Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://uranus.sai.msu.ru/~seif/apj_706_2_1463.pdf Äàòà èçìåíåíèÿ: Fri Nov 13 16:07:34 2009 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 19:54:08 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: ï ï ï ï ï ï ï ï ï ï ï ï ï ï ï ï ï ï ï

The Astrophysical Journal, 706:1463­1483, 2009 December 1
C

doi:10.1088/0004-637X/706/2/1463

2009. The American Astronomical Society. All rights reserved. Printed in the U.S.A.

DISCOVERY OF PHOTON INDEX SATURATION IN THE BLACK HOLE BINARY GRS 1915+105
Lev Titarchuk1
1

,2 ,3 ,4 ,5

and Elena Seifina6

6

` Dipartimento di Fisica, Universita di Ferrara, Via Saragat 1, I-44100 Ferrara, Italy; titarchuk@fe.infn.it 2 ICRANET Piazzale d. Repubblica 10-12 65122 Pescara, Italy 3 George Mason University Fairfax, VA 22030, USA 4 US Naval Research Laboratory, Code 7655, Washington, DC 20375, USA; Lev.Titarchuk@nrl.navy.mil 5 Goddard Space Flight Center, NASA, Code 663, Greenbelt MD 20771, USA Moscow State University/Sternberg Astronomical Institute, Universitetsky Prospect 13, Moscow, 119992, Russia; seif@sai.msu.ru Received 2009 February 8; accepted 2009 October 19; published 2009 November 12

ABSTRACT We present a study of the correlations between spectral, timing properties, and mass accretion rate observed in X-rays from the Galactic black hole (BH) binary GRS 1915+105 during the transition between hard and soft states. We analyze all transition episodes from this source observed with the Rossi X-ray Timing Explorer, coordinated with Ryle Radio Telescope observations. We show that broadband energy spectra of GRS 1915+105 during all these spectral states can be adequately presented by two bulk motion Comptonization (BMC) components: a hard component (BMC1, photon index 1 = 1.7­3.0) with turnover at high energies and soft thermal component (BMC2, 2 = 2.7­4.2) with characteristic color temperature 1 keV, and the redskewed iron-line (LAOR) component. We also present observable correlations between the index and the normalization of the disk "seed" component. The use of "seed" disk normalization, which is presumably proportional to mass accretion rate in the disk, is crucial to establish the index saturation effect during the transition to the soft state. We discovered the photon index saturation of the soft and hard spectral components at values of 4.2 and 3, respectively. We present a physical model which explains the index­seed photon normalization correlations. We argue that the index saturation effect of the hard component (BMC1) is due to the soft photon Comptonization in the converging inflow close to the BH and that of soft component is due to matter accumulation in the transition layer when mass accretion rate increases. Furthermore, we demonstrate a strong correlation between equivalent width of the iron line and radio flux in GRS 1915+105. In addition to our spectral model components we also find a strong feature of "blackbody (BB)-like" bump whose color temperature is about 4.5 keV in eight observations of the intermediate and soft states. We discuss a possible origin of this "BB-like" emission. Key words: accretion, accretion disks ­ black hole physics ­ radiation mechanisms: non-thermal ­ stars: individual (GRS 1915+105) Online-only material: color figures

1. INTRODUCTION The study of the characteristic changes in spectral and variability properties of X-ray binaries is proved to be a valuable source of information on the physics governing the accretion processes and on the fundamental parameters of black holes (BHs). The simultaneous study of the spectral and timing evolution of a BH source during a state transition has been a subject of many investigations (see references in a review by Remillard & McClintock 2006). Fender & Belloni (2004), hereafter FB04, introduced a classification of the spectral states in GRS 1915 +105 and studied the spectral state evolution. Using X-ray colors (hardness ratio) they introduced three spectral states. State A: in which the strong blackbody (BB)-like component of color temperature 1 keV dominates in the overall spectrum and little time variability is detected. State B: similar to state A but substantial red-noise variability on scales >1 s occurs in this state. State C: the spectra are harder than those in states A and B. Photon indices of the power-law components vary from 1.8 to 2.5. White­red-noise (WRN) variability on scales >1stakes place in this state. Furthermore, FB04 discussed the connection between states A, B, ­C observed in GRS 1915 with the three "canonical" states in black hole candidates (BHCs) also identified by their timing and spectral properties. 1463

At a low-luminosity state, the energy spectrum is dominated by a hard Comptonization component combined (convolved) with a weak thermal component. The spectrum of this low (luminosity) hard state (LHS) is presumably a result of Comptonization (upscattering) of soft photons, originated in a relatively weak accretion disk, off electrons of the hot ambient plasma (see, e.g., Sunyaev & Titarchuk 1980). Variability in LHS is high (fractional rms variability is up to 40%) and presented by a flat-top broken power-law (WRN) shape, accompanied by quasi-periodic oscillations (QPOs) in the range of 0.01­30 Hz, observed as narrow peaks in the power density spectrum (PDS). In high soft state (HSS), a photon spectrum is characterized by a prominent thermal component which is probably a signature of a strong emission coming from a geometrically thin accretion disk. A weak power-law component is also present at the level of not more than 20% of the total source flux. In the HSS, the flat-top variability ceases, QPOs disappear, and the PDS acquires a pure power-law shape. The total variability in HSS is usually about 5% fractional rms. The intermediate state (IS) is a transitional state between LHS and HSS. Note that in addition to LHS, IS, and HSS sometimes very soft state (VSS) is observed in which the BB component is dominant and the power-law component is either very weak or absent at all. The bolometric luminosity in VSS is a factor of 2­3 lower than that in HSS. FB04 concluded that probably all three states A, B, C of GRS 1915+105 are instances of something similar to the HSS/


1464

TITARCHUK & SEIFINA

Vol. 706

IS observed in other BHC systems, associated to the high accretion rate value for this source, although during the hardest intervals LHS might be sometimes reached. We come to the similar conclusions analyzing spectral and timing data from GRS 1915+105 obtained by the Rossi X-ray Timing Explorer (RXTE; see below). Close correlations of nearly periodic variability (QPOs) observed during low-hard and ISs with the photon index of the Comptonization spectral component have been reported in multiple state transitions observed from accreting BHs (see Vignarca et al. 2003; Shaposhnikov & Titarchuk 2006, 2007, 2009, hereafter V03, ST06, ST07, and ST09, respectively). The ubiquitous nature of these correlations suggests that the underlying physical processes which lead to the observed variability properties are closely tied to the corona; furthermore, they vary in a well-defined manner as the source makes a transition between spectral states. The fact that the same correlations are seen in many sources, which vary widely in both luminosity (presumably with mass accretion rate) and state, suggests that the physical conditions controlling the index and the low-frequency QPOs (LFQPOs) are characteristics of these sources. Moreover, they may be an universal property of all accreting compact systems, including neutron sources too (see Titarchuk, & Fiorito 2004, hereafter TF04 and Titarchuk & Shaposhnikov 2005). When a BH is in LHS, radio emission is also detected and a jet is either seen or inferred (Fender 2001). Several models are successful in reproducing the energy spectrum from the radio domain to the hard X-rays (see, e.g., Markoff et al. 2001; Vadawale et al. 2001; Corbel & Fender 2002; Markoff et al. 2003; and Giannios 2005). The multiplicity of models that can fit well the time average spectrum of Galactic BHs indicates that this alone is not enough to distinguish the most realistic one among them. X-ray timing features can be the key features to finally finding the common physical connection between the corona, the accretion disk, and the jet radio emission in BHs. There is a big debate in the literature on the origin of QPO frequencies (see, e.g., Remillard & McClintock 2006) and its connection with the radio emission. Migliari et al. (2005) reported on correlation between radio luminosity and X-ray timing features in X-ray binaries containing a number of low-magnetic-field neutron stars and one BH GX 339-4. They showed that in the LHS, radio luminosity is correlated with the LFQPO. Note that ST09 demonstrated that in the LHS of Galactic BHs, LFQPO changes by order of magnitude, from 0.2 to 2 Hz whereas the photon index has almost the same value of 1.5. Below we show that in GRS 1915+105 the photon index monotonically increases with LFQPO and disk mass accretion rate, although the radio luminosity does not correlate with LFQPO and X-ray luminosity in the whole range of spectral states, from low hard to high soft through ISs. Recently, Kylafis et al. (2008) suggested a model which explains how the QPO phenomenon is related to an appearance of radio flares (jets). Below (see Section 3) we present details of our observational study of the QPO connection with the X-ray and radio flaring activity in GRS 1915+105. In LHS and IS which we consider in our study, only a small part of the disk emission component is seen directly. The energy spectrum is dominated by a Comptonization component presented by a power law. To calculate the total normalization of the "seed" disk BB component we model the spectrum with a generic Comptonization model (BMC XSPEC model, see details in Titarchuk et al. 1997) which consistently convolves a disk BB with a Green's function of the Compton corona to

produce the Comptonization component. We argue that the disk emission normalization calculated using this approach produces a more accurate correlation with respect to the correlation with the direct disk component which was obtained using the additive model, multicolor disk plus power law (see, e.g., McClintock & Remillard 2006). This paper is a continuation of the study of index­QPO and index­seed photon normalization correlations in BH sources started in ST07 and ST09. Particularly, here we present a study of the index­seed photon normalization (disk flux) correlation observed from GRS 1915+105 when it evolves from LHS to HSS. The description of RXTE data set used is given in Section 2. We have analyzed a broader sample of state transitions from GRS 1915+105 and we found a diverse phenomenology for index evolution through a transition. In Section 3, we provide a detailed description of state transitions analyzed in this study. In Section 4, we discuss and interpret the results of our observational study. Specifically in Section 4, we consider the effect of the bulk motion Comptonization (BMC) in the inner part of the accretion flow on the index evolution during a state transition. Also we show that the index saturation effect is a direct consequence of the existence of this inner BM region and, therefore, can be considered as an observational signature of the converging flow (BH). Furthermore, in Section 4 we discuss the TF04 model and the Monte Carlo simulations by P. Laurent & L. Titarchuk (2009, in preparation) in which the observable index evolution with m has been already predicted. Conclusions follow in Section 5. 2. OBSERVATIONS AND DATA REDUCTION In the present Paper, we have used publicly available data of the RXTE Observatory obtained from 1997 January to 2006 April. In total, our study includes 107 observations made at different BH spectral states (LHS, IS, HSS) of the system. Data sets were selected to represent a complete rise-middledecay track of bright X-ray activity episodes behavior along bright radio flaring events (S15 GHz 250 mJy). Therefore, we have chosen powerful ( 250 all-sky monitor counts s-1 ) flaring episodes of GRS 1915+105 with a good coverage of simultaneous radio/X-ray observation. In the past, some of these data of spectral transitions in GRS 1915+105 were analyzed by Trudolyubov et al. (1999), Trudolyubov (2001), Muno et al. (1999), Reig et al. (2000), ST07, and Rodriguez et al. (2008) for the 1997­1998 and 2005­2006 transitions, respectively. Standard tasks of the LHEASOFT/FTOOLS 5.3 software package were utilized for data processing using methods recommended by RXTE Guest Observer Facility according to The RXTE Cook Book (http://heasarc.gsfc.nasa.gov/docs/ xte/recipes/cook_book.html). For spectral analysis, we used Proportional Counter Array (PCA) Standard 2 mode data, collected in the 3­20 keV energy range. The standard dead-time correction procedure has been applied to the data. To construct broadband spectra, data from High-Energy X-Ray Timing Experiment (HEXTE) detectors have also been used. We subtracted background corrected in off-source observations. Only HEXTE data in the 20­150 keV energy range were used for the spectral analysis in order to exclude the channels with largest uncertainties. The HEXTE data have been re-normalized based on the PCA. The data are available through the Goddard Space Flight Center (GSFC) public archive (http://heasarc.gsfc.nasa.gov). In Tables 1­6, we list groups of observations covering the complete dynamical range LHS­(IS)­HSS­(IS)­LHS of the source evolution during flaring events. We present here period ranges


No. 2, 2009

DISCOVERY OF PHOTON INDEX SATURATION IN THE BB BINARY GRS 1915+105

1465

Table 1 Best-fit Parameters of Spectral Analysis of PCA and HEXTE Observation of GRS 1915+105 in 3­150 keV Energy Range During Rise 1997 Transitiona Observational ID 20402-01-11-00 20402-01-12-00 20402-01-13-00 20402-01-16-00 20402-01-19-00 20402-01-20-00 20402-01-21-00 20402-01-21-01 20402-01-23-00 20402-01-24-00 20402-01-28-00 20402-01-29-00 20402-01-30-01 20402-01-31-00 20402-01-31-02 20402-01-33-00 20402-01-34-01 20402-01-35-00 20402-01-36-00 20402-01-37-01 20402-01-38-01 20402-01-39-02 20402-01-41-01 20402-01-41-02 20402-01-43-00 20402-01-45-00 MJD (day) 50462.06 50471.06 50477.87 50501.88 50517.04 50524.92 50533.83 50534.90 50548.45 50561.13 50586.68 50589.49 50596.20 50602.53 50604.61 50617.54 50621.80 50636.62 50639.62 50641.48 50649.42 50658.51 50679.30 50679.37 50688.24 50698.65 1 = 1 - 1 0.867(9) 0.844(7) 0.92(2) 0.83(1) 0.62(1) 0.85(1) 0.94(1) 0.95(1) 0.91(3) 0.72(3) 1.59(2) 0.95(2) 1.59(1) 1.79(2) 1.80(5) 1.99(6) 1.92(4) 1.85(6) 1.99(9) 1.90(9) 1.9(1) 2.0(1) 2.0(1) 2.00(9) 2.00(5) 1.90(1) log (A1 )
b

Nbmc1 , 2 L39 /d10 0.115(3) 0.1105(2) 0.116(1) 0.1014(5) 0.101(1) 0.092(2) 0.100(1) 0.101(1) 0.112(2) 0.116(5) 0.2001(2) 0.159(1) 0.200(8) 0.223 (5) 0.218(7) 0.22(4) 0.22(1) 0.24(4) 0.20(1) 0.245(9) 0.370(9) 0.26(2) 0.417(6) 0.459(2) 0.342(5) 0.231(2)

2 = 2 - 1 1.84(7) 1.9(2) 1.9(3) 2.0(4) 1.8(2) 1.9(3) 2.0(4) 2.0(2) 2.0(1) 2.0(3) 2.34(3) 2.0(2) 2.3(1) 2.6(1) 2.6(1) 3.2(2) 3.1(1) 3.10(9) 3.2(1) 3.09(9) 3.20(5) 3.2(3) 3.15(5) 3.20(1) 3.20(7) 3.20(2)

log (A2 )

b

Nbmc2 , 2 L39 /d10 0.009(3) 0.009(2) 0.002(5) 0.008(2) 0.002(1) 0.01(2) 0.009(5) 0.009(2) 0.01(3) 0.01(4) 0.109(1) 0.01(3) 0.110(2) 0.118(5) 0.127(6) 0.24(1) 0.17(1) 0.21(5) 0.256(5) 0.24(1) 0.30(1) 0.303(2) 0.422(8) 0.429(1) 0.236(3) 0.236(2)

EW (eV) ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... 63 (8) 61(10) 78(10) 82(10) 124(14) 151(13)

Ela or (keV) ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... 6.1 ± 6.1 ± 6.1 ± 6.09 ± 6.12 ± 5.78 ±

Fluxc 13.59 13.28 13.12 12.01 12.16 11.51 11.34 11.30 11.55 14.27 18.59 18.00 20.45 23.56 23.67 30.21 27.18 31.54 32.24 34.38 38.59 38.15 46.47 49.90 41.87 31.22



2 red

(dof) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73)

0.106(4) 0.102(3) 0.030(6) 0.100(6) 0.137(2) 0.077(8) 0.050(3) 0.025(3) -0.082(8) 0.095(4) -0.178(8) 0.033(3) 0.32(2) 0.07(4) -0.11(8) 0.13(6) 0.49(9) 0.030(3) -0.40(1) -0.26(9) -1.66(8) -0.58(7) -1.82(9) -1.40(8) -0.19(6) -0.22(9)

2.0 2. 0 2. 0 2. 0 2. 0 2. 0 2. 0 2. 0 2. 0 2. 0 2.0 2. 0 2. 0 0.09(6) 0.8(1) 0.18(1) -0.18(2) 2.0 2.0 2.0 2.0 2. 0 0.7(1) 2.0 2.0 0.47(5)

0. 7 1. 5 1. 0 0. 8 0. 6 1. 4

1.09 1.06 1.01 1.07 1.02 1.05 1.09 1.06 1.09 1.08 1.07 1.07 1.03 1.01 1.08 1.01 1.07 0.97 0.97 1.03 1.10 1.12 1.10 1.08 1.01 0.96

Notes. Parameter errors (put in parentheses) correspond to 1 confidence level. a The spectral model of the continuum is wabs (bmc + bmc highecut + laor ). b When the parameter log (A) > 1.0, this parameter is fixed to a value 2.0 (see comments in the text). c Spectral flux in the 3­150 energy range in units of â10-9 erg s-1 cm-2 . Table 2 Best-fit Parameters of Spectral Analysis of PCA and HEXTE Observation of GRS 1915+105 in 3­150 keV Energy Range During Middle 1997 Transition Observational ID 20402-01-45-02 20402-01-45-03 20402-01-46-00 20402-01-48-00 20402-01-50-00 20402-01-50-01 20402-01-51-00 20402-01-52-01 20402-01-52-02 20402-01-53-00 20402-01-53-02 20402-01-54-00 20402-01-55-00 MJD (day) 50696.21 50700.25 50703.41 50720.59 50735.54 50737.40 50743.29 50751.68 50751.75 50752.01 50756.41 50763.20 50769.22 1 = 1 - 1 1.84(1) 1.96(4) 2.00(7) 2.00(7) 1.74(7) 1.720(2) 1.704(2) 2.00(4) 2.00(8) 2.00(2) 2.00(2) 1.99(8) 0.8(3) log (A1 )
a

Nbmc1 , 2 L39 /d10 0.33(1) 0.23(4) 0.25(2) 0.240(6) 0.222(8) 0.236(1) 0.226(1) 0.240(6) 0.320(2) 0.32(1) 0.39(1) 0.412(6) 0.01(6)

2 = 2 - 1 2.98(2) 3.19(6) 3.1(2) 3.09(3) 3.2(2) 1.9(4) 2.0(3) 3.2(1) 3.2(1) 3.00(7) 3.20(1) 3.20(3) 3.15(3)

log (A2 )

a

Nbmc2 , 2 L39 /d10 0.379(2) 0.245(8) 0.13(2) 0.24(1) 0.323(9) 0.02(2) 0.04(5) 0.271(6) 0.482(5) 0.43(1) 0.32(2) 0.369(7) 0.426(4)

EW (eV) 133(5) 192(8) 87(10) 164(13) 125(19) 102(16) 102(13) 348(19) 342(20) 315(10) 53(12) 58(16) 101(13)

Ela or (keV) 6.31 6.10 6.25 6.46 6.48 6.47 6.47 6.31 6.44 6.10 6.18 6.19 6.25 ± ± ± ± ± ± ± ± ± ± ± ± ± 0.08 0.08 0.08 0.05 0.01 0.04 0.03 0.01 0.01 0.01 0.07 0.08 0.09

Fluxb 39.45 33.49 26.96 39.44 20.69 19.19 20.56 34.60 49.68 48.71 43.76 45.09 35.88



2 red

(dof) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73)

0.35(9) -0.1(2) 0.02(1) 0.77(4) 0.16(5) 2.0 1.9(6) 0.15(9) 0.06(1) -0.46(1) -0.45(3) 2.0 0.08(6)

0.56(1) 0.49(2) 0.8(4) 0.15(2) 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 0.322(1)

1.05 1.01 1.09 1.08 1.02 1.02 1.09 1.02 1.04 0.99 1.01 1.02 1.02

Notes. Parameter errors (put in parentheses) correspond to 1 confidence level. a When the parameter log (A) > 1.0, this parameter is fixed to a value 2.0 (see comments in the text). b Spectral flux in the 3­150 energy range in units of â10-9 erg s-1 cm-2 .

MJD = 50462­51081 and MJD = 53382­53852, as different types (samples) of bright X-ray activity, with transitions between hard and soft states. Two selected data sets have different patterns of radio/X-ray behavior and of light-curve shapes. We also use public GRS 1915+105 data from the ASM on-board RXTE (Swank 1999). The ASM light curves (2­ 12 keV energy range) were retrieved from the public RXTE/ ASM archive at HEASARC (http://xte.mit.edu/ASM_lc.html).

The monitoring Ryle Radio Telescope (RT; 15 GHz) data during the 1997­2006 period were kindly provided by Guy Pooley. The technical details of the RT are described by Pooley & Fender (1997). 2.1. Spectral Analysis
2.1.1. BMC and Iron Line Components of the Model Spectrum

The broadband source spectra were modeled in XSPEC with an additive model consisting of two BMC: a BMC with


1466

TITARCHUK & SEIFINA

Vol. 706

Table 3 Best-fit Parameters of Spectral Analysis of PCA and HEXTE Observation of GRS 1915+105 in 3­150 keV Energy Range During Decay 1997 Transition Observational ID 30184-01-01-00 30703-01-14-00 30402-01-09-00c 30402-01-10-00 30402-01-11-00 30703-01-15-00 30703-01-16-00 30402-01-12-01 30182-01-02-00 30182-01-04-01 30402-01-17-00 30703-01-33-00 30703-01-35-00 MJD (day) 50908.00 50909.87 50912.88 50914.39 50923.26 50925.88 50931.67 50945.01 51003.21 51006.21 51067.62 51071.90 51081.81 1 = 1 - 1 1.81(2) 1.74(1) 1.50(2) 1.50(1) 2.00(3) 1.79(5) 1.79(4) 1.69(2) 1.801(2) 1.0(4) 1.520(3) 1.141(7) 0.88(3) log (A1 )
a

Nbmc1 , 2 L39 /d10 0.260(1) 0.236(1) 0.20(3) 0.1949(8) 0.24(1) 0.22(3) 0.21(3) 0.206(4) 0.223(2) 0.169(8) 0.200(1) 0.172(1) 0.152(2)

2 = 2 - 1 2.0(3) 2.0(5) 2.09(5) 2.09(2) 3.2(1) 2.1(1) 2.10(9) 2.10(8) 2.0(3) 2.49(6) 2.1(3) 2.0(4) 2.1(2)

log (A2 )

a

Nbmc2 , 2 L39 /d10 0.009(1) 0.007(2) 0.11(1) 0.111(9) 0.23(3) 0.10(1) 0.09(2) 0.094(2) 0.009(1) 0.125(2) 0.009(2) 0.002(3) 0.009(1)

EW (eV) 284(18) 317(20) 290(18) 411(30) 308(20) 316(10) 326(8) 290(10) 434(11) ... ... ... ...

Ela or (keV) 5.75 6.39 5.7 5.7 5.6 6.4 6.4 5.7 6.41 ± ± ± ± ± ± ± ± ± ... ... ... ... 1. 2 0.05 0. 3 0. 4 0. 2 0. 2 0. 3 0. 5 0.01

Fluxb 24.90 22.85 23.89 22.92 31.28 23.79 23.52 22.17 26.11 24.23 21.15 19.52 18.18



2 red

(dof) (73) (73) (71) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73)

0.32(2) 0.55(2) -0.32(9) -0.32(8) 1.0(2) 0.37(6) 0.37(6) 0.42(3) 2.0 -1.953(8) 0.6(1) 0.19(1) 0.18(4)

2. 0 2. 0 2.0 2.0 2. 0 2. 0 2.0 -0.5(1) 2. 0 2.0 2. 0 2. 0 2. 0

1.09 1.30 1.04 1.02 1.23 1.01 1.07 1.06 0.99 1.03 0.98 1.07 1.02

Notes. Parameter errors (put in parentheses) correspond to 1 confidence a When the parameter log (A) > 1.0, this parameter is fixed to a value 2.0 b Spectral flux in the 3­150 energy range in units of â10-9 erg s-1 cm-2 c the data are fitted with the wabs*(bmc + bmc*highecut + laor + bbody)

level. (see comments in the text). . model; see values of the best-fit BB color temperature and EW in Table 7.

Table 4 Best-fit Parameters of Spectral Analysis of PCA and HEXTE Observation of GRS 1915+105 in 3­150 keV Energy Range During Rise 2005 Transition Observational ID 80701-01-48-00 90701-01-46-00 80701-01-37-00 90701-01-49-00 90105-05-03-00 90105-05-03-04 90105-05-03-05c 91701-01-04-00c 90105-07-01-00 90105-07-02-00 90105-07-03-00 91701-01-08-00 91102-01-01-00 91412-01-01-00 91701-01-10-00 91701-01-10-01 90105-08-02-00 90105-08-03-00 91701-01-11-00 91701-01-12-00 91701-01-13-00 91701-01-17-00 91701-01-19-00 91701-01-20-00 91701-01-24-00 91701-01-25-00 90105-04-01-00 90105-04-03-00 90105-04-03-01 MJD (day) 53382.39 53400.35 53416.34 53422.30 53442.98 53444.01 53444.08 53456.28 53472.92 53473.05 53473.97 53486.12 53488.72 53500.32 53501.06 53501.12 53503.87 53504.71 53508.06 53515.07 53520.05 53547.93 53562.92 53570.78 53600.89 53606.84 53640.38 53641.43 53641.51 1 = 1 - 1 1.78(5) 2.00(1) 2.00(4) 2.00(1) 1.9(1) 1.981(6) 1.982(4) 1.981(5) 1.980(6) 1.981(5) 1.98(1) 1.96(6) 1.60(5) 1.99(2) 2.0(1) 1.9(1) 1.90(3) 2.00(1) 1.70(3) 0.8(1) 0.8(1) 0.7(1) 0.7(1) 0.7(2) 0.7(1) 0.7(2) 0.7(1) 1.8(6) 1.8(7) log (A1 )
a

Nbmc1 , 2 L39 /d10 0.125(1) 0.292(4) 0.164(1) 0.152(4) 0.204 (3) 0.191(3) 0.187(1) 0.185(2) 0.144(1) 0.147(5) 0.147(2) 0.16 (3) 0.123(3) 0.261(1) 0.338(2) 0.210(4) 0.358(1) 0.242(2) 0.343(2) 0.001 (5) 0.05 (6) 0.06 (2) 0.09 (5) 0.05 (3) 0.123 (6) 0.09 (5) 0.04 (5) 0.215(4) 0.215(4)

2 = 2 - 1 2.1(4) 2.90(1) 2.92(3) 3.00(2) 2.92(4) 2.7(0.1) 2.79(6) 2.79(9) 2.85(6) 2.80(6) 2.80(8) 3.1(2) 3.0(1) 3.10(1) 3.20(2) 3.20(2) 3.20(1) 3.20(1) 3.10(1) 3.197(7) 3.200(3) 2.75(3) 3.20(1) 2.57(1) 2.75(9) 2.75(7) 2.16(2) 3.00(1) 3.00(1)

log (A2 )

a

Nbmc2 , 2 L39 /d10 0.009(2) 0.335(1) 0.157(4) 0.158(4) 0.1574(6) 0.111(2) 0.112(1) 0.110(1) 0.112(1) 0.133(3) 0.135(8) 0.38(3) 0.4705(5) 0.3400(9) 0.456(1) 0.433(4) 0.4901(6) 0.278(1) 0.451(1) 0.3480(1) 0.430(1) 0.193(1) 0.4633(7) 0.151(1) 0.124(1) 0.132(1) 0.1335(4) 0.282(3) 0.284(3)

EW (eV) ... 91(10) 594(10) 626(15) 435(15) 402(12) 378(10) 256(15) 55.6(10) 75.3(6) 210(8) 132(11) 74(12) 50(16) ... 68(9) 104(10) 180(12) 69(10) 90(10) 120(10) 95(10) 100(10) 125(10) 100(10) 100(10) 127(5) 114(12) 108(11)

Ela or (keV) 6.35 5.5 6.56 6.45 6.45 6.46 6.33 6.57 6.55 6.39 6.13 6.32 6.19 6.10 6.3 6.11 6.13 6.4 6.4 7.1 7.0 7.0 7.0 7.1 6.40 6.11 6.32 ... ± ± ± ± ± ± ± ± ± ± ± ± ± ... ± ± ± ± ± ± ± ± ± ± ± ± ± ± 0.01 0. 2 0.02 0.01 0.01 0.01 0.01 0.01 0.07 0.01 0.04 0.08 0.09 0.04 0. 1 0.01 0.04 0. 1 0. 1 0. 2 0. 1 0. 3 0. 3 0. 2 0.09 0.07 0.01

Fluxb 6.25 29.94 24.03 22.30 24.37 23.08 22.67 22.69 22.07 22.36 22.61 40.68 38.69 39.35 51.99 45.30 57.53 37.13 56.21 24.18 37.02 11.24 23.69 69.91 5.90 6.22 6.64 33.96 32.26



2 red

(dof) (73) (73) (73) (73) (73) (73) (71) (71) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73)

-0.96(1) -0.21(2) 2. 0 0.26(9) 2.0 2.0 2.0 2.0 2.0 1.0(2) 2. 0 2. 0 -0.47(7) -1.10(8) -1.11(9) -1.06 (8) -0.86 (6) -0.86(7) -1.7(9) 0.3(1) 0.9(4) 2.0 -0.9(5) -0.9(3) -0.95(5) -0.7(3) 0.4(1) 0.70(4) 0.48(4)

2.0 1.1(4) 2. 0 0.75(5) 2. 0 0.880(2) 0.88(5) 0.884(1) 0.251(1) 0.272(3) 2. 0 -0.02(3) -0.0030(5) 0.235(9) 0.34(1) 2. 0 0.802(8) 0.8(2) 2. 0 2.0 2.0 -1.106(1) 0.256(7) -0.537(8) -0.627(1) -0.626(1) -0.530(4) 1.1(2) 1.1(1)

1.05 0.96 1.03 1.01 1.06 1.09 0.96 0.97 0.99 0.99 0.96 1.01 1.08 1.02 1.11 1.39 1.09 1.02 1.04 1.09 1.10 0.99 1.02 1.18 1.24 1.25 1.35 1.08 1.01

Notes. Parameter errors (put in parentheses) correspond to 1 confidence level. a When the parameter log (A) > 1.0, this parameter is fixed to a value 2.0 (see comments in the text). b Spectral flux in the 3­150 energy range in units of â10-9 erg s-1 cm-2 . c These data are fitted with the wabs*(bmc*highecut + bmc + laor + bbody) model; see values of the best-fit parameters in Table 7.

high-energy cutoff (BMC1 component) and BMC2 component: wabs (bmc + bmc highecut). We also use a multiplicative wabs model taking into account an absorption by neutral material. The wabs model parameter is an equivalent hydrogen column NH .A

systematic error of 1% has been applied to the analyzed X-ray spectra. The BMC model describes the outgoing spectrum as a convolution of the input "seed" BB-like spectrum, whose


No. 2, 2009

DISCOVERY OF PHOTON INDEX SATURATION IN THE BB BINARY GRS 1915+105

1467

Table 5 Best-fit Parameters of Spectral Analysis of PCA and HEXTE Observation of GRS 1915+105 in 3­150 keV Energy Range During Middle 2005 Transition Observational ID 91701-01-31-00 91701-01-33-01 91701-01-33-00 91701-01-34-00 91701-01-34-01 91701-01-35-00 90105-06-03-01 90105-06-03-00 90105-06-03-02 91701-01-38-00 91701-01-38-01 91701-01-39-00 91701-01-39-01 92092-01-01-01 92092-02-01-01 92092-03-01-01 91701-01-39-01 91701-01-41-01 91701-01-42-00 91701-01-42-01c 91701-01-43-00c MJD (day) 53646.76 53659.98 53661.70 53669.69 53669.75 53674.59 53694.90 53695.03 53695.30 53696.56 53696.63 53703.57 53703.63 53704.94 53706.00 53707.68 53711.55 53718.56 53723.47 53723.54 53730.41 1 = 1 - 1 2.00(2) 1.60(1) 1.99(1) 1.42(1) 1.7(2) 1.50(5) 1.70(3) 1.89(2) 1.7(1) 1.80(8) 2.07(3) 1.70(2) 1.734(7) 1.74(3) 1.91(2) 1.7(3) 1.8(1) 1.77(2) 1.701(5) 1.91(1) 1.91(4) log (A1 )
a

Nbmc1 , 2 L39 /d10 0.17(1) 0.087(3) 0.112(2) 0.055(1) 0.084(2) 0.116(2) 0.345(1) 0.261(4) 0.282(6) 0.35(1) 0.685(4) 0.159(3) 0.170(1) 0.195(3) 0.116(3) 0.36(2) 0.23(2) 0.184(1) 0.111(2) 0.129(3) 0.15(2)

2 = 2 - 1 3.0(1) 2.90(6) 2.78(1) 2.51(1) 1.98(4) 3.00(2) 3.200(1) 3.20(1) 3.18(3) 3.12(2) 2.7(2) 2.90(1) 2.98(3) 2.9(1) 2.9(1) 3.00(6) 2.9(1) 2.1(5) 2.90(6) 2.70(6) 2.7(2)

log (A2 )

a

Nbmc2 , 2 L39 /d10 0.176(8) 0.143(2) 0.403(1) 0.189(1) 0.185(1) 0.3127(6) 0.552(1) 0.417(5) 0.389(5) 0.479(1) 0.290(3) 0.273(4) 0.163(1) 0.256(2) 0.229(1) 0.49(1) 0.26(2) 0.009(1) 0.110(1) 0.1644(5) 0.14(9)

EW (eV) ... 110(13) 92(8) 135(10) ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...

Ela or (keV) ... 6.35 ± 0.07 6.30 ± 0.01 6.32 ± 0.01 ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...

Fluxb 22.77 25.55 36.88 21.21 21.14 23.31 62.96 49.38 48.40 51.76 57.06 31.30 27.10 28.45 22.75 56.72 34.48 19.95 20.57 20.57 20.29



2 red

(dof) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (71) (71)

2.0 2.0 2.0 2.0 -0.57(7) -0.55(2) -1.4(2) -0.56(7) -0.70(8) -1.1(1) 0.68(8) -0.22(9) 0.35(2) 0.05(1) 2.0 -1.6(1) -0.33(5) 0.85(3) 0.8(1) 2.0 2.0

2. 0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 0.235(1) 2. 0 2.0 0.471(1) 2. 0 -0.363(1) 0.470(1) 2. 0 2. 0 2.0 2.0 2. 0

1.03 1.03 0.99 1.08 1.04 1.01 1.02 1.01 1.07 0.99 0.98 1.11 1.05 1.05 0.95 1.08 0.96 1.06 1.13 1.05 0.91

Notes. Parameter errors (put in parentheses) correspond to 1 confidence level. a When the parameter log (A) > 1.0, this parameter is fixed to a value 2.0 (see comments in the text), b Spectral flux in the 3­150 energy range in units of â10-9 erg s-1 cm-2 . c These data are fitted with the wabs*(bmc*highecut + bmc + laor + bbody) model; see values of the best-fit parameters in Table 7. Table 6 Best-fit Parameters of Spectral Analysis of PCA and HEXTE Observation of GRS 1915+105 in 3­150 keV Energy Range During Decay 2005­2006 Transition Observational ID 91701-01-46-00c 91701-01-49-00c 91701-01-50-00c 91701-01-22-00 92702-01-01-00 92702-01-01-01 92702-01-02-01 92702-01-03-00 92702-01-05-00 92082-01-05-00 92702-01-07-00 92702-01-08-00 92702-01-08-01 92702-01-08-02 90105-02-04-00 MJD (day) 53753.43 53771.51 53778.30 53794.31 53803.28 53803.35 53809.30 53815.19 53829.20 53834.84 53844.16 53851.10 53851.17 53851.23 53852.15 1 = 1 - 1 1.70(1) 1.69(1) 1.65(1) 1.68(2) 1.40(2) 1.50(3) 1.353(9) 1.49(2) 0.75(3) 1.03(1) 1.10(2) 0.94(3) 1.13(1) 1.04(4) 1.09(2) log (A1 )
a

Nbmc1 2 L39 /d10 0.284(7) 0.221(8) 0.143(2) 0.143(2) 0.12(1) 0.11(1) 0.085(2) 0.094(4) 0.125(1) 0.140(2) 0.163(1) 0.127(2) 0.132(1) 0.128(1) 0.125(1)

2 = 2 - 1 3.00(2) 2.41(3) 2.54(7) 2.00(7) 1.8(2) 1.8(1) 1.74(5) 1.21(6) 1.9(5) 2.0(6) 1.9(4) 2.0(3) 2.0(4) 1.9(3) 1.8(6)

log (A2 )

a

Nbmc2 , 2 L39 /d10 0.3807(8) 0.0978(2) 0.1055(5) 0.081(2) 0.083(4) 0.081(7) 0.0798(4) 0.063(2) 0.005(3) 0.009(2) 0.005(3) 0.01(2) 0.008(3) 0.01(3) 0.009(2)

EW (eV) 108(10) ... ... ... ... ... ... ... ... ... ... ... ... ... ...

Ela or (keV) 5.15 ± 0.06 ... ... ... ... ... ... ... ... ... ... ... ... ... ...

Fluxb 38.60 16.00 17.48 18.60 19.35 19.11 16.64 17.46 16.98 16.05 14.92 14.50 14.74 14.03 14.28



2 red

(dof) (71) (71) (71) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73) (73)

0.6(1) 2.0 2.0 2.0 2.0 2.0 2.0 2.0 0.14(1) 0.13(1) -0.15(2) 0.06(1) 0.09(1) 0.05(2) 0.15(1)

2. 0 2. 0 0.38(1) -0