E
The Astrophysical Journal Supplement Series, 142:145–151, 2002 September
# 2002. The American Astronomical Society. All rights reserved. Printed in U.S.A.
ACETONE: LABORATORY ASSIGNMENTS AND PREDICTIONS THROUGH 620 GHz FOR THE
VIBRATIONAL-TORSIONAL GROUND STATE
Peter Groner
Department of Chemistry, University of Missouri–Kansas City, Kansas City, MO 64110-2499
Sieghard Albert,1 Eric Herbst,2 and Frank C. De Lucia
Department of Physics, Ohio State University, Columbus, OH 43210-1106
Frank J. Lovas
Optical Technology Division, National Institute of Standards and Technology, Gaithersburg, MD 20899-8441
and
Brian J. Drouin and John C. Pearson
Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 91109-8099
Received 2002 January 23; accepted 2002 April 16
ABSTRACT
The acetone molecule has a complex rotational spectrum arising from the internal rotation of two methyl
tops and their interaction with each other and the rigid-body rotation of the molecule. Here we report the
measurement and assignment of a large number of new rotational-torsional transitions of acetone in the
vibrational-torsional ground state at frequencies of up to 620 GHz. The newly measured lines have been combined with previously measured transition frequencies and fitted by a procedure used successfully for
dimethyl ether. From the fit to the global data set, we have extracted parameters that have enabled us to predict accurately the frequencies and intensities of over 10,000 lines of possible astronomical importance at frequencies of up to 620 GHz. The detection of interstellar acetone was first reported by Combes et al. based on
two transitions with partially resolved internal rotation structure. The present work will aid further interstellar studies to verify acetone as an interstellar constituent.
Subject headings: ISM: molecules — molecular data — molecular processes — radio lines: ISM
On-line material: machine-readable tables
dant interstellar ions. Making what appeared to be reasonable assumptions about the rates of the reactions and the
amount of methyl ion, Combes et al. (1987) were able to
show that the calculated abundance ratio of acetone to acetaldehyde is approximately equal to what they surmised to
exist in Sgr B2. The rate of reaction (1), however, had not
yet been studied in any detail.
In 1990, Herbst, Giles, & Smith (1990) investigated the
radiative association between methyl ions and acetaldehyde
theoretically, and they successfully tested their theory with
experimental results on the analogous ternary reaction in
which the collision complex is stabilized by an inelastic collision with a third body. They found that reaction (1) is much
slower than previously estimated and that the amount of
interstellar acetone produced by the sequence of reactions
(1) and (2) is far too little to explain the results of Combes et
al. (1987).
More recently, interferometric studies of Sgr B2 have
shown that saturated (hydrogen rich) oxygen- and nitrogencontaining organic molecules are typically, but not universally, confined to a particular region associated with highmass star formation known as Sgr B2(N-LMH) that was
originally called the ‘‘ compact region in Sgr B2(N) ’’ (Miao
et al. 1995; Mehringer et al. 1997; Pei, Liu, & Snyder 2000).
Sgr B2(N-LMH) is one of many small regions rich in saturated organic and inorganic molecules embedded in giant
interstellar clouds. Known as hot cores, these condensates
possess gas densities and temperatures in excess of standard
values—densities are typically 106 cm3 rather than 103–104
cm3, and temperatures are closer to 300 K than to 10 K
1. INTRODUCTION
Acetone (CH3COCH3), a molecule with two internal
rotors, was first reported to be in the interstellar medium by
Combes et al. (1987), who tentatively detected weak and
blended rotational lines in the source Sgr B2. They estimated a column density of 5 1013 cm2, corresponding to
an average fractional abundance with respect to H2 of
5 1011 . The suggestion that acetone had been detected in
Sgr B2 was given some further credence by Snyder et al.
(1997). Very recently, Snyder et al. (2002) reported the
detection of 13 new acetone emission features assigned to 20
acetone transitions based on the measured and calculated
frequencies of the present study.
Combes et al. (1987) suggested that their estimated abundance of acetone in Sgr B2 can be explained by gas-phase
formation and destruction pathways. Formation occurs by
the radiative association reaction between methyl ion and
interstellar acetaldehyde,
þ
CHþ
3 þ CH3 CHO ! CH3 COCH4 þ h ;
ð1Þ
followed by dissociative recombination with electrons,
CH3 COCHþ
4 þ e ! CH3 COCH3 þ H ;
ð2Þ
while destruction occurs by reaction with a variety of abun1 Present address: Laboratory for Physical Chemistry, ETH Zurich,
HCI Hönggerberg, CH-8093 Zurich, Switzerland.
2 Also at the Department of Astronomy, Ohio State University, Columbus, OH 43210-1106.
145
146
GRONER ET AL.
(Blake et al. 1987; Sutton et al. 1995; Ikeda et al. 2001; Liu,
Mehringer, & Snyder 2001). Although more distant than
the better studied hot-core sources in Orion (the Hot Core
and Compact Ridge), Sgr B2(N-LMH) is a good object for
the detection of complex molecules because it has so much
material. Whether or not acetone will prove to be a standard
hot-core molecule or to extend over a wider region like glycolaldehyde (Hollis et al. 2001) is not yet clear; the results of
Snyder et al. (1997), as well as the new results (Snyder et al.
2002), indicate that acetone is more abundant in the direction toward Sgr B2(N-LMH).
The picture that has emerged concerning molecular formation in hot-core regions is one in which grain surface
chemistry plays a major role in a previous era of low temperatures (10 K) before the onset of star formation. It is necessary to invoke surface processes because gas-phase ionmolecule chemistry is generally unable to produce saturated
species in sufficient abundance (Brown, Charnley, & Millar
1988). At low temperatures, hydrogen atoms, which migrate
rapidly upon sticking to surfaces, can probably convert condensed phase CO into methanol (Charnley, Tielens, &
Rodgers 1997); other surface processes leading to even more
complex saturated species have been discussed (Brown
1990; Hollis & Churchwell 2001). The grain mantles are
eventually lost via thermal evaporation from rising temperatures due to star formation or by sputtering caused by
intermittent shock waves. If methanol is the most complex
species formed on the grains, it can then be at least partially
converted into more complex species in the gas (Blake et al.
1987; Millar, Herbst, & Charnley 1991; Charnley, Tielens,
& Millar 1992; Caselli, Hasegawa, & Herbst 1993; Charnley
et al. 1995). This picture of the chemistry of hot-core regions
has not yet been put on a quantitative basis because of
uncertainties in the surface and gas-phase chemistry of production of molecules more complex than methanol and
because of an incomplete understanding of the role of
shock waves. It is therefore important to look carefully at a
saturated species such as acetone and determine how its
abundance varies spatially compared with other oxygencontaining organic molecules found in Sgr B2, including
HCOOH (formic acid), CH2 OHCHO (glycolaldehyde),
HCOOCH3 (methyl formate), and CH3 COOH (acetic
acid).
To observe acetone unambiguously in space requires
knowledge of its spectrum on the ground. The rotationaltorsional spectrum of acetone has been studied previously
in the laboratory at microwave frequencies by a variety of
groups (Weatherly & Williams 1952; Swalen & Costain
1959; Nelson & Pierce 1965; Peter & Dreizler 1965; White
1975; Oldag & Sutter 1992). In 1986, Vacherand et al.
(1986) extended the frequency range of the measured spectrum to 300 GHz (a wavelength of 1 mm). We have been
engaged in a program to measure the spectrum of acetone
into the submillimeter-wave region of the spectrum. Here
we report our completed work on the rotational spectrum
of acetone in its vibrational-torsional ground (VTG) state.
We have measured a large number of lines and blends
through 600 GHz with three different methods and fitted
these and most previously measured lines and blends to a
high degree of precision. The rotational-torsional parameters determined in the fit have been used to predict the frequencies and intensities of a large number (11,000) of
unmeasured lines through 620 GHz in frequency and
J ¼ 60 in rotational quantum number.
Vol. 142
2. THEORY
To fit the data and to predict the spectrum of acetone, an
effective rotational Hamiltonian for molecules with two
periodic large-amplitude motions (Groner 1997) was used.
Its application to the rotational spectrum of dimethyl ether
has been described (Groner et al. 1998). From a theoretical
standpoint (effective rotational Hamiltonian and group
theory), acetone and dimethyl ether are exactly alike. Both
have the b principal inertial axis coinciding with the C2 symmetry axis of the molecule, and both have two equivalent
methyl internal rotors whose axes are in the a-b principal
inertial plane. Therefore, the calculations for acetone were
performed in exactly the same manner with the same program. The principal difference between these molecules
occurs in the size of the barrier to internal rotation of a single methyl group. For dimethyl ether and acetone, the barriers are 903.4(70) cm1 (Lovas, Lutz, & Dreizler 1979) and
251.4(26) cm1 (Groner 2000), respectively (type A uncertainties with coverage factor k ¼ 1 are given in parentheses).
As a consequence, the effects of the interaction between
internal and overall rotation are much larger in acetone.
In the theoretical treatment of these interactions, the symmetry numbers 1 and 2 have been used to distinguish sets
of internal rotation basis functions. Because of the threefold
periodicity of the internal rotations of the methyl groups in
acetone, the symmetry numbers k can assume any of the
values 0, 1, or 2. They are related to the conventional designations of the torsional substates derived from the original
investigation of the symmetry properties (Myers & Wilson
1960). Of all possible combinations of 1 and 2 ,
(1 ; 2 Þ ¼ ð0; 0Þ is nondegenerate and corresponds to the
conventional AA notation for one of the torsional substates.
The combinations ð0; 1Þ, ð1; 0Þ, ð0; 2Þ, and ð2; 0Þ are fourfold degenerate and correspond to the EE substate. The
doubly degenerate pairs ð1; 1Þ, ð2; 2Þ and ð1; 2Þ, ð2; 1Þ correspond to the AE and EA substates, respectively. Because
of the interactions, the substates have slightly different rotational energy levels. This leads to rotational transitions that
are split into four different components, one for each substate. These quartets are not always completely resolved;
the components ð1; 1Þ and ð1; 2Þ, in particular, blend very
often into a single band resulting in characteristic triplets
for many transitions. For dimethyl ether, the resulting quartets are either collapsed into a single blended line or spread
out over up to 10 MHz. For acetone, the width of the quartet splitting extends from a few megahertz to more than
1 GHz.
As a consequence of the symmetry properties, the different rotational quantum numbers and torsional substates
have different spin statistical weights, which are factors that
take into account the degeneracies of the torsional substates
and the allowed nuclear spin functions. In the ð0; 0Þ substate the spin weights are 6 (Ka Kc : ee $ oo) and 10
(Ka Kc : oe $ eo), in the ð0; 1Þ substate the spin weight
is 16, in the ð1; 1Þ substate the spin weights are 2
(Ka Kc : ee $ oo) and 6 (Ka Kc : oe $ eo), and in the ð1; 2Þ
substate the spin weight is 4 (Myers & Wilson 1960).
3. EXPERIMENT AND SPECTRAL ANALYSIS
The experiments at Ohio State University were accomplished with the Fast Scan Submillimeter Spectroscopic
Technique (FASSST) system discussed by Petkie et al.
No. 1, 2002
ACETONE
(1997). This technique uses a voltage tunable backward
wave oscillator (BWO) as the primary source of radiation.
The BWO is scanned rapidly (104 MHz s1) through a
large frequency range, and frequencies are calibrated optically by use of a Fabry-Perot cavity. There is no need for
active frequency stabilization because of the fast sweep and
data acquisition. The detectors are liquid helium–cooled
InSb bolometers. With this system it is possible to measure
literally thousands of spectral lines per second, with a frequency accuracy of 0.1 MHz. Spectral transitions of acetone were measured in the regions 260–345, 486–515, and
592–611 GHz. About 750 transition frequencies and blends
(with rotational quantum number J of up to 60) belonging
to the VTG state were assigned.
The experiments at the Jet Propulsion Laboratory (JPL)
were accomplished with a klystron-based spectrometer driving a harmonic generator. Tone burst modulation and a
liquid helium–cooled InSb detector were used to produce
and detect second-derivative line shapes. Friedl et al. (1995)
described this spectrometer system in detail. Over 100 transition frequencies were measured in regions between 370
and 430 GHz, as well as in some gaps of the FASSST measurements between 290 and 316 GHz.
At the National Institute of Standards and Technology
(NIST), spectral measurements were carried out with a
Fabry-Perot cavity, Fourier-transform microwave spectrometer (Lovas & Suenram 1987; Suenram et al. 1989) of
the Balle-Flygare type (Balle & Flygare 1981). A pulsed solenoid valve was used to produce a supersonic molecular
beam from a mixture of about 1% by volume acetone
entrained in argon carrier gas at a total pressure of 100 kPa
(1 atm) behind a 0.5 mm nozzle orifice and injected at the
147
midpoint of the Fabry-Perot cavity and perpendicular to
the microwave standing wave field. Molecular beam pulses
with 200–400 ls duration were employed with repetition
rates of up to 35 Hz. The acetone beam was polarized by a
short microwave pulse when the microwave frequency was
near resonant (D < 400 kHz) with a rotational transition
of acetone. The free induction decay signal from the cavity
was digitized in 0.5 ls increments for 512 channels. Typically, 200 pulses were signal averaged, after a background
microwave pulse was subtracted from each signal pulse, to
yield signal-to-noise ratios of 100 or more. The averaged
data were Fourier transformed to obtain the power spectrum in the frequency domain with a resolution element of
3.9063 kHz point1. Molecular transitions observed had
line widths of 106 of the pump frequency, i.e., 10 kHz at 10
GHz, and the frequency measurement uncertainties were
estimated to be 4 kHz in most cases (type A with coverage
factor k ¼ 1), which was the resolution element for the digitization time described above. Some transitions exhibited
partially resolved spin-spin hyperfine structure, which
caused larger uncertainties in those measurements. A total
of 32 frequencies between 11 and 25 GHz were measured
with this technique.
The lines of the VTG state measured during this investigation as well as lines reported by previous investigators
(Peter & Dreizler 1965; Oldag & Sutter 1992; Vacherand et
al. 1986) and 35 unassigned microwave frequencies from a
NASA report (White 1975) have been combined into a
global data set of 1175 frequencies and blends. These lines
are listed in Table 1 along with the torsional substate, asymmetric top rotational quantum numbers J, Ka , Kc , and
source. (Table 1 can be found in its entirety in the electronic
TABLE 1
Assigned and Fitted Transition Frequencies of Acetone in the Vibrational-Torsional Ground State
ta
J0
Ka0
Kc0
J 00
Ka00
Kc00
Frequency
(MHz)
OC
(MHz)
Uncertaintyb
(MHz)
Sourcec
11........
12........
01........
00........
11........
00........
11........
01........
12........
01........
00........
12........
12........
11........
01........
00........
12........
11........
01........
00........
3
3
3
3
3
3
2
3
2
2
2
3
1
1
1
1
2
2
2
2
2
2
2
2
3
3
1
3
1
1
1
3
1
1
1
1
2
2
2
2
1
1
1
1
0
0
1
0
1
1
1
0
1
1
1
1
1
1
1
1
3
3
3
3
3
3
2
3
2
2
2
3
0
0
0
0
2
2
2
2
1
1
1
1
2
2
0
2
0
0
0
2
0
0
0
0
1
1
1
1
2
2
2
2
1
1
2
1
2
2
2
1
0
0
0
0
2
2
2
2
10730.960
10749.920
10751.662
10762.709
11073.970
11199.280
11252.502
11257.731
11265.255
11272.459
11286.144
11468.300
15038.511
15064.927
15074.075
15096.330
15615.732
15736.238
15750.855
15827.738
0.400
0.023
0.009
0.007
0.019
0.040
0.004
0.016
0.014
0.005
0.013
0.054
0.003
0.000
0.001
0.000
0.002
0.003
0.006
0.006
0.0
0.03
0.01
0.01
0.03
0.03
0.004
0.01
0.004
0.004
0.004
0.03
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
O
P
O
O
P
P
N
O
N
N
N
P
N
N
N
N
N
N
N
N
Other Sourcesc
P
P
P
P
P
P
O, P
P
P
P
O, P
O, P
P
P
O, P
O, P
Note.—Table 1 is published in its entirety in the electronic edition of the Astrophysical Journal Supplement. A portion is shown
here for guidance regarding its form and content.
a Torsional substates: 00 ¼ AA, 01 ¼ EE, 11 ¼ AE, 12 ¼ EA.
b Estimated experimental uncertainty (coverage factor k ¼ 1; see Table 2 for type A or B). A value of 0.0 in the uncertainties
column signifies that the transition has been assigned but not included in the fit. A total of 1002 assigned frequencies were used in
the fit.
c O: Oldag & Sutter 1992; P: Peter & Dreizler 1965; N: Fourier transform (FT) microwave data at NIST (this work).
148
GRONER ET AL.
Vol. 142
TABLE 2
Sources and Uncertainties of Frequency Data
Source
Type of Data
Code
Estimated Uncertainty
(coverage factor k ¼ 1)
(MHz)
Peter & Dreizler (1965)........
Vacherand et al. (1986)........
Vacherand et al. (1986)........
Oldag & Sutter (1992) .........
White (1975) .......................
NIST (this work).................
JPL (this work) ...................
FASSST (this work)............
FASSST (this work)............
Microwave
Double resonance microwave
Millimeter wave
FT Microwave
Microwave
FT microwave
Submillimeter wave (klystron)
Submillimeter wave <400 GHz
Submillimeter wave >400 GHz
P
V
V
O
W
N
J
F
F
0.03 (type B)
0.05 (type B)
0.1 (type B)
0.01 (type B)
As quoted (0.2–0.02; type B)
0.004 or 0.008 (type A)
0.1 (type B)
0.1 (type B)
0.2 (type B)
edition of the Astrophysical Journal Supplement.) The torsional substates are identified by the symmetry numbers (1,
2). Their relation to the conventional notation is discussed
in x 2. The origin of the frequency entered into the table is
indicated by the code in the column headed ‘‘ source.’’ Other
investigations with different or less precise frequencies are
coded in the column headed ‘‘ other sources.’’ The codes are
defined in Table 2. For lines without severe torsional-rotational interactions, the transitions obey b-type asymmetric top selection rules (DJ ¼ 0, 1; DKa ¼ 1; 3; . . .;
DKc ¼ 1; 3; . . .; see Gordy & Cook 1984), and the torsional substate does not change.
For large rotational quantum numbers J, the energy levels of asymmetric rotors are practically degenerate for equal
and high Ka (a-type degeneracy) or for equal and high Kc (ctype degeneracy, low Ka ). The c-type degeneracy persists for
all torsional components of acetone. The a-type degeneracy
holds only for the ð0; 0Þ and ð1; 1Þ torsional states, but not
for the ð0; 1Þ and ð1; 2Þ states. As a consequence, many of
the lines measured in the millimeter- and submillimeterwave region are superpositions of degenerate transitions. In
most instances of c-type degeneracies, only one of the
degenerate transitions has been used in the fit and listed in
Table 1.
Weights of individual measurements in the least-squares
fit based on the approach in x 2 were determined from the
estimated experimental uncertainties, which are also listed
in Table 1 and summarized for each source in Table 2. An
experimental uncertainty of 0.0 MHz marks assigned transitions whose frequencies were not used in the least-squares
fit.
As a result of these criteria, 1002 out of 1175 assigned frequencies were used to determine 33 spectroscopic parameters by the least-squares method. The calculated residuals
(observed minus calculated frequencies) are shown in Table
1. The spectroscopic parameters (‘‘ constants ’’) determined
by the least-squares procedure are listed in Table 3. The
dimensionless standard deviation achieved is 1.58. In addition to 14 nontunneling (q ¼ q0 ¼ 0) parameters corresponding to the rotational constants and the quartic and
sextic centrifugal distortion constants, the internal rotation
parameters (unitless) and (degrees) were determined, as
well as four energy tunneling parameters , eight tunneling
parameters associated with the rotational constants, and
TABLE 3
Spectroscopic Parameters for Acetone in the Vibrational-Torsional Ground State
Parameter
Valuea
Parameter
Valuea
...............................................
(deg)......................................
A (MHz) ..................................
B (MHz)...................................
C (MHz) ..................................
DJ (kHz) ..................................
DJK (kHz).................................
DK (kHz)..................................
J (kHz) ...................................
K (kHz)...................................
J (Hz) ....................................
JK (Hz) ..................................
KJ (Hz) ..................................
K (Hz)....................................
J (Hz).....................................
JK (Hz)...................................
K (Hz) ....................................
0.0621760(60)
25.8322(93)
10165.21654(80)
8515.16477(65)
4910.19903(44)
4.9055(25)
3.620(17)
10.245(17)
2.0645(12)
0.7393(56)
0.0506(34)
0.337(20)
0
0.423(20)
0.0254(17)
0.0273(41)
0.2215(83)
10 (MHz).............................
11 (MHz)...........................
11 (MHz).............................
20 (MHz).............................
½A ðB þ CÞ=210 (kHz) ......
½A ðB þ CÞ=211 (kHz) ....
½A ðB þ CÞ=220 (kHz) ......
½ðB þ CÞ=210 (kHz) .............
½ðB þ CÞ=211 (kHz) ...........
½ðB þ CÞ=220 (kHz) .............
½ðB CÞ=410 (kHz) .............
½ðB CÞ=411 (kHz) ...........
½DJ 10 (kHz)..........................
½DJK 10 (kHz)........................
½DK 10 (kHz) .........................
½J 10 (kHz) ..........................
½K 10 (kHz)..........................
763.198(62)
0.0800(83)
1.050(43)
0.767(13)
55.07(64)
1.62(25)
0.87(21)
21.16(56)
1.43(18)
0.31(13)
3.40(27)
0.475(73)
0.03906(34)
0.0998(17)
0.0737(17)
0.01960(18)
0.03427(98)
a
Standard uncertainties are type A, k ¼ 1 (1 ) and are given in units of last digit in parentheses.
No. 1, 2002
ACETONE
five tunneling parameters associated with the quartic distortion constants. For the nontunneling parameters, the symbols of the asymmetric rotor Hamiltonian in Watson’s A
reduction (Watson 1977) have been used. The internal
energy tunneling parameters are designated by qq0 . For the
other tunneling parameters, the notation ½X qq0 has been
used, where X is a linear combination of rotational constants or distortion constants and the subscript designates
the tunneling component.
Two sets of frequencies included in Table 1 (listed with an
estimated uncertainty of 0.0 MHz) were not used in the fit.
The first set consists of frequencies assigned in previous
investigations that do not fit as well as other frequencies
from the same source. Typographical errors, misassignments, or unrecognized blends with unidentified lines are
the most likely causes for the unsatisfactory fit. The second
set consists of frequencies measured in this investigation.
Some of them have nearly degenerate but unresolved transitions whose assignments are quite certain. But most of them
are frequencies of (high Ka , low Kc )–transitions with J > 21
that do not fit very well. A select few tentative assignments
of (high Ka , low Kc )–transitions with J < 20 are also
marked with zero uncertainty. The causes for the bad fit of
some of these lines can only be guessed at. The assignments
of at least the ð0; 0Þ and ð0; 1Þ substate lines seem unquestionable because of their proximity to the predicted positions and the qualitative agreement between predicted and
observed relative intensities. At room temperature, they are
the strongest nondegenerate transitions (both predicted and
observed) at frequencies above 400 GHz. Most likely, some
of the (high Ka , low Kc ) levels involved in these transitions
are perturbed by interactions with levels from the lowest
torsional excited states that are 78 and 125 cm1 above the
VTG state (Groner 2000). Already at J ¼ 20 the stack of
149
energy levels belonging to the ground state begins to overlap
with the stack of the lowest excited state.
4. DISCUSSION
With the spectroscopic constants derived from the fit to
the global data set, the frequencies were predicted for an
additional 11,000 lines belonging to the VTG state through
620 GHz and through rotational quantum number J of 60.
These frequencies are listed in Table 4 (found in its entirety
in the electronic edition of the Astrophysical Journal Supplement) along with the torsional substate t ¼ ð1 ; 2 Þ, the
rotational quantum numbers, the estimated 1 uncertainties (type A, coverage factor k ¼ 1) of the predicted frequencies, the spin weights, the intensities expressed in S values
(Townes & Schawlow 1975), and the energies of the upper
states in units of cm1. Only transitions with S > 0:50,
Eupper < 500 cm1 , and k ¼ 1 uncertainty < 1:00 MHz are
included in Table 4. The upper state energies are measured
relative to the absolutely lowest level (J; Ka ; Kc Þ ¼ ð0; 0; 0Þ
of the substate ð0; 0Þ. They include a small contribution
from the internal rotation [2284 MHz for t ¼ ð0; 1Þ,
4571 MHz for t ¼ ð1; 1Þ, and 4574 MHz for t ¼ ð1; 2Þ].
The assigned and fitted lines listed in Table 1 are included in
Table 4 (but with their predicted frequencies) so that the
additional information concerning these lines (spin weights,
intensities, upper state energies) is readily available for
astronomical use.
The intensities of the lines are given in terms of the S-values so that these numbers must still be multiplied by the
square of the b-direction dipole moment (lb ¼ 2:93 0:03
D; Peter & Dreizler 1965) to obtain the proper intensities. In
addition, they must be multiplied by the appropriate spin
weight.
TABLE 4
Predicted Transition Frequencies of Acetone in the Vibrational-Torsional Ground State
ta
J0
Ka0
Kc0
J 00
Ka00
Kc00
Frequency
(MHz)
Uncertainty
(MHz)b
Spin
Weight
S
Eupper
(cm1)
11........
01........
12........
00........
11........
00........
01........
12........
11........
12........
01........
00........
11........
00........
11........
01........
12........
01........
00........
12........
1
1
1
1
2
2
2
2
3
3
3
3
3
3
2
3
2
2
2
3
1
1
1
1
2
2
2
2
2
2
2
2
3
3
1
3
1
1
1
3
0
0
0
0
0
0
0
0
1
1
1
1
0
0
1
0
1
1
1
0
1
1
1
1
2
2
2
2
3
3
3
3
3
3
2
3
2
2
2
3
0
0
0
0
1
1
1
1
1
1
1
1
2
2
0
2
0
0
0
2
1
1
1
1
1
1
1
1
2
2
2
2
1
1
2
1
2
2
2
1
5253.063
5269.071
5270.904
5276.057
7343.774
7398.953
7399.964
7453.183
10731.360
10749.943
10751.671
10762.702
11073.951
11199.320
11252.506
11257.715
11265.269
11272.464
11286.131
11468.246
0.007
0.002
0.002
0.001
0.009
0.003
0.003
0.007
0.009
0.002
0.002
0.002
0.006
0.005
0.002
0.006
0.002
0.002
0.002
0.015
6
16
4
10
2
6
16
4
6
4
16
10
6
10
2
16
4
16
6
4
1.495
1.496
1.489
1.500
2.012
2.013
1.989
1.926
2.628
2.624
2.633
2.640
2.050
2.041
1.315
1.893
1.313
1.317
1.320
1.693
0.776
0.700
0.776
0.624
2.036
1.887
1.963
2.040
3.588
3.589
3.514
3.440
3.957
3.814
1.791
3.890
1.792
1.716
1.641
3.972
Note.—Table 4 is published in its entirety in the electronic edition of the Astrophysical Journal Supplement. A portion is shown here for guidance regarding its form and content.
a Torsional substates: 00 ¼ AA, 01 ¼ EE, 11 ¼ AE, 12 ¼ EA.
b Calculated (1 ) uncertainty, type A with coverage factor k ¼ 1.
150
GRONER ET AL.
Vol. 142
TABLE 5
Partition Function for Acetone in the Vibrational-Torsional Ground State
State
(0, 0)
AA
(0, 1)
EE
(1, 1)
AE
(1, 2)
EA
Total
Asymmetric Top Approximation
10 K ..........
20 K ..........
50 K ..........
100 Ka .......
200 Ka .......
300 Ka .......
2082.56
5867.96
23148.33
65450.95
183787
326294
4123.59
11682.27
46231.83
130843.32
367540
652572
1020.60
2907.19
11541.75
32696.17
91876
163138
1020.59
2907.18
11541.73
32696.15
91877
163139
8247.34
23364.61
92463.64
261686.59
735080
1305143
8274.89
23404.92
92516.07
261674.95
740129
1359703
a
Sum of states not converged.
Only one transition is listed in cases of c-type or a-type
degeneracy if the predicted frequencies are degenerate
within 0.001 MHz. These cases are identified in Table 4 by
the combined spin weights of the degenerate transitions 16,
32, 8, and 8 for the torsional substates ð0; 0Þ, ð0; 1Þ, ð1; 1Þ,
and ð1; 2Þ, respectively.
For some nearly degenerate energy levels in the substates
ð0; 1Þ (EE) and ð1; 2Þ (EA), mixing and level crossings occur
so that the pseudo–quantum numbers Ka and Kc , defined
according to the standard ordering in asymmetric tops, do
not always represent the actual wave functions very well. As
a consequence, ‘‘ forbidden ’’ c-type transitions occur as follows. True ‘‘ forbidden ’’ c-type transitions occur when the
wave functions mix; they are always accompanied by b-type
transitions between the same pairs of levels. False c-type
transitions arise from level crossings; they lack b-type partners because they are the mislabeled b-type transitions.
The spin weights must be used as factors for line intensities and to determine the total torsional-rotational-nuclear
spin partition function as a function of temperature; numerical values for the overall partition function as well as for
the individual torsional substates (obtained by restricting
the summation to levels belonging to a specific substate) are
listed in Table 5 at selected temperatures. It should be
emphasized that the partition function for the torsional substates ð0; 1Þ, ð1; 1Þ, and ð1; 2Þ includes a factor that depends
on the energy difference of the (J; Ka ; Kc ¼ 0; 0; 0) levels
between the substate and the ð0; 0Þ substate given above. A
reasonable approximation to the numerical values for the
overall partition function can be obtained from the simple
asymmetric top expression (Townes & Schawlow 1975) multiplied by the total spin weight ð2I þ 1Þ6 ¼ 64 ðI ¼ 12 Þ and
divided by the overall symmetry number for the C2v point
group of 2. The result is q ¼ 261:67495T 3=2 ; results at this
level of approximation are also shown in Table 5. This
approximation should be used for temperatures above 100
K because the direct summation of the partition function is
not converged if J is limited to 60. The fractional population
f (the ratio of the state column density to the total column
density) for any rotational-torsional state of energy E is
then given by the expression
f ¼ gs ð2J þ 1Þ
expðE=kTÞ
;
q
ð3Þ
where the symbol gs stands for the appropriate spin weight
as defined in x 2, J is the rotational quantum number, k is
the Boltzmann constant, and T is the temperature in kelvins.
For example, for the upper level of the first transition in
Table 4 [torsional substate (1, 1), J; Ka ; Kc ¼ 1; 1; 0], the
spin weight is 6, and the fractional population at 100 K is
f ¼ 6:80 105 . In performing analyses for line intensities,
astronomers should consider the possibility of blending of
one or more of the quartet of lines for each pair of rotational
quantum numbers. Even if the quartet of torsionally split
lines is totally blended, the total spin weight must still be
considered since it has a rotational dependence as well as a
torsional dependence. For example, if the first quartet of
lines in Table 4 were totally blended, its total spin weight
would be 6 þ 16 þ 4 þ 10 ¼ 36, whereas the total spin
weight of a blended quartet of lines for Ka Kc : ee $ oo
transitions such as 2, 2, 0 $ 1, 0, 1 would be
2 þ 6 þ 16 þ 4 ¼ 28. For a similar reason, the spin weights
of blended lines of c-type or a-type degenerate transitions
are 16, 32, 8, and 8 for the torsional substates ð0; 0Þ, ð0; 1Þ,
ð1; 1Þ, and ð1; 2Þ, respectively.
We would like to thank NASA for their support of laboratory astrophysics at Ohio State University. S. A. thanks
the Alexander-von-Humboldt Stiftung for a Feodor-Lynen
research stipend. A portion of this research was carried out
at the Jet Propulsion Laboratory, California Institute of
Technology, under contract with NASA.
REFERENCES
Gordy, W., & Cook, R. L. 1984, Microwave Molecular Spectroscopy (New
Balle, T. J., & Flygare, W. H. 1981, Rev. Sci. Instrum., 52, 33
York: Wiley)
Blake, G. A., Sutton, E. C., Masson, C. R., & Phillips, T. G. 1987, ApJ,
Groner, P. 1997, J. Chem. Phys., 107, 4483
315, 621
———. 2000, J. Mol. Struct., 550–551, 473
Brown, P. D. 1990, MNRAS, 243, 65
Groner, P., Albert, S., Herbst, E., & De Lucia, F. C. 1998, ApJ, 500, 1059
Brown, P. D., Charnley, S. B., & Millar, T. J. 1988, MNRAS, 231, 409
Herbst, E., Giles, K., & Smith, D. 1990, ApJ, 358, 468
Caselli, P., Hasegawa, T. I., & Herbst, E. 1993, ApJ, 408, 548
Hollis, J. M., & Churchwell, E. 2001, ApJ, 551, 803
Charnley, S. B., Kress, M. E., Tielens, A. G. G. M., & Millar, T. J. 1995,
Hollis, J. M., Vogel, S. N., Snyder, L. E., Jewell, P. R., & Lovas, F. J. 2001,
ApJ, 448, 232
ApJ, 554, L81
Charnley, S. B., Tielens, A. G. G. M., & Millar, T. J. 1992, ApJ, 399, L71
Ikeda, M., Ohishi, M., Nummelin, A., Dickens, J. E., Bergman, P.,
Charnley, S. B., Tielens, A. G. G. M., & Rodgers, S. D. 1997, ApJ, 482,
Hjalmarson, A., & Irvine, W. M. 2001, ApJ, 560, 792
L203
Liu, S.-Y., Mehringer, D. M., & Snyder, L. E. 2001, ApJ, 552, 654
Combes, F., Gerin, M., Wootten, A. S., Wlodarczak, G., Clausset, F., &
Lovas, F. J., Lutz, H., & Dreizler, H. 1979, J. Phys. Chem. Ref. Data, 8,
Encrenaz, P. J. 1987, A&A, 180, L13
1051
Friedl, R. R., Birk, M., Oh, J. J., & Cohen, E. A. 1995, J. Mol. Spectrosc.,
Lovas, F. J., & Suenram, R. D. 1987, J. Chem Phys., 87, 2010
170, 383
No. 1, 2002
ACETONE
Mehringer, D. M., Snyder, L. E., Miao, Y., & Lovas, F. J. 1997, ApJ, 480,
L71
Miao, Y., Mehringer, D. M., Kuan, Y.-J., & Snyder, L. E. 1995, ApJ, 445,
L59
Millar, T. J., Herbst, E., & Charnley, S. B. 1991, ApJ, 369, 147
Myers, R. J., & Wilson, E. B. 1960, J. Chem. Phys., 33, 186
Nelson, R., & Pierce, L. 1965, J. Mol. Spectrosc., 18, 344
Oldag, F., & Sutter, D. H. 1992, Z. Naturforsch., 47a, 527
Pei, C. C., Liu, S.-Y., & Snyder, L. E. 2000, ApJ, 530, 800
Peter, R., & Dreizler, H. 1965, Z. Naturforsch., 20a, 301
Petkie, D. T., Goyette, T. M., Bettens, R. A. P., Belov, S. P., Albert, S.,
Helminger, P., & De Lucia, F. C. 1997, Rev. Sci. Instrum., 68, 1675
Snyder, L. E., Lovas, F. J., Mehringer, D. M., Miao, Y., Kuan, Y.-J., Hollis, J. M., & Jewell, P. R. 2002, ApJ, in press
Snyder, L. E., Miao, Y., Mehringer, D. M., Kuan, Y.-J., Hollis, J. M.,
Jewell, P. R., & Lovas, F. J. 1997, in CO: 25 Years of Millimeter-Wave
Spectroscopy, ed. W. B. Latter et al. (Dordrecht: Kluwer), 460
151
Suenram, R. D., Lovas, F. J., Fraser, G. T., Gillies, J. Z., Gillies, C. W., &
Onda, M. 1989, J. Mol. Spectrosc., 137, 127
Sutton, E. C., Peng, R., Danchi, W. C., Jaminet, P. A., Sandell, G., &
Russell, A. P. G. 1995, ApJS, 97, 455
Swalen, J. D., & Costain, C. C. 1959, J. Chem. Phys., 31, 1562
Townes, C. H., & Schawlow, A. L. 1975, Microwave Spectroscopy (New
York: Dover)
Vacherand, J. M., van Eijck, B. P., Burie, J., & Demaison, J. 1986, J. Mol.
Spectrosc., 118, 355
Watson, J. K. G. 1977, in Vibrational Spectra and Structure, Vol. 6, ed.
J. R. Durig (Amsterdam: Elsevier), 1
Weatherly, T. L., & Williams, D. 1952, J. Chem. Phys., 20, 755
White, W. F. 1975, NASA Tech. Note D-7904, 121