Quantitative rotational spectroscopy for atmospheric applications

1
Quantitative rotational spectroscopy for
atmospheric applications

• G. WLODARCZAK
• Laboratoire de Physique des Lasers, Atomes et
  Molécules (UMR CNRS 8523) and CERLA
• Université de Lille 1, France

2
Goals

• Improvement of existing databases reliable data
  for the analysis of satellite data
• Test values for the theoretical calculations
• Detailed lineshape analysis

3
ODIN satellite
4
ODIN satellite

• launched in february 2001
• Radiometer118.25-119.25 GHz 486.1-503.9 GHz
  541.0-580.4 GHz resolution 0.1 to 1 MHz
• astrophysical observations (CI, H2O, O2, H2S,
  NH3, H2CO, CS, 13CO, H2CS, SO, SO2..)
• aeronomy stratospheric ozone depletion, coupling
  between the upper and lower atmosphere
• Optical spectrometer 280-800 nm and 1270 nm

5
(No Transcript)
6
(No Transcript)
7
(No Transcript)
8
EOS MLS experiment (NASA)

• LaunchAURA platform, jan. 2004 (5 years of
  operation)
• Radiometers 190 GHz (H2O, HNO3), 240 GHz (O3),
  640 GHz (HCl, ClO, N2O), 2.5 THz (OH)
• Coupled with infrared high resolution dynamics
  limb sounder (HIRDLS), infrared tropospheric
  emission spectrometer (TES), UV ozone monitoring
  instrument (OMI)
• Improving our understanding of atmospheric global
  change

9
(No Transcript)
10
JEM/SMILES Experiment

• global mapping of atmospheric trace gases ClO,
  HCl, HO2, HNO3, BrO, O3 isotopic species
• 624.32-626.32 GHz and 649.12-650.32 GHz
• SIS receivers
• launch 2005

11
(No Transcript)
12
MASTER experiment (ESA)MM-wave acquisition for
stratosphere-troposphere exchange research

• Launch ???, ACECHEM baseline mission
• Bands 294.00-305.50 GHz (band B)
• 316.50-325.50 GHz (band C)
• 342.25-348.75GHz (band D)
• 497.00-506.00 GHz (band E)
• 624.00-626.50 GHz (band F)
• Flying demonstrator MARSCHALS (bands B,C,D), oct.
  2003
• Target species O3, H2O, HNO3, CO, N2O, HCl, ClO
• Brightness temperature

13
(No Transcript)
14
Rotational lineshapes
n or Eu, El m G or g, n D or d
n or Eu, El m G or g, n D or d
15
Frequency measurements

• For the target molecules, spectroscopy is well
  known
• But implementation in the databases not always
  done!
• Combined rovibrational analysis of microwave and
  infrared data H2CO, HNO3, O3, CH3Cl, ClONO2,
  HCOOH,

16
Intensity measurements

• Not easy to perform
• For rotational transitions, intensity is a
  function of the permanent dipole moment
• Pb some experimental values rather old
• Dipole moment for isotopic species (HDO, 18O
  species )
• Ab initio calculations

17
Broadening coefficients

• Voigt profile commonly used (convolution of the
  Doppler profile (Gaussian) and the collisional
  profile (Lorentzian)
• collisional linewidth DgL ga-a(T) Pa
  ga-b(T) Pb where a denotes the absorbing gas and
  b the foreign perturbing gas, ga-a(T) and ga-b(T)
  are the pressure broadening parameters for self-
  and foreign-broadenings at the temperature T, and
  Pa and Pb are the partial pressures of absorbing
  and foreign gases respectively.
• Broadening by O2 or N2
• Temperature dependence n exponent
• g(T)g(296) (296/T)n

18
Line broadening measurements

• Classical MMW absorption spectrometers
• Lille, Bologna, JPL, OSU, NAIR, Ibaraki,...
• FIR laser Cambridge,...
• Transient phenomena Kiel, Lille
• Acoustic detection Nizhnii Novgorod
• Resonator Nizhnii Novgorod
• FTIR DLR,

19
Line broadening measurements

• Pressure range usually lt 1 Torr except for FTIR
  and resonator experiments
• Temperature range 200-350 K
• Modulation techniques used (amplitude or
  frequency) see Poster M29 ( frequency modulation)

20
Line broadening measurements

• differences observed for the same transition
  studied by different groups larger than for
  frequency measurements
• pressure measurement
• temperature determination, homogeneity
• baseline substraction, power fluctuations of the
  source
• adsorption of the sample, stability
• need for intercomparison to provide the most
  reliable data

21
Line broadening measurements

• OCS J9-8 transition, self-broadening coefficient
• O12CS 6.022 (18) MHz/Torr ampl. modulation
• O13CS 6.076 (69) MHz/Torr ampl. modulation
• O13CS 6.133 (30) MHz/Torr freq. modulation
• CO J1-0 transition, self-broadening coefficient
• 3.354 (8) MHz/Torr ampl. modulation
• 3.414 (5) MHz/Torr freq. modulation
• from C. Puzzarini et al. J. Mol. Spectrosc. 216,
  428 (2002)

22
Line broadening measurements

• Ex1. CO J3-2 line at 345 GHz, air-broadening
• Lille (BWO) 2.728(17) MHz/Torr
• DLR (FTIR) 2.745(55) MHz/Torr
• Bologna (multiplier) 2.687(18) MHz/Torr (13CO)
• Ex2. Ozone
• See poster J17 for intercomparisons
  (Lille/Bologna) for lines located in the bands C
  and D of MASTER
• See also poster L21

23
HCl, J1-0

• Drouin (JQSRT, 2003, in press) multiplier
• gair3.42(4) MHz/Torr
• Davies et al (J Mol Spectrosc 220, 107-112,
  2003)TuFIR
• gair3.30(6) MHz/Torr
• Pine et al (J Mol Spectrosc 122, 41, 1987) FTIR
• gair3.52(4) MHz/Torr
• Yamada et al (Dijon 2003, poster L22)

24
Pressure broadening of HCl projected at 296 K in
N2 (black) and O2 (grey).
25
N2O Linewidths
26
O2 Linewidths
27
H2O313-220 transition, 183 GHz

• Fig. 5. Measured values of the water line air
  pressure broadening parameter. The authors
  responsible are as follows. (1) Rusk 5 (2)
  Dryagin et al. 3 (3) Frenkel and Woods 6
  (4) Hemmi and Straiton 17 (5)Ryadov and
  Furashov 2 (6) Bauer et al. 4 (7) Bauer et
  al. 14 (8) Goyette and DeLucia 18
  (9)Pumphrey and Buehler 1MLS (10) Pumphrey and
  Buehler 1MAS (11)Krupnov et al. 7 (obtained
  using the calculated apparatus function) (12)
  This work (M. Yu. Tretyakov et al, J Mol.
  Spectrosc, 2003)

28
HNO3
Fig. 7. Comparison of pressure broadening
measurements for species and transitions in Table
1 and Table 3. The horizontal axis are OSU
measurements, and on the vertical axis the solid
circles, the HNO3O2/N2 measurements of JPL
reported in this work the open circles, the
H2OO2/N2 measurements of Refs.9 and13 and
the solid square, the COHe measurement,
Refs.11 and12. The error bars are 5
29
HNO3 Linewidths
30

LINE SHIFTS

• Can have a high impact on the retrieval
• Smaller than infrared values very few reliable
  data in the MMW range
• Ex1. CO J5-4 at 576.3 GHz
• self6 (3) kHz/Torr (Markov 2002)
• 45.8 (214) kHz/Torr (Yamada 2003)
• Ex2. CO J1-0 at 115 GHz
• dself-27 (24) kHz/Torr (Fabian 1997)
• Generally too small to be determined only
  measurable if d gt1/20 g
• Measured for NH3, H2O (307(4) kHz/Torr for the
  110-101 transition at 556 GHz, -70(20) kHz/Torr
  for the 313-220 transition at 183 GHz), HCl

31
HCl lineshift J1-0 line
Drouin, JQSRT, 2003 dair 0.146 MHz/Torr
32
Theoretical calculations

• Complex Robert-Bonamy formalism generally used
• Semi-classical model using curved trajectories
  based on the isotropic part of the intermolecular
  potential
• Total potential VT Ve Va-a, where Ve is the
  electrostatic potential and a short range
  atom-atom component (Lennard-Jones 6-12)
• Various temperature dependence models tested
  (when the temperature range is large)
• See Posters J10 (C2H2), M27 (H2O) for CRB
  calculations, M24 (semi-empirical approach)

33
(No Transcript)
34
(No Transcript)
35
Ozone
For all these transitions the calculated
lineshift is less than 10 kHz/Torr
36
K-Dependence of the pressure broadeningCH3CN,
J12-11 (220 GHz)
37
approximate theoretical model When
dipole-quadrupole interaction is only considered,
the Birnbaum model (1) leads to the following
simple expression for the pressure broadening-para
meter (1) G. Birnbaum, Adv. Chem.
Phys. 12, 487 (1967).
38
Conclusion

• Need for complementary measurements (see also
  conclusions of the San Diego meeting, oct. 2001
  www.atmoschem.jpl.nasa.gov)
• Intercomparison of the data obtained by various
  (or identical) techniques
• Dont claim unrealistic error bars
• Need for an unified up-to-date database (with IR
  data)
• Non resonant absorption continuum (H2O)
  measurements in all spectral ranges including MMW
  (see posters H10 and O20)

39

40
Line shape analysis

• Departure from Voigt profile observed in
  rotational transition lineshapes for O3, N2O, CO,
  NO, CH3Cl, CH3CN, HCN,..
• Deviations observed for some time in
  rovibrational spectra
• Systematic deviation of the broadening
  coefficients (up to 2 )
• Influence on the retrieval not yet considered
  (may be not negligible in upper stratosphere)

41
Line shape analysis

• velocity/speed changing collisions (Dicke
  effect) lead to a Galatry profile (soft
  collisions, magtmb) or Rautian profile (strong
  collision, maltmb)
• Galatry model two parameters, G0 ,relaxation
  rate, and b, diffusion rate
• speed dependence of the relaxation rate G speed
  dependent Voigt profile
• quadratic dependence G(va) G0 G2 ((va /
  va0)2 -3/2)
• two parameters G0 and G2
• from the analysis of the residuals, both models
  are equivalent but
• b is not linear with pressure when pressure
  increases (theory predicts that G0 , G2 and b
  are linear with pressure)

42
(No Transcript)
43
Line shape analysis

• Conclusion speed dependent profile more
  appropriate
• In case of small profile narrowings, we can
  write
• b/3G2 1(3G2/aG0)(G0/kvao)2
  
• Doppler regime G0 lt kvao b/3G2 1, both
  profiles equivalent
• Collisional regime G0 gt kvao b is non linear
  versus pressure and becomes very large and for a
  limit pressure no line fitting is possible
  (observed for HCN/CH3Br in infrared)
• See J-F dEu, B. Lemoine, F. Rohart, J. Mol.
  Spectrosc. 212, 96-110 (2002) and poster J14
• Confirmation by theoretical calculations of the
  speed dependence of the relaxation rates
• Full agreement with previous studies on the J1-0
  line of N2O

44
(No Transcript)
45
Ackowledgments

• CNRS Programme National de Chimie Atmosphérique,
  Programme National de Planétologie
• European Spatial Agency
• B. Bakri, J-F dEu, D. Priem
• J-M Colmont, F. Rohart, J. Demaison
• A. Perrin (Orsay)
• G. Cazzoli, L. Dore, C. Puzzarini (Bologna)
• M. Birk, G. Wagner (DLR)
• C. Verdes, S. Buehler (Bremen)
• B.J. Drouin (JPL)