A New Black Hole Mass Estimate for Obscured Active Galactic Nuclei

0.19 MB·English

by  Takeo Minezaki

PreprinttypesetusingLATEXstyleemulateapjv.5/2/11
Online-only material: color figures
A NEW BLACK HOLE MASS ESTIMATE FOR OBSCURED ACTIVE GALACTIC NUCLEI
Takeo Minezaki1 and Kyoko Matsushita2
ABSTRACT
We propose a new method for estimating the mass of a supermassive black hole, applicable to
obscuredActive GalacticNuclei(AGNs). Thismethod estimatesthe blackholemassusingthe width
5
of the narrowcore ofthe neutralFeKα emission line in X-raysand the distance of its emitting region
1
0 from the black hole based on the isotropic luminosity indicator via the luminosity scaling relation.
2 We collect the line width data of the neutral FeKα line core for seven type-1 AGNs and seven type-2
AGNs obtained by the Chandra High Energy Transmission Grating Spectrometer, which affords the
n
best spectral resolution currently available. Assuming the virial relation between the locations and
a
the velocity widths of the neutral FeKα line core and the broad Hβ emission line, the luminosity
J
scalingrelationoftheneutralFeKαlinecoreemittingregionisestimated. Wefindthatthefullwidth
9
at half maximum of the neutral FeKα line core falls between that of the broadBalmer emission lines
2
and the corresponding value at the dust reverberation radius for most of the type-1 AGNs and for

all of the type-2 AGNs. This suggests that significant fraction of photons of the neutral FeKα line
]
A core originates between the outer BLR and the inner dust torus in most cases. The black hole mass
M estimated with this method is then compared with other black hole mass estimates, such
G BH,FeKα
asthebroademission-linereverberationmassM forthetype-1AGNs,themassM based
BH,rev BH,H2O
. on the H O maser and the single-epoch mass estimate M based on the polarized broad Balmer
h 2 BH,pol
lines for the type-2 AGNs. We find that M is consistent with M for the most of the
p BH,FeKα BH,rev
- type-1 AGNs and with MBH,pol for all of the type-2 AGNs. We also find that MBH,FeKα is correlated
o wellwithM forthetype-2AGNs. TheseresultssuggestthatM isapotentialindicator
BH,H2O BH,FeKα
r ofthe blackhole massespeciallyfor obscuredAGNs. Incontrast,M forwhichthe same virial
t BH,FeKα
s factorasforM andM isadoptedis systematicallylargerthanM by aboutafactor
a of ∼5, and theBpHo,rsesvible origBinHs,paolre discussed. BH,H2O
[
Subject headings: black hole physics — galaxies: nuclei — galaxies: active — galaxies: Seyfert —

1 X-rays: galaxies — polarization
v
2
2 1. INTRODUCTION scured AGNs is important; however, the precise mea-
5 surement is not straightforward. The position-velocity
The correlation between the mass of the supermas-
7 diagram of the molecular disk surrounding the black
sive black holes in the centers of galaxies and the
0 hole, obtained by the VLBI observation of the H O
bulge stellar mass or the velocity dispersion (e.g., 2
1. Kormendy & Richstone 1995; Magorrianet al. 1998; maser, is considered to be one of the most precise
0 Kormendy & Ho 2013) strongly suggestthat the growth estimate for the black hole mass (e.g., Miyoshi et al.
5 of the central supermassive black hole and the evolu- 1995). However, precise alignment between the equa-
1 torialplane of the molecular disk and the observer’sline
tion of the galaxies are connected. Obscured Active
: of sight is required for the disk H O maser to be ob-
v GalacticNuclei(AGNs)areimportanttargetsforstudy- 2
servable; thus the number of applicable targets is very
i ing the AGN-galaxy coevolution because a large num-
X limited. The single-epoch mass estimate based on the
ber of host galaxies can be investigated in detail being
luminosity scaling relation of the broad emission-line
r free from the strong central emission from the AGN it-
a region (BLR) (e.g., Laor 1998; McLure & Jarvis 2002;
self. Moreover, they are of special importance because
Vestergaard& Peterson 2006) measuring the widths of
some types of obscured AGNs such as the ultralumi-
the polarized broad Balmer emission lines would be
nous infrared galaxies (ULIRGs) have long been antic-
widely available for type-2 AGNs. However, a signifi-
ipated to be in the middle of the evolutionary sequence
cantfractionoftype-2AGNsdoesnotpresentthepolar-
of AGNs (e.g., Sanders et al. 1988; Sanders & Mirabel
izedbroadBalmeremissionlines(Tran2001). Moreover,
1996; Hopkins et al. 2008). The “new type AGNs” re-
this type of measurement might be unavailable for the
centlyfoundbyUeda et al.(2007),wherethecentralen-
ULIRGs and the new-type AGNs, whose BLR is largely
gine is thought to be deeply obscured by a dust torus
covered by obscuring materials. A precise estimate of
with an unusually smallopening angle,are also interest-
the black hole mass that is widely availablefor obscured
ing for its nature and role in the AGN evolution.
AGNs is desired.
Consequently, measuring the black hole mass in ob-
The fluorescent iron Kα line (hereafter, FeKα
line) is an almost ubiquitous emission line ob-
1Institute of Astronomy, School of Science, University servable for both obscured and unobscured AGNs
of Tokyo, 2-21-1 Osawa, Mitaka, Tokyo 181-0015, Japan;
(Yaqoob & Padmanabhan 2004; Nandra et al. 2007;
[email protected]
2Department of Physics, Tokyo University of Science, 1-3 Fukazawa et al.2011). Therest-frameenergyofthenar-
Kagurazaka, Shinjuku-ku,Tokyo162-8601, Japan row core of the FeKα emission line is close to that of
2 Minezaki et al.
neutralorlow-ionization-stateirons. Consideringitsnar- target AGNs and their data are listed in Table 1.
rowlinewidth,theemittingregionissupposedtobenot For the type-2 AGNs, four target AGNs whose black
very close to the black hole, such as the outer parts of holemassesaremeasuredbytheVLBIobservationofthe
the accretion disk and more distant matter. The large H OmaserareselectedoutofeightAGNs,forwhichthe
2
equivalent width indicates both large covering fraction FWHM of the neutral FeKα line core was measured by
and large optical depth for the emitting region, which Shu et al.(2011). In addition,onetype-2 AGNfromthe
suggests that the neutral FeKα line core originates in 12AGNsofShu et al.(2010)andfivetype-2AGNsfrom
the dust torus surrounding the central engine in accor- the eight AGNs of Shu et al. (2011) are also selected,
dancewiththeunifiedmodelofAGNs(e.g.,Awaki et al. for which the FWHMs of the polarized broad Balmer
1991). emission lines are available to obtain the single-epoch
When the width of an emission line and the distance black hole mass estimate on the basis of the luminosity
of its emitting region from the black hole are obtained, scalingrelationofthe BLR.Intotal,seventype-2 AGNs
theblackholemasscanbeestimatedassumingthevirial are selected because three objects in the 4 + 6 type-2
relation. Jiang et al. (2011) assumed that the location AGNs are common. The type-2 target AGNs and their
of the emitting region of the neutral FeKα line core cor- data are listed in Table 2.
responds to the near-infrared K-band reverberation ra-
diusfortheinnerregionofthedusttorus,andestimated 3. RESULTS
the black hole mass of 10 type-1 AGNs on the basis of 3.1. Velocity Width of the Neutral FeKα Line Core
the FeKαline widths andthe luminosityscalingrelation
First, we compare the velocity width of the neutral
of the dust reverberation radius to the optical V-band
FeKα line core with that of the broad Balmer emission
luminosity by Suganuma et al. (2006). They examined
lines, and that corresponding to the dust reverberation
theblackholemassonthebasisoftheneutralFeKαline
radius to examine the location of its emitting region.
corebycomparingtheblackholemassobtainedfromthe
InFigure1,weplottedtheFWHMoftheneutralFeKα
broad emission-line reverberation; however, the correla-
linecoreagainstthoseofthe broadHβ emissionlineand
tion between them was statistically insignificant, which
the polarized broadBalmer emission lines for the type-1
could be attributable to small sample size and large un-
and type-2 target AGNs, respectively. We also plotted
certainties in the FeKα line widths.
thelinethatrepresentstheFWHMatthedustreverbera-
Recently, Greene et al. (2010b) and Koshida et al.
tionradiusobservedintheKband,whichisestimatedby
(2014) presented the luminosity scaling relations of the
scalingtheFWHMofthebroadHβ emissionlineaccord-
broad Hβ emission-line reverberation radius and the
ing to the virial relation based on the systematic differ-
dust reverberation radius to isotropic luminosity indi-
ence of the reverberationradii of the broadHβ emission
cators that are unbiased by the obscuration, such as
line and the near-infrared dust emission (Koshida et al.
the luminosity of the [O IV] λ25.89 µm emission line
2014).
L , and that of the Swift BAT hard X-ray (14–195
[OIV] As shown in Figure 1, we find that the FWHM of the
keV)bandL (e.g.,Mel`endez et al.2008;Rigby et al.
BAT neutral FeKα line core falls between that of the broad
2009; Diamond-Stanic et al. 2009). Using these radius-
Balmeremissionlinesandthecorrespondingvalueatthe
luminosityscalingrelations,itbecomespossibletoderive
dustreverberationradiusforallofthetype-2AGNs,and
theradiusoftheemittingregionoftheneutralFeKαline
for most of the type-1 AGNs except the only one outlier
core for obscured AGNs.
of NGC 7469. This suggests that significant fraction of
In this paper, we estimate the black hole mass of both
photons of the neutral FeKα line core originate between
obscured and unobscured AGNs on the basis of the line
the outer BLR and the inner dust torus in most cases.
width of the neutral FeKα line core and the luminos-
ity scaling relation of its emitting regionradius with the
3.2. Luminosity Scaling Relation of the Neutral FeKα
[O IV] λ25.89 µm emission-line luminosity. Then, the
Line Core Emitting Region
estimated black hole mass was examined by comparing
themwithotherblackholemassestimatesthathavebeen Asanextstep,wescaledtheradius-luminosityscaling
established. We assume the cosmology of H = 73 km relation of the broad Hβ emission line according to the
0
s−1 Mpc−1, Ω = 0.27, and Ω = 0.73 according to systematic difference between the FWHMs of the neu-
m Λ
Spergel et al. (2007) throughout this paper. tral FeKα line core and the broad Hβ emission line for
the type-1 AGNs assuming the virial relation. It is esti-
2. TARGETS
mated by the linear regression analysis where the slope
We collect the line width data of the neutral FeKα is fixed to unity, and the best-fit linear regression is
line core obtained by the Chandra High Energy Trans- logFWHM = logFWHM −0.186 (±0.088) with
FeKα Hβ
mission Grating Spectrometer (HETGS; Markert et al. an additional scatter of 0.11 dex in both directions for
1994), which affords the best spectral resolution cur- the reduced χ2 to achieve unity. Then, the emitting re-
rently available in the FeKα energy band (∼ 39 eV, or gion radius of the neutral FeKα line core is estimated
∼1860kms−1 infull width athalfmaximum; FWHM), as
and we further select targetAGNs with other mass esti- logr =logr −0.37 (±0.18) (1)
FeKα Hβ
mates of their central supermassive black holes.
For the type-1 AGNs, seven target AGNs whose black according to the virial relation. Koshida et al. (2014)
hole masses are measured by the reverberation of the estimatedtheluminosityscalingrelationofthereverber-
broad emission lines are selected out of 12 AGNs, for ation radii to the luminosity of the [O IV] emission line
which the best FWHM constraint for the neutral FeKα in the form of logr = α+ 0.5logL[OIV]/1041 erg s−1,
line core was obtained by Shu et al. (2010). The type-1 as α = −1.94 (±0.07) and α = −1.28 (±0.07) for the
A New Estimate for Black Hole Mass 3
reverberation radius of the broad Hβ emission line and Finally, the black hole mass M estimated from
BH,FeKα
the dust reverberation radius observed in the K band, the neutral FeKα line core is compared with other esti-
respectively. By scaling the former relation according matesoftheblackholemassforthesametargettoexam-
to Equation (1), the luminosity scaling relation of the inetheconsistencybetweenthem. Forthetype-1AGNs,
emitting region radius of the neutral FeKα line core is we comparedM with the broademission-line re-
BH,FeKα
estimated as verberationmass,M ,andforthe type-2AGNs,we
BH,rev
compare it with the mass M based on the H O
logrFeKα =−1.57(±0.19)+0.5logL[OIV]/1041 ergs−1 . maserandthesingle-epochmaBsHs,Hes2tOimateMBH,pol bas2ed
(2) onthepolarizedbroadBalmerlines. TheM iscal-
BH,pol
The radius-luminosity scaling relation for the neutral culated in the same way as the M , except that
BH,FeKα
FeKα line core of Equation (2) is located between the the radius-luminosity scaling relation of the broad Hβ
reverberation radii of the broad Hβ emission line and emission line presented by Koshida et al. (2014) is used.
the dust torus emission in the K band as indicated by TheM ,M ,M ,andtheirreferencesare
BH,rev BH,H2O BH,pol
Figure 1, and we use it for the estimation of the black listed in Tables 1 and 2.
hole mass in the next section. In Figure 2, M is plotted against other black
BH,FeKα
hole mass indicators such as M , M , and
BH,rev BH,H2O
3.3. Estimating the Black Hole Mass from the Neutral M for the type-1 and type-2 target AGNs. We
BH,pol
FeKα Line Core then calculated the Pearson’s correlation coeffient, and
also performed linear regression analysis to M
Next, we estimate the black hole mass of both type- BH,FeKα
and the other black hole mass indicators. The resul-
1 and type-2 target AGNs using the neutral FeKα line
tant values are listed in Table 3, and the best-fit linear
core. The emitting region radius r is estimated
FeKα
regressionlines are also presented in Figure 2.
from L and the radius-luminosity scaling relation
[OIV] Forthe type-1 AGNs, M and M are con-
of Equation (2) because the hard X-ray flux seems to sistent within ±2σ exceptBfHo,rFeNKGαC 7469B.HT,rehve best-fit
be attenuated in some heavily obscured AGNs in the
linear regressionwhere the slope is fixed to unity is esti-
targets. The L value is taken from Liu & Wang
[OIV] mated as
(2010),onthebasisoftheoriginaldataofMel`endez et al. log(MBH,FeKα/107.5M⊙)=−0.15(±0.11)+log(MBH,rev/107.5M⊙),
(2008) and Diamond-Stanic et al. (2009) for most of the (4)
targets; however, it is recalculated from the [O IV] which indicates that the systematic difference between
emission-line flux of Diamond-Stanic et al. (2009), as- them is within a factor of several tens percent. How-
sumingtheredshift-independentdistancetakenfromthe ever, the correlation between MBH,FeKα and MBH,rev is
NASA/IPACExtragalacticDatabaseforverynearbyob- notclearaswiththeprecedingstudy(Jiang et al.2011).
jects. In addition, the L of NGC 2110 is calculated ExcludingtheclearoutlierNGC7469,thePearson’scor-
[OIV]
fromthe dataofWeaver et al. (2010). We thenestimate relation coefficient is calculated as R = 0.701, which in-
the black hole mass assuming the virial relation as dicates that the confidence levelof the correlationis less
than 90%. The best-fit linear regressionfor them is esti-
r σ2 mated as
MBH,FeKα =f FeKαGFeKα (3) log(MBH,FeKα/107.5M⊙)=−0.19(±0.17)+1.52(±0.92)×log(MBH,rev/107.5M⊙
(5)
where f is the virialfactor andσ is the velocitydis- with an additional scatter in both directions for the re-
FeKα
persionof the neutralFeKα line core. We adopt f =5.5 duced χ2 to achieve unity. The large uncertainty in the
following the references of the broad emission-line re- slope of the best-fit linear regressionalso indicates no or
verberation for the type-1 target AGNs (Onken et al. a weak correlation between them.
For the type-2 AGNs, M are also consistent
2004; Peterson et al. 2004; Bentz et al. 2006, 2007; BH,FeKα
with M within ±2σ. The best-fit linear regression
Denney et al. 2010; Grier et al. 2012). The velocity dis- BH,pol
persion is calculated as σFeKα = FWHMFeKα/2.35 be- for MBH,FeKα and MBH,pol is estimated as
cause a Gaussian emission-line profile is used for the log(MBH,FeKα/107.5M⊙)=−0.13(±0.13)+1.05(±0.25)×log(MBH,pol/107.5M⊙
spectralfittingoftheneutralFeKαlinecoreinShu et al. (6)
(2010,2011). TheL andtheM forthetype- which indicates that the systematic difference between
[OIV] BH,FeKα
1 and type-2 target AGNs are listed in Tables 1 and 2, MBH,FeKα and MBH,pol is also within a factor of several
respectively. tens percent. In addition, MBH,FeKα and MBH,pol are
The FWHM of the neutral FeKα line core spreads be- well correlated and show an approximately proportional
tween that of the broad Hβ line and that correspond- relationship, since the slope of the best-fit linear regres-
ing to the dust reverberation radii in most cases. This sion is positive more than 3σ uncertainties and close to
scatter causes a systematic uncertainty in the r de- unity. The Pearson’s correlation coefficients are calcu-
FeKα
rived from the radius-luminosity scaling relation, and lated as R = 0.831, which indicates that the confidence
also yields ±0.3 dex or a factor ∼ 2 uncertainty in this level of the correlation is more than 95%.
blackholemassestimates,MBH,FeKα,includingthemea- In contrast to the good agreement with MBH,pol,
surementerrorsof the line widths in the data. As in the MBH,FeKα seems to be systematically larger than
case of NGC 7469, in which the FWHM of the neutral MBH,H2O althoughtheyshowaclearcorrelationwiththe
FeKα line is significantly larger than that of the broad Pearson’s correlation coefficient of R = 0.960 indicating
Hβ line, M could be sometimes overestimated. theconfidencelevelofmorethan95%. Thebest-fitlinear
BH,FeKα regressionis estimated as
3.4. Comparison with Other Black Hole Mass Indicators log(MBH,FeKα/106.5M⊙)=0.70(±0.12)+1.57(±0.33)×log(MBH,H2O/106.5M⊙
(7)
4 Minezaki et al.
which indicates that M is systematically larger Contraryto these studies, Kuo et al.(2011) estimated
BH,FeKα
than M by about a factor of ∼ 5. We note that the radius of the BLR from the absorption-corrected
BH,H2O
MBH,pol is also larger than MBH,H2O by a similar factor luminosity in the 2–10 keV X-ray band L2−10keV via
for the three targets, NGC 1068, NGC 4388, and Circi- the luminosity scaling relation of the BLR to estimate
nus,forwhichbothM andM areestimated. M of4type-2AGNswhoseM wasobtained
BH,H2O BH,pol BH,pol BH,H2O
Tosummarizetheresults,M isconsistentwith bythem,andpresentedthatM andM were
BH,FeKα BH,H2O BH,pol
M for most of the type-1 AGNs and with M consistent with each other. In fact, the three targets,
BH,rev BH,pol
for all of the type-2 AGNs. M is correlated well NGC 1068,NGC 4388,andCircinus,for whichthe large
BH,FeKα
with M and M for the type-2 AGNs, al- systematic difference between M and M is
BH,pol BH,H2O BH,pol BH,H2O
thoughM isasystematicsmallerthantheothers. presented in this paper, are included in the targets of
BH,H2O
These results suggest that the black hole mass estimate Kuo et al. (2011). The reason for the discrepancy be-
usingtheneutralFeKαlineisapotentialindicatorofthe tween these studies is uncertain, but a difficulty in es-
black hole mass, especially for obscured AGNs. timating the luminosity of the central engine L2−10keV
for most of the targets with very large X-ray absorbing
4. DISCUSSION columndensity could be suspected, althoughit wasesti-
mated with particular attention.
4.1. Systematic Difference of the H O maser mass from
2 Assuming both M and M are reliable es-
the Other Black Hole Mass Indicators BH,rev BH,H2O
timates for the black hole mass, M and M
BH,FeKα BH,pol
The systematic difference of M implies an am- are considered to overestimate the black hole mass in
BH,H2O
biguous issue. Since M is considered to be the type-2 AGNs. The systematically larger M for
BH,H2O BH,FeKα
most precise estimate for the black hole mass, then, the type-2 AGNs implies that the width of the neutral
through the relations of M with M and FeKα line core is systematically larger for the large in-
BH,H2O BH,FeKα
M , another reliable estimate M is suspected clination angle for type-2 AGNs. If the emitting region
BH,pol BH,rev
to considerably overestimate the black hole mass that is of the neutral FeKα line core shows an equatorial mo-
difficult to be accounted for by the uncertainties in the tion, then the virial factor f in Equation (3) decreases
virialfactorf (5.5inthispaperandpresentedbyOnken as the inclination angle increases. For example, f = 2
et al. 2004; 5.2 by Woo et al. 2010; 2.8 by Graham et at the edge-on view, which is smaller than f = 5.5 for
al. 2011; 4.31 by Grier et al. 2013). In this section, the type-1 AGNs, is derived assuming an equatorialcircular
systematic differences between those black hole mass es- motion. However, it is not so small as to account for
timates and the possible origins of these differences are the difference of about a factor of ∼5. Outflow motions
discussed. toward the equatorial plane might broaden the width of
In fact, the systematic difference between M the neutral FeKα line core for type-2 AGNs.
BH,pol
and M has been indicated by preceding studies. On the other hand, the systematically larger M
BH,H2O BH,pol
Greene et al. (2010a) measured the stellar velocity dis- forthe type-2AGNscannotbe attributedtotheinclina-
persion σ∗ in the centers of the galaxies hosting type- tionofAGNsfromthepointofviewoftheunifiedmodel
2 AGNs whose M was obtained, and found that ofAGNs;whenthebroademissionlineisscatteredtothe
BH,H2O
M tended to be smaller than the black hole mass observerat the polar regionwell above the height of the
BH,H2O
that was estimated from σ∗ and the MBH–σ∗ relation of dust torus, the viewing angle of the BLR from the scat-
local elliptical galaxies. On the other hand, Zhang et al. tering region would be similar to that from the observer
(2008) estimated the radius of the BLR from the [O III] for type-1 AGNs. Non-virial motions such as outflows
luminosityviatheluminosityscalingrelationoftheBLR in the scattering region are possible to be contributed,
to estimate M of 12 type-2 AGNs, and found that asYoung et al.(2007)proposedtoexplainthebroadHα
BH,pol
M tended to be larger than the black hole mass emissionlineinthepolarizedspectrumofthequasarPG
BH,pol
that was estimated from σ∗ via the MBH–σ∗ relation. 1700+518. However, a common or closely related mech-
According to these results, M would be systemati- anismtobroadenthewidthsofbothoftheneutralFeKα
BH,pol
cally larger than M , as is the case in this study. line core and the polarized broad Balmer emission lines
BH,H2O
Recently, Ho & Kim (2014) presented that the virial systematicallyfortype-2AGNsisrequiredtoexplainthe
factor f of the reverberation-mappedAGN is systemati- approximateagreementbetweenM andM .
BH,FeKα BH,pol
callydifferentbetweentwobulgetypesofthehostgalax-
ies, as f = 6.3±1.5 for classical bulges and ellipticals, 4.2. Location of the Neutral FeKα Line Core Emitting
and f = 3.2±0.7 for psuedobulges. In fact, the black Region
hole mass of the AGNs with psuedobulges tends to be As described in Section 3, the FWHM of the neutral
smaller than that with classical bulges and ellipticals as FeKα line core falls between the FWHM of the broad
shown in their Figure 2. Greene et al. (2010a) also ar- Balmer emission lines and the corresponding values at
guedthatthemassoftheblackholehostedbylater-type the dust reverberation radius in most cases, which in-
and lower-mass galaxies tended to be less massive with dicates that the emitting region of its major fraction is
larger scatter than that estimated from the MBH–σ∗ re- locatedbetweentheouterBLRandtheinnerdusttorus.
lation for the elliptical galaxies. Since our type-2 AGN Itis suggestedby previousstudies for someofthe target
targets with the H O maser observation tend to have a AGNs. ForNGC4151,Takahashi et al.(2002)compared
2
less massive black hole than the others if M is thevariationofthecontinuumandtheexcessfluxaround
BH,H2O
correct,then,itispossiblethatthesystematicdifference the neutral FeKα line and presented that the emitting
of the target selection partly contributes to the system- region of the excess flux has an extent of 1017 cm, or
aticdifferencebetweenM andtheotherblackhole 40lightdays,whichis consistentwithorslightlysmaller
BH,H2O
mass indicators. thanthedustreverberationradius(Koshida et al.2014).
A New Estimate for Black Hole Mass 5
For NGC 5548,Liu et al. (2010) reported that the emit- error of the black hole mass estimated from the neutral
tingregionoftheneutralFeKαlinewaslocatedbetween FeKα line.
thetworeverberationradiiofthebroadHβ emissionline Inthisstudy,Weusedthebestspectralresolutiondata
andthenear-infrareddustemissionbasedonitsfluxvari- currently available, but the spectral resolution and sen-
ation and line width. sitivity are still limited. As a result, the number of the
The BLR and the dust torus are considered not to targets is small and the relatively large uncertainties re-
be decoupled from each other but to constitute a con- main in the FWHM data, which would make it difficult
tinuous structure, and the transition zone of the BLR toexaminethelocationoftheneutralFeKαlineemitting
and the dust torus would be located between the two region in more detail. In the near future, the substan-
reverberation radii (Nenkova et al. 2008; Koshida et al. tial progress is expected by the ASTRO-H X-ray satel-
2009, 2014). In addition, recent reverberation observa- lite (Takahashi et al.2010), whichis capable ofunprece-
tionsoftheopticalFeIIemissionlinespresentedthatits dentedenergy-resolutionspectroscopywithsuperiorsen-
reverberationradiuswascomparableto oratmosttwice sitivity at FeKα band. It will enable us to study more
that of the broad Hβ emission line (Barth et al. 2013; detail on the origin of the neutral FeKα line core by
Chelouche et al. 2014), which shows the presence of low examining the line profile of the neutral FeKα line and
ionization iron atoms there. its time variation, then it will become able to measure
If the FeKα emitting region constitutes a tight strati- the black hole mass with improved accuracy for a large
fied structure with the BLR and the dust torus, a good number of obscured AGNs. The hard X-ray luminosity
correlation between the FWHMs of the neutral FeKα with precise absorption correction is obviously valuable
line and the broad Balmer emission lines is expected. for the isotropic luminosity indicator because the scal-
However, as shown in Figure 1, there appears no clear ing relations of the reverberation radii to the hard X-
correlation between them as has been reported by pre- ray luminosity are tighter than to the [O IV] luminosity
vious studies (Nandra 2006; Shu et al. 2010, 2011). A (Koshida et al. 2014).
simpleexplanationforitwouldbethattheneutralFeKα
line is produced in different origins suchas the outer ac-
We thank T. Kawaguchi, M. Yoshida, K. Kawabata,
cretion disk, the BLR, and the innermost dust torus,
and Y. Fukazawa for useful discussions and comments.
and the mixing ratio is different from target to tar-
We also thank the anonymous referee for valuable com-
get (e.g., Yaqoob & Padmanabhan 2004; Nandra 2006;
mentstoimprovethemanuscript. Thisresearchhasbeen
Bianchi et al. 2008; Liu et al. 2010; Ponti et al. 2013).
partly supported by the Grants-in-Aid of Scientific Re-
On the other hand, Shu et al. (2011) suggested that the
search(22540247and 25287062)of the Ministry of Edu-
neutral FeKα line may originate from a universal region
cation, Science, Culture and Sports of Japan, and has
at the same radius with respect to the gravitational ra-
made use of the NASA/IPAC Extragalactic Database
dius of the central black hole, because its FWHM is al-
(NED),whichisoperatedbythe JetPropulsionLabora-
most constant and independent of the black hole mass.
tory, California Institute of Technology, under contract
Inanycase,thoseuncertaintieswillresultinthepossible
with the National Aeronautics and Space Administra-
tion.
REFERENCES
Awaki,H.,Koyama,K.,Inoue, H.,&Halpern,J.P.1991, PASJ, Ho,L.C.,&Kim,M.2014, ApJ,789,17
43,195 Hopkins,P.F.,Hernquist,L.,Cox,T.J.,&Kereˇs,D.2008,
Barth,A.J.,Pancoast, A.,Bennert,V.N.,etal.2013,ApJ,769, ApJS,175,356
128 Jiang,P.,Wang, J.,&Shu,X.2011,ScienceChinaPhysics,
Bentz,M.C.,Denney,K.D.,Cackett, E.M.etal.2006, ApJ, Mechanics,andAstronomy,54,1354
651,775 Kormendy,J.,&Richstone,D.1995,ARA&A,33,581
Bentz,M.C.,Denney,K.D.,Cackett, E.M.etal.2007, ApJ, Kormendy,J.,&Ho,L.C.2013,ARA&A,51,511
662,205 Koshida,S.,Yoshii,Y.,Kobayashi,Y.,etal.2009,ApJ,700,L109
Bianchi,S.,LaFranca,F.,Matt,G.,etal.2008,MNRAS,389, Koshida,S.,Minezaki,T.,etal.2014, ApJ,788,159
L52 Kuo,C.Y.,Braatz,J.A.,Condon,J.J.,etal.2011,ApJ,727,20
Chelouche, D.,Rafter,S.E.,Cotlier,G.I.,Kaspi,S.,&Barth, Laor,A.1998,ApJ,505,L83
A.J.2014,ApJ,783,L34 Liu,Y.,Elvis,M.,McHardy,I.M.,etal.2010, ApJ,710,1228
Denney,K.D.,Peterson,B.M.,Pogge, R.W.,etal.2010, ApJ, Liu,T.,&Wang,J.-X.2010,ApJ,725,2381
721,715 Lodato,G.,&Bertin,G.2003, A&A,398,517
Diamond-Stanic,A.M.,Rieke,G.H.,&Rigby,J.R.2009, ApJ, Magorrian,J.,Tremaine,S.,Richstone,D.,etal.1998,AJ,115,
698,623 2285
Fukazawa, Y.,Hiragi,K.,Mizuno,M.,etal.2011, ApJ,727,19 Markert,T.H.,Canizares,C.R.,Dewey,D.,etal.1994,
Graham,A.W.,Onken,C.A.,Athanassoula, E.,&Combes,F. Proc.SPIE,2280,168
2011, MNRAS,412,2211 McLure,R.J.,&Jarvis,M.J.2002,MNRAS,337,109
Greene,J.E.,Peng,C.Y.,Kim,M.,etal.2010,ApJ,721,26 Mel`endez,M.,Kraemer,S.B.,Armentrout,B.K.,Deo,R.P.,
Greene,J.E.,Hood,C.E.,Barth,A.J.,Bennert,V.N.,Bentz, Crenshaw,D.M.,Schmitt,H.R.,Mushotzky,R.F.,Tueller,J.,
M.C.,Filippenko,A.V.,Gates,E.,Malkan,M.A.,Treu,T., Markwardt,C.B.&Winter,L.2008,ApJ,682,94
Walsh,J.L.,&Woo,J.-H.2010,ApJ,723,409 Miyoshi,M.,Moran,J.,Herrnstein,J.,etal.1995,Nature,373,
Greenhill,L.J.,Moran,J.M.,&Herrnstein,J.R.1997, ApJ, 127
481,L23 Moran,E.C.,Barth,A.J.,Kay,L.E.,&Filippenko,A.V.2000,
Greenhill,L.J.,Booth,R.S.,Ellingsen,S.P.,etal.2003, ApJ, ApJ,540,L73
590,162 Moran,E.C.,Barth,A.J.,Eracleous,M.,&Kay,L.E.2007,
Grier,C.J.,Peterson, B.M.,Pogge,R.W.,etal. 2012, ApJ, ApJ,668,L31
755,60 Nandra,K.2006,MNRAS,368,L62
Grier,C.J.,Martini,P.,Watson,L.C.,etal.2013,ApJ,773,90
6 Minezaki et al.
Nandra,K.,O’Neill,P.M.,George,I.M.,&Reeves,J.N.2007, Suganuma,M.,Yoshii,Y.,Kobayashi,Y.,etal.2006,ApJ,639,
MNRAS,382,194 46
Nenkova, M.,Sirocky,M.M.,Nikutta,R.,Ivezi´c,Zˇ.,&Elitzur, Takahashi,K.,Inoue, H.,&Dotani,T.2002,PASJ,54,373
M.2008,ApJ,685,160 Takahashi,T.,Mitsuda,K.,Kelley,R.,etal.2010,Proc.SPIE,
Nishiura,S.,&Taniguchi,Y.1998, ApJ,499,134 7732,
Oliva,E.,Marconi,A.,Cimatti,A.,&Alighieri,S.D.S.1998, Tran,H.D.1995,ApJ,440,565
A&A,329,L21 Tran,H.D.2001,ApJ,554,L19
Onken,C.A.,Ferrarese,L.,Merritt,D.,etal.2004,ApJ,615,645 Ueda,Y.,Eguchi,S.,Terashima,Y.,etal.2007,ApJ,664,L79
Peterson,B.M.,Ferrarese,L.,Gilbert,K.M.,etal.2004,ApJ, Vestergaard,M.,&Peterson,B.M.2006,ApJ,641,689
613,682 Weaver,K.A.,Mel´endez,M.,Mushotzky, R.F.,etal.2010,ApJ,
Ponti,G.,Cappi,M.,Costantini,E.,etal.2013,A&A,549,A72 716,1151
Rigby,J.R.,Diamond-Stanic,A.M.,&Aniano,G.2009,ApJ, Woo,J.-H.,Treu,T.,Barth,A.J.,etal.2010,ApJ,716,269
700,1878 Yaqoob,T.,&Padmanabhan, U.2004,ApJ,604,63
Sanders,D.B.,Soifer,B.T.,Elias,J.H.,etal.1988,ApJ,325,74 Young,S.,Hough,J.H.,Efstathiou,A.,etal.1996, MNRAS,
Sanders,D.B.,&Mirabel,I.F.1996,ARA&A,34,749 281,1206
Shen,Y.,&Ho,L.C.2014, Nature,513,210 Young,S.,Axon,D.J.,Robinson,A.,Hough,J.H.,&Smith,
Shu,X.W.,Yaqoob,T.,&Wang, J.X.2010,ApJS,187,581 J.E.2007,Nature,450,74
Shu,X.W.,Yaqoob,T.,&Wang, J.X.2011,ApJ,738,147 Zhang,S.-Y.,Bian,W.-H.,&Huang,K.-L.2008,A&A,488,113
Spergel,D.N.,Bean,R.,Dor´e,O.,etal.2007,ApJS,170,377
A New Estimate for Black Hole Mass 7
10000
]
1
-s

m
k
[

α
K
e
F
M
H
W
F
1000
1000 10000
-1
FWHM [km s ]
Balmer
Fig.1.— The FWHM of the neutral FeKα line core against that of the broad Balmer emission lines. The open circles (colored blue
inthe online version) represent the broad Hβ emission linefor the type-1 AGNs, and the filled circles (colored red inthe online version)
represent the polarized broad Balmer emissionlines for the type-2 AGNs. The solidlinerepresents unity, and the dashed line represents
theFWHMattheK-banddustreverberationradius(Koshidaetal.2014). (Acolorversionofthisfigureisavailableintheonlinejournal.)
8 Minezaki et al.
9
10
8
] 10
n
u
s
M
[

α
K
e
F
H,
B
M 107
6
10
6 7 8 9
10 10 10 10
M ,M ,M [M ]
BH,rev BH,pol BH,H O sun
2
Fig. 2.— TheblackholemassestimatebasedontheneutralFeKαlinecore,MBH,FeKα,andotherblackholemassestimates. Theopen
circles(coloredblueintheonlineversion)representthebroademission-linereverberationmass,MBH,rev forthetype-1AGNs. Thefilled
boxes(coloredgreenintheonlineverseion)representtheblackholemassbasedontheVLBIobservationoftheH2Omaser,MBH,H2O,and
thefilledcircles(coloredredintheonlineversion)representthesingle-epochmassestimatebasedonthepolarizedbroadBalmeremission
lines, MBH,pol, for the type-2 AGNs. The solidlinerepresents unity, and the dashed , dot-dashed, and dotted lines (colored blue, green,
andred,respectively,intheonlineversion)representthebest-fitlinearregressionlinesofMBH,FeKα toMBH,rev,MBH,H2O,andMBH,pol,
respectively. (Acolorversionofthisfigureisavailableintheonlinejournal.)
A New Estimate for Black Hole Mass 9
TABLE 1
List ofthe targettype-1 AGNs
Object FWHMFeKα log L[OIV] MBH,FeKα FWHMHβ MBH,rev refsa
kms−1 ergs−1 107M⊙ kms−1 107M⊙
3C120 2230+2280 42.42 17+51 2539±466 6.70+0.60 1
−1650 −16 −0.60
NGC3516 3180+880 40.95 6.2+3.9 5175±96 3.17+0.28 2
−670 −2.3 −0.42
NGC3783 1750+360 40.73 1.46+0.66 3093±529 2.98+0.54 3
−360 −0.54 −0.54
NGC4151 2250+400 40.66 2.23+0.86 4711±750 4.57+0.57 4
−360 −0.66 −0.47
NGC5548 2540+1140 40.96 4.0+4.4 6396±167 6.54+0.26 5
−820 −2.2 −0.25
Mrk509 2910+2590 41.85 15+38 2715±101 14.3+1.2 3
−1250 −10 −1.2
NGC7469 4890+2770 41.30 22+32 2169±459 1.22+0.14 3
−1700 −13 −0.14
References. —(1)Grieretal.(2012);(2)Denneyetal.(2010);(3)Petersonetal.
(2004);(4)Bentzetal.(2006);(5)Bentzetal.(2007).
a ReferencesforFWHMofthebroadHβ emissionline,FWHMHβ,andtheblackhole
massfromthebroademission-linereverberation,MBH,rev.
TABLE 2
List ofthe targettype-2 AGNs
Object FWHMFeKα log L[OIV]a MBH,FeKα FWHMpol refsb MBH,pol MBH,H2O refsc
kms−1 ergs−1 107M⊙ kms−1 107M⊙ 107M⊙
NGC1068 2660+410 41.67 9.5+3.2 3030 1 6.7 0.80+0.03 8
−440 −2.9 −0.03
NGC2110 2320+810 40.76 2.5+2.1 2150±950d 2 1.2+1.3 ··· ···
−800 −1.5 −0.8
Mrk3 3140+870 41.93 18+11 6000±500 3 35+6 ··· ···
−660 −7 −6
NGC4388 2430+620 41.05 3.9+2.2 3100±400 4 3.4+0.9 0.84+0.02 9
−590 −1.7 −0.8 −0.02
NGC4507 2200+910 40.97 2.9+2.9 5000 5,6 8.1 ··· ···
−720 −1.6
NGC4945 2780+1110 38.69 0.34+0.32 ··· ··· ··· 0.14 10
−740 −0.16
Circinus . 1710+190 40.16 0.69+0.16 3300 7 1.4 0.17+0.03 11
−170 −0.13 −0.03
References. —(1)Nishiura&Taniguchi(1998);(2)Moranetal.(2007);(3)Tran(1995);(4)Youngetal.
(1996);(5)Moranetal.(2000);(6)Shuetal.(2011);(7)Olivaetal.(1998);(8)Lodato&Bertin(2003);(9)
Kuoetal.(2011);(10)Greenhilletal.(1997);(11)Greenhilletal.(2003).
a ThesamedistanceasthatoftheMBH,H2O referenceisadopted whenavailable.
b ReferencesforFWHMofthepolarizedbroadBalmeremissionlines,FWHMpol.
c ReferencesfortheblackholemassbasedontheH2Omaserobservation, MBH,H2O.
dTheaverageFWHMofthepolarizedHβemissionlineandthecorecomponentofthepolarizedHαemission
line.
TABLE 3
Correlation coefficientsandlinear regressions to MBH,FeKα
MBH,ref na Rb logMBH,0 c ac bc
[M⊙]
MBH,revd 6 0.701 7.5 −0.19±0.17 1.52±0.92
−0.15±0.11 1.0(fixed)
MBH,pol 6 0.831 7.5 −0.13±0.13 1.05±0.25
−0.12±0.11 1.0(fixed)
MBH,H2O 4 0.960 6.5 0.70±0.12 1.57±0.33
0.73±0.14 1.0(fixed)
a Thenumberofthedata.
b ThePearson’scorrelationcoefficient.
c The fitted model is log(MBH,FeKα/MBH,0) = a + b ×
log(MBH,ref/MBH,0),whereaandbaretheparameterstobefitted.
d NGC 7469 was excluded from the calculations of the correlation co-
efficientandthelinearregression.