.2 μm emission extracted from the IRS SH spectra.

4.1.1. Reflection Nebulae

In the case of NGC 7023, we obtained a single set of observations for which it was necessary to rotate the FLITECAM image to get the NW and S PDRs to overlap with respect to the IRAC image (Croiset et al. 2016). In the case of NGC 2023, we obtained three sets of observations. Each of these three FLITECAM frames had a significantly high median on-frame background level that is subtracted from each frame (755, 357, and 224 MJy sr⁻¹), after which they are combined in a single image. We checked the absolute calibration for both sources in two ways. The first is by comparison with ISO-SWS observations.⁷ In particular, we checked the absolute flux calibration of the ISO-SWS spectra with IRAC 3.6 μm observations by comparing the observed IRAC 3.6 μm emission with the expected value in the IRAC 3.6 μm bandpass based on the ISO-SWS spectrum. We found a consistency of 90% and 98% for NGC 7023 and NGC 2023, respectively, indicating that the ISO-SWS spectra are well calibrated in an absolute sense. Subsequently, we compare the FLITECAM emission observed in the ISO-SWS aperture with the expected FLITECAM emission based on the ISO-SWS spectrum. We found an agreement of 100% and 104% for NGC 7023 and NGC 2023, respectively, indicating that the FLITECAM images are well calibrated. The second is by comparison with photometry of their central star. Specifically, for NGC 2023, we performed photometry on the central star, HD 37903, in each frame after background subtraction, which resulted in a flux density of 0.38 ± 0.02 Jy on average. Similarly for NGC 7023, we performed photometry on the central star, HD 200775, and found a flux density of 11.56 ± 0.03 Jy. This is consistent with the spectra of HD 37903 and HD 200775 given in Sellgren (1984), again demonstrating that our FLITECAM data are properly calibrated.

We estimate the PAH contribution to the total flux observed in this FLITECAM filter from AKARI spectral observations of NGC 7023 coinciding with the NW PDR (Pilleri et al. 2015). We do this by multiplying both the continuum-subtracted 3.3 μm band and the total spectrum with the FLITECAM 3.3 μm filter response curve and subsequently integrating over the filter range for each AKARI pointing. Dividing these quantities, we found that the PAH contribution in the 3.3 μm filter is 59%–74% of the total emission (which comprises both PAH and dust continuum emission). As both continuum and PAH features are due to UV pumped fluorescence from large molecules (Allamandola et al. 1989) and both species require similar sizes to be excited, it is reasonable to assume that the feature-to-continuum ratio will remain constant within a given source. Moreover, ISO-SWS observations of NGC 2023 and NGC 7023 show that both sources have very similar feature-to-continuum emission ratios for the 3.3 μm band (e.g., Moutou et al. 1999, their Figure 3); thus, the total PAH contribution in the FLITECAM filter will be approximately the same.

4.1.2. Orion

We investigate the absolute calibration of the Orion FLITECAM 3.3 μm data by comparing them with IRAC 3.6 μm data. First, we estimate the expected ratio of the FLITECAM to IRAC observations for the Orion Bar. To this end, we use the ISO-SWS spectra across the Orion Bar.⁸ We multiply these ISO-SWS spectra with both filter response curves and find that the average FLITECAM-to-IRAC ratio equals 2.7 ± 0.3. Next, we compare the observed surface brightness of the FLITECAM 3.3 μm and the IRAC 3.6 μm images (after applying the IRAC 3.6 μm extended source correction; Figure 4). The weighted line of best fit to the data has a slope of 3.82 ± 0.02 and a y-intercept of 342.1 ± 0.8 MJy sr⁻¹. We thus find a scaling factor of 1.4 ± 0.2, which is equal to the observed FLITECAM-to-IRAC ratio divided by the expected FLITECAM-to-IRAC ratio. Hence, to calibrate the FC (FLITECAM 3.3 μm) observations, we use the following scaling relation:

$\begin{eqnarray}&&{FC}(\mathrm{scaled)}=({FC}-342.1)/1.4.\end{eqnarray} \tag{ 1 }$

**Figure 4.** The observed surface brightness in the FLITECAM data and the IRAC 3.6 μm data (after extended source correction). All stellar sources within the FOV have been masked. Surface brightness values of the FLITECAM observations within the ISO-SWS apertures are indicated by magenta triangles, and expected mean surface brightness values (from multiplying the SWS spectra with each filter response curve) are shown as black diamonds. A weighted linear fit is given by the green line.
Download figure:
Standard image High-resolution image

Applying this relationship to the observed Orion FLITECAM 3.3 μm data yields values that agree with the expected FLITECAM 3.3 μm values based on the SWS spectra within 12%.

To further validate this approach, we also apply it to the RNe. The predicted FLITECAM-to-IRAC ratio is determined by multiplying the AKARI spectra of NGC 7023 (Pilleri et al. 2015) with the respective filters. We find that the mean flux density of the spectrum multiplied with the 3.3 μm filter is about 2 times greater than multiplying with the IRAC 3.6 μm. A similar ratio is expected for NGC 2023 based on the similarity of the IR spectra of these sources (e.g., Moutou et al. 1999; Verstraete et al. 2001). Using the same-sized apertures as the Orion pointings in both RNe, we are able to reproduce the predicted FLITECAM-to-IRAC flux ratios of 2. This shows that the FLITECAM observations of the RNe are in agreement with predicted values.

We estimate the PAH contribution to the total flux observed in the FLITECAM filter from the ISO-SWS Orion Bar D2 position by multiplying the filter response curve with the Bar D2 spectrum (e.g., Peeters et al. 2002; van Diedenhoven et al. 2004). We find that the total PAH contribution to the 3.3 μm filter is about 71% of the total emission (which comprises both PAH and dust continuum emission).

4.2. Spitzer

We obtained the Spitzer-IRS data in reduced form (for details on the reduction process, see Peeters et al. 2012 and Shannon et al. 2015 for the RNe and Boersma et al. 2012 for the Orion data). To measure the 11.2 μm PAH emission, we defined and subtracted a local spline continuum with anchor points at 10.85 and 11.65 μm (as in Peeters et al. 2017) from the spectrum at each pixel in our spectral maps or, in the case of the Orion IRS SH data, from the spectrum at each pointing. We performed a simultaneous Gaussian fitting procedure to the 11.0 and 11.2 μm emission features in each continuum-subtracted spectrum in order to separate these features (e.g., Stock et al. 2014, 2016; Peeters et al. 2017). The flux of the 11.2 μm emission band is then determined by integrating these continuum-subtracted spectra over the wavelength range given above and subtracting the 11.0 μm band flux.

We obtained the Spitzer-IRAC data in reduced form for NGC 2023 and the Orion Nebula (Fleming et al. 2010; Megeath et al. 2012) and retrieved them from the Spitzer Heritage Archive in the case of NGC 7023. To approximate the 7.7 μm PAH emission, we converted the IRAC 8.0 μm images (MJy sr⁻¹) to flux densities (W m⁻² sr⁻¹) by multiplying by the wavelength range covered by the 7.7 μm emission band of ∼1.0 μm (e.g., Peeters et al. 2002), again assuming a nominal flat spectrum. The relative PAH contribution in this filter is about 89% of the total emission in the diffuse interstellar medium (Stock et al. 2014) and between 72% and 80% across the S ridge in NGC 2023 over the distance range for which we have 11.2 μm IRS SH measurements (Appendix).

In order to compare the spatial distribution of the different PAH bands for the RNe, the FLITECAM and IRAC 8.0 μm images were regridded to the IRS SH pixel scale (2 farcs 3 × 2 farcs 3). Both images were additionally averaged over 2 × 2 pixels, set by the IRS spatial resolution. Bright stellar sources in the IRS FOVs and pixels for which there was no 3σ detection of the 11.2/3.3 PAH ratio were masked out. In the case of Orion, we determined the median flux density for both FLITECAM and IRAC 8.0 μm images within the IRS aperture at each pointing. In this way, the PAH emission at 3.3, 7.7, and 11.2 μm is compared at five different pointings within the Orion region.

5. Results

We determined the 11.2/3.3 and IRAC 8.0/11.2 (as probed by IRAC 8 μm/11.2 μm) PAH intensity ratios spatially where images of each band were present. For NGC 2023 and NGC 7023, these ratios were measured where each of our observations overlap; this resulted in an FOV of the same size as the IRS SL maps. For Orion, we determined these ratios in the IRS SH apertures covered by the FLITECAM frame. We consider the ratio maps and cuts obtained for each source individually.

5.1. NGC 2023

Figure 5 shows the 11.2/3.3 and IRAC 8.0/11.2 PAH intensity ratios in the southern PDR of NGC 2023. Contours of the 12.3 μm H₂ line are overplotted in white to emphasize the S and SSE emission ridges. These bright filaments are thought to be edge-on corrugations of the cloud material that surrounds the bowl-shaped cavity (Field et al. 1994). The central star is beyond the upper right edge of this map.

We find minima in the 11.2/3.3 PAH intensity ratios at the peak of the H₂ emission and the 3.3 μm emission near the S ridge. In the SSE ridge, the minimum is slightly displaced (in the direction toward the star) compared to the H₂ peak intensity, again coinciding with the nearby peak of the 3.3 μm emission. Another minimum in the 11.2/3.3 ratio behind the H₂ peak centered at pixel (5, 11) is also apparent. In general, we find an increase in this ratio as we move toward the star and further into the molecular cloud behind both ridges. Moving inward toward the star from the S ridge H₂ peak, there is a local minimum shown in cyan surrounded by green that extends westward from the SSE ridge toward the far right of the image and is centered at pixel coordinates (15, 18). This ratio then proceeds to rise moving further inward toward the top right of the frame.

Overall, the IRAC 8.0/11.2 PAH intensity ratio exhibits a stratified morphology with its highest values closest to the central star. In addition, it exhibits a narrow local peak just above the H₂ peak within the S ridge, as well as in the lower right portion of the S ridge. The SSE ridge, in contrast, has a minimum in this ratio. Below the S ridge into the molecular cloud, there is a rise in this ratio.

Figure 7 shows a cross-cut taken across the S ridge emission peak (see Figure 1 for the orientation of the cut) of the normalized flux densities of the 3.3, IRAC 8.0, and 11.2 μm emission, as well as the 11.2/3.3 and IRAC 8.0/11.2 ratios. Note that the 3.3 μm and the 11.2/3.3 profile have a shorter range beyond the PDR front, due to the lack of a 3σ detection with FLITECAM. All three emission features peak at the PDR front (as traced by the peak intensity of the H₂ emission, at 78 farcs 6 from the illuminating source). However, the 3.3 μm emission drops fastest in intensity away from this PDR front, followed by the 11.2 μm emission. Interestingly, the IRAC 8.0 μm emission shows a plateau from ∼50'' to 70'', before dropping off as well closer to the central star. This overall radial profile is reflected in the 11.2/3.3 and IRAC 8.0/11.2 PAH intensity ratios, which show, in general, a smooth decrease with increasing distance from the star until the PDR front, after which they start to increase again. We note that the 11.2/3.3 ratio decreases quickly from ∼4 ± 1 to ∼2.5 ± 0.5, going from 45'' to 55'' away from the star. It then tends to be relatively constant between 55'' and 70'' before dropping to a minimum of ∼1.8 ± 0.1 near the PDR front. In contrast, the IRAC 8.0/11.2 PAH intensity ratio displays a slower decrease moving away from the star, until it becomes relatively flat between 65'' and 75''. At the PDR front, the IRAC 8.0/11.2 PAH ratio shows a very small peak, which thus corresponds to the local maximum at the PDR front as discussed above (see also Figure 5).

5.2. NGC 7023

Figure 6 shows 11.2/3.3 and IRAC 8.0/11.2 ratios in the NW PDR of NGC 7023. Note that these images have been rotated to align with the astrometry of NGC 7023 H₂ data from Lemaire et al. (1996). Contours of the 2.124 μm H₂ line are overplotted in white to probe the PDR front. The central star is positioned beyond the bottom left edge of this image. The 11.2/3.3 intensity ratio is at a minimum all along the PDR front. Moving toward the star, this ratio rises while also showing some smaller local fluctuations on top of the overall increase with respect to the PDR front. In addition, it does not trace very well the ring structure observed in the IRAC 8.0 μm and FLITECAM 3.3 μm images. Behind the PDR front, the ratio rises, although an S/N decrease limits the extent to which this can be traced reliably.

Overall, the IRAC 8.0/11.2 ratio decreases in an almost parallel layered structure with distance from the central star and does not show the ring structure as seen in the IRAC 8.0 emission. While the lowest values are found in regions sandwiching the PDR front, local maxima are apparent along the PDR front itself relative to its immediate surrounding area (around pixels (5, 20) and (13, 18)). Behind the PDR front, away from the central star, the ratio seems to rise again.

Figure 7 shows a cross-cut across the NW PDR (see Figure 2 for the orientation of the cut) of the normalized fluxes of the 3.3, IRAC 8.0, and 11.2 μm emission, as well as the 11.2/3.3 and IRAC 8.0/11.2 intensity ratios. Each of the emission features peaks close to or at the PDR front (as traced by the peak intensity of the H₂ emission, at 46 farcs 3 from the illuminating source). In NGC 7023, as was also found for NGC 2023, the 3.3 μm emission decreases more sharply away from the PDR front compared to the 8.0 and 11.2 μm emission. Moving toward the star, both the 3.3 and IRAC 8.0 μm intensities show an initial decrease followed by a plateau between ∼30'' and 40'' before decreasing again close to the star. This plateau in emission is likely due to the ring of emission seen south of the PDR front detected at both wavelengths (Figure 2). The 11.2 emission shows a linear decrease in intensity, as the portion of the ring structure facing the illuminating star is much fainter in this band. The 11.2/3.3 radial profile shows a decrease away from the star up to a distance of 25''. The ratio then increases up to a maximum at ∼38'' owing to the filaments composing the ring structure perpendicular to the PDR front becoming increasingly luminous in the 11.2 μm. Beyond this peak, there is another decrease to a minimum located just beyond the PDR front at 50'', after which the ratio increases again. The IRAC 8.0/11.2 radial profile shows a different overall trend. It peaks closest to the illuminating star and shows a significant decrease away from the star down to a minimum that occurs at the same distance as the maximum in the 11.2/3.3 radial profile, corresponding to the emergence of the 11.2 μm emission within the ring structure. This again can be attributed to the steadily increasing prominence of the ring structure in the 11.2 μm emission relative to the other bands that show a more uniform emission distribution here. With increasing distance from the central star, the IRAC 8.0/11.2 ratio increases again until ∼50'', after which it drops.

We note that while the relative trends in 11.2 μm/3.3 μm and IRAC 8.0 μm/11.2 μm ratios are quite similar between this study and Croiset et al. (2016), the absolute values of the ratios are not in agreement. This arises as a result of the following differences. First, Croiset et al. (2016) used the FORCAST 11.1 μm broadband filter with an FWHM of 0.95 μm. These authors estimated that the continuum emission contributes 20% to the observed emission in the FORCAST filter and, as a consequence of this low contribution, represents the 11.2 μm PAH flux by the observed FORCAST 11.1 μm flux. In contrast, we measure the integrated 11.2 μm band strength from the IRS SH cube (Section 4.2). The PAH emission contributes 80% of the observed emission at 11.2 μm at the PDR front. Due to the FORCAST filter width, this results in a lower (integrated) PAH contribution. Indeed, we find that the integrated flux accounts for 58% of the PDR front emission within the FORCAST bandwidth, decreasing to 35% at 20'' from the illuminating star. In addition, we note that the portion of the ring-like structure facing the illuminating source is not clearly visible in our 11.2 μm PAH map (extracted from IRS SH spectra), whereas it is clearly evident in the FORCAST 11.2 μm image presented in Figure 1 of Croiset et al. (2016). We attribute this to our subtraction of the dust continuum emission from the IRS SH spectra to isolate the 11.2 μm PAH emission. Summarizing, different methods of representing the 11.2 μm PAH emission strength and the subtraction (or not) of the underlying dust continuum give rise to differences in the 11.2 μm PAH strength between this work and Croiset et al. (2016). Second, the absolute surface brightness of the IRAC 8.0 μm observations used in each study differs. Specifically, the surface brightness at the NW PDR of the archival IRAC 8.0 μm observation used here is ∼3–4 times brighter than in the IRAC 8.0 μm observation used in Croiset et al. (2016). The origin of this discrepancy is unknown, and so we elected to use the archival IRAC 8.0 μm observation. As a consequence of these differences with the 11.2 μm and IRAC 8.0 μm measurements, our IRAC 8.0 μm/11.2 μm ratios are a factor of 7 higher than those reported in Croiset et al. (2016). Despite these discrepancies, we emphasize that the relative trends found in both studies are consistent.

5.3. Orion

Figure 8 shows the normalized flux densities of the 3.3 μm, IRAC 8.0 μm, and 11.2 μm PAH emission, as well as the 11.2 μm/3.3 μm and IRAC 8.0 μm/11.2 μm ratios as a function of distance from the illuminating source for each IRS pointing. Each of the three emission features is brightest near the Bar and decreases steeply out to the M1 pointing. Note that all three of these features vary in lockstep across the Orion Bar. Subsequently, moving outward toward the Veil, the IRAC 8.0 μm and 11.2 μm exhibit a slower rate of decrease with distance, reaching a minimum at the M1 pointing. Differences in the relative decline of the band intensity are better probed with their intensity ratios. Across the I1–I4 positions, the 11.2 μm/3.3 μm ratio remains constant within the associated uncertainties at 2.3 ± 0.2 and decreases at the M1 position to a ratio of 1.4 ± 0.1 owing to the larger relative decrease in the 11.2 μm band. In contrast, the IRAC 8.0 μm/11.2 μm intensity is high at the I4 position and decreases down to a minimum at the I1 position. This is followed by an increase back to a maximum at the M1 position, again due to the 11.2 μm emission showing the sharpest relative decrease at this pointing.

6. Discussion

Relative intensity variations of the 3.3 to 11.2 μm bands trace the average size of the observed PAH population (e.g., Schutte et al. 1993; Ricca et al. 2012). Indeed, upon absorption of a UV photon of a given energy, a smaller PAH molecule, with less vibrational modes in comparison to a larger molecule, will obtain a higher internal temperature as more energy is being distributed per mode. Smaller PAHs will thus emit preferentially in the 3.3 μm band compared to the 11.2 μm band. Both the 3.3 and 11.2 μm emission bands originate in neutral PAH species, which eliminates any dependence on the PAH charge state. Both bands also originate in CH modes and are therefore not sensitive to variations in the H/C ratio of the emitting species. In addition, these bands span the largest range in wavelength of the major PAH bands (excluding the 12.7 μm, which is of mixed charge state; e.g., Peeters et al. 2012, 2017; Shannon et al. 2015, 2016), maximizing the spectral sensitivity to internal temperature.

Ricca et al. (2012) employed the NASA Ames PAH IR spectroscopic database (Bauschlicher et al. 2010) to calculate the intrinsic emission spectrum of PAHs with different sizes for an average photon energy of 6 and 9 eV. These authors find (i) a clear inverse relationship between the 3.3 μm/11.2 μm PAH band ratio and PAH size for PAHs of the coronene and ovalene family and (ii) that a higher absorbed average photon energy will increase the 3.3 μm/11.2 μm ratio for a given PAH species (their Figure 16). Hence, comparing the observed 3.3 μm/11.2 μm PAH intensity ratios with these calculations can then provide an estimate for the expected size distribution of the PAH population.

In contrast, the IRAC 8.0 μm/11.2 μm ratio traces the ionization balance of the PAH population. Indeed, the IRAC 8.0 μm channel covers the 7.7 and 8.6 μm PAH emission bands that are due to ionized PAH molecules, while the 11.2 μm emission band arises from neutral PAH species. Hence, this ratio yields another probe of the interaction between FUV radiation and the PAH population.

Croiset et al. (2016) investigated the size and charge distribution of the PAH population in NGC 7023. These authors followed the approach by Ricca et al. (2012) to determine the average size of the PAH population based on the 11.2 μm/3.3 μm PAH intensity ratio using an average absorbed photon energy of 6.5 eV (their Figure 5). They report that the average PAH size reaches a minimum of 50 carbon atoms at the PDR front and thus increases away from the PDR front toward both the molecular cloud and the illuminating star, reaching sizes up to 70 carbon atoms. In addition, these authors report that the charge distribution becomes increasingly dominated by ionized PAHs with decreasing distance from the central star, consistent with previous results (e.g., Joblin et al. 1996; Verstraete et al. 1996; Berné et al. 2007).

Comparing the (relative) PAH intensity radial profiles for NGC 7023 and NGC 2023 (Figure 7), we report very similar trends in both RNe that are consistent with the results of Croiset et al. (2016): (i) the intensity of each emission component increases with distance from the illuminating source and peaks at the PDR front before decreasing again into the molecular cloud, (ii) the size of the PAH population reaches its minimum at the PDR front, and (iii) the charge state of the PAH population decreases from the illuminating source toward the PDR front. As we probe a different configuration with our pointing toward the Orion Bar, its trends, as described in Section 5, are clearly distinct compared to those of the RNe. We now consider the specific properties of the size distribution and charge balance of the PAH population for each source individually.

6.1. Average PAH Size Distribution

We derive the average PAH sizes for our sample based on the 11.2 μm/3.3 μm ratios as observed with FLITECAM and IRS. We remind the reader that some uncertainty exists related to their absolute values (cf. FLITECAM–SWS comparison, Section 5) and we did not apply a correction for the PAH contribution in the FLITECAM filter. Both will affect the derived average PAH sizes. Consequently, trends in the average PAH size are trustworthy, while its absolute value is indicative.

NGC 7023.—To determine the average PAH size in NGC 7023, we apply the emission model shown in Figure 5 of Croiset et al. (2016). The minimum in the 11.2 μm/3.3 μm ratio of ∼1.7 ± 0.1 at the NW PDR front (Figures 6 and 7) corresponds to an average PAH size of ∼50 ± 2 carbon atoms. Closer to the illuminating star, the 11.2 μm/3.3 μm ratio rises up to 2.9 ± 0.2, corresponding to PAHs with ∼65 ± 5 carbon atoms. This is followed by a decrease to a ratio of 1.8 ± 0.1 or PAHs with ∼50 ± 2 carbon atoms before rising again into the cavity surrounding the star. As detailed in Section 5, the trends found in the 11.2 μm/3.3 μm ratio and consequently the average PAH size for the NW region of NGC 7023 are overall similar to those reported by Croiset et al. (2016), while absolute values differ owing to the different methods used to measure the 11.2 μm emission.

NGC 2023.—We use the Kurucz stellar model used by Andrews et al. (2015) and the emission model of (Croiset et al. 2016, which takes the absorption cross section into account) to obtain an average photon energy of 7.3 eV. Using the dependence of the 11.2 μm/3.3 μm PAH ratio on size as determined by Ricca et al. (2012, their Figure 16), we obtain an average PAH size of 75 ± 5 carbon atoms for a minimum value of the 11.2 μm/3.3 μm ratio of ∼1.8 ± 0.1 at the S and SSE ridges (Figures 5 and 7). Moving toward the star, the 11.2 μm/3.3 μm ratio rises up to ∼4 ± 1, corresponding to a size of ∼110 ± 20 carbon atoms.

Orion.—By using the emission model of Croiset et al. (2016) with a Kurucz stellar model of T_eff = 39,000 K and log(g) = 4 for θ¹ Ori C (Kurucz 1993), we obtain an average photon energy of 8.1 eV.⁹ The 11.2 μm/3.3 μm ratios in all regions probed SE of the Orion Bar vary between approximately 2.3 ± 0.2 and 1.4 ± 0.1.¹⁰ This corresponds to an average PAH size ranging from 70 ± 5 to 85 ± 5 carbon atoms based on the 11.2 μm/3.3 μm PAH ratio and size relationship as determined by Ricca et al. (2012).

Considering these three sources, we thus find that (i) the average PAH sizes in NGC 7023 are the smallest, with its maximum average PAH size being similar to the minimum average PAH size both in NGC 2023 and SE of the Orion Bar; and (ii) the range of average PAH size across the FOV is largest toward NGC 2023 (Table 2). Comparing these results with the physical conditions of the sources (Table 2), we note that for the RNe the average PAH size increases when the intensity of the radiation field, as measured by G₀, increases.

Table 2. Physical Conditions and Average PAH Size for Our Sample

Source	Coordinates^a	G₀ ^b	n_e ^c	T_gas	γ ^d	Average PAH Size		Refs
	(J2000)		(cm⁻³)	(K)	(×10^b)	Minimum	Maximum

NGC 7023	(21:01:33.3; +68:10:17.60)	2600	0.64	250	20	50 ± 2	65 ± 5	1
NGC 2023	(05:41:40.39; −02:16:02.48)^e	1.5 × 10^d	16	750	8			2
	(05:41:37.99; -02:16:35.26)^f	4000				75 ± 5	110 ± 20	2
Orion Bar	(05:35:19.9; -05:25:09.82)^g	4 × 10^d	8	500	88	70 ± 5	85 ± 5	3, 4

Notes.

^aPosition at which the physical conditions were derived.^b G₀ is the integrated 6–13.6 eV radiation flux in units of the Habing field = 1.6 × 10⁻³ erg cm⁻² s⁻¹. ^cElectron density in the PDR (cm⁻³) estimated by n_e ≈ (C/H)n_H = 1.6 × 10⁻⁴ n_H using the carbon abundance of Sofia et al. (2004).^dIonization parameter $\gamma ={({G}_{0}/{n}_{e})({T}_{\mathrm{gas}}/1000\,{\rm{K}})}^{0.5}.$ $\gamma ={({G}_{0}/{n}_{e})({T}_{\mathrm{gas}}/1000\,{\rm{K}})}^{0.5}.$ ^eCorresponds to the peak IR emission located outside our IRS FOV.^fCorresponds to the S ridge position.^gOrion Bar position 4, which corresponds to the ISO-SWS D5 position. References: (1) Chokshi et al. 1988; (2) Steiman-Cameron et al. 1997; (3) Tauber et al. 1994; (4) Galliano et al. 2008.

Download table as: ASCII Typeset image

6.2. PAH Charge Distribution

NGC 7023.—As discussed in Section 5, the values of the IRAC 8.0 μm/11.2 μm ratio are significantly higher than those derived by Croiset et al. (2016). Yet the relative ionization of the PAH population in NGC 7023 shows the same overall decreasing trend with distance from the illuminating star, as well as a local peak along the PDR front as reported by Croiset et al. (2016).

Another tracer for the PAH ionization balance is the 11.0 μm/11.2 μm PAH ratio (e.g., Hudgins & Allamandola 1999; Rosenberg et al. 2011; Shannon et al. 2016; Peeters et al. 2017), as the 11.0 and 11.2 μm PAH emission is due to solo-CH out-of-plane bending modes in, respectively, cationic and neutral PAHs (e.g., Hudgins & Allamandola 1999; Hony et al. 2001; Bauschlicher et al. 2008). A cross-cut of the 11.0 μm/11.2 μm PAH ratio in NGC 7023 shows a clear decrease with distance from the illuminating source (see Figure 9). This is fairly consistent with the cross-cut of the IRAC 8.0 μm/11.2 μm ratio in NGC 7023 shown in Figure 7.

NGC 2023.—Regarding the degree of ionization, we noticed the existence of a local maximum in the S ridge, whereas no similar peak is seen at the SSE ridge (Figure 5). The Two Micron All Sky Survey (2MASS; Skrutskie et al. 2006) observations of NGC 2023 detect a stellar source embedded within the south ridge located ∼2 farcs 5 to the west of our line cut through this region (see Figure 1), which possibly gives rise to this increased PAH ionization balance.

There are clear difference between the IRAC 8.0 μm/11.2 μm ratio and the 11.0 μm/11.2 μm PAH ratio in NGC 2023 (see Figure 9). The 11.0 μm/11.2 μm PAH ratio does not show the local maximum at the PDR front seen in the IRAC 8.0 μm/11.2 μm, but a local minimum just behind this point. Additionally, the 11.0 μm/11.2 μm PAH ratio profile may reflect the bowl-shaped geometry of the cavity, where the relative ionization decrease is largest at the edge-on PDR front. There is a slight rise in the 11.0 μm/11.2 μm PAH ratio moving into the molecular cloud, but it is nowhere near as pronounced as the rise found in the IRAC 8.0 μm/11.2 μm ratio of NGC 2023 (see Figure 7). The discrepancy between the IRAC 8.0 μm/11.2 μm ratio and the 11.0 μm/11.2 μm PAH ratio seen beyond the PDR front suggests that the increasing continuum contribution of the IRAC 8.0 μm makes the IRAC 8.0 μm/11.2 μm ratio a less reliable tracer of PAH ionization into the molecular cloud (e.g., Appendix).

Orion.—The radial profile of the PAH ionization as traced by the IRAC 8.0 μm/11.2 μm ratio does not resemble that of the 11.0 μm/11.2 μm PAH ratio reported by Boersma et al. (2012; see Figure 10). This discrepancy may arise from, among others, a possible changing PAH contribution to the IRAC 8.0 μm band and/or from PAH dehydrogenation. As discussed in Section 3.2, the former cannot be the main driver of the observed variation in the IRAC 8.0 μm/11.2 μm ratio. In contrast, PAH dehydrogenation will influence the 11.0, 11.2, and 8.6 μm band differently, as the 11.0 and 11.2 μm bands are due to solo CH groups, while the 8.6 μm band probes all CH groups in large compact PAHs. We note that the IRAC 8.0 μm band comprises both the 7.7 and 8.6 μm bands, with the 7.7 μm band dominating. Hence, PAH dehydrogenation may cause the IRAC 8.0 μm/11.2 μm and 11.0 μm/11.2 μm tracers for PAH ionization to diverge. Boersma et al. (2012) invoke PAH dehydrogenation for these observations to explain the discrepancy between the 11.0 μm/11.2 μm PAH ratio and the observed change in the hardness of the radiation field as traced by the [S iv] 10.5 μm/[S iii] 18.7 μm line ratio reported by Rubin et al. (2011).

**Figure 10.** The Orion Nebula radial profiles of the PAH ionization balance as traced by the IRAC 8.0 μm/11.2 μm ratio (red, this work, top) and the 11.0 μm/11.2 μm PAH ratio (green, Boersma et al. 2012, top) and of the hardness of the radiation field as traced by the S⁺⁺⁺/S⁺⁺ ratio (blue, Rubin et al. 2011, bottom) as a function of distance to the illuminating source. The associated uncertainties are indicated as error bars of the same color as the respective data.
Download figure:
Standard image High-resolution image

Additionally, Peeters et al. (2017) have shown that the 6.2 and 7.7 μm bands are spatially distinct from the 8.6 and 11.0 μm bands in NGC 2023. They found that the 11.0 and 8.6 μm bands peak much closer to the illuminating source in NGC 2023 than the 6.2 and 7.7 μm bands. Although all of these bands are attributed to ionic PAH species, the spatial diversity found suggests that the 6.2/7.7 μm and 8.6/11.0 μm bands correspond to distinct groups of PAH ions. Since the 7.7 μm band will have a significantly larger contribution within IRAC 8.0 μm than 8.6 μm, we conclude that IRAC 8.0 μm/11.2 μm and 11.0 μm/11.2 μm cannot both be used as identical tracers of relative PAH ionization.

6.3. Photochemical Evolution of PAHs

The observed variation of the average PAH size gives further evidence of significant ongoing photoprocessing of these molecules. Indeed, for both RNe NGC 7023 and NGC 2023, we find that the average PAH size distribution in each source and variations in average size between the two RNe are strongly dependent on the radiation field strength, G₀. Specifically, we find that the minimum average PAH size is found at the PDR front and that the average PAH size increases upon closer approach to the illuminating stars as G₀ increases. Likewise, smaller average PAH sizes are found toward NGC 7023 (compared to NGC 2023), which has also the lowest radiation field intensity (Table 2). Enhanced intensity of the radiation field thus leads to increased PAH photoprocessing, resulting in the destruction of smaller PAH species as predicted by Allain et al. (1996). This, in turn, gives rise to a larger average PAH size. Together with the enhanced emission of fullerenes near the illuminating source (Sellgren et al. 2010; Berné & Tielens 2012; Peeters et al. 2012), these results give support to the hypothesis of the formation of fullerenes from PAH molecules. Specifically, extreme levels of UV radiation promote the complete dehydrogenation of PAHs (>60 carbon atoms) that subsequently fold into loose cage structures upon losing C₂ atoms until they reach stability as C₆₀ molecules (Berné & Tielens 2012; Zhen et al. 2014; Berné et al. 2015). The interpretation of the average PAH size distribution southeast of the Orion Bar is more complicated than that of the RNe. We find no real variation in the average PAH size (within the associated uncertainties) with distance from the illuminating source, except for the M1 pointing. The geometrical nature of the region delineated by the Orion Bar and the southeastern Veil is much more complex than the edge-on morphology of the PDRs found in the RNe. The Bar is an edge-on, compressed shell at the edge of a bowl created by the Trapezium Cluster (Salgado et al. 2016). In our positions out to the Veil, we see a face-on PDR at the surface of the molecular cloud (Pabst et al. 2019). In the case of an edge-on PDR, the gas is transported into the cavity by the PDR evaporation flow, and hence the material closer to the star is the "same" material as farther away except that it is more thoroughly processed by UV photons. In contrast, in the Veil, we are looking face-on, and hence we are integrating over the history of this processing.

7. Conclusions

We present spatial maps of the 3.3 and 11.2 μm PAH intensities, the IRAC 8.0 μm emission, and the 11.2 μm/3.3 μm and IRAC 8.0 μm/11.2 μm emission ratios in three prominent MIR-bright sources—the RNe NGC 2023 and NGC 7023, and the region southeast of the Orion Bar—based on observations from FLITECAM on board SOFIA, as well as archival Spitzer data. These emission ratio maps provide a measure of the average PAH size distribution and the relative PAH ionization, respectively.

Both RNe exhibit similar trends in these intensity ratios: (i) the average PAH size increases upon approaching the illuminating source, while it is at a minimum along the PDR front; (ii) the relative PAH ionization increases with proximity to the illuminating source; and (iii) In contrast to the 11.0 μm/11.2 μm PAH ratio, the IRAC 8.0 μm/11.2 μm ratio is not a suitable tracer of the relative PAH ionization into the molecular cloud region because of an increasing continuum contribution to the IRAC 8.0 band. In contrast, the region southeast of the Orion Bar exhibits little to no variation in the average PAH size and a peculiar radial profile in the relative PAH ionization: a decrease away from the illuminating source up to 4 farcm 5, followed by a increase at ∼5'. This radial profile does not resemble that of the 11.0 μm/11.2 μm PAH ratio.

We derive an average PAH size of 75 carbon atoms at the PDR front up to 110 carbon atoms nearest to the illuminating source for NGC 2023; 50 carbon atoms at the PDR front, which increases up to 65 carbon atoms near the cavity region surrounding HD 200775 for NGC 7023; and a range of 70 to 85 carbon atoms for the southeast Orion region. For the RNe, larger average PAH sizes correlate with an increase in G₀. In addition, the average PAH size in NGC 7023 is lower at all points in comparison to NGC 2023, whereas NGC 2023 has the widest range of average PAH sizes in the FOV considered. These results support the interpretation of the rise in the UV field intensity with proximity to the illuminating source driving the photochemical evolution of the PAH population that destroys all but the largest species and promote the formation of more stable fullerene molecules. Due to the lack of variation in the average PAH size in the region southeast of the Orion Bar, combined with its complex geometry, we are unable to make any significant conclusions about the photochemical evolution of PAHs here. Future studies of these astronomical sources using spaceborne telescopes with higher spatial resolution instruments such as the James Webb Space Telescope will allow these trends of PAH size and relative ionization to be refined in much more detail.

The authors thank the referee, Kris Sellgren, for reviewing this work and providing in-depth feedback. We are especially thankful to Olivier Berné for his role in the SOFIA proposal, sharing his data, and much insightful feedback. We thank Tom Megeath, Christiaan Boersma, Bavo Croiset, and Brian Fleming for sharing their data with us. We also thank Amy Fare for sharing her PAH emission model calculations. C.K. acknowledges support from an Ontario Graduate Scholarship (OGS). E.P. acknowledges support from an NSERC Discovery Grant and a SOFIA grant. Studies of interstellar PAHs in Leiden are supported through the Spinoza premie of NWO and Horizon 2020 funding awarded under the Marie Skłodowska-Curie action to the EUROPAH consortium (grant No. 722346).

Based (in part) on observations made with the NASA-DLR Stratospheric Observatory for Infrared Astronomy (SOFIA). SOFIA is jointly operated by the Universities Space Research Association, Inc. (USRA), under NASA contract NAS2-97001, and the Deutsches SOFIA Institut (DSI) under DLR contract 50 OK 0901 to the University of Stuttgart. This work is based (in part) on observations made with the Spitzer Space Telescope, which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under a contract with NASA. This publication makes use of data products from the Two Micron All Sky Survey, which is a joint project of the University of Massachusetts and the Infrared Processing and Analysis Center/California Institute of Technology, funded by the National Aeronautics and Space Administration and the National Science Foundation. This work has made use of data from the European Space Agency (ESA) mission Gaia (https://www.cosmos.esa.int/gaia), processed by the Gaia Data Processing and Analysis Consortium (DPAC, https://www.cosmos.esa.int/web/gaia/dpac/consortium). Funding for the DPAC has been provided by national institutions, in particular the institutions participating in the Gaia Multilateral Agreement.

Appendix: Relative PAH Contribution to the IRAC 8.0 μm Emission in a PDR

The IRAC 8.0 μm emission is often used as a tracer for PAH emission (e.g., Hogg et al. 2005; Smith et al. 2007; Stock et al. 2014). Here, we estimate the relative contribution of the individual emission components contained within the IRAC 8.0 μm filter within the South NGC 2023 and the Northwest NGC 7023 IRS SL spectral cubes (e.g., Stock et al. 2016; Peeters et al. 2017; Stock & Peeters 2017). The spectral components in these spectra within the IRAC 8.0 μm filter include the 7.7 and 8.6 μm emission features; the 8 μm bump and 5–10 μm plateau, both of which are attributed to emission of PAH and related species; the 6.9 μm H₂ line; and the underlying dust continuum (where we follow the decomposition given by Peeters et al. 2017). We multiply spectral maps of these emission components by the IRAC 8.0 μm filter response function and integrate over the relevant wavelength range of each feature. Subsequently, we determine the fractional contribution of each spectral component to the total IRAC 8.0 μm flux for a given spectrum.

Figure A1 presents these relative contributions to the IRAC 8.0 μm bandpass along radial cuts across the S ridge PDR front in NGC 2023 and the NW PDR front in NGC 7023 in the direction to the illuminating sources in both RNe. We find that along the radial cut, extending from inside the cavity to the PDR front, the PAH-related species account for ∼80% of the emission within the IRAC 8.0 μm. Beyond the PDR front into the molecular cloud in NGC 2023, PAH-related emission drops down to ∼73% of the total emission. Thus, we can effectively state that the IRAC 8.0 μm filter is a strong tracer of PAH emission within PDR environments but becomes a slightly less reliable tracer in molecular-cloud-dominated regions.

Tracing PAH Size in Prominent Nearby Mid-Infrared Environments

Article metrics

Permissions

Author e-mails

Author affiliations

ORCID iDs

Dates

Abstract

1. Introduction

2. Astronomical Sources