Acta Polytechnica doi:10.14311/AP.2014.54.0271 Acta Polytechnica 54(4):271–274, 2014 © Czech Technical University in Prague, 2014 available online at http://ojs.cvut.cz/ojs/index.php/ap SEARCH FOR ISOLATED BLACK HOLES: PAST, PRESENT, FUTURE Sergey Karpov∗, Grigory Beskin, Vladimir Plokhotnichenko Special Astrophysical Observatory of Russian Academy of Sciences ∗ corresponding author: karpov@sao.ru Abstract. The critical property of a black hole is the presence of an event horizon. It may be detected only by means of a detailed study of the emission features of its surroundings. The temporal resolution of such observations has to be comparable to rg/c, which is in the 10−6–10 s range, depending on the mass of the black hole. At SAO RAS we have developed the MANIA hardware and software complex, based on the panoramic photon counter, and we use it in observations on the 6 m telescope for searching and investigating the optical variability of various astronomical objects on time scales of 10−6–103 s. We present here the hardware and methods used for these photometric, spectroscopic and polarimetric observations, together with the principles and criteria for object selection. The list of objects includes objects with featureless optical spectra (DC white dwarfs, blazars) and long microlensing events. Keywords: black holes, accretion, observations. 1. Accretion onto isolated stellar-mass black holes Even though more than 60 years have passed since the theoretical prediction of black holes as astrophysical objects [1], in a certain sense they still have not been discovered. The features of a black hole — the compactness (the size for mass M is close to the Schwarzschild radius rg = 2GMc2 ) and mass larger than 3M� — are necessary but not sufficient features. The key property of a black hole is the presence of an event horizon instead of a usual surface. It is necessary to detect and study the emission generated very close to an event horizon to decide whether the horizon is present in a given compact and massive object. This is a very complicated task that cannot be easily performed in X-ray binaries and AGNs due to the high accretion rate and, consequently, the high optical depth of the accreting gas. At the same time, single stellar-mass black holes which accrete an interstellar medium of low density (10−2–1 cm−3) are the ideal case for detecting and studying an event horizon [2]. The analysis of existing data on possible black hole masses and velocities in comparison with the interstel- lar medium structure shows that in the majority of cases in the Galaxy (> 90%) the dimensionless accre- tion rate ṁ = Ṁc2/Ledd can not exceed 10−6–10−7 [3]. For typical interstellar medium inhomogeneity the cap- tured specific angular momentum is smaller than that on the BH last stable orbit, and the accretion is al- ways spherically-symmetric [4, 5]. The only possible exception is the case of a slowly-moving (V < 10 km/s) black hole in cold dense clouds of interstellar hydrogen (n ∼ 102–105 cm−3, T ∼ 102 K), where the accretion may become disk-like and the accretion rate is high enough to provide luminosities up to 1038–1040 erg/s. The plasma in the accretion flow is collisionless all the way down to the event horizon, and the continuity of the flow is provided by the magnetic field. Correct treatment of adiabatic heating, the major accretion flow heating mechanism in spherically-symmetric ac- cretion flow, shows that it is 25 % more efficient than the flow for ideal gas[3]. Due to it the plasma temper- ature in the accretion flow grows much faster and the electrons become relativistic earlier – the radius where electrons become relativistic Rrel = rrel/rg ≈ 6000, in contrast to Rrel ≈ 1300 in [6] and Rrel ≈ 200 in [7]. The accretion flow is much hotter, and our estimation of “thermal” luminosity L = ηṀc2 = 9.6 · 1033M310n 2 1(V 2 + c2s) −3 16 erg/s (1) is significantly higher than the estimates derived by Ipser & Price [7] LIP = 1.6 · 1032M310n 2 1(V 2 + c2s) −3 16 erg/s (2) and by Bisnovatyi-Kogan & Ruzmaikin [6] LBKR = 2 · 1033M310n 2 1(V 2 + c2s) −3 16 erg/s, (3) while the optical spectrum shape is nearly the same (see Fig. 1). The efficiency of accretion as a function of accretion rate is shown in Fig. 4. Correct treatment of the magnetic field behaviour in accretion flow is an extremely complicated task (see, for example, [8] and references therein). The basis of our analysis is the assumption of energy equipartition in the accretion flow (Shvartsman’s theorem, see [2]) which determines the radial dependencies of both infall velocity and magnetic field strength. A direct consequence of this assumption is the necessity of magnetic energy dissipation at the rate defined as the difference between the rates of magnetic energy 271 http://dx.doi.org/10.14311/AP.2014.54.0271 http://ojs.cvut.cz/ojs/index.php/ap S.Karpov, G.Beskin, V.Plokhotnichenko Acta Polytechnica Figure 1. Decomposition of a single black hole (with the mass 10 M�) emission spectrum into thermal and nonthermal parts. The accretion rate is 1.4 · 1010 g/s, which corresponds to ṁ = 10−8. increase for a purely frozen-in magnetic field and the rate for equipartition [6]. In the accretion models proposed earlier this dis- sipation runs continuously[9] in the turbulent flow and its mechanism is not examined in detail. We considered conversion of the magnetic energy in the discrete turbulent current sheets [10] as a mechanism providing such dissipation, and studied the observa- tional consequences of this process. These include the generation of various modes of plasma oscillations (ion-acoustic and Lengmur plasmons mostly) and the acceleration of electrons, which is very important for the observational appearances of the whole accretion flow. The beams of the accelerated electrons emit their energy due to the motion in the magnetic field and generate an additional nonthermal component in addition to synchrotron emission of thermal particles (see Figs. 1 and 2). A large fraction of black hole emission is generated inside the 2rg sphere, and so carries information about the physical conditions of space-time very close to the event horizon. An important property of nonthermal emission is its flaring nature — the electron ejection process is discrete, and the typical light curve of a single beam is shown in Fig. 3. The light curve of each flare has a stage of fast intensity increase and a sharp cut-off, and its shape reflects the properties of the magnetic field and space-time near the horizon. A study of these flares in black hole emission therefore provides a way to probe extreme space-time regions directly very close to the horizon. 2. Observational appearance of an isolated black hole The black hole at a 100 pc distance (a sphere with this radius must contain several tens of such objects, see [11]) looks like a 15–25m optical object (due to the “thermal” spectral component) with a strongly Figure 2. Spectra of the accretion flow onto the 10 M� black hole for the various accretion rates. variable counterpart in high-energy spectral bands (“nonthermal” component) [3]. The hard emission consists of flares, the majority of which are generated inside a 5rg distance from the BH. These events have durations of ∼ rg/c (∼ 10−4 s), a rate of 103–104 flares per second, and an amplitude of 2–6 %. The BH variable X-ray emission can be detected by modern space-borne telescopes. The optical emission consists of a quasi-stationary “thermal” part and a low-frequency tail of nonthermal flaring emission. The rate and duration of optical flares are the same as X-ray flares, while their ampli- tudes are significantly smaller. Indeed, the contribu- tion of the nonthermal component to the optical emis- sion is approximately 2 · 10−2 for ṁ = 10−8–10−6, so the mean amplitudes of optical flares are 0.04–0.12 %, while the peak amplitudes may be 1.5-2 times higher and may reach 0.2 %. Certainly, it is nearly impos- sible to detect such single flares, but their collective power reaches 18–24m and may therefore be detected in observations with high time resolution (< 10−4 s) by large optical telescopes[3]. Of course, the variability of the BH emission is related not only to the electron acceleration processes described here. Additional variability may be result from plasma oscillations from various kinds, or other types of instabilities. The time scale of such variability may be from rg/c till rc/c (where rc = 2GMV 2+c2s is the gas capture radius [12]), i.e., from microseconds till years. 3. The search for isolated stellar mass black holes in the optical band The most striking property of the accretion flow onto a single black hole is its inhomogeneity — the clots of plasma act as probes testing the space-time proper- ties near the horizon. The characteristic timescale of emission variability is τv ∼ rg/c ∼ 10−4–10−5 s and this short stochastic variability may be considered as a distinctive property of a black hole as the smallest 272 vol. 54 no. 4/2014 Search for Isolated Black Holes Figure 3. Internal structure of a flare as a reflec- tion of the electron cloud evolution. The prevailing physical mechanisms defining the observed emission are denoted and typical durations of the stages are shown. possible physical object with a given mass. Its param- eters — spectra, energy distribution and light curves – carry important information on space-time properties of the horizon [3]. The general observational appearance of a single stellar-mass black hole at typical interstellar medium densities — its brightness and featureless optical spec- trum — is the same as other optical objects without spectral lines — DC-dwarfs and ROCOSes (Radio Ob- jects with Continuous Optical Spectra, a subclass of blazars) [13]. The suggestion that isolated BHs can be among them forms the basis of the observational pro- gramme in search of isolated stellar-mass black holes — MANIA (Multichannel Analysis of Nanosecond In- tensity Alterations). It uses photometric observations of candidate objects with high time resolution, special hardware and data analysis methods [14, 15], and is based on the fact that fast variability is the critical property of isolated black hole emission. In observations of 40 DC-dwarfs and ROCOSes us- ing the 6-meter telescope of the Special Astrophysical Observatory and the standard high time resolution photometer based on photomultipliers, only upper limits for variability levels of 20–5 % on the timescales of 10−6–10 s respectively were obtained, i.e., BHs were not detected [15–17]. In the new stage of the MANIA experiment, since the end of 1990s, we have developed the multichan- nel panoramic spectro-polarimeter (MPPP) based on the position-sensitive detector (PSD) with 1 µs time resolution ([18, 19]). Such detectors use the set of microchannel plates (MCP) for electron multiplica- tion and a multi-electrode collector to determine the incoming photon position. The PSD used in our ob- servations has the following parameters: quantum efficiency of 10 % in the 3700–7500 A range (S20 pho- tocathode), MCP stack gain of 106, spatial resolution of 70 µm (0.21′′ for the 6 m telescope), 700 ns time resolution, 7 · 104 pixels with 22 mm working diameter, and 200–500 counts/s detector noise. The “Quan- tochron 4-480” spectal time-code convertor with 30 ns Figure 4. Efficiencies of the synchrotron emission of thermal and non-thermal electron components of the accretion flow. time resolution and 106 counts/s maximal count rate is used as an acquisition system. This equipment al- lows us to study 20–22m objects for 1-hour exposure (under good weather conditions) with microsecond temporal resolution [20, 21]. In recent years, the population of objects with featureless optical spectra and without known local- izations has been extended by means of the cross- correlation of surveys of various wave bands (from radio to gamma) and follow-up spectroscopic observa- tions [13, 22]. These objects are the major targets of a new stage of the MANIA experiment. In addition, some evidence has recently appeared that the isolated stellar mass black holes may be among the unidentified gamma-ray sources [23, 24]. Another large class of candidate objects for isolated black holes are those where independent estimation of the mass is possible, e.g., long-lasting MACHO microlensing events [25]. Another possibility is a black hole in a binary system with a white dwarf (though, strictly speaking, this type of black hole is not isolated, it accretes from interstellar gas only, and therefore may be considered in the same way). This type of binary may be detected by means of its periodic brightness amplification on the tens of seconds time scale due to gravitational microlensing [26]. By using the technical equipment of SDSS (23m limit in a 6-square-degree field), roughly 15 such objects may be detected in the course of 5 years. It is clear that in such a system it is easy to estimate the mass of the black hole. Acknowledgements This work has been supported by INTAS grant No 04-78- 7366, by the Russian Foundation for Basic Research (grant No 04-02-17555), and by the Russian Science Support Foundation. S.K. has also been supported by a grant from the Dynasty foundation. G.B. thanks Landau Network- Cenro Volta and the Cariplo Foundation for a fellowship and the Brera Observatory for hospitality. 273 S.Karpov, G.Beskin, V.Plokhotnichenko Acta Polytechnica References [1] J. R. Oppenheimer, et al. On Continued Gravitational Contraction. Physical Review 56:455–459, 1939. doi:10.1103/PhysRev.56.455. [2] V. F. Shvartsman. Halos around ”Black Holes”. Astronomicheskij Zhurnal 48:479–488, 1971. [3] G. M. Beskin, et al. Low-rate accretion onto isolated stellar-mass black holes. A&A 440:223–238, 2005. arXiv:astro-ph/0403649 doi:10.1051/0004-6361:20040572. [4] H. Mii, et al. Ultraluminous X-Ray Sources: Evidence for Very Efficient Formation of Population III Stars Contributing to the Cosmic Near-Infrared Background Excess? ApJ 628:873–878, 2005. arXiv:astro-ph/0501242 doi:10.1086/430942. [5] G. Beskin, et al. Search for the event horizon by means of optical observations with high temporal resolution. In V. Karas, et al. (eds.), IAU Symposium, vol. 238 of IAU Symposium, pp. 159–163. 2007. doi:10.1017/S1743921307004899. [6] G. S. Bisnovatyi-Kogan, et al. The Accretion of Matter by a Collapsing Star in the Presence of a Magnetic Field. ApSS 28:45–59, 1974. doi:10.1007/BF00642237. [7] J. R. Ipser, et al. Synchrotron radiation from spherically accreting black holes. ApJ 255:654–673, 1982. doi:10.1086/159866. [8] I. V. Igumenshchev, et al. Three-dimensional Magnetohydrodynamic Simulations of Spherical Accretion. ApJ 566:137–147, 2002. arXiv:astro-ph/0105365 doi:10.1086/338077. [9] G. S. Bisnovatyi-Kogan, et al. Influence of Ohmic Heating on Advection-dominated Accretion Flows. ApJL 486:L43–L46, 1997. arXiv:astro-ph/9704208 doi:10.1086/310826. [10] L. A. Pustilnik. Stability of Accretion Models. ApSS 252:353–362, 1997. [11] E. Agol, et al. X-rays from isolated black holes in the Milky Way. MNRAS 334:553–562, 2002. arXiv:astro-ph/0109539 doi:10.1046/j.1365-8711.2002.05523.x. [12] H. Bondi, et al. On the mechanism of accretion by stars. MNRAS 104:273–282, 1944. [13] S. A. Pustilnik. The list of Radio Objects with Purely Continuum Optical Spectra. Preliminary Analysis of Their Features. Soobshcheniya Spetsial’noj Astrofizicheskoj Observatorii 18:3–41, 1976. [14] V. F. Shvartsman. The MANIA [Multichannel Analysis of Nanosecond Intensity Alterations] experiment. Astrophysical problems, mathematical methods, instrumentation complex, results of the first observations. Soobshcheniya Spetsial’noj Astrofizicheskoj Observatorii 19:5–38, 1977. [15] G. M. Beskin, et al. The Investigation of Optical Variability on Time Scales of 10−7-102 s: Hardware, Software, Results. Experimental Astronomy 7:413–420, 1997. [16] V. F. Shvartsman, et al. The Results of Search for Superrapid Optical Variability of Radio Objects with Continuous Optical Spectra. Astrofizika 31(3):685–690, 1989. [17] V. F. Shvartsman, et al. A Search for 0.5-MICROSECOND to 40-SECOND Optical Variability in DC White Dwarfs. Soviet Astronomy Letters 15:145–149, 1989. [18] V. Debur, et al. Position-sensitive detector for the 6-m optical telescope. Nuclear Instruments and Methods in Physics Research A 513:127–131, 2003. arXiv:astro-ph/0310353 doi:10.1016/S0168-9002(03)02233-2. [19] V. Plokhotnichenko, et al. The multicolor panoramic photometer-polarimeter with high time resolution based on the PSD. Nuclear Instruments and Methods in Physics Research A 513:167–171, 2003. arXiv:astro-ph/0310354 doi:10.1016/S0168-9002(03)02249-6. [20] V. L. Plokhotnichenko, et al. High-temporal resolution multimode photospectropolarimeter. Astrophysical Bulletin 64:308–316, 2009. doi:10.1134/S1990341309030109. [21] V. G. Debur, et al. High temporal resolution coordinate-sensitive detector with gallium-arsenide photocathode. Astrophysical Bulletin 64:386–391, 2009. doi:10.1134/S1990341309040087. [22] G. Tsarevsky, et al. A search for very active stars in the Galaxy. First results. A&A 438:949–955, 2005. arXiv:astro-ph/0502235 doi:10.1051/0004-6361:20042274. [23] N. Gehrels, et al. Discovery of a new population of high-energy γ-ray sources in the Milky Way. Nature 404:363–365, 2000. [24] N. La Palombara, et al. X-ray and optical coverage of 3EG J0616-3310 and 3EG J1249-8330. Mem. Soc. Astron. Italiana 75:476, 2004. arXiv:astro-ph/0403705. [25] D. P. Bennett, et al. Gravitational Microlensing Events Due to Stellar-Mass Black Holes. ApJ 579:639–659, 2002. arXiv:astro-ph/0109467 doi:10.1086/342225. [26] G. Beskin, et al. Detection of compact objects by means of gravitational lensing in binary systems. A&A 394:489–503, 2002. 274 http://dx.doi.org/10.1103/PhysRev.56.455 http://arxiv.org/abs/astro-ph/0403649 http://dx.doi.org/10.1051/0004-6361:20040572 http://arxiv.org/abs/astro-ph/0501242 http://dx.doi.org/10.1086/430942 http://dx.doi.org/10.1017/S1743921307004899 http://dx.doi.org/10.1007/BF00642237 http://dx.doi.org/10.1086/159866 http://arxiv.org/abs/astro-ph/0105365 http://dx.doi.org/10.1086/338077 http://arxiv.org/abs/astro-ph/9704208 http://dx.doi.org/10.1086/310826 http://arxiv.org/abs/astro-ph/0109539 http://dx.doi.org/10.1046/j.1365-8711.2002.05523.x http://arxiv.org/abs/astro-ph/0310353 http://dx.doi.org/10.1016/S0168-9002(03)02233-2 http://arxiv.org/abs/astro-ph/0310354 http://dx.doi.org/10.1016/S0168-9002(03)02249-6 http://dx.doi.org/10.1134/S1990341309030109 http://dx.doi.org/10.1134/S1990341309040087 http://arxiv.org/abs/astro-ph/0502235 http://dx.doi.org/10.1051/0004-6361:20042274 http://arxiv.org/abs/astro-ph/0403705 http://arxiv.org/abs/astro-ph/0109467 http://dx.doi.org/10.1086/342225 Acta Polytechnica 54(4):271–274, 2014 1 Accretion onto isolated stellar-mass black holes 2 Observational appearance of an isolated black hole 3 The search for isolated stellar mass black holes in the optical band Acknowledgements References