XMM-NEWTON AND CHANDRA OBSERVATIONS OF THE FILLED-CENTER SUPERNOVA REMNANT G63.7+1.1

Figure 1(a) displays the Chandra image in the 0.1–10.0 keV range, smoothed using a two-dimensional Gaussian function (σ = 2''). Several bright X-ray sources are seen in the vicinity of G63.7+1.1. These sources will be discussed further in Sections 4 and 5.2. Contours obtained from 21 cm data obtained with the Dominion Radio Astrophysical Observatory's Synthesis Telescope as part of the Canadian Galactic Plane Survey (CGPS) are overlaid for reference.

Zoom In Zoom Out Reset image size

Here, wavdetect was used to create a list of sources from the Chandra 0.1–10.0 keV data. Using dmfilth, the point sources were then removed and the remaining holes filled, using a Poisson distribution to estimate the amount of diffuse emission that would be in the holes based on the emission surrounding each point source. Then aconvolve was used to smooth the image with a two-dimensional (2D) Gaussian distribution (σ = 20'' in each direction). Figure 1(b) shows the resulting image of the diffuse emission, as seen by Chandra in 29.8 ks, after removal of point sources. The diffuse X-ray emission peaks at α(2000) = 19^h47^m530, δ(2000) = +27°43'516, while the peak of the radio emission is located at α(2000) = 19^h47^m579, δ(2000) = +27°44'226, offset from the X-ray peak by 72''. While the brightest portion of the X-ray nebula is located near the radio peak, all of the observed X-ray emission from the nebula is located southwest of the radio peak.

We obtained a more sensitive XMM-Newton observation with the aim of studying the diffuse emission that was marginally detected in the Chandra data. The XMM-Newton data (0.3–10.0 keV) are shown in Figure 2(a) after smoothing the image using a 2D Gaussian function with σ = 12''.

Zoom In Zoom Out Reset image size

We then removed all point sources in the data to study the morphology of the nebula in X-ray. The holes left behind in the data were filled, as for the Chandra data (using dmfilth). At the same time, the gaps between CCDs were filled using emission on either side of the gap to generate an estimate of the emission in the missing regions. The data from each detector (MOS1, MOS2, pn) were processed separately until the final stages of the image creation, when they were summed into one image, as shown in Figure 2(b). The X-ray nebula measures approximately 32 × 42, elongated in the northeast–southwest direction and brightest in the northeast. We discuss the nature of the observed emission further in Section 5.

The XMM-Newton observation provided a higher number of net counts from the nebula than the Chandra observation (3208 versus 1162), and we focus primarily on these data to study the nebula's spectrum. We comment on the corresponding Chandra spectrum toward the end of this section.

Initially, we created a spectrum from the entire X-ray nebula. An elliptical region containing the diffuse X-ray emission was defined (as shown in Figure 2(a)), and counts from the point sources falling within the nebula were excluded from the spectrum. However, because source 2 (located at the peak of the X-ray emission) was not detected as a point source in the Chandra data and may consist largely of some compact feature in the nebula rather than emission from a point source, it was not excluded from this first fit. The background selected was an elliptical annulus surrounding the source region, excluding the point sources falling within that region (as shown in Figure 2(a)). The net 0.5–10.0 keV count rates for G63.7+1.1 are 1.35 ×10⁻² ct s⁻¹ (MOS1), 1.18 × 10⁻² ct s⁻¹ (MOS2), and 3.13 × 10⁻² ct s⁻¹ (pn). An absorbed power-law fit in the energy range 0.8–8.0 keV (see Table 3) results in N_H = 1.63 (1.12–2.12) × 10²² cm⁻², Γ = 1.75 (1.44–2.15), a normalization of 7.6 (4.5–13.9) × 10⁻⁵ photons keV⁻¹ cm⁻² s⁻¹, and a reduced ${\chi }_{\nu }^{2}$ = 1.11 (ν = 94 degrees of freedom), with the intervals quoted at the 90% confidence level. The corresponding unabsorbed 0.5–10.0 keV flux⁹ is 4.6 (1.0–5.8) × 10⁻¹³ erg cm⁻² s⁻¹. The inferred unabsorbed X-ray luminosity of the nebulas is L_X ∼2 × 10³³${D}_{6}^{2}$ erg s⁻¹, where D₆ = D/(6 kpc) is the approximate distance to G63.7+1.1 (R. Kothes et al. 2016, in preparation). This spectrum is shown in Figure 4 (top left). The column density found here with the absorbed power law was used in the point-source fitting (Section 5.2) to search for the neutron star powering the nebula and when fitting the smaller regions of diffuse emission (discussed below). We note that fitting the entire nebula (with source 2 removed) results in model parameters consistent (within error) with those above: N_H = 1.68 (1.06–2.53) × 10²² cm⁻², Γ = 1.79 (1.39–2.28), norm = 7.6 (4.1–15.9) × 10⁻⁵ photons keV⁻¹ cm⁻² s⁻¹, ${\chi }_{\nu }^{2}$ = 1.23, ν = 106, and absorbed 0.5–10.0 keV flux = 2.6 × 10⁻¹³ erg cm⁻² s⁻¹.

Zoom In Zoom Out Reset image size

Table 3.  Spectral Analysis of the Diffuse X-ray Emission

	Entire PWN	East PWN (src 2 Omitted)	East PWN (src 2 Included)	Central PWN	West PWN
# observed counts	12967	3362	3805	4716	3911
# net counts	3208	993	1222	1165	575
# cts/bin	100	50	50	50	50
NH (×1022 cm−2)	1.63 (1.12–2.12)	1.63 (frozen)	1.63 (frozen)	1.63 (frozen)	1.63 (frozen)
Γ	1.75 (1.44–2.15)	1.60 (1.37–1.84)	1.58 (1.38–1.78)	1.76 (1.54–1.98)	1.91 (1.21–2.70)
norma	7.58 (4.50–13.94) E-5	2.27 (1.67–3.02) E-5	2.9 (2.2–3.7) E-5	3.3 (2.4–4.3) E-5	1.3 (0.5–2.7) E-5
${\chi }_{\nu }^{2}$	1.107	1.046	0.877	0.963	1.425
dof	94	56	64	80	66
Flux (erg cm−2 s−1)	2.25E-13	1.03E-13	1.34E-13	1.15E-13	3.69E-14
unabs.fl.(erg cm−2 s−1)	3.57E-13	1.59E-13	2.07E-13	1.94E-13	6.77E-14
Lum.b (erg s−1)	1.6E33	6.8E32	9.0E32	8.4E32	2.9E32
Plot	Figure 4(a)	Figure 4(b)	...	Figure 4(c)	Figure 4(d)

Notes. All intervals are quoted at the 90% confidence level.

^aNormalization on the power law has units of photons keV⁻¹ cm⁻² s⁻¹. ^bAssuming a distance to G63.7+1.1 of 6 kpc.

Download table as:  ASCIITypeset image

As a further check on the global spectrum, a spectrum of the entire nebula was created from the Chandra data. This spectrum of the diffuse X-ray emission was grouped to a minimum of 20 counts per bin, and an energy range of 0.7–6.0 keV was selected (outside this region there were no counts left after background subtraction). The observed count rate is 0.283 ct s⁻¹, and the background count rate is 0.244 ct s⁻¹, resulting in a net (source) count rate of 0.039 ct s⁻¹. An absorbed power-law fit results in N_H = 1.14 (0.41–2.27) × 10²² cm⁻², Γ = 1.44 (0.77–2.40), a normalization of 6.7 (≤24.2) × 10⁻⁵ photons keV⁻¹ cm⁻² s⁻¹, ${\chi }_{\nu }^{2}$ = 1.05, ν = 49, an absorbed flux of 2.4 × 10⁻¹³ erg cm⁻² s⁻¹, and an unabsorbed flux of 3.7 × 10⁻¹³ erg cm⁻² s⁻¹ (L_X = 1.6 × 10³³ ${D}_{6\;\mathrm{kpc}}^{2}$ erg s⁻¹). We conclude that, within error, the Chandra spectrum gives model parameters consistent with those for the XMM-Newton spectrum. The short exposure of the Chandra observation prevents us, however, from performing spatially resolved spectroscopy of the PWN.

To perform a spatially resolved spectroscopy study of the PWN, we use the XMM-Newton data. The diffuse emission was then divided into three regions (shown in Figure 2(b)) to map any variations in the photon index across the nebula. Any point sources falling within the regions of interest were omitted from the data (with the exception of source 2 in the eastern region, where fits both with and without the source were performed). The background used for all three regions was an elliptical annulus, as shown in Figure 2(a), the same as the background used for the entire nebula. We expect to observe steepening (larger photon index, Γ) as we move away from the candidate pulsar location because of synchrotron losses. Assuming the location of source 2 is the pulsar location, while the error bars on the fit parameters are large, we see this general trend for G63.7+1.1. Moving from east to west across the nebula, we find a photon index of 1.6 (1.4–1.8), 1.8 (1.5–2.0), and 1.9 (1.2–2.7) for each of the regions. The fit parameters are given in Table 3, and the power-law fit to each region is shown in Figure 4. We note that keeping source 2 in the eastern spectrum yields fit parameters consistent within error with those found when source 2 is omitted (Table 3).

As we move westward across the nebula, there is an excess of counts at low energies in both the MOS and pn data, indicating an increasing need for an additional (soft) model component (see Figure 4(d)). This component may be due to unrelated material in the foreground or thermal emission from an undetected SNR component (shell or ejecta). To test for this possible scenario, we add an Astrophysical Plasma Emission Code (APEC, Smith et al. 2001) component to the model for the western region of the nebula. We find that an additional thermal component improves the fit (${\chi }_{\nu }^{2}$ = 1.02, ν = 64, compared to ${\chi }_{\nu }^{2}$ = 1.43, ν = 66 for the single power-law model; F-test probability ∼7 × 10⁻⁶), with the power-law parameters consistent with those for the single-component power-law fit (Γ = 1.7${}_{-0.7}^{+0.7}$, normalization = 1.1${}_{-0.7}^{+1.2}$ × 10⁻⁵ photons keV⁻¹ cm⁻² s⁻¹). The additional thermal component is characterized by a temperature kT = 0.054${}_{-0.010}^{+0.005}$ keV, with a normalization = 120${}_{-100}^{+800}$ cm⁻⁵ and a corresponding observed flux (despite being poorly constrained) at ∼9% of the total observed flux. Given the low temperature of this thermal component, it is unlikely to be due to the SNR shell or an interaction between the PWN and a reverse shock, and it may instead be due to foreground or unrelated material. A deeper exposure is needed to either confirm or refute this scenario.

The XMM-Newton diffuse spectra were also fit with alternate backgrounds to test the effect of the background on the spectra. We find that, while there is some variation in the model parameters, all are consistent (within error) with the results shown in Table 2.

While one or more sources may correspond to stars or extragalactic sources, we fit the spectrum of each observed point (or compact, in the case of source 2) source of interest to test consistency with a source located in G63.7+1.1. Spectra were extracted from each of MOS1, MOS2, and pn separately and then fit simultaneously. The background for each point source was defined to be an annular region surrounding the point source, with a maximum radius equal to twice the radius of the source region. Here, evselect was used to create the source and background spectra, and then rmfgen and arfgen were used to generate the response matrices; grppha was used to bin the spectra to some minimum number of counts per bin.

Absorbed power-law and absorbed blackbody fits for each point source listed in Table 1 and labeled in Figure 3 are presented in Table 4. The column density, N_H, was allowed to vary since the distance to each source is unknown. Because of the relatively small number of counts, we also fit with N_H = 1.63 × 10²² cm⁻² (a value derived from the X-ray nebula) in order to better constrain the fit parameters in the event that the source is indeed associated with the nebula. All error bars quoted are 90% confidence intervals.

Table 4.  Spectral Analysis of X-ray Sources Detected in G63.7+1.1

XMMSrc #	1	2	3	4a	5	6
# Observed Counts (0.3–10.0 keV)	221	301	120	217	120	720
# Net Counts (0.3–10.0 keV)	82	94	68	155	34	537
Minimum # cts/bin	5	10	5	5	5	10
	wabs*power with NH free
NH (×1022 cm−2)	${0.90}_{-0.81}^{+1.25}$	8.3${}_{-8.3}^{+21.5}$	4.8${}_{-4.2}^{+14.1}$	2.3${}_{-1.4}^{+1.8}$	0${}_{-0}^{+1.6}$	1.1${}_{-0.3}^{+0.3}$
Γ	${9.3}_{-5.4}^{+12.1}$	${3.0}_{-3.2}^{+6.2}$	${1.5}_{-1.6}^{+3.9}$	${2.0}_{-0.9}^{+1.2}$	${3.8}_{-2.1}^{+6.0}$	${10.0}_{-1.8}^{+2.3}$
norm (photons keV−1 cm−2 s−1)	${9}_{-9}^{+262}$ E-6	${4.8}_{-4.8}^{+284000}$ E-5	${5.4}_{-5.4}^{+5870}$ E-6	${2.2}_{-2.2}^{+9.1}$ E-5	${3.0}_{-3.0}^{+6.7}$ E-7	${1.9}_{-1.0}^{+2.6}$ E-4
${\chi }_{\nu }^{2}$ (ν)	1.16 (35)	0.65 (25)	0.98 (19)	1.06 (38)	1.05 (19)	1.34 (58)
Flux (erg cm−2 s−1, 0.5–10.0 keV)	1.8E-15	1.4E-14	2.2E-14	4.9E-14	9.3E-16	2.1E-14
	wabs*power with NH = 1.63 × 1022 cm−2
Γ	${14.4}_{-3.1}^{+2.6}$	${1.1}_{-0.9}^{+0.9}$	${0.6}_{-0.9}^{+0.8}$	${1.6}_{-0.4}^{+0.4}$	${10.9}_{-3.4}^{+6.3}$	${13.5}_{-0.4}^{+0.4}$
norm (photons keV−1 cm−2 s−1)	${5.3}_{-4.1}^{+5.1}$ E-5	${2.0}_{-2.0}^{+4.2}$ E-6	${1.0}_{-1.1}^{+1.7}$ E-6	${1.2}_{-0.6}^{+0.7}$ E-5	${1.9}_{-1.9}^{+4.8}$ E-5	${7.7}_{-1.4}^{+1.4}$ E-4
${\chi }_{\nu }^{2}$ (ν)	1.14 (36)	0.64 (26)	0.99 (20)	1.04 (39)	0.96 (20)	1.42 (59)
Flux (erg cm−2 s−1, 0.5–10.0 keV)	1.7E-15	2.0E-14	2.6E-14	5.2E-14	4.0E-16	2.0E-14
6 kpc Lum. (erg s−1, 0.5–10.0 keV)b	1.6E34	1.1E32	1.3E32	3.5E32	7.0E33	1.4E36
	wabs*bbody with NH free
NH (×1022 cm−2)	${0.08}_{-0.08}^{+0.83}$	${0}_{-0}^{+22}$	${2.4}_{-2.3}^{+12.6}$	${0.5}_{-0.5}^{+1.1}$	${0}_{-0}^{+3.3}$	${0.48}_{-0.19}^{+0.23}$
kT (keV)	${0.17}_{-0.09}^{+0.09}$	${1.7}_{-1.2}^{+1.5}$	${1.5}_{-0.7}^{+2.0}$	${1.2}_{-0.3}^{+0.4}$	${0.18}_{-0.14}^{+0.20}$	${0.11}_{-0.03}^{+0.02}$
normc	${5}_{-5}^{+812}$ E-8	${2.6}_{-1.6}^{+4.8}$ E-7	${3.1}_{-1.7}^{+7.8}$ E-7	${5.9}_{-1.7}^{+2.1}$ E-7	${3.0}_{-3.0}^{+3582.1}$ E-8	${8.2}_{-8.2}^{+42.0}$ E-6
${\chi }_{\nu }^{2}$ (ν)	1.18 (35)	0.64 (25)	0.97 (19)	1.10 (38)	1.20 (19)	1.18 (58)
Flux (erg cm−2 s−1, 0.5–10.0 keV)	1.9E-15	1.8E-14	1.9E-14	4.3E-14	1.7E-15	2.1E-14
	wabs*bbody with NH = 1.63 × 1022 cm−2
kT(keV)	${0.066}_{-0.030}^{+0.026}$	${1.3}_{-0.5}^{+1.0}$	${1.7}_{-0.6}^{+1.6}$	${1.0}_{-0.2}^{+0.3}$	${0.09}_{-0.03}^{+0.04}$	${0.052}_{-0.003}^{+0.003}$
normc	${6.9}_{-6.9}^{+16000}$ E-4	${2.3}_{-1.3}^{+2.4}$ E-7	${3.1}_{-1.8}^{+7.1}$ E-7	${6.4}_{-1.5}^{+1.7}$ E-7	${1.2}_{-1.2}^{+86.3}$ E-5	${1.3}_{-0.7}^{+1.3}$ E-1
${\chi }_{\nu }^{2}$ (ν)	1.27 (36)	0.62 (26)	0.93 (20)	1.15 (39)	0.94 (20)	1.70 (59)
Flux (erg cm−2 s−1, 0.5–10.0 keV)	1.2E-15	1.5E-14	2.0E-14	3.8E-14	3.2E-16	1.6E-14
6 kpc Lum. (erg s−1, 0.5–10.0 keV)b	1.3E34	7.8E31	9.7E31	2.3E32	7.3E32	6.0E35

Notes. All intervals are quoted at the 90% confidence level.

^aSource 4 was fit using MOS1 and MOS2 data only because the source was located on the edge of a pn CCD. ^bSource luminosity, assuming it is at the same distance as the PWN. ^cWhere bbody norm = ${L}_{39}/{D}_{10}^{2}$ with L₃₉ = L/(10³⁹ erg s⁻¹) and D₁₀ = D/(10 kpc).

Download table as:  ASCIITypeset image

Source 1 has a low column density in either fit when compared to the column density of the nebula (Section 5.1), which argues against source 1 being associated with G63.7+1.1 and points to source 1 being a foreground source. Furthermore, the photon index for source 1 with N_H frozen to the nebula's value is unrealistically high (14.4${}_{-3.1}^{+2.6}$), and the blackbody temperature is much too low (0.066${}_{-0.030}^{+0.026}$ keV) for a neutron star powering a PWN. Overall, the power-law fit with N_H frozen gives a very poor fit below 0.8 keV and above 2 keV, and the blackbody fit gives a very poor fit above 1 keV. Additionally, the luminosity derived for source 1 if it were at the distance of G63.7+1.1 (1.6 × 10³⁴ ${D}_{6\;\mathrm{kpc}}^{2}$ erg s⁻¹ with the power-law model and 1.3 × 10³⁴${D}_{6\;\mathrm{kpc}}^{2}$ erg s⁻¹ with the blackbody model) is greater than the luminosity of the entire nebula (1.6 × 10³³ ${D}_{6\;\mathrm{kpc}}^{2}$ erg s⁻¹), unreasonable for a neutron star powering a PWN. Creating a spectrum for source 1 from the Chandra data gives parameters that are consistent with, but not as well constrained as, the XMM-Newton data.

Fitting source 2 with either a power-law or a blackbody model results in an acceptable fit. The column density found for source 2 in either fit, while having a large uncertainty, is consistent with the column density found for the nebula. Additionally, the photon index found for source 2 (1.1 ± 0.9 with N_H frozen) is consistent with emission from neutron stars powering other PWNe (see, e.g., Table 2 in Kargaltsev & Pavlov 2008). The observed flux from source 2 is 2.0 × 10⁻¹⁴ erg cm⁻² s⁻¹, and its luminosity, assuming the same distance as the PWN, is 1.1 × 10³² ${D}_{6\;\mathrm{kpc}}^{2}$ erg s⁻¹, an order of magnitude lower than the nebula's luminosity (1.6 × 10³³ ${D}_{6\;\mathrm{kpc}}^{2}$ erg s⁻¹). This is also consistent with what is often observed in X-rays from other PWNe (see Table 2 in Kargaltsev & Pavlov 2008). We note that source 2 appears to be the best candidate for the neutron star powering the PWN, but a longer exposure is needed to confirm this scenario. We caution the reader that the flux from any point source may be significantly lower than the flux presented here if the bulk of the source 2 emission is in fact due to a compact structure in the PWN, such as a torus formed at the location of the wind termination shock.

The other sources studied that lie within the X-ray nebula are all much farther from the peak of the diffuse emission. Source 3 has a high, but poorly constrained, column density; with an N_H fixed to that of the PWN, its spectrum has a very hard power-law index (Γ = ${0.6}_{-0.9}^{+0.8}$). This source may be an active galactic nucleus. Source 4 has an N_H consistent within error with that of G63.7+1.1, and when fixed to that of the PWN, its photon index, Γ = 1.6 ± 0.4, is consistent with that expected from a neutron star. However, its position with respect to the nebula makes its association unlikely. Source 5 has a very low column density (i.e., consistent with zero) and very poor quality fits above 2 keV, and it is also located at the edge of the faint diffuse emission. Source 6 is very bright and very soft, with a column density of (1.1 ± 0.3) × 10²² cm⁻², lower than that of G63.7+1.1. Also arguing against sources 5 or 6 as the powering engine of G63.7+1.1 is that both sources yield luminosities greater than that of the entire nebula if they are assumed to be at 6 kpc (7.0 × 10³³ ${D}_{6\;\mathrm{kpc}}^{2}$ erg s⁻¹ and 1.4 × 10³⁶ ${D}_{6\;\mathrm{kpc}}^{2}$ erg s⁻¹, respectively). In conclusion, source 2 (3XMM J194753.4+274357/CXO J194753.3+274351) appears to be the best candidate for a pulsar search in G63.7+1.1.

In X-rays, G63.7+1.1 is approximately 32 × 42, elongated in the northeast–southwest direction and brightest in the northeast (Figure 2). At a distance of 6 kpc, the diameter of the 8' radio nebula is 14 D₆ pc, and the 32 × 42 X-ray nebula is ∼5.6 × 7.3 D₆ pc, where D₆ = D/(6 kpc). This implies that G63.7+1.1 is a large PWN (at least at radio wavelengths). This size, however, is not the largest estimated for a PWN to date. It is comparable to the radio size of G327.1–1.1 (14 pc, Temim et al. 2009) and smaller than the estimated radio sizes of the PWNe G76.9+1.0 (29 pc×35 pc, Arzoumanian et al. 2011) and CTB 87 (28 pc diameter, Matheson et al. 2013).

In Figure 5 we show the position of G63.7+1.1 in relation to the surrounding material. The multiwavelength view of the surrounding material will be the focus of a follow-up paper. We note here that the radio nebula (21 cm continuum, white contours) fits inside a cavity in a ridge of H i (colored red) running northeast to southwest, with a faint H i shell at the western boundary of the nebula. The X-ray nebula (colored blue) is within the southwest portion of this hole, and the radio nebula is extended in the north–south direction, corresponding to a lower density in the surrounding material (which would have allowed more expansion of the PWN). We also note that there is dense molecular material (¹²CO, colored green in Figure 5) at the southeastern and western limit of the nebula, approximately at the boundary of the X-ray emission, and that the peak of the X-ray emission (source 2 in Table 1) falls in a depolarized gap running north to south through the 21 cm nebula. More detailed images with a wider field of view and a discussion of the source's distance (5.8 ± 0.9 kpc, based on the systemic velocity of +13 km s⁻¹ determined using both the H i and CO data) will be presented in a follow-up paper (R. Kothes et al. 2016, in preparation).

Zoom In Zoom Out Reset image size

Why we do not see any X-ray emission in the northeast remains an open question. Since G63.7+1.1 is evolving in a dense region, it may have already evolved enough that the SNR is in the Sedov phase and the reverse shock has interacted with the majority of the PWN. The shock-heated material may then have sufficiently cooled that it is no longer emitting at X-ray wavelengths. It is also possible that the soft X-ray emission expected from a reverse-shock/PWN interaction or the SNR is largely obscured by the surrounding or intervening material detected in CO. We note that the radio peak is not at the geometrical center of the radio nebula. The distance from the radio peak to the southwestern edge is greater than the distance from the peak to the northeastern edge. A possible explanation for this structure is that if the PWN was crushed some time ago, and the reverse shock hit first from the southwest and later from the northeast, then the PWN would have had more time to reexpand to the southwest. This picture may be further complicated by the three-dimensional nature of the PWN and line-of-sight effects. We note that at the limit of the X-ray nebula the X-ray emission does not have a sharply defined edge and merges into the background. A deeper exposure will allow the detection of the full extent of the X-ray emission.

Without a detection of the pulsar powering the PWN or an SNR shell detection, estimating the age of G63.7+1.1 is difficult. We note that the age estimate below has a large uncertainty, and we scale our subsequent results by the age to make the effect on the derived properties obvious.

Using an average 2D speed for pulsars of 246 ± 22 km s⁻¹ (Hobbs et al. 2005), and assuming the pulsar's birth location as the location of the radio peak and the pulsar's current location as the location of the X-ray peak (source 2),¹⁰ we estimate a rough age for G63.7+1.1 as ∼8 kyr (for a distance of 6 kpc). While this assumption on birth location does not necessarily hold true for all systems with a reverse-shock interaction and will be further addressed in our upcoming radio study of this remnant, we note that this age estimate is comparable to those for other evolved PWNe: Vela X (11 kyr, La Massa et al. 2008), G327.1–1.1 (18 kyr, Temim et al. 2009), including the shell-less PWNe G76.9+1.0 (9 kyr, Arzoumanian et al. 2011), and CTB 87 (∼10–15 kyr, Matheson et al. 2013).

Vela X, the PWN powered by the Vela pulsar (PSR B0833-45), is an example of an evolved PWN in a complicated evolutionary stage, where a reverse shock from the host SNR is colliding with the PWN (Blondin et al. 2001; Mattana et al. 2011). Its X-ray emission is described by a power law with a photon index of ∼1.4–1.6, and its pulsar has a spin-down luminosity of $\dot{E}\;=\;6.9\times {10}^{36}$ erg s⁻¹ and a characteristic age of 11.4 kyr (Dodson et al. 2007), all similar to our estimates for G63.7+1.1 and its putative engine (see Section 6.3). If the PWN is being crushed and interacting with the reverse shock, the radio nebula is a "relic," as seen in other remnants such as G327.1–1.1 (Temim et al. 2009) and CTB 87 (Matheson et al. 2013). The detection of teraelectronvolt emission overlapping the radio emission would support this scenario.

We conclude that G63.7+1.1 appears to be a PWN that has evolved enough for the reverse-shock interaction with the PWN to have already occurred or currently be in progress. This interaction at a young age could be due to the high density of the surrounding medium (Kothes et al. 2016, in preparation).

As discussed in previous sections, source 2 marks a candidate location for the neutron star powering G63.7+1.1. In the absence of a pulsar detection, however, we can crudely estimate the properties of the neutron star from the X-ray luminosity of the nebula since PWNe are calorimeters for pulsars (see, e.g., Olbert et al. 2003; Arzoumanian et al. 2008; Kargaltsev & Pavlov 2008; Tüllmann et al. 2010 for a review and references therein). The unabsorbed flux of G63.7+1.1 with the XMM-Newton best fit to the entire PWN, in the 2 − 10 keV band, is ${F}_{\mathrm{unabs},2-10\mathrm{keV}}\;=\;2.9\times {10}^{-13}$ erg cm⁻² s⁻¹, implying a luminosity of ${L}_{{\rm{X}},2-10\mathrm{keV}}\;=\;1.2\times {10}^{33}\;{D}_{6}^{2}$ erg s⁻¹. Estimating the spin-down energy loss of the putative pulsar using the empirical relationship ${L}_{{\rm{X}},2-10\mathrm{keV}}\;=\;{10}^{-19.6}{\dot{E}}^{1.45}$ (Li et al. 2008), we obtain $\dot{E}\;=\;2.1\times {10}^{36}\;{D}_{6}^{1.38}$ erg s⁻¹, implying an efficiency in converting the spin-down luminosity to synchrotron emission of ${\eta }_{{\rm{X}}}\equiv {L}_{{\rm{X}}}/\dot{E}\approx 0.0006\;{D}_{6}^{0.62}$. We estimate the characteristic age as $\tau \;=\;27{D}_{6}^{-0.95}$ kyr using ${L}_{{\rm{X}},2-10\mathrm{keV}}\;=\;{10}^{42.4}{\tau }^{-2.1}$ (Li et al. 2008). This is only a few times larger than the SNR's age estimate given in Section 6.2. We caution the reader that these values of $\dot{E}$ and τ give a rough indication of the putative pulsar properties. For example, there is an observed spread in the L_X–$\dot{E}$ relationship (e.g., Possenti et al. 2002). Furthermore, we know that most pulsars' characteristic ages differ from those of their hosting SNRs (see, e.g., Rogers & Safi-Harb 2016 and references therein). Timing observations in radio and X-ray are needed to determine the pulsar spin properties (including its rotation period and magnetic field). Furthermore, a deep search and spectroscopic studies of the SNR shell would give a much more complete picture of the G63.7+1.1 system. A multiwavelength modeling of the remnant, taking into account the PWN's radio and X-ray spectrum and size and using a model for its evolution inside an SNR (e.g., Gelfand et al. 2009), is beyond the scope of this work and will be addressed in a future study.