| l ‘H i 1 » H | CARBON HALOGEN DISFANCES EN WTHYL CHLORIDE AND METHYL BROMIDE Than:- {0" Ha th'no 0‘ M. 5. MICEKAN STATE HNIYEKSITY Jerry Daniel Kelly. 1964 ' LIB R A R Y Michigan State University W’ MICHIGAN STATE UNIVERSITY EAST LANSING. MICHIGAN ABSTRACT CARBON HALOGEN DISTANCES IN METHYL CHLORIDE AND METHYL BROMIDE by Jerry Daniel Kelly A brief history of the methyl halides is presented, followed by an introduction to isotope effects and the concept of non—bonded inter— The theory of rotational Spectra is discussed including an The 'actions. explanation of centrifugal distortion and quadrupole coupling. Kraitchman equations as applied to symmetric top molecules and their application in determining molecular parameters is discussed. A descrip- tion of the microwave spectrometer used in this investigation is included. The method of preparation of CD3Cl and CD38r is described. The ground state rotational constants of 13CD335C1, 13CD337C1, 13CD37QBr and 13CDsalBr have been determined. The transition frequencies of 12CD335C1, 12CD337C1, 12CD37QBr, 12CD3alBr have been remeasured and the rotational From the rotational constants of the deuterated constants redetermined. methyl halides, the C-Cl distance in CD3Cl and the C-Br distance in CD3Br were determined to be 1.780h 3 and 1.9377 3 respectively. A literature Search was made to determine the best rotational constants of 12CH335Cl, 12CH337C1, 13CH335Cl, 13CH337C1, and 12CH379Br, 12CH3318r, 13CH379Br, 13CH3518r measured by previous researchers. Appreciable discrepencies in the rotational constant of 13CH337C1 in two different papers required remeasurement of the transition frequencies to determine the best value. Calculations of the carbon halogen internuclear dis- tance allowed a comparison with the deuterated species. It was deter- mined that deuterium substitution reduces the carbon halogen bond length by 11 x 10—4 X in methyl bromide and 8 x 10'4 X in methyl chloride. CARBON HALOGEN DISTANCES IN METHYL CHLORIDE AND METHYL BROMIDE By Jerry Daniel Kelly A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF‘SCIENCE Department of Chemistry 196D ACKNOWLEDGMENT The author wishes to express his gratitude for the assistance and encouragment of Professor R. H. Schwendeman throughout the course of this problem. ii To my wife Beverly iii TABLE OF CONTENTS Page I. HISTORICAL BACKGROUND . . . . . . . . . . . . . . . . . 1 II. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . 5 III. mom OF ROTATIONAL SPECTRA . . . '. . . . . . . . . . . 7 3.1 Energy of Rotation . . . . . . . . . . . . . . 7 3.2 Centrifugal Distortion . . . . . . . . . . 9 3.3 Quadrupole Coupling-—First Order . . . . . . . lO 3.h Quadrupole Coupling--Second Order . . . . . . . 11 3.5 Stark Eerct . ...... . . . . . . 12 3.6 The Determination of Atomic Coordinates . . . . 13 IV. DESCRIPTION OF‘THE MICROWAVE SPECTROMETER . . . . . . . 16 b.1 Introduction . . . . . . . . . . . . . . . . . 16 h.2 Reflex Klystron . . . . . . . . . . . . . . . . 16 b.3 Wavemeter . . . . . . . . . ...... . . . . 18 h.h Sample Cell . . . . . . . . . . . . . . . . . . l8 h.5 Square Wave Generator . . . . . . . . . . . . . l8 b.6 Crystal . . . . . . . . . . . . . . . . . . l9 h.7 Sampling Technique . . . . . . . . . . . . . . l9 h.8 Frequency Measurements ..... . . . . . . . 20 V. DETERMINATION OF STRUCTURE OF METHYL CHLORIDE AND METHYL BROMIDE . . . . . . . . . . ...... 22 5.1 Introduction . . . . . . . . . . . . . . . . . 22 5.2 Sample Preparation . . . . . . . . . . . . . . 22 5.3 EXperimental Procedure . . . . . . . . . . . . 2b 5.h Examination of the Spectra ...... . . . . 27 5.5 Molecular Parameters . . . . . . . . . . . . . 27 5.6 Error Analysis . . . . . . . . . . . . . . . . 35 VI. DISCUSSION AND CONCLUSIONS . . . . . . . . . . . . . . . 37 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . LO iv LIST OP‘TABLES Table Page I. Methyl chloride and methyl bromide parameters . . . . . 2 II. Summary of methyl chloride and methyl bromide data . . A III. Centrifugal distortion constants of methyl halides . . 9 IV} Rotational constants of deuterated methyl chloride and methyl bromide from known parameters . . . . . . . 25 V. Approximation of hypothetical unsplit frequencies . . . 26 VI. Calculated vs. experimental frequencies of deuterated methyl chloEIde and methyl bromide . . . . . . . . . . . 26 VII. Experimentally determined constants of deuterated methyl chloride . . . . . . . . . . . . . . . . . . . . 28 VIII. Experimentally determined constants of deuterated methyl bromide . . . . . . . . . . . . . . . . . 29 IX. Carbon-halogen bond distances of deuterated methyl bromide and methyl chloride . . . . . . . . . . . 31 X. Best values of methyl chloride and methyl bromide from references . . . . . . . . . . . . . . . . . . . 33 XI. Carbon—halogen bond distances of methyl chloride and methyl bromide . . . . . . . . . . . . . . . . . . . . . . . . 3h XII. Constants in the error analysis . . . . . . . . . . . . 36 XIII. Carbon-halogen bond distances in methyl chloride and methyl bromide computed assuming 0.00005 3 decrease in the C-X bond distance upon substitution of heavier isotopic species . . . . . . . . . . . . . . . . . . . 39 LIST OF FIGURES Figure Page 1. Diagram of the microwave spectrometer . . . . . . . . . l7 I. HISTORICAL BACKGROUND Immediately following World War II vast amounts of 1.25 cm, radar equipment were made available to research laboratofiies in educational institutions. The availability of this equipment at low cost made it possible for many research workers to investigate microwave absorption in gases. It had been realized since the work of Cleeton and Williams(1) in 1933 that ammonia absorbs radiation in the microwave region. This absorption was attributed to transitions between energy levels split by the 6 Irilocalories potential barrier hindering the inversion of ammonia. One of the early investigators, W. D. Hershberger(2), selected a number of gases which he believed to be most likely to absorb micro- waves. The requirements he believed to be necessary for absorption were that the molecule have a dipole moment, and that there exist the possibility for hindered rotation of a methyl group with respect to some other group about a single carbon-carbon bond. It was suspected that in the ethyl halides, transitions between energy levels Split by hindered rotation should also lie in the microwave region. This hypo- thesis involving hindered motion was found to be inadequate when it was learned that the methyl halides absorbed microwave radiation. Absorp— tion in this region was then attributed to the rotational motion of the molecule in the lowest vibrational states. Previously, structural parameters of the methyl halides had been determined by infrared and electron diffraction experiments and discrep- ancies existed between the two methods(3). The greater precision of 2 microwave Spectroscopy combined with existing infrared data reduced the discrepancies considerably. The first frequency measurements of the rotational spectra of the methyl halides were made by Gordy, Simmons and Smith in 19h7(h). Their work indicated the existence of hyperfine structure due to the inter— action of the Spinning nucleus with the surrounding charge distribution of the molecule. This interaction, known as quadrupole coupling, had been predicted by Kellogg, Rabi, Ramsey and Zacharias in 19h0(5). The values of the moments of inertia Ib and the quadrupole coupling constants qu found by Gordy, it. 21. are listed in Table I. The internuclear distances were determined using moments of inertia Ia evaluated from infrared data. Table I. Methyl chloride and methyl bromide parameters.a Compound Ib(amu A?) qu(Mc) C-X (A) 0(cm) l2c1+13,35c1 38.016 -69 1.79 .055 x io‘z‘L 12CH337Cl 38.558 -56 .015 x 10’24 ”Gag/9131* 52.716 573 .17 x 10"24 1.91; lzcwsaler 52.957 185 .m x 10‘24 aReference h. In 19h? Gordy, Simmons and Smith (6,?) developed an instrument for multiplying stable 10 mc/Sec frequencies. The 10 mc/sec signal was monitored by radio station NNV maintained by the National Bureau of Standards. This new measuring device enabled them to measure frequencies to seven significant figures compared to a maximum.of four obtained previously with the cavity wave meter. In addition to more precise measurements using the frequency multiplier, it was 3 determined from discrepancies between measured and theoretical fre— quencies that second order quadrupole effects were significant in methyl bromide and methyl iodide. When, as in methyl iodide, the effect of quadrupole coupling was large compared to the pure rotational frequency, it was observed that second-order effects were of the magni- tude of several megacycleS/Sec. Improved measuring techniques plus consideration of first and second-order quadrupole effects gave the results listed in Table II(6,7). The work of Gordy, Simmons and Smith initiated an intense study of the methyl halides by microwave spectroscopy. Much of the work was published in the form of "Letters to the Editor" and provided only a minimum of information. Tables I, and II are a summary of all the available data on methyl bromide and methyl chloride reported since 19h7. A Table II. Summahy of methyl chloride and methyl bromide data Compound Rgiig- O(Mc/sec)a B(Mc/Sec) qu Ib(amu A2)b 120H335Cl 6 26585.77 —75.13 38.022 7 38.026 8 -7505 10 13292.95 11 ’7b077 l2 —7bob-O 1b 13292.86 ~7h.9 lZCH337Cl 6 26176.37 -59.03 38.618 7 38.619 8 -58.03 10 13088.2h 38.612 13 26176.26 -58.93 130H335Cl 9 25592.83 39.199 16 12796.2 lsCHWCl 9 25179.52. 110. 11.9 16 12590.0 120H379Br 6 38272.h0 577.0 52.82h 7 52.827 no 9568.188 52.819 13 9568.20 577.15 In 9568.20 577.3 15 9568.37 577.3 120H3§isr 6 38126.97 h82.0 53.02u 7 53.028 10 9531.8h5 53.023 13 9531.8h u82.16 18 9531.82 h82.h 15 9532.20 h82.h 13CH379Br’ 17 9119.507 577.3 13CH3BlBr 17 9082.86 u82.h aJ a 0 __> 1 transition for CH3C1 and J = 1_a> 2 for eager. b I has been converted from the original unit of gm.cm2 to amu A2 for comparison with values given below. II. INTRODUCTION Isotope Effects and.N0n—bonded Interactions It has been established from theoretical considerations and eXper- imental examples that a reaction which involves the removal of a hydro- gen atom in its rate determining step will show a marked difference between the rate of removal of hydrogen and deuterium. The difference in rates, referred to as a primary isotope effect, is not necessarily restricted to hydrogen but is most prominent when hydrogen and its isotopes are involved. A proposed explanation of primary isot0pe ef- fects is that they are due to the difference in zero point energies of the isotopic species. The lower zero point energy of deuterium requires that a greater energy be expended to dissociate the atom than is re- quired for hydrogen. Therefore the rate of the reaction involving the isotopically substituted Species is less. Kinetic studies(l8) have indicated that the effect of substitution on reaction rates is not limited to an isotope directly involved in the reaction but may be extended to other sites. This has been attrib- uted to the effect of the change in the zero point energy on the ampli- tude of vibration. The unusually large change in the amplitude of vibration upon substitution of deuterium for hydrogen.makes hydrogen atoms the most important source of such secondary isotope effects. Bartell(l9) has proposed that deuterium substitution in ethane would lower the non-bonded repulsions thereby decreasing the carbon- carbon bond length. From calculations and electron diffraction exper- iments, he predicts and finds a decrease in the carbon—carbon bond 5 6 length of .OOA X with an estimated accuraqy of approximately .001 X. The precise frequency measurements obtainable with microwave spectroscopy make it the ideal method for determining small variations in the internuclear distances. From the frequenqy measurements the geometry of the molecule and subsequently the internuclear distances can be determined. The restriction that a molecule have a dipole moment prohibits the measurement of absorption frequencies of ethane by microwave radiation but the same principles should also apply to molecules with a dipole such as the methyl halides. Replacement of the hydrogen atoms by deuterium should reduce the non-bonded repulsions and decrease the carbon-halogen bond distance. The decrease in the carbon—halogen bond distance with substitution is expected to be less than the decrease in the carbon-carbon distance in ethane due to the number of replaceable hydrogen atoms involved, but it is believed that the effect would still be measureable. III. THEORY OF‘ROTATIONAL SPECTRA 3.1 Energy of Rotation For a discussion of their rotational energy molecules may be classi- fied according to the moments of inertia about their principal axes. The moment of inertia about an axis is defined as I = g’miri2 where ri is the perpendicular distance of mass mi from the axis. The principal axis system (a, b, c system) is located so that its origin is at the center of mass of the molecule and so that g? miaibI = .fi‘ miaici = 2? m.bici = 0. The three principal moments are called Ia, Ib and IC 1 and defined so that Ia 5 Ib _<_ 10. With this convention all molecules may be grouped as follows: Linear molecules H II 1...; H II 0 b c’ a IC = Ib + Ia Planar molecules Ia %'1b = IC Prolate symmetric top molecules Ia = Ib f’IC Oblate symmetric top molecules Ia f'Ib f'Ic Asymmetric top molecules Ia = Ib = IC Spherical top molecules. The restrictions for each type of molecule lead to different solutions for the rotational energy. The assumption that a molecule acts as a rigid rotor with no vibration-rotation interaction simplifies the energy expressions and yields a reasonably accurate solution. The potential energy of the molecule due to interactions with other molecules is con- sidered negligible because of the low pressures (10—4 atm.) used in microwave Spectroscopy. The classical expression for the energy of a rigid rotating body is w =(1/2)mr?-oa2 = IZwZ/ZI = PZ/ZI, where I is 8 the moment of inertia about the axis of rotation, P is the angular momentum, andIDJ is the angular velocity. The quantization of angular momentum requires that P2 = (h/er)2 J(J+D; therefore ha BUZI w a J(J+l) = hBJ(J + 1) where J can be zero or a positive integer and B is called a rotational constant. For diatomic molecules the axis of rotation is perpendicular to the axis of the molecule and h h I B = = "" C——§) BWZI Buzu r0 where u, the reduced mass, is $12; and r0 is the internuclear distance. 1 2 Molecules classified as symmetric t0pS include all structures having two equal moments of inertia. Methyl chloride is an example of a pro- late top, whereas chloroform is an oblate—top molecule. The more complex structure of the symmetric top requires the intro— duction of another quantum.number K where (5%)K is the projection of the angular momentum on the unique axis of the molecule, and K = 0, t l, i 2, . . . i J. The energy level expressions are as follows: W P we hBJ(J + l) + h(A — B)KZ for a prolate top molecule, hBJ(J + 1) + h(C - B)K2 for an oblate top molecule. The selection rules for allowed transitions for a symmetric top are AJ - 1, AK = 0. Application of these selection rules gives for the observed transition frequencieS\Q = 2B(J + l), where J is the quantum number of the lower energy state involved in the transition. 3.2 Centrifugal Distortion Centrifugal distortion is distortion of the molecule due to the centrifugal force of rotation. If the effects of centrifugal distor- tion are included the expression for the energy levels of a prolate top molecule becomes = ' _ 2 _ 2 2 _ 2 _ 4 wp hBJ(J+l) +1181 B)K hDJJ (J+l) hDJKJ(J+1)K hDKK where DJ, DJK, and DK are centrifugal distortion constants related in a complicated way to the geometry and electronic structure of the molecule. Energy level differences previously considered to be simple multiples of J+1 for the rigid rotor, tend to decrease slightly with an increas- ing value of J. The frequency of a transition for a symmetric top with centrifugal distortion is 9 J—->J+l = 2(J+l)(B - KZDJK) — hDJ(J+l)3 The DJ term results from the stretching which arises from the ro~ tation about the B axis, the DK term compensates for stretching due to rotation about the molecular axis, and DJK is the result of interaction of the two rotations. Numerical values of the stretching constants determined for the methyl halides are given in Table III. Table III. Centrifugal distortion constants of the methyl halidesa Compound BO(Mc/sec) DJ(KC/S€C) DJK(Kc/sec) 120113301 13292.86 18.1 198 12CH379Br 9568.20 9.9 128.3 12CH351Br 9531.82 9.7 127.11 aReference 21. 1O 3.3 Nuclear Quadrupole Hyperfine Structure-~First Order. In addition to the angular momentum due to rotation, some molecules have angular momentum resulting from the spin of one or more nuclei. The spinning nucleus being a charge distribution in motion produces a varying magnetic field which may interact with the rotation. As for any spinning system, the square of the angular momentum must be I(I+l)h2 where I, the spin quantum number, is integral or half integral. A spin of 1/2 leads to no interaction between the spin of the nucleus and the rotation of the molecule. For atoms with I greater than 1/2, the inter- action of the nuclear quadrupole moment with the surrounding charge distribution causes splitting of the rotational lines. The states are labeled by the total angular momentum quantum number P which has the values F = J+I, J+I-l.... J—I. The splitting of the rotational lines in symmetric top molecules with one quadrupole nucleus is proportional 2 to qu where Q is the quadrupole moment of the nucleus and q = <59 Ea E>Z av is the curvature of the electrostatic potential energy in the direction of the unique axis measured at the quadrupole nucleus. It has been shown by Townes and Dailey (21) that the field gradient along the unique axis is determined most strongly by the valence electrons of the atom in question. The electrons in closed shells have no effect due to spherical symmetry, assuming the distortion due to polarization is negligible. The spherical symmetry of the s orbitals eliminates any effect whereas the low penetrating power of d orbitals would appear to minimize any effect they might have in contributing to a field at the nucleus. AS indicated by Townes and Dailey, the only alternative is that the field gradient is produced by the p orbitals in the valence 11 Shell. Therefore an evaluation of the field gradient would be a measure of the p character of the bond. The first order energy expression for the levels Split by quadru- pole interactions of a symmetric top is W = W + W(l) where W is the rotational energy R Q R W61)" BQQ[3%I§':'T5' - I] f(IJF) rum = (7330833 - J_I+1)J(._I_+1)) c = F(F+l) - I(I+1) — J(J+l) The above equations developed by Bardeen and Townes(22) also apply to linear molecules which are a special case of symmetric t0ps with K = O. The frequency of a rotational transition is thus proportional to AWR + AWél). The change in rotational energy, AWR, is commonly refer- red to as the "hypothetical unsplit frequenqy". Since AWéJ) is often much smaller than AWR the transition occurs as a multiplet of lines approximately centered about the hypothetical unsplit frequenqy. The quadrupole energy expression for the asymmetric top is more complex, involving three field gradients instead of only one. A com- plete treatment for all types of molecules may be found in Townes and Schalow(23). 3.h Nuclear Quadrupole Hyperfine Structure--Second Order. When the hyperfine splitting becomes large relative to the hypo- thetical unsplit frequency, calculated first order quadrupole effects do not agree with eXperimentally determined values. This difference which may be as large as several megacycles per second may be evaluated by second order pertubation theory. Hr. __~fiF—_—F'_ 12 For a symmetric t0p the energy resulting from second order perturbation is of the form 2 I 11‘” = Z: I(IJP‘MF IHQI IJ FMFM Q w — w J'K' JK J'K' where M? is the orientation quantum number, HQ is the part of the Hamiltonian which represents the quadrupole energy and J‘ = J+l or J+2. The equations derived by evaluating the squared matrix elements are of the form(23) 2 ‘(I, J) F; MP iHQi I, J+l, F, MF)‘ = BquK 2 K2 [F F'I'I _ I 1+1 ~ J J+2 ]2 [ 81(21—-1)J(J+2)J [ 1 - W2] (iggméjm) L ) x (1+J +F+2) (J +F-I +1) , and 2 K1. J, F, MFI HQI I, 1+2, F, Mr). = [ 3208 ]2 (F+I+J+3)(F+I+J+2) 161(21-1)T2‘J® (2J+1) (2.7+5T ><.(J+I-F+2)(J+I-F+l)(J+F—I+2) x-(J+F-I+1)(I+F-J)(I+F-J—l) Tables enabling easy calculation of W62ihave been prepared(23). 3.5 Stark Effect In the absence of an external electric or magnetic field, the ro- tational energy levels are (2J+1) fold degenerate. The introduction of an electric field removes the degeneracy and allows 2J+l orientations of the total rotational angular momentum with reSpect to the direction of the applied field. The component of the total rotational angular 13 momentum in the direction of the field is designated.by'Mb(g%) where the quantum number MJ = 0, i 1, i 2, .... iJ. The selection rules for allowed transitions are determined by the geometry of the sample cell. If the applied electric field is parallel to the electric field of the microwave radiation, the selection rule is AM, = 0, but if the static electric field is perpendicular to the elec- tric field of the microwave radiation the selection rule is AMJ = i 1. The instrument at M.S.U. is constructed such that AMJ = 0. For J = 0 to J = 1 transitions only one stark lobe, M: = 0 to MJ = 0, was observed, as would be expected. 3.6 Determination of Atomic Coordinates. For prolate symmetric top molecules such as the methyl halides, . . h the A rotat1onal constant 15 8535; where Ia is the moment of inertia of the molecule about the symmetry axis. The B and C rotational con— h stants are both equal to 3535; where Ib is the moment of inertia about an axis perpendicular to the symmetry axis passing through the center of mass of the molecule. By definition, the moment of inertia about the b axis is lb = Z mMai2 + 012) 1 where for a prolate top the a axis is the molecular axis. For an atom located on the symmetry axis, the b and c coordinates are zero. The presently accepted method of determining bond distances from rotational constants is the substitution method(2h,3l) in which atomic coordinates are determined from differences in moments of inertia of different isotopically substituted molecules. 1b The equations for the coordinates of a substituted atom which is on the symmetry axis of a symmetric top molecule are particularly simple. Isotopic substitution of the sth atom on the symmetry axis of a sym- metric top molecule shifts the center of mass along the a axis by Aa and I changes the mass of the molecule by Ams. The moment of inertia lb, of the isot0pica1ly substituted molecule is and Ci’ ai and mi are the parameters of the original molecule. Substi— tuting into the eXpansion for ' = 2 2 2 2 lb 1? mi [(a1 + Aa) + C1 1 + Ams[(aS + Aa) + cS ] Since the sth atom is on the Symmetry axis cS = 0. The a; are measured from the center of mass of the substituted Species. Therefore I I _ = 2% miai — 0 RE nhlai + Aa) + Ams(aS + Aa) Since miai = 0 is the center of mass condition for the parent i Species and since :2, mi = M, the mass of the parent, 1 (M + Ams) Aa + AmsaS = 0 or Aa = - AmsaS M + AmS where M' = mass of the substituted Species. Substituting the expression for Aa into the expression for 1g and simplifying finally leads to b b A s I- where p, = w It should be emphasized that a3 is the coordinate of the substituted atom in the axis system of the unsubstituted or parent species. There- fore isotopic substitution at different sites gives the coordinates of the substituted atoms all in the same coordinate system. Thus the carbon halogen distance, for example, is simply aX - ac. In the methyl halides and deuterated methyl halides using the present experimental data and that previously given, there are a suf— ficient number of isotopic species to allow four different choices for the parent or unsubstituted species for each carbon-halogen distance to be determined. IV. DESCRIPTION OF THE MICROWAVE SPECTROMETER h.l Introduction Instrumentation for work in the microwave region is rather differ- ent from that used in all other regions. Microwave frequencies (1011 qycles/Sec) which lie between radio frequencies (106 cycles/Sec) and visible radiation (1015 qycles/Sec) require techniques similar to those used in generating radio waves rather than Optical methods. The basic microwave spectrometer consists of a klystron power supply and a klystron to produce monochromatic radiation, a wave guide which acts as a sample holder and a Silicon crystal to detect the sig- nal. The signal is amplified and displayed on a recorder or oscillo— scope. A block diagram of the components is shown in Figure l. 11. 2 Reflex Klystron The reflex klystron is an electron tube capable of modulating electron velocities. An electron beam emitted from the cathode is accelerated to pass through a resonant cavity. Further acceleration in the cavity depends on the phase of the oscillation of the fields in the cavity itself. The electrons emerge from the cavity into a drift region where bunching occurs as faster electrons tend to over— take slower ones. The reflector which is from -150 to -h00 volts negative with re- Spect to the cathode, returns the electrons to the cavity. If the re— turning electrons are properly bunched and in phase with the cavity oscillations which originally accelerated the electrons, the oscilla- tions are enhanced. Microwave radiation is conducted from the cavity 16 17 u. wkwcsofibwdm 02030.. 0. :4 a: jaw+c< _I lo>wuwww A Tofiamo I “MEN: um fl L meUmoEomO Jammy Lw+tu W85 FIIi— WCRV T E/mew goth:— X Do as. .Lorotswfl - QxOX L :2 06¢ 1 £63 11 radon , 8:2. 33X. .330me Lokotcow .gwwumnfifl To 0: o EVER sorowka 86550 E33003 500...); m 5.36 8965 xiowmw . \i t 32 9.69:») 8:). :86: we; «$885.83 m12m .I . c _ c :P e8:12.10:etheresoeo goedohmnol o ow to or are . o o: a _ €912.80 3.0.2.. o llmemfo a L 18 by means of a waveguide outlet. The oscillations tend to be highly monochromatic, and the klystron is tunable over a frequency range of approximately 20% of the center frequenqy. b.3 Wavemeter The wavemeter is a qylindrical cavity of which the dimensions can be varied by a micrometer screw. The cavity is coupled to the wave guide by a thin iris. The geometry of the wave meter is such that the fields in the desired cavity mode and in the wave guide are parallel. As the klystron sweeps through the frequency to which the wavemeter is tuned a fraction of the energy is absorbed which is indicated on the oscilloscope as a decrease in the microwave power. h.h Sample Cell The sample cell consists of asilver lined rectangular pipe (wave- guide) capable of transmitting microwave power. The cross sectional dimensions are 0.h0 in. x 0.90 in. Along each of the narrow sides of the sample cell is a Teflon tape which holds a silver septum insulated from and parallel to the wider side. Mica windows at each end, of the cell and a sample inlet system permit the waveguide to be used as the sample holder. h.5 Sguare Wave Generator The square wave generator is an electronic device capable of charging and discharging the septum in the waveguide to voltages up to 1200 volts in less than one microsecond at 5 microsecond intervals. The output of the generator is applied to the stark electrode in the waveguide. 19 The generator serves two purposes. The electric field modifies the absorption frequencies in a way which depends on the dipole moment of the sample, and second, the application of the square wave to the Stark electrode is a means of modulation of the absorption of the sample. The sample is alternately exposed to and isolated from the field at 5 microsecond intervals so that when the klystron frequency is identical to the sample absorption frequency, a small part of the microwave fre— quenqy will be modulated due to switching on and off the absorption of the sample by the changes in Stark field. 11.6 Crystal The crystal rectifier consists of a small silicon wafer with a fine pointed, tungsten wire contact. Contact resistance of the crystal is greater in one direction than the other. The current voltage char- acteristic is non linear near the origin so that any applied alternating current is rectified. Typical of this type of crystal is the IN26 used with K band wave guide. This type of non—linear device can be used for detection, for mixing two signals, and for frequenqy modulation. h.7 Samplinngechnique The sampling technique is determined.by the availability and the reactivity of thesample. For samples which do not react with glass, metal, or stopcock grease a vacuum system attached to the waveguide sample cell is satisfactory for introducing the sample to the system. The system is evacuated, then flooded with sample and partially re- evacuated to optimum pressure, generally about 10-4 atmospheres. If only a small amount of sample is available, it may be recovered from the cell. A reactive sample is allowed to diffuse into one end of the 20 wave guide while evacuating the waveguide from the other end. This allows for a continual sweeping out of any decomposed or reacted sample. h.8 Frequengy Measurements Approximate frequency measurements, within 10—20 megacycles, are obtained using the cavity wavemeter (section h.2) which is attached to the waveguide. When the klystron frequency is identical to the cavity resonance frequenqy, a small amount of energy is absorbed causing a dip on the Y axis of the oscilloSCOpe. Exact frequency measurements are referred to a Manson RD 1h0 high stability one-megaqycle/second, crystal—controlled oscillator with a frequency stability of one part in 108 per day. The crystal—controlled oscillator is standardized by beating its tenth harmonic against the 10 Mc/Sec carrier of radio station WWV from the National Bureau of Standards. The output of the one—megaqycle oscillator is connected to the input of a GertSjiAM-lA VHF interpolator where the Signal is amp- lified and multiplied. The output of the multiplier is mixed with the output of a l—2 megacycle oscillator (LFO). The frequency of the LFO is determined by using only those frequencies which give a stable Lissajous figure when referenced to one megacycle per second. These frequencies may be set to better than one qycle per second. The sum of the multiple of one megacycle and the LED frequency controls the output of the AM-lA interpolator. This output frequency which is be- tween 20 and h0 Mc/sec is then multiplied and mixed with a ten Mc/Sec signal in a second instrument, the GertxxiFM-LA UHF oscillator. The output of the Gertsch FM-hA is controlled by this multiplied and mixed signal. Finally the output of the FM-hA (500-1000 Mc/sec) is multiplied 21 and mixed with the microwave signal on a silcon crystal in the waveguide. The frequency of an absorption line is measured by the difference be— tween the unknown absorption frequency and a multiple of the known frequency of the Gertsch UHF oscillator. The frequenqy difference is measured by either a Haflicrafters SX—62—A or Collins 51J-h receiver. The Collins is generally used for fine measurements because of the ease of reading the dial calibrated to .001 megacycle. The Hallicrafters is tuned much more rapidly and is used for determining initial locations. The exact frequenqy of an absorption line is determined by moving the frequency difference marker until it is centered on the absorption line. Actually two frequency markers are generated corresponding to positive or negative frequenqy differences. The positive or high fre— quency marker is being observed if moving the receiver dial to a higher frequency causes the marker to shift to higher frequencies. With the high frequency marker centered on the absorption line the receiver read— ing is added to the standard frequency. When the low frequency marker is used the receiver reading is subtracted from the standard frequenqy. ~A time delay between the 100 kc signal and the beat marker reaching the oscillosc0pe requires that measurements be made while sweeping the field in both directions to average out the frequenqy discrepancies. This method is capable of giving frequency measurements with an absolute precision of 0.01 Mc/sec at 30,000 Mc/sec. V. DETERMINATION OF STRUCTURE OF‘METHYL CHLORIDE AND METHYL BROMIDE 5.1 Introduction In undertaking the study of the structural parameters of the methyl halides the following steps were taken: 1. Preparation of the deuterated carbon—l3 methyl halides. 2. Calculation of the rotational Spectra using an approximate structure. 3. Assignment of absorption lines. A. Determination of quadrupole coupling effects, rotational con— stants and moments of inertia. 5. Calculation of internuclear distances from moments of inertia. 6. Determination of error. The purpose of the research was to determine the effect of deuter— ation on the carbon halogen bond distance. Consequently a comparison was made of the results of the present experimental work with those of previous research on the methyl halides. 5.2 Sample Preparation The preparation of 13CD3X was accomplished in the following steps: 1. Conversion of Ba13C03 to 13C02 2. Reduction of 1300.2 to 13CD30H 3. Halogenation of 13003011. The procedure used in the preparation of 13CO2 was that of Nystrom and Yanko.(26) The apparatus consisted of a 100 ml round bottom flask attached to a vacuum manifold through a pressure compensating funnel and a plug of glass wool. Attached to the manifold was a gas collec— tion flask in a liquid nitrogen trap. 22 23 A mixture of 3.0 gms of 60% Balscos, 3.5 ng BalZC03 and 12 ml of water was frozen out in the 100 ml round bottom flask. The system was evacuated and 25 ml of 70% H0104 was added to the Ba005 mixture through the pressure compensating funnel. The freezing bath was removed and the system was allowed to reach room temperature slowly. The generator was then closed off to prevent the transfer of the remaining water. The C02 was collected in the gas bulb at -l960C. Reduction of 13C0? to 13CD30H A modification of the procedure by Nystrom and Yanko(26) was used in the reduction of 13C02 with LiA1D4. A solution of 2.h gms of 99.8% LiAlD4 and 250 cc of diethyl carbi— tol was stirred in a 500 ml, 3 necked round bottom flask for 6 hours. A stoppered dropping funnel containing 60 gms of h—butyl carbitol and a thermometer were inserted in the three necked flask. The system was attached to the vacuum manifold and evacuated to 10 mm Hg. The previous- ly prepared lsC02 was admitted to the system and allowed to react by surface adsorption. After 12 hours the p—butyl carbitol was added through the dropping funnel. Six hours later the system was heated to 600C and occasionally partially evacuated through two traps cooled by liquid nitrogen. The CDSOH was collected in the traps and transferred to a gas collection bulb. Conversion of 13CD30H to CDSX The same general procedure was followed in the preparation of both 1300301 and 13CD3Br. A 50 m1 flask was attached to the vacuum manifold thru a cold finger attached to a liquid nitr0gen trap. The 50 ml bulb contained three grams 2h of PC15 (PBr3) which was frozen out. The system was evacuated and 0.3 ml of previously prepared l3CD30H was distilled onto the PC15 (PBr3). A dry ice acetone solution was prepared in the cold finger. All other traps were removed to maintain a continual refluxing of the 13CD30H. After 6 hours of refluxing, one liquid nitrogen bath was placed on the cold trap and one on a U tube in the vacuum manifold. The cold finger was allowed to warm to room temperature and the system was again evac— uated through the two cold traps. The samples were immediately trans- ferred to a gas collection bulb. All the bands in the infrared Spectra belonged to the methyl deuterides. 5.3 Experimental Procedure The molecular dimensions used to determine the approximate rota- tional constants were taken from Townes and Schawlow(23). The values used were compensated for the change in the H-C-H bond angles and C-H bond distances due to the substitution of deuterium for hydrogen. The shortening of the bond was assumed to be 0.009 A and the bond angle was assumed to increase by 12'. The parameters used were as follows: C-D (X) D—C—D angle C—X (8) 00301 1.10l 110031! 1.781 CD3Br 1.10l 111026' 1.939 The Cartesian coordinates of each atom were calculated using the carbon atom as the arbitrary origin. The translation of the coordinate system to the center of mass and the subsequent determination of the rotational constants were performed on the MISTIC digital computer at Michigan State University. The completed transformation resulted in the values of B for the deuterated methyl halides given in Table IV. 25 Table IV. Rotational constants of deuterated mgthyl chloride and methyl bromide from known parameters Compound B(Mc) Compound B(Mc) 12CD379Br 7716.77 12CD335Cl 108u3.b8 lacosalar 7683.36 120D337C1 10660.08 13CD37QBr 7lu2.96 130033501 10555.89 13CD331Br 7l09.22 13cos37c1 10370.56 aReference 23. Using the B values from Table IV the following relation was used to minimize the error in approximating the hypothetical unSplit frequencies: B13CD3X(SpeC) = Bl3CD3X(CalC) - B12CD3X(CalC) + BlZCD3X(ObS) where the BlZCD3X(ObS) were the rotational constants for the deuterated, C-l2 methyl halides as determined by Simmons and Goldstein(27). B(Mc 120033501 108ll.88 12CD337C1 10658.u3 12CD37913r 771h.57 12CD351Br 7681.23 The hypothetical unsplit frequencies (J = 0 -—> 1) were calculated as shown in Table V. The halogens in each case have a spin of 3/2 re— quiring a further calculation of the quadrupole coupling constant due to the interaction of the Spin of the halogen nuclei with the field gradient. From the equation developed by Bardeen and Townes(22), and the nuclear quadrupole coupling coefficient determined by Gordy, Simmons and Smith(6) the calculated spectroscopic frequencies are shown in Table VI. 26 Table V. Approximation of hypothetical unsplit frequencies Compound Th (MC/Sec) 130D335CL 21108.58 13CD33701 20739. 80 13CD379Br lh88l.52 13CD381Br lh81h.18 Table VI. Calculated vs. experimental frequencies of deuterated methyl chloride and methyl bromide. Compound F-+> F' ‘9 (1) a (unsglit) AWQ ~Y calc. ‘Yexp. 13CD33501 3/2 - 1/2 21108.58 + 18.60 21127.18 21128.2n 3/2.— 3/2 - lh.88 21093.70 2109h.67 3/2 — 5/2 + 3.72 21112.30 21113.33 130033701 3/2-— 1/2 20737.80 lh.6h 20752.h5 20753.u6 3/2 — 3/2 - 11.72 20726.09 20727.06 3/2 — 5/2 2.93 207h0.73 207h1.68 13CD37QBI‘ 3/2 — 1/2 111881.52 4113.65 11.173187 1A739.37 3/2 — 3/2 111..92 111996.11. 111998.29 3/2 — 5/2 — 28.73 111852.79 118511.39 13CD351Br 3/2- 1/2 lh8lh.l8 -119.95 lh69h.23 1h695.5h 3/2 - 3/2 95.96 1h9lo.1t 1l911.77 3/2 — 5/2 - 23.99 111790.19 111791.116 aBased on structure of CH3X with appropriate changes for CH to CD and HCH to DCD. 27 5.h Examination of the_§pectra A survey of the predetermined regions of absorption indicated all lines to be within 1.5 Mc of the calculated values. Each J = 0 +> 1 transition occurred as a set of three lines due to quadrupole splitting. Once the preliminary survey was completed each line was measured at 0pti- mum pressure and Stark voltage. A minimum.of two measurements was made on each line, in each case reversing the sweep of the oscilloscope and averaging the measurements. The measured frequencies are also shown in Table VI. 5.5 _Determination of Molecular Parameters Using the quadrupole coupling constants determined by Gordy, Simmons and Smith, it was possible to calculate approximately the effects of second order quadrupole coupling for the carbon—l3 deuterated methyl halides. From these values and the measured frequencies it was possible to calculate improved values of the quadrupole coupling constants using the relation (1) = _ _ <2) AWQ \) V0 AWQ where V is the measured frequency, and V o is the hypothetical unsplit frequenqy. Having obtained the average AWél) for the three lines it was pos- sible to recalculate AWéZ) and subsequently the hypothetical unsplit frequency which is equal to 2B for the J = 0 to J = 1 rotational transi- tion. IB was determined from B in Mc/Sec with the equation: IB = éggéil amu A2 The calculated constants are included in Tables VII and VIII. 28 .NN mononuwom .xnoz manhm Q Rim: m\m N\m m\m o rnmdas.ar ma.mmcoa c©.oamam mm.mm- mHWmOmHN m\m $493 w} m\m $8326 on.aacnm N\m 1r n\m I‘ I‘ I‘ o 36656.2 $.3on E2868 3.1:- $66ch m\m A. m\m omnofim m} I «\m £8326 Bianca m\m Ar m\m s 383.3 amacmofi E598 ham- BARON m\m AI m\m BARON N} Au m\m $8326 mméfim a} A1 m\m m amemae.as oe.smmoa am.aosnm oc.sa- no.saona m\m 11 n\m £538 m} I «\m £8326 SEEN a} I m\m s 883.5 2.362 maonma hem- 5.82m m\m A: m\m i 84QO NE A1 Sm 58326 $.ch «\m A: m\m o Simeon 3.3ng $.8ch 8.1:- $66ch Qm AI m\m 3.82m a} I N\m £8326 . m IIWum US a“ H om 0?. vow hoMWDWWhm cofipfimcmnh UCSoQEoo opfinoflzo finance pobmnopnmp mo mpcmpmcoo pmcwsnobop hfiflmpcosfinonxm .HH> canoe 29 .Nm ooconomomn .xnos mfichm mcfinom N} I «\m 2.42.8 m\m A: m\m o Baghmc $.53 mdsmaom MRS mmamcom n\m A: m\m m A: H n 1. Slang a} I m} 8.5320 66.938 NE. A. m\m 8.95% m\m A: N\m o oemflmfio amid: 3.38m Rim 3.38m m\m A: m\m $.8on m\m I a} m I Tu. mmdmwom N} AI m\._... mnnmnowau 2.3:: m\m AI m\m . scans $.81: amass 8.6% Ede: S I N} I amine: a} I m\m chmooflo order: QT. «\m n diamdéo 3.31:. madame: nodnm 3.3%: m\m Alm\m I Rama: a} I ma Esmeoflo mmnmmmu a} I «\m s aggro Ramon 3.83s does 5.315 «a I «\m I 8&de a} I «\m sonmnofio $623 «\m A: m\m s. omaamao 8.1:: magma Home 2.33 _ S I N} I team S I ma Romeoao IntSm ca E E om or» 606 zomwaannm do: want. Eggnog @3595 7.3.58 popmnopsop .Ho mpcmpmcoo pocwsuobop fimpcoefihomxm .HHH> magma 30 From the equation a = [u—lfiAIle/I2 derived in Section 3.6, the The coordinates of the carbon atom and the halogen were determined. use of four isotopic species permit each one in turn to be used as The AT is the the parent compound with its appropriate isotopes. b difference in the moment of inertia of the substituted speciesand the chosen parent molecule. The sum of the coordinates of the carbon atom, which lies below the center of mass and the chlorine atom above the center of mass is the internuclear distance. Only absolute values of the coordinates are obtained from this calculation.‘ The sign of the coordinate must be determined from other available information. The calculated bond distances from the experimentally determined mo— ments of inertia for methyl chloride and methyl bromide are given in Table IX. Identical calculationswere performed to determine the carbon halogen distances in the ordinary methyl chloride and bromide. The ro- tational constants used in the calculations were determined from a literature search of all original experimental work since l9h5. The following criteria were used in the order listed to determine the best B value: 1. For any compound in which three measurements agreed within 0.03 Me, the average of the three values was used. 2. The value of B used was the average determined for any one compound. The B values of 13CH37QBr and 13CH381C1 determined by Matlock, Glockner and Bianco(9) were substantially different from the values determined by Dailey, May and Townes(16). The latter B value was reported in two Separate papers by Townes and Kisliuk(28,29) with reference to the 31 Table IX. Carbon—halogen bond distances of deuterated methyl bromide and methyl chloride Compounda Coordinate C-X (8) 12CD335C1 13c033501 1.138739 1.780h0h 12CD335C1 13CD337Cl .6t5665 12CD337C1 13c0337c1 .666756 '-'—-'—- 1.780L17 12c0337c1 12CD335C1 1.113661 1300335c1 13CD337C1 1.158197 ———- 1.780380 l~'5c133~'35c:1 lZCD335c1 .622192 13CD337C1 13CD335C1 .6h2983 ———- 1.7801131 13CD337C1 12CD337C1 1.137hh8 12CD379Br 13CD379Br 1.556215 1.937662 12CD379Br 13CD351Br .381hb7 12CD3313r 12CD379Br .373788 1.937666 12CD331Br 13CD3slBr 1.563922 13CD379Br 12CD37QBr 1.5ho276 1.937668 13CD379Br 13CD331Br .397378 13CD331Br 13CD379Br .389h55 “—- 1.937680 13CD381Br lscnselar 1.588225 aThe parent compound is underlined. 32 original work. Frequency measurements listed in the two papers dif- fered considerably, although the rotational constants were in perfect The three lines were remeasured in this laboratory and the agreement. rotational constant was calculated. The measured quadrupole split— tings and the rotational constants were found to compare favorably with the values of Matlock, Glockner and Bianco. Consequently these values were used. Table X lists the values used in the calculation of the carbon halogen internuclear distances. The calculated values for the carbon halogen bond length in methyl chloride and methyl bromide are shown in Table XI . 33 Table X. Best values of methyl chloride and methyl bromide from references Compound ‘9() B(Mc/sec) IB(amuR) References Used 12CH335C1 26585.78 13292.89 38.030180 6, 10, 13, 1h 12CH337C1 26176.37 13088.19 38.62u97h 6, 10, 13 13CH335C1 25592.82 12796.21 39.506276 9, 16 13CH337C1 25197.51 12589.77 b0.15u109 9, 16 12CH379Br 9568 20 52.83uu93 10, 13, lb 12CH381Br 9531.8b 53.036035 10, 13, 1b 13CH379Br 9119.51 55.13u009 17 13CH381Br 9082.86 55.657689 17 Table XI. Carbon-halogen bond distances of methyl chloride and methyl 3b bromide Compounda Coordinate C - X (E) 12CH379Br 13CH379Br 1.617923 1.938832 IZCHE'ZQBI‘ lZCHsalBr .320909 13CH381Br lZCHsalBr 1.60780a 88 5 1.93 7 13CH331Br 13CH379Br .331071 13CH379Br 13CH381Br .338038 1 938863 13CH379Br 12CH379Br 1.600825 12CH351Br 120H379Br .318226 1 9388h5 12CH331Br 13CH351Br 1.62l619 ' 3 1201133501 13CH3 501 1.22h833 1 781291 1201133501 12CH337C1 .556h85 88 120H337C1 13CH337C1 1.2861 1 781273 12CH337C1 12CH335C1 .535085 3 37 . 80 21 13CH335C1 1 CH3 c1 5 S 1 781255 13CH335C1 12C1—1335C1 1.2007311 13 35 , 86th 13CH337C1 CH3 01 55 1.781233 13CH337C1 12CH337C1 1.222579 aParent compound is underlined. \ 35 5.6 Error Analysis To determine the experimental uncertainty in the internuclear dis- tance an equation relating the distance to the necessary hypothetical unsplit frequencies was derived and differentiated. An experimental uncertainty of i 0.03 Mc/sec in the frequency measurements was assumed for the calculations. If a1 is the a coordinate of a substituted atom on the symmetry axis and I and Il are the moments of inertia of the parent and sub— stituted species, respectively, it was shown above that a1 = [ugl (IZ - 1)]1/2 If the two atoms are on Opposite sides of the center of mass - r = a1 + a2 = [ull (ll - 1)]1/2+ [ugl (12 - 1)]1/2. 11/ 512 "3 l - Ml Keél - 1%)]1/2 + [qu mg, - [2911 K61 -%,-)]1/2 + [21121 “$2 -é)]1/2 2K(v- v1) 1/2 [2K(v- V2) 1/2 + ”1VV1 szvz Here K = 505531 a.m.u.—RZ-Mc/sec. Since ‘V, b5, and ‘02 are approxi— mately equal _ 2&1 1/2 2&2 1/2 I‘ " [plvz] + [szz] Differentiation of the above equation for' r then gives r = (R1 +R2)8V _R18V1 'RZSVZ’ where R1 - (EYE—1372) 1/2 K and R2 = (W2) The pr0pagated uncertainty is then 36 1/2 2r = [(R1 +122)st 2 + 13125le + R22 519 2211/2. The values of R1 and R2 calculated from the hypothetical unsplit fre- quencies of the methyl chlorides are shown in Table XII. The propagated experimental uncertainty in the internuclear distances assuming 0.03 _ o Mc/sec uncertainty in the frequenqy measurements is b.7 x 10 5 A. Table XII. Constants in the error analysis 12CH335C1 13CH335C1 12CH337C1 “—1/2 1.0008137 .721523 Kl/Z 711.01 1 26585.78 25592.112 26176.38 .1/2 163.05 159.98 161.79 1 A- 993.36 1109.110 1 [1.1/2 31.52 20.23 1 R 6.12 x 10‘4 6.80 x 10‘4 1 R.2 36.31 x 10_8 b5.6 x 10-8 1 R1 + R2 12.92 x 10‘4 (R1-1R2)2 16.69x10‘7 VI. DISCUSSION AND CONCLUSIONS In the previous section it was shown that the carbon-chlorine dis- tance decreases 0.0008 R and the carbon—bromine distance decreases 0.0011 R upon deuteration of the appropriate methyl halide. The question then, is what is the Origin of the decrease in bond distance. First to be examined are possible experimental uncertainties. It was also shown in the previous section that the strictly experimental uncertainty in the carbon—halogen distances was approximately 6.6 x 10—5 X, a small fraction of the observed decrease. No account has been taken of the possible effects of centrifugal distortion. However, an estima— tion of the effect shows it to be completely negligible. The second and most likely source of the shorter distances is a vibration-rotation interaction. The distances determined are substitu- tion parameters, and though they are often believed to be close to equilibrium values, some differences always remain. Thus if the average C-Cl distance in the deuterated methyl chlorides is less than in the corresponding hydrogen compounds, one wuuld expect the decrease to appear in the substitution parameters. The changes of 0.0008 g in the C—Cl distance and 0.0011 X in the C-Br appear to be consistent with the changes in average bond length suggested by Bartell (l9). Bartell predicted a decrease of 0.003 X in the average C—C bond length upon complete deuteration of ethane. His prediction was based on a consideration of the relaxation of non—bonded repulsions which would occur upon deuteration owing to the smaller ampli- tude of vibration of deuterium atoms relative to hydrogen. In the methyl -37- -38- halides only three hydrogen atoms are involved, and one would expect the effect to be smaller. The lower force constant of the carbon—bromine stretch in CHsBr (2.3 x 105 dyne/cm 1g. 3.8 x 105 dyne/cm for 0-01 in 011301 (30)) and the larger size of the bromine atom are consistent with the greater shorten- ing found for C-Br than for C-Cl. Another important vibration-rotation effect can be ruled out. Laurie (32) has shown that average interatomic distances between non— hydrogenic atoms decrease upon substitution of a heavier isotopic species for one or both of the atoms. This decrease is very small, approximately 0 0.00005 A, but is important because it is greatly magnified by the substitution method. To determine the effect of this decrease a series of calculations were made on l2C191335Cl, 120H337C1, 13CH335C1, l3CH337CI, and the corre- Sponding deuterated methyl chlorides. The moments of inertia of all the species were computed using an assumed structure which was (a) the same for all Species, and (b) which was the same for all species except that the 12€37Cl and 13035Cl bond lengths were 0.00005 A shorter, and the 1303701 bond was 0.0001 A shorter than the 12C35Cl bond length. The moments of inertia from these calculations were then used to compute the C-Cl bond distances by the substitution method. The re- sults of these calculations and similar ones for CH3Br are shown in Table XIII. If Laurie's suggestion is correct, and if 0.00005 3 is a reasonable value to use, the C-Cl distance in CHscl is apparently 0.0018 0 A longer than the value shown in Table XI and the C—Cl distance is o . 0.0020 A longer than the value shown for CD301 in Table IX. Thus we -39- 0 have recovered 0.0002 A of the apparent shortening but no reasonable parameters would enable recovery of much more. A similar conclusion is reached for CH3Br. Table XIII. Carbon—halogen bond distances in methyl chloride and methyl bromide computed assuming 0.00005 A decrease in the C-X bond distance upon substitution of heavier isotopic species. Parent Species C-X (Assumed) C-X (Calculated) 1201133501 1.78100 1.77923 120033501 1.78100 1.77903 12CH379Br 1.93900 1.93611 12003791311 1.93900 1.93595 A final possibility for the apparent decrease in bond length upon deuteration is a failure of the Born-Oppenheimer approximation. The only way to determine the cause of the decrease would be a completely accurate calculation of vibration-rotation effects, or by a determina- tion of the equilibrium.moments of inertia for all of the species. Of these two possibilities the latter seems much more likely, but still represents an extremely difficult experimental problem. 10. 11. 12. 13. 11. 15. 16. 17. 18. 19. 20. REFERENCES Cleeton, c. B. and N. H. Williams, Phys. Rev. 15, 231 (1931). Hershberger, w. 0., J. Applied Physics 11, 195 (1916). Neilson, A. H. and H. H. Neilson, Phys. Rev. 56, 271 (1939). Gordy, W., J. W. Simmons and A. G. Smith, Phys. Rev. 11, 311L (1919). Kellogg, J. M. B., I. I. Rabi, N. F. Ramsey, and J. R. Zacharias, Phys. Review 51, 677 (1910). Gordy, w., J. w. Simmons and A. 0. Smith, Phys. Rev. 11, 213 (1918). Gordy, w., J. w. Simmons and A. G. Smith, Phys. Rev. 11, 1216A (1918). Karplus R. and A. H. Sharbaugh, Phys. Rev. 15, 889L (1919). Matlack, G., G. Glockher, D. R. Bianco and A. Roberts, J. Chem. Phys. 18, 332 (1950). Simmons, J. W. and W. B. Anderson, Phys. Rev. 89, 338 (1950). Geshwind, S., R. Gunther Mohr and C. H. Townes, Phys. Rev. 81, 2881 (1951). Johnson, C. M., W. Gordy and R. Livingston, Phys. Rev. §5, 1219L 1951. Kraitchman, J. and B. P. Dailey, J. Chem. Phys. 11, 1177 (1951). Thomas, J. W. Orville, J. T. Cox and W. Gordy, J. Chem. Phys. 11, 1718 (1951). Sharbaugh, A. H. and J. Mattern, Phys. Rev. 15, 1102 (1919). Dailey, B. P., J. M. May and C. H. Townes, Phys. Rev. 15, l36L (19h9)o Simmons, J. W. and W. 0. Swan, Phys. Rev. 89, 289 (1950). Shiner, V. J., Jr., J. Am. Chem. Soc. 11, 5285 (1952). Bartell, L. S., J. Chem. Phys. 26, 3195 (1962). Thomas, W. J. Orville, J. T. Cox and W. Gordy, J. Chem. Phys. 11, 1718 (1951). -10- 21. 22. 23. 21. 25. 26. 27. 28. 290 30. 31. 32. -11- T0wnes, C. H. and B. P. Dailey, J. Chem. Phys. 11, 782 (1919). Bardeen, J. and C. H. Townes, Phys. Rev. 15, 97 (1918). Townes, C. H. and A. L. Schalow, Microwave Spectroscgp , McGraw Hill 00., New York, 1955. Kraitchman, J., Am. J. Physics 11, 17 (1952). Goldstein, H., Classical Mechanics, Addison Wesley Press Inc., Reading, Mass., 1950. Nystrom, R. P., and W. Yanko, J. Am. Chem. Soc. 10, 111 (1918). Simmons, J. w. and I. H. constein, J. Chem. Phys. 29, 122 (1952). Kisluik, P. and C. H. Townes, Nat. Bureau Standards Circ. 518 (1952). Kisluik, P. and C. H. Townes, J. Research, Nat. Bureau Standards 11, 611 (1950). Linnet, J. W., J. Chem. Phys. 8, 91 (1910). Costain, C. C., J. Chem. Phys. 19, 861 (1958). Laurie, V. W. and D. R. Herschbach, J. Chem. Phys. 37, 1687 (1962); V. W. Laurie, Symposium on Molecular Structure—and Spectroscopy, Columbus, Ohio, June, 1963. CH.BTRV UREA 97 1‘4? I MICHIGAN STATE UNIVERSITY LIBRARIES 2 II II NW 889 31293 03083