Published papers

We mapped the high-velocity compact cloud CO-0.40-0.22 in 21 molecular lines in the 3 mm band using the Nobeyama Radio Observatory 45 m radio telescope. Eighteen lines were detected from CO-0.40-0.22. The map of each detected line shows that this cloud has a compact appearance (d ≃ 3 pc) and extremely broad velocity width (ΔV ≃ 100 km/s). The mass and kinetic energy of CO-0.40-0.22 are estimated to be 10^(3.6) Msun and 10^(49.7) erg, respectively. The representative position-velocity map along the major axis shows that CO-0.40-0.22 consists of an intense region with a shallow velocity gradient and a less intense high-velocity wing. Here, we show that this kinematical structure can be attributed to a gravitational kick to the molecular cloud caused by an invisible compact object with a mass of ~10^5 Msun. Its compactness and the absence of counterparts at other wavelengths suggest that this massive object is an intermediate-mass black hole.

The circumnuclear disk (CND) is a potential reservoir for material accreting into the central parsec of our Galaxy. Two giant molecular clouds (GMCs; +20 km s-1 and +50 km s-1 clouds) are adjacent to the CND. Although several authors have suggested the physical connection between these GMCs and the CND, no compelling evidence of physical connections has been found to date. Recently we have performed deep OTF mapping observations of the CND with the NRO 45 m radio telescope and the ASTE 10 m telescope in the HCN J=1-0, SiO J=2-1, CS J=2-1, HCO+ J=4-3 lines, etc. We discovered a bridge in the l-b-V space which connects the CND and the +20 km s-1 cloud. We interpret this as a gas streamer from outside to the CND.

We have performed unbiased line surveys at the 3 mm band toward the Galactic circumnuclear disk (CND) and Sgr A* using the NRO 45 m radio telescope. We have obtained three wide-band spectra that cover the frequency range from 81.3 GHz to 115.8 GHz, detecting 50 lines from 30 species. Based on the line intensities, we classified the detected lines into three categories: the CND, GMC, HBD types. We suggest that the large molecules have been destroyed in the vicinity of the nucleus by high energy photons and/or cosmic ray particles from the nucleus.

We performed a 3 mm band line survey toward CO-0.40-0.22, a small cloud with an extremely large velocity width (Δ V˜90 km s-1) in the central molecular zone of our Galaxy, using the Mopra 22 m telescope. We surveyed the frequency range between 76 GHz and 116 GHz detecting 54 lines from 32 molecules. Analyzing line profiles carefully, we concluded that CH3OH, HC3N, H2CS, SiO, and SO lines are good probes for this cloud. We also have performed deep OTF mapping observations of CO-0.40-0.22 with the NRO 45 m telescope in these probes. Spatial-velocity behaviors of these probes show that this cloud consists of an intense component with shallow velocity gradient and a less intense high-velocity wing. This kinematical structure can be explained by the gravitational kick of a molecular cloud by an invisible compact object with a mass of ˜105 M⊙.

The high velocity dispersion compact cloud CO-0.30-0.07 is a peculiar molecular clump discovered in the central molecular zone of the Milky Way, which is characterized by its extremely broad velocity emissions (∼145 km s-1) despite the absence of internal energy sources. We present new interferometric maps of the cloud in multiple molecular lines in frequency ranges of 265269 GHz and 276280 GHz obtained using the Submillimeter Array, along with the single-dish images previously obtained with the ASTE 10 m telescope. The data show that he characteristic broad velocity emissions are predominantly confined in two parallel ridges running through the cloud center. The central ridges are tightly anticorrelated with each other in both space and velocity, thereby sharply dividing the entire cloud into two distinct velocity components (+15 and +55 km s-1). This morphology is consistent with a model in which the two velocity components collide with a relative velocity of 40 km s-1 at the interface defined by the central ridges, although an alternative explanation with a highly inclined expanding-ring model is yet to be fully invalidated. We have also unexpectedly detected several compact clumps (≲0.1 pc in radius) likely formed by shock compression. The clumps have several features in common with typical starforming clouds: high densities (106.57.5cm-3), rich abundances of hot-core-type molecular species, and relatively narrow velocity widths apparently decoupled from the furious turbulence dominating the cloud. The cloud CO 0.300.07 is possibly at an early phase of star formation activity triggered by the shock impact

It is widely accepted that black holes with masses greater than a million solar masses (M⊙) lurk at the centres of massive galaxies. The origins of such ‘supermassive’ black holes (SMBHs) remain unknown1, although those of stellar-mass black holes are well understood. One possible scenario is that intermediate-mass black holes (IMBHs), which are formed by the runaway coalescence of stars in young compact star clusters2, merge at the centre of a galaxy to form a SMBH3. Although many candidates for IMBHs have been proposed, none is accepted as definitive. Recently, we discovered a peculiar molecular cloud, CO–0.40–0.22, with an extremely broad velocity width, near the centre of our Milky Way galaxy. Based on the careful analysis of gas kinematics, we concluded that a compact object with a mass of about 105M⊙ is lurking in this cloud4. Here we report the detection of a point-like continuum source as well as a compact gas clump near the centre of CO–0.40–0.22. This point-like continuum source (CO–0.40–0.22*) has a wide-band spectrum consistent with 1/500 of the Galactic SMBH (Sgr A*) in luminosity. Numerical simulations around a point-like massive object reproduce the kinematics of dense molecular gas well, which suggests that CO–0.40–0.22* is one of the most promising candidates for an intermediate-mass black hole.

We discovered two small high-velocity compact clouds (HVCCs) in the HCN J = 4-3 and J = 3-2 maps of the central 20 pc of our Galaxy. Both HVCCs have broad velocity widths ({{Δ }}V≳ 40 km s-1) and compact sizes (d˜ 1 {pc}), and originate from the dense molecular clouds in the position-velocity space. One of them has a faint counterpart in a Paschen-α image. Their spatial structure, kinematics, and absence of luminous stellar object are compatible with the notion that each of the small HVCCs is driven by the plunge of an invisible compact object into a molecular cloud. Such objects are most likely inactive, isolated black holes.

CO-0.40-0.22 is a peculiar molecular cloud that is compact and has an extraordinary broad velocity width. It is found in the central molecular zone (CMZ) of our Galaxy. In this direction, there is another cloud with an H2O maser spot at a lower velocity. A collision with this low-velocity cloud could be responsible for the broad velocity width of CO-0.40-0.22. We performed phase-referencing very long baseline interferometry (VLBI) astrometry with VERA and detected the annual parallax of the H2O maser spot in the low-velocity cloud to be 0.33 ± 0.14 mas, which corresponds to a distance of {3.07}-0.91+2.22 {kpc} from the Sun. This implies that the low-velocity cloud is located in the Galactic disk on the near side of the CMZ.

Quasi-periodic oscillations (QPOs) are believed to be indirect evidence for black holes. Several authors have reported detections of QPOs from Sgr A*, the nucleus of our Galaxy, in infrared and X-ray wavelength during flare-ups. Miyoshi et al. (2011) reported a tentative detection of QPOs in the 43 GHz light curve of Sgr A* obtained with the Very Long Baseline Array (VLBA). To confirm their detection, we reanalysed their VLBA data very conservatively. The 43 GHz flux was calculated for every 15 seconds by assuming a two-dimensional Gaussian-shape spatial structure. The Lomb-Scargle periodogram of the 43 GHz flux just after a millimeter wave flare of Sgr A*, shows three apparent peaks at 10.2, 14.6 and 32.1 min. Two of them are barely consistent with the previously reported QPOs. Using the resonant oscillation model, we estimated the spin parameter of the Sgr A* black hole to be 0.56 assuming the mass of 4.3 × 106 M ⊙.

We have been conducting flux monitoring observations of Sgr A* at 8 GHz and 2 GHz using the NICT Kashima-Koganei VLBI system (109 km baseline) since mid-February 2013. The primary objective of the monitoring is a search for flux variation which is expected to be caused by the interaction between the G2 cloud and the accretion disk. Until 2013 September 22, we observed Sgr A* for 39 days, five hours on each day. Four quasars (NRAO 530, PKS 1622-253, PKS 1622-297, PKS 1921-293) were also observed as flux calibrators every 6 minutes. No significant change nor variation has been detected in the 8 GHz flux density of Sgr A*. The 8 GHz flux density was 0.81 ± 0.07 Jy (preliminary), while no significant 2 GHz emission was detected by our system. We will continue monitoring as often as possible until at least 2014 May. Copyright © International Astronomical Union 2014.

We have been monitoring the flux density of Sagittarius A* (Sgr A*) at 22 GHz since DOY=42 (11 Feb. 2013) with a sub-array of the Japanese VLBI Network in order to search the increase of 22-GHz emission from Sgr A* induced by the interaction of the G2 cloud with the accretion disk. The flux densities observed until DOY=322 (18 Nov. 2013) are consistent with the previously observed values before the approaching of the cloud. We have detected no large flare during this period. Copyright © International Astronomical Union 2014.

High-velocity compact clouds (HVCCs) are a population of molecular clouds which have compact appearance (d < 10 pc) and large velocity width (Δ V > 50 km s-1), and are found in the central molecular zone of our Galaxy. We performed a 3 mm band line survey toward CO-0.40-0.22, a spatially unresolved HVCC with an extremely large velocity width (Δ V ≃ 90 km s-1), using the Mopra 22 m telescope. We surveyed the frequency range between 76 GHz and 116 GHz with a 0.27 MHz frequency resolution. We detect at least 54 lines from 32 molecules. Many line profiles show a prominent peak at vLSR ~ 70 km s-1 with very large velocity width, indicating they are emitted by the HVCC. Detections of largish molecules are indicative of non-equilibrium chemistry. We extracted some prominent lines based on velocity structure, intensity ratios, and PCA analyses. Shock diagnostic lines (SiO, SO, CH3OH, HNCO) and dense gas probes (HCN, HCO+) appear to be prominent. Excitation analysis of CH3OH lines show an enhancement in Trot in the negative high-velocity end of the profile. These results suggest that CO-0.40-0.22 has experienced a shock, acceleration, compression, and heating in the recent past. Copyright © International Astronomical Union 2014.

We have performed unbiased spectral line surveys at the 3 mm band toward the Galactic circumnuclear disk (CND) and Sgr A∗ using the Nobeyama Radio Observatory 45 m radio telescope. The target positions are two tangential points of the CND and the direction of Sgr A∗. We have obtained three wide-band spectra that cover the frequency range from 81.3 GHz to 115.8 GHz, detecting 46 molecular lines from 30 species, including 10 rare isotopomers and 4 hydrogen recombination lines. Each line profile consists of multiple velocity components which arise from the CND, +50 km s-1and +20 km s-1giant molecular clouds (GMCs), and the foreground spiral arms. We define the specific velocity ranges that represent the CND and the GMCs toward each direction, and classify the detected lines into three categories: the CND, GMC, HBD types, based on the line intensities integrated over the defined velocity ranges. The CND and GMC types are the lines that mainly trace the CND and the GMCs, respectively. The HBD types possesses the both characteristics of the CND and GMC types. We also present lists of line intensities and other parameters, as well as intensity ratios, which must be useful to investigate the difference between the nuclear environments of our Galaxy and others.

We report the results of millimeter-wave molecular line observations of the Tornado Nebula (G357.7-0.1), which is a bright radio source behind the Galactic center region. A 15′ × 15′ area was mapped in the J = 1-0 lines of CO, 13CO, and HCO+ with the Nobeyama Radio Observatory 45 m telescope. The Very Large Array archival data of OH at 1720 MHz were also reanalyzed. We found two molecular clouds with separate velocities, V LSR = -14 km s-1 and +5 km s-1. These clouds show rough spatial anti-correlation. Both clouds are associated with OH 1720 MHz emissions in the area overlapping with the Tornado Nebula. The spatial and velocity coincidence indicates violent interaction between the clouds and the Tornado Nebula. Modestly excited gas prefers the position of the Tornado "head" in the -14 km s-1 cloud, also suggesting the interaction. Virial analysis shows that the +5 km s-1 cloud is more tightly bound by self-gravity than the -14 km s-1 cloud. We propose a formation scenario for the Tornado Nebula; the +5 km s-1 cloud collided into the -14 km s-1 cloud, generating a high-density layer behind the shock front, which activates a putative compact object by Bondi-Hoyle-Lyttleton accretion to eject a pair of bipolar jets. © 2014. The American Astronomical Society. All rights reserved..

We report the detection of extremely broad emission toward two molecular clumps in the Galactic central molecular zone. We have mapped the Sagittarius C complex (-0.°61 < l < -0.°27, -0.°29 < b < 0.°04) in the HCN J = 4-3, 13CO J = 3-2, and H13CN J = 1-0 lines with the ASTE 10 m and NRO 45 m telescopes, detecting bright emission with 80-120 km s-1 velocity width (in full-width at zero intensity) toward CO-0.30-0.07 and CO-0.40-0.22, which are high velocity compact clouds (HVCCs) identified with our previous CO J = 3-2 survey. Our data reveal an interesting internal structure of CO-0.30-0.07 comprising a pair of high velocity lobes. The spatial-velocity structure of CO-0.40-0.22 can be also understood as a multiple velocity component, or a velocity gradient across the cloud. They are both located on the rims of two molecular shells of about 10 pc in radius. Kinetic energies of CO-0.30-0.07 and CO-0.40-0.22 are (0.8-2) × 1049 erg and (1-4) × 1049 erg, respectively. We propose several interpretations of their broad emission: collision between clouds associated with the shells, bipolar outflow, expansion driven by supernovae (SNe), and rotation around a dark massive object. These scenarios cannot be discriminated because of the insufficient angular resolution of our data, though the absence of a visible energy source associated with the HVCCs seems to favor the cloud-cloud collision scenario. Kinetic energies of the two molecular shells are 1 × 1051 erg and 0.7 × 1051 erg, which can be furnished by multiple SN or hypernova explosions in 2 × 105 yr. These shells are candidates of molecular superbubbles created after past active star formation. © 2014. The American Astronomical Society. All rights reserved..

We mapped molecular gas toward the supernova remnant W44 in the HCO + J = 1-0 line with the Nobeyama Radio Observatory 45 m telescope and in the CO J = 3-2 line with the Atacama Submillimeter Telescope Experiment 10 m telescope. High-velocity emission wings were detected in both lines over the area where the radio shell of W44 overlaps with the molecular cloud in the plane of the sky. We found that the average velocity distributions of the wing emission can be fit by a uniform expansion model. The best-fit expansion velocities are 12.2 ± 0.3 km s-1 and 13.2 ± 0.2 km s-1 in HCO+ and CO, respectively. The non-wing CO J = 3-2 component is also fit by the same model with an expansion velocity of 4.7 ± 0.1 km s-1. This component might be dominated by a post-shock higher-density region where the shock velocity had slowed down. The kinetic energy of the shocked molecular gas is estimated to be (3.5 ± 1.3) × 1049 erg. Adding this and the energy of the previously identified H I shell, we conclude that (1.2 ± 0.2) × 10 50 erg has been converted into gas kinetic energy from the initial baryonic energy of the W44 supernova. We also found ultra-high-velocity CO J = 3-2 wing emission with a velocity width of ∼100 km s-1 at (l, b) = (+34.°73, -0.°47). The origin of this extremely high velocity wing is a mystery. © 2013. The American Astronomical Society. All rights reserved..

This paper reports the discovery of a helical molecular cloud in the central molecular zone (CMZ) of our Galaxy. This "pigtail" molecular cloud appears at (l, b, V LSR) ≃ (-0°7, + 0°.0, - 70 to -30kms-1), with a spatial size of ∼20 × 20 pc2 and a mass of (2-6) × 105 M⊙. This is the third helical gaseous nebula found in the Galactic center region to date. Line intensity ratios indicate that the pigtail molecular cloud has slightly higher temperature and/or density than the other normal clouds in the CMZ. We also found a high-velocity wing emission near the footpoint of this cloud. We propose a formation model of the pigtail molecular cloud. It might be associated with a magnetic tube that is twisted and coiled because of the interaction between clouds in the innermost x 1 orbit and ones in the outermost x 2 orbit. © 2012. The American Astronomical Society. All rights reserved.

We present high-resolution CO J = 3-2 maps of the Galactic center region, taken with the ASTE 10m telescope. We have collected approximately 30,000 spectra with a 34″ grid spacing. The mapping area is roughly -1°.8 < l < +35 and -08 < b < +0°.9, which includes the central molecular zone and Bania's Clump 2, covering almost the full extent of the molecular gas concentration in the Galactic center. These CO J = 3-2 images show a behavior similar to the CO J = 1-0 images with the same resolution. Molecular gas in the Galactic center shows a higher J = 3-2/J = 1-0 intensity ratio (0.7) than the gas in the spiral arms in the Galactic disk (∼0.4). The CO J = 3-2/CO J = 1-0 luminosity ratio is 0.71. We see several regions with very high J = 3-2/J = 1-0 intensity ratios exceeding 1.5, including the Sgr A, l = +13, l = -04, and l = -1°.2 regions. A number of small spots of high ratio gas are also found. Many of them have large velocity widths, indicating that they are spots of hot molecular gas shocked by unidentified supernovae and/or winds from massive stars. © 2012. The American Astronomical Society. All rights reserved.