The INTEGRAL view of 511 kev annihilation line in our Galaxy

The INTEGRAL view of 511 kev annihilation line in our Galaxy G. De Cesare, IASF-INAF, Rome giovanni.decesare@iasf-roma.inaf.it

Outline I. First detection of positrons in cosmic rays II. Galactic positrons as revealed by 511 kev annihilation line III. A possible 511 kev emission from the microquasar 1E 1740.7-2942 in 1990 IV. The INTEGRAL gamma-ray observatory V. SPI key results on the diffuse Galactic 511 kev emission VI. IBIS search for Galactic 511 kev point sources VII. Discussion VIII. Conclusion

First detection of a positron in cosmic rays The positron is the antiparticle of the electron. It has a positive electric charge +e (e = 1.602177 10-19 C), a spin 1/2 and the same mass as the electron (m e = 511 kev/c 2 ); Its existence was postulated by Paul Dirac in 1928 as a consequence of the quantum theory of the electron; The first experimental evidence of positrons dates 1932, when ground cosmic ray positrons were discovered by Carl D. Anderson 6 mm lead plate positron track The first positron track identified in a cloud chamber

Galactic positrons as revealed by 511 kev annihilation line First detection The 511 kev em issio n f ro m p o sitro ns annihilation from the direction of the Galactic center was first discovered in the early 1970s, by balloon-born detectors with low energy resolution (Johnson et al. 1972) The line was clearly identified a few years later with high energy resolution Ge detectors (Leventhal et. Al., 1978)

Galactic positrons as revealed by 511 kev annihilation line The energy distribution Annihilation of positron can take place by direct annihilation in two opposite 511 kev photons or by formation of a prositronium, i.e. a bound state made by an electron and a positron. The positronium can be formed with two total spin (S) state: S=0 para-positronium (pps) -> two 511 kev photons S=1 ortho-positronium (ops) -> three photon with a continuum of energies up to 511 kev By measuring the intensities of the 511 kev line and of the Ps continuum one can derive the fraction of positrons that annihililate by positronium formation: I 3γ 3 4 3f P s I 2γ 2(1 f Ps )+ 1 4 2f P s Leventhal 1978 f Ps = 8I 3γ/I 2γ 9+6I 3γ /I 2γ

A possible 511 kev emission from microquasar 1E 1740.7-2942 in 1990 The 1 E 1740.7 microquasars has been discovered by Einstein X-ray telescope (Herz and Grindlay, 1984). Its position is about 1 degree from the Galactic center. It is reported in the INTEGRAL/ IBIS in hard X catalogue with a flux of (29.8 ± 0.1) mcrab (A.J. Bird, 2007) During the 1990 October 13 the SIGMA telescope, aboard the GRANAT space observatory, discovered that this source was exhibiting a remarkable feature in its emission spectrum (See e.g. Bouchet et al. 1991) This transient feature appeared during a 13 hours obser vation and then possibly in two further occasions but at a less significant level. The line is transient (13 hours) and very broad (hudred kev) The 511 kev feature of 1E 1740.7-2942 was not confirmed by OSSE measures (Jung et al. 1995).

The INTEGRAL gamma-ray observatory Scientific instruments on-board INTEGRAL, the second ESA gamma-ray mission (the first being COS-B launched in 1975) was launched by PROTON from Baikonur in Kazakistan on 17th October 2002. INTEGRAL s scientific goal are met by fine spectroscopy w ith imaging capabilities and accurate positioning of the celestial sources. SPI and IBIS, the two gamma ray instrument aboard the INTEGRAL satellite, are optimized for spectroscopy and imaging, respectively. Both instrument are based on coded mask imaging.

The INTEGRAL gamma-ray observatory SPI and IBIS capabilities The spectrometer SPI is based on 19 high energy resolution Germanium detection units. The imager IBIS position sensitive detector is based on two detection plane (about 20 000 detection units): ISGRI (128 x 128 = 16384 Detection Units) PICsIT (64 x 64 = 4096 CsI Detection Units) SPI detectors SPI IBIS Energy range 20 kev - 8 MeV 15 kev - 10 MeV Field of 16 o 9 o x 9 o (fully View (corner to coded) corner) 19 o x 19 o (partially coded, 50 %) Energy resolution 2.33 kev @ 1.33 MeV ISGRI: 8 % at 100 kev IBIS PICsIT: 10 % at 1MeV Angular resolution ~ 2.0 o 12

The INTEGRAL gamma-ray observatory Coded mask imaging A pinhole camera concept It not easy to focalize gamma ray photons, but imaging is also possible without any lens or mirror. The coded mask imaging is an extension (with larger aperture) of the pinhole camera concept. In a coded mask system it is possible to obtain for a given point source a simultaneous measure of the source and background count rate. A coded aperture camera working principle: a 3D scheme in the left panel, a cut view in the right panel. The white mask elements are transparent to gamma rays, while the black ones are opaque. The recorded count rate in each pixel of the detection plane (shadowgram) is the summation of contributions from each source flux modulated by the mask. IBIS/ISGRI map of the Galactic central region in the 20-40 kev energy range Pro: large field of view Con: high background

SPI key results on the diffuse Galactic 511 kev emission A summary Intensity: The total rate of positrons annihilation observed observed in gamma-rays is about 2 x 10 43 s -1. Most of it comes from the bulge. Morphology: The bulge/disk ratio of e+ annihilation is B/D ~ 1.4 About half of the disk emission can be explained by the observed radioactivity of 26 Al (1800 kev line) Variability: There is not evidence of 511 kev flux variability. Spectroscopy: The emission is made by a narrow (~ kev) line and a 3-gamma ortho-positronium continuum. The energy spectrum can be explained by positronium decay (f Ps = (100 ± 2) %, Churazov et al. 2011) Initial energy: The positrons injection energy must be less then ~ 3 MeV (otherwise the emission from inflight annihilation would exceed the observed gamma ray flux measured by COMPTEL)

SPI key results on the diffuse Galactic 511 kev emission A possible correlation with the LMXBs population Some evidence of an asymmetry of the 511 kev emission along the Galactic longitude, possibly correlated with the spacial distribution of the hard X (E > 20 kev) LMXBs detected by the imager IBIS aboard INTEGRAL, has been reported The SPI sky map in the 511 kev electron-positron annihilation line and the sky distribution of the hard X (E > 20 kev) LMXBs detected by IBIS (Weidenspointner et al., 2008) But a different analysis of the same SPI data do not confirm this asymmetry (Bouchet et al., 2008, Churazov et al., 2011)...

SPI key results on the diffuse Galactic 511 kev emission Flux profile and spectrum SPI flux profile along the Galactic plane in 3 energy band (511 kev band in the center), obtained with a simple scan ( light bucket imaging ). The gray bars show the positions of a series of scans perpendicular tio the Galactic plane. (Churazov et. al 2011). Broad 30-1000 kev spectrum of the Galactic Bulge emission measured by SPI (Churazov et. al 2011).

SPI key results on the diffuse Galactic 511 kev emission Possible astrophysical sources Basically, any model has to explain how generate 2 x 10 43 s -1 positrons with an initial energy of the order of MeV and why they are concentrated in the Galactic bulge. I mention here some possible solutions: Radioactivity from stellar nucleosynthesis: positrons can be emitted by the β + decay of instable nucley, that are produced either hydrostatically in massive stars or explosively in novae or SN explosions. Astrophysically important positron emitting radioactivities X-ray binaries and microquasars: e+e- pairs can be created either in the hot inner accretion disk, in the X- ray corona surrounding the disk, or at the base of the jet. If jets consist of e+ e- plasma or e- p plasma, e + e- pairs can be also created by the jet interaction with the ISM or with the companion stars. The supermassive black hole in our Galaxy (Sgr A*): this source was proposed in the 1980s on the basis of the variability found in the HEAO-3 data. This variability was not confirmed by the subsequent measures. Therefore this model requires that (1) the positrons have time to file the bulge before annihilate and (2) that the Sgr A* activity was much higher in the past then now. Dark matter: Positrons from dark matter decays have been discussed as a possibility, mainly because a rather spheroidal, symmetric, bulge-centered annihilation emission morphology.

IBIS search for Galactic 511 kev point sources The data sample and analysis With our sample, we reach a 10 Ms exposure in the Galactic center region. The data set consists of all the IBIS data accumulated until April 2007. In addition, our sample includes all the Core program data until April 2008. All the data that we have reduced correspond to 39413 IBIS pointings each lasting about 2000 s., i.e. a total observing time of about 80 Ms. The IBIS pointings are also called Science Windows (ScWs). The IBIS data consists in a large set of small Data Unit, named Science Windows (ScWs). Each ScW, associated to one IBIS pointing, corresponds to one observation of ~2000 s. A sky map is obtained by a mosaic of n ScWs. The exposure map for the IBIS all sky 511 kev data analysis. The data set consists in about 5 year of observations. The exposure obtained in the Galactic bulge (yellow and white colors) is 10 Ms (De Cesare, 2011) The data reduction is CPU time consuming: to increase sensitivity a large sample is needed. Our data sample consists in about 5 years of data, which corresponds to ~ 40 000 ScWs, each one has some cost in term of computing time. The data analysis at 511 kev requires calibration: we need to evaluate the energy resolution, the background count rate, and the effective area.

IBIS search for Galactic 511 kev point sources Physical Monte Carlo simulation This tool is used to estimate the IBIS detector effective area and sensitivity at 511 kev The flowchart of IBIS Monte Carlo. The code is based on the CERN GEANT library to reproduce the interaction of the gamma-rays with IBIS detectors and passive materials. An IBIS Monte Carlo simulation with a 60 kev mono-energetic beam, simulating an on-axis point source. As expected, the ISGRI detector co u nt m a p ( le ft p a n e l), t h e s o - cal le d shadowgram, reproduces the IBIS mask. After decoding the detector map, the point source is shown in the right panel.

IBIS search for Galactic 511 kev point sources Instrumental background long term monitoring and detector effective area The 511 kev line sensitivity basically depends on the background count rate and on the effective area. ISGRI background count rate vs revolution Solar activity Simulating an on-axis source with the calibrated Monte Carlo, we estimate the following peak effective area at 511 kev: A I eff = (27.1 ± 0.1) cm 2

IBIS search for Galactic 511 kev point sources The 511 kev line sensitivity: a first estimation The sensitivity is taken to mean as the minimum flux that a source can have with a appreciable chance to be detected with a given exposure. Thus I define the sensitivity as the flux detected at 2 sigma of significance level. We measure: Count = Source + Background Then: Source = Count - Background Assuming B >> S, the source signal to noise is given by (Goldwurm et al., 2003): Signal Noise = Reconstructed Source Count T otal Count = S T (S + B)T S T B Using the background measure and the effective area (MC estimated), for an exposure T = 10 Ms we estimate for IBIS/ISGRI a 2 sigma sensitivity of the order of 10-4 ph cm -2 s -1 The instrument gamma-ray background is mainly due to the cosmic interaction with the the satellite and the upper atmosphere. At the beginning of the INTEGRAL observation, we measure for ISGRI a background count rate of: B I 511 kev = (7.85 ± 0.06)counts/s rev. 50 If one assumes a source with a 511 kev flux F = 10-3 ph cm -2 s -1 (the total flux measure by SPI) the source count is S = F x A eff = 0.027 counts/s. The source to background ratio is S/B = 0.34 %.

IBIS search for Galactic 511 kev point sources The 511 kev line sensitivity: Monte Carlo simulation rev. 50 rev. 600 A simulation of a 511 kev point source as seen by IBIS/ISGRI during a pointing in the early (rev. 50), left panel, and late (rev. 600), right panel, orbits of the INTEGRAL satellite. The signal level of the detected 511 kev source goes to 10 σ in the first image to 7 σ in the second one. These images have been obtained adding to the real data the data generated by the IBIS Monte Carlo (one million 511 kev photons as input) and then processing the result with the standard analysis software until the image level. The sensitivity with an exposure of 10 Ms estimated by MC simulation

IBIS search for Galactic 511 kev point sources The 511 kev line sensitivity: Crab detection 40-100 kev 40-100 kev 431-471 kev 491-531 kev The Crab nebula detected by IBIS with an exposure of 3 Ms The Crab nebula as detected by IBIS/ ISGRI in the soft gamma-ray domain with a 3 Ms of exposure. Left and right top panel: the ISGRI 40-100 kev shadowgram accumulated during one ScW and the decoded sky image. Left bottom panel: sky mosaic in 431-471 kev (colorbar: 2-8 σ). Right bottom panel: sky mosaic in 491-531 kev (colorbar: 3-4 σ). The IBIS 511 kev sensitivity as estimated with the Crab data

IBIS search for Galactic 511 kev point sources The 511 kev line sensitivity map Processing about 5 year of IBIS data I do not find in a 511 kev full sky map any evidence of point source. The best flux limit constraint in this map is equal to 1.6 10 4 ph cm 2 s 1 for 1 E 1740.7-2942 and GRS 1758-2258, the two IBIS microquasars located near the center of the Galaxy. For the sources with lower exposure, as for example Sco X-1 that has a flux upper limit of 3.7 10 4 ph cm 2 s 1, our observations gives a less tight constraint on the 511 kev emission.

IBIS search for Galactic 511 kev point sources LMXBs 511 kev upper limits De Cesare 2011 Sky distribution of the the LMXBs detected by IBIS above 20 kev. The red point corresponds to the LMXBs with b < 10 degrees, while for the blue ones b > 10 degrees. A possible correlation between this source distribution and the SPI 511 kev m a p h a s b e e n p o i nt e d o u t by Weidenspointner et al., 2008

IBIS search for Galactic 511 kev point sources Microquasars 511 kev upper limits The possibility to detect a 511 kev radiation due to the jet positrons hitting the atmosphere of the companion star, in the case of misaligned microquasars, is discussed by Guessoum et al. 2006. De Cesare 2011

IBIS search for Galactic 511 kev point sources The Galactic Center monitoring A possible 511 kev emission from the Galactic Center has been also monitored during the INTEGRAL visibility periods at Ms timescale, without finding any signal. The flux upper limit are reported here: De Cesare 2011 In data analysis a sky map has been generated for each INTEGRAL revolution, therefore also monitoring the IBIS 511 kev sky in day-timescale.

Discussion Low Mass X-ray Binaries The observed distribution of LMXRBs in the Galaxy is strongly peaked towards the central regions, similar to that of the 511 kev emission In terms of total energy the LMXBs population could explain the 511 kev flux f = n 4πd 2 d = 10 kpc n =2 10 43 e + s 1 (taking into account of the positron fraction) f SP I (511 kev ) = 10 3 ph cm 2 s E e+ = 511 kev = 10 6 erg L 511 kev =2 10 37 erg/s This value is 50 times lower than the total luminosity of the LMXBs (Grimm et al., 2002) L tot (LMXB) = 10 39 erg/s

Discussion Positronium continuum 3-gamma emission The proof of positrons annihilation in compact object could be found by detecting the gamma rays produced in the 3-gamma positronium decay. The signal depends on the positronium fraction, being maximum when all the positrons annihilate by the positromium formation. I 3γ 3 4 3f P s I 2γ 2(1 f Ps )+ 1 4 2f P s f Ps =1 f Ps = 8I 3γ/I 2γ 9+6I 3γ /I 2γ I 3γ I 2γ = 9 2 Depending on the positronium fraction the ratio between continuum and line goes from 0 to 9/2 We note also that the gamma ray sensitivity is better at lower energies...therefore one possibility is to analyze all the INTEGRAL archival data looking for some positronium component in a galactic compact object.

Conclusions By reducing 5 year of the IBIS imager data, I do not find any evidence of 511 kev lines feature from BH sources. The 2 sigma flux upper in the Galactic center region, with an exposure of 10 Ms, is estimated at a level of 1.6 x 10-4 ph cm -2 s -1. In principle it will be possible to obtain a much lower flux limits with the application of the gamma-ray lenses in astronomy. In particular analyzing data at day-mounts-year timescales, IBIS does not not detect any persistent or transient 511 kev emission from 1E 1740.7-2942 (from this source a transient emission with SIGMA was reported in 1990). The Galactic 511 kev emission measured by SPI spectrometer (narrow line, large positronium fraction, diffuse emission) is generally explained by positrons propagation in the interstellar medium before they annihilate away from the birth place. There is some evidence of an asymmetry of the 511 kev emission along the Galactic longitude, possibly correlated with the spacial distribution of the hard X (E > 20 kev) LMXBs detected by IBIS. I have shown in this presentation that also in terms of energy LMXRs are possible candidates. In principle, it is possible to search for the 3 gamma continuum in Galactic BH. We note that the intensity of this component depends on the positronium fraction, being 9/2 of the 511 kev line intensity if all positrons annihilate by formation of positronium atoms.