ACCEPTEDBYAPJ PreprinttypesetusingLATEXstyleemulateapjv.04/17/13 NOEVIDENCEOFINTRINSICOPTICAL/NEAR-INFRAREDLINEARPOLARIZATIONFORV404CYGNIDURING ITSBRIGHTOUTBURSTIN2015:BROADBANDMODELINGANDCONSTRAINTONJETPARAMETERS Y.T.TANAKA1,R.ITOH2,M.UEMURA1,Y.INOUE3,C.C.CHEUNG4,M.WATANABE5,K.S.KAWABATA1,Y.FUKAZAWA1, Y.YATSU6,T.YOSHII6,Y.TACHIBANA6,T.FUJIWARA6,Y.SAITO,6,N.KAWAI6,M.KIMURA7,K.ISOGAI7,T.KATO7,H.AKITAYA1, M.KAWABATA2,T.NAKAOKA2,K.SHIKI2,K.TAKAKI2,M.YOSHIDA1,M.IMAI5,S.GOUDA5,Y.GOUDA5,H.AKIMOTO8, S.HONDA8,K.HOSOYA8,A.IKEBE8,K.MORIHANA8,T.OHSHIMA8,Y.TAKAGI8,J.TAKAHASHI8,K.WATANABE8,D.KURODA9, 6 T.MOROKUMA10,K.MURATA11,T.NAGAYAMA12,D.NOGAMI7,Y.OASA13,K.SEKIGUCHI14 1 AcceptedbyApJ 0 2 ABSTRACT r We present simultaneous optical and near-infrared (NIR) polarimetric results for the black hole binary a V404 Cyg spanning the duration of its 7-day long optically-brightest phase of its 2015 June outburst. The M simultaneousR and K -band light curves showed almost the same temporal variation except for the isolated s (∼30minduration)orphanK-bandflareobservedatMJD57193.54.Wedidnotfindanysignificanttemporal s 4 variationofpolarizationdegree(PD)andpositionangle(PA)inbothRandK bandsthroughoutourobserva- s 1 tions,includingthedurationoftheorphanNIRflare. WeshowthattheobservedPDandPAarepredominantly interstellar in origin by comparing the V404 Cyg polarimetric results with those of the surroundingsources E] within the 7′×7′ field-of-view. The low intrinsic PD (less than a few percent) implies that the optical and NIRemissionsaredominatedbyeitherdiskoroptically-thicksynchrotronemission,orboth. Wealsopresent H the broadband spectra of V404 Cyg during the orphan NIR flare and a relatively faint and steady state by h. includingquasi-simultaneousSwift/XRT andINTEGRALfluxes. Byadoptingasingle-zonesynchrotronplus p inverse-Comptonmodelaswidelyusedinmodelingofblazars,weconstrainedtheparametersofaputativejet. - BecausethejetsynchrotroncomponentcannotexceedtheSwift/XRTdisk/coronaflux,thecutoffLorentzfactor o intheelectronenergydistributionisconstrainedtobe<102,suggestingparticleaccelerationislessefficientin r thismicroquasarjetoutburstcomparedtoAGNjets. Wealsosuggestthattheloadingofthebaryoncomponent t s insidethejetisinevitablebasedonenergeticarguments. a [ Keywords:polarization—binaries: general—radiationmechanisms: non-thermal—stars: jets—infrared: stars—stars: individual(V404Cyg) 3 v 2 1. INTRODUCTION (∼108- 1010M⊙) BHs. Polarimetryin the opticaland near- 1 infrared(NIR)bandsisapowerfulmethodtounveiltheemis- Arelativisticcollimatedoutflow(knownasajet)emerging 3 sion mechanism and investigate the magnetic field structure from a black hole (BH) is ubiquitously observed in various 1 inside the jet (e.g., Marscheretal. 2008). In Active Galac- spatial scales from stellar mass (∼10M ) to supermassive 0 ⊙ tic Nuclei (AGN) and Gamma-Ray Burst (GRB) jets, it is . 1 [email protected] well accepted that the optical and NIR lights are produced 0 1HiroshimaAstrophysicalScienceCenter,HiroshimaUniversity,1-3-1 by high-energy electrons through optically-thin synchrotron 6 Kagamiyama,Higashi-Hiroshima739-8526,Japan emission. High degrees of linear polarization in the opti- 1 2Department of Physical Sciences, Hiroshima University, Higashi- cal/NIRbandobservedfromAGNandGRBjets(∼10%upto : Hiroshima,Hiroshima739-8526,Japan 30–40%;seee.g.,Ikejirietal.2011;Ueharaetal.2012)indi- v 3InstituteofSpaceandAstronauticalScience,JAXA,3-1-1Yoshinodai, catethatthesynchrotronprocessisoperating,withhighlyor- i Chuo-ku,Sagamihara,Kanagawa252-5210,Japan X 4SpaceScienceDivision,NavalResearchLaboratory,Washington,DC deredmagneticfieldsintheemissionregions. Measurements r 20375-5352,USA of polarization position angle (PA) are also useful to deter- a 5DepartmentofCosmosciences,GraduateSchoolofScience,Hokkaido mine the magnetic field direction at the emission region. In University,kita10,Nishi8,Kita-ku,Sapporo,Hokkaido060-0810,Japan addition,the detectionofa polarizationPA swingmayman- 6Department of Physics, Tokyo Institute of Technology, 2-12-1, ifestfromthepresenceofahelicalmagneticfieldalongajet Ohokayama,Tokyo,Japan 7DepartmentofAstronomy,KyotoUniversity,KitashirakawaOiwake- or curved structure indicating the global jet geometry (e.g. cho,Sakyo-ku,Kyoto,Kyoto606-8502,Japan Marscheretal.2008;Abdoetal.2010). 8CenterforAstronomy,UniversityofHyogo,407-2Nishigaichi,Sayo, It is also known that the jet emission in stellar-mass BHs Hyogo679-5313,Japan appearintheNIRband.Whileemissionintheoptical-bandis 9Okayama Astrophysical Observatory, National Astronomical Obser- dominatedbytheaccretiondisk,anexcesswithrespecttothe vatoryofJapan,Asakuchi,Okayama719-0232,Japan 10InstituteofAstronomy,GraduateSchoolofScience,TheUniversity Rayleigh-Jeanstailofthediskblackbodycomponentisoften ofTokyo,Mitaka,Tokyo181-0015,Japan foundintheNIR-band. Hence,iftheNIRemissionisdueto 11GraduateSchoolofScience,NagoyaUniversity,Furo-cho,Chikusa- optically-thin synchrotron emission, a high polarization de- ku,Nagoya464-8602,Japan gree (PD) is theoretically expected. However, reliable opti- 12GraduateSchoolofScienceandEngineering,KagoshimaUniversity, cal/NIR polarimetric measurements for Galactic BH binary Kagoshima890-0065,Japan jetsarestillverylimitedmainlyduetothedifficultyinelim- 13FacultyofEducation,SaitamaUniversity,Sakura,Saitama338-8570, Japan inating the interstellar polarization caused by large amounts 14National Astronomical Observatory of Japan, Mitaka, Tokyo 181- of dust clouds in our Galaxy in the target directions. An- 8588,Japan otherreasonforthepaucityofintrinsicpolarizationmeasure- 2 ments is that the objects are often not sufficiently bright to typical exposure time of each frame was 15 s. To remove perform polarimetry. In this regard, we note that clear NIR theinstrumentalpolarization(p=0.78%)andtocalibratethe polarizationwas detected from Cyg X-2 and Sco X-1 based polarization PA, we used past MSI data of the two unpolar- onspectro-polarimetry(Shahbazetal.2008). izedstars(BD+323739andHD212311;Schmidtetal.1992) An opportunity to study the BH binary jet through opti- andthree strongly-polarizedstars (HD 154445,HD 155197, cal/NIRpolarimetrywaspresentedwhenV404Cygni(a.k.a., andHD204827;Schmidtetal.1992)obtainedonMJD57167 GS 2023+338; hereafterV404 Cyg) producedan exception- and57169. We also confirmedthe polarizationefficiencyof ally bright outburst in June 2015. This object is one of ∼99.7%usingapolarizerandflat-fieldlamp. the famous low-mass X-ray binaries (LMXBs) because a similar huge outburst was detected in 1989 with intensive 2.3. Swift/XRT multi-wavelength observations performed at that time (e.g., WeanalyzedX-raydataforV404CygtakenwithXRTon- Makino 1989). The distance is accurately determined as board the Swift satellite using HEASoft version 6.16. The 2.39±0.14kpcfromparallaxmeasurementusingastrometric Swift/XRTdataanalyzedhereweretakenonMJD57193and VLBI observations (Miller-Jonesetal. 2009). In this paper, 57194 (observation IDs: 00031403040 and 00031403046), aimedatstudyingthenon-thermaljetemissionandconstrain- which were almost simultaneous (within less than 1 hour) ing the physicalparametersin a microquasarjet, we present with the Kanata/HONIRmulti-bandphoto-polarimetrydata. resultsoflinearpolarizationmeasurementsintheopticaland Cleaneventsofgrade0–12withinthesourcerectangleregion NIRbandsforV404CygduringthebrightestoutburstinJune (becausethe observationwas in window-timingmode) were 2015 performedby the Kanata 1.5 m and Pirka 1.6 m tele- selected. After subtractingbackgroundcountsselectedfrom scopes in Japan. Observations and data reductions are de- bothsidesofthesourcewithrectangleshapes,the0.5–10keV scribedin§2. Weshowtheresultsin§3,andtheimplications events were utilized for spectroscopy. We generated ancil- ofourfindingsarepresentedin§4. lary responsefiles with the xrtmkarftool. Using XSPEC version 12.8.2, we roughly fit the data by assuming a disk 2. OBSERVATIONSANDDATAREDUCTIONS blackbodypluspower-lawmodel,bothmodulatedbyGalac- 2.1. Kanata/HONIR tic absorption (i.e., wabs*(diskbb+pow)),and converted thedeabsorbedspectratoνF fluxes. We performed simultaneous optical and NIR imaging po- ν larimetry for V404 Cyg using the Hiroshima Optical and 3. RESULTS Near IR camera (HONIR; Akitayaetal. 2014) mounted on Fig. 1 (top panel) shows the V-band light curve of theKanata1.5-mtelescopeinHigashi-Hiroshima,Japan.The V404 Cyg during the bright outburst (Kimuraetal. 2016) data presented here were taken on MJD 57193 and 57194. after the detection of burst-like activities by Swift/BAT, The HONIR polarization measurements utilize a rotatable Fermi/GBM,andMAXI/GSCon2015June15(MJD57188) half-wave plate and a Wollaston prism. We selected R and C (Negoroetal. 2015; Kuulkersetal. 2015). The optical flux K bandsasthetwosimultaneousobservingfilters. Tostudy s increased by ∼ 7 mag compared to the quiescent level thewavelengthdependenceofpolarizationpropertiesforthe (V =18.6mag,Wagneretal.1991)withamaximumaround target,wealsotookVR I JHK imagingpolarimetricdataon C C s MJD 57194 and the highest flux level continued for about MJD57194.50–57194.55(the“C”subscriptsarehereinsup- one week. During this brightest phase, the source showed pressed). Each observation consisted of a set of four expo- large-amplitude(asmuchas3mag)andshort-timevariability. sures at half-wave plate position angles of 0.◦0, 22.◦5, 45.◦0, Fig. 1 (two bottom panels)illustrate intra-nightvariationsof and 67.◦5. Typical exposures in each frame were 30 s and theR-andK-bandfluxes,polarizationdegrees(PDs),andpo- 15 s for the R- and K-bands, respectively, but these expo- s s larizationpositionangles(PAs)measuredbyKanata/HONIR sures were sometimes modified depending on weather con- and Pirka/MSI on MJD 57193and 57194(correspondingto ditions (decreased when seeing became good, and increased 2015 June 19 and 20). Note that the Kanata/HONIR R- whencirruscloudspassedoverthetarget). VI-andJH-band and K-band photometric and polarimetric observations are exposuresonMJD57194were30sand15s,respectively. s strictly simultaneous. On MJD 57193, the HONIR observa- To calibrate these data, we observed standard stars on tionswereinterruptedbycloudyweatherandstoppedaround MJD 57195 that are known to be unpolarized (HD 154892) MJD 57193.64, while Pirka/MSI continuously obtained R- and strongly-polarized (HD 154445 and HD 155197; bandphotometricandpolarimetricdataover∼5hours. Turnsheketal. 1990; Wolffetal. 1996). We thereby con- On the whole, the simultaneous R- and K-band light firmed that the instrumentalPD is less than 0.2% and deter- s curvesshowedsimilartemporalvariations. However,around mined the instrumental polarization PA against the celestial MJD 57193.54, a flux increase is evident only in the K coordinategrid.Absolutefluxcalibration,whichisneededto s band, while no corresponding enhancement was observed construct the optical and NIR SEDs, was performed by ob- in the R-band. During this NIR flare, the K-band PD and servingstandardstarsonthephotometricnightMJD57200. s PA did not show any significant variation despite the pro- nounced flux change. The K-band PD and PA were con- 2.2. Pirka/MSI stant at 1.4±0.1% and 9.◦1±s2.◦2, respectively, throughout We also performed optical R-band imaging polarimetry theHONIR observationsthatnight. Thatsame night, theR- monitoring for V404 Cyg using the Multi-Spectral Imager bandlightcurveshowedarapidandlarge-amplitudedecrease (MSI;Watanabeetal.2012)mountedonthe1.6-mPirkatele- around MJD 57193.64 and then gradually recovered. Dur- scopelocatedin Hokkaido,Japan, fromMJD 57190–57193. ingtheopticaldip,thePDandPAwereconstantanddidnot TheMSIobservationswereperformedinasimilarmannerto show any significant variations. Similarly, the R-band PDs theKanata/HONIRones,namelytheMSIutilizesarotatable andPAsmeasuredbyPirka/MSIonthisnightremainedcon- half-waveplateandaWollastonprism,andaseriesoffourex- stant at 7.77±0.01% and 6.◦19±0.◦03, respectively. Note posureswere taken for each polarization measurement. The thatthereis asmalldiscrepancybetweenthe Pirka/MSIand 3 Figure1. Top:V-bandlightcurveofV404Cygduringthemostactivephasein2015June(takenfromKimuraetal.2016). Horizontaldashedlineindicates thequiescentfluxlevelofV=18.6(Wagneretal.1991).ThehatchedorangerectanglesindicatetheKanata/HONIRobservationperiodsshownindetailinthe bottommulti-panelplots.Bottomleftpanels:Kanata/HONIRandPirka/MSIpolarimetricobservationsinMJD57193.NotethattheR-bandmagnitudesshown inthefirstpanelareoffsetby3.0magforillustrativepurposes(i.e.,theactualR-bandmagnitudesare3.0magfainter).Thetwoblackarrowsandgrayrectanglein thefirstpanelindicatethetimerangeswhenKanata/HONIRVRIJHKsandSwift/XRTX-rayspectrawereconstructed,respectively(seealsoleftpanelsinFig.5 andFig.6).Bottomrightpanels:Sameasbottomleftpanels,butforthedatainMJD57194.AblackhorizontalarrowindicatestheperiodwhenKanata/HONIR multibandpolarimetricobservationswereperformed(seealsoFig.4aswellasrightpanelsinFig.5andFig.6). Kanata/HONIR R-band polarimetric results (see Table 1). clouds(NIRobservationsaremoreheavilyaffectedbyclouds Thiscouldbeduetoalackofcross-calibrationbutoursubse- comparedtotheoptical). quentdiscussionisunaffectedbythissmalldifference. Finally, we note that in addition to the photo-polarimetric Onthenextnight,theobservingconditionswererelatively datapresentedherewealsoobtainedPirka/MSIR-banddata gooduntilMJD57194.7andweobtained∼4.5-hoursofcon- onMJD57190,57191,and57192.AsshowninTable1,these tinuous, simultaneous R- and K-band photometric and po- datashowedthatthepolarizationparametersoftheobjectre- s larimetric data for V404 Cyg. As shown in Fig. 1 (bottom- mainedconstantatPD=7.7- 7.9%andPA=6.◦2- 6.◦6over rightpanel),theHONIRphotometriclightcurvesinR-andK thethreenightsdespitedramaticvariabilityofthetotalfluxof s bandsexhibitedquitesimilartemporalprofiles,includingthe ∼2.0mag. two‘dips’aroundMJD 57194.68. ThePDsandPAsinboth To investigate the polarization properties of the sky re- bandswereagainconstantoverthe∼4.5-hourduration,even gion in the direction of V404 Cyg, we also analyzed the overthecourseoftwoobservedfluxdips.Inaddition,thePDs Kanata/HONIR R-band data taken on MJD 57194 (selected andPAs in each bandwere unchangedfromthose measured becausetheobservingconditionwasmuchbettercomparedto onthepreviousnight(MJD57193).Wenotethesporadicna- MJD57193)anddeterminedthePDsandPAsforthebright- tureoftheK-bandpolarimetricdatapoints(namely,PDand estfieldstarswithintheHONIRfield-of-view(FoV).There- s PA) after MJD 57194.64 were due to the passage of cirrus sults are displayed in Fig. 2. We found that the PAs of not 4 150 120 g] 90 e d e [ gl 60 n a n o ati 30 z ari ol 0 P -30 -60 0 2 4 6 8 10 Polarization degree [%] Figure3. R-bandPDandPAforV404Cyg(showninblue)comparedto those observed from each source within the Kanata/HONIR field-of-view (showninred);cf.,Fig.2. Figure2. R-bandPDandPAforV404Cyg(showninblueatcenter)and forsurrounding sources (shown inred)within the7′×7′ Kanata/HONIR field-of-view. omatlholnsaoclotays,tlodVtdbeh4ussee0sprt4sivtaceeCmldtoyheuagetdd,rssiberimuellaoctitctaliiaavolstreneolld.yetvhMlbeaeelrostgsw.rueeTeroPrehvoDneeusrnoVe,dftfi4hi∼0nne4gd7mion.C8ebgy%ajsgescucflatroesenra,ddrVslhPyt4hoDi0wens4deEwCidcayearatrgtlhee-, Polarization degree [%] 10 are the likely cause of the polarized emission in this sky di- 1 rVthe4cet0io4obnCseayrngvd)edaarbePoDluotcahanatdelfdPobAfeytfhooenrdsVut4rhr0eo4udnuCdsytingcgliosoubdnj.oetcTtihsnit(sriinsnucslgiucgdebisnutgst Angle [deg] 6400 V R I J H Ks ivlnaartrgeiraesbtflleeullxParDvosarraiignatidinoP.nAsTs(hfseoerehVytwp4o0ot4hbeCosytitsgomiosbspsueaprnvpeeoldsrteienvdeFnbiygd.uthr1ie)n.gnoWthnee- Polarization 200 also plot in Figure 3 the measured PA as a function of PD -20 for each object within the FoV. Two clusterings of the data 1 Wavelength [um] areclearlyvisible:objectswithverysmallPDandawidePA rangeover180◦arelikelylocatedinfrontoflocaldustclouds, Figure4. NearlysimultaneousVRIJHKs-bandPDandPAmeasurementsof while those with relatively large PDs of typically ∼5% and V404CygobservedwithHONIRfromMJD57194.51–57194.54. broadlysimilarPAsof0◦–30◦arepositionedbeyondthedust clouds. We note that the slightly greater PD of V404 Cyg (∼8%) with respect to the surrounding objects (∼5%) im- 4.1. BroadbandspectrumofV404Cyg pliesasmalllevelofintrinsicpolarizationforV404Cygofat The simultaneous Kanata/HONIR R- and K-band light s mostafewpercent. curves showed almost the same temporal evolution, except Furthermore,weshowtheVRIJHKs-bandPDsandPAsof theorphanKs-bandflarewhichpeakedaroundMJD57193.54 V404 Cyg in Fig. 4 indicating a steep PD decrease toward and lasted for ∼30 mins (see Fig. 1). Apart from this or- longerwavelengthswithconstantPAsoverthesixobservation phanflare(whichisdiscussedindetaillater),thequitesimilar bands.Thispolarizationbehaviorissimilartothatofahighly R- and K-band light curves naturally leads to an interpreta- s reddened star, suggesting that the polarization is interstellar tionthattheNIRandopticalemissionscomefromthesame origin. Detailed studyof the interstellar dustbased on these component(or have the same origin). There are two possi- multibandpolarimetricdatawillbereportedinaforthcoming ble optionsto explainthe opticalandNIR emissions: oneis paper(Itohetal.inpreparation).Fromtheseobservationalre- a disk origin and the other is from a jet. To gauge which is sults, we considerthemeasuredpolarizationofV404Cygis the most plausible, we constructeda broadbandspectrum of predominantlycontaminatedbyinterstellar dustbetweenthe V404CygfromtheradiotoX-raybandsinν-F representa- ν object and the Earth. The low intrinsic PD (less than a few tion(Fig.5). Notethattheradiofluxesarenotsimultaneous, percent)impliesthattheopticalandNIRemissionsaredom- while the Kanata and Swift/XRT data were obtained within inatedbyeitherdiskoroptically-thicksynchrotronemission, a one-hour timespan. The Kanata optical and NIR fluxes orboth. were dereddened by assuming A =4.0 and R =3.1. The V V A value of 4.0 was derived by Casaresetal. (1993) from V 4. DISCUSSION the spectral type of companionstar and B- V colors, which 5 103 103 102 102 101 101 100 100 F [Jy]ν 10-1 F [Jy]ν 10-1 10-2 10-2 10-3 MJD 57191.952 10-3 MJD 57191.952 MJD 57194.944 MJD 57194.944 MJD 57195.941 MJD 57195.941 10-4 MMJJDD 5577119968..993383 10-4 MMJJDD 5577119968..993383 10-5 10-5 108 1010 1012 1014 1016 1018 1020 108 1010 1012 1014 1016 1018 1020 Frequency [Hz] Frequency [Hz] Figure5. Left: Quasi-simultaneous (within1hour)ν- Fν plottedmeasurementsofV404CygobservedbyKanata/HONIRandSwift/XRT onMJD57193. Thesimultaneous Ks-andR-bandfluxesmeasuredduringtheorphanKs-bandflareareshownwithmagentastars. Right: Sameastheleftpanel, butforthe Kanata/HONIRandSwift/XRTdatatakenonMJD57194. Alsoshowninbothpanelsarepreliminarynon-simultaneous2.3to21.7GHzRATAN-600fluxes fromTrushkinetal.(2015a). Forillustrativepurposesonly,dashedlinesindicatingvariousspectralindices(α=0and- 1.0inboth,plusadditionalα=0.2in leftpanel;Fν∝να)areshown. wasalsoconfirmedbysubsequentstudies(e.g.,Shahbazetal. spectralbreakbetweentheK andRbandsoftheflaringemis- s 2003; Hynesetal. 2009). After we corrected the observed sioncomponentcausedbythetransitionofsynchrotronemis- fluxesforextinctionusing A =4.0, we founda slightly ris- sionfromoptically-thicktooptically-thinregimes,ifpresent, V ing,butalmostflat(α∼0)shapeintheopticalandNIRspec- would make the contribution of the flaring component neg- trum.Thisimpliesoptically-thicksynchrotronemissionfrom ligible with respect to the baseline flux, as observed. The anouterjet.Indeed,the∼2- 20GHzradiospectrumobtained quasi-simultaneousF Kanata/HONIRand Swift/XRT spec- ν about 1.5 days before our observation (on MJD 57191.95) tra measured on MJD 57194.52, together with the RATAN- canbesmoothlyextrapolatedtotheKanata/HONIRspectral 600non-simultaneousradiofluxesarealsoshown.Theradio, data assuming a F ∝ν0.2 form (see Fig 5). However, the optical/NIR, and X-ray spectra would be reasonably under- ν flat optical/NIR spectral shape can also be interpreted in a stoodasoptically-thicksynchrotronemission fromthe outer diskmodel(seee.g.,Kimuraetal.2016,ExtendedDataFig- jet, disk emission, and disk plus corona emissions, respec- ure6therein).Thus,itisdifficulttodeterminetheoptical/NIR tively. emission mechanismsolely from its spectralshape. No evi- dence of intrinsic linear polarization in the R- and Ks-bands 4.2. Propertiesofthejet are allowed in both scenarios because both optically-thick To constrain the jet parameters and physical quantities in synchrotronandblackbodyradiationonlygenerateweaklin- the emission region, we attempted to model the spectral en- earpolarizationoforder.10%. ergy distribution (SED) of V404 Cyg by using a one-zone Here, we focus on the orphan K-band flare which lasted s synchrotron plus synchrotron self-Compton (SSC) model for only ∼ 30 mins at MJD 57193.54. The observed red (Finkeetal.2008),whichiswidelyusedforblazarSEDmod- color and short duration imply synchrotron emission from eling(e.g.,Abdoetal.2009,2011;Tanakaetal.2014,2015). a jet as the most plausible origin of the flare. Indeed, the WeshowinFig.6(leftpanel)thequasi-simultaneousbroad- K-band peak flux of the flare reached ∼2.0 Jy, which was s bandSEDofV404CygduringtheK-bandflare. Thismod- the same level measured during the giant radio and sub-mm s eling assumes that a single emission region is located at the flaresobservedbyRATAN-600andSubMillimeterArrayon inner-most part of the jet. Hence, the optically-thick syn- MJD 57198.933 and MJD 57195.55 (Trushkinetal. 2015b; chrotron emission from an outer jet, which has a flat spec- Tetarenkoetal. 2015), respectively. As shownin Fig.5 (left panel),anextrapolationoftheradiospectrumobservedduring trumofFν ∝ν0 observedintheradiouptoNIRbandisnot modeled,whiletheoptically-thinsynchrotronandSSCemis- thegiantflareatGHz-frequenciesonMJD57198.933nicely sions at optical frequencies and higher are fitted. However, connects to the K-band peak flux by assuming a flat spec- s tral shape (i.e., F ∝ ν0). More interestingly, even during in the currentcase, we now knowthatthe opticaland X-ray ν emission are from a disk and disk plus corona, respectively. the orphan flare, the K-band PD remains showed no signif- s WethereforeregardtheKanata/HONIR,Swift/XRT,andIN- icant temporal variation, indicating the NIR emission is not TEGRAL data points (taken from Fig. 3 of Rodriguezetal. strongly polarized. This result would be reasonably under- 2015) as upper limits for the jet emission. Another con- stood if the jet synchrotron emission in the K band is still s straint comes from the Kanata/HONIR observation that the in the optically-thick regime. We therefore conjecture that K-band emission is not significantly polarized even during thisorphanK-bandflareisproducedbyoptically-thicksyn- s s the orphan flare. This indicates that the K-band emission chrotronemissionfromanouterjet. Iftheoptically-thicksyn- s is still in the optically-thick regime and that the break fre- chrotronemissionextendsuptotheR-bandwithaflatspectral quency(definedasν )isduetosynchrotronselfabsorption shapeandthebaselineR-bandfluxisnotashighastheflaring SSA (SSA). The transition from the optically-thick to optically- component of ∼2.0 Jy (after extinction correction, assum- thin regime is above the K-frequency band, thus ν & ingA =4.0),itsignificantlycontributestotheR-bandfluxas s SSA well,VmakingtheflarevisiblealsointheR-bandlightcurve.A 1.4×1014Hz,andweadoptavalueofνSSA=3.0×1014Hz. We also assume that the synchrotron peak flux is 2 Jy as 6 10-4 10-4 10-5 10-5 10-6 10-6 10-7 10-7 -2-1 s]m 1100--98 -2-1 s]m 1100--98 νν F[erg c 1100--1101 ν νF[erg c 1100--1101 10-12 10-12 10-13 MJD 57191.952 10-13 MJD 57191.952 MJD 57194.944 MJD 57194.944 10-14 MMMJJJDDD 555777111999568...999433183 10-14 MMMJJJDDD 555777111999568...999433183 10-15 10-15 109 1011 1013 1015 1017 1019 1021 1023 1025 109 1011 1013 1015 1017 1019 1021 1023 1025 Frequency [Hz] Frequency [Hz] Figure6. Left:BroadbandspectrumofV404Cyginν- νFν representation.TheKanata/HONIRandSwift/XRTfluxeswerequasi-simultaneouslyobservedon MJD57193,whileRATAN-600andINTEGRALdatawerenotsimultaneous. One-zonesynchrotronandSSCemissionsarealsodrawnbysolidredandblue lines,respectively. Blackdashedlineindicatestheoptically-thicksynchrotronemissionfromanouterjet,withaspectralshapeofFν∝ν0.Right:Sameasthe leftpanel,butforthequasi-simultaneousKanata/HONIRandSwift/XRTfluxesobservedonMJD57194. observed by Kanata/HONIR (and also by RATAN-600 on 2014). Since dominant cooling processes for electrons of MJD 57198.933). We can then derive the magnetic field B γ=γ =102arebothsynchrotronandSSC,thecoolingtime cut and the size of the emission region R by using the standard is estimated as t = 3γm c/ 8σ γ2 U +U , where cool e T B sync formulae for synchrotron absorption coefficient and emis- U = B2/8π and U = L /4(cid:0)πR2c a(cid:0)re the ener(cid:1)g(cid:1)y densi- sitivity (e.g., Rybicki&Lightman 1979; Chatyetal. 2011; B sync sync ties of magnetic field and synchrotron photons, respectively Shidatsuetal.2011), (e.g.,Finkeetal. 2008). Thus, we obtainη∼106 bysetting 3s+10 - 2 - 4 E=γcutmec2andB=105G.Thisismuchlargerthanη∼10in B≈1×105 νSSA 2s+13 Fν 2s+13 D 2s+13 Gau(sb1sl,a)zarjets(e.g.,Rachen&Mészáros1998),indicatingmuch (cid:18)3×1014Hz(cid:19) (cid:18)2Jy(cid:19) (cid:18)2.4kpc(cid:19) longer acceleration times and inefficient acceleration in this -(s2+7s+8) s+6 2(s+6) microquasarjet. R≈5×108 νSSA 2(2s+13) Fν 2s+13 D 2s+13 cm,(2)Note that the electron power-law index of 2.2 we ob- (cid:18)3×1014Hz(cid:19) (cid:18)2Jy(cid:19) (cid:18)2.4kpc(cid:19) tained from SED modeling has already been modified by rapid synchrotron and SSC cooling. In the current situa- where s is the power-law index of electron distribution (see tion, because the magnetic field is strong (B ∼ 1×105 G) below) and D is the distance to V404 Cyg. Here we as- and emission region is small (R ∼ 5×108 cm), we need sumedalmostequipartitionbetweenmagneticfieldandelec- to consider the following three energy loss processes: adi- tron energy density, which was confirmed by the following abatic cooling (this is also equivalent to particle escape SEDmodeling. from emission region), synchrotron cooling, and SSC cool- The electron energydistribution is assumed to have a sin- ing. Note here that we can neglect synchrotron cooling gle power-law shape with exponential cutoff as dN/dγ = for electrons of γ . 40 because the optically-thick regime Kγ- sexp(- γ/γcut)forγmin≤γ≤γmax,whereγistheelectron is below ν =3×1014 γ/40 2 B/105G Hz (e.g., Piran Lorentzfactor,Kistheelectronnormalization,sisthepower- SSA 2004). Cooling timescale(cid:0)s for t(cid:1)he(cid:0)se proces(cid:1)ses are estimated lawindex,γcutisthecutoffenergy,γminandγmaxarethemini- as t &R/c∼1.7×10- 2 R/5×108cm and t ∼5.1× mumandmaximumelectronenergiesandarerespectivelyset ad SSC to 1 and 106. The jet inclination angle of V404 Cyg is es- 10- 2γ- 1 Usync/3×108erg(cid:0)cm- 3 - 1, respe(cid:1)ctively (e.g., Piran timatedas∼55◦ (e.g.,Shahbazetal.1994;Kharghariaetal. 2004;Ch(cid:0)atyetal.2011). Theref(cid:1)ore,high-energyelectronsof 2010)andassumingajetvelocityof0.9c,thecorresponding γ &3 rapidlylose their energyvia SSC emission and hence Doppler beaming factor is δ =[Γ(1- βcosθ)]- 1 ∼0.9. We theelectronpower-lawindexbecomessteeperby onepower ofE,ifinjectionofhigh-energyemittingelectronscontinued canthereforesafelyneglectrelativisticbeamingeffects. overafewtensofminutes(whichcorrespondstotheflaredu- Bychangingtheparametersoftheelectronenergydistribu- rationobservedbyKanata/HONIR).Namely,theelectronen- tion, we calculated the resultant synchrotronand SSC emis- sions. The calculatedmodelcurvesare shownin Fig. 6 (left ergydistributionatγ &3isalreadyinafast-coolingregime, whichindicatesthattheoriginal(orinjected)electronpower- panel)andallthemodelparametersaretabulatedinTable2. law index is 1.2. This is much smaller than the standard The derived parameters for the electron energy distribution are K =4.5×1040, s=2.2, and γ =102. Importantly, the power-law index of 2.0 derived by the first-order Fermi ac- cut celerationtheory(e.g.,Blandford&Ostriker1978). Swift/XRT data allowed us to constrain the cutoff energy as We can now derivethe total energyin electronsand mag- 102 becauselargerγcut violatesthese upperlimitsinthe soft neticfieldasW =m c2 γmaxdγγdN/dγ=9.9×1034ergand X-rayband. Thisimpliesthatparticleaccelerationinthismi- e e γmin croquasar jet is not very efficient. The electron energy dis- W = 4/3 πR3U = 4R.7×1035 erg, respectively, thus the B B tributioncutoffenergyisdeterminedbythebalancebetween jet is P(cid:0)oyn(cid:1)ting-flux dominated by a factor of ∼5. The jet accelerationandcoolingtimes,wheretheaccelerationtimeis powerin electrons(L ) andmagneticfield (L ) iscalculated e B definedastacc=ηE/(eBc)byusinganelectronenergyE and asLi=2πR2βcUi (i=e,B), whereUe=We/ 4πR3/3 isthe parameterη,thenumberofgyrationsanelectronmakeswhile electron energy density, β = 0.9 is assume(cid:0)d, and th(cid:1)e fac- doubling its energy (e.g., Finkeetal. 2008; Muraseetal. 7 tor of 2 is due to the assumption of a two-sided jet (e.g., tion, particularlyin relationto AGN jets canbe summarized Finkeetal. 2008). Then, we obtain L =7.8×1036 erg s- 1 asfollows. e andL =3.7×1037ergs- 1,andthesummedpower(L +L ) amounBts to 4.5×1037 erg s- 1. On the other hand, wee caBn 1. The SSA frequency and peak flux density enable us to estimate the magnetic field strength and size of alsocalculatethetotalradiatedpowerusingtheSEDmodel- theemissionregionbyassumingequipartitionbetween ingresultasL =L +L =7.0×1037 ergs- 1, whichis rad sync SSC magnetic field and relativistic electrons. The derived largerthanthesummedLe+LB=4.5×1037 ergs- 1. Thisin- magneticfieldofB∼105 Gaussismuchstronger,and dicatesthatthePoyntingflux(LB)andLe arenotsufficientto sizeoftheemissionregionofR∼108cmmuchsmaller, explainLrad andanotherformofpowerisrequired. Thesim- comparedtoAGNjets(typicallyB∼1- 10Gaussand plestandmostprobablesolutionistoassumethatthejetcon- R∼1017- 18cm,seee.g.,Ghisellinietal.(2010)). tainsenoughprotonswhichhavelargerpowerthanL (e.g., rad Sikora&Madejski2000;Ghisellinietal.2014;Tanakaetal. 2. Based on modeling of the broadband spectrum of 2015;Saitoetal.2015). Thisisanother(thoughindirect)ev- V404 Cyg, we found an upper limit to the cutoff idenceofabaryoncomponentinamicroquasarjet. Notethat Lorentzfactorofelectronsof∼100. Becausethecut- wereachedtheaboveconclusionbasedonjetenergeticsargu- off is determined by the balance of the acceleration ment,butthebaryonicjetinamicroquasarhasalreadybeen andcoolingtimes,thisresultimpliesalongeraccelera- claimedbyadifferent,independentmethodbasedonthede- tiontimeofη∼106inthismicroquasarjet,suggesting tectionofblue-shiftedemissionlinesintheX-rayspectrafor electronaccelerationismuchlessefficientcomparedto SS433and4U1630- 47(Kotanietal.1994;DíazTrigoetal. AGNjets(thattypicallyshowη∼10). 2013). By assumingthat the jet containsone cold protonper one 3. Theoriginal(orinjected)power-lawindexoftheelec- relativistic (emitting) electron, we can derive the total en- tron energydistributionwas derivedas s=1.2. In the ergyofcoldprotonsasW =m c2 γmaxdγdN/dγ=5.4×1037 SED modeling of AGN jets, electrons are assumed to erg, where m is the propton mpasRsγ.miTn his corresponds to the haveabrokenpower-lawshape. Thepower-lawindex p cold proton power (L ) of 2.1×1039 erg s- 1 by using the belowthebreakLorentzfactorγbreak(typicallyγbreak∼ p 100- 1000)isestimatedass∼1(e.g.,Ghisellinietal. relation of L = 2πR2βcU (e.g., Finkeetal. 2008), where p p 2010). Hence, s=1.2derivedhereinV404Cygjetis Up = Wp/ 4πR3/3 is the energy density of cold protons comparableto thatderivedin AGN jets, implyingthat and β = 0(cid:0).9 is ass(cid:1)umed. We therefore obtain the total jet same acceleration mechanismoperatesin these differ- powerLjet=Le+LB+Lp andradiativeefficiencyofthejetas entsystems. L /L ∼3%(seealsoTable2). rad jet DuringthemoderateandsteadystateonMJD57194,there 4. To account for the total radiated power of the jet of were no observational constraints on ν due to the domi- V404Cyg,acoldprotoncomponentisrequiredinside SSA nance of the disk component in the optical and NIR bands, thejet. ThisisthesamesituationasinAGNjets. hence we assume that it remained the same as that during thebrightflare,ν =3.0×1014 Hz. Weestimatedthesyn- 5. DuringthebrightflareonMJD57193,thejetradiative chrotronpeakfluxSSaAs0.2Jy,becausesuchafluxlevelwasob- efficiency(ǫrad≡Lrad/Ljet=Lrad/ Le+LB+Lp ) isde- servedintheGHzbandduringthehighstate(Trushkinetal. rived as ∼3%. This is roughlyc(cid:0)omparable to(cid:1)that of 2015a)andanextrapolationtotheNIRbandwithaflatshape AGN and GRB jets (∼10%, see e.g., Nemmenetal. (α=0) seems reasonable. We thereby obtained the follow- (2012);Ghisellinietal.(2014)). ing estimates of B ∼ 2×105 G and R ∼ 2×108 cm (see Equations(1)and(2)). Importantinformationaboutthenon- thermal jet emission, which should be included in the SED We appreciate the anonymousreferee’s constructive com- modeling, comes from the INTEGRAL detection of an ad- ments that helped to improve the manuscript. We thank ditional power-law component of Γ = 1.54+0.24 in the hard Dr. Trushkin and Dr. Rodriguez for providing us with their - 0.45 X-ray band (Rodriguezetal. 2015). The Kanata/HONIR, RATAN-600andINTEGRALdata,respectively. Thisworkis Swift/XRT, and INTEGRAL (exponential cutoff power-law supported by the Optical & Near-Infrared Astronomy Inter- componentdominantupto∼100keV)datapointsaretreated University Cooperation Program, and the MEXT of Japan. as upper limits. Based on these assumptions and the multi- We acknowledge with thanks the variable star observations wavelength data, we calculated the broadband non-thermal from the AAVSO International Database contributed by ob- jetemissionbyacceleratedelectronsusingtheone-zonesyn- servers worldwide and used in this research. YTT is sup- chrotronand SSC model. The resultis shown by solid lines ported by Kakenhi 15K17652. MU is supported by Kak- in Fig. 6 (right panel) and the model parameters are tabu- enhi 25120007. CCC is supported at NRL by NASA DPR lated in Table 2. We obtained the same parameter values of S-15633-Y.MWissupportedbyKakenhi25707007. s=2.2andγ =102fortheelectronenergydistribution,but cut the electron normalization K =7.0×1039 is smaller due to REFERENCES thefainterjetflux,aswasobservedintheradioband.Wealso Abdo,A.A.,Ackermann,M.,Ajello,M.,etal.2009,ApJ,707,55 foundL /L ∼2,suggestingagainthejetisslightlyPoynting- B e —.2010,Nature,463,919 flux dominated. More interestingly, the total radiated power —.2011,ApJ,736,131 is again larger than the summed electron and magnetic field Akitaya,H.,Moritani,Y.,Ui,T.,etal.2014,inSocietyofPhoto-Optical powers in the jet (see Table 2), implying the presence of a InstrumentationEngineers(SPIE)ConferenceSeries,Vol.9147,4 Blandford,R.D.,&Ostriker,J.P.1978,ApJ,221,L29 baryoniccomponentevenduringthefainterstate. Casares,J.,Charles,P.A.,Naylor,T.,&Pavlenko,E.P.1993,MNRAS, OurdiscussionofthejetpropertiesofV404Cyginthissec- 265,834 8 Chaty,S.,Dubus,G.,&Raichoor,A.2011,A&A,529,A3 Rybicki,G.B.,&Lightman,A.P.1979,RadiativeProcessesin DíazTrigo,M.,Miller-Jones,J.C.A.,Migliari,S.,Broderick,J.W.,& Astrophysics(Wiley) Tzioumis,T.2013,Nature,504,260 Saito,S.,Stawarz,Ł.,Tanaka,Y.T.,etal.2015,ApJ,809,171 Finke,J.D.,Dermer,C.D.,&Böttcher,M.2008,ApJ,686,181 Schmidt,G.D.,Elston,R.,&Lupie,O.L.1992,AJ,104,1563 Ghisellini,G.,Tavecchio,F.,Foschini,L.,etal.2010,MNRAS,402,497 Shahbaz,T.,Dhillon,V.S.,Marsh,T.R.,etal.2003,MNRAS,346,1116 Ghisellini,G.,Tavecchio,F.,Maraschi,L.,Celotti,A.,&Sbarrato,T.2014, Shahbaz,T.,Fender,R.P.,Watson,C.A.,&O’Brien,K.2008,ApJ,672, Nature,515,376 510 Hynes,R.I.,Bradley,C.K.,Rupen,M.,etal.2009,MNRAS,399,2239 Shahbaz,T.,Ringwald,F.A.,Bunn,J.C.,etal.1994,MNRAS,271,L10 Ikejiri,Y.,Uemura,M.,Sasada,M.,etal.2011,PASJ,63,639 Shidatsu,M.,Ueda,Y.,Tazaki,F.,etal.2011,PASJ,63,785 Khargharia,J.,Froning,C.S.,&Robinson,E.L.2010,ApJ,716,1105 Sikora,M.,&Madejski,G.2000,ApJ,534,109 Kimura,M.,Isogai,K.,Kato,T.,etal.2016,Nature,529,54 Tanaka,Y.T.,Stawarz,Ł.,Finke,J.,etal.2014,ApJ,787,155 Kotani,T.,Kawai,N.,Aoki,T.,etal.1994,PASJ,46,L147 Tanaka,Y.T.,Doi,A.,Inoue,Y.,etal.2015,ApJ,799,L18 Kuulkers,E.,Motta,S.,Kajava,J.,etal.2015,TheAstronomer’sTelegram, Tetarenko,A.,Sivakoff,G.R.,Young,K.,Wouterloot,J.G.A.,& 7647,1 Miller-Jones,J.C.2015,TheAstronomer’sTelegram,7708,1 Makino,F.1989,IAUCirc.,4782 Trushkin,S.A.,Nizhelskij,N.A.,&Tybulev,P.G.2015a,The Marscher,A.P.,Jorstad,S.G.,D’Arcangelo,F.D.,etal.2008,Nature,452, Astronomer’sTelegram,7667,1 966 Trushkin,S.A.,Nizhelskij,N.A.,&Tsybulev,P.G.2015b,The Miller-Jones,J.C.A.,Jonker,P.G.,Dhawan,V.,etal.2009,ApJ,706,L230 Astronomer’sTelegram,7716,1 Murase,K.,Inoue,Y.,&Dermer,C.D.2014,Phys.Rev.D,90,023007 Turnshek,D.A.,Bohlin,R.C.,Williamson,II,R.L.,etal.1990,AJ,99, Negoro,H.,Matsumitsu,T.,Mihara,T.,etal.2015,TheAstronomer’s 1243 Telegram,7646,1 Uehara,T.,Toma,K.,Kawabata,K.S.,etal.2012,ApJ,752,L6 Nemmen,R.S.,Georganopoulos,M.,Guiriec,S.,etal.2012,Science,338, Wagner,R.M.,Bertram,R.,Starrfield,S.G.,etal.1991,ApJ,378,293 1445 Wolff,M.J.,Nordsieck,K.H.,&Nook,M.A.1996,AJ,111,856 Piran,T.2004,ReviewsofModernPhysics,76,1143 Rachen,J.P.,&Mészáros,P.1998,Phys.Rev.D,58,123005 Rodriguez,J.,CadolleBel,M.,Alfonso-Garzón,J.,etal.2015,A&A,581, L9 Table1 Dailyopticalandnear-infraredpolarizationdegree(PD)andpositionangle (PA). R-band Ks-band MJD PD(%) PA(deg) PD(%) PA(deg) Instrument 57190 7.85±0.11(1.34/25) 6.59±0.35(1.09/25) – – Pirka/MSI 57191 7.72±0.05(0.62/20) 6.33±0.22(1.04/20) – – Pirka/MSI 57192 7.76±0.02(0.51/22) 6.39±0.10(0.74/22) – – Pirka/MSI 57193 7.77±0.01(0.74/226) 6.19±0.03(0.72/226) – – Pirka/MSI 8.03±0.06(0.24/16) 8.23±0.25(0.29/16) 1.43±0.10(1.45/14) 9.13±2.18(1.73/14) Kanata/HONIR 57194 7.96±0.02(0.58/44) 7.92±0.07(0.32/44) 1.50±0.04(0.78/24) 10.36±0.72(0.69/24) Kanata/HONIR Note.—Shownintheparenthesesarecorrespondingreducedχ2anddegreeoffree- domwhentheobservationswerefittedwithaconstantvalue(i.e.,assumingnotemporal variation). Table2 Modelparameters. Parameter Symbol MJD57193 MJD57194 Breakfrequency[Hz] νSSA 3×1014 3×1014 MagneticField[G] B 1.4×105 1.8×105 Sizeofemissionregion[cm] R 5.3×108 1.7×108 Jetvelocity[c] β 0.9 0.9 Electrondistributionnormalization[electrons] K 4.5×1040 7.0×1039 ElectronPower-lawIndex s 2.2 2.2 MinimumElectronLorentzFactor γmin 1.0 1.0 CutoffElectronLorentzFactor γcut 102 102 MaximumElectronLorentzFactor γmax 106 106 Synchrotronluminosity[ergs- 1] Lsync 2.8×1037 4.3×1036 SSCluminosity[ergs- 1] LSSC 4.1×1037 7.2×1036 Totalradiationluminosity[ergs- 1] Lrad 6.9×1037 1.2×1037 JetPowerinMagneticField[ergs- 1] LB 3.7×1037 6.3×1036 JetPowerinElectrons[ergs- 1] Le 7.8×1036 3.7×1036 JetPowerinColdProtons[ergs- 1] Lp 2.1×1039 2.0×1039 JetRadiativeEfficiency[%] ǫrad ∼3 ∼1