J. Phys. Chem. A 2010, 114, 12021–12027
12021
Detection of Vibrational Bending Mode ν8 and Overtone Bands of the Propargyl Radical, ˜ 2B1 HCCCH2 X Xu Zhang and Stanley P. Sander Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak GroVe DriVe, Pasadena, California 91109-8099, United States
Adam Chaimowitz Chemistry Department, Pomona College, 645 N. College AVenue, Claremont, California 91711, United States
G. Barney Ellison Department of Chemistry and Biochemistry, UniVersity of Colorado, Boulder, Colorado 80309-0215, United States
John F. Stanton Institute for Theoretical Chemistry, Department of Chemistry, UniVersity of Texas, Austin, Texas 78712, United States ReceiVed: June 17, 2010; ReVised Manuscript ReceiVed: September 7, 2010
Infrared (IR) absorption spectra of matrix-isolated HCCCH2 have been measured. Propargyl radicals were generated in a supersonic pyrolysis nozzle, using a method similar to that described in a previous study (Jochnowitz, E. B.; Zhang, X.; Nimlos, M. R.; Varner, M. E.; Stanton, J. F.; Ellison, G. B. J. Phys. Chem. A 2005, 109, 3812-3821). Besides the nine vibrational modes observed in the previous study, this investigation ˜ 2B1 out-of-plane bending mode (ν8) at 378.0 ((1.9) cm-1 in a cryogenic argon detected the HCCCH2 X matrix. This is the first experimental observation of ν8 for the propargyl radical. In addition, seven overtone and combination bands have also been detected and assigned. Ab initio coupled-cluster anharmonic force field calculations were used to guide the analysis. Furthermore, ν12, the HCCCH2 in-plane bending mode, has been assigned to 333 ((10) cm-1 based on the detection of its overtone (2ν12, 667.7 ( 1.0 cm-1) and a possible combination band (ν10 + ν12, 1339.0 ( 0.8 cm-1). This is the first experimental estimation of ν12 for the propargyl radical. Introduction The propargyl radical (HCCCH2) is an important species in high temperature flames as well as the low temperature atmosphere of Saturn’s moon, Titan.1-4 It is produced by H-abstraction from CH3CCH and CH2CCH2, and by the addition of CH2 a˜1A1 to HCtCH.5 Propargyl is a delocalized radical ˜ 2B1).6,7 Recombinaand can be represented by HCd ¨ Cs ¨ CH2 (X tion of propargyl radicals produces benzene, which is considered one of the key reactions in generating aromatic species in both combustion and Titan’s atmosphere.4,8 Due to the importance of propargyl, several experimental studies have been conducted to determine its rotational constants,9 electric dipole moment,10 ionization energy,11 and electron affinity.12,13 However, the molecular structure of the radical has never been determined experimentally; microwave spectroscopy has shown only that it is planar.9 The vibrational frequencies of the propargyl radical have been studied previously.10,12-19 Of its 12 fundamental vibrational modes, 10 have been detected. The two low frequency bending modes ν8 and ν12 have not previously been observed. Table 1 lists the results from earlier experimental studies. Among them, the most recent study6 by Jochnowitz et al. measured nine fundamental * To whom correspondence should be addressed. E-mail: xu.zhang@ jpl.nasa.gov.
modes in an Ar matrix, using a heated supersonic nozzle to generate the propargyl radicals. In addition, the study also presented ab initio coupled-cluster anharmonic force field calculations and predicted the frequencies for the two far-IR bending modes (ν8 ) 398 cm-1 and ν12 ) 338 cm-1). The goal of this work is to measure the two far-IR bending modes (ν8 and ν12) and some overtone/combination bands of the propargyl radical. In this paper, we report the detection of ν8 and seven overtone/combination bands, using a method described by Jochnowitz et al.6 We also report an inferred experimental frequency of ν12 based on the observation of its overtone 2ν12 band and a tentative assignment of the ν10 + ν12 combination band. Experimental Section A. Radical Beam and Matrix IR Spectroscopy. We use a hyperthermal nozzle to produce intense beams of propargyl radical. The experimental apparatus has been described previously.20,21 Briefly, the nozzle consists of a resistively heated SiC tube (1 mm inside diameter, 2 mm outside diameter) at the output of a pulsed solenoid Parker General Valve (Series 9). The hyperthermal nozzle can be heated as high as 1800 K to thermally dissociate an appropriate precursor. Because the residence time in the hyperthermal nozzle is estimated20,29 to be about 30-100 µs, few radical-radical byproducts are
10.1021/jp105605f 2010 American Chemical Society Published on Web 10/25/2010
12022
J. Phys. Chem. A, Vol. 114, No. 45, 2010
Zhang et al.
TABLE 1: Earlier Experimental Studies of HCCCH2 Vibrational Modes mode a1
b1
b2
ν (cm-1)
methods
ref
3308.5 ( 0.5 3322.2929 ( 0.0020 3322.15 ( 0.01 3310 3308.8 3307 3028.3 ( 0.6 3026 1935.4 ( 0.4 2080 1440.4 ( 0.5 1440 1061.6 ( 0.8 686.6 ( 0.4 687.17603 ( 0.00062 688 483.6 ( 0.5 490 ( 10 NA 3111 1016.7 ( 0.4 620 ( 2 647.3 NA
Ar matrix CW color center laser spectroscopy CW color center laser spectroscopy/He nanodrops N2 matrix Ar matrix Ar matrix Ar matrix Ar matrix Ar matrix Ar matrix Ar matrix Ar matrix Ar matrix Ar matrix time-resolved IR diode laser spectroscopy N2 matrix Ar matrix CH2dC ) CHs photodetachment
6 18, 19 10 14 16 15 6 15 6 15 6 15 6 6 17 14 6 12, 13
Ar matrix Ar matrix Ar matrix time-resolved IR diode laser spectroscopy
15 6 6 17
description
1
C-H st
2
sym CH2 st
3
CtC st
4
CH2 scissors
5 6
C-C st CH2CCH umbrella
7
C-H out-of-plane bend
8 9 10 11
CH2CCH out-of-plane bend asym CH2 st C-C x CtC in plane bend C-H in plane bend
12
CH2CCH in-plane bend
observed. Upon exiting the nozzle, the heated molecules undergo a supersonic expansion since they are entrained by the Ar buffer gas. The experimental setup in this work is similar to that described previously,20,21 except that here the SiC tube is glued into an alumina (Al2O3) base that is sealed with an O-ring to the faceplate of the pulsed valve.11 The previous setup has the SiC tube sitting directly on the faceplate of the valve. The advantage of the current configuration is that heat transfer from the nozzle to the valve is decreased, which improves the performance and longevity of the valve. It also prevents the leakage of gas through the gap between the SiC tube and the valve faceplate in the previous configuration. Two different propargyl radical precursors were used in this work: HCtCCH2Br and HCtCCH2CH2ONO. The propargyl bromide precursor was purchased from Aldrich Chemical Co. while 1-butyn-4-nitrite (HCtCCH2CH2ONO) was synthesized by a standard method.6 Propargyl itself is produced by heating the precursors, viz.: ...
...
HCtCsCH2sBr + ∆ f HCdCsCH2 + Br•
(1)
HCtCsCH2sCH2sOsNdO + ∆ f [HCCCH2CH2O• + •NO] f
(2)
...
...
HCdCsCH2 + HCHO + •NO These two precursors are chosen because of their availabilities and their relatively weak bond energies28 (D298(C3H3-Br) = 62 kcal/mol, D298(C3H3CH2O-NO) = 42 kcal/mol, and D298(C3H3-CH2O) = 15 kcal/mol). The hyperthermal nozzle was mounted to the vacuum shroud of an Advanced Research System (ARS) two-stage closed-cycle helium cryostat, approximately 2.5 cm away from the cryogenic CsI window. Room temperature gas mixtures were created by seeding the degassed vapor of the precursor in argon (about 0.05% mole fraction of the precursor). The hyperthermal nozzle was operated with an approximate 1 ms pulse width and a stagnation pressure of 1.1 atm. The pressure drop in the
stagnation reservoir (1.2 L) was measured using a capacitance monometer to determine the gas throughput (average about 0.4 mmol/minute of the gas mixture). Beams of propargyl radicals in Ar were deposited onto the cold CsI window at about 10 K. After dosing the matrix substrate for a few hours, the CsI window was cooled to about 4.5 K. Subsequently, the infrared spectrum of the sample was measured using a Nicolet Magna 560 Fourier transform infrared spectrometer with a deuterated L-alanine-doped triglycine sulfate (DLaTGS) detector. The ARS cryostat is equipped with a pair of CsI side windows that are traversed by the IR beam. The IR spectra were collected at 0.5 and 1 cm-1 resolutions. The duration of signal averaging for a typical experiment was several hours (about 10-20 h). B. Electronic Structure Calculations. Positions of the fundamental and two-quantum overtone and combination levels of propargyl were estimated using vibrational second-order perturbation theory22 (VPT2) in conjunction with force fields calculated with coupled-cluster theory. In particular, the CCSD(T) treatment23 of electron correlation was used in the frozen-core approximation, along with the atomic natural orbital (ANO) basis sets of Almlo¨f and Taylor.24 The values given in Table 2 were calculated using harmonic force constants taken from a calculation done with the 5s4p3d2f1g (carbon) and 4s3p2d1f (hydrogen) contractions (ANO2), and the remainder of the VPT2 parameters (cubic and quartic force constants, rotational constants and Coriolis zeta matrices) taken from a calculation using the smaller 4s3p2d1f (carbon) and 4s2p1d (hydrogen) contractions (ANO1). All quantum chemical calculations were done with the CFOUR program system.25 Calculations of the vibrational levels were done with the GUINEA module of CFOUR, which does a numerical sumover-states treatment of VPT2. In the course of the harmonic derivative analysis,26 a reasonably strong resonance was found involving the ν4 and 2ν6 levels. The ANO2/CCSD(T) harmonic positions for these levels are 1466 and 1336 cm-1, a difference of 130 cm-1, but the force constant that couples them is quite large (φ466 ) -183 cm-1) and the straightforward VPT2 estimates of these level positions (along with the related 2ν4, ν4 + 2ν6 and 4ν6 levels) are significantly affected by this
Vibrational Bending and Overtone Bands of HCCCH2 2B1
J. Phys. Chem. A, Vol. 114, No. 45, 2010 12023
˜ 2B1 Fundamental Vibrational Modes and Overtone/Combination Bands from This Worka TABLE 2: HCCCH2 X Ar matrix IR/10 K mode a1
b1 b2
1 2 3 4 5 6 7 8 9 10 11 12
description C-H st sym CH2 st CtC st CH2 scissors C-C st CH2CCH umbrella C-H out-of-plane bend CH2CCH out-of-plane bend asym CH2 st C-C x CtC in plane bend C-H in plane bend CH2CCH in-plane bend
2ν12 2ν7 ν6 + ν 8 ν10 + ν12 2ν6 2ν10 2ν4
-1
CCSD(T) υ (cm )
A (km mol-1)
3308.3 ( 0.5 3028.2 ( 0.6 1935.2 ( 0.4 1440.1 ( 0.8 (T) 1061.6 ( 0.8 686.5 ( 0.7 483.6 ( 0.7 378.0 ( 1.9 na 1016.8 ( 0.6 619.5 ( 1.4 333 ( 10
3318.9 3035.7 1921.2 1465.1 (F) 1056 689.6 483.5 397.5 3116.9 1017.4 615.8 335.3
38.7 3.3 5.6 0.5 (F) 6.8 37.4 46.3 7.9 0.8 2.2 47.5 5.3
667.7 ( 1.0 972.5 ( 0.9 1053.9 ( 0.9 1339.0 ( 0.8 (T) 1369.2 ( 0.6 2029.7 ( 0.5 2863.3 ( 0.8
670 961.9 1083.2 1351.6 1370.6 (F) 2032.0 2859.8 (F)
0.2 10.6 0.2 0.5 3.4 (F) 0.3 0.2 (F)
ν (cm )
-1
a T: tentative assignment; F: Fermi resonance is treated in the calculation by diagonalization of an effective Hamiltonian (VPT2+F). Note that ν12 has been assigned to 333 ((10) cm-1 based on the detection of its overtone (2ν12, 667.7 ( 1.0 cm-1) and a possible combination band (ν10 + ν12, 1339.0 ( 0.8 cm-1). Detailed analysis can be found in the Discussion section.
interaction. Therefore, a deperturbation and diagonalization procedure was used to obtain the estimates shown in Table 2 for the assigned ν4, 2ν4, and 2ν6 levels in the spectrum. This procedure, in which a dressed effective Hamiltonian is constructed and subsequently diagonalized, is a common way of dealing with Fermi resonance and is denoted27 as VPT2+F. Results ˜ 2B1 radical has 12 vibrational modes (5a1 The HCd ¨Cs ¨CH2 X x 3b1 x 4b2); all of them are IR active. Of the two propargyl precursors, HCtCCH2Br and HCtCCH2CH2ONO, propargyl bromide produces a cleaner (fewer bands due to contaminants) radical beam in this work. The main contaminants when using HCtCCH2Br as the precursor are benzene and propargyl bromide (the unpyrolyzed precursor). Production of benzene indicates that secondary chemistry is occurring in the pyrolysis nozzle. In the case of the HCtCCH2CH2ONO precursor, more contaminants have been observed, including benzene, HCtC-CH3, HCtC-CH2-CH2-CtCH (1,5-hexadiyne), which are also secondary reaction products. In addition, HCtCCH2CH2OH, the starting material for HCtCCH2CH2ONO synthesis, has also been detected. Figures 1-7 show the IR absorption spectra of propargyl radical using propargyl bromide as precursor in different wavelength regions. A summary of our assignments for the vibrational modes of the propargyl radical, {ν}i, is listed in Table 2, along with the ab initio predictions for the level positions, {υ}i. Each of the Figures 1-7 is discussed below. Figure 1a and 1b shows the spectra in the low frequency fingerprint region, from 370 to 500 cm-1. Figure 1a illustrates the spectra (3 traces) of propargyl generated from propargyl bromide pyrolysis. Two bands in this region are detected at 378.0 ( 1.9 and 483.6 ( 0.7 cm-1. The strong feature at 483.6 cm-1 has been observed previously in an Ar matrix6 and assigned to the CH out-of-plane bending mode (b1), ν7. A gasphase photodetachment study12,13 also assigned ν7 to be 490 ( 10 cm-1. The weak band at 378.0 cm-1 has not been detected before. According to the ab initio calculations, the fundamental
frequency for the HCd ¨Cs ¨CH2 out-of-plane bending mode (b1), ν8, is estimated at 398 cm-1. Consequently, the band at 378.0 cm-1 is assigned to ν8. This assignment has also been confirmed by the comparison of band intensity changes from one propargyl spectrum to another with different amount of propargyl radicals (intensity comparison or progression method). Figure 1a demonstrates the intensity comparison for ν7 and ν8. The band intensity of ν7 increases from trace I to trace II to trace III. Meanwhile, the band intensity for ν8 progresses in the similar trend: increases from trace I to II to III. Both of these features have also been observed when using HCtCCH2CH2ONO as the precursor. However, since HCtCCH2CH2OH is present as a contaminant in the propargyl spectra from HCtCCH2CH2ONO pyrolysis, the propargyl ν8 band overlaps with the HCtCCH2CH2OH band at 383.5 cm-1, which results in a broadening of the ν8 band. Figure 1b illustrates the propargyl spectra from propargyl bromide pyrolysis (black trace) and HCtCCH2CH2ONO pyrolysis (green trace), along with the HCtCCH2CH2OH spectrum from an authentic sample (purple trace). Spectra from other side products such as benzene, 1,5hexadiyne, fulvene, dimethylene cyclobutene, etc. have also been collected to confirm our assignment. The issue that the experimental value for ν8 is about 20 cm-1 lower than the CCSD(T) prediction will be discussed in a later section. Figure 2 depicts the spectrum of another fingerprint region from 610 to 700 cm-1. Here three bands are detected for propargyl, 619.5 ( 1.4, 667.7 ( 1.0, and 686.5 ( 0.7 cm-1. Two of them, ν11 ) 619.5 cm-1 and ν6 ) 686.5 cm-1, have been observed previously in an Ar matrix6 and with timeresolved IR diode laser spectroscopy in the gas phase.17 The ν6 band has also been detected in a N2 matrix.14 In this work, the weak signal at 667.7 cm-1 is assigned to 2ν12, according to the CCSD(T) prediction of this overtone band at 670 cm-1. In addition, using HCtCCH2CH2ONO as the precursor along with the intensity comparison method also supports the assignment of this band to propargyl. Bands that belong to unpyrolyzed propargyl bromide, CO2, and benzene can also be seen in this spectrum. In Figure 3, four bands are assigned to propargyl
12024
J. Phys. Chem. A, Vol. 114, No. 45, 2010
Zhang et al.
Figure 2. The fingerprint region from about 610 to 700 cm-1 of the propargyl radical produced by the thermal decomposition of propargyl bromide (HCtCCH2Br) at roughly 1300 K. The experimental frequencies {ν} are marked by bullets (•). The CCSD(T) calculated anharmonic frequencies {υ} are shown as red sticks. The intensity of the anharmonic overtone 2υ12 band is calculated to be small, so for display purposes it is enhanced by a factor of 10 and is marked with an asterisk (*).
Figure 1. The low frequency fingerprint region (from about 370 500 cm-1) of the propargyl radical. The experimental frequencies {ν} are marked by bullets (•). The CCSD(T) calculated anharmonic frequencies {ν} are shown as red sticks. (a) The propargyl radical generated by the thermal decomposition of propargyl bromide (HCtCCH2Br). Experimental conditions for Trace I (yellow) and Trace II (black) are: 0.05% gas mixture of C3H3Br in Ar and about 1300 K nozzle temperature; the band intensity differences from trace I to II are resulted from different amount of gas mixture deposited on the substrate. The experimental condition for trace III (blue) are: 0.1% gas mixture of C3H3Br in Ar and about 1400 K nozzle temperature; the higher concentration of the gas mixture generates more propargyl radical but even more secondary products (which can be seen as some weak features in trace III but not present in traces I and II). (b) The propargyl radical generated by the thermal decomposition of 1) propargyl bromide (HCtCCH2Br) at roughly 1300 K (black trace); 2) butyne nitrite (HCtCCH2–CH2ONO) at approximately 900 K (green trace). The purple trace is the spectrum of 3-butyn-l-ol (HCtCCH2–CH2OH) from its authentic sample.
radical; two of them are fundamentals, ν10 ) 1016.8 ( 0.6 cm-1 and ν5 ) 1061.6 ( 0.8 cm-1, which have been observed previously.6 The strong feature at 972.5 ( 0.9 cm-1 is assigned to the overtone of ν7 (2ν7), which was also assigned by Jochnowitz et al.6 The intensity comparison method and the spectrum from HCtCCH2CH2ONO pyrolysis have also been used to show that these features come from propargyl. The weak feature at 1053.9 ( 0.9 cm-1 is attributed to the combination bands, ν6 + ν8, based on the CCSD(T) anharmonic frequency calculations as well as the intensity comparison method. The small peak at 1073 cm-1 is from an unknown contaminant; it is not assigned to ν6+ν8 due to: (1) it does not appear in the pyrolysis spectra of the other precursor, HCtCCH2CH2ONO; and (2) its intensity does not satisfy the intensity comparison
Figure 3. The fingerprint region from about 950 to 1100 cm-1 of the propargyl radical produced by the thermal decomposition of propargyl bromide (HCtCCH2Br) at roughly 1300 K. The experimental frequencies {ν} are marked by bullets (•). The CCSD(T) calculated anharmonic frequencies {υ} are shown as red sticks. The intensity of the anharmonic combination bands, υ6 + υ8, is calculated to be small, so for display purposes it is enhanced by a factor of 10 and is marked with an asterisk (*).
method. Weak signals from the contaminants such as benzene and propargyl bromide can also be seen in this spectrum. Figure 4 is the spectrum in the fingerprint region from 1335 to 1466 cm-1. Three bands are observed for propargyl radical. The very weak feature at 1440.1 ( 0.8 cm-1 is tentatively assigned to the fundamental ν4 mode, which has been detected earlier6,15 in an Ar matrix. The strong peak at 1369.2 ( 0.6 cm-1 has also been observed in Ar matrices previously6,15 and was assigned to the overtone (2ν6) of the intense CH2CCH umbrella mode (b1), ν6. The weak band at 1338.9 ( 0.8 cm-1 has never been detected; here it is tentatively assigned to the combination band of ν10 + ν12. In addition, the propargyl spectrum with the HCtCCH2CH2ONO precursor as well as the intensity comparison method also supports these assignments. Figure 5 illustrates the propargyl spectra from 1920 to 2050 cm-1 and the 2840-2880 cm-1 region. The ν3 ) 1935.2 ( 0.4
Vibrational Bending and Overtone Bands of HCCCH2 2B1
Figure 4. The fingerprint region from about 1335 to 1466 cm-1 of the propargyl radical produced by the thermal decomposition of propargyl bromide (HCtCCH2Br) at roughly 1300 K. The experimental frequencies {ν} are marked by bullets (•). The CCSD(T) calculated anharmonic frequencies {υ} are shown as red sticks. The intensities of the anharmonic combination band, υ10 + υ12, as well as the fundamental υ4 mode are calculated to be small, so for display purposes they are enhanced by a factor of 10 and are marked with an asterisk (*).
Figure 5. The spectra from the 1920 to 2050 cm-1 and 2840-2880 cm-1 regions of the propargyl radical generated by the thermal decomposition of propargyl bromide (HCtCCH2Br) at roughly 1300 K. The experimental frequencies {ν} are marked by bullets (•). The CCSD(T) calculated anharmonic frequencies {υ} are shown as red sticks. The intensities of the anharmonic overtone bands, 2υ10 and 2υ4, are calculated to be small, so for display purposes they are enhanced by a factor of 10 and are marked with an asterisk (*).
cm-1 band has been detected previously.6 The weak band at 2029.7 ( 0.6 cm-1 has not been reported before; here it is assigned to the overtone (2ν10) of the C-C x CtC in-plane bending mode (b2), ν10. Again, the propargyl spectrum from HCtCCH2CH2ONO precursor as well as the intensity comparison method also confirms this assignment. The feature at 2863.3 ( 0.8 cm-1 is assigned to the overtone (2ν4) of the CH2 scissors mode (a1), ν4, based on the CCSD(T) calculation as well as the intensity comparison method. Figures 6 and 7 illustrate the CH stretching regions of the propargyl radical. There are three CH stretching modes in HCd ¨Cs ¨CH2. Figure 6 shows the CH2 symmetric stretching mode (a1), ν2, measured at 3028.2 ( 0.6 cm-1. The CH2 asymmetric stretching mode (b2), υ9, is calculated to be a very weak band at 3116.9 cm-1; it is not observed in this work due to the low sensitivity of the DLaTGS detector in this wavelength region. In Figure 7, the strong peak at 3308.3 ( 0.5 cm-1 is attributed to the C-H stretching mode (a1), ν1. Both ν1 and ν2
J. Phys. Chem. A, Vol. 114, No. 45, 2010 12025
Figure 6. The CH stretching region from about 3010 to 3130 cm-1 of the propargyl radical produced by the thermal decomposition of propargyl bromide (HCtCCH2Br) at roughly 1300 K. The experimental frequencies {ν} are marked by bullets (•). The CCSD(T) calculated anharmonic frequencies {υ} are shown as red sticks. The intensity of the anharmonic fundamental υ9 mode is calculated to be small, so for display purposes it is enhanced by a factor of 10 and is marked with an asterisk (*). The fundamental ν9 band is not detected here in this work.
Figure 7. The CH stretching region from about 3300 - 3350 cm-1 of the propargyl radical produced by the thermal decomposition of propargyl bromide (HCtCCH2Br) at roughly 1300 K. The experimental frequencies {ν} are marked by bullets (•). The CCSD(T) calculated anharmonic frequencies {υ} are shown as red sticks.
have been detected by Jochnowitz et al. in an Ar matrix.6 All three CH stretching modes, ν1, ν2, and ν9, have been observed by Korolev et al. in an Ar matrix with positions at 3307, 3026, and 3111 cm-1, respectively.15 The ν1 mode has also been detected in Ar and N2 matrices previously.14,16 Two high resolution studies10,18 using CW color center laser spectroscopy have been fit, yielding band origins of ν1 to be 3322.2929 ( 0.0020 cm-1 in a free jet18 and 3322.15 ( 0.01 cm-1 in a helium nanodroplet.10 Discussion The results of this study are in good agreement with earlier studies for bands of propargyl that have previously been observed and assigned. The only previous assignment of a fundamental that might be considered to be somewhat uncertain is for ν4, which was assigned to a feature at 1440 cm-1 both by
12026
J. Phys. Chem. A, Vol. 114, No. 45, 2010
Jochnowitz et al. and again here. The reason for caution is that there is a relatively strong Fermi resonance between this level and the overtone of ν6, which was overlooked in the theoretical analysis of the previous study. Without treatment of this resonance, the locations of these two levels are 1445.3 and 1390.4 cm-1, with the overtone (at lower energy) having significantly greater intensity (4.1 vs < 0.01 km mol-1). The calculated value of 1445.3 cm-1 certainly helped direct the assignment of this very weak feature at 1440 cm-1. However, when the resonance between the two levels is properly treated via the VPT2+F approach, the two level positions change to those in Table 2, 1465.1 and 1370.6 cm-1, with qualitatively similar relative intensities. The latter value is in quite good agreement with the assignment for the strong 2ν6 feature, but the revised estimate of the fundamental is now some 25 cm-1 away from the assigned position of ν4. It is possible that this assignment is in error (and hence is marked as tentative in Table 2); another possible candidate for ν4 is the unassigned feature at 1445.4 ( 0.6 cm-1. Although the calculations suggest that the ν12 fundamental is located below the nominal range of our detector, the present study allows us to establish its position with a reasonable degree of certainty. Specifically, the overtone of ν12 is assigned here to a band at 667.7 cm-1 and the combination ν10 + ν12 to 1338.9 cm-1. The ab initio calculations of the positions of these two bands, 670.0 and 1351.6 cm-1 respectively, can be used to estimate the location of ν12 and also serve as a consistency check of sorts for these assignments. Specifically, the error in the calculated position of a two-quantum level should be, in the absence of resonance effects, about equal to the sum of the errors in the associated fundamental positions. That is, if 2ν12 is at 670.0 cm-1, then the fundamental location should be at about 335.3 - 0.5 (670 - 667.7) ) 334.1 cm-1. Similarly, the position of ν12, as inferred from the ν10 + ν12 assignment, is 335.3 + y cm-1, where y ) (1339.0 -1351.6 - 1016.8 + 1017.4) ) -12.0, that is, 323.3 cm-1. However, the latter assignment of ν10 + ν12 ) 1338.9 cm-1 has to be regarded as somewhat tentative because it is found only very weakly in the spectrum. Taken together, an estimate of 333 ( 10 cm-1 seems appropriate for ν12. For the overtone 2ν7, the calculated position is 10 cm-1 below the observed feature, despite the fact that the corresponding fundamental (for which the assignment is quite certain) is within 0.5 cm-1 of the calculated position. Given the nature of ν7, this can be understood by noting that this mode, which corresponds to an out-of-plane wagging motion of the acetylenic CH group, has a very strong quartic anharmonicity. This is reminiscent of allene and analogous to the propargyl resonance structure that • places the unpaired spin on the ethynyl carbon (HCdCdCH2). In propargyl radical, the quadratic and (diagonal) quartic constants that correspond to this mode are 482 and 4246 cm-1, respectively. The value of 2ν7 calculated from VPT2 is quite sensitive to the magnitude of this large quartic constant; adjusting it upward by about 20 cm-1 brings the calculated position of 2ν7 into quite good agreement with the observed level position. The procedure used in this work, where the harmonic force constants are calculated at one level of theory and the anharmonic constants at another, might very well experience problems for this case. Moreover, VPT2, which is well-suited for cases of Morse-like anharmonicity, is less suited for cases like this, where there is a large quartic anharmonicity. Finally, there is also a relatively large discrepancy between the calculated and assigned positions for the ν8 fundamental. This is apparently not due to a basis set effect; the harmonic
Zhang et al. frequencies calculated at the ANO1 and ANO2 levels differ by only 0.02 cm-1 (388.53 and 388.51 cm-1, respectively). VPT2 predicts a positive anharmonicity, while the assigned level position is below the calculated harmonic frequency. Nonetheless, the assignment seems quite certain, and it might well be that this is a failure of VPT2 or perhaps due to a relatively large matrix shift. In the future, this issue might be clarified with gas-phase IR spectra or perhaps spectra obtained in matrix isolation using a neon which usually has smaller gas-matrix shifts. Acknowledgment. This research was carried out by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration (NASA). This work was supported by the grant from NASA planetary atmosphere program and NASA postdoctoral fellowship program. Additional support for this work comes from the US Department of Energy and the Robert A. Welch Foundations (to J. F. S.). G. B. E. acknowledges support from the Chemical Physics Program, United States Department of Energy (DEFG02-93ER14364) and the National Science Foundation (CHE0848606). We would like to thank Prof. Fred Grieman and Pomona College for their support. The authors would also like to thank Dave Nazic for his laboratory support. References and Notes (1) Miller, J. A.; Melius, C. F. Combust. Flame 1992, 91, 21. (2) Fahr, A.; Nayak, A. Int. J. Chem. Kinet. 2000, 32, 118. (3) Tang, W. Y.; Tranter, R. S.; Brezinsky, K. J. Phys. Chem. A 2006, 110, 2165. (4) Wilson, E. H.; Atreya, S. K.; Coustenis, A. J. Geophys. Res.-Planets 2003, 108. (5) Hahn, D. K.; Klippenstein, S. J.; Miller, J. A. Faraday Discuss. 2001, 119, 79. (6) Jochnowitz, E. B.; Zhang, X.; Nimlos, M. R.; Varner, M. E.; Stanton, J. F.; Ellison, G. B. J. Phys. Chem. A 2005, 109, 3812. (7) Kasai, P. H. J. Am. Chem. Soc. 1972, 94, 5950. (8) Miller, J. A.; Klippenstein, S. J. J. Phys. Chem. A 2003, 107, 7783. (9) Tanaka, K.; Sumiyoshi, Y.; Ohshima, Y.; Endo, Y.; Kawaguchi, K. J. Chem. Phys. 1997, 107, 2728. (10) Kupper, J.; Merritt, J. M.; Miller, R. E. J. Chem. Phys. 2002, 117, 647. (11) Minsek, D. W.; Chen, P. J. Phys. Chem. 1990, 94, 8399. (12) Oakes, J. M.; Ellison, G. B. J. Am. Chem. Soc. 1983, 105, 2969. (13) Robinson, M. S.; Polak, M. L.; Bierbaum, V. M.; Depuy, C. H.; Lineberger, W. C. J. Am. Chem. Soc. 1995, 117, 6766. (14) Jacox, M. E.; Milligan, D. E. Chem. Phys. 1974, 4, 45. (15) Korolev, V. A.; Maltsev, A. K.; Nefedov, O. M. Bull. Acad. Sci. USSR DiV. Chem. Sci. 1989, 38, 957. (16) Huang, J. W.; Graham, W. R. M. J. Chem. Phys. 1990, 93, 1583. (17) Tanaka, K.; Harada, T.; Sakaguchi, K.; Harada, K.; Tanaka, T. J. Chem. Phys. 1995, 103, 6450. (18) Morter, C. L.; Domingo, C.; Farhat, S. K.; Cartwright, E.; Glass, G. P.; Curl, R. F. Chem. Phys. Lett. 1992, 195, 316. (19) Yuan, L.; DeSain, J.; Curl, R. F. J. Mol. Spectrosc. 1998, 187, 102. (20) Zhang, X.; Friderichsen, A. V.; Nandi, S.; Ellison, G. B.; David, D. E.; McKinnon, J. T.; Lindeman, T. G.; Dayton, D. C.; Nimlos, M. R. ReV. Sci. Instrum. 2003, 74, 3077. (21) Friderichsen, A. V.; Radziszewski, J. G.; Nimlos, M. R.; Winter, P. R.; Dayton, D. C.; David, D. E.; Ellison, G. B. J. Am. Chem. Soc. 2001, 123, 1977. (22) Mills, I. M. Modern Spectroscopy: Modern Research; Rao, C. K., Mattews, C. W., Eds.; Academic Press: New York, 1972; Vol. 1, pp 115. (23) Raghavachari, K.; Trucks, G. W.; Pople, J. A.; Head-Gordon, M. Chem. Phys. Lett. 1989, 157, 479. (24) Almlöf, J.; Taylor, P. R. J. Chem. Phys. 1987, 86, 4070. (25) Stanton, J. F.; Gauss, J.; Harding, M. E.; Szalay, P. G. CFOUR, Coupled Cluster Techniques for Computational Chemistry, 2009; http:// www.cfour.de. (26) Matthews, D. A.; Stanton, J. F. Mol. Phys. 2009, 107, 213. (27) Schneider, H.; Vogelhuber, K. M.; Schinle, F.; Stanton, J. F.; Weber, J. M. J. Phys. Chem. A 2008, 112, 7498. (28) The bond energy, D298(HCCCH2-Br), is not known because ∆fH298(HCCCH2Br) has not been measured. The heat of formation of
Vibrational Bending and Overtone Bands of HCCCH2 2B1 propargyl bromide has been estimated [Ritter, E. R.; Bozzelli, J. W. Int. J. Chem. Kinet. 1991, 23, 767–778, by using Benson’s method of Group Equivalents. Benson, S. W. Thermochemical Kinetics; 2nd ed.; WileyInterscience: New York, 1976]. A value of ∆fH298(HCCCH2Br) ) 49 kcal/ mol has been reported [Kern, R. D.; Chen, H.; Kiefer, J. H.; Mudipalli, P. S. Combust. Flame 1995, 100, 177–184]. Negative ion studies have produced the heat of formation for propargyl radical itself, ∆fH298(CH2CCH) ) 82.5 ( 3.0 kcal/mol [Robinson, M. S.; Polak, M. L.; Bierbaum, V. M.; DePuy, C. H.; Lineberger, W. C. J. Am. Chem. Soc. 1995, 117, 6766– 6778]. Consequently, one finds D298(HCCCH2-Br) ) 62 kcal/mol. Likewise the bond energies of the nitrites must be estimated. Standard tables list ∆fH298(CH3ONO) [Pedley, J. B.; Naylor, R. D.; Kirby, S. P. Thermochemistry of Organic Compounds; 2nd ed.; Chapman and Hall: New York, 1986] and the heat of formation of methoxy radical has been found to be ∆fH298(CH3O) ) 5.0 ( 0.5 kcal/mol [Ruscic, B.; Boggs, J. E.; Burcat, A.; Csaszar, A. G.; Demaison, J.; Janoschek, R.; Martin, J. M. L.; Morton, M. L.; Rossi, M. J.; Stanton, J. F.; Szalay, P. G.; Westmoreland, P. R.;
J. Phys. Chem. A, Vol. 114, No. 45, 2010 12027 Zabel, F.; Berces, T. J. Phys. Chem. Ref. Data 2005, 34, 573–656]. Consequently it is true that D298(CH3O-NO) ) 42.7 ( 0.6 kcal/mol. A reasonable estimate for D298(HCCCH2CH2O-NO) is 42 ( 2 kcal/mol. The fragmentation energy of the alkoxy radical, HCCCH2CH2O f HCCCH2 + CH2O, will be roughly 15 kcal/mol. The dissociation energy of ethoxy radical is known to be D298(CH3-CH2O) ) 12 ( 1 kcal/mol [ Ruscic, B.; Boggs, J. E.; Burcat, A.; Csaszar, A. G.; Demaison, J.; Janoschek, R.; Martin, J. M. L.; Morton, M. L.; Rossi, M. J.; Stanton, J. F.; Szalay, P. G.; Westmoreland, P. R.; Zabel, F.; Berces, T. J. Phys. Chem. Ref. Data 2005, 34, 573–656]. Because the stabilization energy of the propargyl radical is about 10 kcal/mol [Jochnowitz, E. B.; Zhang, X.; Nimlos, M. R.; Varner, M. E.; Stanton, J. F.; Ellison, G. B. J. Phys. Chem. A 2005, 109, 3812– 3821] one expects the HCCCH2-CH2O bond energy to be about that of ethoxy radical or less. (29) Daily, J. W. Private communication, 2010.
JP105605F
