An ultraviolet-optical flare from the tidal disruption of a helium-rich stellar core

TLDR: A luminous ultraviolet–optical flare from the nuclear region of an inactive galaxy at a redshift of 0.1696 is reported and it is determined that the disrupted star was a helium-rich stellar core, modulo a factor dependent on the mass and radius of the star disrupted.

Abstract: The flare of radiation from the tidal disruption and accretion of a star can be used as a marker for supermassive black holes that otherwise lie dormant and undetected in the centres of distant galaxies. Previous candidate flares have had declining light curves in good agreement with expectations, but with poor constraints on the time of disruption and the type of star disrupted, because the rising emission was not observed. Recently, two ‘relativistic’ candidate tidal disruption events were discovered, each of whose extreme X-ray luminosity and synchrotron radio emission were interpreted as the onset of emission from a relativistic jet. Here we report a luminous ultraviolet–optical flare from the nuclear region of an inactive galaxy at a redshift of 0.1696. The observed continuum is cooler than expected for a simple accreting debris disk, but the well-sampled rise and decay of the light curve follow the predicted mass accretion rate and can be modelled to determine the time of disruption to an accuracy of two days. The black hole has a mass of about two million solar masses, modulo a factor dependent on the mass and radius of the star disrupted. On the basis of the spectroscopic signature of ionized helium from the unbound debris, we determine that the disrupted star was a helium-rich stellar core.

Figures

Figure 2 Ultraviolet-optical light curve. The GALEX NUV and PS1 gP1-, rP1-, iP1-, and zP1-band light curves of PS1-10jh (with the host galaxy flux removed), plotted against logarithmic time since the peak (top) and since the disruption (bottom). The curves (shown with solid lines scaled to the flux in the GALEX and PS1 bands) were determined from the best fit of the gP1-band light curve to a numerical model20 for the mass accretion rate of a tidally disrupted star with a polytropic exponent of 5/3. For each of the four optical bands, we independently performed a least-squares fit of the model for a 106M⊙ black hole to the light curve from −36 to 58 rest-frame days from the peak, with the time of disruption, a vertical scaling factor, and a time stretch factor as free parameters. The GALEX and PS1 photometry at t > 240 rest-frame days since the peak is shown binned in time in order to increase the signal-to-noise ratio. The dates of multiple epochs of MMT spectroscopy are marked with an S, and the date of the Chandra X-ray observation is marked with an X.

Figure 4 Offset of PS1-10jh from the mean x and y position of the host galaxy centroid measured in the nightly stacked images before the event. Solid points show the centroid of the host galaxy before the event, and X symbols show the centroid of PS1-10jh, in each of the 4

Figure 3 Spectral energy distribution. SED of PS1-10jh during nearly simultaneous GALEX ultraviolet and PS1 optical observations (with the host galaxy flux removed) at two epochs (rest-frame days -19 to -17 and 242 to 247 from the peak of the flare). Flux densities have been corrected for Galactic extinction of E(B − V ) = 0.013 mag. The ultraviolet-optical SED from −19 to 247 rest-frames days from the peak is fitted with a 2.9×104 K blackbody. Orange solid lines show blackbodies with this temperature scaled to the NUV flux densities for the respective epochs. Open symbols show the GALEX and PS1 flux densities corrected for an internal extinction of E(B−V ) = 0.08 mag, and dotted blue line shows the 5.5× 104 K blackbody fit to the de-reddened flux densities for the respective epochs. The upper limit from the Chandra observation on 2011 May 22.96 UT assuming a spectrum with a photon index of Γ = 2, typical of an AGN, is plotted with a thick black line. The X-ray flux density expected from an unobscured AGN with a comparable NUV flux is plotted for comparison with a thick grey line28. Also shown are the u, g, and r band flux densities measured from aperture photometry with the Liverpool Telescope29 on 2011 Sep 24 UT, after subtracting the host galaxy flux measured by SDSS. Errors, 1σ.

Figure 8 Fits of the gP1-band light curve of PS1-10jh from −38 to 58 rest-frame days from the peak to models for the mass accretion rate of tidally disrupted stars of different polytropic exponent γ.

Figure 6 3000 4000 5000 6000 7000 Rest Wavelength (Å)

Figure 1

Full text

arXiv:1205.0252v1 [astro-ph.CO] 1 May 2012
An ultraviolet-optical flare from the tidal disruption of a
helium-rich stellar core
S. Gezari
1
, R. Chornock
2
, A. Rest
3
, M. E. Huber
4
, K. Forster
5
, E. Berger
2
, P. J. Challis
2
, J. D.
Neill
5
, D. C. Martin
5
, T. Heckman
1
, A. Lawrence
6
, C. Norman
1
, G. Narayan
2
, R. J. Foley
2
, G. H.
Marion
2
, D. Scolnic
1
, L. Chomiuk
2
, A. Soderberg
2
, K. Smith
7
, R. P. Kirshner
2
, A. G. Riess
1
, S.
J. Smartt
7
, C.W. Stubbs
2
, J.L. Tonry
4
, W. M. Wood-Vasey
8
, W. S. Burgett
4
, K. C. Chambers
4
, T.
Grav
9
, J. N. Heasley
4
, N. Kaiser
4
, R.-P. Kudritzki
4
, E. A. Magnier
4
, J. S. Morgan
4
, & P. A. Price
10
1
Department of Physics and Astronomy, Johns Hopkins University, 3400 North Charles Street,
Baltimore, MD 21218, USA.
2
Harvard-Smithsonian Center for Astrophysics, 60 Garden Street, Cambridge, MA 02138, USA.
3
Space Telescope Science Institute, 3700 San Martin Drive, Baltimore, MD 21218, USA.
4
Institute for Astronomy, University of Hawaii, 2680 Woodlawn Drive, Honolulu HI 96822, USA.
5
California Institute of Technology, 1200 East California Blvd., Pasadena, CA 91125, USA.
6
Institute for Astronomy, University of Edinburgh Scottish Universities Physics Alliance, Royal
Observatory, Blackford Hill, Edinburgh, EH9 3HJ, UK.
7
Astrophysics Research Centre, School of Mathematics and Physics, Queen’s University Belfast,
Belfast, BT7 1NN, UK.
8
Pittsburgh Particle Physics, Astrophysics, and Cosmology Center, Department of Physics and
Astronomy, University of Pittsburgh, 3941 O’Hara Street, Pittsburgh, PA 15260, USA.
9
Planetary Science Institute, 1700 East Fort Lowell, Tucson, AZ 85719, USA
10
Department of Astrophysical Sciences, Princeton University, Princeton, NJ 08544, USA.
The flare of radiation from the tidal disruption and accretion of a star can be used as a
marker for supermassive black holes that otherwise lie dormant and undetected in the cen-
tres of distant galaxies
1
. Previous candidate flares
2–6
have had declining light curves in good
agreement with expectations, but with poor constraints on the time of disruption and the type
of star disrupted, because the rising emission was not observed. Recently, two ‘relativistic’
candidate tidal disruption events were discovered, each of whose extreme X-ray luminosity
and synchrotron radio emission were interpreted as the onset of emission from a relativistic
jet
7–10
. Here we report the discovery of a luminous ultraviolet-optical flare from the nuclear
region of an inactive galaxy at a redshift of 0.1 696. The observed continuum is cooler than
expected for a simple accreting debris disk, but the well-sampled rise and decline of its light
curve follows the predicted mass accretion rate, and can be modelled to determine the time
of disruption to an accuracy of two days. The black hole has a mass of about 2 million solar
masses, modulo a factor dependent on the mass and radius of the star disrupted. On the basis
of the spectroscopic signature of ionized helium from the unbound debris, we determine that
the disrupted star was a helium-rich stellar core.
When the pericenter of a star’s orbit (R
p
) passes within the tidal disruption radius of a
massive black hole, R
T
≈ R
⋆
(M
BH
/M
⋆
)
1/3
, tidal forces overcome the binding energy of the
1

star, which breaks up with roughly half of the stellar debris remaining bound to the black hole
and the rest being ejected at high velocity
1
. For black holes above a critical mass, M
crit
≈
10
8
r
3/2
⋆
m
−1/2
⋆
M
⊙
(where r
⋆
= R
⋆
/R
⊙
and m
⋆
= M
⋆
/M
⊙
), the star becomes trapped within
the event horizon of the black hole before being disrupted. The mass accretion rate (
˙
M) in a
tidal disruption event (TDE) can be calculated directly from the orbital return-times of the bound
debris
1,11,12
. For the simplest case of a star of uniform density this yields,
˙
M =
2
3
(
fM
⋆
t
min
)(
t
t
min
)
−5/3
,
where f is the fraction of the star accreted and t
min
is the orbital period of the most tightly bound
debris and, therefore, the time delay between the time of disruption and the start of the flare, which
scales as M
1/2
BH
M
−1
⋆
R
3/2
⋆
for R
p
= R
T
. The radiative output of the accreted debris is less certain,
and depends on the ratio of the accretion rate to the Eddington rate
13
.
The optical transient, PS1-10jh (α
J2000
= 16h09min28.296s, δ
J2000
= + 53 deg 40
′
23.52
′′
),
was discovered on 2010 May 31.45 UT in the Pan-STARRS1
14
(PS1) Medium Deep Survey by our
two independent image-differencing pipelines. The densely sampled (cadence,∆ ∼ 3 d) optical
light curves of PS1-10jh in the g
P1
, r
P1
, i
P1
, and z
P1
bands (Supplementary Information) follow the
rise of the transient to its peak in the g
P1
band on 2010 July 12.31 UT and its subsequent decline
until 2011 September 1.24 UT (Supplementary Table 1). PS1-10jh was discovered independently
as a transient, near-ultraviolet (NUV ) source at the 20σ level by the
GALEX
15
Time Domain
Survey (TDS) on 2010 June 17.68 UT within 2.5± 3.0 arcsec of the PS1 location, and was detected
in ten more epochs of TDS observations between then and 2011 June 10.68 UT (Supplementary
Table 2). No source is detected in a deep coadd of all the TDS epochs in 2009, with a 3σ upper
limit of > 25.6 mag implying a peak amplitude of variability in the NUV of > 6.4 mag. See the
Supplementary Information for details on the PS1 and
GALEX
photometry.
PS1-10jh is coincident with the centre of a galaxy within the 3σ positional uncertainty (0.036
arcsec; Supplementary Information) with rest-frame u, g, r, i, and z photometry from SDSS
16
and
K photometry from UKIDSS
17
fitted with a galaxy template
18
with M
stars
= (3.6 ± 0.2) × 10
9
M
⊙
and M
r
= −18.7 mag, where M
stars
is the galaxy stellar mass and M
r
is the absolute r-band
magnitude. The mass of the central black hole as determined indirectly from locally established
scaling relations
19
is 4
+4
−2
× 10
6
M
⊙
. We obtained five epochs of optical spectroscopy at the location
of PS1-10jh between 2010 June 16.33 and 2011 September 4.23 UT with the 6.5-m MMT (Sup-
plementary Table 3). The continua in the spectra are well modelled by the combination of a galaxy
host at redshift 0.1696 (luminosity distance, d
L
= 816 Mpc) with a stellar population with an age
of 1.4 − 5.0 Gyr, depending on the chosen metallicity, and a fading hot blackbody component with
T
BB
≈ 3 × 10
4
K (Fig. 1).
The spectra show no narrow emission lines that would be indicative of star formation or
an active galactic nucleus (AGN). We obtained a 10 ks, 0.2-10 keV X-ray observation, using the
Chandra
X-ray Observatory, at the location of PS1-10jh on 2011 May 22.96 UT, and detected no
source above the background with a 3σ upper limit of L
X
(0.2 − 10 keV) < 5 .8 × 10
41
ergs s
−1
for an unobscured AGN spectrum. The X-ray faintness and extreme NUV variability amplitude
of PS1-10jh strongly disfavour its origin in an AGN, and its prolonged brightness in the ultraviolet
2

strongly disfavours its origin in a supernova (Supplementary Information).
The rise and decay of the light curve of PS1-10jh is well described by numerical simulations
for the mass return rate from a star that is tidally disrupted at R
p
= R
T
with an internal structure
parameterised by a polytropic exponent of 5/3 characteristic of a fully convective star or a degen-
erate core
20
(Fig. 2). The decline from the peak is too steep to be fitted by simulations of a more
centrally concentrated stellar structure, such as one that is characteristic of a solar-type star (Sup-
plementary Information). There are systematic differences between the light curve and the model
during the early rise (more than −44 rest-frame days before the peak) and the late decay (more than
240 rest-frame days after the peak) which could imply a stellar structure more complex than one
described by a single polytrope. The mass of the black hole is determined from the stretch factor
of 1.38 ± 0.03 applied to fit the model of a 10
6
M
⊙
black hole to the light curve, which implies that
the time of disruption was 76 ± 2 d before the peak and M
BH
= (1.9 ± 0.1) × 10
6
m
2
⋆
r
−3
⋆
M
⊙
.
The most constraining property of PS1-10jh is the detection of very broad high-ionisation
He II emission at wavelengths of λ = 4, 686
˚
A(full-width at half-maximum, 9, 000 ± 700 km
s
−1
) and λ = 3, 203
˚
Athat fade in time along with the ultraviolet-optical continuum. The lack of
Balmer line emission in the spectra requires an extremely low hydrogen mass fraction, of < 0.2
(Supplementary Information), which cannot be found in the ambient interstellar medium or in a
passiveaccretion disk. This is the strongest evidence that PS1-10jh must be fuelled by the accretion
of a star that has lost its hydrogen envelope, either through stellar winds or tidal interactions with
the central supermassive black hole. The broad width of the line is also what is expected from
the velocities of the most energetic unbound stellar debris in a tidal disruption event
21
, that is
v
max
∼ 1 × 10
4
(M
BH
/10
6
M
⊙
)
1/6
(R
T
/R
p
)r
−1/2
⋆
m
1/3
⋆
km s
−1
.
We measure the SED of the flare over time from the nearly simultaneous PS1 optical and
GALEX
ultraviolet observations (with the host galaxy flux removed; Fig. 3). The pre-peak SED is
fitted with a blackbody with T
BB
= (2.9 ± 0.2) × 10
4
K, consistent with the blackbody component
seen in the spectra. However, the temperature fit is very sensitive to internal extinction. If we
correct for the maximum internal extinction of E(B − V ) = 0.08 mag allowed by the observed
He IIλ = 3, 203
˚
A, λ = 4, 686
˚
Aemission, the best-fit temperature increases to ( 5.5 ± 0.4) × 10
4
K. In fact, we know that the photo-ionizing continuum must have T
BB
>
∼
5 × 1 0
4
K 22 rest-frame
days before the peak in order to produce enough λ < 228
˚
A photons to photoionise the He II
λ =4,686
˚
Aline observed with a luminosity of (9 ± 1) × 10
40
ergs s
−1
. The late-time SED can be
fitted with the same temperature as the pre-peak SED. We note that the observed continuum tem-
perature, and even the maximum temperature allowed by possible de-reddening, are considerably
cooler than the temperature of ≈ 2.5 × 10
5
(M
BH
/10
6
M
⊙
)
1/12
r
−1/2
⋆
m
−1/6
⋆
K expected from mate-
rial radiating at the Eddington limit at the tidal radius
13
. This discrepency is also seen in AGN
22
and may imply that the continuum we see is due to reprocessing of some kind
22,23
On the basis of the arguments above, we assume that the observed temperature is a lower
3

limit, T
BB
>
∼
3 × 10
4
K. The peak bolometric luminosity is thus
>
∼
2.2 × 10
44
ergs s
−1
and the total
energy emitted from integrating under the light-curve model is
>
∼
2.1 × 10
51
ergs, corresponding
to a total accreted mass (M
acc
) of
>
∼
0.012(ǫ/0.1)
−1
M
⊙
, where ǫ is the efficiency of converting
matter into radiation.
The internal structure and helium-rich abundance of the star derived from the light curve
and the spectra can be consistently modelled by the tidally stripped core of a red giant (precursor
to a helium white dwarf) that had a main-sequence mass of M
⋆
>
∼
1M
⊙
in order to have evolved
off the main sequence in less time than the age of the stellar population (< 5 Gyr). This tidal
stripping mechanism has been invoked to explain the hot stars in the Galactic Centre
24
, and the
rate of tidal disruption of tidally stripped stars is likely to be higher than solar-type stars
25
. The
mass of the black hole from the light curve fit depends on the mass and radius of the star at the
time of disruption. Using M
⋆
∼ 0.23M
⊙
and R
⋆
∼ 0.33R
⊙
(measured for a red giant core that
was stripped in a binary system
26
), and assuming that the evolution of the core is similar to one
that is tidally stripped, we find that f = M
acc
/M
⋆
>
∼
0.058 (approaching f
>
∼
0.1 as measured
in simulations
27
), that M
BH
= (2.8 ± 0.1) × 10
6
M
⊙
and that the peak luminosity approaches the
Eddington luminosity of the supermassive black hole (L
peak
>
∼
0.6L
Edd
).
4

Figure 1
0
2
4
6
f
λ
(10
−17
erg cm
−2
s
−1
Å
−1
)
T
BB
= 30,000 K
2.5 Gyr SSP model
BB + 2.5 Gyr model
PS1−10jh
Day −22
a
3000 4000 5000 6000 7000
0
1
Difference
He II λ4686He II λ3203
0.0
0.2
0.4
0.6
0.8
1.0
1.2
f
λ
(10
−17
erg cm
−2
s
−1
Å
−1
)
Day 254b
3000 4000 5000 6000 7000
Rest Wavelength (Å)
−0.2
0.0
0.2
Difference
Optical spectrum. MMT optical spectra (black) of PS1-10jh obtained −22 (a) and +254 (b)
rest-frame days from the peak, expressed in terms of flux density. Each continuum is fitted
with a combination (magenta) of a stellar population 2.5 Gyr old and a fading blackbody
with a temperature of ∼ 3 × 10
4
K determined from the ultraviolet-optical spectral energy
distribution (SED). He II emission at λ = 4, 686
˚
A(Fowler series n = 4 → 3) is detected
above the continuum model and fitted with a Gaussian with a full-width at half-maximum
of 9, 0 00 ± 700 km s
−1
and L = (9 ± 1) × 10
40
ergs s
−1
(plotted with a green line in the early
epoch (a)). Residual emission above the continuum model is also detected at ∼ 3, 200
˚
A which is coincident with the location of the He IIλ3203 (Fowler series n = 5 → 3) line,
5

Frequently asked questions

Q1. What are the contributions in this paper?

In this paper, the authors present a survey of the authors of this paper: A.S.Chornock, A.C.Chambers, T.N. Grav, J.M.Neill, D. C. Neill, C. N. Heasley, N. S. Nelson, R.-P. Kudritzki, E. A.Charlton, G. H. Nielsen, A Soderberg, K. J. Smith, R. P. Radev, S. Sermanet, R-P 

Q2. What is the purpose of the PS1 system?

The PS1 system is developing the Transient Science Server (TSS) which automatically takes the nightly stacks, creates image differences with reference images created from deep stacks, carries out PSF fitting photometry on the image differences, and returns catalogues of variable and transient candidates. 

Q3. What is the accretion rate of a super-Eddington galaxy?

For sub-Eddington accretion rates, the luminosity should follow the decline of the massreturn rate, which depends on the internal structure of the star at early times, but approaches an n = 5/3 power-law after a few times tmin for all stellar types. 

Q4. How much NH does one need to obscure the AGN emission during the flare?

in order to obscure the AGN hard X-ray emission during the flare, assuming a standard intrinsic αox, one requires NH ∼ 1024 cm−2. 

Q5. How long did the data take to binne?

In order to improve the signal-to-noise (S/N) in the photometry at late-times (t > 240 rest-frame days after the peak) in the figures, the authors binned the data into time intervals of 30 days. 

Q6. What is the tidal disruption radius of a black hole?

When the pericenter of a star’s orbit (Rp) passes within the tidal disruption radius of a massive black hole, RT ≈ R⋆(MBH/M⋆)1/3, tidal forces overcome the binding energy of thestar, which breaks up with roughly half of the stellar debris remaining bound to the black hole and the rest being ejected at high velocity1. 

Q7. How many ks of light are visible in PS1-10jh?

The most constraining property of PS1-10jh is the detection of very broad high-ionisation He II emission at wavelengths of λ = 4, 686 Å(full-width at half-maximum, 9, 000 ± 700 km s−1) and λ = 3, 203 Åthat fade in time along with the ultraviolet-optical continuum. 

Q8. How many decays can be expected for a fixed RBB?

For L = 4πR2BBσT 4 BB, if L ∝ Ṁ ∝ (t/tmin) −5/3, then on the Rayleigh-Jeans tail, for a fixed RBB one expects an n = 5/12 decay21, 49. 

Q9. What is the recombination time of the unbounddebris?

Since the number density of the unbounddebris is high21, n ∼ 3× 1013M1/66 β −5m −2/3 ⋆ r 3/2 ⋆ (t/36 d)−3 cm−3, the recombination time is short compared to the flare timescale, τrec = (neαB)−1 ∼ (n 1+2[n(H0)/n(He+)] 

Q10. What is the mass accretion rate in a tidal disruption event?

The mass accretion rate (Ṁ ) in a tidal disruption event (TDE) can be calculated directly from the orbital return-times of the bound debris1, 11, 12. 

Q11. How many epochs of optical spectroscopy of PS1-10jh?

The authors obtained five epochs of optical spectroscopy of PS1-10jh using the Blue Channel36 and fiberfed Hectospec37 spectrographs on the 6.5-m MMT. 

Q12. How does the temperature fit to the He II?

If the authors correct for the maximum internal extinction of E(B − V ) = 0.08 mag allowed by the observed He IIλ = 3, 203Å, λ = 4, 686Åemission, the best-fit temperature increases to (5.5 ± 0.4) × 104 K. 

Q13. What is the flux density of the spectrum?

MMT optical spectra (black) of PS1-10jh obtained −22 (a) and +254 (b) rest-frame days from the peak, expressed in terms of flux density. 

Q14. What is the bolometric luminosity evolution of the UV and optical continuum?

A possible explanation for both the constant shape of the UV and optical SED and the linear scaling of the UV and optical light curve with the predicted bolometric luminosity evolution of the TDE, is that the UV and optical continuum is a ”pseudo-continuum” whose shape is determined by atomic reprocessing. 

Q15. How many ergs s1 cm2 is the flux of a?

This corresponds to a flux of < 7.2 × 10−15 ergs s−1 cm−2 when corrected for Galactic extinction with NH = 3.1E(B − V )1.8 × 1021 cm−2 = 7.2 × 1019 cm−2 and assuming a Γ = 2 energy spectrum typical of an unobscured AGN, or LX(0.2 − 10)keV< 5.8 × 1041 ergs s−1. 

Q16. What is the stretch factor for the UV and optical continuum?

Without the constraints from the rise and decay of the light curve, the values for the stretch factor and the time of disruption can vary widely. 

Q17. How much lower is the X-ray to UV luminosity ratio?

The upper limit to the X-ray to UV luminosity density ratio 260− 270 rest-frame days from the peak is 20 times lower than observed in broad-lined AGNs of a comparable NUV luminosity28, and argues strongly against an association of the flare with an AGN. 

Q18. What is the bolometric luminosity of the UV and optical continuum?

In such a scenario, the UV and optical SED shape remains fixed even if the photoionising continuum is cooling with time (its shape is determined by a velocity-blurred reflection spectrum and not the temperature of the photoionising continuum), and the UV and optical light follows the decay of the bolometric luminosity since it is the result of the reflection, absorption, and re-emission of the photoionising continuum. 

Q19. What is the average cadence of observations in the yP1 band?

The typical Medium-deep cadence of observations cycles through the gP1, rP1, iP1 and zP1 bands every 3 nights, with observations in the yP1 band close to the full moon. 

Q20. Who designed the observations and the transient detection pipeline for GALEX TDS?

Contributions S.G. designed the observations and the transient detection pipeline for GALEX TDS, and measured the UV photometry of PS1-10jh. 

Q21. How much energy is emitted from the light curve?

The peak bolometric luminosity is thus >∼ 2.2×10 44 ergs s−1 and the total energy emitted from integrating under the light-curve model is >∼ 

Q22. Why does the yP1 band have an additional uncertainty?

The authors do not include the yP1 band photometry which has an additional uncertainty of ∼ 0.05 mag in the zeropoint due to the lack of an SDSS comparison.