.. «J. qu- 1 w-n<---. ...... STUDIES OF CYCLOPROPYL METHYL KETONE CYANODIFLUOROPHOSPHENE DIFLUOROCYANAMIDE AND BUTYRALDEHYDE BY MOLECULAR ROTATIONAL SPECTROSCOPY Thesis for the Dégree of; ER D MLCHIGAN STATE UNIVERSTTY PETER LA! SUN LEE ' ’ 19T1 LIBR?)1 WY Michigan x. are Univers. / This is to certify that the thesis entitled STUDIES OF CYCLOPROPYL METHYL KETONE, CYANODI FLUOROPHOSPHINE , DI FLUOROCYANAMIDE , AND BUTYRALDEHYDE BY MOLECULAR ROTATIONAL SPB CEDEEEQTT Peter Lai-Sun Lee has been accepted towards fulfillment of the requirements for .Ph.__D_..__ degree in Jhemislly @z/AVLML,LLLE Major professor mew.— \l‘l {It‘ll‘ III . lllllllll l ) ‘ ABSTRACT STUDIES OF CYCLOPROPYL METHYL KETONE, CYANODIFLUOROPHOSPHINE, DIFLUOROCYANAMIDE, AND BUTYRALDEHYDE BY MOLECULAR ROTATIONAL SPECTROSCOPY BY Peter Lai-Sun Lee The applications, theoretical treatments, and experi- mental aspects of microwave rotational spectroscopy are briefly surveyed. Results of investigations of cyclopropyl methyl ketone, cyanodifluorophosphine, difluorocyanamide, and butyraldehyde are presented. Rotational transitions in the ground-vibrational state and three excited states of the acetyl torsional mode of cis-cyclopropyl methyl ketone have been assigned. From an analysis of the ground state spectrum the height of the barrier to internal rotation of the methyl group is deter- mined to be V3 = 1180120 cal/mole, and the dipole moment components are “a = 0.47 D and uh = 2.58 D, so that the total dipole moment pt = 2.62 D. In spite of exhaustive effort no spectrum could be assigned to a second species. The rO and partial rs structures of PFZCN have been determined from analyses of observed moments of inertia of PF CN, PF 13CN, and PF CISN. The bond distances and bond 2 2 2 0 angles for the rO structure are as follows: CN = 1.158 A, Peter Lai-Sun Lee pc = 1.810 A, PF = 1.567 A, PCN = 172.0°, FPF = 99.1°, and CPF = 97.2°. The quadrupole coupling parameters in the CN bond-ax1s system are Xbond = -4.71 MHz, nbond = 1.03, and 0 = 8.5°. The dipole moment components are found to be pa = 2.27 D, uC = 1.27 D, and pt = 2.39 D. The r0 and partial rs structures of NFZCN were similarly determined from observed moments of inertia of NFZCN, NF213CN, NFZCISN, and lsNFZCN. The structural parameters derived from the r analysis are: CN = 1.154 X, 0 NC = 1.391 K, NF = 1.396 A, NCN = 171.2°, CNF = 105.0°, and FNF = 103.20. The dipole moments are “a = 1.03 D, uC = 0.39 D, and “t = 1.10 D. Two rotamers of butyraldehyde, trans-trans and gauche-trans, together with their vibrational satellites from the ethyl torsional mode have been assigned. The dihedral angle 0 for the gauche-trans species is found to be 101:2°. The dipole moments for the trans-trans species are “a = 1.602 D, ub =2.040 D, and u = 2.594 D. For the t gauche-trans species, “a = 0.855 D, “b = 2.233 D, 0C = 0.73 n, and “t = 2.500 D. STUDIES OF CYCLOPROPYL METHYL KETONE, CYANODIFLUOROPHOSPHINE, DIFLUOROCYANAMIDE, AND BUTYRALDEHYDE BY MOLECULAR ROTATIONAL SPECTROSCOPY BY Peter Lai-Sun Lee A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1971 | IT itf’l “lirlltlllllrl ID? T T kftfll‘r'kll‘rlt‘ [. {Bruit-[Ill To Patricia ii ACKNOWLEDGMENTS I wish to thank Professor P. H. Schwendeman for his supervision of this work. Collaboration with Professor K. Cohn in some phase of this work is also appreciated. Financial assistance from the National Science Foundation is acknowledged. iii TABLE OF CONTENTS Page I. INTRODUCTION .................................... 1 II. THEORETICAL TREATMENT ........................... 3 2.1 Moments of Inertia ........................ 3 2.2 Hamiltonian of the Rigid Rotor ............ 4 2.3 Stark Effect .............................. 7 2.4 Determination of Structural Parameters ...... 10 III. EXPERIMENTAI ...................................... 13 3.1 Microwave Spectrometer .................... 13 3.2 Detection, Modulation, and Amplification.. 15 3.3 Frequency Measurement ..................... 17 3.4 Microwave-Microwave Double Resonance ...... 18 3.5 Stark Effect Measurement .................. 20 3.6 Relative Intensity ........................ 20 IV. CYCLOPROPYL METHYL KETONE ....................... 21 4.1 Introduction .............................. 21 4.2 Spectrum .................................. 23 4.3 Barrier to Internal Rotation .............. 33 4.4 Dipole Moment ............................. 37 4.5 Discussion ................................ 39 V. CYANODIFLUOROPHOSPHINE .......................... 42 5.1 Introduction .............................. 42 5.2 Experimental .............................. 42 5.3 Spectra ................................... 43 5.4 Molecular Structure ....................... 49 5.5 Nuclear Quadrupole Coupling Constants ...... 53 5.6 Dipole Moment ............................. 55 5.7 Discussion ................................ 58 iv TABLE OF CONTENTS Page VI. DIFLUOROCYANAMIDE ............................. 60 6.1 Introduction ............................ 60 6.2 Experimental ............................ 61 6.3 Spectra ................................. 61 6.4 Molecular Structure ..................... 65 6.5 Dipole Moment ........................... 73 6.6 Discussion .............................. 75 VII. BUTYRALDEHYDE ................................. 78 7.1 Introduction ............................ 78 7.2 Spectra for trans-trans and gauche-trans Butyraldehyde ........................... 79 7.3 Dipole Moments.... ....................... 87 7.4 Discussion .............................. 92 REFERENCES ......................................... 93 APPENDIX A--STARK EFFECT CALIBRATION OF OCS 98 TABLE 10. 11. 12. LIST OF TABLES Assumed Structural Parameters and Rotational Parameters of Cyclopropyl Methyl Ketone ....... Cartesian Coordinates for cis—Cyclopropyl Methyl Ketone in the Principal Axis System.... A-level Double Resonance Connections for the Ground State of cis-Cyclopropyl Methyl Ketone. Frequencies of Transitions for cis-Cyclopropyl Methyl Ketone in the Ground and First Excited State ......................................... Rotational Parameters for cis-Cyclopropyl Methyl Ketone ................................. Frequencies of A-level Transitions of cis- Cyclopropyl Methyl Ketone in the Second and Third Excited State ........................... Comparison of P Values for Cyclopropane Derivatives....YY ............................. Internal Rotation Parameters for the Ground Torsional State of cis-Cyclopropyl Methyl Ketone ..... - .................................. Barrier to Internal Rotation for Some CH3COX Compounds ..................................... Stark Effect Data for cis-Cyclopropyl Methyl Ketone ........................................ Comparison of Dipole Moments for Some Mole- cules Containing a Carbonyl Group ....... - ..... Frequencies of Ground State Rotational Tran- sitions for Isotopic Species of PFZCN ........ vi Page Z4 25 27 28 30 31 34 36 36 37 38 44 LIST OF TABLES--continued TABLE 13. 14. 15. l6. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. Ground State Rotational Parameters of PF 2CN, PF213CN, and PF2 C‘SN .......................... Frequencies of Transitions in Vibrationally Excited States for Isotopic Species of PFZCN.. Rotational Parameters for the Vibrationally Excited States of PF, CN, PF213CN, and PFzC‘SN. Cartesian Coordinates in the Principal Axis System of PFzCN ............................... Structural Parameters of PFZCN ................ Quadrupole Coupling Constants of PFZCN ........ Frequencies of the Transitions used in the Determination of the Nuclear Quadrupole Coup- ling Constants of PFZCN ....................... Comparison of Observed and Calculated Stark Shift Slopes for PFZCN ........................ Comparison of Observed and Calculated Frequen— cies of Ground State Rotational Transitions for Isotopic Species of NFZCN ................. Observed Rotational Parameters for the Isotopic Species of NFZCN .............................. Frequencies of Observed Transitions of the Excited States of NFZCN ....................... Rotational Parameters for the Excited States of Normal NFZCN..: ............................ Structural Parameters from Analyses of the Isotopic Species of NFZCN ..................... Cartesian Coordinates in the Principal Axis System of NFZCN ............................... Molecular Parameters of NF2 CN and Related Molecules ..................................... vii Page 46 47 48 51 52 52 S4 56 62 64 66 67 68 69 72 LIST OF TABLES--continued TABLE 28. 29. 30. 31. 32. 33. 34. 35. 36. Stark Coefficients and Dipole Moment for NFZCN. Dipole Moments of NFZCN and Related Molecules.. Observed Transitions for trans-trans Butyr- aldehyde ....................................... Observed Rotational Parameters for trans-trans Butyraldehyde .................................. Comparison of AP Values for trans-trans Butyraldehyde anaC Related Molecules .......... Observed Transitions for gauche-trans butyr- aldehyde ....... . ............................... Observed Rotational Parameters for gauche-trans Butyraldehyde .................................. Stark Coefficients and Dipole Moments for Butyraldehyde .................................. Stark Shift for the 0+1 Transition of OCS ...... viii Page 74 74 84 85 86 88 89 91 99 FIGURE 10. 11. 12a. 13. LIST OF FIGURES The Cross-section of an Absorption Cell ........ Block Diagram of a Stark Modulation Spectro- meter .......................................... Schematic of a Three Level System for Double Resonance ...................................... Projection of cis-Cyclopropyl Methyl Ketone in its Plane of Symmetry .......................... Projection of anti-gauche Cyclopropyl Methyl Ketone in the ab Plane ......................... Change of Rotational Constants as a Function of the Vibrational Level ....................... Plots of the ITS-H13 and HS-Hlu Distances versus the Dihedral Angle ............................. Plot of Hs-Hlu Distance versus T ............... The Structure of PFZCN with likely Orientation of the Dipole Moment ........................... Projection of NFZCN in its Symmetry Plane ...... Orientation of the Dipole Moment Vectors for Some NFZX and PFZX Molecules ................... Projections of trans-trans and trans-gauche Butyraldehyde in the ab Plane .................. Projections of gauche—trans and gauche-gauche-l Butyraldehyde in the ab Plane .................. (A—C)/2 vs. K Plot for trans-trans Butyraldehyde. ix Page 14 16 19 22 22 32 4O 4O 57 71 76 80 81 83 LIST OF FIGURES--continued FIGURE Page 14. (A-C)/2 vs. K Plot for gauche-trans Butyr- aldehyde ........................................ 83 15. Determination of the Dihedral Angle for gauche- trans Butyraldehyde from a Comparison between Calculated and Observed Rotational Constants.... 90 16. Stark Shift for the 0+1 Transition of OCS ........ 100 of e1 I‘a TO (0 Sp The Cu 11 f1“ I. INTRODUCTION Gas-phase rotational spectroscopy of polar molecules in the microwave region has been developed within the last two decades. Prior to 1940 there was only one paper reported in the literature; Cleeton and Williams (1) used a semi-optical method to study the inversion spectrum of ammonia. The ex- tensive research in microwave electronics and its application to radar during World War II were instrumental in the early development of microwave rotational spectroscopy. Since then the advances in this field have been both rapid and impressive. Many reviews and books (2-8) on this subject have appeared since 1948. The application of rotational spectroscopy to the study of molecular dynamics is well known and has been discussed elsewhere (9). Briefly, the interaction of electromagnetic radiation (in the 5 GHz to 600 GHz frequency range) with the rotational motion of polar molecules at very low pressures (of the order of 1-100 microns) gives rise to the rotational spectrum, from which a number of types of molecular infor- mation can be determined. These include very accurate mole— cular structures, dipole moments, nuclear quadrupole coup- ling constants, barriers to internal rotation, vibrational frequencies, stereochemical conformations, and energy separations of rotational isomers. The work reported here was undertaken to study some of these molecular properties for the molecules cyclopropyl methyl ketone, cyanodifluorOphosphine, difluorocyanamide, and butyraldehyde. The reasons for studying these molecules and some background information concerning each of them are given in the introductions to Chapters IV, V, VI, and VII, where the results of the present studies are given. II. THEORETICAL TREATMENT 2.1 Moments of Inertia One of the most attractive applications of quantum mechanics is in the area of rotational spectroscopy. Theo- retical treatments of the rigid rotor problem have been given by a number of people (10-13) and can be found in the books referred to earlier (4-8). The moment of inertia of a rigid molecule about any axis through the center of mass of the molecule is defined by I = Z m-r- (2-1) where mi is the mass of the ith atom, and r1 is the distance of the ith atom normal to the axis. The molecular structure can then be described by a second-rank symmetric tensor of the form I I I xx xy xz I = I I I (2-2) YX YY YZ I I I zx zy zz wh e I " Z ( 2 + 2) t (2‘3) er xx 1111 yi zi , e c., = - . -4 Ixy X mixiyi , etc (2 ) The off-diagonal elements of the inertia tensor may be elimi— nated by means of a similarity transformation which is equi- valent to a rotation of the inertial axes. The diagonal ele- ments then become the principal moments of inertia and the axes become the principal inertial axes. Since the trace of the inertial tensor is not changed by a rotation of axes, IXX + Iyy + IZZ = Ia + lb + I , (2—5) where Ia, Ib, and IC are the diagonal elements after the trans- formation. It is conventional to choose the principal moments of inertia such that 1:1 a i I . (2-6) b C To a good approximation a rigid molecule with three principal moments of inertia serves as a relatively successful model for predicting many molecular properties, including those to be reported in this work. A computer program which calculates the rotational parameters from bond distances and bond angles has been written by Schwendeman (14). 2.2 Hamiltonian of the Rigid Rotor The eigenvalue equation for a rigid rotor in a principal axis system is given by HT = ET (2-7) where H is the rotational Hamiltonian operator, T is the wavefunction, and E is the energy. For an asymmetric rotor, 2 2 2 H = Pa/ZIa + Pb/ZIb + PC/ZIC (2-8) with Pa, Pb, P as the components of the total angular mom- C entum along the principal axes a, b, c. In the molecule- fixed (center—of—mass) axis system + Pg + P2, (2-9) The commutation relationships between the above are ( Pb , pc ) = -ihPa ( pc , pa ) = -1npb . (2-10) ( Pa , Pb ) =-1hPC 2 (P.Pq)=o (a=a.b,c). In a representation (10,15-16) which simultaneously diag- onalizes Pa and P2 with eigenvalues hK and h2J(J + 1), the matrix elements of the squares of the angular momentum com- ponents in (2-10) are, 2 2 2 2 (PC).J,K;J,K= (Pb)J,K;J,K= 15h (JU +1) ‘ K ) 2 2 2 (Pc)JK;J,K+2= ”(Pb)JK:J,K+2 = T“ {( J(J *1) ' K(K * 1) )X 0101+ 1) - (K + 11(K + 2) )1‘2 2 2 2 (P3)J,K:.T,K= T1 K (2‘11) (1>2)J,K;J,K = h2J(J + 1). . (2-12) The quantum number J must be a positive integer or zero and K = 0, :1, ..... , tJ. The calculation of the energy levels of an asymmetric rotor has been described in detail by Ray (12) and by King, Hainer, and Cross (13). Ray's approach makes use of an asym- metry parameter,K, defined by K = (2B -A - C)/A - C ; (2-13) K varies from -1 for a prolate symmetric top (A>B=C) to +1 for an oblate symmetric top (A=B>C). The computation of the eigenvalues of H, except for the very low J levels, is a rather involved process. To simplify this process a computer program, EIGVALS, written by Hand and Schwendeman for calcu- lating and fitting the rotational spectrum has been in use at this laboratory for several years. For a symmetric top the off-diagonal elements in the Hamiltonian are zero, and expressions for the energy levels can be given in closed form. For a prolate top it is found that E = hBJ(J + 1) + h(A - B)K2, (2-14) while for an oblate top E = hBJ(J + 1) + h(C - B)K2. , (2-15) Thus the rotational state of a symmetric top is specifed by JK. For an asymmetric top, however, K is no longer a good quantum number and thus is not a useful state label. King, Hainer, and Cross showed that it is possible to label the state of an asymmetric rotor with the pseudo quantum numbers K_1K+1, which are the absolute values of K which would be obtained if the B rotational constant were changed to the prolate and oblate symmetric top limits, respectively. The rotational transitions are classified as Q-branch, R-branch, and P-branch according as AJ=0, AJ=+1, and AJ=-1, respectively. The dipole selection rules for K_1 and K+1 are given below: Axis parallel to AK_1 AK+1 dipole moment a ( least ) even odd b ( intermediate ) odd odd c ( greatest ) odd even 2.3 Stark Effect In the presence of a static electric field the rotational motion of the molecule is perturbed. This perturbation is attributed to the interaction of the molecular dipole with the external electric field. Each energy level splits into a number of sublevels of different energies; the resultant appearance of fine structure in the rotational spectrum is known as the Stark effect. The interaction energy of this perturbation is given by . “E = -J-€ (2-16) where U and 3 are the dipole moment and electric field vectors, respectively. The rotational Stark effect is classified as first order or second order according to whether the change in energy is proportional to the first or second power of the field. A first-order Stark effect requires a degeneracy or near degen- eracy of rotational energy levels and is typically seen for symmetric top molecules, near symmetric top molecules, and for symmetric-top-like energy levels of asymmetric top molecules. For most levels of most asymmetric tops the Stark effect is second order and may be calculated by ordinary second-order perturbation theory. || 131(2): 2 (2-17) A rigorous treatment of the Stark effect for an asymmetric rotor has been presented by Golden and Wilson (18). In this treatment Equation (2-17) is written as .(2)= , 2 2 . 2 2 2 _ E1 ( a Alquq c ) + ( g quuq e )M (2 17a) (0 = a.b.C) where Aiq and Biq are functions of J,e is the electric field strength, and Mh is the projection of the rotational angular momentum in the direction of the electric field. Since M=J, J-l,...., -J, and E£2)depends on M2 , each level is split into M+1 components by the application of a field. The Michigan State University spectrometers are constructed with the microwave electric field parallel to the Stark field. Thus, the selection rule is AM=0, and the frequencies in the presence of the field are given by the relation 0 =vo + V (2-18) where vois the zero-field frequency and 2 2 2 Vs =8 Zq (AAq + M ABq)uq. (2-19) In this equation AAq and ABQ are the differences in Aiq and Biofor the energy levels involved in the transition. It may be shown that for transitions between non-degenerate levels the relative intensities of the Stark components are given by 1 (AJ = 0) cc M2 (2-20) 1 (11.1: 1) «(1+1f- A12. (2-21) J is the lower level involved in the transition. The assignment of a low J transition is often indicated by its characteristic Stark pattern. 10 2.4 Determination of Structural Parameters Structure analysis from spectroscopic data is generally carried out on the basis of a rigid model in which vibrational effects are either ignored or averaged. Significant differences arise when the "effective” structure derived from the observed moments of inertia is compared with the equilibrium structure. These differences are normally attributed to zero-point vib— rations, isotopic effects, and other vibration-rotation inter- actions. The vibrational effects on structure analysis have been treated by Herschbach and Laurie (19-21) and by Merino and Oka (22-24). As pointed out by Herschbach and Laurie, the contribution of zero-point vibrations to the observed moments of inertia can be as much as 1%, which is several orders of magnitude larger than the experimental precision (0.001%). Molecular geometries derived from microwave spectroscopy are commonly referred to as r r0, r or (rz) structures. e’ S’ The re parameters are those obtained from equilibrium moments of inertia, rO parameters are those obtained from ground- state rotational constants, rS parameters are those obtained from Kraitchman's equations (described below) and or rZ parameters are those obtained from average moments of inertia. From one point of view re structures are the most desirable because all isotopic species of a molecule have the same rC structure. However, equilibrium moments of inertia are very difficult to obtain, and consequently have been obtained for 11 only a few small molecules. From another point of view or rZ structures are more desirable because chemical proper- ties are probably more closely related to average parameters. However, or rz structures differ for different isotopic species which makes it difficult to combine moments of inertia for different species. As a result, most structures reported are rO and r structures. 5 In an effort to determine a structure free of vibration- rotation interactions, Kraitchman (25) first proposed the simple moment relationships, which were subsequently used by Costain (26) to develop the substitution method. A modified form of the Kraitchman's equation is given below: P' -P P' - P ,/ 'asl = { u'l ( péa _ Pad )( 1 + 88 BB )( 1 + YY YY )}2, -P P - P 88 do YY Ga (2-22) with a, B, y = a, b, c in cyclic order; Péa = principal second moment of the substituted species. The rs parameters obtained in this way provide internal consistency, particularly in cases where an excess number of isotopic species have been studied and where redundancy relations introduced by symmetry elements are present. Costain has also shown that in simple molecules where comparison can be made, the rs structure is fairly close to the re structure. The zero-point average structure, designated as by Herschbach and Laurie and as rZ by Merino and Oka, assumes that all atoms are frozen at their average positions. The relative ordering of internuclear distances determined 12 from the above methods has been deduced for a number of diatomic molecules (19) as Z S 6 Recently, it has been shown that structures determined by electron-diffraction (27) can be compared to those obtained by spectroscopic methods provided that the differences due to thermal motions are considered. In this work determination of the average structures for cyanodifluorophosphine and difluorocyanamide are not possible, because detailed force field calculations have not yet been done, though a somewhat incomplete normal coordinate analysis has been reported for the latter (28). III. EXPERIMENTAL 3.1 Microwave Spectrometer Advances in the experimental aspects of microwave spec- troscopy have been many and have resulted in new applications to structural chemistry and molecular physics. For example, the magnetron first used as the radiation source by Cleeton and Williams (1) has been replaced by reflex klystron tubes, which are in turn gradually being replaced by backward wave oscillators and a variety of solid state devices. A microwave spectrometer consists of a tunable radiation source capable of generating a practically monochromatic beam, devices for attenuation and measurement of frequency, an absorption cell, and a detection and amplification system. There are two distinctive characteristics of microwave spec- troscopy. First, the wavelength is the same order of magnitude as the spectrometer components. As a result, the propagation of radiation is controlled by waveguide devices. Waveguide is hollow metal pipe whose cross-section varies with the freq— uency region being studied. There are four X-band (0.400” x 0.900" inside dimensions) absorption cells currently in use at the MSU microwave laboratory. A cross-section of one of the sample cells is given in Figure l. The second distinctive characteristic of microwave spec- troscopy is the very small absorption coefficient which 13 14 __,‘ 1.: 0.062 -—ev 0.03] 1.000 —)T ~e——— 0.050 0.062 0.015 -LT— 1. T T }<——- 0.500 —->| Figure l. The Cross-section of an Absorption Cell. (Dimensions in inches) The waveguide, usually made of brass or copper, varies from 4 ft. to 12 ft. long. It is made vacuum tight by means of grooved flanges fitted with a lightly greased O-ring and a mica window (0.002" thick). Ordinarily, the Stark septum is made a few inches shorter than the length of the waveguide. An inlet in the form of a slit in the broad face of the waveguide is sealed by means of a Kovar-glass connection to the vacuum line. 15 necessitates the use of signal modulation in order to increase the sensitivity of the spectrometer. The Hughes-Wilson Stark modulation spectrometer (29) is now used by almost every worker in this area of research. At MSU two such spectrometers with 90 kHz and 100 kHz square-wave modulation can be assembled with either reflex klystron oscillators or backward wave oscil- lators as signal sources in the X, P, and K band regions. A block diagram of a Stark modulation spectrometer is shown in Figure 2. In addition, a digitalized Hewlett-Packard Molecular Rotational Resonance (MRR) spectrometer equipped for the R-band region (26.5 GHz-40 GHz) was acquired recently. 3.2 Detection, Modulation, and Amplification The microwave radiation reaching the detector is rectified into a DC component proportional to the total microwave power, and an AC component which arises from the Stark modulation and hence contains the absorption information. The detector con- sists of a silicon diode in either a narrow band (tunable) or wide band (non-tunable) waveguide mounting. The AC and DC components are separated at the input stage of a narrow-band preamplifier with the former being amplified and fed to the phase sensitive detector, while the latter serves as a monitor for microwave power. When the AC component and a reference signal from the square-wave generator are in phase or out of phase, a DC signal proportional to the former is obtained. The polarity of the DC signal is determined by the relative Ill ).[fll|\[u[[i|\ll|~[?ll?| invilfil ’IpTIIb 51‘ 16 .AouoEopuuoam cofipmfiswoz xwmum m we Emhmmflo xoofim .m ohzmfim mcbumhma Amu MMHZDOU moe »Saa:m an. CA on >m moeHeHmzmm zumemmu +nzpeamm op mmHmumm 1# oHauco:oohm Hmcmfim .xucoscohw madmasu posoa cu“; Q > A mm .mmm V .oocmcommm mansom pom Empmxw Ho>oq oopch m we ofiumEmgom AC A: E S U) Q. ——~ .m whamflm 20 signal 0 was very dependent on the modulation frequency and s on the phase setting of the phase-sensitive detector. Thus, the best operating conditions could be achieved by using a pair of standard transitions to tune the spectrometer. 3.5 Stark Effect Measurement All Stark effect measurements were made with a Fluke 412B high DC source connected to the septum through the square- wave generator. The DC voltage offset the base of the square wave. The spacing of the sample cell was calibrated with the 0-1 transition of OCS; the revised value of the dipole moment ( uOCS= 0.7152D ) determined recently by Muenter (36) was used. The result of a typical calibration is given in Appendix A. 3.6 Relative Intensity The relative populations of two vibrational states can be expressed by the Boltzmann equation, Nl/No = exp(—AE/kT). (3-1) In this equation AB is the energy separation between the two states and kT has its usual physical significance. The relative pepulations and hence the energy separation AF can be estimated from a comparison of the intensities of two similar rotational transitions in the two vibrational states under identical exp- erimental conditions. Energies of low frequency vibrational modes may be determined in this manner. IV. CYCLOPROPYL METHYL KETONE 4.1 Introduction Recently, microwave studies of several carbonyl deriva- tives of cyclopropane have been undertaken in this laboratory. Results of the investigations of cyclopropanecarboxaldehyde and cyclopropanecarboxylic acid fluoride have just been re- ported (37-38). In both cases the torsional potentials governing the twofold rotation of the carbonyl group with respect to the cyclopropane ring were deduced and discussed in terms of the electronic effect of the orbitals of the axial atoms and the steric effect of the non-bonded atoms. It was felt that additional information about this potential function could be learned from a study of cyclopropyl methyl ketone, provided that two rotamers were present in appreciable amounts. A gas-phase, electron-diffraction study of cyclopropyl methyl ketone (39) showed that about 80% of this compound is present at room temperature with the CO bond eclipsing the cyclopropane ring (Figure 4). The remaining 20% was attributed to an anti-gauche rotamer, in which the -COCH3 group was rotated 1500 out of the symmetry plane (Figure 5). But the uncertainty reported in this study, 115%, was relatively large and there was some question as to whether the minor species was an anti-gauche or a trans rotamer. In addition, a feature 21 22 010 H5,H7 l C C2,C3 L 9 /////r ””////’ H12 H60H8 C1 C11 H139H1u H4 Figure 4. Projection of cis-Cyclopropyl Methyl Ketone in its Plane of Symmetry (e- 0).' b Hiu H13 HS H12 C11 C2 H6 1 7 CI C9 C3 1 010 Ha Ha Figure 5. Projection of anti-gauche Cyclopropyl Methyl Ketone in the ab Plane(6= 150°). 23 not observed in the electron-diffraction study was the inter- nal rotation of the three hydrogen atoms in the —COCH3 group. A microwave study could be expected to provide a value for the potential barrier hindering this motion. Unfortunately, it was not possible to assign a spectrum to a second species. However, the principal species was de- finitely assigned as a cis rotamer, the barrier to internal rotation was determined, and the dipole moment was estimated. 4.2 Spectrum The preliminary structural parameters of cycloprOpyl methyl ketone were transferred from acetone (40), cyclopropyl chloride (41), and the electron-diffraction study. The rotational constants were calculated with the computer pro- gram ST. The molecular parameters, atomic coordinates, and resulting moments of inertia are given in Tables 1 and 2. The approximate spectrum calculated by means of the computer program EIGVALS then served as a guide in the initial search for an assignment. The spectrum of cyclopropyl methyl ketone is extremely rich, with transitions practically every few MHz. An intensive effort to locate transitions having a characteristic Stark effect proved to be futile, except for the single transition at 15214.78 MHz. However, a series of strong and closely spaced doublets were observed. The initial assignment was made when the lowest two members, 606-615 and 707-716’ of the 24 Table l. Assumed Structural Parameters and Rotational Parameters of Cyclopropyl Methyl Ketone, Bond Distancesa and Bond Anglesb CICZ 1.515 C1C2C3 60. C2C3 1.515 HSCZH6 116. CH 1.085 HECICg 117.5 Cng 1.507 C1C90 127.4 CO 1.222 0C9C11 115.5 CgCll 1.505 H9C11H12 110. Rotational ParametersC Eli Anti-gauche Ia 69.9193 75.6000 1b 197.4172 181.3175 IC 223.9544 218.6807 K -0.877967 -0.782242 3 in A. b in degrees. 02 C moments of inertia are in u-A 25 Table 2. Cartesian Coordinates for cis-Cyclopropyl Methyl Ketone in the Principal Axis Systema Atom a b c C1 -0.55847 -0.65187 0.00 C2 -l.69110 0.01038 -0.75750 C3 -1.69110 0.01038 0.75750 H“ -0.54056 -l.73672 0.00 HS —1.47484 0.94980 -1.25543 H6 -2.40370 -0.63884 -1.25543 H7 -1.47484 0.94980 1.25543 He -2.40370 ~0.63884 1.25543 C9 0.76684 0.06549 0.00 O10 0.95642 1.27270 0.00 C11 2.01067 -0.78536 0.0000 H12 2.89700 -0.15953 0.0000 H13 2.03574 -1.41857 -0.88071 H1“ 2.03574 -l.41857 0.88071 a The coordinates shown here were calculated from the assumed structural parameters given in Table 1. 26 0,.1"71,.1-1 their relative splittings. The frequencies of the A-level series J in the X—band region were identified through components of the two transitions were used to determine the rotational constant combinations (B-C)/2 and {A-(B+C)/2}. Determination of the effective rotational constants A , B , and C of the A species was completed after a few R—branch transitions were assigned by means of double resonance, as shown in Table 3. The E—level transitions were observed, fit, and confirmed in a similar manner. A comparison of the ob- served and calculated frequencies for a number of transitions is given in Table 4, while the rotational parameters are pre- sented in Table 5. The lowest frequency vibrational mode in cyclopropyl methyl ketone probably arises from the torsional motion about the C-C bond joining the acetyl group and the cyclopropane ring. Transitions belonging to the lowest three excited states of this mode have been observed and identified, and the freq- uencies are shown in Tables 4 and 6. To a good approximation the methyl group internal rotation splittings observed in both the ground and first-excited torsional state are the same. Effective rotational parameters for the excited states are also given in Table 5. That the lowest-frequency vibration is relatively free from vibration-rotation interaction can be demonstrated by the linearity of plots of the rotational constants 1:. the vibrational quantum number, as shown in Figure 6. Rough relative intensity measurements of several transitions in 27 Table 3. A-level Double Resonance Connections for the Ground State of cis-CycloprOpyl Methyl Ketone. Pump Pumpa Signal Signala Transition Frequency Transition Frequency 211+ 220* 12737.56 211+ 322 28516.32 2121 221* 14709.68 212+ 321 29591.58 606+ 615 9060.63 606+ 717 35358.91 707+ 716 10792.80 616+ 7o7 31501.77 a Frequency in MHz. * E-level was also pumped and observed. 28 woscwucou flee.oVom.m flfim.o-vwo.amoofi flmo.ovwm.e Afic.c-vfie.meK©H .mm 1m th.ova.H mmfi.o-vom.momNH mwm.onm.H 540.0-me.moamfi mam Hm flofi.oVoo.o- flmH.o-Voo.KwOHN fioH.oVHH.c- mmfi.o-vom.wNOHN mom is fleo.onN.K ASN.0-Vmo.ameH flmo.onK.o mmo.ovKo.mmmmH was is flKH.oVew.o- floo.oVHm.mmeH nmfi.onw.o- flmo.o-vfiK.ameH .oe 1m flKo.oVeo.H flmH.o-me.mmwNN .se om fleo.one.NH Aeo.oVwK.eHNmH Nam 2m mem.oVNN.K- flofi.ono.HNNMH mem.o-vom.c- heo.o-vam.NNmMH Ham 1m meN.o-UNm.mN nco.ovwo.moaefi HNN 1N flow.oVHo.om- flea.o-vefi.mmomfi ”ON.O-Vwe.aH- hmo.oVom.KmeH CNN 2N mwm.o-vmm.om Aoo.ovmfi.efimmm Ham 2H nfim.o-voa.mfi- AHH.onN.HoNeN cam 1H mofi.oVoe.H moo.ovfiK.omoeH ASH.oVom.H afioo.ovmm.KOHeH NLN 0H nm>-<>v «9 nm>-4>v M> :ofiufimcmhh Hu> > .oumum powwoxm umpfim paw pancho ecu ca ocouox H>guoz Hypogaofiuxu-mfiu how mcofiuflmcmhh mo mowuceSUohm .v oHpmH 29 Hmcoflumpom .m ofipme ca ohm mucmumcoo .mmfifimkr UmmeSUHmU WSCfiE ®®>Hmmflo Ohm mommfiufimhmm Cw mmSHm> m .Nmz mc.ou mfl xucwmppouc: woumeumo ”N22 :fl 2 Awfi.ovee.m ham.ovfic.NKHNH hmc.ovom.N meo.ovem.NmNNH Lam mam nom.o-veo.m Amm.o-vma.KKHmH ”64.0-Voe.m Amm.o-vmm.fiemmfi mam mom nKN.OvHe.N ham.ovwm.mNmHH hmH.onN.N hem.0vfie.momfifi saw him ASN.0-VNN.N hom.ovoa.oowNH hHm.c-VKH.N “mo.o-vmo.mowmfi 22m mow mmm.ovmm.m floo.ovaa.HAKHH fimN.oVNm.N nON.OvKo.mmmHH mNK 21K mKo.o-Voo.N floa.o-vom.NeKOH flmN.o-Vew.H ASH.oVom.NQAOH LHK 20K ”om.oveo.m mmfi.o-va.emmHH flKN.ono.N n00.0vmm.oqomfi .No mac hmfi.o-voo.fi hmo.o-voq.efiom AmH.o-voo.H flmH.ono.ocom mac 206 flm>-<>v <> fim>-<>v <9 cofipwmcmhk :fle.p:66V a magma: 30 Table 5. Rotational Parameters8 for cis-Cyclopropyl Methyl Ketone. Parameter v=0 v=1 v=2 v=3 AA(MHz) 7204.35 7172.70 7141.76 7112.38 BA 2630.90 2633.34 2635.68 2638.04 cA 2301.14 2304.67 2308.12 2311.84 AA 0.81 0.88 AB 0.12 0.07 AC 0.07 0.01 Ia(U°A2) 70.1488 70.4583 70.7635 71.0559 1b 192.0922 191.9142 191.7438 191.5723 1C 219.6197 219.2836 218.9554 218.6032 paab(u-AZ) 170.7816 170.3698 169.9679 169.5599 Pbb 48.8381 48.9138 48.9876 49.0434 PCC 21.3106 21.5444 21.7760 22.0125 K -0.865492 -0.864966 -0.864466 -0.864115 3 b Estimated uncertainty in rotational constants is 0.05 MHz; AA = AA - AE’ etc. Paa = (Ib + IC - Ia)/2, etc. Table 6. Frequencies of A-level Transitions of cis- Cyclopropyl Methyl Ketone in the Second and Third Excited State. v=2 v=3 Transition v* 0* 313 + 404 15896.41(0.00) 303 + 41“ 22849.98(-0.19) 41k + 423 15676.98(-0.27) 41“ + 505 21144.14(—0.17) 514 + 523 12215.46(-0.25) 12127.34(—0.34) 515 + 52“ 16526.87(-0.64) 615 + 62“ 11865.40(-0.03) 62k + 633 22399.63(0.16) 616 + 707 31630.01(—0.24) 707 + 716 10691.72(-0.42) 11685.59(O.13) 11743.10(0.33) 12093.60(0.76) 11604.06(0.61) 11663.13(-0.26) * Estimated uncertainty =:0.05 MHz; given in the parenthesis. ( YAobs YAcal ) are 32 MHz .—_—_ ____ __ ——1~ . AC r—- ———1 6 0 —— -1\A —# AB —- ——4 2 0 _— .— 1 2 3 v = vibrational level Figure 6. Change of Rotational Constants (in MHz) as a Function of the Vibrational Level. -AA has been reduced by a factor of 10. 33 the ground and first excited states lead to a value of 90:10 cm for the excitation energy of this mode. The best evidence that the species assigned has a plane of symmetry is a comparison of the out-of—plane second moment of cis-cyclopropyl methyl ketone with corresponding values for other cyclopropane derivatives. For this comparison, shown in Table 7, the out-of—plane second moment of cyclopropyl methyl ketone was reduced by subtracting the effect of the out-of- plane hydrogens of the methyl group. 4.3 Barrier to Internal Rotation To compute the internal rotation splittings Herschbach's formulation (42) was used. In this formulation the effective Hamiltonian for each torsional state is written as H = Ho + F 2 w52¥99(“) (4-1) V0 with the following definitions: Ho = AD; + 80% + cp: A = hz/Zla, etc. Pg = Component of total angular momentum along the g pr1nc1pa1 ax1s €90 = Zap8A8/12 Ag = Direction cosine between the tOp axis and the g principal axis g = The principal axes a, b, or c Ia = Moment of inertia of the top about its symmetry axis 2 T = 1 - XgAgIa/Ig .cofiumoficseaoe epm>flhm acmsevcezzum .m .m paw mcflwneu .o .m 34 n .nnm u>>m mwcnanou Hezuo you .mcmam we use ohm :uflzz asmwm fissuos one Eopm macaw somepwzz mm: vasanou many mocflm N\NH.m n we u a ocewex Hxspee Hzmoumofiuzo-mfio you .uceEoE ecouem ocmfla-mo-u:o u a . <.: cw w No a om.mfi mammopaofiuxufixcfl>-mcmpe mm ow.mH ewfihosaw wwum ofifixxonhmuecmmopmoHuzu-mcmhe mm nn.mfi ewwAOSHw @Hum u“fixxonemeechOHaoHuxu-mfiu mm on.mfi ewxnewfimxonumuecmaogaoHuxu-mcmue um on.mH omxcewfimxonhmuechOMQOHoxu-mfiu Macs mflgh mn.mfi ecouex axguez Hunchaofiuzu-mfiu >> oecehomem d mazoofioz >> .mo>fium>flpem mamaoemofiozu pom mosflw> mm we somwquEou .n ofinmh 35 F = hZ/ZrI a v = Torsional quantum number 0 = A or E sublevel WVOrm = Perturbation coefficient The internal rotation splittings for many transitions were calculated by means of a desk calculator and later checked by a computer program INROT. If it is assumed that V3>>V6, only Ia’ A , A, B, C, and V3 are variable parameters. Of these, the rotational constants can be obtained independently from the spectrum and Ag can be calculated from the assumed structure. In the cis-species, only Aa and A are non-zero b because of the plane of symmetry and Ab2=n/2 - 138 The moment of inertia of the methyl group Ia varies little from molecule to molecule and was assumed to be 3.12 u-Az. The remaining adjustable parameter is V3, the threefold barrier to internal rotation. The value of V3 is then determined empirically from the A- and E-level splittings. The best value obtained for V3 was 1180120 cal/mole. All internal rotation parameters used in the present calculation are listed in Table 8, and a comparison of the value of V3 for this molecule with that for several molecules is given in Table 9. 36 Table 8. Internal Rotation Parameters for the Ground Torsional State of cis-Cyclopropyl Methyl Ketone. Assumed 02 IQ = 3.12 U’A Ta = 0.8254 6 = 34.40 F = 167.986 GHz Determined A0 = 7203.34 MHz 5 = 32.74 B0 = 2630.82 V3 = 1180120 cal/mole C0 = 2301.11 Table 9. Barrier to Internal Rotation for Some CH COX Compounds. 3 Molecule V3(ca1/mole) Reference acetaldehyde 1162 43 acetyl bromide 1305 44 acetyl chloride 1296 45 acetyl cyanide 1270 46 acetyl fluoride 1041 47 methyl vinyl ketone 1250 48 cyclopropyl methyl ketone 1180 this work 37 4.4 Dipole Moment As mentioned earlier, a major complication in the initial search for an assignment was the absence of resolvable Stark effect in most transitions. Examination of all the low J transitions observed revealed only one which had a Stark effect that was suitable for dipole moment measurement. The sole transition was the 313-322 line with two Stark lobes at 15214.78 MHz. These two lobes were used to determine uaand “b with the results shown in Table 10. The total dipole moment vector makes an angle of approximatelylO with the CO axis. As a comparison, the dipole moments of several related mole- cules are presented in Table 11. Table 10. Stark Effect Data for cis-Cyclopropyl Methyl Ketone. Transition M *(dv/dEz) 313 T 322 2 -30.8 3 T 3 3 -128 8 13 22 I N ”a: 0.47 $0.17 D .58 $0.26 D u = 2.62 10.26 D Uh— * 2 unit in MHz/(kvolt/cm) ; uOCS = 0.7152 D assumed. 38 Table 11. Comparison of Dipole Moments for Some Molecules Containing a Carbonyl Group. Molecule u(D) Reference Acetaldehyde 2.69 43 Propienaldehyde (cis) 2.52 49 Butyraldehyde (trans-trans) 2.59 This thesis (gauche-trans) 2.50 Cyclepropanecarboxaldehyde (cis) 2.74 (trans) 3.26 37 Acetone 2.90 40 Z-Butanone (trans) 2.78 50 Cyclopropyl Methyl Ketone (cis) 2.62 This work 39 4.5 Discussion A major goal of this investigation was to determine the barrier to internal rotation of the methyl group in both the cis- and anti-gauche species. The cis-species, though obs- cured by a high density of lines, was eventually identified and the internal rotation barrier was determined. The search for the second species was exhaustive. An extensive Stark effect search was centered around regions where strong a- type transitions were predicted from several structures as- sumed for the trans and gauche-like species. No assignment consistent with any such structure could be made. Attention was then focused on doublets that were not assigned to the cis-species. Fittings were made by varying the values of V3, but the results were again negative. Finally, double-search, double resonance techniques were utilized, and the results were also unfavorable. In view of the fact that transitions in the cis-species could readily be observed through the third vibrationally ex- cited state, it seemed plausible that the anti-gauche species, if present at all, could not amount to more than 5%, the lower limit set by Bartell (39). The non-bonded distances HS-H (location shown in Figure 4), plotted as and Hs-H 13 1'4 functions of the dihedral angle 0 and of the methyl group rotation and illustrated in Figures 7 and 8, respectively, show that at O=150o (predicted for the anti-gauche species) the Hs-Hlt distance is considerably shorter than the Van 40 4.00 l | I l 1 60° 120° 180° O 0 Figure 7. Plots of the Hs-H13 and Hs-Hlt Distances (A) versus the Dihedral Angle O. 111111 20° 60° T Figure 8. Plot of Hs—Hlt Distance (A) versus the Internal Rotation r for the Methyl Group in anti-gauche Cyclopropyl Methyl Ketone. 41 der Waals radii of 2.40 A. Nevertheless, the failure to observe the spectrum of the anti-gauche species is no proof that it does not exist. The apparent absence of a second species may be due to a relatively high energy separation from the cis—species, or it may be due to a severe vibration-rotation interaction which would make a search requiring a rigid—rotor spectrum for confirmation impossible. As shown in Table 9, the value of the barrier height determined for cis-cyclopropyl methyl ketone is comparable to those reported for other CH3COX molecules. Wilson (51) has suggested the existence of an intrinsic barrier for a characteristic group such as the CH3CO— group. From this Table it is apparent that the barrier height for the CH3CO- combination is essentially independent of the substituent X. V. CYANODIFLUOROPHOSPHINE 5.1 Introduction A microwave study of cyanodifluorophosphine, PFZCN, was undertaken in response to the current interest in the structural and bonding aspects of substituted difluoro— phosphines (52-54). Of particular interest in this study were the P-C bond distance and the PCN angle. The results of a determination of the crystal structure of phosphorous tricyanide, P(CN)3, by X—ray diffraction (55) indicated that the average PCN angle is about 171230, rather than the 1800 that would be predicted by simple bonding theory. Miller et a1 (56) speculated that this non-linearity was due to the close packing in the crystal lattice, and that it would not occur in free molecules. 5.2 Experimental The synthesis and characterization of cyanodifluoro- phosphine was first reported by Rudolph and Parry (57). Samples of PFzCN, PF213CN, and PFZCISN were prepared (in collaboration with Dr. K. Cohn) by the reaction, PFZI + CuCN + PFZCN + CuI . (5.1) 42 43 1 13 S For the C and N species, the corresponding cuprous cyanides were obtained from the appropriately enriched KCN following the procedure of Barber (58). The enriched samples contained 15% PF213CN and 25% PFZCISN, respectively. In each case, PFZCN was confirmed through its mass spectrum. The mass spectrum also revealed the presence of PFZI as the major im- purity, apparently because PFZCN and PFZI have about the same volatility. Although PFZI has a fairly rich spectrum, its interference in the spectrum of PFZCN has been minimal. 5.3 Spectra The spectrum of PFZCN was initially calculated from an assumed structure. The ac plane was predicted to be a plane of symmetry, so that a- and c-type transitions were predicted. The normal species was readily assigned when the three a- type J=1+2 transitions were identified through their charac- teristic Stark effect. iSome c—type transitions were subse- quently located. Spectra for the 13C and 15N species were assigned in the same fashion. The frequencies of a number of transitions and the rotational parameters for the three species in the ground state are listed in Tables 12 and 13. Transitions belonging to excited states of the lowest frequency vibrational mode of the three species were also observed, fit, and are given in Table 14. The corresponding rotational parameters are presented in Table 15. At dry ice 44 wcucou flaa.o-vmm.KHANN flfim.ovflo.ommfim flNH.o-Vao.HHoNN ANN.oVoo.OONKH mmo.ovam.fiemKH hmo.ovoo.mmmofl hmm.o-vfie.mmamfi mHo.o-VmH.ooaoH mwo.oVom.HNaHH noo.ovfim.NmSOH flHN.o-Vmo.mmNHH flaw.o-vmw.mKNmN mmN.one.oNKHN flafl.o-vfio.amemm floo.o-vfiN.oq0KH mHH.one.emeH Ame.ovma.momafi “mo.oVNN.wemoH nmN.0vem.omoafi flac.c-V~H.ANNNH flHH.ono.©Nm0H Amm.o-vNe.oomHH mom.o-vmo.oommm floc.o-vmm.ofimfim m0fi.0vmm.omomm AmN.oVKm.KHwKH AmH.o-VNm.mKemH hom.oVHw.ommKH Amo.o-vefi.mmeofi “40.0me.memafi Amm.oVow.aemNH flam.oVoo.eNOHH aflmfi.oVem.MHoHH 2 game m” zu NEE mH zuwwm cofiuwmcmpe .zu~mm we mefloomm uflmOHOmH Mom meofluwmcmeh Hmaoflpmuom eumum poncho we mofiocosconm .NH oHQMH .mH ofinwb ca ohm mueueEmumm Homewumpom .NIZ mo.OA ma mcofiufimcmeu we>pemno map ca zucflmupooca .mofiocoscehm ecu we mesfim> woumfisofiwu use we>pemLo ceezumg meocehomwflw ohm memogucohmo cw whenezc 45 m Amo.c-va.HoHKH ism + 20H AmH.oUNK.omSOH 02H + 000 Amo.ovmm.omemm th.oUmm.moNom Amo.o-vNH.mmaom .2m + .24 ASN.0-VwH.mmOAN Amm.o-vmm.mmNKN mam + .24 m0fi.o-vmm.NHNmN floN.o-VNm.NmmmN Amm.o-va.momom mam 1 Nae mmo.o-vmm.omHKN mom.o-vma.mmKKN mmN.o-VKm.momKN mom + .04 mam.o-vmm.40KmN flom.ovmm.ammsm flmo.cvmm.ommem 224 + Nam zmsoada zomsaaa zomaa cofiefimcmch 2m.@.#COUVNH OHQNH: 46 Table 13. Ground State Rotational Parameters of PFZCN, PF213CN, and PFZCISN. Parameter PFZCN PF213CN PFZCISN Aa 7403.56 7403.92 7400.48 B 3253.01 3219.36 3135.39 c 2590.47 2569.07 2515.62 lab 68.2612 68.2579 68.2897 1b 155.3562 156.9802 161.1847 IC 195.0905 196.7156 200.8952 paac 141.0927 142.7190 146.8951 pbb 53.9977 53.9967 54.0001 PCC 14.2635 14.2612 14.2896 K -0.724691 -0.730998 -0.746510 a in MHz. Uncertainty in the rotational constants is 0.05 MHz. 0 O in u-Az; conversion factor: 505376 MHz-u-Az. 0 C ' . 2. = _ , in u A , Paa (Ib+Ic Ia)/2, etc. .Nmzmo.OH ohm mofiocoswopw we>hemno esp :H mofipcwmupeucs .mH efinmh :fi ohm whouosmpmo Hmcoflumuom .mefio:o:cehw peumfisofimo mamas we>uompo ohm memezpcouma ca muonssc 47 m flwo.o-ume.oaoem mac + .am flmH.onm.amoom mac.ovmm.mfieom mam 4 Nae moc.onm.HemKN floo.onm.mNmKN mam + sue oe.omeN fleH.one.oNHaN .Nm + Mae oe.aemmm flKH.o-VOK.awNom mom.oVHe.owmom .sm 4 nae ee.awHKN fimN.ono.maKKN floo.oumm.emowN hwc.onH.eamKN mom + .04 mNo.onm.memeH mmm.o-vmfi.mfimefi Mam + Nam n00.0vfio.OOOHH amNH.o-Vem.NeoHH Nam + 22H Hn> Hu> Nu> Hu> zmaoama zonaaaa zeaaa :ofiufimemcb .zuama we mofloeom ofioouomH new moumum wepfioxm zfifimcofiumpnfl> cw mcofipfimcwpe mo mefiocmsuepm .QH oflpmh 48 C in u-AZ; Paa = (Ib+IC—Ia)/2, etc. Table 15. Rotational Parameters for the Vibrationally Excited States of PFZCN, PF213CN, and PFZCISN. PFZCN PF2‘3CN PFZCISN Parameter v=1 v=2 v=1 v=1 Aa 7377.22 7351.29 7367.08 ‘7387.22 B 3265.50 3275.49 3231.60 3147.72 C 2592.40 2594.84 2571.04 2517.31 lab 68.5049 68.7466 68.5992 68.4122 Ib 154.7624 154.2906 156.3856 160.5528 IC 194.9455 194.7619 196.5648 200.7603 Paac 140.6015 140.1529 142.1756 146.4505 Pbb 54.3440 54.6090 54.3892 54.3098 'PCC 14.1609 14.1376 14.2101 14.1023 K -O.718653 -0.713801 -0.724539 -0.741098 a in MHz. Uncertainty in rotational constants is 0.05 MHz. in u-Az; conversion factor: 505376 MH2°U°A2. 49 temperature the relative intensity ratio between the first excited state and the ground state is approximately 0.4, corresponding to an energy separation of about 120 cm'1 The assignment of the infrared spectrum of this compound is incomplete (57). The frequencies of two modes- the rocking and out-of—plane wagging motion of the -PF2 group— are be- lieved to lie below 200 cm‘l. By analogy with PFZNH2(54) the lowest-frequency mode observed in the microwave spectrum is probably the out-of—plane wagging motion of the PF2 group. In addition, rotational transitions arising frem molecules in the first excited state of a second mode were also observed in the microwave spectrum but not assigned. The relative intensity of these transitions is intermediate between those of the first and the second excited state of the lowest fundamental. 5.4 Molecular Structure In PFZCN there are five atoms and hence a total of fifteen atomic coordinates needed to fix the structure. As a result of the plane of symmetry, the b coordinates of the P, C, and N atoms are zero and the b coordinate of one fluorine atom is the negative of that of the other. Also, the a and c coordinates of the two fluorine atoms are the same. To fix the remaining nine coordinates there are available three second moments of the normal species, two second moments for each of the labeled species, and the two 50 first-moment relations, 2 miai=0 and Zmici=0,and the product- of-inertia relation, Zmiaici=0. Thus, there are ten pieces of information from which to determine nine coordinates. The nine coordinates were determined in two ways. An "r0" structure was obtained by fitting the nine moments of inertia of the three species, the first moment relations, and the product of inertia. A ”semi-rs" structure was obtained by using the Kraitchman equations to obtain the a and c coordinates of the C and N atoms, followed by fitting the remaining coordinates to the second moments of the normal species, the first moment relations, and the product of inertia. The coordinates for the two structures are shown in Table 16 and the corresponding bond distances and bond angles are shown in Table 17. Except for the c-coordinate of the carbon atom the agreement between these two methods is quite good. To examine the effect of the large uncertainty in the c-coerdinate of carbon atom on the overall structural para- meters, a series of ro structures were calculated by vary- ing the carbon c coordinate from -0.006 to 0.010, while the other coordinates were held constant. The results indicated that variation in the structural parameters were insignificant. In comparison with other substituted difluorophesphines, the P-F distance and FPF angle appear to be normal, whereas the P—C bond length and PCN angle were unexpected. The P-C bond length (1.810 A) was slightly shorter than the sum of the covalent bond radii (1.87 A). The non-linearity of the PCN 51 Table 16. Cartesian Coordinatesa in the Principal Axis System of PFZCN, Atom Method a b c NK 0.453210.0033 0.00 0.528410.0028 p K 0.452810.0033 0.00 0.5291i0.0028 NK 0.928410.0016 1.1921t0.0013 -0.370410.0041 Fl K 0.9286i0.0016 1.1921i0.0013 -0.3702i0.0041 NK 0.9284i0.0016 -1.1921i0.0013 -0.3704i0.0041 F2 K 0.928610.0016 -1.1921i0.0013 -0.3702i0.0041 NK -l.2798:0.0012 0.00 0.0068+0.0160b C K -1.2798i0.0012 0.00 0.0000i0.0160b NK -2.424810.0006 0.00 -0.169410.0089 N K -2.4248i0.0006 0.00 -0.166210.0089 a 0 in A. The uncerta1nties in the coord1nates include an experimental contribution and a vibration-rotation contribution estimated to be dq = 0.001S/q. The uncertainty in this coordinate was estimated from 6c carbon { Z(m.6c_) 11/m 1 1 1 carbon 52 Table 17. Structural Parameters of PFZCN, a Parameter r semi-r o S P-F l.567:0.003 l.567i0.003 P-C l.810:0.007 1.81210.007 CEN l.158:0.004 1.157:0.004 < ll 2 2 2 (X cos 9 -X sin 0 )/(cos 9 -sin20 ) 22 aa 2 CC 2 Z 2 (5’3) X ll . 2 2 . 2 2 (X s1n 9 -X cos 9 )/(51n 0 -cos 0 ) xx aa 2 cc 2 z z 54 Table 19. Frequencies3 of the Transitions used in the Determination of the Nuclear Quadrupole Coupling Constants of PFZCN. Transition F + F v u b obs hyp 101 + 202 2 + 3 11613.94(0.06)c 11613.87(0.06)d l + 2 11613.94(0.02) 11613.81(0.02) 0 + l 11612.52(-0.01) 11613.79(-0.02) 111 12 11024.66(0.01) 11024.44(0.01) 11023.20(-0.06) 11024.36(-0.06) 110 11 12349.86(0.01) 12349.52(0.01) 12348.53(0.18) 12349.69(0.18) 221 22 17530.81(0.03) 17530.48(0.03) 17529.31(0.02) 17530.47(0.02) 827 26 10148.67(-0.07) 10148.28(0.09) 10147.36(0.09) 10148.28(0.09) a in MHz. b v = the hypothetical frequency derived by subtracting C numbers in parentheses are (v d numbers in parentheses are (Yhyp ' V bs cal Ycal)‘ the calculated quadrupole contribution from the observed frequency. 55 In these equations 02 is the angle between the CN bond (z-axis) and the a-axis, ny is equal to Xbb’ the out-of- plane component. In comparison with HCN and CH3CN,Xbond of the CN bond of PFZCN is essentially that expected for a pure triple bond. In contrast, the observed asymmetry (x - Xxx) of 5.50 MHz strongly suggests the presence of YY some perturbation to the bonding. 5.6 Dipole Moment The results of an analysis of Stark effect measurements on PFZCN are shown in Table 20. As expected, the utotalof 2.39D is considerably smaller than the m4.0D obtained for organic cyanides (59-60). This decrease may be attributed to partial cancellation of the CN bend moment by the PFZ group in the c-direction. As shown below, a similar decrease is seen upon comparison of the dipole moment of difluoro- cyanamide with that of cyanamide. The orientation of the dipole moment vector in PFZCN makes an angle of about 320 with the a-axis and is shown in Figure 9. 56 Table 20. Comparison of Observed and Calculated Stark Shift Slopes for PFZCN, Transition M (dv/dEz) a (dv/dEz) obs cal 101 + 202 0 “22.67 '23.29 101 + 2 02 1 40.03 39.89 211 + 312 l -10.61 -10.35 211 + 312 2 -40.33 -39.78 212 + 313 l 13.15 13.15 212 + 313 2 48.53 48.40 2 2 u = 4.109 D “a = 2.027t0.011 D 2 2 u = 1.619 D NC = 1.272:0.028 D it = 2.393:0.024 D a in MHz(kvolt/cm)'2 ,u 0.7152 D assumed. a 57 .pceEoz efioowm ecu we cofiumucewwo xfioxfifi cuwz zummm we oHSuospum one 11 .m ohsmflw 58 5.7 Discussion The slight decrease of the P-C bond length and the non- linearity of the PCN angle in PFZCN may be discussed from two points of view. One approach assumes "back bending" by the lone pair electrons in the phosphorous atom to the pTr orbital with appropriate symmetry in the CN bond. The bond- ing in sulfur dicyanide (61) and methyl thiocyanate (62) has also been discussed in terms of this reasoning. Since the phOSphorous lone pair electrons are directed away from the two fluorine atoms, the bend of the PCN angle in this direction could be viewed as the result of this back bond- ing. It might be argued that the bend away from the fluorine atoms could be due to the electrostatic repulsion by the PF2 group. However, in sulfur dicyanide, where no such steric repulsion is expected, the SCN angle is also slightly bent. The second approach invokes (p-d)Tr bonding. In PFZCN the 3dXZ or 3dyz orbitals of the phosphorous atom can par- ticipate in partial bonding with the p1T electron systems of the CN bond. Although (p-d)7T bonding would lead to a shorter P-C bond length, it is hard to see why the PCN angle is bent. Furthermore, there is a question as to whether the 3dXZ or 3dyz orbitals might interact with the two fluorine atoms. In either case, a structure having some ionic character, N=C=PF2 is assumed. Suffice to say, the intuitive arguments presented here are by no means quantitative. Detailed force field calculations or other quantum mechanical treatments 59 are needed before a unique correlation of the structural and bending relationships of molecules such as PFZCN can be given. VI. DIFLUOROCYANAMIDE 6.1 Introduction In the preceeding study of PFZCN it was found that the PCN angle deviated from linearity by about 80 and was tilted toward the lone pair electrons of the phosphorous atom. The tilt was attributed to participation of the lone pair electrons in partial n-bonding with the CEN group. But further studies were needed to yield more convincing evidence in support of this hypothesis. Difluorocyanamide, NFZCN, appeared to be an ideal choice for such a study. In addition, the degree of non—planarity of the -NF2 group of this compound could be compared with the ~NH2 group of NHZCN, since there has been considerable controversy concerning the exact geometry of the -NH2 group in the latter (63-69). An infrared study of NFZCN was reported (28) following its first synthesis (70) in 1966. In this report eight funda- mentals were directly observed and assigned in the 245-4000 cm.1 region. The remaining mode, v attributed to an a" motion 9, of the -NF2 out-of—plane wagging, was predicted to lie around 190 cm-1. In view of the fact that the lowest mode of vib- ration in NHZCN has been well established and assigned as the inversion mode of the ~NH2 group, it seemed pertinent to examine also the nature of such a mode in NFZCN in the micro- wave region. 60 61 In the present study the microwave spectra of four species of NFZCN have been examined and assigned, from which the struc- tural parameters have been determined and the non-planarity of the -NF2 group definitely confirmed. Several low frequency vibrational modes were also observed, two were assigned, and rough intensity measurements were made. 6.2 Experimental NFZCN was prepared by fluorination of aqueous NHZCN in a buffered solution (70). The isotopically labelled species were obtained by reacting KCN with Br2 to yield BrCN, which upon mixing with NH3, gave the labelled NH2CN in quantitative yields (71). In the case of the 15N species, both the -C15N and -15NF2 species were obtained in roughly equal proportions when prepared from either 15NH3 or KC’SN. The various species were confirmed by mass spectra prior to the microwave study. All samples were prepared in collaboration with Dr. K. Cohn, Mr. Henry Carels, Jr., and Mr. David Detzler. 6.3 Spectra The a-type transitions in the normal and other isotopic species of NFZCN were assigned without difficulty. Some tran— sitions belonging to the normal species exhibited small quad- rupole splittings, but no attempt to resolve them was made. Likewise, the c-type transitions were not measured because of their relative weakness. The frequencies of the ground state transitions and the rotational parameters of the four species 2 6 .©.u:ou 500.02Km.ommmm flfio.oSmo.NeNNm fleo.oVwK.moemN flwH.oUKo.aeHmN flao.onw.eHweN “mo.ovoo.momfim nwm.o-vma.mommm Amfi.oVeo.KNmNN mmH.oVam.KHeeH Amo.oVKe.OKNeH floN.oveN.mKNmH fleH.onm.mmeKN fiaH.o-me.mmHHm Aeo.o-veo.flmawm mKo.o-Vme.omeNN Amfi.ovmm.mfioeN AmN.OVme.HOKoN mNN.o-VeH.OHOMN Amm.OVea.omeN fleN.ovom.eKoaH floH.oUNm.memH fiwN.oUee.meeH n0fi.0vee.fififiwm AOH.O-V~m.eeomm heo.ovfim.mKNaN nKm.oVam.meomm “KH.oVaH.emeeN neH.oVeK.NONHN fleo.cvwm.wKemN hHe.oVNo.mNeNN mOO.Oer.mmmeH Ame.onm.meeH hmfi.ovmm.mommfi flmH.oVao.HmmwN fiKH.onN.KNmNm th.cVOH.NmeaN mKH.oVam.eHmmm Aafi.onm.meeN mHH.OVam.oemHm flmH.onm.eeme flKo.oVwo.memNN hoe.oVKe.meeeH mmm.onK.NmNeH flefi.ovfim.OHmmH :H NN :0 mm Na mn Hm mo Hg 61 NA 0»! no ma "N mo HNJ a“ N“ (‘J on No o” HH “0 N 20 mzmfl N zmflu wz N 20 m2 ma zuaaz :ofiuwmcmhp .zUamz mo mofiuomm UMQOpOmH pom meowpwmcmhh Hmcoflumuom oumum vcsoeo we mefiocesceam neeumfisofimu use mwo>pemno mo comwpmoEou .HN manmh 6 .sz a“ mefiocoSUopm weumfisofimo mSCME ©e>gemLo ohm momocucehmo :fi monam> .sz mo.ow u >ucfimuheoc: woumeflumo ”N22 cfl n m mmo.o-vom.oaomm ANQ.O-V~N.Kwamm AKH.O-VNN.meom Amo.onK.meNm hmo.o-vNa.eNHem mNH.o-Vaa.meomm flmfi.o-vae.fimmam AMH.onK.KNmHm Amo.o-vme.afiaem flmH.o-VmK.aNmmm flHH.o-Vom.Koeom ”Ho.oVom.HOKNm Amo.oVNH.mmHmm moN.o-UmH.mmoam nefi.o-vem.emwom fiNH.oUmK.memNm mHm Al mom AI mm? Al Ma? T 1“? 10V NNM Nam zuaaz ma 2 oaaz mH 2o aaz ma zuaaz cofiuflmcmhe :A.e_peoevflm magma: 64 .N<.:.NIZ onmmomnpouomm :oflmpo>cou o m emmmme.o- wHNome.o- mmmame.o- mafieme.o- e Keae.e mKKe.e maes.e mame.e 66a aONa.me Hame.me mKNe.me eeNQ.me pea mwNa.mOH aeeH.mHH oeHK.a0H aoow.w0H aaa Kmem.emH oeam.wmfi emefi.mmfi mKNN.emH 6H emme.mHH N4N©.KHH mmwfi.efifi NoeN.mHH EH eKHH.om aeom.ae omaw.me Kmmm.ae aH mN.eKNm Ka.ewfim Ke.KmNm mw.eKNm o 4K.Keee mm.eame 00.6Nee mo.Neee m Hm.mwQOH we.eNHOH aa.wNHOH we.omHOH < zoaazaa zaauaaz zealaaz zoaaz Aeanaaaa .zuamz mo mofiuoom owmouomH ecu pom poemhmo Hmcoflumuom wo>pomno .NN ofinmh 65 are given in Tables 21 and 22, respectively. Vibrational satellites arising from several low frequency modes of vibration were present in the vicinity of each main line, and the frequencies of transitions in the first two excited states of the lowest frequency mode (hereby designated as v) and in the first excited state of a second mode(v') are listed in Table 23. The rotational parameters of these vibrational excited states are given in Table 24. On the basis of observed inertial defects v definitely arises from an anti-symmetric a” motion, and v' appears to belong to a symmetric a' motion. The vibrational frequencies of v and v' estimated from rough intensity measurements are 150 cm"1 and 220 cm". 6.4 Molecular Structure The observed second moments of the normal and isotopic species of NFZCN are more than adequate for determination of the nine non-zero atomic coordinates in the principal inertial axis system. Since four different combinations of the moments were possible, a series of structural calculations were per- formed and the results are given in Table 25. The atomic coordinates corresponding to calculation No. I are presented in Table 26. In the Kraitchman calculations the in-plane coordinates of the N-CEN skeleton were completely determined by means of the Kraitchman equations (Chapter 11), leaving the remaining 66 Table 23. Frequencies3 of Observed Transitions of the Excited States of NFZCN. Transition v = l v = 2 v' = l 101 202 15342.80(0.44)b 15373.76(0.06) 111 212 l43l3.65(-0.01) 110 211 16716.39(-O.15) 202 303 22604.43(0.32) 22638.66(-0.04) 220 321 23941.50(0.32) 24017.73(0.32) 212 313 21369.32(0.02) 21397.23(0.11) 211 312 24958.09(0.l8) 25033.85(0.11) 24952.11(0.79) 221 322 23271.83(0.00) 23327.88(-0.l8) 303 40“ 29482.00(-0.16) 29511.15(-0.54) 29544.61(0.00) 321 422 32441.94(-0.07) 313 41“ 28324.05(0.06) 28356.23(0.02) 28379.08(0.01) 312 413 33037.02(0.26) 33128.90(—O.40) 322 423 30895.00(-0.14) 40“ 505 36062.74(-0.39) 36154.30(0.00) 41“ 515 35173.28(-0.l3) 35207.47(0.23) 35248.36(—0.01) a in MHz; estimated uncertainty is :0. b 05 MHZ. values in parentheses are observed minus calculated frequencies in MHz. 67 Table 24. Rotational Parameters for the Excited States3 of Normal NFZCN. Parameter v=1 v=2 v'=l A (MHz) 10102.70 10079.64 10168.81 B 4479.50 4496.87 4474.38 C 3278.05 3279.15 3287.38 Ia(u°A2) 50.0239 50.1383 49.6986 Ib 112.8199 112.3840 112.9489 IC 154.1695 154.1180 153.7323 Paa 108.4828 108.1819 108.4913 pbb 45.6868 45.9362 45.2410 PCC 4.3371 4.2021 4.4576 K -0.647911 -0.641873 -0.655014 a The lowest frequency vibrational mode is designated by the quantum number v and the other mode is represented by v'. 68 Table 25. Structural Parameters from Analyses of the Isotopic Species of NFZCN. N0. Methode CEN N-C N—F NCN CNF FNF Ia NK 1.154 1.391 1.396 171.2 105.0 103. K 1.163 1.380 1.399 174.5 105.5 102. b NK 1 163 1.369 1.396 175.0 106.4 103. 11 K 1.163 1.376 1.391 175.1 106.2 103. C NK 1.150 1.393 1.396 170.7 104.9 103. III K 1.157 1.387 1.399 173.6 105.3 102. d NK 1.156 1.389 1.396 171.1 104.9 103. IV K 1.165 1.380 1.399 174.7 105.5 102. a 13 15 15 From parent, C, C N, NFZ. 13 15 Parent, C, and C N. C 13 15 Parent, C, and NF d 15 15 Parent, C N, and NF K = Kraitchman calculation; NK = no Kraitchman calculation. Bond distance in A and bond angle in degrees. 69 Table 26. Cartesian Coordinatesa(A) in the Principal Axis System of NFZCN. Atom Methodb a b c NK 0.9629:0.0062 0.00 -0.0585t0.1000C C K 0.9611:0.0062 0.00 -0.0913:0.1000C NK 2.1017:0.0029 0.00 0.1251:0.0480 N K 2.1009:0.0029 0.00 0.1382t0.0434 NK -0.3599:0.0167 0.00 -0.4873t0.0123 N K -o.3600:0.0167 0.00 -0.4914:0.0122 NK -0.9460:0.0063 1.0934t0.0055 0.1520:0.0387 F1 K -0.9467:0.0063 1.0934:0.0055 0.1551:0.0387 NK -0.9460:0.0063 -1.0934:0.0055 0.1520:0.0387 F2 K -0.9467:0.0063 -1.0934:0.0055 0.1551:0.0387 a From calculation No.1 described in Table 25. The uncertainties in the coordinates include an experimental contribution and a vibration-rotation contribution estimated to be do = 0.0015/q. NK K no Kraitchman fit( ro structure); Kraitchman fit ( partial r in NFZCN). s The uncertainty in this coordinate was estimated from the expression 2 1 6c = { X(m,6c,) ié/m 1 1 carbon carbon 70 three coordinates of the fluorine atoms to be determined with five pieces of information. The out of plane coordinate b of the fluorine atoms was determined from the relation pbb = ZmFbF2 where mF is the mass of a fluorine atom; the other two in plane coordinates were obtained by least squares from the first moment and cross-product conditions as well as P33 and PCC of the normal species. In the No Kraitchman fits (the rO structures) the least squares principle was utilized to determine the nine coord- inates from a total of nine observed second moments plus the two first moment and the cross product conditions. Table 25 shows that the Kraitchman and No Kraitchman results are practically identical for calculation No.11, but differ appreciably for the others. The CEN distance is in excellent agreement with values for other cyanide compounds. The N-C distance, however, is about 0.06 A shorter than that of CHaNFz but 0.04 A longer than in NHZCN. The N-F distance, the CNF angle, and the FNF angle appear to be normal. The NCN angle is noted to differ from 1800 by the same amount as was found for the PCN angle of PFZCN. The final structure of NFZCN, as given in Figure 10, is compared with that of several related molecules in Table 27. 71 1f .ocmfia ShuoEESm mu“ ca zuamz we cofluoemoem .oH acsmsa 72 .eaeMLHaaa an ac .efiaamou .o .o eea .eaefleaam .e .aafixe .e .e m .vn .wem .mn .mem .Nn .mom 6 w o m .mo~vcmenouwmux o: u xx A a Hafieaeav amazeefiaex n e ”zoaazaa .zaaoaaz .zomLaaz .zoaaz mo DEL 0:“ seem eexae n .moehuow cw ofiucm wcop mam « cfi oocmumww ccom m m.NOH o.~o~ v.0AN.moH w.oww.~ofl m2m o.qoa m.owm.mofl e.oum.moH wzu o.ow~.flna m.o“m.wna zuz Hum.fi ooq.H qu.H moo.ouomm.a moo.owmmm.fi m-z oem.fi meq.fi moc.o«amm.~ ooo.owomm.fi 0-2 ooH.H moo.ouva.H woo.owmoH.H 2mm x2 x a a . m h N . N .m . a N a heuesmhm m7u :7 o m2 e m2: on? mp n/u dz a moflsoofioz weumfiem wcm zu~mz we mmhoueEmpmm Amazuefioz .nm ofinmb 73 6.5 Dipole Moment Accurate measurements of the Stark effect of the 10,-202 and 111-212 transitions were made and the results are summarized in Table 28. The total dipole moment is 1.10:0.06 D and makes an angle of 200 with the a-axis. The values of the dipole moments for some related PFZX and NFZX molecules are compared in Table 29, and it is at once seen that for those in which X=F and CN, a sizable decrease of dipole moment is noted on going from a phosphorous com- pound to its nitrogen analogue. On the other hand, this trend is reversed and smaller when X= H and CH3. The dipole moments of NF3(72) and HNF2(73) have been interpreted in terms of bond moments. Pierce (74) pointed out that a model based on bond moments alone could not ex- plain the dipole mements of NF3, N(CH3)3(75), and NFZCH3(74). However, the discrepancy could be accounted for if bond polarizabilities (inductive effects) were taken into account. For a comparison of phosphorous and nitrogen compounds, as in Table 29, the dipole moments can be explained in terms of bond moments, probably because of the large electro- negativity differences involved. The key to an explanation of the comparisons in Table 29 is the contribution of the lone pair electrons on nitrogen and phosphorous. If it is assumed that the nitrogen lone pair represents a sizable negative charge, or a bond dipole 74 Table 28. Stark Coefficients and Dipole Moment for NFZCN, 2 a 2 a Transition M (dv/dE ) (dv/dE ) obs cal 101 + 202 0 -5.05 -4.92 101 + 202 1 5.54 5.59 101 + 202 l 4.55 4.36 2 2 “a = 1.060 D “a = 1.03010.03 D 2 2 “c = 0.154 D “c = 0.39:0.08 D ea = 20.80 it = l.10210.06 D a in MHz(kvolt/cm).2 Table 29. Dipole Moments of NFZCN and Related Molecules. Molecule fl Ref. Molecule D Ref. (D) (D) PF3 1.03 72 NF3 0.23 72 HPF2 1.32 52 HNF2 1.93 73 (311,121:2 2.06 a CH3NF2 2.57 74 CNPFz 2.39 b CNNFz 1.10 b a E. G. Cedding and R. H. This work. Schwendeman, private communication. 75 in the direction of the lone pair, and that the contribution of the phosphorous lone pair is considerably smaller, the trends can be rationalized. When comparing PFZX and NF2X compounds it is apparent that the lone pairs tend to reduce the dipole moments when X= F or CN, whereas they tend to increase the moments when X= H or CH3. To demonstrate this the orientations of the dipole moment vectors in some of these molecules are shown in Figure 11. 6.6 Discussion The microwave study of NFZCN has yielded several chemi- cally interesting results. First, the shortening of the N-C distance and the tilt of the N-CEN angle tend to confirm the participation of the lone pair electrons of the nitrogen atom in the n-bonding of the CEN group; the result is the partially ionic structure +NF2=C=N-. However, the contri— bution of this structure to the equilibrium configuration should be slightly less than in NHZCN, since the N-C distance is 0.04 A shorter in the latter. Second, the larger degree of non-planarity in NFZCN compared to NHZCN is demonstrated by the observed FNF angle of 1030, compared to 113.50 for the HNH angle in NHZCN. This difference may be attributed to the presence in the fluorine atoms of lone pair electrons which repel the nitrogen lone pair. .mofinoefioz mem com wa2 eEom New mpouoe> ucoEoz oHoowQ may we sewumucoflho .HH ossmwm I w L / N N a z a a o fl \3 1 I 1 111 \ / .. .11 /\J\ m- N u z 3 k\\\/\ / m .1». . ‘ 2 NIH: _ . : NaaNu Naasz Nazmzo 6 7 . a ----y a Z \\ llllll A1 \ / N H N M \\ a a a a \ \ \ \ :R I : :Naz :Naa 77 Third, the —NF2 out-of-plane wagging mode (v9), pre- dicted to be at 190 cm'l, is assigned to the first mode which is estimated here at 150 cm'l. The -NF2 rocking l motion (v6), observed at 258 cm’ in the infrared region, may be assigned to the second mode estimated here at 220 cm“. With the structural parameters new accurately determined a detailed force field calculation of this compound is needed to yield additional information about its electronic structure. Finally, the discussion above of the dipole moments of the PFZX and NFZX molecules suggests that the dipole moment of NFZCA should be quite close to that of NF,. A microwave study of NFZCA is thus suggested to test this point by comparing its dipole moment with that of PFZCA (53) already completed in this laboratory. VII. BUTYRALDEHYDE 7.1 Introduction Many studies of rotational isomerism in acyl compounds by nuclear magnetic resonance spectroscopy (76), electron- diffraction (77), infrared spectrosc0py (78), and microwave techniques have been published recently. These include microwave studies of propionyl compounds such as propional- dehyde (49), propionyl fluoride (79), 2-butanone (50), and ethyl formate (80). For the butyryl and higher aliphatic series, asymmetry in the potential functions for the rotations about the a C-C and B C-C bonds should, in principle, give rise to at least four and probably five different rotamers. These isomers, given in Figures 12a and 12b for butyraldehyde, can be generated from rotations about these two bonds by 00 and To, and are designated as the trans-trans, trans-gauche, gauche-trans, gauche-gauche-l and gauche-gauche-Z isomers. Studies of butyraldehyde by infrared spectroscopy (81) and NMR spectroscopy (76, 82) have confirmed the presence of the trans-trans and trans-gauche rotamers. The enthalpy difference for the CH0 torsional isomers has been found to be about 1 kcal/mole. 78 79 The present investigation was undertaken to examine the microwave spectra of the isomers of butyraldehyde and to determine their relative populations. Transitions in the trans—trans and gauche-trans rotamers have been assigned and dipole moments from the Stark effect measurements have also been determined. An attempt to assign spectra to other species was made but was unsuccessful. Since it did not appear that further searching would lead to success in a reasonable amount of time, the study was abandoned at this point. 7.2 Spectra for trans-trans andgauche-trans Butyraldehyde An initial spectrum for trans-trans form of butyraldehyde was calculated from an assumed structure with all heavy atoms in a plane of symmetry, and with all the C-C and C-H bonds staggered. Spectra for other rotamers were then calculated by appropriate rotations of 0 and r, as shown in Figures 12a and 12b. The observed spectrum for this compound, though not as dense as that for cyclopropyl methyl ketone,is, nevertheless, relatively rich. The initial search for an assignment was centered on the trans-trans species, for which strong a- and b-type transitions were expected. The transition 1 -1 01 10 of the Q-branch series, JOJ-J1 J 1, in the ground—vibrational state was first identified through its extremely fast high 80 trans-trans O /, CI 0 r e = 0° T = 0° b trans-gauche yo 0 = 0° T - 130° Figure 12a. Projections of trans-trans and trans-gauche Butyraldehyde in the ab Plane. H and C atoms are unmarked. 81 ’0 1 a 0=lOO° // 1=0° gauche-trans r=l30° gauche-gauche-l Figure 12b. Projections of gauche-trans and gauche-gauche-l Butyraldehyde in the ab Plane. H and C atoms are unmarked. 82 frequency Stark effect, and was soon followed by other members of the series. A Q-branch plot for this series from which K and (A—C)/2 may be determined is shown in Figure 13. The assignment was completed by identifying several low J a-type, R-branch transitions. The frequencies of transitions and the rotational parameters in the ground state and two vibration- ally excited states for the trans-trans species are listed in Tables 30 and 31, respectively. From a comparison of the change of the out-of—plane principal second moment of trans- trans butyraldehyde with that for some related molecules, it is seen from Table 32 that the excited states must be from the ethyl torsional mode, rather than from the CH0 torsion as was initially expected. With the trans-trans species thus assigned, attention was then directed toward the other dominant species, the trans-gauche species observed by infrared and NMR studies. After an unsuccessful search for transitions having the predicted Stark effect, the attempt to assign transitions in this species was not pursued further. In the meantime, several intense transitions with resolvable Stark effect, not assigned to the trans-trans species, were subsequently fit and assigned to the J1 J 1 - J2 J 2 series of the gauche- trans species. A Q-branch plot of this series, given in Figure 14, shows that the contribution from centrifugal dis— tortion is quite appreciable, in contrast to the situation in Figure 13 for the trans-trans species. Another Q-branch series, J - J , was also correctly predicted and 0,] 1,J-1 ' 83 MHz “ 0.1 "1,.1-1 __) l l -O.956 -0.957 K Figure 13. (A-C)/2 vs. K Plot for trans-trans Butyraldehyde. 1,.1-1 2,J-2 It" , 1 1 -O.763 -0.764 K Figure 14. (A-C)/2 vs. K Plot for gauche-trans Butyraldehyde. 84 .sz mc.oN ma mefiocozcepw wo>pomno :H Sucwmppeoca .sz :N .CCwuumcfihou GUCNCOmmH QHDSOU \AD UmEhwaOU U .Hm eHan cw co>flu mew museumcou Hmcoflumuom .NIZ cfi moflocesoepw weumHSUHmo mscfls we>gemno ohm momeENCehmo a“ mosam> nfim.o-vme.eweefi flHN.oVoN.HKmmH afleK.oVem.ermN efiao.onm.mmeeN flNo.o-Vom.NHKmH mao.o-va.mNaeH flmo.o-umfi.womefi Amo.o-vefi.mmKNH ANN.O-VNN.H64NH U n mKO.C-va.HMKaN Aflo.o-va.mammH afloo.ovow.eowaN mNo.o-Vea.KmeeH flmo.ovom.aommfi floH.oVaN.mNmeH neN.ovfim.mmeNH Amo.o-vmm.moaNH mae.o-vam.eNeNH mmo.o-vmm.oaaefi ACH.ovNe.NNNmH flofi.o-VoK.NommN AHH.OVKo.amNeH aflo.ovafi.maoefi noc.ovem.efiaeN fleo.ova.mNeeH Ame.o-VHK.NKomH Amo.c-vaa.oaKNH n . w wflhfnoh mfiOTmoO m~©i1~m Jamimom mfiV+3ofi NamfauN momTNoN ~HN+NoN oHHi Ho.H Nu> Hu> ou> cofluwmcmhh .mowxcewfimpzusm mcmuu-mcmpu New mcofiuflmcmph we>pemno .om manme 85 Table 31. Observed Rotational Parameters8 for trans—trans Butyraldehyde. Parameter v=0 v=1 v=2 . A(MHz) 15069.42 14909.17 14748.74 B 2555.98 2555.11 2554.62 C 2278.61 2282.59 2287.24 Ia(u-A2) 33.5365 33.8970 34.2657 1b 197.7231 197.7903 197.8281 1C 221.7915 221.4045 220.9548 Paab 192.9891 192.6489 192.2586 Pbb 28.8024 28.7556 28.6962 PCC 4.7341 5.1414 5.5695 K -0.956630 -0.956834 -0.957086 a Estimated uncertainty in rotational constants is $0.05 MHz. b 02 P = %(I + I - I ), etc., unit in U°A 33 b C a .mw .wem 86 0 .0m .uam my .Km .eam 0 .ae .wem n m 00 Hy» u psoEoE wcooem ocmfla-wo-p:o n m . <.: CH No Q -Nzomzu mmN.OH Nae.m eexeeefiaaxeam meaca-meaep -Nzomzo Nam.mfi wae.NN maeaesm-N-mN6 -Nzomzo mwe.a mam.e eNNNEAOE fixeNm-mH6 -ozu ooo.ov mNm.mm oowxgopamxonsmuHzm0No0Huxu-mflu -ozo 0mm.e NNH.e eaeseeefiaeoNaoea-mwe 00 SN aN eeaN :ofiumhnfl> Hn> cu> ofisoefioz oo .mefisuofioz wopmfiom mam ewxgewflmpxusm mcmpu-m:mpp New mesflm> m o4 mo comflymmsou .Nm ofipmh 87 measured. Assignments of the P-branch transitions were made through double resonance connections. The observed transitions and the resulting rotational parameters for this species are presented in Tables 33 and 34. The observed dihedral angle was determined from plots of the calculated rotational con- stants, A, B, and C versus 8, as shown in Figure 15. Transitions from the first excited state, presumably also of the ethyl torsional mode, were also observed and fit, and together with the rotational parameters, are given in Tables 33 and 34, respectively. 7.3 Dipole Moments Results of accurate Stark effect measurements for the trans-trans and gauche-trans rotamers are summarized in Table 35. The weak dependence of the overall dipole moment of this compound upon internal rotation can be noted from the values of D for the two species, which are essentially identical and comparable to those of other carbonyl compounds. 88 Table 33. Observed Transitions for gauche-trans Butyraldehydea. Transition v=0 v=1 b 000 4 111 ll437.08(-0.01) 11489.71(0.l7) 101 4 212 17294.29(0.01) 17320.13(-0.03) 111 4 220 29176.33(0.00) 29333.55(0.09) 110 4 221 28454.00(0.07) 28637.97(—0.05) 211 4 220 l4821.22(0.00) 15118.82(0.04) 212 4 221 16739.69(0.00) l6976.73(—0.06) 212 4 303 14985.31(-0.44) 14741.64(-0.65) 312 4 321 14116.70(O.19) 14424.08(0.06) 303 4 4H 28174.00(-0.21) 28167 45(0.53) 404 4 413 9271.10(-0.21) 9197.08(-0.53) 413 4 422 13474.22(0.52) 13772.90(0.64) 41“ 4 505 28785.95(-1.51)b 28481.02(—1.92)b 505 4 51“ 11713.65(-0.85)b 11533.15(-1.31)b 5“+ 4 523 l3100.63(0.68) 13359.58(0.84) 606 4 615 14838.63(-2.76) 14532.60(—2.94) 615 4 62“ 13184.63(0.51) 13366.9l(0.64) 716 4 725 l3875.47(—0.42) 13938.59(-0.30) 817 4 826 15285.32(—2.68) 15188.51(-2.72) 918 4 927 17488 12(-6.61) l7l93.69(-6.95) a in MHz. Uncertainty in observed frequencies is $0.05 MHz. Confirmed by double resonance connections. 89 Table 34. Observed Rotational Parameters3 for gauche-trans Butyraldehyde. Parameter v=0 v=1 A(MHz) 8508.49 8574.24 8 3588.78 3553.67 0 2928.60 2915.31 0 Ia(u-A2) 59.3967 58.9412 1b 140.8212 142.2124 1 172.5659 173.3525 C paa 126.9952 128.3118 pbb 45.5707 45.0407 0 13.8260 13.9005 CC K -0 763372 -0.774387 3 Estimated uncertainty in rotational constants is $0.05 MHz. b o 2 = L - ° ' . Paa 4(Ih + Ic Ia), etc., unit in u A . 9O .mucmpmcou Hmcoflumpom ©e>pomnc wcm peumfisofimu coozpen cemwpmoEou m Eopm oexgewfimgxusm mcmpp-ono:wm New oHu2< Hmpwonflm ecu we CoNumcHEAmqu ® OQNH COOS com ace . _ _ _ _ d _ 4 TI 1‘- _ In I l .mN wesNfia sz 91 .NmEo\ >xv\N:2 cw wepmfizofimo mscflE J 0 Cmtehomfio Ohm mQWOLHCOPCQ CH mozflm> a N mefiooom mcmpu-e:o:mm N mewooom mcmwu-mcmpp a Nmo.oaoom.N "Na a No.oNMN.o "ea mmm.ouaea a aoo.oa4am.N n a Q mmo.oummm.m an: mwm.vu an a moo.cucqo.m u a ”mofi.w N91 N a mo.o-aw.o "Na 4ma.ou N: N 4 N a ecc.OANoe.H u a ”mam.Nu a N mefiooom m:mpu-o:o:mm mowooom mcmpu-mcmpu n60.0vfim.aN m 446 1 426 ANH.O-Vwm.NH 4 .Na + ale fl4o.o-vmm.4m m 44m + 41m Amm.o-v4m.e4 e 440 + ace flNo.o-VwN.NN 4 44m + 44m ANH.O-me.4H 4 426 4 ace A4H.OVNm.NH m 44m + 42m ”KN.O-VN4.N0 4 444 + .44 . t - n w t w flmN.oV44.me m 42m 44m flfio.o Vmw.m- m 42 44 flmN.o-VaN.mm 4 41m + 44m flmfi.ovme.ma m Nam + 44m m4N.o-vem.HN m 42m + 44m flec.o-vmm.NMH N 44N + NON enmeov me\>e z eofleflaeeae ahaLov Ne\>e z eoHNNmeNee .ewkaowampxusm New mucoEoZ oaoofio mom mucofleflmmeou Memum .mm oHQwH 92 7.4 Discussion A comparison of the relative intensities of the tran- sitions suggests that the trans-trans and gauche-trans rotamers are present in approximately equal amounts. In view of the relatively low barrier height predicted for the ’ CHZ-CH2 rotation this finding is not surprising. Since f“ transitions from the methyl torsional mode were not assigned, the barriers to internal rotation of the methyl group in these two species could not be determined. However, the values for the barrier heights can be expected to be around 3 kcal/mole, similar to those reported for propionaldehyde, ethyl methyl ketone, and l-butene (83). A point of interest to be noted here is that for the gauche-trans species, rotation of the ethyl group out of the symmetry plane has resulted in an increased deviation from the rigid—rotor pattern, an effect which was also observed in ethyl formate and l-butene. REFERENCES 10. 11. 12. 13. REFERENCES C. E. Cleeton and N. H. Williams, Phys. Rev. 45, 234(1934). 22' Gordy, Rev. Mod. Phys. 20, 668(1948). E B. Wilson, Jr., Ann. Rev. Phys. Chem. 2, 151(1951); B. P. Dailey, ibid. 4, 425(1953); P. J. Myers and w. D. Gwinn, ibid. 57 385(1954); 0. R. Lide, Jr., ibid. 13’ 225(1964); W. HT Flygare, ibid. 18, 325(1967); Y. Merino and E. Hirota, ibid. 20, 139(1969); H. D. Rudolph, ibid. 31, 73(19707. W. Gordy, W. V. Smith, and R. F. Trambarulo, Microwave Spectroscopy, Wiley, New York, 1953. M. W. P. Strandberg, Microwave Spectroscopy, Mathuen, London, 1954. C. H. Townes and A. L. Schawlow, Microwave Spectroscopy, McGraw Hill, New York, 1955. T. M. Sugden and C. N. Kenny, Microwave Spectroscopy of Gases, Van Nostrand, Princeton, N. J., 1965. J. E. Wollrab, Rotational Spectra 8 Molecular Structure, Academic Press, New York, 1967. D. R. Lide, Jr., " Adv. in Anal. Chem. and Instr.” Vol. 5, 235, Eds. C. N. Reilly and F. W. McLafferty, Wiley, New York, 1966; E. B. Wilson, Jr., Science, 192, 59(1968). S. C. Wang, Phys. Rev. 34, 243(1929). H. B. G. Casimir, Rotation of a Rigid Body in Quantum Mechanics, Wolters, The Hague, 1931. B. S. Ray, Z. Physik, 78, 74(1932). G. W. King, R. M. Hainer, and P. C. Cress, J. Chem. Phys. _2, 210(1944). 93 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 94 H. Schwendeman, J. Mol. Specty. p, 301(1961). H. Van Vleck, Phys. Rev. 44, 467(1929). 5. Mulliken, Phys. Rev. 59, 873(1941). R J R E. C. Kemble, Fundamental Principles of Quantum Mechanics, Dover, New YorR, 1958. S. Golden and E. B. Wilson, Jr., J. Chem. Phys. 19, 669(1948). D. R. Herschbach and V. W. Laurie, J. Chem. Phys. 37, 1668(1962). V. W. Laurie and D. R. Herschbach, J. Chem. Phys. 41, 1687(1962). D. R. Herschbach and V. W. Laurie, J. Chem. Phys. 40, 3142(1964). T. Oka, J. Phys. Soc.(Japan) 15, 2274(1960). Y. Merino, K. Kuchitsu, and T. Oka, J. Chem. Phys. 49. 1108(1962). T. Oka and Y. Merino, J. Mol. Specty. p, 472(1961); ibid. 4, 9(1962): 44, 349(1963). J. Kraitchman, Am. J. Phys. 24, 17(1953). C. C. Costain, J. Chem. Phys. 49, 864(1958). K. Kuchitsu, T. Fukuyama, and Y. Merino, J. Mol. Str. 4, 463(1968). B. Colthup, Spectrochim. Acta 344, 2167(1967). H. Hughes and E. B. Wilson, Jr., Phys. Rev. ll, 562(1947). Battaglia, et a1, Nuovo Cimento l4, 1076(1959). Yajima and K. Shimoda, J. Phys. Soc.(Japan) 15, 1668(1960). Yajima, ibid. lg, 1709(1961); ibid. 14, 1594(1961). P. Cox, G. W. Flynn, and E. B. Wilson, Jr., N R A A. Battaglia, Arch. Sci.(Geneva) ii, Fasc 17(1960). T T A J Chem. Phys. 4;, 3094(1965). 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 95 R. C. Woods 111, A. M. Ronn, and E. B. Wilson, Jr., Rev. Sci. Inst. 41, 927(1966) J. Muenter, J. Chem. Phys. 48, 4544(1968). H. N. Volltrauer and R. H. Schwendeman, J. Chem. Phys. £4, 260(1971). H. N. Volltrauer and R. H. Schwendeman, J. Chem. Phys. g4, 268(1971). L. S. Bartell, J. P. Guillory and A. T. Parks, J. Phys. Chem. pg, 3043(1965). J. D. Swalen and C. C. Costain, J. Chem. Phys. 44, 1562(1959). R. H. Schwendeman, G. D. Jacob, and T. M. Krigas, J Chem. Phys. 49, 1022(1964). D. R. Herschbach, J. Chem. Phys. 44, 91(1959). R. W. Kilb, C. C. Lin, and E. B. Wilson, Jr., J. Chem. Phys. 29, 1695(1957). L. C. Krisher, J. Chem. Phys. 44, 1237(1960). K. W. Sinnott.J. Chem. Phys. 34, 851(1961). L. C. Krisher and E. B. Wilson, Jr., J. Chem. Phys. 41, 882(1959). L. Pierce and L. C. Krisher, J. Chem. Phys. 44, 875(1959). P. D. Foster, V. M. Rae, and R. F. Curl, Jr., J. Chem. Phys. 44, 1064(1965). S. Butcher and E. B. Wilson, Jr., J. Chem. Phys. S. 49, 1671(1964). 1 Pierce, et al, J. Mol. Specty. 43, 449(1969). E. B. Wilson, Jr., Proc. Natl. Acad. Sci. 4§,8ll(1957); Adv. Chem. Phys. E, 367(1959). R. L. Kuczkowski, J. Am. Chem. Soc. 92, 1705(1968). A. Brittain, J. Smith, and R. H. Schwendeman, to be published. A. Brittain, J. Smith, P. L. Lee, K. Cohn, and R. H. Schwendeman, to be published. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 96 K. Emerson and D. Britten, Acta Cryst. 11, 1134(1964). F. A. Miller, et al, Spectrochim. Acta 21, 775(1964). W. Rudolph, R. C. Taylor, and R. W. Parry, Am. Chem. Soc. fig, 3729(1966). CNN J. Barber, J. Chem. Soc.(London), 79(1943). . C. Costain and B. P. Stoicheff, J. Chem. Phys. 0, 777(1959). H V. W. Laurie, J. Chem. Phys. 41, 1500(1959). C 3 L. Pierce, R. Nelson, and C. Thomas, J. Chem. Phys. 3423(1965). R. G. Lett and W. H. Flygare, J. Chem. Phys. 41, 4730(1967). J. K. Tyler, L. P. Thomas, and J. Sheridan, Proc. Chem. Soc. p 155(1959). D. R. Lide, Jr., J. Mol. Specty. g, 142(1962). D. J. Millen, G. Topping, and D. R. Lide, Jr., J. Mol. Specty. g, 153(1962). W. H. Fletcher and F. B. Brown, J. Chem. Phys. égq 2478(1963). E. M. Pepov and I. P. Yakovlev, Zhurnal Strukurnoi Khimii 9, 54(1968). J. N. Macdonald, D. Taylor, and J. K. Tyler, J. Mol. Specty. 29, 285(1968). T. R. Jones and N. Shepphard, Chem. Cemm., 715(1970). M. D. Meyers and S. Frank, Inorg. Chem. E, 1455(1966). A. Murray and D. L. Williams, Organic Synthesis With Isotopes, Part I, 589; Part II, 1716, InterscTence, New York, 1958. S. N.Gbosh, R. Trambarulo, and W. Gordy, J. Chem. Phys. 24, 308(1953). D. R. Lide, Jr., J. Chem. Phys. gg, 456(1963). L. Pierce, R. G. Hayes, and J. F. Beeson, J. Chem. Phys. 49, 4352(1967). 75. 76. 77. 78. 79. 80. 81. 82. 83. 97 R. Lide, Jr., and D. E. Mann, J. Chem. Phys. , 572(1958). J. Karabatsos and N. Hsi, J. Am. Chem. Soc. 7, 2864(1965). Romers and J. E. G, Creutzberg, Rec. tray. chim. , 331(1956). E. Katon and W. R. Feairheller, Jr., J. Chem. Phys. 4, 144(1966). 5L. lute loom INC 0 U]. o o L. Stiefvator and E. B. Wilson, Jr., J. Chem. Phys. , 5385(1969). M. Riveros and E. B. Wilson, Jr., J. Chem. Phys. , 4605(1967). 919913.01 Sbrana and V. Schettino, J. Mol. Specty. 44, 100(1970). J. M. Lehn and J. J. Riehl, J. Chim. Phys. 62, 573(1965). S. Kondo, E. Hirota, and Y. Merino, J. Mol. Specty. g_, 4711(1968). APPENDIX W APPENDIX A STARK EFFECT CALIBRATION OF OCS- The second order Stark shift of the 0+1 transition of f‘ OCS is given by Av = 4(u2 cz)/15Bh where u, e, B, and h have the usual physical meaning. After rearrangement and insertion of the appropriate Equation (A-l) is reduced to Av/c2 = 5.68463 MHz/(kvolt/cm)2 = m theor Experimentally, the unit of m is in MHz/kvoltz; theor hence, the ratio m has dimension cmz. But theor/mexp convenience, the convers1on factor , mexp/mtheor’ is the EIGVALS program. i [It a (A-1) constants, for used in The results of a calibration of the "R" sample cell, presented in Table 36 and Figure 16, yielded a value 4.5476 for the conversion factor. 98 of lilll I'll 99 Table 36. Stark Shift for the 0+1 Transition of OCS. Volt(DC) yo1t2x10'“ Av(MHz) 0.0 0.00 0.0 100.0 1.00 0.272 160.0 2.56 0.680 200.0 4.00 1.056 260.0 6.76 1.771 300.0 9.00 2.355 360.0 12.96 3.378 400.0 16.00 4.165 460.0 21.16 5.496 500.0 25.00 6.482 600.0 36.00 9.328 660.0 43.36 11.268 Observed slopea = MHz/(kvolt)2 = 25.85153 Conversion factor = 25.85135/5.68463 = 4.5476 a From a least squares fit. 100 0.34 .meo we eofleflmeaee H+o ace 264 Nosem 4445a JICHXNPHO> 0.0m o.om o.oH .3 338.5 NH N $2 l .l 1. "Eli 1"] i II. A II 11114 El... It‘ll m: a dilll‘ TV [lililv‘llil Lilia .