Table Of ContentFLUX AND SPECTRAL VARIATIONS OF
1E1740.7−2942 OVER THE YEARS 2003−2012
5
1
0
2
n
a Manuel Castro∗, Flavio D’Amico, Francisco Jablonski, João Braga
J
InstitutoNacionaldePesquisasEspaciais(INPE),SãoJosédosCampos,SãoPaulo,Brazil
4
1 E-mail: [email protected]
]
E
Theblackholesystem1E1740.7–2942isusuallythebrightesthardX-raysource(above20keV)
H
near the Galactic Center, but presents some epochs of low emission (below the INTEGRAL
.
h
detectionlimit,forexample). Inthiswork,wepresenttheresultsofstudieson1E1740.7–2942
p
- over 10 years, using the instruments ISGRI/IBIS and JEM-X, both on board the INTEGRAL
o
observatory. WefitthespectrawithboththecompTTandcutoffplmodels. Accordingtothefits
r
t
s andtakingthemeanvalueoverthe10years,wehaveobtainedaplasmatemperatureintherange
a
[ ∼20–90keV, and an average powerlaw index of 1.41 (σ=0.25). We have also made a Lomb–
Scargle periodogram of the flux in the 50–20keV band and found two tentative periods at 2.90
1
v and3.99days. Wepresentherethepreliminaryresultsofthisongoingwork.
4
0
4
3
0
.
1
0
5
1
:
v
i
X
r
a
10thINTEGRALWorkshop: "ASynergisticViewoftheHighEnergySky"-Integral2014,
15-19September2014
Annapolis,MD,USA
∗Speaker.
(cid:13)c Copyrightownedbytheauthor(s)underthetermsoftheCreativeCommonsAttribution-NonCommercial-ShareAlikeLicence. http://pos.sissa.it/
1E1740.7−2942withINTEGRAL ManuelCastro
1. Introduction
The source 1E1740.7–2942 is a black hole candidate nearby the center of our Galaxy. It
was discovered by the Einstein satellite [1] and it was thoroughly studied by several high energy
missions like SIGMA/GRANAT, INTEGRAL, Suzaku and Chandra [2, 3, 4, 5]. It was also the
subjectofarecentstudybyourgroup[6],wherewemodelledthesoft–to–hardX-rayspectrumas
dueto(thermal)Comptonization.
TheINTEGRALsatellite[7]hasalreadyprovidedanuniformobservationdatabaseof1E1740.7–
2942. Thisdatabaseissuitableforastudyofanypossiblelong–termvariabilityofthesource. One
interestinggoalofsuchastudyisthesearchformodulationsintheX-raylightcurvetorevealany
possible orbital period. Reported orbital periods for 1E1740.7–2942 [8, 9] are still tentative, and
a more precise determination of this period would be very important in the search for a (longer
wavelength)counterpartof1E1740.7–2942,achallengingtaskthatremainstobeaccomplished.
In this work we show the results of an ongoing effort to determine the orbital period of
1E1740.7–2942 (if there is one). We have made use of the INTEGRAL database consisting of
all of the 1E1740.7–2942 observations from 2003 up to 2012. We have found two marginally
significanttentativeperiodsat∼2and∼4days. Inthenextsectionswedescribeourdatabase,our
datareductionandanalysis,anddiscussourresults.
2. DataReductionandAnalysis
From the public archive of INTEGRAL, we downloaded all of 1E1740.7–2942 observations
starting at 2003 up to 2012. 1E1740.7–2942 spends most of its time in the so-called canonical
low–hard state (LHS), where the fraction of the total flux is higher in the hard X-ray band than in
the soft X-ray band. Accordingly, the source flux is below the sensitivity of JEM-X detectors in
their energy range, being preferably monitored in the band covered by the IBIS telescope. As a
result, Figure(1) shows how our retrieved 314 observations database is distributed between the 3
INTEGRALimaging(unit)telescopes.
Figure1:Ourdatabaseof314retrievedobservationsof1E1740.7–2942. Inthevastmajorityofcases(212)
the source was imaged by IBIS alone. In 14 occasions the source was imaged by the 2 JEM-X units plus
IBIS.In54withIBISplusonlytheJEM-X1unitand34withIBISplusJEM-X2unit.
Spectra were extracted for each individual revolution using an automatic extraction script
basedontheOSA10.0software. InsomecasesthespectraarebuiltonlywithIBISdata,covering
2
1E1740.7−2942withINTEGRAL ManuelCastro
the20–200keVenergyband. WhenthesourcewasalsoimagedbyJEM-X,aspectrumfrom5–200
keV could be extracted with the aid of our automatic script. After spectrum extraction, data were
fitted also automatically with the aid of a Tcl script. For this preliminary study, we fitted the 20–
200orthe5–200keVspectra(whenpresent)withtwomodels. ThefirstoneisasimplecutoffPL,
suitable,forinstance,todetermineinwhichstatethesourcewasobserved[10]. Thesecondmodel
is compTT, following the procedures of our previous study of the source [6]. In 14 occasions, as
is shown in Figure (1), the spectrum extraction script needs to run three times (first run is always
for IBIS), since both JEM-X units provided data. After fitting the data, our Tcl script saves the
fitparameters(witherrorestimates)andcalculatedfluxes. Webuiltourdatabasewiththiskindof
datafiles. Inthisfirststageofourstudy,forsimplicity,errorsinthefluxmeasurementsareassumed
tobewithin±5%.
3. Results
The1E1740.7–2942fluxhistoryofourdatabaseisshowninFigure(2). InFigure(3)weshow
thecorrelationbetweenthefluxesinthetwohighestenergybandsofthisstudy. Thehistogramdis-
tributionsforthe4bandsweadoptedinthisstudyarepresentedinFigure(4). Thepowerlawindex
andplasmatemperaturedataforourdatabasearepresentedinFigures(5)and(6),respectively.
We have also produced a Lomb-Scargle periodogram using the R software environment [11].
ThisisshowninFigure(7).
4. Discussion
Our results have confirmed, as it is widely known, that 1E1740.7–2942 spends most of its
timeinitscanonicalLHS,asderivedfromthepowerlawindexofafitwithcutoffPL(seeFigure5).
Spectralfittinginourdatabasefrom20to200keVor5to200keV(whenavailable)usingcompTT
models,yieldsanaveragetemperatureoftheelectronicplasma,which(thermally)Comptonizesthe
soft seed photons, of ∼50keV (see Figure6). The ∼300 points in our database were helpful to
better constrain this value. It must be emphasized, however, that at this point of our preliminary
studythedataforderivingΓandkT arebasedonverylowfluxes(seeFigure2)sotheuncertainties
e
aresomewhathigh.
Asmentionedabove,oneofthemaingoalsinthisworkistosearchforpossiblebinaryperiods.
Ourdatabaseisoneofthebestforsuchpurposes,since1E1740.7–2942spendsmostofitstimein
the LHS, i.e., the flux of 1E1740.7–2942 is higher in the hard X-ray band (E>20keV). Previous
searchesuptodatewherelimitedbydatabasesinsofterbands. INTEGRALandthetelescopeswe
used in this study, JEM-X and IBIS, are also imaging instruments and consequently 1E1740.7–
2942 flux is not affected by source confusion. Previous studies of searches for 1E1740.7–2942
periodshavemadeuse(mostly)ofdatacollectedbynon-imaginginstrumentsandatthesoftX-ray
band(E<20keV).
Our results, which can be seen in (Figure7), show the presence of periods at approximately
2.90 and 3.99days. Even though relatively well above local noise level in the periodogram, the
second period must be taken with caution since it is quite close to a multiple of the one-day
sampling period used. In addition, the 2.90-day period produces a folded light curve that is not
3
1E1740.7−2942withINTEGRAL ManuelCastro
111111000000
111111EEEEEE 111111777777444444000000......777777−−−−−−222222999999444444222222 IIIIIInnnnnntttttteeeeeeggggggrrrrrraaaaaallllll//////JJJJJJeeeeeemmmmmm−−−−−−XXXXXX ddddddaaaaaattttttaaaaaa
999999 bbbbbblllllluuuuuueeeeee iiiiiissssss 555555−−−−−−222222000000 kkkkkkeeeeeeVVVVVV aaaaaannnnnndddddd mmmmmmaaaaaaggggggeeeeeennnnnnttttttaaaaaa iiiiiissssss 111111000000−−−−−−222222000000 kkkkkkeeeeeeVVVVVV
888888
777777
666666
555555
1111 444444
−−−−
ssss 333333
2222 222222
−−−−
mmmm
1177
g cg cg cg c 1155 1r1reeEEdd 11iiss77 442200 00..−−7755−−0022 99kk44ee 22VV IIaannnntteeddgg gg rraarreell//eeIIBBnnII SSiiss aa55ll00oo−−nn22ee00 dd00 aa kkttaaeeVV
rrrr 1133
eeee
1111
0000
1111 99
−−−−
0000 77
1111
55
n n n n
33
i i i i
xxxx 11
uuuu
FlFlFlFl 11117777
11115555 1111EEEE 1111777744440000 ....7777−−−−222299994444 2222 IIIInnnntttteeeegggg rrrraaaallll ddddaaaattttaaaa
ccccoooolllloooorrrr sssscccchhhheeeemmmmeeee ffffoooolllllllloooowwwwssss pppprrrreeeevvvviiiioooouuuussss pppplllloooottttssss
11113333
11111111
9999
7777
5555
3333
1111
5555 2222 5555 4444 5555 6666 5555 8888 5555 11110000 5555 11112222 5555 11114444 5555 11116666 5555 11118888 5555 22220000 5555 22222222 5555 22224444 5555 22226666 5555 22228888 5555 33330000 5555
OOOObbbbsssseeeerrrrvvvvaaaattttiiiioooonnnn nnnnuuuummmmbbbbeeeerrrr
Figure2: Thefluxhistoryof1E1740.7–2942basedonourdatabase.
very clean, which introduces uncertainties that we are still analysing. It must also be emphasized
thatmuchmoresensitivestudiesusingRXTE[9]haveshownaperiodof12.73±days.
5. Conclusion
We have analysed data from the INTEGRAL mission to build flux and spectral history of
1E1740.7–2942from2003to2012. Ourdataanalysisisautomaticanddividedintwomainfronts.
FirstweuseascripttoautomaticallyreducethedatausingtasksoftheOSA10.0software. Second,
a Tcl script fits the data with XSPEC–aided commands. We collected, when possible, flux in 4
energybands. Ourdatashow,basedonourfitswithcutoffPL,thatthesourceisobservedtobein
its canonical LHS state, where the flux in the hard X-ray band is greater than in the soft one. Our
characterization ofthe 1E1740.7–2942 spectra withcompTTshows anaverage kT of 50keV. A
e
Lomb–Scargleperiodogramanalysishasshowntwotentativeperiodsof2.90and3.99day. These
resultsmustbetakenwithcautionduetouncertaintiesthatarestillbeingaddressed.
4
1E1740.7−2942withINTEGRAL ManuelCastro
18
16
1E 1740.7−2942 INTEGRAL/IBIS data
1
−
s 14
2
−
m
c
g 12
er
0
1
−0 10
1
ux in 8
V fl
e
0 k 6
0
2
o
0 t 4
5
2
0
0 1 2 3 4 5 6 7 8 9
20 to 50 keV flux in 10−10 erg cm−2 s−1
Figure3: Thefluxesinthetwohighestenergybandsinthisstudy. Shownintheplotalso(inblue)isthe
regionofaverage±1σ forbothbands. Thisshowstheregion(influx)ofpreferredfluxesforthe(canonical)
highstatewhere1E1740.7–2942iscommonlyobserved.
35
mean=6.70 1E 1740.7−2942
30 median=6.60 5−20 keV flux
25 σ=1.14 102 data points by Jem−X
20
15
10
5
25
1E 1740.7−2942 mean=3.90
20 1100−2 2d0a tkae Vpo filnutxs by Jem−X mσ=e0d.i7a4n=3.80
15
10
y
c 5
n
e 65
u
q 55 1E 1740.7−2942 mean=5.1
Fre 234555 2207−4 5d0a tkae Vpo filnutxs by IBIS mσ=e1d.i6an=5.2
15
5
55 1E 1 740.7 −294 2 27 6 dat a poi nts b y IBIS
50−200 keV flux mean=8.7
45 median=8.6
35 σ=3.4
25
15
5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
−10 −2 −1
Flux (10 erg cm s )
Figure4: Histogramdistributionforfluxesinthe4energybandsconsideredinthisstudy.
5
1E1740.7−2942withINTEGRAL ManuelCastro
50
1E 1740.7−2942
mean=1.48
median=1.48
40
σ=0.22
30
y
c
n
e
qu
e
Fr
20
10
0
1.0 1.2 1.4 1.6 1.8 2.0 2.2
Γ
Figure5: HistogramdistributionforthepowerlawindexΓderivedfromacutoffPLfitinXSPEC.
60 1E 1740.7−2942
INTEGRAL/IBIS data
median=46
mean=49
σ=18
50
40
y
c
n
e
qu
e 30
Fr
20
10
0
20 40 60 80 100
kT, the compTT plasma temperature (keV)
Figure6: HistogramdistributionfortheplasmatemperaturekTderivedfromacompTTfitinXSPEC.
6
1E1740.7−2942withINTEGRAL ManuelCastro
2.90 days
0.5
0.4
1
−
z 3.99 days
H
2
n)
a
e 0.3
m
s/
m
er, (r 0.2
w
o
P
0.1
0
0.00 0.10 0.20 0.30 0.40 0.50
Frequency (days−1)
Figure 7: Lomb-Scargle periodogram of the 50–200keV flux showing two tentative possible periods for
1E1740.7–2942(seetextfordetails).
References
[1] Hertz,P.,Grindlay,J.,TheEinsteingalacticplanesurvey-StatisticalanalysisofthecompleteX-ray
sample,ApJ,278,137(1984)
[2] Syunyaev,R.,Gilfanov,M.,Churazov,E.,etal.,TwoHardX-RaySourcesNeartheGalacticCenter-
theKnownSource1E1740.7-2942andtheNewlyDiscoveredSourceGRS:1758-258,Sov.Astron.
Lett+.17.50(1991)
[3] Bouchet,L.,delSanto,M.,Jourdain,E.,etal.,UnveilingtheHighEnergyTailof1E1740.7-2942
WithINTEGRAL,ApJ,693,1871(2009)
[4] Reynolds,M.,Miller,J.,SuzakuObservationsoftheGalacticCenterMicroquasar1E1740.7-2942,
ApJ,716,1431(2010)
[5] Gallo,E.,Fender,R.,Chandraimagingspectroscopyof1E1740.7-2942,Mont.Not.R.Astron.Soc.,
337,869(2002)
[6] Castro,M.,D’Amico,F.,Braga,J.,etal.,ConfirmingthethermalComptonizationmodelforblack
holeX-rayemissioninthelow-hardstate,A&A,569,A82(2014)
[7] Winkler,G.,Courvoisier,T.J.-L.,DiCocco,G,.etal.,TheINTEGRALmission,A&A,411,L1(2003)
[8] Smith,D.,Heinfl,W.,Swank,J.,etal.,X-RayTimingin1E1740.7-2942andGRS1758-258,ApJ,
489,L51(1997)
[9] Smith,D.,Heindl,W.,Swank,J.,OrbitalandSuperorbitalPeriodsof1E1740.7-2942andGRS
1758-258,ApJ,578,L129(2002)
7
1E1740.7−2942withINTEGRAL ManuelCastro
[10] Remillard,R.,McClintock,J.,X-RayPropertiesofBlack-HoleBinaries,Annu.Rev.Astron.Astr.,44,
49(2006)
[11] RCoreTeam,R:ALanguageandEnvironmentforStatisticalComputing,RFoundationforStatistical
Computing,Vienna,Austria,2008
8