Article pubs.acs.org/JPCA
Study of the Q Branch Structure of the 14N and 15N Isotopologues of the ν4 Band of Ammonia Using Frequency Chirped Quantum Cascade Lasers Geoffrey Duxbury,* David Wilson, Kenneth Hay, and Nigel Langford Department of Physics, SUPA, The University of Strathclyde, John Anderson Building, 107 Rottenrow E, Glasgow G4 0NG, Scotland, U.K. ABSTRACT: Intrapulse quantum cascade (QC) laser spectrometers are able to produce both saturation and molecular alignment of a gas sample owing to the rapid sweep of the radiation through the absorption features. In the QC lasers used to study the 14N and 15N isotopologues of the ν4 band of ammonia centered near 1625 cm−1, the variation of the chirp rate during the scan is very large, from ca. 85 to ca. 15 MHz ns−1. In the rapid chirp zone the collisional interaction time of the laser radiation with the gas molecules is short, and large rapid passage effects are seen, whereas at the slow chirp end the line shape resembles that of a Doppler broadened line. The total scan range of the QC laser of ca. 10 cm−1 is sufficient to allow the spectra of both isotopologues to be recorded and the rapid and slow interactions with the laser radiation to be seen. The rapid passage effects are enhanced by the use of an off axis Herriott cell with an effective path length of 62 m, which allows a buildup of polarization to occur. The effective resolution of the chirped QC laser is ca. 0.012 cm−1 full width at half-maximum in the 1625 cm−1 region. The results of these experiments are compared with those of other studies of the ν4 band of ammonia carried out using Fourier transform and Laser Stark spectroscopy. They also demonstrate the versatility of the down chirped QC laser for investigating collisional effects in low pressure gases using long absorbing path lengths.
■
Mills,9 although the most recent version of the molecular line parameters of NH3 in the HITRAN 2008 database10 indicated that no information on the 15NH3 isotopologue for the 5−7 μm region. Huang, Schwenke, and Lee11,12 have developed a new global potential energy surface for 14NH3 and 15NH3 and have used it to predict spectroscopic assignments. In their paper II,12 Table 4, they have given their most recent calculations of the vibrational band origins of the ν4 bands of both 14NH3 and 15NH3. In the course of our developments of spectroscopic applications of quantum cascade (QC) lasers,13 we have investigated the Q branch region of both the 15NH3 and the 14NH3 isotopologues between 1630.7 and 1621.3 cm−1. As the ν4 band origins of the 14N isotopologue are symmetric “s”, 1624.8901 cm−1, and antisymmetric, “a”, 1626.0091 cm−1, and of the 15N isotopologue are “s”, 1621.7389 cm−1, and “a”, 1622.8168 cm−1,5 some similar, but shifted, transitions of the two isotopologues may be recorded. An isotope shift technique, VISTA, has been developed by Lees, Li, and Xu14,15 for assigning transitions in combination and overtone bands of 15NH3 in the 1.5 μm region relative to the equivalent assigned bands of 14NH3. We have used a variant of this method for the identification of some of the transitions.
INTRODUCTION The high resolution infrared spectrum of ammonia has been studied by a variety of methods since the pioneering laser Stark spectroscopy experiments of Shimizu.1,2 Initially, electric resonance methods were used because of their ability to allow Doppler and sub-Doppler resolution to be made.3,4 Once high resolution Fourier transform spectroscopy methods were developed, they began to be used to allow wide bandwidth infrared spectra to be recorded. A study by Sasada, Hasegawa, Amano, and Shimizu5 in 1982 allowed the information from microwave spectra, laser Stark spectroscopy, and Fourier transform spectra of 14NH3 and 15NH3 to be combined. More recently, as higher resolution Fourier transform spectrometers have been developed, the majority of the studies of high resolution infrared molecular spectra have been carried out using Fourier transform spectroscopy. A good example of the use of Fourier transform spectroscopy (FTS) has been given by a number of high resolution studies of the line positions and intensities of the 5−7 μm region of the absorption spectrum of the ν4 and 2ν2 bands of 14NH3. These include a very detailed analysis of the vibration−rotation−inversion spectrum by Cottaz et al.,6 the analysis of self-broadening, self-shift, and self-mixing by Aroui et al.,7 and measurements of line mixing and transition dipole moments by Aroui et al.8 Although the spectrum of the ν4 and 2ν2 bands has been studied in great detail, there has been much less work on the equivalent region of the absorption spectrum of 15NH3. The most complete studies of this region were published in 1982 by Sasada et al.5 and by Di Lonardo, Fusina, Trombetti, and © 2013 American Chemical Society
Special Issue: Oka Festschrift: Celebrating 45 Years of Astrochemistry Received: December 15, 2012 Revised: April 12, 2013 Published: April 12, 2013 9738
dx.doi.org/10.1021/jp3123665 | J. Phys. Chem. A 2013, 117, 9738−9745
The Journal of Physical Chemistry A
■
The transition dipole moment for the ν4 band of 14NH3, measured by Aroui et al. is 0.0420 D for the symmetric (s) transition and 0.0394 D for the antisymmetric (a) transition. The transition dipole moments of the ν4 band of 15NH3 measured by Malathy Devi et al.16 using a tunable diode laser spectrometer are about 18% larger than those derived for 14NH3 by Aroui et al.8 However, as only 33 transitions in the ν4 band system were used to measure the line intensities and nitrogen broadening coefficients, part of this difference may arise owing to the use of a smaller sample of absorption lines. From the standpoint of the use of the isotope shift method, the estimated error of the line intensities may be ca. 20%. The effective resolution of our pulsed QC laser spectrometer, based upon the measured full width at half-maximum (fwhm) of the absorbance lines is 0.012 cm−1, so that more of the absorption features are resolved than in the FTS measurements by Sasada et al.5 and Di Lonardo et al.9 where the resolution ranged from 0.06 to 0.08 cm−1. In our FTS measurements of the infrared absorption spectrum of 15NH3 we used a Bomem DA003 spectrometer with the resolution of 0.02 cm−1. In Figure 1a we show the central part of the ν4 and 2ν2 band system of 15NH3, recorded using the Bomem DA003, and in
Article
EXPERIMENTAL DETAILS The experimental measurements described in this paper were made using pulsed Quantum Cascade (QC) lasers employing the intrapulse method.13 A top hat current pulse of duration between 1000 and 2000 ns was applied to a distributed feedback (DFB) QC laser operating at 6.1 μm. with a drive voltage in the range 8−9 V. The temperature range used was set via a Peltier cooler and was varied from −30 to +45 °C. As rapid Joule heating of the laser and the distributed feedback (DFB) grating occurs, the laser frequency sweeps rapidly to lower frequency, a frequency downchirp. Here a 2 μs (2000 ns) top hat pulse, with a drive voltage of 9 V, and a repetition frequency of 5 kHz, has been used to drive the DFB QC laser 61. The rate of change with time of the frequency downchirp ranges from ca. 80 MHz ns−1 at the beginning of the pulse to ca. 15 MHz ns−1 at its end. The resultant spectrum recorded in the time domain is transformed to one in the frequency domain by making use of the etalon fringes recorded during the down-chirp. The resultant change in the chirp rate of the spectrum recorded at the highest pressure is shown in Figure 2a. Figure 2b shows a
Figure 1. Fourier transform spectrum of the Q branch region of the ν4 band of 15NH3 recorded using a Bomem DA003 spectrometer. The path length was 10 cm and the gas pressure ca. 10 Torr. The resolution was 0.02 cm−1. (a) Overview of the entire Q branch, showing the location of the QC laser spectra. (b) Expanded view of the band center showing the total tuning range of the frequency chirped QC spectrometer.
Figure 2. Examples of the rapid passage structure visible on the chirped frequency spectra induced by DFB QC laser 61. A 2 μs (2000 ns) top hat pulse has been used to drive the laser. The laser base temperature was −20 °C. The path length in the astigmatic Herriot cell is 62 m. Spectrum i 0.49 Torr, mainly 15NH3, drive voltage of 8.5 V, laser repetition frequency 2 kHz. Spectrum ii drive voltage of 9 V, laser repetition frequency of 5 kHz, (a) 0.24 Torr, mainly 15NH3 and (b) 0.24 Torr, mainly 14NH3.
Figure 1b the location of the region scanned by the QC laser over its full temperature tuning range. In the Experimental Details, we describe the functioning of pulsed, down-chirped quantum cascade (QC) laser spectrometers, and also the effect of down-chirped QC laser pulses on the types of absorption spectra recorded. The Analysis and Results contains a detailed comparison of the high resolution QC laser spectra of 15NH3 alnd 14NH3 in the central part of Q branch, where the spectra of the two isotopologues overlap, and shows examples of the effects of the rapid wavenumber down-chirp upon the line shape of the absorption lines.
comparisons between a spectrum where 14NH3 is dominant and one in which the main absorber is 15NH3. The rapid slowing of the scan rate of the laser frequency during the down chirp has a major effect on the shape of the absorption lines observed with an intrapulse spectrometer, as may be seen in Figure 2a. At the high wavenumber side (LHS) the scan rate is 9739
dx.doi.org/10.1021/jp3123665 | J. Phys. Chem. A 2013, 117, 9738−9745
The Journal of Physical Chemistry A
Article
Table 1. Summary of the Band Origins and Principal Molecular Constants of the ν4 = 1 States of 14NH3 and 15NH3, Including the Magnitude of the Origin Shifts of the Symmetric “s” and Antisymmetric “a” Components, Derived from the Paper by Sasada, Hasegawa, Amano, and Shimizu5 a 14
NH3
T′eff B′ (C′ − B′)eff Δ(C′−B′) DJ′ DJK′ a
15
s
a
NH3
1624.8901(12) 10.18005 (13) −2.50501 −0.31281 1.0355(41) × 10−3 −1.9801(84) × 10−3
1624.0091(13) 10.17370(89) −2.48915 −0.30252 0.9694(74) × 10−3 −1.817(17) × 10−3
s
a
1621.7389(10) 10.15498(123) −2.51524(23) −0.31270(18) 1.406(39) × 10−3 −1.9814(90) × 10−3
1622.8168(11) 10.15344(36) −2.50134(31) −0.30754(45) 0.9766(64) × 10−3 −1.832(15) × 10−3
T′eff = T′ + (C′ − B′) − DK′ − 2(Cζ′ + ηK′ + τK′) . (C′ − B′)eff = (C′ − B′) − 2DK′ + 3HK′ − Cζ′ − 3ηK′ − 5τK′.
approximately 80 MHz ns−1 and large emission spikes, followed by free induction decay signals, are observed at the longer wavelength side of each absorption line. This behavior is owing to the rapid passage of the chirped pulse though each absorption line. However, at the low wavenumber side (RHS), where the scan rate of the chirped pulse is ca. 20 MHz ns−1, the shape of an absorption lines resembles that of a Doppler broadened line. Effects of the Rapid Passage Caused by the downChirped QC Laser Pulse. The frequency down-chirp of laser 61 varies from ca. 86 MHz/ns at the beginning of the chirped pulse to ca. 10 MHz/ns at the end of the scan. The measured value of the Doppler broadened line at the end of the scan is ca. 300 MHz full width at half-maximum (fwhm), which is approximately twice that of the calculated Doppler line width of ca. 145 MHz. As a result of the high initial chirp rate, few collisions occur during the rapid initial chirp rate whereas a significant number occur at the slow end of the scan. These effects are due to the time dependence of the rate of the passage through the absorption line from very rapid to slow. In previous papers13,17 we have demonstrated that this behavior may be modeled using numerical solutions of the coupled Maxwell Bloch equations.17 Following the original interpretation of these effects in NMR spectroscopy by Ernst,18 we describe the effects on the line shape of the frequency down-chirp in terms of a normalized sweep rate. This quantifies the degree by which the frequency chirp rate departs from the slow sweep limit. The normalized sweep rate, a, is defined as
a=
dν /dt γ1γ2
14
NH3 the buffer gas, or vice versa. In the long path length refocusing cell, with an effective path length of 62 m, both the polarization, P, and the field, E, build up with each pass through the refocusing system. Hence the Rabi frequency increases with increasing propagation length. This behavior is similar to that which we have observed previously17 in collisions between nitrous oxide and the asymmetric form of carbon dioxide 18O12C16O. When carbon dioxide was used as a buffer gas, the rapid passage induced gain spikes are rapidly quenched, as shown in ref 17. Comparison of the Spectra of the Q Branch Regions of the ν4 Bands of 14NH3 and 15NH3 Recorded Using the Chirped Pulse QC Laser 61. The first detailed comparison of the Q-branch structure of the ν4 band of 14NH3 and 15NH3, including the magnitude of the origin shifts was made by Sasada and his colleagues5 and is summarized in Table 1. Although a much more comprehensive analysis was made of the 2ν2/ν4 system of 14NH3 by Cottaz and colleagues,6 we have chosen to derive the shifts from the earlier data because there is no equivalent recent study of the 2ν2/ν4 system of 15NH3. Although many of the absorption lines of 15NH3 can be identified using the data of Sasada et al.5 and Di Lonardo et al.,9 a considerable number of lines are overlapped at the resolution of their Fourier transform spectrometers, and also in Figure 1. By using a variant of the isotopic line shift technique of Lees, Li, and Xu,14,15 we have been able to assign transitions that were labeled as overlapped in the earlier papers of of Sasada et.al5 and Di Lonardo et al.9 In Table 2 we have included both columns of line positions, and their associated intensities, of the shifted lines, as well as a column containing the line positions measured from the absorption lines shown in Figures 3 and 4. Although the spectra of the sample of 15NH3 were recorded using two different sets of gas pressures of 14NH3 and of 15NH3, a residual spectrum of 14NH3 may be seen in all of the spectra of 15 NH3 recorded in the long path length absorption cell. This occurs partly due to the gas sample of 15NH3 containing some residual 14NH3, and partly due the adsorption of 14NH3 on to the glass walls of the long path length gas cell. The effect of this impurity is analyzed by comparing a spectrum of 14NH3 only with the spectra of the 15NH3 samples, and also by comparing two different sets of spectra of 14NH3 and of 15NH3. The transmission spectra have been converted to absorbance to facilitate the comparison of the pressure dependence of the intensity of the absorption lines. As we have seen in Figure 2, there is a very large variation of the chirp rate during the frequency down-chirp of the laser. In particular, large rapid passage signals, which are evident at the beginning of each scan, are almost absent at the slow end of the down-chirp. As the scan ranges of successive spectra overlap, this variation of behavior, from large rapid passage effects to quenching of the chirp induced ringing, may be seen in most of the spectra. Although the effective resolution of the
(1)
where dv/dt is the frequency chirp rate, γ1 is the population decay rate, and γ2 is that of the polarization. In a similar way the degree of power induced saturation may be quantified using a normalized saturation rate, s, defined as s=
(μ12 E(t ))2 γ1γ2
(2)
In molecules the population decay rate, γ1, and the polarization decay rate, γ2, are usually assumed to be approximately equal, such that we may set γ1 ≈ γ2 = γ, where γ is the pressure broadening coefficient. The value of γ depends upon both the pressures of the chromophore and of the particular buffer gas used, so that we may write γ = γ cpc + γ bpb
(3)
where c is the chromophore and b is the buffer gas. In the experiments described in this paper 15NH3 is the chromophore and 9740
dx.doi.org/10.1021/jp3123665 | J. Phys. Chem. A 2013, 117, 9738−9745
The Journal of Physical Chemistry A
Article
Table 2. Q-Branch Structure of the Center of the ν4 Bands of 14NH3 and 15NH3a 14
line position/ cm−1 1630.4574 1630.3088 1630.2339 1630.2220 1629.5936 1629.3133 1629.1909 1629.1133 1628.9694 1628.9226
1628.6105
1627.9390 1627.8274 1627.6291 1627.3215
1627.0478
1626.9319 1626.6716 1626.3522 1626.1289
1625.6093 1625.5171 1625.4651
1624.9661 1624.8955 1624.8628
1624.3903
1624.0161 1623.5591
line identity
intensity/ cm molecule−1
p
1.060 × 10−20 1.613 × 10−20
Q(1,1) s Q(7,0) s
r
r
4.508 × 10−21 7.890 × 10−22
r
−22
s Q(8,1) r s Q(11,3)
a Q(10,4)
a
r
−21
Q(8,2)
r
Q(7,1)
r
Q(5,0)
s
−20
r
Q(6,1)
r
Q(8,2) a Q(7,2) r s Q(3,0) r a Q(2,0) s
r
× × × ×
Q(9,3) s Q(7,2) r a Q(4,1)
a
r
Q(6,2)
r
Q(4,1)
line identity
1630.254
1630.3319
3.653 × 10−22
s
1629.841 s,d
1629.8492
1.043 × 10−20
s
1629.336 s,d
1629.4832
2.121 × 10−21
a
1629.002
1629.1408
1.298 × 10−22
a
1628.828 s,d 1628.726 1628.674 s
1628.9073 1628.8123 1628.7342
1.129 × 10−20 2.117 × 10−21 8.232 × 10−21
p a Q(2,1) r s Q(9,1) r a Q(8,0)
1628.261 s,d 1628.022
1628.2793 1628.1270
1.120 × 10−20 9.441 × 10−22
s
water 1627.700 s,db
0,1,0−0,0,0 1627.7010
202−111 1.012 × 10−20
a
1627.604
1627.5733
1.347 × 10−21
a
1627.303 s,d 1627.138 s,db
1627.3074 1627.1588
1.060 × 10−20 1.613 × 10−20
s
1627.049 s,db 1626.956
1627.08391 1627.0720
4.508 × 10−21 7.890 × 10−22
s
1626.026 1625.974 s,db 1625.866
1626.0409 1625.9633 1625.8550
2.153 × 10−21 2.906 × 10−20 4.421 × 10−21
1625.581 s
1625.7726
3.954 × 10−21
1625.392 s,d 1625.187 sb
1625.4605 1625.4353
8.350 × 10−21 1.020 × 10−22
s
1624.779 s,d 1624.579 s,d
1624.7890 1624.6697
4.429 × 10−20 1.436 × 10−20
s
1624.222 1624.163 s,d
1624.1160 1624.1715
1.182 × 10−22 5.882 × 10−20
s
1623.9 s,d water
1623.8978 010−000
1.359 × 10−20 211−202
p
Q(11,2)
p
Q(3,1)
r
Q(9,1)
p
Q(3,1)b
s
s
p
Q(2,1) Q(10,2)
r
p
Q(1,1)
r
Q(10,0)
p
Q(1,1) Q(7,0)
r
s
r
Q(8,1) Q(11,3)
r
10−21 10−21 10−20 10−20
−20
r
s
4.215 8.222 6.767 6.423
r
r
intensity/ cm molecule−1
NH3 line shifted down by 3.15 cm−1
1.359 × 10−20
8.474 × 10−21
Q(8,3)
Q(5,1) r s Q(1,0)
s
5.882 × 10
r
a
s
4.429 × 10−20 4.082 × 10−20
r
Q(4,0)
refs 5, 9, 10
14
8.350 × 10−21
1.958 × 10−21
a
exp NH3 line pos / cm−1
NH3
4.421 × 10 3.954 × 10−21
r
s Q(10,3)
b
15
−21
8.678 × 10 2.906 × 10−20 2.153 × 10−21
a(9,3)
s
7.642 × 10
r
Q(7,1) r a Q(6,0) r s Q(9,2) a
s
15
NH3
r
Q(9,2) Q(6,0) r a Q(8,2) s
a
a
r
r
Q(9,3)
1.901 × 10 4.706 × 10−20 r
Q(7,1) a Q(3,2) p
4.384 × 10−21 7.485 × 10−21 2.448 × 10−20 r
a
Q(5,0) Q(6,1)
r
1.310 × 10−20 a
a
Q(3,1) Q(4,0)
r
2.224 × 10−20 8.182 × 10
−20
9741
s
r
Q(6,1)
dx.doi.org/10.1021/jp3123665 | J. Phys. Chem. A 2013, 117, 9738−9745
The Journal of Physical Chemistry A
Article
Table 2. continued 14
line position/ cm−1 1623.3521
a
15
NH3
line identity
intensity/ cm molecule−1
r
2.332 × 10−20
Q(3,1)
exp NH3 line pos / cm−1 1623.334 1623.303 1623.227 1622.954
1622.7090 1622.5977 1622.5937 1622.4265 1622.2999 1622.0870 1621.9290 1621.7612
1621.6178 1621.3590
s
r
Q(6,2)
r
Q(3,1) r a Q(7,3) s
s
10−20 10−21 10−20 10−20
8.586 × 10
r
1.754 × 10−20
r
1.558 × 10−20 6.259 × 10−22
q
2.100 × 10−21
r
−20
Q(7,0)
Q(2,1)
1.427 × 10
sb,db sb,db sb,db s,d
NH3
refs 5, 9, 10
14
intensity/ cm molecule−1
line identity
10−21 10−20 10−20 10−20
r a Q(7,2) r s Q(3,0) r a Q(5,1) r a Q(2,0)
NH3 line shifted down by 3.15 cm−1
× × × ×
1623.521 1623.2022 1623.2021 1622.9789
8.222 6.767 2.049 6.423
water
0,1,0−0,0,0
423−330
1622.309 s,d
1622.3671
1.901 × 10−20
s
1622.295 s,d
1622.3151
4.706 × 10−20
s
1621.931 s
1621.8161
4.384 × 10−21
s
1621.727 1621.660 s,b
1621.8154 1621.7455
2.030 × 10−21 7.485 × 10−21
r a Q(9,4) r s Q(7,2)
1621.601 s
1621.7128
2.488 × 10−20
a
r
Q(5,1)
−21
Q(8,3)
a Q(2,1) r a Q(10,3)
s
× × × ×
r
a Q(5,2)
s
1.192 8.338 2.101 1.581
b
15
r
Q(1,0)
r
Q(9,3)
r
Q(4,1)
a
The line positions and intensities are derived in part from earlier data of Sasada et al.5 and of Di Lonardo et al.,9 and also from equivalent patterns in NH3 by using the method of Lees et al.15,16 This translates the 14NH3 lines in HITRAN10 to the lower wavenumber/cm−1 range of the 15NH3 system, by using a common isotopic shift of 3.15 cm−1. Key: s, Sasada et al.; sb, Sasada et al. blended; d, diLonardo et al.; db, diLonardo et al. blended. bKey: s, Sasada et al.; sb, Sasada et al. blended; d, diLonardo et al.; db, diLonardo et al. blended. 14
spectrometer is ca. 0.012 cm−1, some overlap, or partial overlap of lines may occur, in particular with lines associated with the two isotopologues occurs. Some of these ambiguities have been resolved by comparing the spectra where the amplitudes of the spectra have been varied by changing the gas pressure within the gas cell. The frequency ranges scanned are varied by changing the base temperature of the QC lasers from −30 °C, maximum start frequency, to +45 °C, minimum start frequency. The duration of the chirp and the drive current are also used to control the scan ranges. The range of values used in the experiments is given in Table 2. The first series of experiments were carried out using identical cell pressures of 0.24 Torr, for both 14NH3 and 15NH3, and a path length of 62 m. The spectra were recorded in a series of overlapping scans, and direct comparisons can be made between the absorbances of 14NH3 and 15NH3 shown in each spectral region. These start at a wavenumber of 1630 cm−1 in Figure 3a, the next region to that shown in Figure 2, and show the large change in the line shape from high to low frequency chirp rate. The 15NH3 spectra contain a contribution from 14NH3, but the converse is not the case. We attribute this difference to the composition of the sample of 15NH3 containing some 14NH3. In Figure 3b, −20 °C, the baseline curves upward on the right-hand side due to absorption by the 0,1,0−0,0,0, 211−212, line of water, H216O. This absorption occurs in the laboratory owing to the long free space path length before the laser beam enters the Herriot cell. This becomes more evident in Figure 3b, −10 °C, where the water line has its maximum absorption at line center, and dominates the absorbance profile. In the following figures, (d) and (e), in the range from 0 to 10 °C, the baselines are almost linear and the two isotopologues are easily distinguished. In the next two figures, (f) and (g), where the base
temperatures are 20 and 30 °C, respectively, a second strong water line, 0,1,0−0,0,0, 211−202, dominates the absorbance at 30 °C. Finally, in Figure 3h the baselines are almost linear, except that a much weaker line of water 0,1,0−0,0,0, 423−330, may be seen at ca. 1622.6 cm−1. Most of the attenuation is due to water vapor in the laboratory atmosphere. This absorption can vary greatly between very weak to quite strong. This leads to not only the observation of strong water lines but also the production of rapid oscillations, as the ratio of the intensities of water is equivalent to the division by a very small number. For the spectra shown in Figure 4, a turbomolecular pump was used to try to rid the cell of residual water, and the experiments were carried out using a much shorter time scale. When the experiments for Figure 4 were carried out, the absorption due to water in the laboratory air was almost the same for the reference scans of the evacuated cell, and for those containing the 14N and 15 N isotopologues. This led to the elimination of the contribution due to water except when the absorption of the laser light by the residual external water vapor was complete. A range of scans similar to those in Figure 3 are recorded, but using much lower pressures of 14NH3 and 15NH3. A low gas pressure of 20 mTorr was used to record the spectrum of 14NH3, and a much larger pressure of 201 mTorr was used for 15NH3. The result is that the residual 14NH3 lines in the spectrum of 15NH3 are almost identical in intensity to those recorded in the low pressure gas spectrum of 14NH3, making the absorption lines of 15NH3 much easier to identify. As the residual water vapor concentration in the Herriott cell had been reduced, the interference effects are minimized over most of the regions where interference with water occurs. The penalty paid for this is exaggerated oscillatory structure close to the center of the water absorption lines. 9742
dx.doi.org/10.1021/jp3123665 | J. Phys. Chem. A 2013, 117, 9738−9745
The Journal of Physical Chemistry A
Article
Figure 3. Comparison of the temperature dependent spectra of 15NH3 and of 14NH3 recorded at a gas pressure of 0.24 Torr, using DFB QC laser 61. A 2 μs (2000 ns) top hat pulse, with a drive voltage of 9 V, and with a laser repetition frequency of 5 kHz, has been used to drive the laser. The path length in the astigmatic Herriot cell is 62 m. Upper spectrum: 15NH3. Lower spectrum: 14NH3. QC laser temperatures: (a) −30 °C, (b) −20 °C, (c) −10 °C, (d) 0 °C, (e) 10 °C, (f) 20 °C, (g) 30 °C, and (h) 45 °C.
of the arQ(4,0) line of 14NH3 and the spQ(1,1) line of 15NH3, which are easily resolved with a separation of 0.014 cm−1. The results of these experiments, shown in Figures 3 and 4, have been summarized in Table 2, which contains an analysis of these data.
At a lower gas pressure, within the absorption cell, the change in line shape from the rapid passage limit to collision dominated behavior is more marked in Figure 4 than in Figure 3. In Figure 4d the high resolution of the chirped laser system is seen by the resolution 9743
dx.doi.org/10.1021/jp3123665 | J. Phys. Chem. A 2013, 117, 9738−9745
The Journal of Physical Chemistry A
Article
Figure 4. Comparison of the temperature dependent spectra of 15NH3 and of 14NH3 recorded using a 62 m path length. The laser parameters are as in Figure 3. Upper spectrum: 210 mTorr 15NH3. Lower spectrum: 21 mTorr 14NH3. QC laser temperatures: (a) −30 °C, (b) −20 °C, (c) −10 °C, (d) 0 °C, (e) 10 °C, (f) 20 °C, (g) 30 °C, and (h) 45 °C.
■
spectra of different isotopologues such as 14NH3 and 15NH3. This versatility of chirped pulse QC lasers has already been shown by experiments on the chemical composition of reactive plasmas carried out in the Ashfold group at Bristol University,19 as well as
CONCLUSIONS The purpose of this paper is to demonstrate the role that chirped pulse QC lasers can play in molecular spectroscopy of small molecules, in particular to provide rapid information about 9744
dx.doi.org/10.1021/jp3123665 | J. Phys. Chem. A 2013, 117, 9738−9745
The Journal of Physical Chemistry A
Article
(17) Tasinato, N.; Hay, K. G.; Langford, N.; Duxbury, G.; Wilson, D. J. Time Dependent Measurements of Nitrous Oxide and Carbon Dioxide Collisional Relaxation Processes by a Frequency Down-Chirped Quantum Cascade Laser: Rapid Passage Signals and the Time Dependence of Collisional Processes. J. Chem. Phys. 2010, 132, 164301. (18) Ernst, R. R. Sensitivity Enhancement in Magnetic Resonance. Advances in Magnetic Resonance; Academic Press: New York, 1966, Vol. 2, 1−135. (19) Cheesman, A.; Smith, J. A.; Ashfold, M. N. R.; Langford, N.; Wright, S.; Duxbury, G. J. Applications of a Quantum Cascade Laser for Time-Resolved, in Situ Probing of CH4/H2 and C2H2/H2 Gas Mixtures during Microwave Plasma Enhanced Chemical Vapor Deposition of Diamond. J. Phys. Chem. A 2006, 2821−2828. (20) Hancock, G.; Horrocks, S. J.; Ritchie, G. A. D.; van Helden, J. H.; Walker, R. J. Time Resolved Detection of the CF3 Photofragment using Chirped QCL Radiation. J. Phys. Chem. A 2008, 112, 9751−9757.
the study of time dependent kinetics by the Ritchie group in Oxford University,20 and measurements of the time dependence of collisional relaxation by Tasinato and his colleagues.13,17
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We thank the United Kingdom Engineering and Physical Research Council for research studentships to D.W. and to K.H. We are also grateful to the Leverhulme Trust for the award of an Emeritus Fellowship to G.D., and to the EU Accord program for the evaluation of new types of chirped frequency QCLs.
■
REFERENCES
(1) Shimizu, F. Dipole Moment of NH3 in the ν2 Excited State. J. Chem. Phys.Lett. 1969, 51, 2754−55. (2) Shimizu, F. Stark Spectroscopy of NH3 ν2 Band by 10-μ CO2 and N2O Lasers. J. Chem. Phys. 1970, 52, 3572−3576. (3) Weber, W. H.; Terhune, R. W. Stark Tuned Lamb Dip Spectroscopy of the ν4 and 2ν2 Bands of 14NH3. J. Chem. Phys. 1983, 78, 6422−6436. (4) Duxbury, G. Infrared Vibration-Rotation Spectroscopy; John Wiley & Sons Ltd.: Chichester, U.K., 2000. (5) Sasada, H.; Hasegawa, Y.; Amano, T.; Shimizu, T. High-Resolution Infrared and Microwave Spectroscopy of the ν4 and ν2 Bands of 14NH3 and 15NH3. J. Mol. Spectrosc. 1982, 96, 106−130. (6) Cottaz, C.; Kleiner, I.; Tarrago, G.; Brown, L. R.; Margolis, J. S.; Poynter, R. L.; Pickett, H. M.; Fouchet, T.; Drossart, P.; Lellouch, E. Line Positions and Intensities in the 2ν2/ν4 Vibrational System of 14NH3 Near 5−7 μm. J. Mol. Spectrosc. 2000, 203, 285−309. (7) Aroui, H.; Laribi, H.; Orphal, J.; Chelin, P. Self-Broadening, SelfShift and Self-Mixing in the ν2, 2ν2 and ν4 Bands of NH3. J. Quant. Spectrosc. Radiat. Transfer. 2009, 110, 2037−2059. (8) Aroui, H.; Broquier, M.; Picard-Bersellini, A.; Bouanich, J. P.; Chevalier, M.; Gherissi, S. Absorption Intensities, Pressure-Broadening and Line Mixing Parameters of Some Lines of NH3. J. Quant. Spectrosc. Radiat. Transfer. 1998, 60, 1011−1023. (9) Di Lonardo, G.; Fusina, L.; Trombetti, A.; Mills, I. M. The ν2, 2ν2, 3ν2, ν4, and ν2+ν4 Bands of 15NH3. J. Mol. Spectrosc. 1982, 92, 298−325. (10) Rothman, L. S.; Gordon, I. E.; Barbe, A.; Benner, D. C.; Bernath, P. F.; Birk, M.; Boudon, V.; Brown, L. R.; Campargue, A.; Champion, J.P.; et al. The HITRAN 2008 Molecular Spectroscopic Database. J. Quant. Spectrosc. Radiat. Transfer. 2009, 110, 533−572. (11) Huang, X.; Schwenke, D. W.; Lee, T. J. Rovibrational Spectra of Ammonia. I. Uprecedented Accuracy of a Potential Energy Surface Used With Nonadiabatic Corrections. J. Chem. Phys. 2011, 134, 044320. (12) Huang, X.; Schwenke, D. W.; Lee, T. J. Rovibrational Spectra of Ammonia. II. Detailed Analysis, Comparison, and Prediction of Spectroscopic Assignments For 14NH3, 15NH3, and 14ND3. J. Chem. Phys. 2011, 134, 044321. (13) Duxbury, G.; Langford, N.; Hay, K. G.; Tasinato, N. Quantum Cascade Laser Spectroscopy: Diagnostics to Non-Linear Optics. J. Mod. Opt. 2009, 56, 2034−2048. (14) Li, L.; Lees, R. M.; Xu, L.-H. External cavity Tunable Diode Laser Spectra of the ν1+2ν4 Stretch-Bend Combination Bands of 14NH3 and 15 NH3. J. Mol. Spectrosc. 2007, 243, 219−226. (15) Lees, R. M.; Li, L.; Xu, L.-H. New VISTA on Ammonia in the 1.5 μm Region: Assignments for the ν1+2ν4 Bands of 14NH3 and 15NH3 by Isotopic Shift Labelling. J. Mol. Spectrosc. 2008, 251, 241−251. (16) Malathy Devi, V.; Narahari Rao, K.; Pracna, P.; Urban, S. Intensities in the ν4 Band of 15NH3. J. Mol. Spectrosc. 1990, 146, 18−24. 9745
dx.doi.org/10.1021/jp3123665 | J. Phys. Chem. A 2013, 117, 9738−9745
