A nominal H2 column density of CW Leo can be calculated this way: assuming a constant mass loss rate
Mlr = 2x10^-5 M_sun/yr, constant outflow velocity Vexp = 14.5 km/s and outer radius of R = 10¹⁷ cm, we have
    N(H2) = Mlr * (R/Vexp) / PI / R^2 / m_H₂ 
          = Mlr / PI / R / Vexp / m_H2 
Here m_H₂ is mass of a H₂ molecule. Therefore, average H2 column density is a function of source size under consideration. If we are comparing a species that extends out to the whole CSE (with R = 10^17 cm, e.g., CO) with H2, then the average N(H2) = ~4x10^20 cm^-2 within that sphere. If we are comparing a centrally confined species (with R = 10^14 cm, e.g., CS,v=1) with H2, then the average N(H2) = ~4x10^23 cm^-2 within the small volumn.

background field star optical spectroscopy showed that the H+H2 column density at an offset of 37" from the central star of CW Leo is ~ 2x10²¹ cm^-2. DIB absorptions are weak. K I resonance lines are stronge, with broad line width of ~ 30 km/s (FWHM = 29.45 km/s from their spectral plot. my speculation: it means the size of the K I component of the CSE is >> 37", because the steady outflow velocity is 14.5 km/s.) (from Kendall et al., 2002A&A...387..624K)

Detected CO, 13CO, J=1-0 and HCN, H13CN, J=1-0 lines from CW Leo using the NRAO 36-foot 11m telescope. The source diameter determined by 5-point mapping is 136"+-30" for CO and 40" for HCN. (from Wilson et al., 1973ApJ...183..871W)

CO and 13CO, J=1-0 lines were observed using NRAO 11m telescope to CW Leo 1975 March. (from Kuiper et al., 1976ApJ...204..408K)

CO,J=2-1 was mapped using NRAO 11m telescope towards CW Leo, CIT 6, CRL 2688 and NGC 7027. The line intensity was unexpectedly high for all four stars. The FWHM source size was 54" for CW Leo (after deconvolution with a beam width of 26"). (from Wannier et al., 1979ApJ...230..149W)

NRAO 36 foot (11m) telescope detected CO, J=1-0. (from Cummings et al., 1980ApJ...235..886C)

The asymmetry of CO 1-0 and 2-1 line profiles was analysed. The existance of turbulance is responsible for the missing blue wing of optically very thick lines. (from Huggins & Healy, 1986ApJ...304..418H)

CO,J=7-6 was detected using bolometer of 2.24m telescope of the University of Hawaii during March of 1987. (from Wattenbach et al., 1988A&A...202..133W)

Strip maps of CO,J=1-0 and 2-1 were done using the Onsala 20m and NRAO 12m telescope. CO emission was found out to a radius of 120". The deconvolved FWHM source sizes were: 23" for CO,J=1-0 measured by OSO, 33" for CO,J=1-0 measured by NRAO 12m, 36" for CO,J=2-1 measured by NRAO 12m (calculated from their table 1 data). The best fit model had mass loss rate of 4x10^-5 Msun/yr and dust-gas momentum transfer efficiency of Q=2.35x10^-2, CO/H2 = 6X10^-4. (from Huggins et al., 1988ApJ...332.1009H)

IRAM 30m mapped CW Leo in CO,J=1-0 and 2-1. The deconvolved FWHM source sizes were 38.7" for J=1-0 and 16" for J=2-1. A two-shell model was used to derive better fitting to observational data: a hot core with r < 4.2", mass loss rate = 2.5x10^-5 Msun/yr, dust-gas momentum transfer efficiency Q = 8x10^-2; an extended outer region with r > 4.2", mass loss rate = 4x10^-5 Msun/yr, Q = 0.6x10^-2. (from Truong-Bach, 1991A&A...249..435T)

The JCMT single dish (beam size 15") mapping of CO,J=3-2 showed a FWHM source size of 23", so a Gaussian source size can be derived as 13". The maximum brightness temperature at the source center was about 38 K, lower than that of HCN (53 K), which was thought to be because CO is collisionally excited while HCN is mainly radiatively excited. (from Williams and White, 1992A&A...266..365W)

CO,J=3-2 and 13CO,J=3-2 were observed using CSO (10.4m). (from Wang et al., 1994ApJS...95..503W)

The ISO LWS spectrum (43-197um) of CW Leo show a lot of emission lines on the top of smooth continuum emission (mainly in the long wavelength part where continuum emission is weaker). They are identified to be high J rotational lines of CO (Jup = 14-39) and a forest of HCN rotational lines in the ground and  different vibrational states (nu1=1, nu2=1,2^0,2^2, nu3=1) with Jup = 18-48. The molecular emission comes from the hot and dense part of the CSE.(from Cernicharo et al., 1996A&A...315L.201C)

Asymmetry found at large scale. Exccess emission is found between -18.3 to -14.3 km/s in the red wing of the CO 1-0 line profile of IRC +10216 on June 25, 1996 and confirmed on January 15-16, 1997, but is absent on April 3-4, 1997 and June 3-4, 1998 (using IRAM 30 telescope). Comparing with other observations, they found that the CO 1-0 line shapes are very similar over most epoches with the exception that the June 1996 and January 1997 spectra clearly show the excess emission near the red wing. This red excess is found in CO 2-1 line profiles by quantitative comparison as well, but less prominent. This trantient feature could be interpreted as localized episodic ejection from the star. But this feature can be found only by high spectral resolution data (100 KHz or better). They performed spectral mapping of the CO J=1-0 and 2-1 and HCN 1-0 lines using the IRAM single dish in April 1997. Offset spectra show distinct features in -18 to -14 km/s (corresponding to the excess found in June 1997), -34 to -22 km/s, and -40 to -34 km/s ranges. There is good correspondace between CO 1-0 and 2-1 lines for these features. The spatial distribution of these features are not spherical. The peak of HCN 1-0 line is about 5-10 arsec away from the center, with a position angle of 45 deg. (Groenewegen & Ludwig, 1998A&A...339..489G)
(Figures from left to right: (1) Red part of CO 1-0 spectra, from top to bottom on June 1996, January 1997, April 1997 and June 1998; (2) comparing CO 1-0 (left) 2-1 (right) at different epoches: From top to bottom Sep. 1988, Aug. 1991, Dec. 1994, Jan. 1995, May 1996, June 1996, Jan. 1997, April 1997, June 1998; (3) Spectral maps of CO 1-0 upto 110 arcsec from the central star (78 kHz resolution), North is up, East is right; (4) same as (3) but for CO 2-1 (312 kHz resolution); (5) same as (3) but for HCN 1-0; )
__
__

JCMT single dish mapping of multi-transitions of CO and 13CO to CW Leo, together with detailed heat balance (dust-gas collision, photoelectric heating, cooling by 12CO, 13CO, HCN, H2 emission and adiabatic cooling) and LVG modeling, constrained the parameters of the star as: L* = (1.0-1.5)x10^4 Lsun, distance = 100-150 pc, mass loss rate = 1.5+-0.3 x 10^-5 Msun/yr, dust to gas ratio = 700+-100, 60um opacity = 250 cm^2/g, CO/H2 = 1.1x10^-3. But the fits to CO,J=3-2,7-6 and 13CO,J=6-5 were not satisfactory. Enhanced mass loss rate might be necessary to fit the maps at r > 50". (from Groenewegen et al., 1998A&A...338..491G)

BIMA interferometer masaic plus NRAO 12M single dish data revealed multiple shell like structure in the CO, J=1-0 map. CO is detectable out to radius of 140". The radii of the multiple shells range from 26"-120", while the inter-shell timescales range from 1300-2900 years. The shells could be formed by episodic enhancement of mass loss rate. (from Fong et al., 2003ApJ...582L..39F)
(figure: upper left -- masaic CO channel maps; middle -- radial intensity profile; lower right -- background subtracted CO channel maps.)

BIMA plus NRAO 12M combined imaging survey of CO,J=1-0 for a sample of AGB-PPN-PN sources were performed. For CW Leo, the CO emission extends out to a radius of 190". The FWHM diameter is 37.5" (measured from their PV diagram contour). The CO envelope looks quite spherical, only with a little elongation in S-N direction in the central region in the systemic velocity channel. The source morphologies were classified into three classes: spherical/elliptical/shell sources, disk sources, and structured outflow sources. The mass loss rates are correlated with the CSE sizes for circular and outflow sources, but not for disk sources whose source sizes are determined by disk evolution. (from Fong et al., 2006ApJ...652.1626F)
(figure: distribution of the 38 mapped sources on L-R diagram.)

They detected for the first time CO v=1, J=3-2, 2-1 line from CW Leo, using SMA. The line widths are around 8 km/s from their spectral plots. The 2" resolution didn't get the emission region resolved. (from Patel et al., arXiv:0811.4736)

The first detection of metal halides AlCl, Al³⁷Cl, KCl and possibly AlF was reported. The ³⁵Cl/³⁷Cl = 2.3+-0.5 (terrestrial value is 3.1). (from Cernicharo and Guelin, 1987A&A...183L..10C)

source size of AlF is 5"-10" (comparing CSO and IRAM spectra), N = (0.3-1.1)x10¹⁵ cm^-2, equivalent to a relative abundance AlF/H2 = 10^-9. MgF, CaF were searched but not found. It is suggested that F is produced not in SNe, but in the He flash of AGB stars. (from Ziurys et al., 1994ApJ...433..729Z)

AlCl/H2 = 2.2x10-7. Highberger  et al., 2001, ApJ, 562, 790

AlNC was detected in CW Leo for the first time. Source size is >20" and shell like (judged from the U shaped profile, very different from AlCl and AlF). Tex = 60K, Nrot = 9x10¹¹ cm^-2, AlNC/H2 = 3x10^-10. (from Ziurys et al., 2002ApJ...564L..45Z)

IRAM 30m detected Al37Cl, J=17-16 line and estimated an abundance of Al37Cl/H2 = (2-8)x10^-8, ten times higher than previous values. (from Ford et al., 2004ApJ...614..990F)

The first detection of metal halides AlCl, Al³⁷Cl, KCl and possibly AlF was reported. The ³⁵Cl/³⁷Cl = 2.3+-0.5 (terrestrial value is 3.1). (from Cernicharo and Guelin, 1987A&A...183L..10C)

The first detection of MgCN in CW Leo by observations with NRAO 12m and IRAM 30m telescopes. Column density was ~10^12 cm^-2. The abundance ratio MgNC/MgCN = ~22. It could be formed by reactions between Mg and HNC or CN or grain surface chemistry in the outer CSE:
    Mg⁺ + HNC => MgCNH⁺ + h*nu,
    MgCNH⁺ + e^- => MgCN + H.
MgCN may not be directed formed from MgNC, because of the large barrier for isomerization. Therefore, the MgNC/MgCN ratio may reflect the HNC/HCN ratio. (form Ziurys et al., 1995ApJ...445L..47Z)

The first identification of MgNC in IRC+10216 with the help of laboratory measurements. The column density of MgNC is 1-2x10^13 cm^-2, and Tex = 15 K. (from Kawaguchi et al., 1993ApJ...406L..39K)

PdBI imaging showed MgNC,N=8-7 in a patchy hollow shell of radius 15" (4.5x10¹⁶ cm). The shell is displaced by 2" from the compact mm/IR central source, indicating binary star. The similar shell radius of MgNC, C3H and C4H indicates that they may be formed though fast time dependent chemistry such as desorption from dust grains. (from Guelin et al., 1993A&A...280L..19G)
(figure: left upper -- MgNC channel map around Vlsr; left lower -- C4H channel map around Vlsr; right -- channel maps of MgNC.)

Detected 25MgNC and 26MgNC and possibly 26AlF in CW Leo. Isotopic ratios: 24Mg:25Mg:26Mg = 78:11+-1:11+-1 (similar as solar values: 79.0:10.0:11.0). This implies 3 Msun < Minit < 5 Msun.(from Guelin et al., 1995A&A...297..183G)

The first detection of metal halides NaCl, Na³⁷Cl was reported. The ³⁵Cl/³⁷Cl = 2.3+-0.5 (terrestrial value is 3.1). NaCl size is <= 15" by comparing the line profile and width with that of SiS and C4H.Tex and column density N for NaCl are 17K and 5x10¹² cm^-2 (for size of 15") or 16K and 3x10¹³ cm^-2 (for size of 3").(from Cernicharo and Guelin, 1987A&A...183L..10C)

NaCN/H2 = 10^-7. (size of 5") Guelin et al., 1997

BIMA interferometry mapping showed a ring like distribution of C₂H N=1-0 with a diameter of 20"+-4". The peak relative abundance is C2H/H2 = 1.8x10^-6 at radius of 6x10¹⁶ cm. It was concluded that C2H is solely a product of photodissociation from C₂H₂ at a radius > 3x10¹⁶ cm.  (from Bieging and Nguyen-Quang-Rieu, 1988ApJ...329L.107B)

OSO single dish mapping observation of C₂H,N=1-0 (87 GHz) with beam width of 44" revealed the FWHM source sizes of 43". The column density determined from modeling is 1.6x10^15 cm^-2. (from Truong-Bach et al., 1987A&A...176..285T)

Several lines of C2H around 87 GHz were detected in the first run of IRAM 30m telescope during May 1985. (from Lucas et al., 1986A&A...154L..12L)

NRAO 11m + OSO 20m + Bell 7m single dish observation identified C₃H with Jup = 7/2, 9/2, 13/2, 15/2 in ^2^PI_1/2 ladder and Jup=9/2 in ^2^PI_3/2 ladder in CW Leo. With fine structure level splitting, it was found that Tex = 8 K within each ²PI_1/2 ladder and 52 K across ladders. The reason was thought to be that far-IR transitions across ladders are only weakly allowed. The source size is about 84", column density is 2.8x10^13 cm^-2. The excitation of C₃H is similar as that of SiCC in this star. (The source size is determined by two beam method: the line temperature in the 40" beam of Onsala telescope is 2.4 times tronger than in the 130" beam of the Bell telescope, assuming uniform intensity distribution, a source size of 84" were derived.) (my comments: if we assume the source has a gaussian intensity distribution, the FWHM source size would be 97".) The column density of C₃H is one order of magnitude smaller than that of C₂H and C₄H, and so they concluded that acetylenic chain radicals with one H atom and even number of C is more abundant than that with odd number of C. (from Thaddeus et al., 1985ApJ...294L..49T)
(figure: rotational diagram that shows the difference between the two fine structure states.)

PdBI imaging showed C3H,^2^PI_1/2,J=9/2-7/2 in a patchy hollow shell of radius 15" (4.5x10¹⁶ cm). The shell is displaced by 2" from the compact mm/IR central source, indicating binary star.(from Guelin et al., 1993A&A...280L..19G)

Quantum mechanical calculations using matrix Hartree-Fork model gave theoretical rotation constants of Be = 4753 MHz for C4H and Be = 4955 MHz for C3N. (from Wilson & Green, 1977ApJ...212L..87W)

Four pairs of C4H doublets were detected towards CW Leo. The rotational and spin doubling constants derived from the astronomical data are B0 = 4758.48 +-0.10 MHz and gamma = 38.7 +-1.0 MHz. The column density of C4H could be between 4x10^14 and 5x10^15 cm^-2. But the sign of gamma is still unknown due to the low quality of observed data. If gamma is positive, the upper component will be 10% stronger than the lower; if gamma is negative, the inverse is true. (from Guelin et al., 1978ApJ...224L..27G)

NRAO 36 foot (11m) telescope observed two pairs of well separated fine structure lines of C4H near 10 GHz. In principle, for each pair of fine structure lines, the high-frequency member to low-frequency member line ratio depends on the sign of the spin-rotation constant gamma: it is > 1 for gamma positive; <1 for gamma negative. The observed ratio is 1.06+-0.16 and 1.09+-0.23 for the N=10-9 and 11-10 pairs respectively, demonstrating that gamma is positive for C4H. (from Cummings et al., 1980ApJ...235..886C)

The pair of fine structure lines C4H,J=10-9 around 95 GHz were detected in the first run of IRAM 30m telescope during May 1985. (from Lucas et al., 1986A&A...154L..12L)

The first detection of fine structure rotational lines pairs from C4H,nu7=1^1 were reported. The nu7=1^1 state lies ~ 300 K about ground state.  Trot = 35+-5 K, N = 1.1(+-0.4)x10^15 cm^-2 (this is the value converted from the paper by adopting new dipole moment of 0.87D from CDMS). (NOTE: Now we know from CDMS database that these lines are from nu7=2^0 state!!). (from Guelin et al., 1987A&A...182L..37G)

Laboratory measurement of C4H rotational lines were performed and rotational constants and line frequencies were calculated for ground and vibrational states. 22 unidentified lines were assigned to C4H and its vibrational states. The lines detected by Guelin et al. (1987A&A...182L..37G) were now assigned to nu7=2^0. The low lying of the varous vibrational states of C4H indicates that the spin-orbital coupling is very strong and the coupling constant gamma is negative. The detected nu7 = 2^2 lines in CW Leo seem to be much weaker than nu7 = 1^1 and 2^0 lines, the pumping mechanism could be directly radiative pumping from the ground state to the nu7 = 1^1 and 2^0 states, because the nu7 = 2^2 is not radiatively connected with the ground vibrational state. (from Yamamoto et al., 1987ApJ...323L.149Y)

PdBI imaging showed C4H, N=10-9 in a patchy hollow shell of radius 15" (4.5x10¹⁶ cm). The shell is displaced by 2" from the compact mm/IR central source, indicating binary star. (from Guelin et al., 1993A&A...280L..19G)
(figure: upper -- MgNC channel map around Vlsr; lower -- C4H channel map around Vlsr)

The N=9-8 line of C4H was mapped by BIMA. The ring-like distribution has a diameter of 30", with brightness temperature of 0.88 K at the peak. The half-peak ring thickness was estimated to be 15". The peak abundance was derived to be C4H/H2 = 1.8x10^-6 at the peak (r=16.7"), a factor of 5 higher than previous model predictions. (from Dayal & Bieging, 1993ApJ...407L..37D)
(Figure: C4H Vlrs channel map.)

The rotational lines of the carbon chain anion C4H- were detected in CW Leo. The similarity of line profiles fo C4H and C4H- indicates the coexistence of the two in the CSE. For C4H, Trot = 35 K, N = 3x10^15 cm^-2; for c4H-, Trot = 23+-2 K, N = 7.1+-0.2x10^11 cm^-2, 4200 times lower than C4H. The parameters of C6H- were also calculated from their observations to be Trot = 27+-2 K, N = 4.1+-1.5x10^12 cm^-2. The neutral C6H (2PI3/2) has Trot = 31+-2 K, N = 6.6+-2.0x10^13 cm^-2. Therefore, the abundance ratios are C6H-/C6H = 1/16, C4H-/C6H- = 1/6. They also calculated the radiative association rates of electron with C4H and C6H: Kratt(C4H) = 1.6x10^-10 cm^3/s, Kratt(C6H) = 4x10^-8 cm^3/s. C8H- was predicted to be detectable in CW Leo because C8H has been detected and the radiative association rate is expected to be larger for larger molecules. (from Cernicharo et al., 2007A&A...467L..37C)

Four doublets in the ²PI_3/2 ladder of C5H were detected using IRAM 30m telescope. The Trot = 14 K for ²PI_1/2 ladder and 24 K for ²PI_3/2 ladder. The column density averaged over the 25" beam was ~10^13 cm^-2, 40 times lower than HC5N. The two fine structure states have different excitation temperture, and the excitation temperature of HC5N was found to be close to that of the ²PI_3/2 ladder, which was interpreted using multilayer excitation model, because the considered HC5N lines have similar energy levels as the considered C5H lines in the ²PI_3/2 ladder. (from Cernicharo et al., 1986A&A...167L...5C)
(figure: left -- rotational diagram of C5H, showing difference between the two fine structure states; right -- energy level diagram of the two states of C5H and that of HC5H that show the levels involved in observed transitions.)

N(C5H) = 1.5 x 10^14 cm^-2 (new determination with size of 25").(from Cernicharo et al., 1987A&A...181L...1C)

IRAM single dish spectral line profiles of C6H indicates a uniform disk source size of 25" (flat top with 30" beam and U shape with 22" beam). Rotation diagram analysis gave Trot = 30 K, N = 3 x 10^14 cm^-2 (in the two fine structure states: ²PI_3/2 and ²PI_1/2). The two fine structure states of C6H showed different but parallel distribution in the rotation diagram. This was explained as that the two states were not in equilibrium with each other. They also showed that the column densities of different species decreases with increasing number of atoms (among CN, C3N, C2H, C3H, C4H, C5H, C6H, HCN, HC3N, HC5N, HC7N, HC11N, and HNC).(from Cernicharo et al., 1987A&A...181L...1C)
(figure: rotation diagram of C6H that show difference between the two fine structure lines.)

The rotational lines of the carbon chain anion C4H- were detected in CW Leo. The similarity of line profiles fo C4H and C4H- indicates the coexistence of the two in the CSE. For C4H, Trot = 35 K, N = 3x10^15 cm^-2; for c4H-, Trot = 23+-2 K, N = 7.1+-0.2x10^11 cm^-2, 4200 times lower than C4H. The parameters of C6H- were also calculated from their observations to be Trot = 27+-2 K, N = 4.1+-1.5x10^12 cm^-2. The neutral C6H (2PI3/2) has Trot = 31+-2 K, N = 6.6+-2.0x10^13 cm^-2. Therefore, the abundance ratios are C6H-/C6H = 1/16, C4H-/C6H- = 1/6. They also calculated the radiative association rates of electron with C4H and C6H: Kratt(C4H) = 1.6x10^-10 cm^3/s, Kratt(C6H) = 4x10^-8 cm^3/s. C8H- was predicted to be detectable in CW Leo because C8H has been detected and the radiative association rate is expected to be larger for larger molecules. (from Cernicharo et al., 2007A&A...467L..37C)

First detection of C7H in CW Leo was reported. It's concluded that the acetylenic chain radicals with even number of C are more abundant than that with odd number of C. (from Guelin et al., 1997A&A...317L...1G)

First detection of 10 lines of C8H in CW Leo was reported. (from Cernicharo & Guelin, 1996A&A...309L..27C)

Eight transition of H2CCCC (an isomer of diacetylene) between 80-135 GHz were detected by IRAM 30m in CW Leo. Tex = 20 +-3 K, column density averaged over a beam of 25" was N = (1.6+-0.4)x10^13 cm^-2. (from Cernicharo et al., 1991ApJ...368L..43C)

Detection of CH3CCH was reported. (from Agundez et al., 2008A&A...479..493A)

NRAO 36 foot (11m) telescope detected CN, J=1-0 series. (from Cummings et al., 1980ApJ...235..886C)

OVRO (beam 30") single dish mapping of CN in CW Leo and TMC-1, TMC-2. The deconvolved source diameter of CW Leo was about 70". The CN can be formed by photodissociation of HCN in such a chain: HCN + h*nu => CN, CN + h*nu => C I, C I + h*nu => C II. Model calculation for CW Leo (with Mlr = 1.5x10^-4 Msun/yr) showed that peak CN abundance occurs at r = 1.7x10^17 cm, and falls to 1/e at r = 2.8x10^17 cm. Modeled CN column density is 4.8x10^15 cm^-2, including those CN in the far radius where they are not excited. (from Wootten et al., 1982ApJ...257..151W)

The first detection of 13CN in space by Bell 7m telescope (FWHM beam size 1.8') were reported in CW Leo, Sgr B2 and Ori A. A isotopic ratio 12CN/13CN = 10 were derived (for the same source size as CN: 70"), indicating a column density of N(13CN) = (1-5)x10^14 cm^-2. (from Gerin et al., 1984A&A...136L..17G)

OSO single dish mapping observation of CN,N=1-0 (113 GHz) with beam width of 44" revealed the FWHM source sizes of 77". The column density determined from modeling is 5.4x10^15 cm^-2. (from Truong-Bach et al., 1987A&A...176..285T)

PdBI + IRAM 30m single dish mapped CS,J=2-1, SiS,J=5-4, CN,N=1-0 and SiCC,4(2,3)-2(2,2), found incomplete ring like distribution of CN and SiCC. The CN ring show an axis along PA of 20 degree. The ring size of CN measured from the contour map was 39". (from Lucas et al., 1995Ap&SS.224..293)
(Figure: maps of molecules.) 

BIMA interferometer mapping showed that CN appears in a hollow shell with diameter of 32". Modeling showed that the peak abundance of CN is 3.85x10^-6 at around 2.8x10^16 cm. (from Dayal & Bieging, 1995ApJ...439..996D)

Quantum mechanical calculations using matrix Hartree-Fork model gave theoretical rotation constants of Be = 4753 MHz for C4H and Be = 4955 MHz for C3N. (from Wilson & Green, 1977ApJ...212L..87W)

NRAO 43m telescope detection of C3N N=2-1 seems to be too weak than expected, compared with other transitions in literature. (from Bell et al., 1992ApJ...400..551B)

The pair of C3N,N=9-8 fine structure lines around 89 GHz were detected in the first run of IRAM 30m telescope during May 1985. (from Lucas et al., 1986A&A...154L..12L)

BIMA interferometer mapping of C3N,N=11-10 showed that it has a ring-like brightness distribution with diameter of 36". (from Bieging & Taffala, 1993AJ....105..576B)
(Figure: left -- C3N channel maps; right -- C3N integrated Vlsr map averaged for the spin doublets.)
_

BIMA interferometer mapping showed a ring like distribution of HNC,J=1-0 with a diameter of 34". The peak relative abundance is HNC/H2 = 2.3x10^-8 at a radius of 4x10¹⁶ cm. It was concluded that HNC is produced by photochemistry reactions involving HCN at radius > 3x10^16 cm. (from Bieging and Nguyen-Quang-Rieu, 1988ApJ...329L.107B)

HC14N and H13C14N J=1-0 lines were detected in CW Leo, using the NRAO 36-foot (11m) telescope. (from Morris et al., 1971ApJ...170L.109M)

Detected CO, 13CO, J=1-0 and HCN, H13CN, J=1-0 lines from CW Leo using the NRAO 36-foot 11m telescope. The source diameter determined by 5-point mapping is 136"+-30" for CO and 40" for HCN. (from Wilson et al., 1973ApJ...183..871W)

Onsala 20m telescope (FHPBM = 41"-52") mapping of HCN,J=1-0, HNC,J=1-0, H13CN,J=1-0 gave deconvolved gaussian source diameters of 41" for HCN, 34" for H13CN and 34" for HNC. The derived abundances are 1.1x10^-5 (assuming 12C/13C=40) for HCN, 2.75x10^-7 for H13CN, 3.9x10^-8 for HNC. (from Olofsson et al., 1982A&A...107..128O)

BIMA interferometer mapping of HCN,J=1-0 showed that there exists a wide non-guassian skirt in addition to centrally peaked spherical distribution. The FWHM source size is 20", and it extends to a FWHM of 45" at 2 K brightness contour. The excitation of HCN is controlled by radiative pumping through 14um vibrational band. There is a HCN density enhancement at a radius around 4x10^16 cm. The outer boundary of HCN is at around 10^17 cm, determined by photodissociation. (from Bieging et al., 1984ApJ...285..656B)

The possible maser line HCN,(0,2^0,0),J=1-0 around 89 GHz were detected in the first run of IRAM 30m telescope during May 1985. (from Lucas et al., 1986A&A...154L..12L)

The vibrationally exicted HCN lines (02^{0}0) and (01^{1c,1d}0),J=3-2 were observed in space for the first time (in Ori-KL and/or CW Leo). Through modeling, they found HCN/H2 abundance distribution in the CSE of CW Leo: x(0,1,0) = 3x10^-6 * r^-4 and x(0,2,0) = 7x10^-7 * r^-4.(from Ziurys & Turner, 1986ApJ...300L..19Z)

A survey of HCN millimeter maser towards 85 carbon stars resulted in 6 new HCN maser sources. All masers sources have mass loss rate between 10^-6 to 10^-5 Msun/yr and have similar IR properties. Time variation occurs on month scales. (from Lucas et al., 1988A&A...194..230L)

The strong possible maser emission HCN (01^{1c}0),J=2-1 was detected by IRAM 30m. They detected all HCN lines: v=0,J=J=3-2,2-1,1-0; (01^{1c,1d}0),J=3-2,2-1; (02^{0}0),J=3-2,2-1,1-0; (02^{2c,2d}0),J=3-2. The flux density is about 400 Jy. The possible maser pumping mechanism could be IR dust emission pumping or line overlap in IR. (from Lucas & Cernicharo, 1989A&A...218L..20L)

HCN and HCCCN were observed using OSO 20m, NRO 45m and CSO 10m telescopes towards 8 carbon stars (including CW Leo). The 14N/15N was found to be 5300 for CW Leo (assuming 12C/13C = 48). 29SiO/28SiO = 0.065+-0.004. (from Wannier et al., 1991ApJ...380..593W)

The JCMT single dish (beam size 15") mapping of HCN,J=4-3 showed a FWHM size of 20", so a Gaussian source size can be derived as 13". The maximum brightness temperature at the source center was about 53 K, higher than that of CO (38 K), which was thought to be because CO is collisionally excited while HCN is mainly radiatively excited. (from Williams and White, 1992A&A...266..365W)

IRAM PdBI interferometer mapping of HCN,(0,2^0,0),J=1-0 line in CW Leo didn't significantly resolve it. The continuum is marginally resolved. Gaussian fit of the continuum visibility curve gives a FWHM diameter of 1.1"+-0.2". Assuming an upper limit of source size of 1", the lower limit of brightness temperature of the line was 2000K, indicating a weak maser action. The measured continuum peak position was (RA,DEC,J2000): 09h47m57.394s, 13d16m43.65s to an accuracy of 0.05" in both directions. (from Lucas & Guilloteau, 1992A&A...259L..23L)

HCN,J=4-3 was observed using CSO (10.4m). (from Wang et al., 1994ApJS...95..503W)

BIMA interferometer mapping showed that the J=1-0 lines of HCN and H13CN appear concentrated towards the central star, with a FWHM diameter of 18" and 16.4" respectively. The column density of H13CN was derived by modeling to be 7.8x10^-7. Adopting a 12C/13C = 35+-5 from literature, the HCN abundance was obtained to be 2.7x10^-5. (from Dayal & Bieging, 1995ApJ...439..996D)

The ISO LWS spectrum (43-197um) of CW Leo show a lot of emission lines on the top of smooth continuum emission (mainly in the long wavelength part where continuum emission is weaker). They are identified to be high J rotational lines of CO (Jup = 14-39) and a forest of HCN rotational lines in the ground and different vibrational states (nu1=1, nu2=1,2^0,2^2, nu3=1) with Jup = 18-48. The molecular emission comes from the hot and dense part of the CSE. LVG model yields the best fit to the spectrum with HCN/CO = 0.1, corresponding to abundance of HCN/H2 = 3x10^-5. In far-IR, lambda>70um, the luminosity of HCN lines is 0.44 Lsun, higher than 0.28 Lsun of CO, and so HCN is the major coolant at FIR in CW Leo. (from Cernicharo et al., 1996A&A...315L.201C)
(figure: left -- ISO LWS spectrum; right -- continuum subtracted FIR line spectra.)

The HCN,(04^00),J=9-8 laser transition was detected in CW Leo. Because the lower energy level is 4200 K above the ground state, this laser must arise from the inner most part of CSE. The laser may be chemically pumped. A vibrational temperature of 1000K was derived by fitting line intensities from 01^{1}0,02^{0}0,02^{2}0,04^{0}0,11^{1}0 states.(from Shilke et al., 2000ApJ...528L..37S)

SiO,J=5-4,8-7 and HCN,J=3-2,4-3 were detected towards CW Leo and other M, S, C type AGB stars. V Cyg was found to show possible maser emission in HCN,(0,1^1c,0),J=3-2,4-3 lines. HCN in M stars is suggested to be produced in the central region by shock chemistry. HCN/SiO ratio increases with increasing mass loss rate for M and S stars. (from Bieging et al., 2000ApJ...543..897B)

HHT observations showed that HCN 01^{1c}0,J=3-2, 4-3 showed maser action in five carbon stars: R Scl, V384 Per, R Lep, Y CVn, and V Cyg (out of 12 survey stars). (from Bieging, 2001ApJ...549L.125B)

Laser emission near 891 GHz was found from the J=10-9 between 11^10 and 04^00 vibrational states of HCN towards three carbon stars CW Leo, CIT 6 and Y CVn. The laser line may arises from the very inner hot CSE region. Together with another laser near 805 GHz from the J=9-8 transition within the 04^00 state, they will be ideal targets for ALMA to resolve the dust formation region in future. Temporal variation in the maser lines is evident with repeated observations half year apart. (From Schilke & Menten, 2003ApJ...583..446S)

Sub-arcsec mapping of H13CN,J=8-7 in CW Leo by the SMA showed a diameter < 1". Excitation analysis showed the abundance of H13CN/H2 = 4x10^-7 in the inner wind (r<5x10^15 cm). Then adopting a 12C/13C = 50, the HCN abundance is 2x10^-5. The abundances do not change much out to the outer CSE. Also shown a single dish observations of J=1-0, 3-2, 4-3 and 8-7 lines. It was concluded that non-equillibrium chemistry in the extended atmosphere doesn't affect the abundance of H13CN much. (from Schoier et al., 2007ApJ...670..766S)

Onsala 20m telescope (FHPBM = 41"-52") mapping of HC3N,J=10-9 gave deconvolved gaussian source diameters of 30". The derived abundance is HC3N/H2 = 2.1x10^-6. (from Olofsson et al., 1982A&A...107..128O)

One line H13CCCN,J=11-10 around 97 GHz were detected in the first run of IRAM 30m telescope during May 1985. (from Lucas et al., 1986A&A...154L..12L)

HC3N,J=12-11 was mapped using Owens Valley Interferometer to show a ring distribution with diameter of 25". (no map shown). (from Likkel et al., 1987BAAS...19..755L)

BIMA interferometry mapping showed a ring + central component distribution of HC3N,J=10-9 with a ring diameter of 25"+-4". The peak relative abundance is HC3N/H2 = 1.5x10^-7 at peak radius of 4.5x10^16 cm. It was supposed that, in radius < 3x10^16 cm, HC3N may be made by neutral reaction 
    HCNH⁺ + C₂H₂ -> H₂C₃N⁺ + H₂, 
followed by dissociative recombination. But in radius > 3x10^16 cm, it could be created by radical reaction
    C₂H₂⁺ + HCN -> H₂C₃N⁺ + H,
where C₂H₂⁺ is photodissociation product of C₂H₂. Longer acetylene chains in the outer CSE can be made in the similar way as the second reaction. (from Bieging and Nguyen-Quang-Rieu, 1988ApJ...329L.107B)

HCN and HCCCN were observed using OSO 20m, NRO 45m and CSO 10m telescopes towards 8 carbon stars (including CW Leo). The 14N/15N was found to be 5300 for CW Leo (assuming 12C/13C = 48). 29SiO/28SiO = 0.065+-0.004. (from Wannier et al., 1991ApJ...380..593W)

BIMA interferometer mapping of HC3N,J=12-11 showed that it has a ring-like brightness distribution with diameter of 28". (from Bieging & Taffala, 1993AJ....105..576B)
(figure: left -- HC3N channel maps; right -- integrated Vlsr HC3N map)
_

Two beam method was used to determine the Gaussian source size of HC3N,Jup=8,9,10,11,12 to be 51". HC3N show warm and cool components on its rotation diagram, with Trot = 12.7 K and 48 K respectively. (from Bell, 1993ApJ...417..305B)
(figure: rotation diagram of HC3N that shows two components. Note that the two highest transitions were calculated with source sizes of 18.5" and 18.0" (for lower and higher), while the other lower transitions were calculated with biger source size of 40".)

IRAM 30m single dish mapping of HC3N, J=16-15, 18-17, and 24-23 showed centrally peaked distribution, quite different from lower transitions that showed ring-like distribution. The center of HC3N distribution is shifted from the shell center of C4H, as reported for MgNC observations by Guelin et al. (1993). It was concluded that HC3N abundance in CW Leo should be higher than previously determined and should peak at a smaller radius. The apparant source sizes are 35" for J=16-15, 34" for J=18-17, 25" for J=24-23. With the beam size of 13", the Gaussian source sizes can be derived as 32.5", 31.4"and 21.4" respectively.
(from Audinos et al., 1994A&A...287L...5A)
(Figures: The single dish maps of HCN,J=16-15,18-17,24-23 and C4H,J=nu7 2PI_3/2, left to right and top to bottom.)

HCCNC, J=9-8 was detected in CW Leo for the first time using ARO 12M telescope. The line intensity indicates column density of N=8.4x10^12 cm^-2. The abundance ratio HCCNC/HC3N = ~130, which is 2-7 times higher than in TMC-1. The detection of HCCNC means ion-molecule reactions are active in the CSE, while the large abundance ratio indicates other chemical processes such as neutral-neutral reactions also contribute to the production of HC3N. (from Gensheimer, 1997ApJ...479L..75G)

NRAO 43m telescope observation of HC5N in IRC +10216, Sgr B2 and TMC-1. Adopted C5H source size for IRC +10216 is 50" (from literature). Trot = 13(+12-4) K. No column density was given. (from Jennings and Fox, 1982ApJ...254..111J)

Comparison of IR and collisional excitation of HC5N indicates that IR excitation through bending mode bands (nu7, nu8) is more efficient than collision. IR excitation agree with observations in both high and low J. The abundance of HC5N was estimated to be HC5N/H2 = 2.5x10^-7. (from Deguchi & Uyemura, 1984ApJ...285..153D) 

NRAO 140 foot (43m) telescope didn't find the nu10 and nu11 bending state lines of HC5N. This was explained by the low excitation temperature of HC5N. Together with literature data, the rotation diagram of ground state lines showed two components: 
    a cooler components with diameter of 33", Trot = 13 K, N = 3.2x10^14 cm^-2 and 
    a hotter component with diameter of 50", Trot = 25 K, N = 3.7x10^14 cm^-2.
(from Bell et al., 1992ApJ...396..643B)
(figure: rotation diagram that shows two components. The temperatures of the dotted lines are adopted from literature.)

Two beam method was used to determine the Gaussian source size of HC5N to be 30". Trot = 35 K. (from Bell, 1993ApJ...417..305B)

Effelsberg telescope (beam size of 40" at 23.7GHz) observation of HC7N resulted in Tex = 133/FWHM^0.7 K and column density of 1.5x10^14 cm^-2. (from Nguyen-Q-Rieu et al., 1984A&A...138L...5N)

NRAO 43m telescope observation of HC9N in IRC +10216 revealed its Trot = 12+-3 K. Assuming a source size of 50", the column density can be derived as N = 4.0x10^13 cm^-2. (from Bell et al., 1992ApJ...400..551B)

Detection of HC11N was reported by the observation during March and May 1981 at the Haystack Observatory. The relative abundance is HC11N/H2 = 7.0x10^-8. HC11N was the heaviest molecule detected in space at that time. (from Bell et al., 1982Natur.295..389B)

Detection of CH2CHN was reported. (from Agundez et al., 2008A&A...479..493A)

Detection of CH2CN was reported. (from Agundez et al., 2008A&A...479..493A)

Detection of 4 rotational lines of HCP in CW Leo was reported by the observation in Juanuary of 2007 with the IRAM 30m telescope. This is the third phosphorus bearing species found in space (the first two are PN and CP). A rotational temperature of 61 K and a column density of 3x10^14 cm^-2 were derived by assuming a source diameter of 12". Chemistry and radiative transfer calculations reproduced the observed line profiles quite well. HC3P was predicted to be detectable in carbon rich CSE. (from Agundez et al., 2007ApJ...662L..91A)

Detection of 5 rotational lines of CCP in CW Leo was made by the ARO 12M observation . This is the fifth phosphorus bearing species found in space. Parameters: Nrot = 1.2x10^-12 cm^-2, Trot = 21 K, [CCP]/[H2] = 10^-9, source radius = 40" (from modeling). (from Halfen et al., 2008ApJ...677L.101H)

The J_k=1_0-0_0 line of PH3 was detected in IRC +10216 and CRL 2688 and marginally in VY CMa for the first time. The PH3 line in IRC +10216 show a composite profile: normal broad line (FWHM=28 km/s) with a narrow component (FWHM=4 km/s) at the center. If both components are thermal lines, the abundances of PH3 are 5x10^-8 in a region with r<150 R* and ~5x10^-7 in a region with r< 8 R*. (from Tenenbaum & Ziurys, 2008ApJ...680L.121T)

Effelsberg telescope (beam size of 40" at 23.7GHz) observation of NH3 (1,1) and (2,2) resulted in Tex = 133/FWHM^0.7 K and column density of 4.2x10^13 cm^-2. (from Nguyen-Q-Rieu et al., 1984A&A...138L...5N)

Odin detection of ortho-H2O 1_10-1_01 at 557 GHz and the first detection of ground level rotation line of ortho-NH3 1_0-0_0 at 572 GHz were reported. It confirmed the SWAS detection of the same ortho-H2O line. Abundances of ortho-H2O/H2 = 2.4x10^-6 and ortho-NH3/H2 = 1x10^-6 were derived. (from Hasegawa et al., 2006ApJ...637..791H)

Detection of new species in the CSE of CW Leo: CH2CHCN, CH2CN, H2CS, CH3CCH and C3O using IRAM 30m and ARO 12M telescopes. The abundances can be interpreted by gas phase chemistry in the outer CSE. (from Agundez et al., 2007Ap&SS.tmp..247A)

Detection of new species in the CSE of CW Leo: CH2CHCN, CH2CN, H2CS, CH3CCH and C3O using IRAM 30m and ARO 12M telescopes. The abundances can be interpreted by gas phase chemistry in the outer CSE. (from Agundez et al., 2007Ap&SS.tmp..247A)

Detection of new species in the CSE of CW Leo: CH2CHCN, CH2CN, H2CS, CH3CCH and C3O using IRAM 30m and ARO 12M telescopes. The abundances can be interpreted by gas phase chemistry in the outer CSE. (from Agundez et al., 2007Ap&SS.tmp..247A)

One line C33S,J=2-1 around 97 GHz were detected in the first run of IRAM 30m telescope during May 1985. (from Lucas et al., 1986A&A...154L..12L)

NRAO 12m telescope detected CS v=1 and v=0,J=5-4 lines from CW Leo. CS rotational lines are excited by absorption of 8um photons from dust and so CS lines are obviously varying with time. Abundance at the center were derived to be 2x10^-6 for v=0 state and 1x10^-7 for v=1 state. (from Turner, 1987A&A...182L..15T)

The JCMT single dish (beam size 15") mapping of CS,J=7-6 showed a FWHM size of 23", so a Gaussian source size can be derived as 17.4". The maximum brightness temperature at the source center was about 38 K, lower than that of HCN (53 K), which was thought to be because CO is collisionally excited while HCN is mainly radiatively excited. (from Williams and White, 1992A&A...266..365W)

CS,J=7-6 was observed using CSO (10.4m). (from Wang et al., 1994ApJS...95..503W)

PdBI + IRAM 30m single dish mapped CS,J=2-1, SiS,J=5-4, CN,N=1-0 and SiCC,4(2,3)-2(2,2), found incomplete ring like distribution of CN and SiCC. The CN ring show an axis along PA of 20 degree. The FWHM source size of CS,J=2-1 measured from the contour map was 7", then with the beam size of 3.6", a gaussian source size can derived as 6". A clumpy ring at radius of 15" can be seen.(from Lucas et al., 1995Ap&SS.224..293)
(Figure: maps of molecules.) 

Polarization of 1% was detected in CS, J=2-1 towards CW Leo. It could be caused by the non-sphericity of the CSE. PA of the polarization is -70 deg, roughly perpendicular to the elongation of the CSE. In another star CRL 2688, 5.1(+-1.5)% (1 sigma) and <0.9% (3 sigma) was found for CS and HCN respectively. (from Glenn et al., 1997ApJ...487L..89G)

The vibrationally excited transitions of CS,v=1,(J=2-1, 3-2, 5-4, 6-5, 7-6) have been detected towards CW Leo. All lines show narrow line width (~ 10km/s), and so should originate from the dust formtion region. The J=3-2 was particularly narrow (1~2 km/s) and could be a weak maser line. The narrow line widths indicate a source size of <0.5". The column density in v=1 state was (2-6)x10^16 cm^-2 (very uncertain because points on rotation diagram were on a curve, instead of a straight line!).(from Highberger et al., 2000ApJ...544..881H)

SMA observation of CS,J=14-13. The 680 GHz continuum emission mainly comes from photosphere, with a peak flux of 3.8 Jy agree well with previous observation at 89 and 242 GHz according to Snu ~ nu^2 for optically thick emission. CS showed a centrally peak image in the map with 2" resolution, but didn't show the 15" ring structure found by previous observation of J=2-1 line. The measured diameter on the Vsys map is 2.5", with beam of 2", a Gaussian source size is 1.5". (from Young et al., 2004ApJ...616L..51Y)

With an assumed source size of 25", IRAM 30m observation gave Trot(C2S) = 14+-3 K, N = 1.5 x 10^14 cm^-2. The column density of both C2S and C3S is 20-80 times lower than CS. This means a new carbon chain series C_nS had been detected, in addition to other 3 known chains: H(CC)_nCN, C_nH, C_nN. (from Cernicharo et al., 1987A&A...181L...9C)

With an assumed source size of 25", IRAM 30m observation gave Trot(C3S) = 28+-4 K, N = 1.1 x 10^14 cm^-2. The column density of both C2S and C3S is 20-80 times lower than CS. This means a new carbon chain series C_nS had been detected, in addition to other 3 known chains: H(CC)_nCN, C_nH, C_nN. (from Cernicharo et al., 1987A&A...181L...9C)

The C3S, J=4-3 was detected using NRAO 40 foot (43m) telescope. The rotation diagram of C3S show two components with Trot = 17 K and 47 K respectively. The column density was 6.7x10^13 cm^-2. A plot of HC(2n+1)N, C(n)S, C(n) was given to show the decreasing trends of column density with increasing number of atoms. (from Bell et al., 1993ApJ...417L..37B)
(figure: left -- rotation diagram C3S and C5S; right -- decreasing column density with number of atoms in species.)

Detection of new species in the CSE of CW Leo: CH2CHCN, CH2CN, H2CS, CH3CCH and C3O using IRAM 30m and ARO 12M telescopes. The abundances can be interpreted by gas phase chemistry in the outer CSE. (from Agundez et al., 2008A&A...479..493A)

IRAM 30m telescope measurement of SiC J=4-3, 6-5 by Cernicharo et al., 1989ApJ...341L..25C gave N_SiC = 6x10¹³ cm^-2 using a source diameter of 30". They mentioned that the Tex might be different within and between Omega ladders. (from Cernicharo et al., 1986A&A...167L...9C)

The mm spectral of SiC radical in its 3PI ground state was discussed and detected in CW Leo for the first time. The source diameter of the detected emission is larger than 54". (from Cernicharo et al., 1989ApJ...341L..25C)

Nine of previously unidentified lines were identified to the first radical ring molecule SiCC. (from Thaddeus et al., 1984ApJ...283L..45T)

The first detection of cm line of SiCC: 1(1,0)-0(0,0) was reported. (from Snyder et al., 1985ApJ...290L..29S)
(figure: the rotation diagram)

IRAM 30m detected lines from ²⁹SiCC and ³⁰SiCC and their spectroscopic parameters were calculated. Isotopic ratios had been determined: ²⁸Si/²⁹Si = 19.0+-1.5, ²⁸Si/³⁰Si = 30.7+-2.3, ²⁹Si/³⁰Si = 1.44+-0.15. (from Cernicharo et al., 1986A&A...167L...9C)

One line SiCC,4(2,2)-3(2,1) around 95 GHz were detected in the first run of IRAM 30m telescope during May 1985. (from Lucas et al., 1986A&A...154L..12L)

Astronomical and laboratory study of Si13CC. (from Cernicharo et al., 1991A&A...246..213C)

Nobeyama Millimeter Array observation of SiCC,4(0,4)-3(0,3) show an incomplete shell structure with a diameter of 30". Chemistry calculations showed that SiCC is mainly produced in the shell region by ion-molecule reactions.(from Takano et al., 1992PASJ...44..469T)
(figures: left -- channel maps of SiCC; right -- Vsys map between -22.7 and -31.7 km/s)
_

Use BIMA mapped SiCC 4_2,2-3_2,1 at 95.57936 GHz. The emission ragion is an asymmetric shell with rough radius of 14". V_LRS = -23.8 km/s. (from Gensheimer et al., 1992ApJ...388L..31G)

PdBI + IRAM 30m single dish mapped CS,J=2-1, SiS,J=5-4, CN,N=1-0 and SiCC,4(2,3)-2(2,2), found incomplete ring like distribution of CN and SiCC. The CN ring show an axis along PA of 20 degree. The ring size of SiCC measured from the contour map was 27". There is also a central component in the SiCC map. (from Lucas et al., 1995Ap&SS.224..293L)
(Figure: maps of molecules.) 

BIMA mapping was combined with NRAO 12M single dish data to show that SiCC,4(2,2)-3(2,1) in CW Leo show a shell-like distribution. (They adopted a distance of 290 pc.) The SiCC emission is distributed in a shell with inner and outer radius of ~2x10^16 cm (4.6") and ~6x10^16 cm (13.8"). Abundance is SiCC/H2 = 1.4x10^-6. (from Gensheimer et al., 1995ApJ...439..445G)
(figure: The conbined channel maps showing shell structure with bipolar symmetry.)

Tentative detection of 5 rotational lines from SiCC,nu3=1 state was reported. The abundance ratio N(nu3=1)/N(nu3=0) < 0.024. (from Gensheimer & Shyder, 1997ApJ...490..819G)

Seven rotational lines of the planar and highly polar ring molecule SiC3 were detected in CW Leo using the ARO 12M telescope. Assuming a shell with diameter of 40", the column density was found to be 4.3x10^12 cm^-2. Rotational temperature is about 10-20 K within K-ladders and 50 K across K-ladders, similar as SiCC. (from Apponi et al., 1999ApJ...516L.103A)
(figure: the rotation diagram of SiC3)

SiS J=6-5 and 5-4 lines are detected for the first time in CW Leo using the NRAO 36-foot 11m telescope. Also detected is the SiO,J=2-1 line. (from Morris et al., 1975ApJ...199L..47M)

The SiS, v=0, J=1-0 and 2-1 lines were detected in CW Leo. The blue peak of J=1-0 line is too strong than prediction and maser emission is suspected. (from Grasshoff et al., 1981A&A...101..238G)

Onsala 20m telescope (FHPBM = 41"-52") mapping of SiS,J=5-4 gave deconvolved gaussian source diameters of 20"-19". The derived abundance is SiS/H2 = 2.8x10^-6. An isotopic ratio was derived as 28Si/29Si = 16+-3 (1 sigma).(from Olofsson et al., 1982A&A...107..128O)

The SiS, J=1-0 maser was found in CW Leo. The maser appear at the blue edge of the broad thermal line and the intensity increase with higher spectral resolution. (from Henkel et al., 1983ApJ...267..184H)
(Figure: line profile of the mm SiS maser)

Multiple transitions of SiS were observed in CW Leo and modelled. IR excitation in inner CSE and photodissociation in the outer CSE are necessary to get good fit. The excitation model of SiS lines predicts time variation. SiS, J=1-0 was found to be maser. The column density is SiS/H2 = 2.4x10^-7. Worse fit of several lines could be explained by IR line overlap with HCN and C2H2. The abundance of CS and SiS in several C stars demonstrated that Si atoms could be all condensed into SiC dust grains while S atoms are not. (from Sahai et al., 1984ApJ...284..144S)

29SiS,J=5-4 and 30SiS,J=5-4 were detected in the first run of IRAM 30m telescope during May 1985. An isotopic ratio of 29Si/30Si =  2.1+-0.5 was derived. (from Lucas et al., 1986A&A...154L..12L)

NRAO 12m telescope observed SiS,v=1,J=12-11, 13-12, 14-13 of CW Leo. All lines are narrower than 29 km/s, and so solely arise from the acceleration zone (smaller than 8 stellar radii, say, r < 1x10^15 cm). The lines are consistent with high excitation temperature > 600 K, but maser action is also possible.  (from Turner, 1987A&A...183L..23T)

Model fitting of BIMA mapping and other single dish data showed that the SiS abundance is SiS/H2 = 7.5x10^-6 within a radius of 3x10^15 cm, but falls to 6.5x10^-7 beyond that. SiS begins to be photodissociated beyond r=4x10^16 cm and there is a cut off at 8x10^16 cm. From their maps, the directly measured source diameter is 21" for SiS,5-4, so deconvolution with the beam size of 8" gives FWHM = 19". For SiS,6-5, the map give a diameter = 20", deconvolusion with a beam size of 13" gives FWHM = 15".(from Bieging & NGuyen-Quang-Rieu, 1989ApJ...343L..25B)

SiS size = 14" (from Kahane et al., 1987, A&A, inpress)

BIMA interferometer mapping of SiS,J=6-5 showed that it has a centrally peaked brightness distribution with FWHM source diameter of 16". Modeling showed that the SiS gas abundance must has dropped drastically when it flow out beyond a radius of ~ 2x10^15 cm. (from Bieging & Taffala, 1993AJ....105..576B)
(figure: left -- SiS channel maps; right -- integrated Vlsr SiS map)
_

PdBI + IRAM 30m single dish mapped CS,J=2-1, SiS,J=5-4, CN,N=1-0 and SiCC,4(2,3)-2(2,2), found incomplete ring like distribution of CN and SiCC. The CN ring show an axis along PA of 20 degree. The FWHM source size of SiS measured from the contour map was 6.7". (from Lucas et al., 1995Ap&SS.224..293)
(Figure: maps of molecules.) 

The SiS,v=1,J=13-12 line was resolved by PdBI and a source size of 0.4" (7x10^14 cm, 10 stellar radii) was derived. (from Lucas, 1997Ap&SS.251..247L)

Maser lines were found in the SiS,v=0, J=11-10, 14-13, 15-14 line profiles towards CW Leo for the first time. These lines and maser could be pumped by IR line overlap with C2H2 and HCN in the inner CSE. Large number of other IR line overlaps indicates that maser action could be found in other higher transitions of SiS and even in isotopologues of SiS. (from Fonfria-Exposito et al., 2006ApJ...646L.127F)
(figure: maser and relatied line profiles and possible line overlaps.)

SiS J=6-5 and 5-4 lines are detected for the first time in CW Leo using the NRAO 36-foot 11m telescope. Also detected is the SiO,J=2-1 line. (from Morris et al., 1975ApJ...199L..47M)

Onsala 20m telescope (FHPBM = 41"-52") mapping of SiO,J=2-1 gave deconvolved gaussian source diameters of <20". The derived abundance is SiO/H2 = (1.1-2.7)x10^-7. (from Olofsson et al., 1982A&A...107..128O)

SiO,v=0,J=2-1 was mapped using BdBI in CW Leo, S star Chi Cyg and 10 M stars. The size of SiO emission region is of a diameter of about 20" in CW Leo. But the SiO lines in other stars seem to be from a region without outflow velocity (size doesn't change with velocity). (from Locus et al., 1992A&A...262..491L)

SiO,J=5-4,8-7 and HCN,J=3-2,4-3 were detected towards CW Leo and other M, S, C type AGB stars. V Cyg was found to show possible maser emission in HCN,(0,1^1c,0),J=3-2,4-3 lines. HCN in M stars is suggested to be produced in the central region by shock chemistry. HCN/SiO ratio increases with increasing mass loss rate for M and S stars. (from Bieging et al., 2000ApJ...543..897B)

SMA mapping of SiO, J=5-4 and multitransition by single dish in CW Leo showed that the abundance SiO/H2 = 1.5x10^-6 in the inner part of the CSE (3-8 R*), one order of magnitude higher than predicted from equillibrium photosphere chemistry. In addition to the compact component, a more extended region (r~2.4x10^16 cm) with lower abundance (1.7x10^-7) is necessary to fit the data. (from Schoier et al., 2006ApJ...649..965S)

Multiple transitions of SiO observed in CW Leo and other C stars showed that the abundence of SiO in these C stars can be several orders of magnitude higher than predictions from equillibrium stellar chemistry. This could be explained by shock chemistry and dust formation. The measured diameter on the map is 6", with the beam size of 2.6", a Gaussian source size = 5.4".
(from Schoier et al., 2006A&A...454..247S)

Five transitions of the first extraterrestrial silicon-nitrogen molecule SiN were detected in the outer CSE of CW Leo. The column density was N = 3.8x10^13 cm^-2. (from Turner, 1992ApJ...388L..35T)

The detection of SiCN lines in the ground fine structure state 2PI1/2 (each line with lambda doubling) by IRAM 30m were reported. The U-shaped line profile indicating a shell like pattern. The abundance was estimated to be SiCN/H2 = 4x10^-9, which is 20 time lower than MgNC. (from Guelin et al., 1993A&A...280L..19G)

Detection of SiCN in CW Leo by IRAM 30m telescope was reported. The abundance is estimated to be SiCN/H2 = 4x10^-9, about 20 times lower than that of MgNC. (from Guelin et al., 2000A&A...363L...9G)

SiNC and SiCN lines were detected by IRAM 30m in CW Leo. The excitation temperature for 15-30 K for SiCN and ~18 K for SiNC. The column density of SiNC is 2x10^12 cm^-2, similar to that of SiCN. Possible ways to make SiNC/SiCN are 
(1) neutral gas reaction:
    Si + HCN => SiCN + H
    Si + HCN => SiNC + H
(2) grain surface chemistry: reaction of Si with HCN on Si substrate can efficently produce both SiCN and SiNC at a low temperature of 12K. (from Guelin et al., 2004A&A...426L..49G)

IRAM 30m detected H2CO lines in CW Leo for the first time. The abundance was H2CO/H2 = 1.3(+1.5-0.8) x10^-8 and H2CO/H2O = 1.1+-0.2x10^-2, in accord with formaldehyde abundance in solar comets. But H2CO in CW Leo could be produced by either comets or photochemistry in the outer envelope, because the source size of H2CO was found to be extended by preliminary mapping. (from Ford et al., 2004ApJ...614..990F)

Odin detection of ortho-H2O 1_10-1_01 at 557 GHz and the first detection of ground level rotation line of ortho-NH3 1_0-0_0 at 572 GHz were reported. It confirmed the SWAS detection of the same ortho-H2O line. Abundances of ortho-H2O/H2 = 2.4x10^-6 and ortho-NH3/H2 = 1x10^-6 were derived. (from Hasegawa et al., 2006ApJ...637..791H)

ARO 12m and IRAM 30m observations detected C3O lines in CW Leo for the first time. The U-shaped line profile indicating extended distribution of C3O in a shell. The total column density was N(C3O) = 1.2x10^12 cm^-2. Model predicts that C3O shell has a ourter radius of ~ 30". (from Ford et al., 2004ApJ...614..990F)

Detection of new species in the CSE of CW Leo: CH2CHCN, CH2CN, H2CS, CH3CCH and C3O using IRAM 30m and ARO 12M telescopes. The abundances can be interpreted by gas phase chemistry in the outer CSE. (from Agundez et al., 2007Ap&SS.tmp..247A)

Line observations using IRAM 30m telescope during Oct. 1985 to May 1987. T_mb = T_A^*/0.6.

C³³S, Si³³S and HN¹³C are first detected in IRC +10216.

Only optically thin lines (showing U-shaped profile or round-topped profile but weak) can be used to calculate isotopic ratio.

Three effects that make line intensity raio different from isotopic abundance ratio: 

Isotope ratios:
-- ²⁸Si/²⁹Si = 18.7(+1.3,-1.0) (terrestrial: 19.6) using SiS J=5-4, SiCC.
-- ²⁸Si/³⁰Si = 28.6(+1.8,-1.5) (terrestrial: 29.8) using SiS J=5-4, SiCC.
-- ²⁹Si/³⁰Si = 1.47(+0.11,-0.09) (terrestrial: 1.52) using SiS J=5-4,8-7, SiCC, SiO.
-- ³²S/³⁴S = 20.2(+2.6,-2.1) (terrestrial: 22.5) using SiS J=5-4.
-- ³²S/³³S = 100(+23,-16)    (terrestrial: 125 ) using SiS J=5-4,8-7.
-- ³⁴S/³³S = 5.7(+0.8,-0.6)  (terrestrial: 5.5) using SiS J=5-4,8-7, CS J=2-1.
-- ¹²C/¹³C = 47(+6,-5)       (terrestrial: 89) using C³⁴S/¹³CS J=2-1,3-2 with above ³²S/³⁴S value
-- ¹⁴N/¹⁵N > 4400            (terrestrial: 277) using H¹³CN/HC¹⁵N

CO, CS and HC3N main isotopic lines are optically thick and can not be used to constrain ¹²C/¹³C ratio. The most accurate way is using double isotopic line ratio C³⁴S/¹³CS, other usuable optically thin lines are: CCH/C¹³CH, HNC/HN¹³C, SiCC/Si¹³CC.

The fact that CCH/C¹³CH, HNC/HN¹³C, SiCC/Si¹³CC give similar ¹²C/¹³C ratio as C³⁴S/¹³CS demonstrated that isotopic fractionation is not important for ¹³C in IRC+10216.

¹³C is enriched by a factor of 2 and ¹⁵N supressed by at least a factor of 15 w.r.t. solar abundance. This is strong evidance for processed material. All other isotopes are near to solar ratios.

CO spectra observations of giant molecular clouds by 7 m Crawford-Hill radio telescope during winter and spring of 1979 and 1980 April-May. 

¹⁸O/¹⁷O = 3.65+-0.15 for Galatic disk, 3.5+-0.2 for Galactic center. (uniform in the whole Galaxy within ~5% variation)
¹⁶O/¹⁸O = 675+-200  for Galactic disk, 250+-100 for Galactic center. (¹⁸O enhanced towards GC)
¹⁶O/¹⁷O = 2460+-750 for Galactic disk, 890+-370 for Galactic center. (¹⁷O enhanced towards GC)
¹⁸O/¹⁷O = 0.33+-0.1 for IRC+10216 (Solar: 5.5) (10 times lower than disk! Solar 40% higher than disk!)
¹⁶O/¹⁸O = 915(+520-380) for IRC+10216 (Solar: 500) (18O is only slight diminished)
¹⁶O/¹⁷O = 302(+140-80) for IRC+10216 (Solar: 2750) (17O is strongly enhanced!)

The combined ratio [¹³CO/C¹⁸O]*[C¹⁸O/¹³C¹⁸O] was used to constrain ¹⁶O/¹⁸O ratio, where ¹³CO and C¹⁸O lines are strong enough to compare only a restricted velocity region to avoid opacity effect while C¹⁸O and ¹³C¹⁸O are compared in the whole velocity range. The velocity range for ¹³CO/C¹⁸O is determined by checking CO/¹³CO line ratio at different velocities, say, only in the velocity range where CO is strong w.r.t. ¹³CO, ¹³CO/C¹⁸O could be free of opacity effect.

The interstellar enhancement of ¹⁸O/¹⁶O towards GC controdicts with the idea that it is mainly produced by by equillibrium hydrogen buring CNO cycle in red giants in which ¹⁸O has very low abundance due to nuclear reactions (see below).

The interstellar enhancement of 17O/16O towards GC may reflect the distribution of stellar populations.

Uniform spatial distribution of ¹⁸O/¹⁷O indicates that the slight destruction of ¹⁸O in red giants has little effect to interstellar ¹⁸O.

The 40% higher ¹⁸O/¹⁷O remains to be an enigma due to large uncertainty in the data.

¹⁸O is produced by ¹⁴N(alpha, gamma)¹⁸F followed by beta+ decay in the hydrogen burning CNO tri-cycle, but only mantain a low equilibrium abundance due to the reaction ¹⁸O(p,alpha)¹⁵N. It can be produced in low temperature regions only, because it will be rapidly destroyed by the reaction ¹⁸O(alpha, gamma)²²Ne when the temperature exceeds ~10⁸ K. Therefore, ¹⁸O should be mainly produced in the core of stars with M < 2M_sun or in red giants, whereas supernovae are not major contributors to the interstellar ¹⁸O.

¹⁷O is produced by both equilibrium and non-equilibrium hydrogen burning.

IRAM 30m telescope line observations of ¹²CO, ¹³CO, C¹⁷O, C¹⁸O during August of 1988. Tmb = TA* / eta (eta = 0.60 around 115 GHz, 0.45 around 230 GHz)

Typical circumstellar envelope line shapes: 
  round top    -- optically thick and unresolved; 
  flat top     -- optically thin and unresolved or optically thick but resolved; 
  double peaks -- optically thin and resolved.
The optical thickness effect can be appraised by comparing the line intensities channel by channel, because the optical thickness increase from line center channels to line edge channels.

Isotopic fractionation happens in the outer envelope where the temperature is low (<35K) and selective photodissociation also happens in the outer envelope where the optical depth is low.

chemical fractionation of 13C: ¹³C⁺ + ¹²CO -> ¹²C⁺ + ¹³CO favors the formation of ¹³CO for temperature below 35K. (Watson et al., 1976ApJ...205L.165W) Another consequency is the increase of C⁺ abundance towards outer layer of the circumstellar envelope.

Determined isotope ratio: (¹²C/¹³C = 44+-3 from literature is used to help determine...)
¹⁶O/¹⁷O = 840(+230-170)
¹⁶O/¹⁸O = 1260(+315-240)
¹⁷O/¹⁸O = 1.47(+0.33-0.25) 

Evolutionary effects on C and O isotope ratios: 
-- Initial solar abundances: 
   ¹²C/¹³C = 89
   ¹⁶O/¹⁷O = 2750
   ¹⁶O/¹⁸O = 500
   ¹⁸O/¹⁷O = 5.5

-- The first dredge-up when a star ascendes RGB brings up H-burning products: 
   ¹²C/¹³C drops to 20-30;
   ¹⁶O/¹⁷O slightly reduced for low mass stars, but can decrease to values as low as 100 for intermediate mass stars (1.4Msun < M < 3.4Msun).
   ¹⁶O/¹⁸O increases less dramatically (to value of 600-1000, depending on star mass)
   ¹⁷O/¹⁸O increases significantly (depending on star mass) 

-- The second dredge-up during the formation of C and O core after He core burning in a massive stars (M > 4 Msun) has little effect on C and O isotope ratios.

-- The third dredge-up during thermal pulses in AGB stage brings up CO core material to surface (less 16O than 12C). H shell burning that takes place between thermal pulses in massive stars converts some 12C into 13C. Therefore, in the star surface:
   ¹²C/¹⁶O increases
   ¹²C/¹³C increases
   ¹⁶O/¹⁸O may increase
   ¹⁶O/¹⁷O may increase
   ¹⁷O/¹⁸O only slightly changed

Observed 12C/13C ratios agree with above evolution theory: it's smaller than solar value but larger in C stars than in O-rich stars. (13C enriched in first dredge-up, but diminishes a little during the third dredge-up.)

Observed 17O/18O ratios agree with above evolution theory: 17O is enriched w.r.t. 18O in all five carbon rich stars (by a factor of 2-15). (17O significantly enriched during the first dredge-up and doesn't change any more.)

Observed 16O/17O is small (agree with first dredge-up) but too small w.r.t. other Carbon and S stars.

(More similar observation of O rich stars indicates O rich stars have 17O/18O <= 0.4, agree with ISM value, and so may be a major provider of interstellar 17O and 18O.)