Multifrequency Behavior of Microquasars in the GeV--TeV era: A review
aa r X i v : . [ a s t r o - ph . H E ] J un Mem. S.A.It. Vol. 75, 282 c (cid:13) SAIt 2008
Memorie della
Multifrequency Behavior of Microquasars in theGeV–TeV era: A review
V. Bosch-Ramon Dublin Institute for Advanced Studies, 31 Fitzwilliam Place, Dublin 2, Ireland e-mail: [email protected]
Abstract.
Microquasars are X-ray binaries that present non-thermal radio jets. E ffi cientparticle acceleration can take place in di ff erent regions of the jets of microquasars. Theaccelerated particles can emit gamma-rays via leptonic or hadronic processes, with a com-plex spectral and temporal behavior. The jet termination region can be also an e ffi cientnon-thermal emitter, as well as, in high-mass microquasars, the region of the binary systemoutside the jet. In this work, I briefly describe the physics behind the non-thermal emissionobserved in microquasars at di ff erent scales, focusing in the GeV and TeV bands. Key words.
X-ray: binaries – Gamma-rays: theory – Radiation mechanism: non-thermal
1. Introduction
Microquasars are X-ray binaries with non-thermal jets (e.g. Mirabel & Rodr´ıguez 1999;Rib´o 2005), being called high-mass micro-quasars when hosting a massive star, andlow-mass microquasars otherwise. The energypowering the non-thermal emission in micro-quasars can be either of accretion or black-hole rotation origin. The magnetic and kineticpower is channelled through a jet launchedfrom the inner regions of the accretion disk(e.g. Blandford & Znajek 1977; Blandford &Payne 1982), and part of this power is even-tually converted into relativistic particles andradiation.For several decades, microquasars wereconsidered strong candidates to gamma-raysources (e.g. Chadwick et al. 1985; seealso Chardin & Gerbier 1989; Levinson &Blandford 1996; Paredes et al. 2000), but they
Send o ff print requests to : V. Bosch-Ramon have not become fully recognized as power-ful gamma-ray emitters until recent years, af-ter the most recent generation of ground-basedCherenkov (HESS, MAGIC, VERITAS) andsatellite-borne instruments ( Fermi , AGILE ) ar-rived. The most relevant cases are the micro-quasars Cygnus X-1 and Cygnus X-3 (Albertet al. 2007; Sabatini et al. 2010; Tavani et al.2009; Abdo et al. 2009a; Sabatini 2011). Othersources, like for instance SS 433, Scorpius X-1 or GRS 1915 − + +
303 (e.g. Albert et al. 2006;Abdo et al. 2009b; Pittori et al. 2009), LS 5039 This source has been detected in GeV and TeVenergies with significances close, but slightly below,5 σ , and thus these detections are still to be firmlyestablished.osch-Ramon: Microquasars in the GeV–TeV era 283 (e.g. Aharonian et al. 2005; Abdo et al. 2009c;Pittori et al. 2009), HESS J0632 +
057 (e.g.Hinton et al. 2009; Falcone et al. 2011; Mold´onet al. 2011), and 1FGL J1018.6 − + + ffi ciently channel accretion or black-hole rotational energy into radiation. In addi-tion, together with this high e ffi ciency, the tem-poral characteristics of the detected radiationmay favor leptonic models, although hadronicmechanisms cannot be discarded. Also, theextreme conditions under which gamma-raysare produced can put restrictions in the emit-ter structure. Morphological studies can bealso of help, since non-thermal processes cantake place not only at the binary scales, butalso far away (e.g. the jet termination region).Although the complexity of microquasar phe-nomenology can make the characterization ofthe ongoing processes di ffi cult, high qualitydata together with semi-analytical modelingcan provide sensible information on the non-thermal physics of the sources. Numerical cal-culations are also important, since they can in-form about the conditions of the backgroundplasma in which emission takes place.In this paper, we briefly review relevantaspects of the non-thermal emission in mi-croquasars. We will focus mainly in the GeVand TeV energy bands, for which photon pro-duction requires extreme conditions in thesesources. In Figure 1, a sketch of the micro-quasar scenario is presented.
2. Non-thermal emission inmicroquasars
Microquasar jets can produce non-thermalpopulations of relativistic particles via di ff u-sive shock acceleration or other mechanismsat di ff erent spatial scales (e.g. Rieger et al. 2007). These particles, electrons, protons oreven heavy nuclei, can interact with the back-ground matter, radiation and magnetic fieldsto produce non-thermal emission from radioto gamma-rays. In the GeV and TeV bands,the most e ffi cient process is inverse Compton(IC), but in systems with very high density re-gions, like SS 433, Cygnus X-3 and possiblyCygnus X-1, proton-proton interactions maybe also relevant. Also, in systems with verydense fields of energetic target photons, pho-tomeson production and even photodisintegra-tion of nuclei may be e ffi cient. The gamma-rayemission can take place at di ff erent scales, al-though certain regions su ff er from strong ab-sorption via pair creation (e.g. deep inside thesystem or at the jet base), and some others may(or may not) lack enough targets (e.g. the jetlargest scales). Below, we discuss farther thenon-thermal phenomena at di ff erent scales inhigh- and low-mass microquasars. For a gen-eral review on the e ffi ciency of leptonic andhadronic processes under typical microquasarconditions, see Bosch-Ramon & Khangulyan(2009) and references therein. Di ff erent emitting regions can be consideredwhen understanding the non-thermal emissionfrom microquasars. Jets are the best accelera-tion sites given the large amount of energy thatthey transport. Di ff erent forms of dissipationcan take place in them through shocks, veloc-ity gradients and turbulence (e.g. Rieger et al.2007), as well as magnetic reconnection (e.g.Zenitani & Hoshino 2001), which can lead togenerate non-thermal particle populations.The jet formation itself, interaction with anaccretion disc wind, or recollimation and in-ternal shocks can accelerate particles at the jetbase. The presence of non-thermal electrons inthe region can lead to the production of gammarays through IC with accretion photons, orwith photons produced by the same electronsvia synchrotron emission (e.g. Bosch-Ramonet al. 2006a). The base of the jet is possiblythe region in which hadronic processes maybe the most e ffi cient, given the high densityof matter and photons in there, and the hard-
84 Bosch-Ramon: Microquasars in the GeV–TeV era
Observere StarB
X−raysgamma−rays
ICsync γ∗ Wind e+ e−pp
Jet/wind interactions radio region
Accretion diskJet e ? Fig. 1.
Illustrative picture of the microquasar scenario (not to scale), in which relevant elementsand dynamical and radiative processes in di ff erent regions are shown (background image fromESA, NASA, and F´elix Mirabel).ness of the latter (e.g. Levinson & Waxman2011; Romero & Vila 2008). However, the lo-cal radiation fields could also strongly sup-press the GeV emission via gamma-ray ab-sorption and pair creation (see, e.g., Romero& Vila 2008; Cerutti et al. 2011). For lowambient magnetic fields, electromagnetic cas-cades can increase the e ff ective transparencyof the source to gamma-rays (Akharonian &Vardanian 1985). An example of a (leptonic)low-mass microquasar spectral energy distri-bution, with its high-energy radiation mainlycoming from the base of the jet, is shown inFig. 2.In high-mass microquasars, the strong ra-diation and mass-loss from the star can ren-der significant non-thermal radiation, in par-ticular at high energies, whereas radio maybe at least partially free-free absorbed. Theconsidered most e ffi cient high-energy chan-nel is typically IC with stellar photons (e.g.Bosch-Ramon et al. 2006b), interaction thatis anisotropic and has specific lightcurve andspectral features (e.g. Khangulyan et al. 2008).Anisotropic IC may be behind the orbital mod-ulation of the GeV lightcurve of Cygnus X-3 V. Bosch-Ramon et al.: On the spectrum of 1E 1740.7 − (RN) Table 2.
Specific parameter values for di erent models.Parameter [units] corona jetAccretion rate [ yr] 1 10 10Corona luminosity [erg s ] 3 10 We have neglected here those components that are not relevantto model the SED of this particular source. The accretion rate,10 yr , has been fixed assuming that the total X-rayluminosity (corona disk) is 0 05 the total accretion rate.The corresponding Eddington accretion rate for a 5 blackhole is 2 10 yr . The magnetic field at the base of thejet and the (one) jet total kinetic luminosity are about 4 10and 4 10 erg s , respectively. The expected fluxes are sev-eral 10 erg cm at 1 GeV, and close to 10 erg cmat 100 GeV. The source would have not been detected above100 MeV, since it is below the lower EGRET limit, and its con-tribution to the nearby source could not be significant. HESSupper limit is at the moment above the computed emission level.For a dominant synchrotron X-ray jet (see Fig. 1), we adoptmore extreme parameter values than those explored above andneglect the corona to try to reproduce the broadband emission.The synchrotron modeling of the X-rays requires a two-sidedjet with jet accr
2, energy dissipation e ciency of 70%,acceleration e ciency of 0.002 and an accretion rate taken tobe a 10% of the Eddington accretion rate quoted above (i.e.10 yr ). The magnetic field and jet total kinetic lu-minosity are 3 10 G and 3 10 erg s . Particularly for thiscase, the magnetic field has been increased to reach the observedX-ray fluxes. For the dissipation e ciency in this case, the pureradiative e ciency is very large, of about a 50%. Radio fluxesare exceeded by more than one order of magnitude, and an elec-tron power-law index of 1.7 is required. Instead of being the re-sult of synchrotron emission, the X-rays might be produced viaIC scattering. When attempting to reproduce the observed X-rayspectrum with an IC jet model with a weak corona, it is not pos-sible to reach the X-ray fluxes through IC of external photonsand or SSC emission because the jet power requirements are toohigh, and it is not possible to keep the particle energy low enoughas to make them radiate just up to a few hundreds of keV forany reasonable acceleration rate with any photon field. In ad-dition, for the SSC model, radio constraints are also violated.We conclude, in the context of our model and after exploring avast range of parameter values, that the corona X-ray dominatedemission reproduces better the observed broadband SED thanthe jet X-ray dominated emission in 1E 1740.7 2942.Finally, we have explored semi-quantitatively the radiationfrom the extended radio emitting jets. A magnetic field of 10all along that jet region and a jet carrying at least 1% of the ac-creted matter was required in our model to generate radio emis-sion up to 2–3 pc, i.e. the size of the radio lobes detected byMirabel et al. (1992) at 8 kpc. This magnetic field is similar towhat is typically found in molecular clouds (Crutcher 1999) andabove the equipartition value with jet matter in those regions. Werecall that 1E 1740.7 2942 could be located within a molecularcloud (Yan & Dalgarno 1997). The magnetic field decreases like 1 as far as cold matter energydensity goes down along the jet. −5 −3 −1 1 3 5 7 9 11 13log (energy photon [eV])2931333537 l og ( ε L ε [ e r g / s ]) cor. dominancejet dominanceVLA RXTEINTEGRAL HESSsync disk corona/sync COR IC SSC
EGRET 2obscured region EGRET 1star
Fig. 1.
Computed broad-band SED for 1E 1740.7 2942 for a dominantX-ray corona (thick solid line) and a dominant X-ray jet (thick long-dashed line) (see Table 2). Data points and observational upper-limitsare also shown. Note that radio constraints are violated for the dominantX-ray jet case. We note that dips at 100 MeV and few GeV are due tophoton-photon absorption in the corona and the disk photon fields re-spectively. At the radio band, the core luminosity at the adopted distanceis shown. The slope at this range is similar to that observed (not shownexplicitly; see Fender 2001, and references therein). At X-rays softgamma-rays,
RXTE and
INTEGRAL data are plotted (Del Santo et al.2005). At high-energy gamma-rays, two upper-limits are shown: onefrom the spectrum of the source 3EG J1744 3011 (EGRET 1), as abso-lute upper-limit (Hartman et al. 1999), and the other obtained from thesensitivity limits of EGRET (EGRET 2) in the region (Hartman et al.1999). The di use emission flux at 100 GeV observed by HESS inthe direction of the source is shown as an upper-limit (Aharonian et al.2006).
4. Discussion and summary
There is a wide debate in the astrophysical community aboutthe possible origin of the X-rays in microquasars since, at thepresent state of knowledge, the corona and the jet scenariosseem to be roughly consistent in some cases with observa-tions (Marko et al. 2005). In 1E 1740.7 2942, the hard X-rayemission is di cult to be explained if coming from a jet sinceit would imply an energetic e ciency in the jets significantlylarger than for the corona emission (see Sect. 3). Actually, in thecontext of our model, synchrotron emission from the jet seemsto be able to explain the low-hard state X-ray spectrum of thesource, but it exceeds largely the observed core radio fluxes.1E 1740.7 2942 appears to show related radio and X-ray vari-ability (Mirabel et al. 1993), although is radio underluminous inthe context of the radio-X-ray luminosity correlation found forblack hole candidate X-ray binaries and associated with the ac-cretion ejection activity (Gallo et al. 2003; Corbel et al. 2003;Fender et al. 2003). This might be due to a particularly radia-tively e cient corona. We remark that radio variations (Mirabelet al. 1993) could hardly explain a flux as large as that predictedwith the synchrotron jet model (there remains the possibility thatother jet models, with a much higher electron minimum energy,may be consistent with observations). All this points to the factthat, although accretion and jet phenomena are probably linked,the dominant component at X-rays can be either the jet or thedisk corona depending on the source.
Fig. 2.
Computed spectral energy distri-bution of the non-thermal emission from1E 1740.7 − ∼ GeV energies).For details, see Bosch-Ramon et al. (2006a). osch-Ramon: Microquasars in the GeV–TeV era 285 seen by
Fermi (Abdo et al. 2009a; Dubus et al.2010). Absorption of TeV emission in the stel-lar photon field is likely to be significant forcompact high-mass systems, like Cygnus X-3and Cygnus X-1. That may be the reason whythe former has not been detected in the TeVrange (see Aleksic et al. 2010 and referencestherein), and why the evidence of detection ofCygnus X-1 by MAGIC may imply an emitteroutside the binary system (see Bosch-Ramonet al. 2008a). As before, for low enough mag-netic fields (see Khangulyan et al. 2008), elec-tromagnetic cascades can increase the e ff ectivetransparency of these two sources (see, e.g.,Bednarek & Giovanelli 2007; Orellana et al.2007; Bednarek 2010). As discussed below, therole of pair creation cannot be neglected in thecontext of broadband non-thermal emission. Atthe binary scales, absorption of GeV photonsis not expected since this band is below thegamma-ray energy threshold for pair creation,around ∼ −
100 GeV for stellar photonspeaking in the UV. Proton-proton, photome-son production and photodisintegration havealso been proposed as possible mechanismsof gamma-ray emission at these scales (e.g.Romero et al. 2003; Aharonian et al. 2006;Bednarek 2005). An example of a (leptonic)spectral energy distribution of a high-mass mi-croquasar is shown in Fig. 3.The interaction of the jets with the stellarwind cannot be neglected in microquasars witha massive companion. The impact of the windon the jet triggers strong shocks, good can-didates for particle acceleration, jet bending,and potentially jet disruption (e.g. Perucho &Bosch-Ramon 2008; Perucho et al. 2010). Thisinteraction can generate high-energy emission(Perucho & Bosch-Ramon 2008), but the spe-cific properties can depend on the level of in-homogeneity of the stellar wind (e.g. Araudo etal. 2009, 2011; also Perucho & Bosch-Ramon,in preparation). Figure 4 shows the densitymap resulting from a 3-dimensional simulationof a microquasar jet interacting with the windof the companion.Far from the binary system, say at milliarc-second to second scales, the jet propagates un-a ff ected by significant external disturbances.However, there are di ff erent mechanisms that may lead to energy dissipation, particle heat-ing / acceleration and subsequent radiation, likevelocity gradients and Kelvin-Helmholzt in-stabilities in the jet walls. Shear accelerationhas been proposed for instance to explain ex-tended emission from large scale jets in mi-croquasars and AGNs (Rieger et al. 2007).All this could generate fresh relativistic parti-cles that could emit in radio by synchrotron.Very powerful ejections could also be brightenough to be detectable, from radio to gamma-rays, far away from the binary (e.g. Atoyan &Aharonian 1999).It is noteworthy that, unless there is not sig-nificant previous jet activity, the wall of a con-tinuous jet, or a transient ejection, are always toencounter diluted and hot jet material. This ma-terial was reprocessed in the jet reverse shock,where jet and ISM pressures balance, and wasswept backwards filling the so-called cocoon.Only the presence of a strong wind, eitherfrom the accretion disc or the star, can cleanthis material out up to a certain distance fromthe microquasar. However, the jet material willunavoidably end up embedded in the cocoonplasma before the reverse shock is reached.The pressure of the cocoon can trigger a recol-limation shock in the jet, which becomes colli-mated and su ff ers pinching. The jet fed cocoondrives a slow forward shock in the ISM, muchdenser and cooler than the jet. This complexdynamical behavior has associated the produc-tion of non-thermal emission, which possiblymay reach gamma-ray energies. An interest-ing situation arises when the microquasar hasa high-mass companion and the proper motionvelocity is & cm s − , in which case the jetcan be completely disrupted before reachingthe ISM, as illustrated in Fig. 5. Farther dis-cussion of jets interacting with the ISM can befound in Bordas et al. (2009), Bosch-Ramon etal. (2011), and references therein.Jets, or their termination region, are not theonly possible emitting sites in microquasars.The inner regions of the accretion structures(e.g. disc, corona / ADAF and the like) mayalso contain non-thermal populations of parti-cles (e.g. Bisnovatyi-Kogan & Blinnikov 1976;Pineault 1982; Spruit 1988; Gierlinski et al.1999; Romero et al. 2010). At the binary sys-
86 Bosch-Ramon: Microquasars in the GeV–TeV era
272 V. Bosch-Ramon et al.: Broad-band emission from microquasars
Concerning annihilation rates inside this cold matter dom-inated jet, for any reasonable set of parameter values the lu-minosity that could be emitted in form of an annihilation lineis too low to be distinguished from the continuum emission.Other models, like the one of Punsly et al. (2000), where a purepair plasma is assumed, could produce detectable annihilationlines.Observable predictions from considering pair creation phe-nomena in our model are presented and discussed briefly inSects. 5 and 6, although we remark that the creation of pairsinside the jet could lead to the appearance of bumps due to theaccumulation of particles at the energies of pair creation. Tointroduce such an e ect properly requires a better knowledgeof the particle injection function, which is beyond the scope ofthis work. Therefore, the high-energy gamma-ray band of thecomputed SEDs probably gives good enough flux estimates, al-though slopes could be slightly di erent as a result of all thesesubtle e ects.For those pairs that are created within the binary system,but outside the jet, the situation is di erent from that of pairscreated inside. Starting with a determinate number of relativis-tic particles in the jet, plus the given jet conditions, one canconsistently derive the SED of the produced radiation in thecompact object RF. Thus, the spectrum is known, and it allowsus to know precisely the number of absorbed photons and cre-ated pairs within the star, the disk and corona photon fields(for previous treatments of this, see Romero et al. 2002). Fromthe previous result, it is possible to roughly estimate the radia-tion that is generated by those pairs through IC interaction withexternal source photons. Although it is a rough estimate, it isfound to be in agreement with more detailed models of theseprocesses (Khangulyan & Aharonian 2005).
5. Application of the model
The di erent radiation components produced in the jet and thepredicted SEDs have been computed for the four specific sce-narios considered here. The e ects of pair creation phenomenadue to the external photon fields interacting with the producedgamma-ray photons are taken into account, and the secondaryradiation produced by the created pairs is estimated. The calcu-lations are performed at the periastron passage, when the com-pact object is in opposition to the observer and the interactionangle between star photons and jet leptons implies more lu-minosity for the star IC component (see Dermer et al. 1992),showing the importance of such an e ect. However, such an an-gle depends on the electron energy, which should be taken intoaccount in more detailed models of the IC interaction (e.g.,Khangulyan & Aharonian 2005).The broad-band SEDs for cases A and B are presented inFigs. 8 and 9 respectively. The strong e ects on the computedSED due to the presence of a massive star can be appreciated.The star IC component is very significant, partially because ofthe specific interaction angle between seed photons and lep-tons at phase 0, and also because the interaction with stellarphotons is more significant at large , when the number ofrelativistic particles is higher (see Fig. 3), than for other pho-ton fields. For , gamma-gamma opacity is very high at VHE. −5 −3 −1 1 3 5 7 9 11 13 log (photon energy [eV]) l og ( ε L ε [ e r g / s ]) predicted SEDstar/star ICsync./SSCcorona/cor. ICdisk/disk ICext. Bremsstr.int. Bremsstr.star disksync. corona Fig. 8.
Case A computed SED for the entire spectrum as it would beobserved. Attenuation of the jet photons due to absorption in the ex-ternal photon fields is taken into account, as well as the IC emission ofthe first generation of pairs created within them. Isotropic luminosityis assumed. The di erent IC, relativistic Bremsstrahlung, synchrotronand other seed photon fields are shown. For the several components,the production SED is shown. The corona photon field is also takeninto account, but its e ects on pair creation and subsequent emissionare overcome by the synchrotron emission. −5 −3 −1 1 3 5 7 9 11 13 log (photon energy [eV]) l og ( [ e r g / s ]) predicted SEDstar/star ICsync./SSCcorona/cor. ICdisk/disk ICext. Bremsstr.int. Bremsstr. star disk sync.corona Fig. 9.
The same as in Fig. 8 but for the case B. The small bumpspresent from beyond 100 MeV come from the IC radiation emittedby those leptons generated by pair creation in the disk and the stellarphoton field. These pair components are not made explicit in the plotfor clarity.
We recall that the disk and corona emission have been assumedto radiate just a few per cent of the accretion power. As accre-tion does not dissipate a significant fraction of the available en-ergy via either disk or corona radiation, the jet can carry moreenergy and matter for the same ejection velocity (and the as-sumptions put forward in Sect. 3.1 are valid). The accelerationciency has been assumed to be high.In Fig. 10, the broad-band SED of case C is shown. We haveincreased the disk and the corona emission, fixing the jet ve-locity. This implies a lighter jet than in the two previous cases.Also, we have modified the acceleration e ciency of the jet
Fig. 3.
Computed spectral energy distribution, from radio to very high energies, for a high-massmicroquasar (see Bosch-Ramon et al. (2006b)).
Fig. 4.
Density map for a high-mass microquasar jet interacting with the stellar wind, which iscoming from the top of the image (see Perucho et al. (2010)).tem scales, and in particular with high-masscompanions, the stellar wind is dense, carriesmagnetic field, and is embedded in a densephoton bath by the star. Therefore, for thosesystems with very e ffi cient particle accelera-tion in the jet, electrons and protons could dif-fuse out of it and radiate their energy in the en-vironment. Also, gamma-ray absorption due topair creation in the stellar photon field can in-ject electrons and positrons in the wind, alsoleading to broadband non-thermal emission,as shown for instance in Bosch-Ramon et al.(2008b). This emission may be actually behinda substantial fraction of the milliarcsecond ra-diation in a TeV emitting microquasar (Bosch-Ramon & Khangulyan 2011). An example ofthis is shown in Fig. 6, in which 5 GHz maps are presented for di ff erent orbital phases in aTeV emitting binary.
3. Final remarks
Microquasar can e ffi ciently accelerate parti-cles up to very high energies and producegamma-rays within and outside the binary sys-tem. It is however unclear currently why somesources emit gamma-rays and others do not.A key point may be the presence of a mas-sive star, which as discussed here can af-fect the jet significantly, with the formationof particle acceleration sites, and also o ff er-ing dense target photon and matter fields, suit-able for gamma-ray production at the binaryscales. Low-mass microquasars could in prin-ciple produce gamma-rays, and a reason for re- osch-Ramon: Microquasars in the GeV–TeV era 287 [mJy/beam] [mJy/beam][mJy/beam] [mJy/beam] Fig. 6.
Computed image, in the direction to the observer, of the 5 GHz radio emission fromsecondary pairs in a TeV binary for di ff erent orbital phases. Units are given in mJy per beam,being the beam size ∼ ff erencebetween high- and low-mass microquasars at these scales, which may be the case account-ing for their di ff erent environments. Acknowledgements.
I want to thank the organiz-ers for their kind invitation. The research lead-ing to these results has received funding fromthe European Union Seventh Framework Program(FP7 / References
Abdo, A. A., et al. 2009a, Science, 326, 1512Abdo, A. A., et al. 2009b, ApJ, 701, L123Abdo, A. A., et al. 2009c, ApJ, 706, L56Acero, F. et al. 2009, A&A, 508, 1135Albert, J. et al. 2006, Science, 312, 1771Albert, J., et al. 2007, ApJ, 665, L51Aleksic, J. 2010, ApJ, 721, 843Aleksic, J. et al. 2011, ApJL, in press [astro-ph /
88 Bosch-Ramon: Microquasars in the GeV–TeV era
Fig. 5.
Density map resulting from a 2-dimensional slab simulation, in which a jetpropagates in an environment characterizedby the microquasar motion in the ISM. Theshocked stellar wind, deflected by the micro-quasar proper motion, comes from the top andimpact the jet from a side (see Bosch-Ramonet al. 2011).Akharonian, F. A. & Vardanian, V. V. 1985,Ap&SS, 115, 31Aharonian, F. A. et al. 2005, Science, 309, 746Aharonian, F., Anchordoqui, L., Khangulyan,D., Montaruli, T. 2006, J. Phys. Conf. Ser.,39, 408 [astro-ph / / / / / osch-Ramon: Microquasars in the GeV–TeV era 289 Paredes, J.M., Mart´ı, J., Rib´o, M., Massi, M.2000 Science, 288, 2340Perucho, M. & Bosch-Ramon, V. 2008, A&A,482, 917Perucho, M., Bosch-Ramon, V., &Khangulyan, D. 2010, A&A, 512, L4Pineault, S. 1982, A&A, 109, 294Pittori, C., et al. 2009, A&A, 506, 1563Rib´o, M. in Future Directions in HighResolution Astronomy: The 10thAnniversary of the VLBA, (ASPC, 2005)340, 421 [astro-ph / ff y, P.2007, Ap&SS, 309, 119Romero, G. E., Torres, D. F., KaufmanBernad´o, M. M., Mirabel, I. F. 2003, A&A,410, L1Romero, G. E., Vila, G. S. 2008, A&A, 485,623Romero, G. E., Vieyro, F. L., Vila G. S. 2010,A&A, 519, 109Sabatini, S. et al. 2010, ApJ, 712, L10Sabatini, S., 2011, these proceedingsSaito, T. et al. 2009, Proc. 31st ICRC [astro-ph / DISCUSSIONWOLFGANG KUNDT’s Comment:
In yourthoughtful review on VHE radiation from mi-croquasars you also addressed the expectedVHE emission region. How realistic was yourcartoon? In my understanding of all the jetsources, their high-energy leptons are createdby magnetic reconnection in the distorted mag-netosphere of its central rotator, and post-accelerated by its outgoing frequency waves,on scales < ∼ that of the Blandford & Rees de-Laval nozzle, some 10 ± cm. This scale wasalso found by Martin Kluczykont for M 87. Itshould exceed that of the accretion disc. VALENT´I BOSCH-RAMON:
In standardmodels of jet formation, the jet launching re-gion is ∼ − R Sch , so ∼ cm for a stellar mass black hole. In that framework,within that region the jet would be magnet-ically dominated and magnetic reconnectionmay be important, but farther, particle accel-eration is likely to take place through kineticenergy dissipation, via some di ff usive acceler-ation process of the Fermi type. Of course, theissue is still open. IMMACOLATA DONNARUMMA:
Howcould the GeV detection of Cygnus X-1 chal-lenge the theoretical interpretation of micro-quasar activity? May you compare the case ofCygnus X-1 with the one of Cygnus X-3?
VALENT´I BOSCH-RAMON: