大約百分之五十的螺旋星系具有棒的結構，此類星系被稱為棒旋星系。在許多棒旋星系的中心區域可以發現星遽增拱心環 。跟據理論預測，星系的棒狀結構能將分子雲輸入星系的中心區域。然而，彌散分子雲的聚集並不表示恆星能在此生成，因為恆星生成需要緻密分子雲作為恆星形成的原料。故彌散分子雲必須透過某些機制形成緻密分子雲以利恆星生成。 Kohno et al. （1999）提出一個模型來解釋緻密分子雲在星遽增拱心環中是如何形成。此模型的產生來自於 Kohno et al. （1999）對於棒旋星系NGC 6951在 CO (J = 1 - 0) and HCN (J = 1 - 0) 的觀測。在這個模型中，彌散分子雲沿著塵埃帶被注入星系中心，接著彌散分子雲在塵埃帶與星遽增拱心環的交接處匯集。然而，在此交會處的氣體擾動太大，以致於緻密分子雲無法在此生成，因此此區僅見 CO (J = 1 - 0)的輻射；部份彌散分子雲將會隨著星遽增拱心環的旋轉被帶入星遽增拱心環中，在星遽增拱心環中，氣體的彌散速度較小，故緻密分子雲可於此區形成。在此模型中，我們見到緻密分子雲的聚集點會位於塵埃帶與星遽增拱心環的交接處匯集的下游。為了應證這個模型，我們觀測另一個棒懸星系NGC 7552.，此星系具有與NGC 6951極為相似的外觀以及結構。

我們利用澳洲望遠鏡緻密陣列 （Australia Telescope Compact Array；ATCA）觀測HCN (J = 1 - 0) 和HCN+ (J = 1 - 0) 的輻射以了解緻密分子雲在星遽增拱心環中的所在處。同時我們也利用次毫米波陣列 （Submillimeter Array；SMA）觀測彌散分子雲的分布以了解拱心環中分子雲的物理性質。

首先，為了解NGC 7552的運動學，我們建立其旋轉曲線，並根據理論估計其中各林布萊德共振區的位置。結果顯示內外林布萊德共振區均可以大略與觀測所見的結構符合。接著我們利用Large Velocity Gradient (LVG) 來估計星遽增拱心環中分子雲的物理性質。結果顯示此區域內的分子雲的密度（nH2 ~ 10^3 – 10^6 cm-3）與溫度 （T > 100 K）均高，意味著這是一個有淺力形成恆星的地方，或是大質量恆星已誕生並加熱此區域內的分子雲。我們接著估計星遽增拱心環內的恆星形成率以及恆星形成效率，結果均顯示在NGC 7552中，恆星形成的活動主要由星遽增拱心環所主導，且其恆星形成的效率極高。

最後，我們討論緻密分子雲和恆星是如何在星遽增拱心環中形成。在我們觀測 HCN (J = 1 - 0) 的結果顯示 HCN (J = 1 - 0) 強度的峰值落在塵埃帶與星遽增拱心環的交接處，此結果與近年來NGC 6951具有較高解虛度與靈敏度的新觀測相符。我們故總結兩個緻密分子雲形成的機制。第一，彌散分子雲匯聚於塵埃帶與拱心環的交接處時，彼此間有較多機會碰撞而形成緻密分子雲。第二，緻密分子雲與恆星形成是藉由重力不穩定性導致分子雲有機會崩塌形成恆星。

在NGC 7552和NGC 6951中， HCN (J = 1 - 0) 和電波連續譜的輻射峰值的所在位置均有位移。我們可藉由此位移估計當今電波連續譜的輻射峰值由於拱心環旋轉，從當今 HCN (J = 1 - 0) 輻射峰值被被帶往現今電波連續譜的輻射峰值所在位置所需的時間。此時間應與大質量恆星形成的生命週期吻合，而此預測也在NGC 6951的觀測中得到驗證。然而在NGC 7552中，此時間竟約十倍小於大質量恆星形成的生命週期。可能的原因有二。首先，由於此區為星遽增區域，故其大質量恆星的生命週期較短。第二是系統誤差造成我們對於拱心環旋轉速度的錯誤估計。

Around 50% of spiral galaxies have bars. Many barred galaxies show circumnuclear starburst rings. Theory predicts that a bar can channel molecular gas towards the central region of the galaxy. However, the collection of molecular gas does not mean stars can form in the molecular clouds. It is because the clouds are not dense enough. Diffuse molecular gas channeled in the circumnuclear starburst ring has to become dense to form stars. Kohno et al. (1999) suggested a picture of formation of dense molecular gas in the circumnuclear starburst ring based on their CO (J = 1 - 0) and HCN (J = 1 - 0) observations of a barred galaxy NGC 6951. In the picture, diffuuse gas is channeled into the central region by the bar along the dust lanes. Gas collects at the intersections of the dust lanes and the circumnuclear ring. However, at the intersections, the gas are too turbulent to form dense molecular gas so stars cannot form at the intersections. Some diffuse gas will be driven into the circumnuclear ring where the velocity dispersion is smaller than that at the intersections. Dense molecular gas then can form through gravitational instability. In this case, the dense molecular gas is at the downstream of the intersections. Stars therefore can form in the dense molecular gas. We tested this picture with NGC 7552, which has almost identical morphology and orientation with NGC 6951.

NGC 7552 is a barred galaxy at a distance of ~22.1 Mpc in the southern hemisphere. The circumnuclear starburst ring is already seen in previous radio continuum and infrared observations. NGC 7552 has the same major morphological features to NGC 6951. NGC 7552 has a strong bar, which extends along the east-west direction. Two spiral arms are seen in NGC 7552. NGC 7552 has two dust lanes along the bar and there is circumnuclear starburst ring at the inner ends of the dust lanes.

To test the picture of Kohno et al. (1999), we need to find out where dense molecular gas forms in the circumnuclear ring. Therefore we observed HCN (J = 1 - 0) and HCO+ (J = 1 - 0) with Australia Telescope Compact Array (ATCA). To determine the physical conditions (density and temperature) of the gas, we also observed in 12 CO (J = 2 - 1) and 13 CO (J = 2 - 1) with the Submillimeter Array (SMA). Then physical conditions can be estimated from line ratio of HCN to CO.

In the first place, we constructed a rotation curve of NGC 7552 with our 12CO (J = 2 - 1) and HI data to understand the kinematics of this galaxy. With the derived rotation curve, we can theoretically predict locations of dynamical resonances in NGC 7552. The inner Lindblad resonances (ILRs) and the outer Lindblad resonance (OLR) are in rough agreement with the observations. The circumnuclear ring lies between the outer inner Lindblad resonance (oILR) and inner inner Lindblad resonance (iILR). The OLR is at or beyond the outermost of spiral arms, which is located at around 10 kpc. Then we calculated physical conditions of the molecular gas in the circumnuclear ring with the Large Velocity Gradient (LVG) approximation and our 13 CO (J = 2 - 1) and HCN (J = 1 - 0) observations. The results show that the molecular gas in the circumnuclear ring is dense (nH2 ∼ 10^3 to 10^6 cm −3 ) and warm (> 100 K). The dense molecular gas implies that the stars can form in the circumnuclear ring since the dense molecular gas is the material of star formation. The warm molecular gas indicates that massive stars indeed have formed in this region and heat the surrounding molecular gas. For these results, we estimated star formation rate (SFR) and star formation efficiency (SFE) in the ring. The SFR is around 10 to 20 Msun in the ring. Moreover, the SFR in NGC 7552 is dominated by the circumnuclear ring and the SFE of 5 × 10^ −9 year −1 is high comparing to normal galaxies and extreme starburst galaxies.

In the last part, we discuss how the dense molecular gas forms in the circumnuclear ring. The result of our HCN (J = 1 - 0) observation shows that HCN forms at the intersections of the dust lanes and the ring. This result conflicts to the conclusion of Kohno et al. (1999) in NGC 6951. At the same time, new higher angular resolution and sensitivity HCN (J = 1 - 0) observation shows that the HCN also forms at the intersections in NGC 6951. We therefore suggested two ways to form molecular gas and stars in the ring. First, the molecular gas accumulates at the intersections where the dust lanes and the ring meet. Giant molecular clouds (GMCs) have more chance to collide with each other to form dense molecular gas. Secondly, the dense gas and stars form through the gravitational instability. We calculated the Toomre Q parameter in the ring. The Toomre Q parameter of 0.14 suggests that the gas in the ring is gravitational unstable. Therefore the gas can collapse to form stars.

In both NGC 7552 and NGC 6951, there is a displacement between radio emission and HCN knots. The deprojected displacement is ~0.5" and ~3.0" in NGC 7552 and NGC 6951, respectively. We estimated the time it would take for the rotating circumnuclear ring to have carried the radio emission knots (the locations of present supernova emission) away from the present location of the HCN knots (the location of present dense molecular gas). The expectation of the timescale is approximately the lifetime of massive stars. The timescale is 3 × 10^6 year in NGC 6951. However, the timescale of 3 × 10^5 year in NGC 7552 is one order shorter than the lifetime of massive stars. The result suggests that the initial mass function (IMF) is top heavy in this starburst region in NGC 7552. The systemic error of derived rotational velocity of the ring may also lead to the shorter timescale in our galaxy.

[1] Aalto, S. et al. 1995, A&A, 300, 369
[2] Athanassoula, E. 1992, MNRAS, 259, 345
[3] Binney, J. & Tremaine, S. 1987, Galactic Dynamics
[4] Bonatto, C. Bica, E. & Alloin, D. 1989, A&A, 226, 2
[5] Brandt, J. C. 1960, ApJ, 131, 293
[6] Buta, R. 1986, ApJS, 61, 609
[7] Claussen, M. J. & Sahai, R. 1992, AJ, 103, 1134
[8] Clemens, M. S. & Alexander, P. 2001, MNRAS, 321, 103
[9] Charles, P. A. & Phillips, M. M. 1982, MNRAS, 200, 263
[10] Combes, F.& Gerin, M. 1985, A&A, 150, 327
[11] Condon, J. J. & Yin, Q. F. 1990, ApJ, 357, 97
[12] Condon, J. J. 1992, ARA&A, 30, 575
[13] Contopoulos, G. & Grosbol, P. 1989, A&ARv, 1, 261
[14] Dahlem, M. 2005, A&A, 429, 5
[15] Diaz, R. et al. 1999, ApJ, 512, 623
[16] Diaz, R. et al. 2003, ApJ, 597, 860
[17] Durrent, F. & Bergeron, J. 1988, A&AS, 75, 273
[18] Engelbracht, C. W. et al. 2005, ApJ, 628, 29
[19] Espada, D. et al. 2009, ApJ, 695, 116
[20] Feinstein, C. et al. 1990. A&A, 239, 90
[21] Ferguson, A. M, N, et al. 1996AJ, 111, 2265
[22] Forbes, D. A. et al. 1994a, AJ, 107, 984
[23] Forbes, D. A.; Kotilainen, J. K. & Moorwood, A. F. M. 1994b, ApJ,433, 13
[24] Freeland, E., Stilp, A.& Wilcots, E. 2009, AJ, 138, 295
[25] Gao, Y. & Solomon, P. M. 2004, ApJ, 606, 271
[26] Gallimore, J. F.; Baum, S. A. & O’Dea, C. P. 1996, ApJ, 464, 198
[27] Hameed, S. & Devereux, N. 1999, AJ, 118, 730
[28] Harwit, M. & Pacini, F. 1975, ApJ, 200, 127
[29] Heckman, T. M. 1980, A&A, 87, 152
[30] Helfer, T. T. & Blitz, L. 1995, ApJ, 450, 90
[31] Jackson, J. M. et al. 1993, ApJ, 402, 173
[32] Kalna js, A. J. 1973, PASAu, 2, 174
[33] Kenney, J. D. P. et al. 1993, ApJ, 418, 687
[34] Kennicutt, R. C., Jr. 1983, ApJ, 272, 54
[35] Kennicutt, R. C., Jr. 1984, ApJ, 287, 116
[36] Kennicutt, R. C., Jr. et al., 1994, ApJ, 435, 22
[37] Kennicutt, R. C., Jr. 1998, ARA&A, 36, 189
[38] Kennicutt, R. C., Jr. et al., 2003, PASP, 115, 928
[39] Kohno, K. et al. 1999, ApJ, 511, 157
[40] Koribalski, B. 1996, ASPC, 106, 238
[41] Kormendy, J. 1979, ApJ, 227, 714
[42] Krips, M. et al. 2007, A&A, 468, 63
[43] van der Kruit, P. C. 1971, A&A, 15, 110
[44] Laine, S. et al. 2002, ApJ, 567, 97
[45] Leitherer, C. & Heckman, T. M. 1995, ApJS, 96, 9
[46] Lin, C. C. & Shu, F. H. 1964, ApJ, 140, 646
[47] Matsushita, S. et al. 1998, ApJ, 495, 267
[48] Matsushita, S. et al. 2010, PASJ, 62, 409
[49] Mazzuca, L. M. et al. 2008, ApJS, 174, 337
[50] Miller, G. E. & Scalo, J. M. 1979, ApJS, 41, 513
[51] Neugebauer, G. et al. 1984, ApJ, 278, 1
[52] Oey, M. S. & Kennicutt, R. C., Jr. 1997, MNRAS, 291, 827
[53] Page, T. 1952, ApJ, 116, 63
[54] Pease, F. G. 1918, PNAS, 4, 21
[55] Perez-Gonzalez, P. G. et al. 2006, ApJ, 648, 987
[56] Planesas, P., Scoville, N.& Myers, S. T. 1991, ApJ, 369, 364
[57] Rieke, G. H. et al. 1980, ApJ, 238, 24
[58] Rubin, V. C.& Ford, W. Kent, Jr. 1970, ApJ, 159, 379
[59] Saikia, D. J. et al. 2002, A&A, 383, 98
[60] Sanders, D. B. & Mirabel, I. F. 1996, ARA&A, 34, 749
[61] Sarzi, M. 2007, MNRAS, 380, 949
[62] Scalo, J. M. 1986, FCPh, 11, 1
[63] Schinnerer, E. et al. 1997, ApJ, 488, 174
[64] Schinnerer, E. et al. 2000, ApJ, 533, 850
[65] Schmitt, H. R. et al. 2006, ApJS, 164, 52
[66] Schommer, R. A. et al. 1988, ApJ, 324, 154
[67] Scoville, N. Z. et al. 1993, PASP, 105, 1482
[68] Staveley-Smith, L. et al. 1996, PASA, 13, 243
[69] Storchi-Bergmann, T.; Wilson, A. S.; Baldwin, J. A. 1996, ApJ, 460,252
[70] Tacconi. L. J. et al. 1997, Ap&SS, 248, 59
[71] Tecza, M. et al. 2000, ApJ, 537, 178
[72] Telesco, C. M. & Decher, R. 1988, ApJ, 334, 573
[73] Toomre, A. 1964, ApJ, 139, 1217
[74] Veilleux, S. & Osterbrock, D. E. 1987, ApJS, 63, 295
[75] Ward, M. et al. 1980, MNRAS, 193, 563
[76] Weedman, D. W.; Feldman, F. R. & Sramek, R. A. 1980, BAAS, 12,872
[77] Zabludoff, A. I. & Mulchaey, J. S. 1998, ApJ, 496, 39