Neutral Pion Emission from Accelerated Protons in the Supernova Remnant W44

Article (PDF Available)inThe Astrophysical Journal Letters 742(2):L30 · November 2011with32 Reads
DOI: 10.1088/2041-8205/742/2/L30 · Source: arXiv
Abstract
We present the AGILE gamma-ray observations in the energy range 50 MeV-10 GeV of the supernova remnant (SNR) W44, one of the most interesting systems for studying cosmic-ray production. W44 is an intermediate-age SNR (~20, 000 years) and its ejecta expand in a dense medium as shown by a prominent radio shell, nearby molecular clouds, and bright [S II] emitting regions. We extend our gamma-ray analysis to energies substantially lower than previous measurements which could not conclusively establish the nature of the radiation. We find that gamma-ray emission matches remarkably well both the position and shape of the inner SNR shocked plasma. Furthermore, the gamma-ray spectrum shows a prominent peak near 1 GeV with a clear decrement at energies below a few hundreds of MeV as expected from neutral pion decay. Here we demonstrate that (1) hadron-dominated models are consistent with all W44 multiwavelength constraints derived from radio, optical, X-ray, and gamma-ray observations; (2) ad hoc lepton-dominated models fail to explain simultaneously the well-constrained gamma-ray and radio spectra, and require a circumstellar density much larger than the value derived from observations; and (3) the hadron energy spectrum is well described by a power law (with index s = 3.0 ± 0.1) and a low-energy cut-off at Ec = 6 ± 1 GeV. Direct evidence for pion emission is then established in an SNR for the first time.
Draft version November 22, 2011
Preprint typeset using L
A
T
E
X style emulateapj v. 11/10/09
NEUTRAL PION EMISSION FROM ACCELERATED PROTONS IN THE SUPERNOVA REMNANT W44
A. Giuliani
1
, M. Cardillo
2,3
, M. Tavani
2,4,5
, Y. Fukui
6
, S. Yoshiike
6
, K. Torii
6
,
G. Dubner
7
, G. Castelletti
7
, G. Barbiellini
8
, A. Bulgarelli
9
, P. Caraveo
1
, E. Costa
2
, P.W. Cattaneo
10
,
A. Chen
1
, T. Contessi
1
, E. Del Monte
2
, I. Donnarumma
2
, Y. Evangelista
2
, M. Feroci
2
, F. Gianotti
9
,
F. Lazzarotto
2
, F. Lucarelli
12
, F. Longo
8
, M. Marisaldi
9
, S. Mereghetti
1
, L. Pacciani
2
, A. Pellizzoni
11
,
G. Piano
2
, P. Picozza
3
, C. Pittori
12
, G. Pucella
13
, M. Rapisarda
2
, A. Rappoldi
10
, S. Sabatini
2
, P. Soffitta
2
,
E. Striani
2
, M. Trifoglio
9
, A. Trois
11
, S. Vercellone
14
, F. Verrecchia
12
, V. Vittorini
2
,
S. Colafrancesco
16,17
, P. Giommi
12
, G. Bignami
15
Draft version November 22, 2011
ABSTRACT
We present the AGILE gamma-ray observations in the energy range 50 MeV - 10 GeV of the super-
nova remnant (SNR) W44, one of the most interesting systems for studying cosmic-ray production.
W44 is an intermediate-age SNR ( 20, 000 years) and its ejecta expand in a dense medium as shown
by a prominent radio shell, nearby molecular clouds, and bright [SII] emitting regions. We extend our
gamma-ray analysis to energies substantially lower than previous measurements which could not con-
clusively establish the nature of the radiation. We find that gamma-ray emission matches remarkably
well both the position and shape of the inner SNR shocked plasma. Furthermore, the gamma-ray
spectrum shows a prominent peak near 1 GeV with a clear decrement at energies below a few hun-
dreds of MeV as expected from neutral pion decay. Here we demonstrate that: (1) hadron-dominated
models are consistent with all W44 multiwavelength constraints derived from radio, optical, X-ray,
and gamma-ray observations; (2) ad hoc lepton-dominated models fail to explain simultaneously the
well-constrained gamma-ray and radio spectra, and require a circumstellar density much larger than
the value derived from observations; (3) the hadron energy spectrum is well described by a power-law
(with index s = 3.0 ± 0.1) and a low-energy cut-off at E
c
= 6 ± 1 GeV. Direct evidence for pion
emission is then established in an SNR for the first time.
Subject headings: ISM: supernova remnants (W44) cosmic rays acceleration of particles
gamma rays: general
1. INTRODUCTION
Providing an unambiguous proof of the cosmic-ray ori-
gin until now has been elusive, despite many decades
of attempts and controversial claims ( e.g., Fermi 1949;
1
INAF-IASF Milano, via E.Bassini 15, 20133 Milano, Italy
2
INAF/IASF-Roma,via Del Fosso del Cavaliere 100, 00133
Roma, Italy
3
Dipartimento di Fisica, Univ.di Roma “Tor Vergata”, via
della Ricerca Scientifica 1, 00133 Roma, Italy
4
Consorzio Interuniversitario Fisica Spaziale (CIFS), villa
Gualino, v.la Settimo Severo 63, 10133, Torino, Italy
5
INFN Roma Tor Vergata, via della Ricerca Scientifica 1,
00133 Roma, Italy
6
Department of Astrophysics, Nagoya University, Chikusa-ku,
Nagoya 464-8602, Japan
7
Instituto de Astronomia y Fisica del Espacio (IAFE,
CONICET-UBA), 1428 Buenos Aires, Argentina
8
Dipartimento di Fisica and INFN, Via Valerio 2, I-34127 Tri-
este, Italy
9
INAF/IASF-Bologna, Via Gobetti 101, I-40129 Bologna,
Italy
10
INFN-Pavia, Via Bassi 6, I-27100 Pavia, Italy
11
INAF, Osservatorio Astronomico di Cagliari, Poggio dei
Pini, strada 54, i-09012 Capoterra, italy
12
ASI-ASDC, Via G.Galilei, I-00044 Frascati (Roma), Italy
13
ENEA-Frascati, Via E.Fermi 45, I-00044 Frascati (Roma),
Italy
14
INAF/IASF-Palermo, Via U.La Malfa 153, I-90146
Palermo, Italy
15
Istituto Universitario di Studi Superiori (IUSS), viale Lungo
Ticino Sforza 56, 27100 Pavia, Italy
16
INAF, Osservatorio Astronomico di Roma, Monte Porzio
Catone, Italy
17
School of Physics, University of the Witwatersrand, Johan-
nesburg Wits 2050, South Africa.
Ginzburg et al. 1964; Torres et al. 2003; Aharonian 2004;
Berezhko et al 2007; Butt 2009). Cosmic-rays are mainly
protons and heavy ions (hadrons) and, in a few percent,
electrons and positrons. Supernova remnants (SNRs) are
ideal candidates for the cosmic-ray production up to en-
ergies near E
knee
= 10
15
eV. The SNR energy output in
the Galaxy can indeed supply the energy budget neces-
sary to maintain the present population of cosmic-rays.
Furthermore the observations of ultra-relativistic elec-
trons support the hypothesis that also protons are accel-
erated in these objects (for a recent review see Reynolds
2008, and references therein). Proving the fact that the
SNR origin of hadronic cosmic-rays is difficult because
of the complexity of the SNR-environment interaction.
From an observational point of view, a direct proof can
be given by an unambiguous detection of the gamma-ray
emission expected from neutral pion decay in hadronic
interactions. However radiation from co-spatially ac-
celerated electrons can mask and sometimes overcome
the expected neutral pion decay signature of proton/ion
emission in the 100 MeV-a few TeV energy range. Re-
cent analysis suggests that several gamma-ray observa-
tions of SNRs can be understood in terms of acceler-
ated hadrons (Castro & Slane 2010; Giuliani et al. 2010;
Abdo et al. 2010b,c,a, 2009). However it is currently not
possible to exclude that the observed γ-ray emission is
produced by leptons alone. The discrepancies between
leptonic and hadronic models are expected to be more
evident at low energies (50 - 100 MeV). Gamma-ray as-
tronomy in this energy band is very challenging because
arXiv:1111.4868v1 [astro-ph.HE] 21 Nov 2011
2 Giuliani et al.
of high background-noise flux and of the strong multi-
ple scattering suffered by electrons originating from γ-
ray events. The AGILE/GRID instrument (calibrated
in the 50 MeV - 10 GeV band), however, has already
shown its ability to provide an energy spectrum starting
at 50 MeV for bright objects (Vercellone et al. 2009; Giu-
liani et al. 2010; Vittorini et al. 2011). In this paper we
report on a low-energy γ-ray and multiwavelength spec-
trum for the SNR W44 in order to constraint the emitting
particle spectrum and discriminate between leptonic and
hadronic models.
2. THE SUPERNOVA REMNANT W44
SNR W44 (G34.7 0.4) is a well studied middle-aged
(20,000 yr) SNR located in the Galactic disk at a dis-
tance of 3 kpc from Earth (Clark et al. 1976; Wol-
szczan et al. 1991). W44 is an ideal system to test the
presence of accelerated hadrons and the interplay be-
tween hadronic and leptonic models. Radio (Castelletti
et al. 2007, and references therein) and X-ray (Wat-
son et al. 1983) mapping of the SNR show a roughly
elliptical shocked shell and a centrally peaked emission
respectively. In the IR band (Reach et al. 2005) it is evi-
dent that the shell is expanding into a dense surrounding
medium (n 100 cm
3
). Wolszczan et al. (1991) dis-
covered the radio pulsar PSR B1853+01 with distance
and age compatible with the SNR. Wootten et al. (1977)
and then Rho et al. (1994) found a massive molecular
cloud interacting with the south-eastern side of the rem-
nant. Evidence for a more complex system of massive
MCs and for their interactions with the remnant, shown
by some features typical of a strong shock, was given by
Seta et al. (2004) and then by Reach et al. (2005). The
MC-SNR interactions were confirmed by the maser OH
(1720 MHz) emission reported by Claussen et al. (1997)
and then by Hoffman et al. (2005).
The first estimation of the spectral radio index varia-
tions as a function of position over the remnant was done
by Castelletti et al. (2007): the eastern limb spectrum
was consistent with a diffusive shock acceleration model
and the spectrum flattening in the westernmost arc con-
firmed the MC-SNR interaction observed in IR and op-
tical band.
Gamma-ray emission from this SNR has been detected
by the Fermi/LAT instrument at energy E>200 MeV
(Abdo et al. 2010a), suggesting the presence of acceler-
ated protons interacting with the surrounding medium.
3. DATA ANALYSIS
AGILE-GRID data were analyzed using the AGILE
Standard Analysis Pipeline. We used γ-ray events fil-
tered by means of the F M3.119 2 AGILE Filter Pipeline
(as described in Vercellone et al. 2008). In order to dis-
criminate between background events and gamma rays,
the GRID and anticoincidence system (ACS) signals are
processed, reconstructed and selected by a dedicated
software (Giuliani et al. 2006). We used the most re-
cent versions of the diffusion model (Giuliani et al. 2004)
and of the calibration files, available at the ASDC site
(www.asdc.asi.it). We created counts, exposure and
Galactic background gamma-ray maps with a bin-size
of 0.02
x 0.02
. In order to derive the source average
flux and spectrum we ran the AGILE point source anal-
ysis software ALIKE (Bulgarelli et al. 2011, based on
the maximum likelihood technique described in Mattox
et al. (1993)) over the whole observing period 2007 July
- 2011 April. Both statistic and systematic uncertainties
are taken into account. The spectrum was obtained by
computing the γ-ray flux in six energy bins selected with
the aim to have a significance σ > 4. In order to study
the source morphology, we obtained an intensity map in-
tegrated over the energy range where the GRID angular
resolution is optimal (E>400 MeV).
4. RESULTS AND DISCUSSION
AGILE detects SNR W44 with a significance of 15.8 σ
as an extended source. Figure 1a shows AGILE gamma-
ray intensity map above 400 MeV of the W44 region
with the 324 MHz VLA radio contours. The gamma-
ray morphology remarkably resembles the quasi-elliptical
pattern of the interior of the radio shell, especially coin-
ciding with the radio brightness enhancements toward
the north-west and south-east regions. Both the radio
pulsar PSR B1803+01 position (Petre et al. 2002) and
its (small) pulsar wind nebula (Giacani et al 1997, and
references therein) are inconsistent with the gamma-ray
morphology detected by AGILE. Moreover, Abdo et al.
(2010a) excluded the presence of a pulsation in the γ-ray
signal. Figure 1b shows the CO emission at a kinematic
velocity compatible with the distance of W44, tracing
the presence of MCs in the W44 surroundings, together
with the gamma-ray contour levels. It can be inferred
that the south-eastern side of the γ-ray source overlaps
with the MC - SNR interaction region. In the northern
part of the shell, the γ-ray and CO emissions are not
correlated, however many studies of the surrounding in-
terstellar medium showed the presence of dense gas not
traced by CO (Reach et al. 2005). Figure 1c displays an
SII map overlaid with X-ray and gamma-ray contours.
The strong sulfur [SII] emission, along with Hα emission,
indicates the presence of shocked gas (Draine & McKee
1993) .
18
4.1. The gamma-ray spectrum
Figure 2 shows the AGILE W44 photon energy spec-
trum for the whole range 50 MeV- 10 GeV. The measured
flux above 400 MeV is F = (16.0 ± 1.2) × 10
8
pho-
tons cm
2
s
1
. In this figure are also shown Fermi/LAT
spectral points (Abdo et al. 2010a) over an energy range
0.2-30 GeV. In the band where the spectra overlap, the
sets of data are compatible at 1σ. In this regard it is im-
portant to stress how AGILE is able to detect the emis-
sion from W44 in the energy range 50 MeV-300 MeV
extending the spectrum to energies substantially lower
than those previously obtained. This spectrum shows a
clear decrement at photon energies lower than 400 MeV,
confirming the expectations based on neutral pion emis-
sion from accelerated protons/ions: a peak energy near
1 GeV for hadron energy spectra flatter than E
2
(e.g.,
Aharonian 2004). The gamma-ray spectrum becomes
steep at higher energies with a photon power-law index
α 3±0.1, in agreement with previous measurements
(Abdo et al. 2010a).
18
As pointed out by Rho et al. (1994) in their X-ray and optical
study of W44, the optical filaments and the X-ray image showing
locally bright emission clumps along the filaments suggest that
both are produced by the interaction between the SN shock front
and regions of enhanced ambient density.
3
Fig. 1.— (a): AGILE gamma-ray intensity map (in Galactic coordinates) of the W44 region (1.1 x 0.75 degrees) in the energy range
400 MeV - 3 GeV obtained by integrating all available data collected during the period 2007 July and 2011 April. The color bar scale is
in units of 10
5
photons cm
2
s
1
pixel
1
. Pixel size is 0.02 degrees with a six-bin Gaussian smoothing. Green contours show the 324
MHz radio continuum flux density detected by the Very Large Array. The white cross indicates the position of the pulsar PSR B1803+01.
Source A does not appear to be associated with W44. (b): combined CO data from the NANTEN Observatory superimposed with the
AGILE gamma-ray data contours above 400 MeV (in white) of the W44 region (map in Galactic coordinates). CO data have been selected
in the velocity range 35 - 45 km s
1
corresponding to a kinematic distance compatible with the W44 distance. (c): the [SII] image of SNR
W44 obtained in 1993 with the 0.6 m Burrell-Schmidt telescope at Kitt Peak National Observatory with a matched red continuum image
subtracted to emphasize the faint nebulosity (adapted from Giacani et al 1997). The overlaid contours trace the gamma-rays emission
detected by AGILE between 400 MeV and 3 GeV (in red) and the X-ray emission detected by ROSAT in the 0.1 - 2.4 keV range (in blue).
4.2. Lepton-dominated models of emission
Leptonic-only models of gamma-ray production have
to satisfy the very well determined spatial and spectral
constraints provided by the radio, optical, and gamma-
ray emissions. For both the Bremsstrahlung and in-
verse Compton cases we tested the class of leptonic-only
models attempting to reproduce the observed gamma-
ray spectrum for different input leptonic spectra (see
Table 1). We used a phenomenological approach
based on two evidences: a) the synchrotron spectrum
implies that the radio electron distribution is well de-
scribed by a power-law over a wide range of energies,
b) the discontinuity in the slope of the gamma-rays
spectrum implies a discontinuity in the emitting par-
ticle spectrum. We assumed three electron distri-
butions that can, in principle, describe this behavior:
1) a power-law with a high energy cut-off, F
1
e(E) =
K
e
E
p
e
E
E
c
, (as in Hendrick et al. 2001), 2) a power-law
with a low energy cut-off, F
2
e(E) = K
e
E
p
e
E
c
E
, (as in
Gabici et al. 2009), 3) a broken power-law, F
3
e(E) =
K
e
E
E
c
p
1
1
2
1 +
E
E
c

p
1
p
2
, (as in Zirakashvili et al.
2007), where E
c
is the cut-off energy and K
e
is the nor-
malization constant.
Since the distributions with a cut-off (1 and 2) fail to
reproduce simultaneously both the radio and γ-ray spec-
trum, we refer to the distribution (3). The best fit to
the gamma-ray data is obtained with E
c
= 1 GeV, and
indices p
1
= 0 and p
2
= 3.3 above and below E
c
, re-
spectively. Synchrotron emission originating from this
distribution can be evaluated for different values of the
average magnetic field. Figure 3 shows the case of the
most “favorable” leptonic-only model characterized by
B = 20 µG (other cases turn out to be even less fa-
vorable). We find that, at high frequencies, the cal-
4 Giuliani et al.
Fig. 2.— Combined AGILE (red) and Fermi/LAT (green) spec-
tral energy distribution (SED) for SNR W44. AGILE points are
in the range 50 MeV- 10 GeV divided into six energy intervals.
Fermi/LAT data span the energy range 0.2 - 30 GeV (from Abdo
et al., 2010).
culated synchrotron spectrum is in strong disagreement
(factor larger than 4) with the radio emission produced
co-spatially to the gamma-ray one (Castelletti et al. 2007,
and Figure 1). Furthermore, the inferred average density,
n= 300 cm
3
, is too large (by a factor of three) com-
pared with the circumstellar medium constraints (Reach
et al. 2005); a lower value for n would be incompatible
with the gamma-ray spectrum fitting. Similar or even
stronger contradictions with the multiwavelength data
apply to other leptonic-only models that we systemati-
cally explored for a large variety of parameters.
In the case of inverse Compton dominated models, two
sources of soft photons are available: the cosmic back-
ground radiation (CBR) and the interstellar radiation
field (ISRF). In the first case, a second peak in the
gamma-ray spectrum is unavoidably expected with a
peak energy E
max
1 TeV, in contradiction with the
upper-limits obtained from TeV Cherenkov telescopes.
In the case of interaction with ISRF, instead, the calcu-
lated synchrotron peak is not compatible with the radio
continuum data for any reasonable value of the magnetic
field in the SNR shell. We can then reliably exclude
leptonic-only models of emission for SNR W44.
4.3. Hadron-dominated models of emission
It is interesting to determine the main physical pa-
rameters of an emission model dominated by hadrons in
the gamma-ray energy range. The gamma-ray emission
was derived assuming that protons interact with the nu-
clei of the ambient medium through pp interactions and
then radiate through π
0
decay (Kelner et al. 2006). In
order to fit the gamma-ray spectrum we tested, as pro-
ton energy distribution, a power law, a broken power
law and a power law with cut-offs. The distribution pro-
viding the best-fit of the gamma-ray spectrum turned
out to be a power-law with a relatively steep spectral
index and a low-energy cut-off, F
p
(E) E
p
e
E
c
/E
,
with p = 3.0 ± 0.1 and E
c
= 6 ± 1 GeV. The inferred
total energy of hadrons producing the observed gamma-
ray spectrum is E
h
= 10
49
10
50
erg (depending on the
average magnetic field and local densities, see Table 1)
corresponding to a fraction of the total SNR shell kinetic
energy 0.01 0.1. This value is a lower limit on
the total energy of hadrons accelerated by W44 during
its entire life, because the most energetic hadrons, likely
accelerated during the early epochs of the SNR life, can
escape from the acceleration site on timescales shorter
than the age of this SNR (Berezinskii et al. 1990; Gabici
et al. 2009). Electrons produce the radio continuum and
shell-like features by synchrotron emission in an aver-
age magnetic field in the range B 10 100 µG. They
also produce Bremsstrahlung and inverse Compton com-
ponents that are subdominant in the gamma-ray energy
range for a standard value of the electron/proton number
ratio χ 0.01. Our best “hadronic model” is shown in
Figure 4. We note that the spectral index found in this
analysis is quite steep in comparison with the expecta-
tions for the spectrum of particles accelerated in SNRs
(e.g. Reynolds 2008) and that the cosmic-ray spectrum
can be altered (softened) by the interaction with sites
where the target gas is confined, for example by molecu-
lar clouds (see e.g. Gabici et al. 2010; Ohira et al. 2011).
A comprehensive theoretical interpretation of this fact is
beyond the scope of this work and will be addressed in a
forthcoming paper.
Fig. 3.— Most favorable of the leptonic-only models, character-
ized by B =20 µG and n= 300 cm
3
. The green curves show the
electron synchrotron (dashed curve), Bremsstrahlung (solid curve),
and IC (dotted curve) contributions. This model is in contradic-
tion with the radio data, and requires an average value of the gas
number density that is too large compared with the value deduced
by radio and optical observations (n 100 cm
3
).
5. CONCLUSIONS
W44 turns out to be an extremely interesting SNR
whose environment and shocked material configuration
favor a detailed testing of hadronic versus leptonic emis-
sion. We find that models of accelerated protons/ions
interacting with nearby dense gas successfully explain
all the observed multifrequency properties of W44. The
total number of accelerated hadrons in W44 is consistent
with an efficiency of a few percent in the conversion of
SNR kinetic energy into cosmic-ray energy.
Within the hadronic scenario, the broad-band spec-
trum is successfully modeled by the gamma-ray emission
resulting from the decay of neutral pions produced by ac-
celerated protons/ions with a spectral index near 3 and a
5
Fig. 4.— Theoretical modeling of the broad-band spectrum of
SNR W44 superimposed with the radio (data points in red color)
and gamma-ray data of Figure 2 (in blue color). TeV upper limits
are also shown.. This is an hadronic model characterized by B
=70 µG and n =100 cm
3
. The yellow curve shows the neutral
pion emission from the accelerated proton distribution discussed
in the text. The green curves show the electron contribution by
synchrotron (dashed curve), Bremsstrahlung (solid curve), and IC
(dotted curve) emissions. The red curve shows the total gamma-
ray emission.
cut-off energy E
c
6 GeV (Figures 2 and 4). Electrons
(a few percent in number) contribute to the synchrotron
radio emission and to a weak Bremsstrahlung compo-
nent. It is interesting to note that the spectral cut-off
energy E
c
and the steep power-law index may indicate
either a “diffusion” effect of high-energy hadrons, or a
suppression of efficient particle acceleration in dense en-
vironments (Uchiyama et al. 2010; Malkov et al. 2011)
as expected in W44. Our results on W44 are comple-
mentary with those obtained in other SNRs, most no-
tably IC 443 (Tavani et al. 2010) and W28 (Giuliani et
al. 2010). Also in these SNRs the complex interaction
of cosmic-rays with their dense surroundings produces
gamma-ray spectra with high-energy cut-offs observed
in the range 1-10 GeV.
We acknowledge stimulating discussions with F. Aha-
ronian. The AGILE Mission is funded by the Italian
Space Agency (ASI) with scientific and programmatic
participation by the Italian Institute of Astrophysics
(INAF) and the Italian Institute of Nuclear Physics
(INFN). Investigation carried out with partial support by
the ASI grant no. I/042/10/0. G.D. and G.C. are mem-
bers of CIC-CONICET (Argentina) and are supported
by CONICET, ANPCyT and UBACYT grants.
REFERENCES
Abdo A. et al., 2009, ApJ Lett., 706, L1L6.
Abdo A. et al., 2010, Science, 327, 1103-1106
Abdo A. et al., 2010, The Astrophysical Journal, 712, 459-468
Abdo A. et al., 2010, The Astrophysical Journal, 718, 348-356
Aharonian F.A., Very high energy cosmic gamma radiation: a
crucial window on the extreme Universe, 2004, World Scientific
Publishing, Singapore.
Aharonian F.A. et al., 2006, Astron. Astrophys., 449, 223-242.
Berezhko E.G. & H.J.Voelk, 2007, Astrophys. Journal, 661,
L175-178.
Berezinskii V.S., Astrophysics of Cosmic Rays, 1990,
Amsterdam:North Holland
Blandford R. & D.Eichler, 1987, Physics Reports, 154, 1-75.
Bulgarelli A. et al., 2011, A&A, (submitted)
Butt Y., 2009, Nature, 460, 701-704.
Castelletti G. et al., 2007, A&A, 471, 537-549
Castro D. & Slane P., 2010, ApJ, 717, 372-378
Clark D.H. & Caswell J.L, 1976, Mon.Not.R.astr.Soc., 174,
267-305
Claussen M.J., Frail D.A. & Goss W.M., 1997, ApJ, 489, 143-159
Draine B.T. & McKee C.F., 1993, ARA&A, 31, 373-432
Drury L.O., 1983, Reports on Progress in Physics, 46, 973-1027.
Fermi E., 1949, Physical Review, 75, 1169-1174
Gabici S. et al., 2009, MNRAS, 396, 1629-1639
Gabici S. et al., 2010, SF2A, 313-317
Giacani E.B. et al., 1997, AJ, 113, 1379-1390.
Ginzburg V.L. & S.I.Syrovatskii, 1964, The Origin of
Cosmic-rays, Pergamon Press.
Giuliani A. et al., 2004, Mem. SaIt, 5, 135-138
Giuliani A. et al., 2006, NIM A, 568, 692-699
Giuliani A.et al., 2010, A&A, 516, L11-L14
Giuliani A. et al., 2010, ApJL, 708, 84-88
Hendrick S.P. et al., 2001, ApJ, 559, 903-908
Hoffman I.M. et al., 2005, ApJ,627, 803-812
Kelner, S. R., Aharonian, F. A., & Bugayov, V. V. 2006,
Phys. Rev. D, 74, 034018
Malkov M.A., Diamond P.H. & Sagdeev R.Z., 2011, Nature
Communication, 2, 194-201
Mattox J. R. et al., 1993, ApJ, 410, 609-614.
Ohira Y. et al., 2011, 410, 1577-1582
Petre R. et al., 2002, 579, 404-410
Ptuskin V.S. & Zirakashvili V.N., 2005, Astron. Astrophys., 429,
755-765.
Reach W.T. et al., 2005, AJ, 618, 297-320
Reynolds S.P., 2008, Ann. Rev. Astron. Astrophys., 46, 89-126.
Rho J. et al., 1994, AJ, 430, 757-773
Seta M. et al., 2004, AJ, 127, 1098-1116
Tavani M. et al., 2008, NIM A, 588, 5262.
Tavani M. et al., 2009, Astron. Astrophys., 502, 995-1013.
Tavani M. et al., 2010, ApJ, 710, L151-L155
Torres D.F., Romero G.E., Dame T.M., Combi J.A. & Butt
Y.M., 2003, Physics Reports 382, 303-380.
Uchiyama Y., Blandford R.D., Funk S., Tajima H. & Tanaka T.,
2010, ApJ, 723, L122-L126
Vercellone S. et al., ApJ, 676, L13-L16
Vercellone S.et al., ApJ, 690, 1018-1030
Vittorini V. et al., ApJ, 732, L22-L27
Watson M.G. et al., 1983, IAUS, 101, 273W
Wolszczan A. et al., 1991, AJ, 372, L99-L102
Wootten H.A., 1977, ApJ,216, 440W
Zirakashvili V.N. & Aharonian F.A., 2007, A&A, 465, 695-702
6 Giuliani et al.
Table 1: Electronic component of hadronic and ”ad hoc” leptonic models
Model < n > < B > E
h
E
e
p
1
p
2
Comments
(cm
3
) (µG) (erg) (erg)
good agreement with data
20 20 2.1 × 10
50
2.0 × 10
49
1.7 3.0 n small
Hadronic E
h
large
100 70 3.3 × 10
49
2.8 × 10
48
1.7 3.0 good agreement with all data
incompatible with the spectrum
300 20 - 7.3 × 10
48
0 3.3 too large medium density
too large E
e
Leptonic incompatible with the radio spectrum: wrong
slope, low-frequency excess
too large medium density
5000 200 - 4.4 × 10
47
0 3.3 too large average B
electron cooling (Bremsstrahlung) timescale too
short with respect to the SNR age
    • "The Chandra X-ray Observatory (CXO) has nearly matched the resolution of radio maps of SuperNova Remnants (SNR) and exhibited electron synchrotron emission from the bounding shock fronts ( that high Mach number shocks stretch and amplify magnetic field as well as accelerate cosmic ray electrons with energy up to ∼ 100 TeV [16]. While, it was hard to doubt that this was accompanied by proton acceleration , direct evidence has only been presented recently by Fermi and AGILE observations which exhibit the predicted " pion bump " in the γ-ray observations in middleaged supernova remnants expanding into dense molecular gas [17, 18] (Figure 4). No less dramatic have been the observations by Atmospheric Cerenkov Telescopes (ACTs) of TeV γ-rays which show evidence for efficient acceleration beyond ∼ 100 TeV (e.g., [19, 20]). "
    [Show abstract] [Hide abstract] ABSTRACT: Physicists have pondered the origin of cosmic rays for over a hundred years. However the last few years have seen an upsurge in the observation, progress in the theory and a genuine increase in the importance attached to the topic due to its intimate connection to the indirect detection of evidence for dark matter. The intent of this talk is to set the stage for the meeting by reviewing some of the basic features of the entire cosmic ray spectrum from GeV to ZeV energy and some of the models that have been developed. The connection will also be made to recent developments in understanding general astrophysical particle acceleration in pulsar wind nebulae, relativistic jets and gamma ray bursts. The prospects for future discoveries, which may elucidate the origin of cosmic rays, are bright.
    Article · Sep 2014
  • [Show abstract] [Hide abstract] ABSTRACT: Recently, the gamma‐ray telescopes AGILE and Fermi observed several middle‐aged supernova remnants (SNRs) interacting with molecular clouds. A plausible emission mechanism of the gamma‐rays is the decay of neutral pions produced by cosmic ray (CR) nuclei (hadronic processes). However, observations do not rule out contributions from bremsstrahlung emission due to CR electrons. TeV gamma‐ray telescopes also observed many SNRs and discovered many unidentified sources. It is still unclear whether the TeV gamma‐ray emission is produced via leptonic processes or hadronic processes. In this Letter, we propose that annihilation emission of secondary positrons produced by CR nuclei is a diagnostic tool of the hadronic processes. We investigate MeV emissions from secondary positrons and electrons produced by CR protons in molecular clouds. The annihilation emission of the secondary positrons from SNRs can be robustly estimated from the observed gamma‐ray flux. The expected flux of the annihilation line from SNRs observed by AGILE and Fermi is sufficient for the future Advanced Compton Telescope to detect. Moreover, synchrotron emission from secondary positrons and electrons and bremsstrahlung emission from CR protons can be also observed by the future X‐ray telescope NuSTAR and Astro‐H.
    Full-text · Article · Mar 2011
  • [Show abstract] [Hide abstract] ABSTRACT: Recent observations reveal that spectral breaks at ∼GeV are commonly present in Galactic γ-ray supernova remnants (SNRs) interacting with molecular clouds and that most of them have a spectral (E2dF/dE) ‘platform’ extending from the break to lower energies. In Paper I, we developed an accumulative diffusion model by considering an accumulation of the diffusive protons escaping from the shock front throughout the history of the SNR expansion. In this paper, we improve the model by incorporating the finite volume of molecular clouds, demonstrate the model dependence on particle diffusion parameters and cloud size, and apply it to nine interacting SNRs (W28, W41, W44, W49B, W51C, Cygnus Loop, IC443, CTB 37A and G349.7+0.2). This refined model naturally explains the GeV spectral breaks and, in particular, the ‘platforms’, together with available TeV data. We find that the index of the diffusion coefficient δ is in the range 0.5–0.7, similar to the Galactic averaged value, and the diffusion coefficient for cosmic rays around the SNRs is essentially two orders of magnitude lower than the Galactic average (χ∼ 0.01), which is a good indication for the suppression of cosmic-ray diffusion near SNRs.
    Full-text · Article · Aug 2011
Show more