Part (i) MICROWAVE SPECTROSCOPY by J. E. Parkin (Department of Chemistry University College London) SINCE the last Annual Report by Shcridan.‘ summaries by Gordy2 on milli-metre wave spectroscopy and by Sheridan3 on microwave spectroscopy have appeared. More recently Flygare4 has published a comprehensive review covering January 1964-1967. A book by Sugden and Kenney’ contains a reference list of work molecule by molecule complete up to 1963. Instrumental advances have been discussed by Gordy2 and She~idan.~ The increasing range of sources and detectors now affords considerable overlap with the i.r. region. Greater sensitivity and stability of detectors including the development of commercial ‘spectrum accumulators’ by which a weak signal is built up with time thus increasing the signal to noise ratio have enabled molecules with very small dipole moments to be studied.The microwave absorption for HC=CD observed by Muenter and Laurie6 with a dipole moment ca. 0.01~ is ample testimony to this. Double-resonance techniques and improved computational procedures have also enabled significant advances to be made particularly in the field of larger molecules. Finally the development of a commercial instrument by the Hewlett-Packard Company,’ with several novel features suggests that the day may be near when microwave spectroscopy becomes a routine chemical research tool. The great sensitivity and selectivity of the technique gives it many advantages over more con-ventional spectroscopic methods for analytical applications as well as for purely spectroscopic investigations.Structure Determinations.-Microwave spectroscopy when applicable, continues to be the most reliable method for the determination of molecular structure and such molecular parameters as dipole moments. quadrupole coupling constants and barriers to internal rotation. During 1967 about eighty papers have appeared dealing in greater or lesser detail with molecules ranging from diatomic free radicals to one molecule with eighteen atoms. Very precise information can be obtained for diatomic molecules including bond lengths accurate to 0.0001 8 or better. It still remains true however that for very few polyatomic molecules have the equilibrium geometrical para-’ J. Sheridan Ann. Reports 1963 160. W. Gordy Pure Appl. Chem. 1965,11,403 J.Sheridan Pure Appl. Chem. 1965,11,455. W. H. Flygare Ann. Rev. Phys. Chem. 1967,18,325. T. M. Sugden and C. N. Kenney ‘Microwave Spectroscopy of Gases’ van Nostrand London, 1965. ‘ J. S. Muenter and V. W. Laurie J . Amer. Chem. SOC. 1964,86,3901. ’ Hewlett-Packard Company. Palo Alto California U.S.A 182 J . E. Parkin meters the so-called re-structure been determined. To do this involves measurement of the rotational constants for several isotopic modifications of the molecule in all the excited vibrational fundamentals. More usually, approximate ro or rs structures are determined. Costain' has discussed in-formally the problems relating to the accuracy of these structures. The critical use of data from electron diffraction and elsewhere will become more important especially when large molecules are considered.Dallinger and Toneman*= report an electron-diffraction study of 1,3-cyclohexadiene in which the micro-wave rotational constants are used as constraints in the structure determination. In dealing with small molecules especially those containing hydrogen, centrifugal-distortion effects become important. Two important papers by Watsong have appeared in which he considers the general problem of deter-mination of centrifugal-distortion constants from rotational energy levels, and the problem of indeterminacy of the constants obtained. He shows in particular that only five quartic distortion constants zapYs are linearly inde-pendent for non-planar molecules rather than six as previously assumed. A few recent analyses have been re-examined" in the light of this finding with significant improvement in the results.No doubt others will follow. Small Molecules and Radicals. Interest in these molecules traditionally the field of the electronic spectroscopist is becoming increasingly apparent see F l ~ g a r e . ~ The S - 0 radical is much studied and Amano Hirota and Morino,' from measurements on the first vibrationally-excited state of 32S-1 60 have determined the equilibrium bond length re to be 1.48108 & 0.00005 A. Recent investigations of Sn0,12 InCl,I3 and GeTe14 (in the latter 19 different isotopic modifications were measured separately) have provided data of similar accuracy. The spectra of KOH and CsOH vapours have been measured by Kuczkowski, Lide and Krisher." The K-0 and Cs-0 bond distances were determined as 2.18 and 2.40 A and the dipole moment of CsOH as 7.1 rf O ~ D favouring a highly ionic bond.A peculiar variation in the rotational constant B with the bending vibration (a ca. 300 cm.-') remained unaccounted for although quasi-linearity (a small potential hump at the linear configuration) and the very large amplitude of the vibration were considered. The combination of microwave data (centrifugal distortion and 1-type doubling constants) and i.r. data (vibrational frequencies and Coriolis coupling constants) in order to provide a reliable force field for the molecule is especially favourable in triatomic molecules where the number of parameters is still C. C. Costain Trans. Amer. Cryst. Assn. 1966 2 157.J. K. G. Watson J . Chem. Phys. 1966,45 1360; 1967,46 1935. G. Dallinger and L. H. Toneman J . Mol. Structure 1967 1 11. l o H. Dreizler 2. Naturforsch. 1966,21a 1719; H. 0. Sorensen J . Mol. Spectroscopy 1967,22,325. l 1 T . Amano E. Hirota and Y. Morino J . Phys. SOC. Japan 1967,22,399. l2 T . Torring 2. Naturforsch. 1967 224 1234. l 3 G. A. L. Delvigne and H. W. de Wijn J . Chem. Phys. 1966,45,3318. l4 J. Hoeft and H. P. Nolting Z . Naturforsch. 1967 224 1121. l 5 R. L. Kuczkowski D. R. Lide jun. and L. C. Krisher J . Chem. Phys. 1966 44 3131; D. R. Lide jun. and R. L. Kuczkowski ibid. 1967,46,4768 Part (i) Microwave Spectr.oscopy 183 small. Work of this kind has been reported on HCN,16 OCSe,17 NSF,18 FCN,I9 and NF, PF, and AsF,.~' Morino and Matsumura2' have deter-mined a new ?-,-structure for o c s rc- = 1.157 A and rcs= 1-5606 8 using measurements on various vibrationally-excited states.Measurements of the 1-type doubling constants enable the Coriolis coupling constants to be deter-mined and a preliminary calculation of the harmonic potential field is given. The variation of these constants in higher vibrational levels will provide data sufficient to determine many of the anharmonic potential constants and work is proceeding along these lines. Other Molecules. Structural and other information for some interesting molecules is summarised in the Table. A number of complete or near-complete structural determinations have been reported. Accurate bond lengths and angles for molecules containing silicon,26-28* 38 germanium,32.and selen-ium23,42 might be noted and some of these papers give interesting correlations with analogous compounds containing carbon and sulphur. Several interesting small-ring compounds4s~ 46 " 9 s 3 9 54 have had their conformation confirmed on the basis of their rotational constants. Internal rotation is a fruitful study with microwave spectroscopy. Molecules can be studied in several excited torsional states from which the torsional frequency and torsional barrier can be determined quite accurately. The methyl torsion in toluenes9 and p-chloro-toluene58 has a six-fold potential and is found to have a barrier of 13.94 cal. mole- '. Flygare4 summarises a large number of methyl torsional barriers and a large amount of theoretical work is currently being carried out on their origin.Kuzckowski and Lide39 report the structure of the interesting molecule PF,BH,. The molecule has a high barrier 3.24 kcal.mole-' and they were able to estimate the dissociation enthalpy of the molecule (2PF,BH3 + 2PF3 + B2H6) as 10.99 kcal. Investigation of the excited states of nitrosobenzeness indicates a barrier of 3.9 k.cal.mole- and the molecule is planar at equilibrium. Thiophenols6 is also planar and demonstrates no torsional splitting. One of the most promising structural determinations recently is the work of Pierce and Beecher61 and Pierce and Nelson62 on the two conformational isomers of cyclohexyl fluoride. This appears to be the first reported microwave study of the cyclohexane ring system. They find the axial fluoride isomer to be more stable than the equatorial fluoride by 400 300 cal.mole-'.The ring geometry is assumed and bond lengths adjusted slightly to fit the observed rotational constants. With such large molecular systems as this there is little likelihood of complete structural determinations but many gross conformational problems of interest to the organic chemist are waiting to be solved. l 6 A. G. Maki and D. R. Lide,jun. J . Chem. Phys. 1967,47,3206. Y. Morino and C . Matsumura h l l . Chem. SOC. Japan 1967,40 1101. '* A. M. Mirri and A. Guarnieri Spectrochim. Acta. 1967,23A 2159. l9 W. J. Lafferty and D. R. Lide jun. J . Mol. Spectroscopy 1967 23,94. *' A. M. Mirri J . Chem. Phys. 1967,47 2823. Y . Morino and C. Matsumura Bull. Chem. SOC. Japan 1967,40,1095.22 M. Lichtenstein V. E. Derr and J. J. Gallagher J . Mol. Spectroscopy 1966,20 381 TABLE. Some Recent Microwave Investigations. Bond lengths rxy are in A; dipole moment p in Debyes; barrier to internal rotation coupling constants determined ; c.d. = centrifugal distortion Molecule Reference Comments H2O SeOF, HCNO HNCO trans-€€NO2 HSiC13 SiF31 SiFC1, CH3F CD3F CHSCN CD3CN CH3Br CD3Br CH31 GeH3CN CF3CH0 CH,FCOF 1,2,5 selenadiazole CH,SiC13 GeH3SiH3 CH3NF2 CF3CHF2 22 23 24 2 4a 2s 26 2 1 28 29 29 30 31 32 33 34 3 5 36 31 26 38 p = 1.884 & 0012 new transitions. rSeO = 1.580 rSeF = 1.727 angles; character. rCH = 1.027 rCN = 1.161 rNo = 1.207, Dipole-moment variation with K. rNo(H) = 1.433 rNO = 1.177 r o H = 0.954, rSiH = 1.4655 rSiCl = 2-0118 p = 0.86, rsI = 2.387 q.c.comparson with other rSiF = 1.520 rSiCl = 2.019 p = 0 4 9 p = 1.8472 1-8682 precise measurement, p = 3.913 3.919 precise measurement, excited states Coriolis interactions. q.c. excited states Fermi resonance. complete structure p = 2-57 V = 4.170, p = 1.65 V = 0.885 double resonance brans = 2.67 pcis = 2-05 V = 19-1.3, p = 1-11 q.c. planar. rCF = 1-335 I, = 1.520 V = 3.51, rGeSi = 2-357 - p = 0.1 c.d. could not TCN = 1.155 rCeC 1*919,.~ = -3.99, p = 1-91 I/ = 0. PF3BH3 CHCS CH3 CH2CHN02 CH SeCH, pyrazole 1-methylcyclopropene bicyclo[ 1,1,0] butane pyridazine 1,3,2-dioxaborolan furfural CH3CH20CH0 3,4-dimethylenecyclo bu tene 1,2 and 1,3 difluorobenzene trans-2,3-epoxybutane spiropentyl chloride nitrosobenzene thiophenol 4-methyl pyridine p-chlorotoluene toluene (CH3)3CCH0 cyclohexyl fluoride CH3CHCHZOH 39 40 4 1 42 4 3 44 4 5 46 47 48 49 5 0 5 1 5 2 5 3 5 4 5 5 5 6 5 1 58 5 9 60 61 rBH = 1.207 rpp = 1.538 rpB = 1.836, rCMeS = 1.815 rcs = 1.680 p = 1.69, p = 3-70 V = 0.3 (nitro) q.c.planar. rCSe = 1-943 rCH = 1.093 p = 1.41, p = 2.214 planar with no symmetry p = 1-55 gauche configuration. structure determined V = 1-39. structure determined p = 0.675. complete structure p = 2.28 q.c. c. complex potential fuction ECis - E,,,, structure p = 0.618 planar. structure p = 2.03 V = 2.444. rotational constants V = 3.9 excited rSH = 1-30 rcs = 1-77 planar no torsional r" = 1.330 rcsc5 = 1.375 p = 4.22, brans = 1.98 pgau&e = 1.81 V = 1.10, p1,2 = 2.59 ~ 1 3 = 1.51 c.d.rCCl = 1.74 rCC = 1.51,q.C. p = 2.70 V6 = 13.51 cal q.c. rCcl = 1.74 v6 = 13.93 cal 9.c. P = 0.374 V = 13.94 cal. p = 2.66 VBut= 1.186 VMe = 2.6 3.5, axial rCF = 1.399 rcc = 1.526 rC 186 J . E. Parkin Analysis of the Stark splittings of favourable lines in the microwave spectrum provides the most accurate method available for dipole-moment measurement. Several erroneous dipole moments still appear in the literature based on solution studies when precise values are available. Analysis of hyperfine structure in the microwave spectrum enables nuclear quadrupole coupling constants to be measured fairly accurately.Here again a large amount of data is accumulating rapidly. Microwave Double-Resonance and Rotational Relaxation.-Double-resonance techniques are proving very useful in several aspects of microwave spectroscopy. By saturating a given transition with microwave power a non-equilibrium population distribution is set up in the two levels involved, and an increase or decrease in intensity in other transitions involving one of 23 I. C. Bowater R. D. Brown and F. R. Burden J. Mol. Spectroscopy 1967,23 272. 24 M. Winnewisser and H. K. Bodenseh Z. Naturforsch. 1967,22a 1724. 24(Q) K. J. White and R. L. Cook J. Chem. Phys. 1967,46 143. *’ A. P. Cox and R. L. Kuczkowski J. Amer. Chem. SOC. 1966,88,5071. 26 M. Mitzlaff R. Holm and H. Hartmann Z. Naturforsch. 1967,22a 1415.27 L. C. Sams jun. and A. W. Jache J. Chem. Phys. 1967,47 1314. R. Holm M. Mitzlaff and H. Hartmann Z. Naturforsch. 1967,22a 1287. 29 P. A. Steiner and W. Gordy J. Mol. Spectroscopy 1966,21,291. 30 Y. Morino and C. Hirose J. Mol. Spectroscopy 1967,24 204. 31 Y. Morino and C. Hirose J. Mol. Spectroscopy 1967 22,99. 32 R. Varma and K. S. Buckton J. Chem. Phys. 1967,46 1565. 33 L. Pierce R. G. Hayes and J. F. Beecher J. Chem. Phys. 1967,46,4352. 34 R. C. Woods J. Chem. Phys. 1967,46,4789. 35 E. Saegebarth and E. B. Wilson jun. J. Chem. Phys. 1967,46,3088. 36 G. L. Blackmann R. D. Brown and F. R. Burden Chem. Phys. Letters 1967,1 379. 37 A. B. Tipton C. A. Britt and J. E. Boggs J. Chem. Phys. 1967,46 1606. 38 A. P. Cox and R. Varma J. Chem. Phys. 1967,46,2007. j9 R.L. Kuscowski and D. R. Lide jun. J. Chem. Phys. 1967,46,357. 40 D. den Engelsen J. Mol. Spectroscopy 1967,22,426. 41 H. D. Hess A. Bauder and H. H. Gunthard J. Mol. Spectroscopy 1967 22 208. 42 J. F. Beecher J. Mol. Spectroscopy 1966 21,414. 43 W. H. Kirchhoff J. Amer. Chem. SOC. 1967,89 1312. 44 A. N. Murity and R. F. Curl jun. J. Chem. Phys. 1967,46,4176. ” M. K. Kemp and W. H. Flygare J. Amer. Chem. SOC. 1967,89,3925. 46 M. D. Harmony and K. Cox J. Amer. Chem. SOC. 1966,88,5049. 47 W. Werner H. Dreizler and H. D. Rudolph Z. Naturforsch. 1967 22a 531. 48 J. H. Hand and R. H. Schwendeman J. Chem. Phys. 1966,45,3349. 49 F. Monnig H. Dreizler and H. D. Rudolph Z. Naturforsch. 1966 21a 1633. ’ O J. M. Roveros and E. B. Wilson jun. J. Chem. Phys. 1967,46,4605.” L. Nygaard E. R. Hansen R. L. Hansen J. Rastrup-Andersen and G. 0. Sorensen Spectrochim. 53 M. R. Emptage J. Chem. Phys. 1967,47 1293. 54 L. M. Woerner and M. D. Harmony J. Chem. Phys. 1966,45,2339. ” Y. Hanyu C. 0. Britt and J. E. Boggs J. Chem. Phys. 1966,45,4725. 56 K. I. Johansson H. Oldeberg and H. Selen Arkiv Fysik 1967 33 313. ” H. D. Rudolph H. Dreizler and H. Seiler Z. Naturforsch. 1967 22a 1738. 58 G. E. Herberich Z. Naturforsch. 1967 22a 761. 59 H. D. Rudolph H. Dreizler A. Jaeschke and P. Wendling Z. Naturforsch. 1967 22a 940. 6o A. M. Ronn and R. C. Woods J. Chem. Phys. 1966,45,3831. 61 L. Pierce and J. F. Beecher J. Amer. Chem. SOC. 1966,88,5406. R. D. Brown F. R. Burden A. J. Jones and J. E. Kent Chem. Comm. 1967,808. Acta. 1967,23A 2813 Part (i) Microwave Spectroscopy 187 the levels is observed.Spectrometers have been d e ~ c r i b e d ~ ~ - ~ ~ incorporating this principle. Unland Weiss and F l ~ g a r e ~ ~ and Cox Flynn and Wilson64 have discussed the application of the technique to problems of assignment and identification. As an example Woods34 in an investigation on fluoral has described a successful search for weak and dispersed R-branch transitions which were connected with known Q-branch transitions. Oka68 has used the technique to demonstrate a ‘forbidden’ AJ = +3 transition 303 t Ooo in ethyl iodide. On pumping at the corresponding frequency with intense microwave power a signal line 404 +- 303 was observed to increase in intensity by some 10%. The weak AJ = 3 transition is allowed by the large iodine nuclear quadrupole.In some even more interesting experiments Oka679 69 has investigated collision-induced transitions in a series of molecules. For ethylene oxide for instance by pumping strongly the 2, t 2, transitions and monitoring the transitions 3, + 31 and 3, t 321 he showed that rotational energy was transferred by collision from the J = 2 levels to the J = 3 levels according to quite strict electric-dipole selection rules. The change in energy was larger at increased pressure due no doubt to the minimisation of wall effects. Similar observations were made with H,CO HCN and H,CCO. In some cases, quadrupole selection rules leading to AJ = 2 transitions appear to be import-ant. In addition he has obtained quantitative rate constants for the various collision processes and demonstrates for instance that for AJ = 1 collision-induced transitions the rate constants are the same order as those for AJ = 0 transitions even though the former involve much more widely-spaced energy levels.He considers this as evidence in support of Anders~n’s’~ theory of rotational resonance proposed to account for pressure broadening in the ammonia spectrum. Unland and Flygare66 have described double-resonance experiments on OCS in which they measured directly the relaxation time between two states. They found that the relaxation times obtained were half as long as those obtained from line-width data also indicating preferred collisional-exchange selection rules. They discuss possible extensions of relaxation experiments in the study of transient species gaseous chemical kinetics and lifetimes of excited vibrational states.Ronn and Wilson’ have confirmed Oka’s findings and have considered several other molecular systems. Gordon7 has contri-6 2 L. Pierce and R. Nelson J . Amer. Chem. SOC. 1966,88 216. 6 3 M. L. Unland V. W. Weiss and W. H. Flygare J . Chem. Phys. 1965,42 2138. 64 A. R. Cox G. W. Flynn and E. B. Wilson jun. J . Chem. Phys. 1965,42 3094. 6 5 R. C. Woods A. M. Ronn and E. B. Wilson jun. Rev. Sci. Instr. 1966 37 927. 66 M. L. Unland and W. H. Flygare J . Chem. Phys. 1966,452421. 6’ T. Oka J . Chem. Phys. 1967,47 13. 6 8 T. Oka J . Chem. Phys. 1966,45752. 6 9 T. Oka J . Chem. Phys. 1966,45,754. ’O P. W. Anderson Phys. Rev. 1949,76647. 7 1 A. M. Ronn and E. B. Wilson jun. J . Chem. Phys. 1967,46,3262. 7 2 R. G. Gordon J . Chem. Phys. 1967,46,4399 I88 J . E. Parkin buted a theoretical study of inelastic collisions between molecules allowing a more quantitative interpretation of these and related experiments. There is no doubt that this technique will provide much more quantitative information on collision processes than has hitherto been available and further results are awaited with interest