Rotational spectrum of three conformers of 3,3-difluoropentane: Construction of a 480 MHz bandwidth chirped-pulse Fourier-transform microwave spectrometer

Journal of Molecular Spectroscopy 261 (2010) 35–40

Contents lists available at ScienceDirect

Journal of Molecular Spectroscopy journal homepage: www.elsevier.com/locate/jms

Rotational spectrum of three conformers of 3,3-difluoropentane: Construction of a 480 MHz bandwidth chirped-pulse Fourier-transform microwave spectrometer Daniel A. Obenchain a, Ashley A. Elliott a, Amanda L. Steber a, Rebecca A. Peebles a, Sean A. Peebles a,*, Charles J. Wurrey b, Gamil A. Guirgis c a b c

Department of Chemistry, Eastern Illinois University, 600 Lincoln Avenue, Charleston, IL 61920, USA Department of Chemistry, University of Missouri-Kansas City, Kansas City, MO 64110, USA Department of Chemistry and Biochemistry, The College of Charleston, 66 George Street, Charleston, SC 29424, USA

a r t i c l e

i n f o

Article history: Received 3 February 2010 Available online 6 March 2010 Keywords: Difluoropentane Broadband Fourier-transform microwave Conformer

a b s t r a c t The rotational spectra for three conformers of 3,3-difluoropentane have been measured using both a newly constructed narrow bandwidth chirped-pulse Fourier-transform microwave spectrometer and a Balle–Flygare resonant cavity Fourier-transform microwave spectrometer. The chirped-pulse instrument produces a microwave pulse spanning up to 480 MHz bandwidth in the 7–18 GHz region by mixing a 1 ls chirped pulse (of up to 240 MHz bandwidth) from an arbitrary function generator with the output from a microwave synthesizer. Rotational spectra for the normal isotopic species and all possible 13C single substitutions were observed for the gauche–gauche and anti-gauche conformers, allowing a Kraitchman substitution structure and an inertial fit structure to be determined. 13C isotopic species and dipole moment components were not measurable for the less intense anti–anti species as a result of partially resolved fine splitting. Details of the new chirped-pulse instrument will be described and the structural results will be presented and compared with ab initio data for 3,3-difluoropentane. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction The recent introduction of broadband chirped-pulse Fouriertransform microwave (CP-FTMW) spectrometers by Pate [1,2] and Cooke [3] has provided a powerful new tool for the assignment of complex microwave spectra. However, the cost of chirped-pulse instruments that are capable of recording a full 11 GHz broadband spectrum with each gas pulse exceed the typical budget of research groups at smaller institutions. Fortunately, the technology is easily adaptable to enable the instrument’s bandwidth to be adjusted to fit within the desired price range [3]. We report here on the construction of a scaled-down CP-FTMW spectrometer that has a bandwidth of up to 480 MHz, providing significant advantages over resonant cavity FTMW spectrometers of the Balle–Flygare type [4]. Our incentive to construct such an instrument lies in its potential for the study of exotic species such as ions or radicals, where it increases the probability for location of spectral transitions, a concept already proven by coupling such an instrument with a laser ablation source [3]. CP-FTMW spectrometers thereby provide a practical solution for simplifying the optimization process for source conditions in pulsed discharge nozzle or laser ablation * Corresponding author. Fax: +1 217 581 6613. E-mail address: [email protected] (S.A. Peebles). 0022-2852/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.jms.2010.03.002

experiments. With good quality ab initio predictions able to provide estimates of rotational constants to within a few percent of experiment, the likelihood of rotational transitions for an unknown species falling within one or two frequency steps of the 480 MHz bandwidth spectrometer is high, thereby greatly reducing the amount of time spent in searching. Even for collection and interpretation of microwave spectra of stable and neutral molecular species, there are clear advantages over a standard Balle–Flygare instrument. For instance, the acquisition of a survey spectrum spanning the full 7–18 GHz region (even a spectrum having a relatively poor signal to noise ratio) offers improved confidence in spectral assignment since it is much more likely that significant spectral features (such as the existence of vibrationally excited states or tunneling or inversion motions) will be promptly identified. We report in this paper on the initial construction and testing of a 480 MHz bandwidth instrument, focusing on the assignment of the rotational spectra of three conformers of the straight chain alkane, 3,3-difluoropentane. This molecule, with total dipole moments of all three conformers predicted in the range 2.5–2.8 D, provides a perfect candidate for testing and optimization of this new system and illustrates the usefulness of a scaled-down broadband microwave spectrometer, at a cost which makes it more affordable for principally undergraduate institutions. The 3,3-difluoropentane has four conformers (arising from rotation about the two C–C single bonds connected to the central

36

D.A. Obenchain et al. / Journal of Molecular Spectroscopy 261 (2010) 35–40

carbon atom – see Fig. 1). In a recent FT-microwave study of pentane [5], only spectra for anti–anti (aa) and anti-gauche (ag) species were assigned, while the gauche–gauche (gg) conformer (with methyl groups on opposite sides of the C–C–C plane) and a sterically crowded gauche–gauche0 (gg0 ) conformer (with methyl groups on the same side of the C–C–C plane) were not observed. This is consistent with the expected order of stability in pentane (where a room temperature distribution of 51% aa, 36% ag, 13% gg is predicted) [5]. Our own recent ongoing studies of a series of straight chain alkanes with Si or Ge substitutions (such as diethylsilane [6,7], diethyldifluorosilane [6], diethylgermane [8], n-butylsilane [9] and n-butylgermane [10]) also makes 3,3-difluoropentane an interesting addition to this series so that structural comparisons and insights should eventually be possible.

2. Experimental Design of the narrow bandwidth CP-FTMW instrument used in this study followed basic concepts introduced by Pate [1] for an 11 GHz bandwidth instrument, and later modifications for a reduced (ca. 4 GHz) bandwidth instrument introduced by Cooke [3]. Fig. 2 shows a schematic of our instrument, along with a list of key components. A basic outline of the instrument’s operation is given below. Briefly, a 1 ls duration chirped frequency pulse from an arbitrary function generator ((component (1), Fig. 2), Tektronix AFG3251) is mixed (4) with the output of an HP8673G microwave synthesizer (2) and then amplified via a 10 W solid state microwave amplifier (5) (Microwave Power, L0618-40-T646), giving a pulse of up to 480 MHz bandwidth centered around the microwave synthesizer frequency. This polarizing radiation then enters the chamber and is broadcast onto the pulsed molecular gas expansion via a wide-band microwave horn antenna ((6), ATM WRD750-442Ò C3). A Kapton window (ATM WRD750-230-G3-G3) ensures that vacuum integrity is maintained between the waveguide feedthrough and microwave horn. The gas sample is pulsed through a Parker Hannifin (General Valve) Series 9 nozzle of 0.8 mm diameter orifice (7), which is aligned perpendicular to the microwave horns and pulses straight into the throat of a 10 in. Varian VHS250 diffusion pump. The vacuum chamber (8), an ISO-250 6-way cross with two 8 in. extensions to accommodate the microwave horn antennae, has a volume of approximately 85 L and maintains an idle pressure of mid-10–7 Torr, increasing to mid-10–5 Torr when gas is pulsed into the chamber. Pulsing rates of 10 Hz are achievable with this vac-

Fig. 1. Structures of the four conformers for 3,3-difluoropentane obtained from ab initio calculations at the MP2/6-311++G(2d,2p) level.

uum chamber although typically 4 Hz were used in this experiment for optimal scope performance. Sheets of microwave absorbing foam (Emerson and Cuming Eccosorb HR-25/ML) placed in the vacuum chamber prevent cavity resonances, although with the relatively low power used, such resonances have not been observed to be a problem. Rotational polarization induced in the molecules by an incoming microwave pulse is detected by a low noise amplifier, LNA ((10), Miteq AMF-5F-08001800-14-10P) on the detection side of the circuit; this LNA is protected by an SPST switch ((9), HP33102A) during the polarizing radiation pulse. Molecular emission is detected immediately after cessation of the polarizing pulse and the free induction decay (FID) is collected over 20 ls, amplified by the LNA, and then this signal is mixed back down (40 ) to radiofrequency (RF) with the split output (3) from the microwave synthesizer (2). This RF signal is finally passed through a 1–500 MHz amplifier ((11), Miteq AU-2A-0150) before being sent to an oscilloscope ((12), Tektronix TDS 5054B) for averaging and fast Fouriertransformation (FFT). The instrument described here is currently configured such that both upper and lower sidebands are collected simultaneously and absolute frequencies of any transitions are determined by carrying out a second measurement at a slightly offset center frequency; these two files are then compared by a simple peak-picking program written in LabView [11] to determine absolute frequencies of the rotational transitions. More sophisticated peak-picking routines are currently being developed. Timing for the gas pulse is provided by a Quantum Composers QC9614+ pulse generator that controls an Iota One valve driver, while a second digital delay/pulse generator (Stanford Research Systems DG-535) synchronized to the QC9614+ nozzle pulse provides timing to both the 10 W amplifier (5) and LNA protection switch (9). All equipment is referenced to a 10 MHz rubidium frequency standard (Stanford Research Systems FS725). Although the reduced bandwidth instrument described here cannot provide the deep averaging that is possible with the broadband instrument designed by Pate [1,2] (in which multiple nozzles and multiple FID’s per gas pulse allow millions of averages to be accumulated over the entire 11 GHz region), spectral intensity in the present study was sufficient to observe spectra for numerous 13 C substituted species, which was adequate for initial assignment of multiple conformers. A sample of 3,3-difluoropentane was prepared by adding 2.15 g (0.025 mol) of a freshly distilled sample of 3-pentanone in 3 mL diglyme to a solution of 0.025 mol of diethylaminosulfur trifluoride in 5 mL of diglyme dropwise at –20 °C under nitrogen. The reaction mixture was warmed to room temperature and stirred for 10 h and was frozen using liquid nitrogen and degassed. The volatile contents were collected in a tube containing 3 mL of NaHCO2 immersed in liquid nitrogen and after warming to room temperature was stirred for 30 min and the volatile material was collected in another tube containing water and stirred for 20 min. Finally the volatile material was collected and purified using trap to trap distillation for a total of three times. The final sample was checked by NMR and IR spectroscopy. To perform the microwave experiment, a small amount of 3,3-difluoropentane vapor was transferred to a 2 L stainless steel tank, pressurized to about 5 atm with He:Ne (17.5%:82.5%, BOC Gases) giving a sample concentration of about 0.4%. This mixture was then delivered to the pulsed nozzle at a constant pressure of about 1.8 bar. Subsequent experiments have proven a sample tank of this concentration to be sufficient to run a 5000 gas pulse sample scan with a 0–240 MHz chirped pulse to provide a spectrum of the entire 7– 18 GHz frequency range (amounting to approximately 24 frequency steps). Numerous improvements can almost certainly be made in terms of optimizing the sample and timing conditions and these are currently being explored.

37

D.A. Obenchain et al. / Journal of Molecular Spectroscopy 261 (2010) 35–40

MW source

2

1

AFG

11

Oscilloscope (5GS/s)

3

4′

RF

12

4

MW

7

10

5 MW

6

6

9 8 1. Arbitrary Function Generator DC-240 MHz (Tektronix AFG3251); 2. Hewlett-Packard microwave synthesizer 2-26.5 GHz (HP8673G); 3. Wilkinson power divider, 2-way, 2-18 GHz (Omni Spectra 204947); 4 and 4′. Double-balanced mixers 5-20 GHz (Miteq DM0520LW1); 5. 10 Watt solid state amplifier (Microwave Power L0618-04-T646); 6. Double ridged microwave horns 7.5–18 GHz (ATM 750-442-C3);7. Parker Hannifin Series 9 solenoid valve, 0.8 mm orifice; 8. Vacuum chamber (Nor-Cal ISO-250 6 way cross, with 8 inch extensions to accommodate horns), pumped by Varian VHS-250 10” diffusion pump and Adixen 2033SD rotary vacuum pump; 9. SPST switch (HP33102A); 10. 8-18 GHz low-noise amplifier (Miteq AMF-5F-08001800-14-10P); 11. 1-500 MHz amplifier (Miteq AU-2A-0150); 12. 500 MHz digital oscilloscope (Tektronix TDS5054B); Fig. 2. Schematic of the 480 MHz bandwidth chirped-pulse Fourier-transform microwave spectrometer. See text for details of operation.

Additional spectral measurements of 13C isotopologues and Stark effect measurements were carried out in this study using a standard Balle–Flygare type resonant cavity instrument [4,12] although initial assignments of all spectra, including those of the 13 C species (observed in natural abundance), were made with the CP-FTMW spectrometer. Stark effect measurements were carried out by application of voltages up to ±5 kV to a pair of steel mesh plates placed 31 cm apart in the resonant cavity instrument’s vacuum chamber, straddling the molecular expansion; electric field calibration was achieved by measurement of the J = 1 0 transition of OCS and assuming a dipole moment of 0.71519(3) D [13]. 3. Results 3.1. Rotational spectra Optimizations were carried out using Gaussian 03 [14] at the MP2/6-311++G(2d,2p) level to predict relative stabilities and rotational constants of all four conformers; results are listed in Table 1 and principal axis coordinates and structural parameters for these optimized structures are available as supplementary data. Both zero-point energy (ZPE) uncorrected and ZPE corrected energies predicted the stability order gg > aa > ag > gg0 although the energy difference between gg and aa conformers is small (18 cm–1 at the ZPE corrected level), so given the sensitivity of these relative energies to changes in basis set and level of calculation, relative stabil-

Table 1 Ab initio rotational constants, relative stabilities and dipole moment components for the four possible conformers of 3,3-difluoropentane. See Fig. 1 for structures. Parameters

Gauche–gauche

Anti–anti

Anti-gauche

Gauche–gauche0

A (MHz) B (MHz) C (MHz) la (D) lb (D) lc (D) ltotal (D) DE (cm1)a DEZPE (cm1)b

2788.3 2398.2 1793.2 0.00 2.53 0.00 2.53 0 0

4829.3 1677.8 1676.3 0.00 2.22 0.00 2.22 59 18

3689.1 1921.2 1709.4 0.94 1.79 1.24 2.37 125 100

2626.9 2557.3 1768.5 2.07 0.00 1.33 2.46 967 914

a Energy (zero-point energy uncorrected) relative to the gauche–gauche conformer. b Energy (zero-point energy corrected) relative to the gauche–gauche conformer.

ities of the three lowest energy conformers are difficult to state unambiguously; this will be explored further in Section 4. Initial scans with the CP-FTMW spectrometer were made around frequencies predicted by ab initio optimizations for the gg conformer and rotational transitions were seen within 100 MHz of these predictions. The high density of transitions observed in some spectral regions enabled spectra belonging to the other two conformers, as well as numerous 13C transitions, to be quickly identified, although small isotopic shifts required concurrent analysis of several species to avoid incorrect assignments. Measured transitions were fitted with a Watson A-reduced Hamiltonian [15] using the SPFIT program [16] and all resulting spectroscopic constants are given in Tables 2–4. Measured transition frequencies are available as supplementary data. The gg (C2 symmetry) and aa (C2v symmetry) species both exhibited pure b-type spectra while a-, b- and c-type transitions were all observed for the C1 symmetry ag species. 13C spectra were measured in natural abundance for both the gg and ag conformers, and rotational constants and planar moments for these are shown in Tables 2 and 3. Trends in both sets of parameters upon isotopic substitution help confirm the nature of the isotopic assignment. Note, for instance, that the close similarity of rotational constants to the parent molecules for 13C substitution at the central carbon atom (C3) in both the gg and ag conformers (Tables 2 and 3) indicate that this atom is close to the center of mass in the ag conformer, with two of its three principal axis coordinates equal to or close to zero; this will be further discussed in Section 3.3. The strongest observed transitions had intensities of around 35 mV, corresponding to signal to noise ratios in excess of 150 for 150 averaged gas pulses. (Extended averaging for 5000 gas pulses reduces the noise level to around 30 lV, increasing the signal to noise ratios to over 500 for stronger lines and making 13C transitions easily visible.) The full-width at half maximum height of transitions measured in this work were typically around 100– 130 kHz, up to three times that as measured on the resonant cavity instrument. Transitions for the gauche–gauche species were typically a little less broad than those for the ag and aa conformers, presumably as a result of partially resolved internal rotation splittings in the latter two species. This fine structure was particularly apparent for the aa species (see Fig. 3) in which some transitions showed clear indications of internal rotation splittings (this is in line with previous observations of the pentane spectrum [5]); in these cases an average frequency for the multiplets was used in

38

D.A. Obenchain et al. / Journal of Molecular Spectroscopy 261 (2010) 35–40

Table 2 Spectroscopic parameters for the gauche–gauche conformer of 3,3-difluoropentane. See Fig. 1 for atom numbering scheme.

a b c d e

Spectroscopic parameter

Parent

13

A (MHz) B (MHz) C (MHz) DJ (kHz)b DJK (kHz) DK (kHz) dJ (kHz) dK (kHz) Dmrms (kHz)c Nd Paa (uÅ2)e Pbb (uÅ2) Pcc (uÅ2)

2798.1353(14) 2354.5078(11) 1772.9048(6) 1.018(35) 4.16(15) 5.34(13) 0.434(17) 0.47(8) 2.5 20 159.5437(2) 125.5133(2) 55.0995(2)

2773.4743(6) 2315.7444(14) 1741.1095(4) 1.018 4.16 5.34 0.434 0.47 2.1 9 163.1400(1) 127.1227(1) 55.0961(1)

C1

a (5)

13

C2

13

C3

(4)

2786.0966(6) 2336.3378(14) 1765.6990(4) 1.018 4.16 5.34 0.434 0.47 3.4 9 160.5698(1) 125.6506(1) 55.7427(1)

2795.8353(7) 2354.7230(14) 1772.0261(10) 1.067(32) 4.16 5.34 0.434 0.47 3.4 9 159.5303(2) 125.6681(2) 55.0933(2)

The pairs of carbon atoms (1 and 5) and (2 and 4) are equivalent by symmetry. Centrifugal distortion constants for the isotopic species were fixed at values obtained for the normal isotopologue. P Dmrms ¼ ½ ðmobs  mcalc Þ2 =N1=2 . N is the number of fitted transitions. P P aa ¼ 0:5ðIb þ Ic  Ia Þ ¼ i mi a2i , etc.

Table 3 Spectroscopic parameters for the anti-gauche conformer of 3,3-difluoropentane. See Fig. 1 for atom numbering scheme.

a b c d

Spectroscopic parameter

Parent

13

13

13

A (MHz) B (MHz) C (MHz) DJ (kHz)a DJK (kHz) DK (kHz) dJ (kHz) Dmrms (kHz)b Nc Paa (uÅ2)d Pbb (uÅ2) Pcc (uÅ2)

3682.6421(7) 1905.1784(5) 1694.5378(4) 0.263(10) 0.174(17) 2.13(8) 0.064(6) 3.3 41 213.1367(9) 85.1034(9) 52.1293(9)

3663.7431(6) 1869.4311(6) 1662.3049(6) 0.263 0.174 2.13 0.064 3.2 9 218.2104(1) 85.8127(1) 52.1280(1)

3653.2711(7) 1898.9857(9) 1686.0120(4) 0.263 0.174 2.13 0.064 4.3 8 213.7716(1) 85.9766(1) 52.3595(1)

3681.1440(8) 1905.2960(9) 1694.2877(5) 0.263 0.174 2.13 0.064 2.6 8 213.1226(1) 85.1615(1) 52.1271(1)

C1

C2

a

c

13

13

3673.5124(7) 1888.0973(7) 1682.0220(4) 0.263 0.174 2.13 0.064 3.9 8 215.2756(1) 85.1836(1) 52.3902(1)

3662.7815(7) 1867.6371(10) 1660.6872(5) 0.263 0.174 2.13 0.064 3.7 9 218.4702(1) 85.8490(1) 52.1279(1)

C4

C5

Centrifugal distortion constants for the isotopic species were fixed at the values obtained for the normal isotopologue. P Dmrms ¼ ½ ðmobs  mcalc Þ2 =N1=2 . N is the number of fitted transitions. P P aa ¼ 0:5ðIb þ Ic  Ia Þ ¼ i mi a2i , etc.

Table 4 Spectroscopic parameters for the anti–anti-conformer of 3,3-difluoropentane.

b

C3

Spectroscopic parameter

Parent

A (MHz) B (MHz) C (MHz) DJ (kHz) DJK (kHz) Dmrms (kHz)a Nb Paa (uÅ2)c Pbb (uÅ2) Pcc (uÅ2)

4819.3323(10) 1667.2385(7) 1661.1272(6) 0.134(17) 1.71(6) 4.1 14 251.2486(1) 52.9901(1) 51.8749(1)

P Dmrms ¼ ½ ðmobs  mcalc Þ2 =N1=2 . N is the number of fitted transitions. P P aa ¼ 0:5ðIb þ Ic  Ia Þ ¼ i mi a2i , etc.

the fit as the splittings were still not sufficiently resolved to assign them separately. Despite broader lineshapes, and a resolution of 80 kHz resulting from the FFT parameters used on the oscilloscope, comparisons of frequency measurements between the CPFTMW and resonant cavity instrument agreed to well within 6 kHz and the precision of the spectral fits (Tables 2–4) was also comparable to that typically obtained using the Balle–Flygare instrument. Subsequent studies using this CP-FTMW instrument have clearly resolved Br and 14N nuclear quadrupole hyperfine structure in 1-bromobutane, 2-bromobutane [17] and difluorosilylisocyanate [18].

221←110 (anti-anti) 16119.0817 MHz

116

220←111 (anti-anti) 16125.2023 MHz

120 124 Offset from center frequency / MHz

128

Fig. 3. Two transitions for the anti–anti-conformer of 3,3-difluoropentane showing internal rotation fine splitting. The horizontal frequency scale shows an offset from the center frequency (16 000.002 MHz). This spectrum is an average of 250 gas pulses and was recorded by mixing a chirped-pulse spanning a narrower frequency range of 50–150 MHz (to improve signal intensity) with the center frequency output from a microwave synthesizer. The transition on the left has a maximum intensity of about 1.9 mV.

3.2. Dipole moments A conventional Balle–Flygare Fourier-transform microwave spectrometer was used for the Stark effect measurements and also for checking the frequency precision of transitions measured on

39

D.A. Obenchain et al. / Journal of Molecular Spectroscopy 261 (2010) 35–40

the chirped-pulse instrument; the former instrument has been described in detail before [12]. Field strengths up to 330 V cm1 were used leading to Stark shifts of up to 1 MHz. Measured Stark coefficients were least-squares fitted to those calculated using secondorder perturbation theory [19] using rotational constants for the parent isotopic species listed in Tables 2 and 3. The resulting dipole moment components are given in Table 5 for both the gg and ag conformers. As discussed earlier, lower intensity of the aa transitions precluded determination of any dipole moment information for that conformer.

Table 6 Kraitchman coordinates (in Å) for the gauche–gauche conformer of 3,3-difluoropentane and a comparison with ab initio principal axis coordinates. Kraitchman

C1, C2, C3

5 4

Ab initio

|a|

|b|

|c|

a

b

c

1.8565(8) 1.0108(15) 0.116(13)i

1.3379(11) 0.3754(40) 0.3945(40)

0.060(25)i 0.8091(19) 0.079(19)i

±1.8295 ±1.0012 0.0000

1.3439 0.3754 0.4050

0:0218 ±0.8150 0.0000

Table 7 Kraitchman coordinates for the anti-gauche conformer of 3,3-difluoropentane and a comparison with the principal axis ab initio coordinates.

3.3. Heavy atom structures of the gauche–gauche and anti-gauche conformers Heavy atom structures for both the gg and ag species were available from two sources, namely via determination of Kraitchman single substitution coordinates [20] for all carbon atoms and also via least-squares fitting of measured moments of inertia for the parent and 13C species to selected structural parameters (using the STRFITQ program [21]). Note that during this least-squares fitting procedure, all parameters relating to fluorine and hydrogen atoms were held fixed at the ab initio optimized values, with only bond distances, angles and dihedrals involving the carbon backbone being allowed to vary. Tables 6 and 7 list principal axis coordinates resulting from application of Kraitchman’s equations using Kisiel’s KRA program [22] for the gg and ag conformers, respectively. It should be noted that several of these coordinates are imaginary, particularly in the case of the anti-gauche species which ab initio calculations predict has at least one small coordinate (<0.1 Å) for three out of five of its carbon atoms (ab initio principal axis coordinates are also listed in Tables 6 and 7). Given the small coordinates of several carbon atoms, the Kraitchman calculation is not particularly instructive for determining bond lengths and angles in the present case although the structure determined from these coordinates has been included in Table 8 (specific conditions for treatment of the imaginary coordinates are outlined in the footnote to that table). Structures of the gg and ag conformers obtained from leastsquares fits are in reasonable agreement with ab initio structural parameters (Table 8). C–C distances around the central carbon atom (C2–C3 and C3–C4) in both conformers are consistently shorter than the terminal C–C distances (C1–C2, C4–C5, a trend reflected in the ab initio values) and agree closely between conformers. Dihedral angles are reproduced to within a degree and most bond

Table 5 Observed and calculated Stark coefficientsa and derived dipole moment components for the gauche–gauche and anti-gauche conformers of 3,3-difluoropentane. Gauche–gauche 221 212 313

110 |M| = 0 101 |M| = 0 202 |M| = 1

Anti-gauche 221 212 221 312 312 404

a

110 101 111 211 211 313

|M| = 0 |M| = 0 |M| = 0 |M| = 0 |M| = 1 |M| = 1

105  (Dm/E2) obs.

105  (Dm/E2) calc.

3.5923 4.4507 3.9424 lb = ltotal = 2.4186(40) D

3.5643 4.4641 3.9526

105  (Dm/E2) obs.

105  (Dm/E2) calc.

0.5533 1.4800 1.3988 0.3510 1.8080 2.4187 la = 0.8933(29) D lb = 1.7286(20) D lc = 1.1826(19) D ltotal = 2.2769(21) D

0.5632 1.4795 1.3962 0.3460 1.8029 2.4227

Stark coefficients are in units of MHz V2 cm2.

Kraitchman

C1 C2 C3 C4 C5

Ab initio

|a|

|b|

|c|

a

b

c

2.2529(7) 0.7959(19) 0.119(13)i 1.4652(10) 2.3095(7)

0.8612(17) 0.9362(16) 0.2418(62) 0.2853(53) 0.8839(17)

0.038(40)i 0.4884(31) 0.048(31)i 0.5163(29) 0.040(38)i

2.2398 0.7881 0.0273 1.4641 2.3008

0.8643 0.9387 0.2532 0.3017 0.8788

0.0255 0.4870 0.0481 0.5079 0.0268

Table 8 r0 structural parameters obtained from least-squares fits of the experimental moments of inertia for the gauche–gauche and anti-gauche conformers of 3,3difluoropentane. See Fig. 1 for atom numbering scheme. Parameter

Inertial fita

Kraitchmanb

Ab initio

Gauche–gauche R(C1–C2) (Å) R(C2–C3) (Å) \(C1–C2–C3) (°) \(C2–C3–C4) (°) s(C1–C2–C3–C4) (°)

1.5202(60) 1.5141(33) 114.20(21) 117.40(42) 57.9(3)

1.515(10) 1.506(10) 114.4(10) 118.5(10) 57.2(10)

1.525 1.509 113.1 117.7 56.7

Anti-gauche R(C1–C2) (Å) R(C2–C3) (Å) R(C3–C4) (Å) R(C4–C5) (Å) \(C1–C2–C3) (°) \(C2–C3–C4) (°) \(C3–C4–C5) (°) s1(C1–C2–C3–C4) (°) s2(C2–C3–C4–C5) (°)

1.5267(53) 1.5165(25) 1.5151(21) 1.5300(62) 113.17(31) 117.03(25) 113.96(34) 177.68(17) 60.77(46)

1.539(21) 1.503(52) 1.554(14) 1.532(44) 111.1(36) 114.4(38) 112.7(32) 176.7(10) 64.9(30)

1.525 1.509 1.509 1.525 112.8 116.8 113.7 177.4 60.4

a The listed parameters refer to a ‘‘best fit” structure obtained from fitting all possible permutations of Ia, Ib and Ic for the isotopic species. The standard deviations of these fits are as follows: gauche–gauche = 0.0112 uÅ2 and antigauche = 0.0078 uÅ2. b Structural parameters obtained from the principal axis coordinates derived from the use of Kraitchman’s equations. For this column, any imaginary coordinates were set to zero before calculation of structural parameters and hence the resulting parameters agree with the inertial fit and ab initio values much more poorly than is typical, hence larger uncertainties are estimated for these parameters.

angles are also well within a degree (more typically within a few tenths of a degree) of the ab initio values.

4. Discussion Although transitions belonging to the gg conformer were considerably more intense than those from the aa or ag conformers, inconsistent linewidths between transitions of different conformers, along with an observed variation in signal intensity as a function of center frequency, has frustrated (at present) any attempts to determine relative abundances for these conformers in our gas expansion. However, it is apparent that the three lowest energy conformers are similar enough energetically to provide sufficient population in the supersonic expansion to allow observation of all three spectra in reasonable intensity.

40

D.A. Obenchain et al. / Journal of Molecular Spectroscopy 261 (2010) 35–40

MP2 calculations predicted all three conformers to lie within 100 cm–1 of each other (Table 1), and with the same order of stabilities as was determined for the silicon analog of this molecule, diethyldifluorosilane [6]. Calculation of relative energies of the four conformers in the silane (albeit with a slightly smaller basis set, at the MP2/6-311+G(2d,2p) level) resulted in relative energies of aa (76 cm–1), ag (81 cm–1), gg0 (299 cm–1) with the gg defined as 0 cm–1 [6]. Although still the least stable conformer in both cases, the gg0 is significantly less stable relative to the gg conformer in difluoropentane (914 cm–1, Table 1) than in diethyldifluorosilane (299 cm–1, Ref. [6]), presumably because methyl group crowding is quite considerably reduced in the latter compound due to longer C–Si bonds. Indeed, an inspection of ab initio optimized structures reveals the methyl group carbons in 3,3-difluoropentane to be separated by 3.30 Å, compared to 3.73 Å in the silane. Interestingly, an independent density functional theoretical calculation of the conformer energies in 3,3-difluoropentane, carried out using Gaussian 03 at the College of Charleston at the B3LYP/ 6-311++G(3df,3pd) level reveals a different order of stabilities, namely aa > ag > gg > gg0 (from most to least stable), seemingly more in line with the results observed for n-pentane, where spectra for only the aa and ag species were observed [5]. The B3LYP relative energies (aa = 0 cm–1, ag = 112 cm–1, gg = 185 cm–1 and gg0 = 901 cm–1) are of similar magnitudes to results from our MP2 calculations, again with gg0 significantly higher in energy than the other conformers, but with the gg conformer significantly destabilized relative to the aa and ag. Clearly an experiment which can reliably determine the populations of different conformers in a supersonic expansion would provide a valuable benchmark for computational calculations, where the relative stabilities can vary widely with different levels of theory. IR and Raman studies of this molecule are currently underway [23] and should provide some insight on the question of relative stabilities. Additionally, systematic computational studies using triple zeta or better quality basis sets would be useful to further explore these stability trends; these are beyond the scope of the current study. Comparison of dipole moment components predicted by ab initio calculation (Table 1) with those obtained from experiment for the gg and ag conformers (Table 5) indicates a relatively good agreement, although all theoretical values are overestimated by a few percent. Comparison of measured dipole moments of 3,3-difluoropentane with diethyldifluorosilane [6] reveals a very similar charge distribution. In the gg conformer the silane analog possessed an approximately 10% smaller dipole moment (2.184(17) D versus 2.4186 (40) D in the present study) while the total dipole moment of the ag conformer of 3,3-difluoropentane (ltotal = 2.2769(21) D) was almost identical with that of the same conformer for the silane (ltotal = 2.2445(33) D; la = 0.790(6) D, lb = 1.8151(25) D, lc = 1.0579(28) D) [6]. Unfortunately, for diethyldifluorosilane, only rotational spectra for the normal isotopic species were measured and so further structural comparison is not possible. 5. Conclusions A newly constructed chirped-pulse Fourier-transform microwave spectrometer with up to 480 MHz bandwidth has been used to observe rotational spectra of three conformers of 3,3-difluoropentane. Two of these spectra were intense enough that it was possible to obtain spectra for single 13C substitutions of each carbon position thereby allowing determination of structural parameters for the carbon atom backbone. Some transitions of the anti–anti conformer spectrum were additionally broadened, presumably from methyl group rotations, preventing assignment of additional isotopic species for this conformer.

Acknowledgments The authors would like to extend their most sincere thanks to Prof. Brooks Pate, Justin Neill and the rest of the Pate group at the University of Virginia for being such gracious hosts during our visits to their lab, for their support and advice on design and construction as well as help in troubleshooting our instrument. The authors also graciously acknowledge assistance from Prof. Steve Cooke and Smitty Grubbs at the University of North Texas on numerous aspects of CP-FTMW spectrometers. This work was supported by the National Science Foundation under CHE0809387. Appendix A. Supplementary data Supplementary data for this article are available on ScienceDirect (www.sciencedirect.com) and as part of the Ohio State University Molecular Spectroscopy Archives (http://library.osu.edu/sites/ msa/jmsa_hp.htm). Supplementary data associated with this article can be found, in the online version, at doi:10.1016/ j.jms.2010.03.002. References [1] G.G. Brown, B.C. Dian, K.O. Douglass, S.M. Geyer, S.T. Shipman, B.H. Pate, Rev. Sci. Instrum. 79 (2008) 053103. [2] G.G. Brown, B.C. Dian, K.O. Douglass, S.M. Geyer, B.H. Pate, J. Mol. Spectrosc. 238 (2006) 200. [3] G.S. Grubbs, C.T. Dewberry, K.C. Etchison, K.E. Kerr, S.A. Cooke, Rev. Sci. Instrum. 78 (2007) 096106. [4] T.J. Balle, W.H. Flygare, Rev. Sci. Instrum. 52 (1981) 33. [5] R.K. Bohn, G.B. Churchill, J. Phys. Chem. A 111 (2007) 3513. [6] S.A. Peebles, M.M. Serafin, R.A. Peebles, G.A. Guirgis, H.D. Stidham, J. Phys. Chem. A 113 (2009) 3137. [7] A.L. Steber, D.A. Obenchain, R.A. Peebles, S.A. Peebles, J.L. Neill, B.H. Pate (2008), unpublished observations. [8] A.L. Steber, R.A. Peebles, S.A. Peebles, J.L. Neill, B.H. Pate (2009), unpublished observations. [9] J.J Oh, D.A. Obenchain, A.A. Elliott, R.A. Peebles, S.A. Peebles (2009), unpublished observations. [10] A.J. Laplante, H.D. Stidham, S.A. Peebles, R.A. Peebles, C. Wurrey, G.A. Guirgis, J. Mol. Struct., manuscript in preparation. [11] LabView version 7.1, National Instruments Corporation, 2004. [12] J.J. Newby, M.M. Serafin, R.A. Peebles, S.A. Peebles, Phys. Chem. Chem. Phys. 7 (2005) 487. [13] J.M.L.J. Reinartz, A. Dymanus, Chem. Phys. Lett. 24 (1974) 346. [14] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.P. Hratchian, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C. Gonzalez, J.A. Pople, Gaussian 03, Revision D.01, Gaussian, Inc., Wallingford, CT, 2004. [15] J.K.G. Watson, Vib. Spectra Struct. 6 (1977) 1. [16] H.M. Pickett, J. Mol. Spectrosc. 148 (1991) 371. [17] J.J. Oh, D.A. Obenchain, S.A. Peebles, R.A. Peebles, unpublished observations. [18] G.A. Guirgis, M.H. Palmer, D.A. Obenchain, R.A. Peebles, S.A. Peebles, manuscript in preparation. [19] R.A. Beaudet, Second-order Stark coefficients were computed with use of the University of Michigan modified version of the original ASYSPEC code, Ph.D. Thesis, Harvard University, 1961. [20] J. Kraitchman, Am. J. Phys. 21 (1953) 17. [21] R.H. Schwendeman, in: D.R. Lide, M.A. Paul (Eds.), Critical Evaluation of Chemical and Physical Structural Information, National Academy of Sciences, Washington, DC, 1974 (The STRFITQ program used in this work is the University of Michigan modified version of Schwendeman’s original code.). [22] Z. Kisiel, Kraitchman coordinates and propagated errors in parameters calculated using the KRA and EVAL code, PROSPE – Programs for Rotational Spectroscopy; . [23] G.A. Guirgis, S.A. Peebles, R.A. Peebles, work in progress.