m REGION µ (DEUTERATED NITRIC ACID) IN THE 11 BAND OF...
1
FIRST ANALYSIS OF THE ν 5 BAND OF DNO 3 (DEUTERATED NITRIC ACID) IN THE 11 μm REGION J. Koubek 1,2 , A. Perrin 2 , H. Beckers 3 , H. Willner 3 1 Institute of Chemical Technology, Department of Analytical Chemistry, Technick´a 5, 166 28, Prague, CZECH REPUBLIC 2 Laboratoire Inter-universitaire des Syst` emes Atmosph´ eriques, CNRS, Universit´ e Paris 12, 61 Av. du G´ en´ eral de Gaulle, 94010 Cr´ eteil Cedex, FRANCE 3 Anorg. Chemistry, University of Wuppertal, D-42119 Wuppertal, GERMANY 1 Introduction Nitric acid (HNO 3 ) plays an important role as a ”reservoir” molecule for both the NO x (ni- trogen oxides) and HO x (hydrogen oxides) species in the stratosphere [1]. These radicals are potentially active contributors to the ozone destruction in the stratosphere through catalytic reactions. For this reason, various isotopic species of nitric acid have been the subject of nu- merous spectroscopic studies [2, 3]. High resolution studies on nitric acid isotopomers in the infrared, submillimeter and centimeter region are referred in the article by Drouin et al. [4] DNO 3 isotopomer is referred particularly by Chou et al. [5] In the infrared, just the ν 9 (O–D torsion) [6] , ν 8 (out of plane NO 2 bend) [7] , ν 7 (O–NO 2 bend) [8] , ν 6 (O–NO 2 stretch) [8] and ν 2 (NO 2 a-stretch) [9] fundamental bands were subjects of high resolution FTIR studies by Tan, Looi, Lua, Maki, Johns and No¨ el. The present analysis describes the first analysis of the ν 5 (NO 2 planar bend) band of DNO 3 in the 11 μm spectral region. 2 Experimental DNO 3 was synthesized from D 2 SO 4 and KNO 3 by using vacuum techniques in Wuppertal. The sample contained traces of HNO 3 ,H 2 O, HDO, D 2 O, NO 2 - these impurities served in calibra- tion of the spectra using HITRAN database. The infrared spectrum of deuterated nitric acid was recorded on the Bruker IFS 120 HR Fourier transform spectrometer of Wuppertal in the 700–1500 cm −1 region at three pressures: 1 Torr, 0.3 Torr and less then 0.1 Torr (cf Figure 1). The instrumental resolution is 0.0022 cm −1 and the maximal spectral resolution is 0.0027 cm −1 (from minimal observed spacing between two lines that is greater than the FWHM of either line). Figure 1: ν 5 band of DNO 3 , spectrum measured in Wuppertal < 0.1 Torr 0.3 Torr 1.0 Torr 3 Theory DNO 3 is planar asymmetric molecule (C S symmetry). The form of the Hamiltonian matrix used for DNO 3 is described in Table 1. The rotational operators for both the v =5 1 and v =7 1 9 1 vibrational diagonal blocks include Watson-type operators written in an I r represen- tation with an A-type reduction. Due to symmetry considerations, A-type and B-type Coriolis resonances are to be considered in the 5 1 ⇔ 7 1 9 1 off-diagonal vibrational operators. The resonance between the 5 1 and 7 1 9 1 levels is indeed very strong. Table 1: Hamiltonian matrix 5 1 7 1 9 1 5 1 H 5,5 complex conjugate 7 1 9 1 H 79,5 = C A + C B H 79,79 v-diagonal operator (rotational operator): H v,v = E v +A v J 2 z +B v J 2 x +C v J 2 y +Δ v K J 4 z −Δ v JK J 2 J 2 z −Δ v J (J 2 ) 2 −2δ v J J 2 J 2 xy −δ v K {J 2 z ,J 2 xy }+... v-off-diagonal operators: H 79,5 = C A + C B : C A = C A1 J z + C A2 {iJ y ,J x } C B = C B1 J x + C B2 {iJ y ,J z } + C B3 J x J 2 + C B4 J x J 2 z + C B5 (J 3 + + J 3 − ) {A, B } = AB + BA; J 2 xy = J 2 x − J 2 y ; J ± = J x ± iJ y Due to the close proximity of the 5 1 and 7 1 9 1 energy levels of 14–DNO 3 , strong perturba- tions are observed in the spectrum. Due to the relative symmetry of the 5 1 ⇔ 7 1 9 1 interacting states (A ′ ⇔ A ′′ ), A-type and B-type Coriolis resonances are to be considered for the energy levels calculation. Through these interactions, the 7 1 9 1 ”dark ” state is populated on behalf of the 5 1 state. Thanks to this line mixing , many ν 7 + ν 9 transitions become observable. This differs completely from the scheme of resonance observed for 14–HNO 3 and 15–HNO 3 , since Fermi and C-type Coriolis resonances are coupling the 5 1 ⇔ 9 2 (A ′ ⇔ A ′ ) resonating states. Figure 2: Energy level ladder for HNO 3 and DNO 3 4 Analysis of the D 14 N 16 O 3 infrared spectra 4.1 Assignment 3000 rovibrational transitions, approximately, of ν 5 band and cca 300 rovibrational transitions of ν 7 + ν 9 were assigned using ground state combination differences with rotational ground state parameters achieved by Drouin et al. [4]. Least squares fit of molecular parameters enabled to synthesize satisfactorily the spectrum. Following figures demonstrate sufficient conformity between observed and calculated spectrum. Figure 3: Hotbands’ region ↑ ν 5 + ν 6 − ν 6 ↑ ν 5 + ν 7 − ν 7 ν 5 + ν 9 − ν 9 ↑ calc 7 1 9 1 calc 5 1 ,7 1 9 1 obs Figure 3 shows the region where three hot bands occur: ν 5 + ν 6 − ν 6 , ν 5 + ν 7 − ν 7 and ν 5 + ν 9 − ν 9 located at cca 881, 883 and 884 cm −1 , respectively. These bands were not considered in the analysis. Figure 4: ν 5 Q branch region calc 7 1 9 1 calc 5 1 ,7 1 9 1 obs Figure 4 shows the ν 5 region of the very dense Q branch, severely mixed with ν 7 + ν 9 tran- sitions. Several J (K c ′′ ) packets of ν 5 transitions (green colour labels) and ν 7 + ν 9 tran- sitions (red colour labels) are sorted with following selection for upper and lower states: J K a ′ = J − K c ′ K c ′ ←− J K a ′′ K c ′′ . Figure 5: Hotbands’ region - detail ↑ ν 5 + ν 6 − ν 6 calc 7 1 9 1 calc 5 1 ,7 1 9 1 obs Figures 5 and 6 show detailed part of the spectrum. In Figure 5, P lines and Q lines belonging to the ν 5 band together with lines from the resonating ν 7 + ν 9 dark bands are evidenced. Figure 6 shows the central part of the ν 5 Q branch. Figure 6: ν 5 Q branch central region calc 7 1 9 1 calc 5 1 ,7 1 9 1 obs 4.2 Fit Line positions and intensities were determined from least squares fit on a rather large set of experimental data as shows Table 2. All values of parameters are in cm −1 , the uncertainty in the last digits (one standard deviation) is given in the parenthesis. Table 2: Molecular rovibrational parameters Parameter 5 1 7 1 9 1 E v 887.657124(81) 882.21072(49) A 0.43244023(96) 0.4317699(37) B 0.3765583(20) 0.3757854(37) C 0.20048282(20) 0.2001817(30) Δ K 1.629(46)×10 −7 2.85(12)×10 −7 Δ JK 0.262(46)×10 −7 –0.43(11)×10 −7 Δ J 2.1928(53)×10 −7 2.258(17)×10 −7 δ K 3.478(17)×10 −7 1.881(59)×10 −7 δ J 0.8929(28)×10 −7 1.209(19)×10 −7 5 1 ⇔ 7 1 9 1 Coriolis interaction parameters Operator, Constant Value J z , C A1 3.4523(51)×10 −2 {iJ y ,J x }, C A2 3.997(12)×10 −4 J x , C B1 7.1521(68)×10 −2 {iJ y ,J z }, C B2 1.3064(22)×10 −3 J x J 2 , C B3 –6.15(23)×10 −7 J x J 2 z , C B4 6.128(62)×10 −6 J 3 + + J 3 − , C B5 1.322(27)×10 −6 The rms deviation of the fit with all weighted lines was about 0.51×10 −3 cm −1 . Still, the assignment of some ν 7 + ν 9 transitions in the ν 5 Q branch is problematic. Table 3: Statistics of the fit fit of 1096 experimental energy levels (979 levels for ν 5 and 117 levels for ν 7 +ν 9 ) 0. 10 −3 cm −1 < 92.7 % of levels < 1. 10 −3 cm −1 1. 10 −3 cm −1 < 4.6 % of levels < 2. 10 −3 cm −1 2. 10 −3 cm −1 < 2.3 % of levels < 4. 10 −3 cm −1 4. 10 −3 cm −1 < 0.4 % of levels < 10. 10 −3 cm −1 Acknowledgements The work of J.K. was supported through the Ministry of Education, Youth and Sports of the Czech Republic (research program LC06071) and a scholarship of the French government (Bourse du Gouvernement Fran¸ cais - Bourse de Doctorat en co-tutelle). References [1] World Meteorological Organization, in “Scientific Assessment of Ozone Depletion: 2006” (WMO, Ed.), Geneva, 2007. [2] A. Goldman, C. P. Rinsland, A. Perrin, and J.-M. Flaud, J. Quant. Spectrosc. Radiat. Transfer 60, 851–861 (1998). [3] A. Perrin, Spectrochim. Acta 54, 375–394 (1998). [4]B. J. Drouin, C. E. Miller, J. L. Fry, D. T. Petkie, P. Helminger, and I. R. Medvedev, J. Mol. Spectrosc. 236, 29–34 (2006). [5] S. G. Chou, D. T. Petkie, R. A. H. Butler, and C. E. Miller, J. Mol. Spectrosc. 211, 284–285 (2002). [6] T. L. Tan, E. C. Looi, K. T. Lua, A. G. Maki, J. W. C. Johns, and M. No¨ el, J. Mol. Spectrosc. 150, 486–492 (1993). [7] T. L. Tan, E. C. Looi, K. T. Lua, A. G. Maki, J. W. C. Johns, and M. No¨ el, J. Mol. Spectrosc. 149, 425–434 (1991). [8] A. G. Maki, T. L. Tan, E. C. Looi, K. T. Lua, J. W. C. Johns, and M. Noel, J. Mol. Spectrosc. 157, 248–253 (1993). [9] T. L. Tan, E. C. Looi, K. T. Lua, A. G. Maki, J. W. C. Johns, and M. Noel, J. Mol. Spectrosc. 166, 97–106 (1994).
Transcript of m REGION µ (DEUTERATED NITRIC ACID) IN THE 11 BAND OF...
FIRST ANALYSIS OF THE ν5 BAND OF DNO3 (DEUTERATED NITRIC ACID) IN THE 11 µm REGION
J. Koubek1,2, A. Perrin2, H. Beckers3, H. Willner3
1Institute of Chemical Technology, Department of Analytical Chemistry, Technicka 5, 166 28, Prague, CZECH REPUBLIC2Laboratoire Inter-universitaire des Systemes Atmospheriques, CNRS, Universite Paris 12, 61 Av. du General de Gaulle, 94010 Creteil Cedex, FRANCE
3Anorg. Chemistry, University of Wuppertal, D-42119 Wuppertal, GERMANY
1 Introduction
Nitric acid (HNO3) plays an important role as a ”reservoir” molecule for both the NOx (ni-trogen oxides) and HOx (hydrogen oxides) species in the stratosphere [1]. These radicals arepotentially active contributors to the ozone destruction in the stratosphere through catalyticreactions. For this reason, various isotopic species of nitric acid have been the subject of nu-merous spectroscopic studies [2, 3]. High resolution studies on nitric acid isotopomers in theinfrared, submillimeter and centimeter region are referred in the article by Drouin et al. [4]DNO3 isotopomer is referred particularly by Chou et al. [5]
In the infrared, just the ν9 (O–D torsion) [6] , ν8 (out of plane NO2 bend) [7] , ν7 (O–NO2
bend) [8] , ν6 (O–NO2 stretch) [8] and ν2 (NO2 a-stretch) [9] fundamental bands were subjectsof high resolution FTIR studies by Tan, Looi, Lua, Maki, Johns and Noel. The present analysisdescribes the first analysis of the ν5 (NO2 planar bend) band of DNO3 in the 11 µm spectralregion.
2 Experimental
DNO3 was synthesized from D2SO4 and KNO3 by using vacuum techniques in Wuppertal. Thesample contained traces of HNO3, H2O, HDO, D2O, NO2 - these impurities served in calibra-tion of the spectra using HITRAN database. The infrared spectrum of deuterated nitric acidwas recorded on the Bruker IFS 120 HR Fourier transform spectrometer of Wuppertal in the700–1500 cm−1 region at three pressures: 1 Torr, 0.3 Torr and less then 0.1 Torr (cf Figure 1).The instrumental resolution is 0.0022 cm−1 and the maximal spectral resolution is 0.0027 cm−1
(from minimal observed spacing between two lines that is greater than the FWHM of eitherline).
Figure 1: ν5 band of DNO3, spectrum measured in Wuppertal
< 0.1 Torr0.3 Torr1.0 Torr
3 Theory
DNO3 is planar asymmetric molecule (CS symmetry). The form of the Hamiltonian matrixused for DNO3 is described in Table 1. The rotational operators for both the v = 51 andv = 7191 vibrational diagonal blocks include Watson-type operators written in an Ir represen-tation with an A-type reduction. Due to symmetry considerations, A-type and B-type Coriolisresonances are to be considered in the 51 ⇔ 7191 off-diagonal vibrational operators. Theresonance between the 51 and 7191 levels is indeed very strong.
Table 1: Hamiltonian matrix
51 7191
51 H5,5 complex conjugate
7191 H79,5 = CA + CB H79,79
v-diagonal operator (rotational operator):
Hv,v = Ev+AvJ2z +BvJ
2x+CvJ
2y +∆v
KJ4z−∆v
JKJ2J2z−∆v
J(J2)2−2δvJJ2J2
xy−δvK{J
2z , J
2xy}+. . .
v-off-diagonal operators: H79,5 = CA + CB:
CA = CA1Jz + CA2{iJy, Jx}
CB = CB1Jx + CB2{iJy, Jz} + CB3JxJ2 + CB4JxJ
2z + CB5(J
3+ + J3
−)
{A, B} = AB + BA; J2xy = J2
x − J2y ; J± = Jx ± iJy
Due to the close proximity of the 51 and 7191 energy levels of 14–DNO3, strong perturba-tions are observed in the spectrum. Due to the relative symmetry of the 51 ⇔ 7191 interactingstates (A′ ⇔ A′′), A-type and B-type Coriolis resonances are to be considered for the energylevels calculation. Through these interactions, the 7191 ”dark” state is populated on behalf ofthe 51 state. Thanks to this line mixing, many ν7 + ν9 transitions become observable.
This differs completely from the scheme of resonance observed for 14–HNO3 and 15–HNO3,since Fermi and C-type Coriolis resonances are coupling the 51 ⇔ 92 (A′ ⇔ A′) resonatingstates.
Figure 2: Energy level ladder for HNO3 and DNO3
4 Analysis of the D14N16O3 infrared spectra
4.1 Assignment
3000 rovibrational transitions, approximately, of ν5 band and cca 300 rovibrational transitionsof ν7+ν9 were assigned using ground state combination differences with rotational ground stateparameters achieved by Drouin et al. [4]. Least squares fit of molecular parameters enabledto synthesize satisfactorily the spectrum. Following figures demonstrate sufficient conformitybetween observed and calculated spectrum.
Figure 3: Hotbands’ region
↑ ν5 + ν6 − ν6 ↑ ν5 + ν7 − ν7
ν5 + ν9 − ν9 ↑
calc 7191
calc 51,7191
obs
Figure 3 shows the region where three hot bands occur: ν5+ν6−ν6, ν5+ν7−ν7 and ν5+ν9−ν9
located at cca 881, 883 and 884 cm−1, respectively. These bands were not considered in theanalysis.
Figure 4: ν5 Q branch region
calc 7191
calc 51,7191
obs
Figure 4 shows the ν5 region of the very dense Q branch, severely mixed with ν7 + ν9 tran-sitions. Several J(Kc
′′) packets of ν5 transitions (green colour labels) and ν7 + ν9 tran-sitions (red colour labels) are sorted with following selection for upper and lower states:J Ka
′ = J −Kc′ Kc
′ ←− J Ka′′ Kc
′′.
Figure 5: Hotbands’ region - detail
↑ν5 + ν6 − ν6
calc 7191
calc 51,7191
obs
Figures 5 and 6 show detailed part of the spectrum. In Figure 5, P lines and Q lines belongingto the ν5 band together with lines from the resonating ν7+ν9 dark bands are evidenced. Figure6 shows the central part of the ν5 Q branch.
Figure 6: ν5 Q branch central region
calc 7191
calc 51,7191
obs
4.2 Fit
Line positions and intensities were determined from least squares fit on a rather large set ofexperimental data as shows Table 2. All values of parameters are in cm−1, the uncertainty inthe last digits (one standard deviation) is given in the parenthesis.
Table 2: Molecular rovibrational parameters
Parameter 51 7191
Ev 887.657124(81) 882.21072(49)
A 0.43244023(96) 0.4317699(37)
B 0.3765583(20) 0.3757854(37)
C 0.20048282(20) 0.2001817(30)
∆K 1.629(46)×10−7 2.85(12)×10−7
∆JK 0.262(46)×10−7 –0.43(11)×10−7
∆J 2.1928(53)×10−7 2.258(17)×10−7
δK 3.478(17)×10−7 1.881(59)×10−7
δJ 0.8929(28)×10−7 1.209(19)×10−7
51 ⇔ 7191 Coriolis interaction parameters
Operator, Constant Value
Jz, CA1 3.4523(51)×10−2
{iJy, Jx}, CA2 3.997(12)×10−4
Jx, CB1 7.1521(68)×10−2
{iJy, Jz}, CB2 1.3064(22)×10−3
JxJ2, CB3 –6.15(23)×10−7
JxJ2z , CB4 6.128(62)×10−6
J3+ + J3
−, CB5 1.322(27)×10−6
The rms deviation of the fit with all weighted lines was about 0.51×10−3 cm−1. Still, theassignment of some ν7 + ν9 transitions in the ν5 Q branch is problematic.
Table 3: Statistics of the fit
fit of 1096 experimental energy levels(979 levels for ν5 and 117 levels for ν7+ν9)
0. 10−3cm−1 < 92.7 % of levels < 1. 10−3cm−1
1. 10−3cm−1 < 4.6 % of levels < 2. 10−3cm−1
2. 10−3cm−1 < 2.3 % of levels < 4. 10−3cm−1
4. 10−3cm−1 < 0.4 % of levels < 10. 10−3cm−1
Acknowledgements
The work of J.K. was supported through the Ministry of Education, Youth and Sports ofthe Czech Republic (research program LC06071) and a scholarship of the French government(Bourse du Gouvernement Francais - Bourse de Doctorat en co-tutelle).
References
[1] World Meteorological Organization, in “Scientific Assessment of Ozone Depletion: 2006”(WMO, Ed.), Geneva, 2007.
[2] A. Goldman, C. P. Rinsland, A. Perrin, and J.-M. Flaud, J. Quant. Spectrosc. Radiat.
Transfer 60, 851–861 (1998).
[3] A. Perrin, Spectrochim. Acta 54, 375–394 (1998).
[4] B. J. Drouin, C. E. Miller, J. L. Fry, D. T. Petkie, P. Helminger, and I. R. Medvedev, J.
Mol. Spectrosc. 236, 29–34 (2006).
[5] S. G. Chou, D. T. Petkie, R. A. H. Butler, and C. E. Miller, J. Mol. Spectrosc. 211,284–285 (2002).
[6] T. L. Tan, E. C. Looi, K. T. Lua, A. G. Maki, J. W. C. Johns, and M. Noel, J. Mol.
Spectrosc. 150, 486–492 (1993).
[7] T. L. Tan, E. C. Looi, K. T. Lua, A. G. Maki, J. W. C. Johns, and M. Noel, J. Mol.
Spectrosc. 149, 425–434 (1991).
[8] A. G. Maki, T. L. Tan, E. C. Looi, K. T. Lua, J. W. C. Johns, and M. Noel, J. Mol.
Spectrosc. 157, 248–253 (1993).
[9] T. L. Tan, E. C. Looi, K. T. Lua, A. G. Maki, J. W. C. Johns, and M. Noel, J. Mol.
Spectrosc. 166, 97–106 (1994).
