1. NH3 measurements in the far-IR
14NH3 Line positions and intensities in the FIR: Comparison of FT-IR measurements to empirical Hamiltonian model predictions.
Keeyoon Sung1, ShanShan Yu1, John Pearson1,
Fridolin Kwabia Tchana2, Laurent Manceron2, Olivier Pirali2,3
1 Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, USA
2 AILES Beamline, Synchrotron SOLEIL and LISA, UMR 7583 CNRS-Univ. Paris Est, Créteil, France
3 ISMO, UMR 8214 CNRS-Université Paris-Saclay, Orsay, France
© 2016 All rights reserved.

2. Motivation and Data acqusition
Primary objectives
Comparison to Calculations
Yu et al. 2010; Pearson et al.2016
Measuring weak lines in the FIR;
esp. ΔK=3 forbidden transitions
Effective pumping to high J/K states
Line intensities are critical in explaining the observed excitation
Experimental conditions
Gas samples
Normal sample
Spectral region
50 – 660 cm-1
Resolution(B)
0.003 cm-1
Pressures
2.0, 6.0 mbar
Path lengths
10.33, m
Temperatures
295 K
Data acquisition
Constantan is a copper-nickel alloy usually consisting of 55% copper and 45% nickel.
2.52 m base White cell

3. NH3 line positions and intensities
Spectrum fitting
Voigt profile assumed
sinc function with FOV correction
Non-linear curve fitting
One spectrum at a time
Line position, intensity, self-pressure broadening, simultaneously retrieved
Averaging the measurements
Grouping within 0.5*resolution
Positions, intensities, widths were averaged
Compiled into one set of list
~ 4820 lines were retrieved
Frequency calibration
18 isolated water features
HITRAN as standards
freq. accuracy ≥ cm-1
Residuals = Observed - Calculated

4. Spectroscopic sample characterization
Filtering out contaminated species
H2O – 343 lines (HITRAN)
15NH3 – lines (15NH3 spectrum from SOLEIL)
Abundances
H2O (0.2%)
15NH3 (0.33%  0.905*INA)
NH2D (0.037%: INA assumed)
Total impurity = 0.57%
NH2D – 530 lines (CDMS;
14NH3 – most of the remaining lines
Cologne Database for
Molecular Spectroscopy
(Muller et al. 2001)
NH2D (ξ= )
15NH3
Δv = cm-1
Ps = 2.7 Torr
cm-1

5. Quantum assignment (I)
Eight bands measured
cf. Energy level diagram
Assignments available from predictions
(Yu et al. 2010)
Synthetic spectra
SPFIT/SPCAT prediction
15NH3 (HITRAN+)
H2O (HITRAN)
Three observed spectra
(optically thin; intermediate, thick)
Interactive quantum assignment
Quantum assignment
Developed a graphical tool
Interactively validated by comparing line intensity
Scanned through the features within matching criteria
Results are compiled band by band
Fig. Inversion-vibration energy levels of 14NH3 up to 2000 cm-1

6. Results and Comparison: gs(a-s)
Line positions
Allowed transitions(A)
Excellent agreement with both
Forbidden transitions(F)
Deviation at high J from H12
Line intensities
(A): high J lines
(F): intermediate J lines
(A): Measured low
(F): Yu et al. shows
large offsets,
up to by a factor of 300.
gs(a-s) band

7. Results and Comparison: v2(a-s)
v2 (a-s) band
Line positions
Allowed transitions(A)
Good agreement with both
but diverging at high J lines
Forbidden transitions(F)
sparse but excellent agreement
Line intensities
(A): Agree better with Yu et al.
(F): Agree better with HITRAN

8. Results and Comparison: v4 (a-s)
v4 (a-s) band
Line positions
Allowed transitions(A)
Excellent agreement with both.
A few outliers at high J
Forbidden transitions(F)
Excellent agreement
Line intensities
(A): Good agreement in general,
but, progressively diverging
toward high J lines
(F): outliers from predictions

9. Results and Comparison: v2 – gs
v2 - gs band
Only high J lines
from P-branch captured
Line positions
Allowed transitions(A)
Very good
except a few outliers.
Line intensities
(A): Agree better than HITRAN
Also compared with
HM86
(Hermanussen et al. 1986)

10. Results and Comparison: v4 – v2
v4 – 2v2 band
Line positions
Allowed transitions(A)
Forbidden transitions(F)
Excellent agreement
Line intensities
(A): Much better agreement
with HITRAN, esp. low J lines
Also compared with
HM86:
(Hermanussen et al. 1986)
2v2 – v2(Used to be a big problem)

11. Results and Comparison: v4 – 2v2
v4 – 2v2 band
This band
Not listed in HITRAN
Line positions
Allowed transitions(A)
Forbidden transitions(F)
Excellent agreement
Line intensities
(A): Measured low
with large spread on K
(F): Sparsely measured
for high J lines and
randomly scattered
2v2(Used to be a big problem

12. Compiled line parameters#
Bands
#Meas.(A; F)
J range
K range
RMS(v)cm-1)
RMS(%dS) 1
gs (a-s)
152; 140
16 ≤ J ≤ 24
0 ≤ J ≤ 22
8.1 2
v2 (a-s)
156;
11 ≤ J ≤ 19
0 ≤ J ≤ 18
7.2 3
2v2(a-s)
164;
2 ≤ J ≤ 18
1 ≤ J ≤ 18
6.0 4
v4 (a-s)
365;
2 ≤ J ≤ 17
7.4 5
v2 – gs
94;
14 ≤ J ≤ 19
6.1 7
2v2 – v2
109;
6 ≤ J ≤ 19
6.7 9
v4 – v2
259;
1 ≤ J ≤ 17
1 ≤ J ≤ 15
8.0 8
v4 – 2v2
445;
6.4
Total
Meas.
2835
3×10-25 ≤ S ≤ 1×10-22 cm/molecule
0 – 2 %
2 – 5 %
5 – 10 %
Assigned
2047
Allowed(A)
1744
Forbidden(A)
303
Unassigned
788
14NH3, NH2D (?)

13. Summary and further work
Obtained far-FIR NH3 spectra
using AILES beamline.
Measured pos. & intensities
for ~2835 NH3 transitions.
Made quantum assignments
for ~2047 transitions from 8 bands
For positions, SPFIT predictions
(Yu et al. 2010, Pearson et al. submitted)
agree to ~0.002 cm-1 or better.
Predicted line intensities for gs(a-s) ΔK=3 transitions are observed to be overestimated by a factor of a few 100.
For the unassigned lines, empirical lower state energies will be derived from cold spectra.
Need Decent Line strengths ~ 20% Or
Accurate line positions
Acknowledgements
Research described in this talk was performed at Jet Propulsion Laboratory, California Institute of Technology, and was supported by the Astrophysics Research and Analysis (APRA) Program under the National Aeronautics and Space Administration. Yu and Sung acknowledge the Synchrotron Soleil for granting us the AILES time.