\

H)

‘1
H

H

“WI

I
|

H
i‘

1

W

I
7

[III

 

A PRELEMENAR‘!’ STUEY 0F ?HE
MICROWAVE SPECTRUM 0F
METHYLCYCLO‘PEEGPANE AND THE.
DETERMi.MT§QN Q? WEE 33$?th
MOMEWT 0F EYCLOMW‘YE» CYANEDE

T'hcsfis few é‘im Deng?“ Q§ M. 5,.
MECHEGAN STATE UNWEESFW
Lois Ceder'berg Leffler

{gag

.2qu

 

LIBRARY

Michigan Stan
University

 

A PRELIMINARY STUDY OF THE MICROWAVE SPECTRUM OF
’METHYLCYCLOPROPANE AND THE DETERMINATION OF
THE DIPOLE MOMENT OF CYCLOPROPYL CYANIDE

BY

Lois Cederberg Leffler

A THESIS

Submitted to the College of Science and Arts
of Michigan State University in partial fulfillment
of the requirements for the degree of

MASTER OF SCIENCE

Department of Chemistry

1959

To "Little Bear"

ii

ACKNOWLEDGMENT

The'author wishes to express gratitude toDr. Richard H.

- Schwendemanfor his willingness to spend long hours answering
questions, tracking down the causes of difficulties with the
instrument, and for his patience in the many explanations re-
quired during the period of time this work was done. - Also
appreciated were'Dr. Harold Hart's suggestions and» literature
references for the preparation of amethylcyc10propane and the
financial aid contributed by the Department of Chemistry in

the form of a graduate teaching assistantship.

iii

A PRELIMINARY STUDY OF THE MICROWAVE SPECTRUM OF
TMETHYLCYCLOPROPANE AND THE DETERMINATION OF
THE DIPOLE MOMENT OF CYCLOPROPYL CYANIDE

BY

Lois Cede rbe rg , L'effler

- AN ABSTRACT

Submitted to the College of Scienceaud-Arts
of Michigan State University in partial fulfillment
of the requirements for the- degree of

MASTER OF SCIENCE

Department of Chemistry

1959

' Approved

 

ABSTRACT

MethylcycloprOPane was prepared using zinc and l, 3—dibromo-
butane;and preliminary calculations were made'which indicated that
the 110-211, 101-202, and 111‘“212 transitions should be visible in the
range of the 2K33 klystron ‘then available. A sweep of the entire
frequency range (22, GOO-25, 000 mcfsecond) was made and four'weak
lines were found at 25056.1, 23870.1, 24139. 3, and 245.32. 2 mc/
second. No consistent assignment could be obtained using any 3 of
these lines.

The Stark effect of certain transitions. in the microwave Spectrum
of cycloproPyl cyanide was studied and the dipole moments about the a
and c axes were determined to be ”A = 3. 88 D and PC = 0. 86 D making
the total dipole moment ”total = 3. 97 D debye. In addition the fre-
quencies of several previously unreported transitions were measured
in an unsuccessful attempt to obtain an accurate. value of the smallest

principal moment of inertia.

TABLE OF CONTENTS

CHAPTER Page
I HISTORICAL BACKGROUND AND THEORY ........... 1

II DESCRIPTION OF THE MICROWAVE SPEC-
TROMETER ....................................... 9
III CALIBRATION OF THE FREQUENCY METER ........ 16
IV METHYLCYCLOPROPANE. . .. . . . . . . . . . .. ......... 20
V CYCLOPROPYL CYANIDE ......................... 24

LIST OF TABLES

TABLE Page
I. Rotational Constants and Structural Parameters of
C'yclOpropyl Cyanide ........... . . . . ...... . . . ....... . 25
II Observed Frequencies for Cyclopropyl Cyanide ........ 26
III Predicted Lines and Frequencies for C-Type
Transitions. . . . . . .................................. 27
IV Transition me313 (20521.1 ms) ...................... 28
V Transition 212—313 (19984.6 mc) ...................... 29
VI Transition 2.03m303 (202-46. 27 mc) ..................... 29
VII Dipole Moments of Ethyl Cyanide, Cyclopropyl '

Cyanide and Ac ronnitrile in Solution and Using Mic ro-
wave Techniques ....................... . ........ . . . 30

vii

F I GU RE
I
II
III
I V
V

LIST OF FIGURES

Page
The Microwave Spectrometer ....................... 10
Reference Frequency Generator ..................... 14
Transition 211—312 (20521.10 mc). . . ........... . . . . . . 32
Transition 212-313 (19984. 6 me) ....... . ...... . ...... 33
Transition 202"303 (20246. 27 mo) .................... 34

viii

CHAPTER I

HISTORICAL BACKGROUND AND THEORY

World War 11 made it possible for research using high resolution
microwave spectroscopy to make giant strides. The military had
arbitrarily chosen and developed 1. 25 cm. radar in an effort to replace
10 cm. equipment, as the latter could not be focussed into a narrow
beam because of its long wavelength. The 1. 25 cm. microwaves could
be sharply focussed but had an extremely short range due to absorption ‘
by .water vapor in the rangeof 1. 0 to. 1. 6 cm. , withmaximum absorption
very close to 1. 25 cm. Sincevthe equipment was needed principally in
the'South'Pacific area, and sincerthis area abounds in .water vapor, the
1. 25 cm. equipment had to be discarded in favor of 3 cm. equipment
which had been found to be satisfactory. This-military defeat proved to
bea boon for subsequent research, as it made‘much 1. 25 cm. equipment
available to scientists (1).

The first experiments with-microwaves were reported in 1933 by
‘ Cleeton and Williams (2) who-determined the absorption of centimeter
radiation by ammonia gas. The- first demonstrations of the high resol-
1utioneavai1ab1e making useeof radar electronics were reported in 1946
by Bleaney and Penrose‘ (3, 4) and by Coles and Good (5), also studies
of absorption of ammonia. - Since 1946'mic rowave spectroscopy of
gases has yielded useful informationin a variety of fields. Its ‘ high
resolution, novel frequency range, and greater accuracy make possible

the- study of a variety of different phenomena. In addition to pure

rotational spectra these include nuclear quadrupole effects, pressure

broadening, and Stark and-Zeemaneffects (6)

, - 1
General Theory

, Abso‘rptionof microwave radiation, by simple. gas-molecules is
largely the result of transitionsbetween the-discrete rotational energy
- states of themolecule in its ground vibrational and ground electronic
energy state. The frequencies-at which the' absorption of microwaves
:takes'placevdepends uponthe- structure- of the-molecule as-may be seen
.from the expression for the absorption frequencies of a diatomic mele-

cule

“)J —)rJ+1 =,~ZB(Jd-1)

, where J. is zero or- apositive integer. which-measures the total angular
.momentum of the-molecule, and-B is. a constant called the "rotational
constant and is~ related tothe average internuclear-distance r0 of the
-molecule by

81? mime T110
. wherethis’ Planck' s constant and-tmg and. may arethemasses of the two
- atomsin the- molecule.

Whereas-the diatomic: molecule canrundergo only end-over-end
rotation, nonelinear-polyatomic: molecules can rotate about any of three
axes, the~principa1 axes. of inertia. Thus the rotational. energy levels
of polyatomiccmolecules-aremuch-more complicated expressions of
. [the internuclearrdistances of themelecule; in'fact being- related

.through the three-principal moments of inertia of themolecule.

 

AA more comprehensivetreatment of the. general theory of microwave
- spectroscopy maybe-found in References. 7, 8' and 9.

l

The rotational energy levels of non-linear polyatomic molecules

may be expressed as follows:

B'C B+ZC
WM = Tz J(J+1)+ (A - ——-—) wa(bp)

 

Here W Jgs the rotational energy of a molecule in the 11th of the 2J+1
energy states with total angular momentum “J J(J+l) h/2 TI . The
quantum number T ranges from -J to +J in integral steps. The
rotational constants A, B and C are related to the principal moments
of inertia IA'B I , and .IC

expressions of the form

,. - ' <
(conventionally IA _<_ IB __ IC) by three

 

The quantities W th’ are a function of the asymmetry parameter bP

given by

1. C-B
”P‘ ZA-B-C

 

and are obtained as the eigenvalues of a non-diagonal matrix. However
in the case of a near prolate symmetric top (a molecule would be a
prolate symmetric top if A > B = C) bp is very small and the quantity

w J,’pmay be obtained by systematic . application of perturbation theory

as a power series in lap,

. __. Z 3 4 5

NJ’t - K +C, bp+Cszz+ Csbp +Csbp +C5bP
The values of K2 and C, through C5 have been tabulated for all energy
levels with J _<_ 40 (10, 11).

The rotational absorption frequencies are given by the Bohr

frequency condition

hd J,’t ;J',"’£ ' = J',L ' A 33?.”

~ Not all transitions occurwith equal probability'however and some are
forbidden. Tables giving values. of the relative probability of transi-
tions between energy levels with J _<_ ,12 have been‘prepared for mole-
cules containing components of the dipolevmoment along any or all of

the threeeprincipal axes (12). The theory shows that the intensity of a
transitionis prOportional to the square of the component of the dipole
moment along a. given axis. Those transitions requiring a dipole moment

- along the A-«axis (the. axis from which-I

A is measured) are calledeA-type

transitions, etc.

It is often useful to replace the index fwith the bipartile index
K-“ K“ where K41 h /2 1r would be the angular momentum of the mole-
cule about the-A axis in the case of a limiting prolate symmetric top

{IA < I = Ic), and K+ 1 h/Z It would be the angularmomentum of the

B
molecule about the (Iv-axis in the case of a limiting oblate symmetric
t0p (IA: IB< 1C). The quantities K4 and K“ must be zero or
positive integers less than or equal to J, whereas 1' ranges from -J to

+J in integral steps. The relation between the two is

A‘: = Kul“ K+1

All the transitions below will be listed as occurring between energy

levels J and J'
KMKH K'-1K'+1

The moments of inertia of the molecule are related to the coordi-

nates of the various nuclei in an arbitrary cartesian coordinate system

as follows .

.. 2 2 V 2 2
I 213mi(yi+si)-M(y +s )
Iyy 213 m1 (xi + :1. ) M(x + 2;. )
_ 2 z _ ' z z
.122 — E mi (xi + yi ) M(x + y )

. .th
where xi, y:, and zi are the coordinates of the 1 atom of mass mi
J.
and the summations extend over all the atoms in the molecule. The
values x , y and z are the coordinates of the center of mass of
m m m

the molecule and M is its total mass. The products of inertia are

given by
I 2-2 m_x, ', - Mx
xy 1 1 ryi mym
I 3-2.3 m,x,Z, - Mx Z
X15 1 1 1 1 m
I 21-2 m. ,z, - M‘ z
yz i 1Y1 1 Ym m
The three principal moments of inertia IA’ IB’ and IC are the eigen-

values of the moment of inertia tensor
[1 I I I
xx xy xz

I_ I I I
- XYWYZ

 

 

lez yz zz}

./

The procedure in analyzing the microwave spectrum of a new

compound is then as follows:

a) Computation of the principal moments of inertia of the
.molecule based upon an assumed structure. The rotational
energy levels and transition frequencies may then be computed
taking into account the region of the spectrum which is avail--
able to the Spectrometer.

b) Search of the microwave Spectrum of the compound for the
predicted absorption lines and careful measurement of the
frequencies of absorption lines in the regions of those

predicted.

c) Assignment of absorption lines to the energy levels
involved and recalculation of the principal moments of
inertia based upon the experimentally determined fre-
quencies. An assignment is usually assumed confirmed when
six or more absorption lines are consistent with the same
three rotational constants to within 0.1 mc/sec (about 5 parts
in 106).

(1) Analysis of the structure of the molecule based uponthe
measured rotational constants. Ordinarily themoments of
inertia of one or-more isotopic species will be required before

very much may be said about the structure of the molecule.

The Stark Effect.

The variation of the energy levels of a molecule in a varying
electric field is known as the'Stark effect. The rotational energy levels
are ordinarily at lea st (2J+l)-fold degenerate corresponding to the
2J+1 orientations of the total angular momentum with respect to an
axis fixed in Space. Upon the application of an electric field the field
direction becomes the Space-fixed axis and because of an interaction
between the applied field and the dipole moment of the molecule the
2J+l different orientations of the. molecule have slightly different
energy levels.

The theory of the Stark effect of asymmetric-rotator molecules
has been examined (13). It predicts a shift in energy levels whichis
proportional to the square of the component of the dipole moment along
a given axis and to the square of the field strength. In-particular for
-molecules with'M = 0 (total angular momentum oriented perpendicular

to the field) the Shift in. energy (A W ) for a molecule with components

Jgt
of the dipolemoment along theA and C axes will be

A
1:”A

C

2 3
,‘E ”c ’E

AW =‘FJ,

z z
E +
J,’t 1:‘.I

where ”A and “C are the components of the dipole moment and E is
. . A . .

the field strength. Th: (giantity FL,‘ is given by x (A)

' J J,’l’;J-l,’t ' J-I-l Jfl: ;J-l,"ci'

F A __ g. Q + . 23"

Lt ‘ (4.13-1) m: ; L1,? (2J+1)@J+3) t J,’B;J-l,"t"

 

 

 

 

(A)
J,’t ; J—lfli ' . f
for the-A-type tranSition of frequency J,’t;J-1,’B '. The summation

The quantity x is the. transition. probability mentioned above
extends over all the transitions with non-zerox.

The absorption frequencies inthe presence of an electric field
-will be the result of a transition between two shifted energy levels and
consequently the shift in frequency, A Q J A: °J"’t"' will be the difference
between the energy level shifts, or ’ ,

A
J: rt

A 3 3 C C
~FJY)I.LAE+(F -F

h __ ‘
A9 " (F my 32’

2 Z
J,t ;J"t' ) "c E '

Therefore the slope, d(hA\) )%(Ez) ,. will be

A c
__g_(_hA9)=AF “Mar (.2.
GIFT—E) A c

Consequently determination of the slope of the frequency vs. square-
of—the-field line for the'M = O Stark component of two or more
different transitions may be used to. determine HA2 and “C2 and the

total dipole moment may be obtained from

2.. z z
,u HA+HC°

In the conventional Stark-modulated microwave spectrometer the

sample gas is subjected to alternate periods of electric field and zero

field. - Since the frequency of alternation is usually very'large

(1009 000 c. p. s.) the normal Spectrum- and the'Stark spectrum appear
. superimposed. The-phaseasensitive detector being sensitive to the
phase as—well as torthe frequency of the incoming. signal is capable of
distinguishing Stark components. from the normal. spectrum. by virtue
of the fact that their signalsarrive at the detector 1800 out of phase.
Consequently the output of the phase-sensitive detector is of one polarity
for absorption. lines in the normal spectrum,. zero for. no-absorption,

' and of oppositepolarity for Stark components. Since themodulating

- fieldmay betvaried at will between» 0 and; about lZSOvolts, determin-
ation of the required slope is accomplished by determining the
frequency of the M = O Stark component at various settings of the

-modulation voltage .

CHAPTER II
DESCRIPTION OF THE MICROWAVE SPECTROMETER

Figure I shows a block diagram of the instrument in use at Michigan
State University in the Department of Chemistry.

The klystron power supply is a- FXR, Type Z815B manufactured
by the‘Electronics and‘rX-Ray Division, FR Machine Works, Inc. ,

- Woodside 77,, New Jersey. It must supply «500 to-2500 volts to the
cathode, depending on the particular klystron being used, -50 to -1000
volts (measured from the cathode) to the reflector of the klystron,

0 to -300 volts to the grid, again measured from the cathode, and 6. 3
volts AC or DC for the heater.

The klystron used for most of this work was a Raytheon”I QK306
requiring 1800 volts at the cathode'at 9- ma, and 6,. 3 volts at . 58
amperes for the heater. Its frequency range was approximately
- 18, 000 to 22, 000 megacycles/second. However, some work was done ‘
with the Raytheon 2K33 klystron w--ose range was 22, 000 tO' 25, 000
megacycles/second.

. The frequency meter used for approximaterfrequency measurements
.was- of the absorptiontype, causing a dip in the transmitted power
, when. the size of the cavity was of the proper dimensions for the
radiation. The frequency/gitshere was a: De Mornay Bonardi, Serial
No. 715-2 720. The calibration chart sent with the meter appeared to
be in. error so a recalibration was made. The results appear in

Chapter III.

 

:(I
Raytheon Manufacturing-Co. , Tube Power'Division, Waltham, Mass.

10

hoaosgpooam @530th one H 0.53%

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. .h-
89%

 

 

 

 

 

rofinmoomfpa new,"
_ manna Ta .5 8m

 

 

 

 

 

  

$5

l.vl;

 

 

 

 

    

, s
_ ”Human _ on p5 Honda H935 .839"
, -m . «8.3.929 M.
- pooqco om n 0 dance - .

. as is: E . a
82 OPOGPQQ him .7 0+
.8. _ . . m

a 3996 v

9 go on .0308 m
nopoopom uopwhocmm Ewom m.
Haws.“ m . 3n 9.." 98m 9530 apoopzmm Adan .m.

.§ 02 F u

i . Ndoo
mesopohom Toupee
98.3 .v- 353% Agmmoooz “HY“...
Jox OOH .ox OOH .
”ago H . _
.n . .x. . i l
.. dnou enemas; 83H and.” hopoz- 335p weaves
539334 ponoama- 8.393 493.5on manage 0.5929
meHm: . W -
8.3 on 08308
:83va mid 91m.
. . h H

 

 

3.5.08
JOE H

 

 

 

 

 

 

 

 

11

- The waveguide is a ten foot length of X~band brass guide with a
silvered inside surface which conducts radiationabove 8. 2 kilo-
megacycles/second and can be used for radiation up to 40 kilomega-
cycles/second. Its inside dimensions are 0.400" x 0. 900" and the
walls are 0. 050”. It is sealed off with-mica windows from the rest
of the instrument and is filled via the vacuum system. The detector
is a silicon rectifier.

Two oscillosc0pes are in use; a Tektronix twin beam experimental
model and a Dumont single beam oscilloscope. The latter is used in
connection with the frequency standard to observe the Lissajous figures
while the Tektronix oscilloscope is used to monitor the absorptions of
the molecule and also to determine the frequency at which the absorption
occurs. The sawtooth voltage‘sweeping across the x—axis of the
oscillosc0pe is simultaneously applied to the reflector electrode of
the klystron varying the microwave frequency uniformly. The signal
from either the phase-«sensitive detector or the diode detector is
applied to the y axis causing the face of the oscilloscope to show a
plot of the amplitude of the amplified A. C. signal at the detector vs.
microwave frequency.

The square wave generator was designed by L. C. Hedrick and
patterned after a model described by him (14). It provides up to 1250
volts of zero based square wave output at 100 KC/second. The square
‘wave voltage is applied to a coin silver septum (O. 032” x 0. 796'”)
which ismounted between Teflon tapes (0. 062" x O. 400"with a 0. 032"
groove 0. 010” deep). The Stark field applied to the sample produces
changes in frequency or Splitting of absorption lines by exerting a

torque on the rotating molecules upon interaction» with their dipole

12

moments. This "Stark effect")which can be used for the determination
of dipole moments as described in the previous chapter,is also extremely
.useful in the differentiation of noise from the absorption line. The
generator exposes the sample to electric fields during alternate 5 p.-
second periods, thus alternating zero-field absorptionand‘ Stark
absorption. Only signals modulated in this manner appear on the face
of the Oscilloscope. From the crystal detector the signal goesthrough
a preamplifier-amplifier system which blocks all frequencies but the
ones near 100 kc. The 100 kc amplifierrused-was adapted from one
designed by C. C.. Costain of the” National Research Council at Ottawa,
Canada, while thepreamp was designed by J. C. Williams of Harvard
University. The amplifier output then may go either to a crystal
diode or to a phase-“splitter and phase-shifter. If the output is sent to
the diode the 100 kilocycle signal is rectified, averaged, and applied
to the input of the oscilloscope. . If sent to the phase-splitter-phase-
shifter combination the phase splitter splits the signal into‘ 2 signals
which are equal but 1800 out of phase, while the phase shifter makes it
possible to make the phase of the incoming signal identical to that of
a reference produced by the square-wave modulator. The net result
is ”that the Stark absorption signal being- 180° out of phase to, the normal
. spectrum is of opposite polarity causing Stark lobes-to be in the opposite
direction to the normal absorption at the oscilloscope or recorder.
The great advantage of this latter system is the rejection of 100 kc
noise components having the wrong phase, as well. as noise having a
frequency different from 100 kc.

If use of the recorder is desired the signal from the phase-sensitive
detector. is sent directly to the recorder and the klystron is driven

mechanically over the frequency range.

13

The frequency standard consists of aHallicrafters’SX-62-A
receiver, a MansonsLaboratories RD-l40 high-stability 1 mo oscillator,
a Gertsch‘products AM- 1A VHF interpolator, anda Gertsch‘FM-4A
microwave frequency multiplier (see'Figure'II). It is- possible to
.measure frequencies to 0. 02imc at 20, 000 mc but with compounds
containing broad absorption-lines considerably less-precision is
possible, 0.1 me at 20, 000 mc seemingly the-limit.

The 1 me oscillator is calibratedby comparison of its 10th
harmonic with the 10 mo carrier of WWV, of the'National Bureau of
Standards. It is adjusted so that the background noise of a receiver
tuned to WWV pulsates less than once every 2 seconds‘indicating an
error in the 1 me oscillator of less thanO. l c.p. s. . Thel mc output
of the‘ RD-140 is then fed to the input of the AM-IA VHF interpolator
where it is amplified and selectively multiplied 19-38 times. The
output of the multiplier is mixed with the output of a 1-2 mc. oscillator
- and the sum frequency used to control the frequency of a 20-40 mc
oscillator. The frequency of the 1-2 mc oscillator (LFO) is determined
by the appearance of a stable Lissajous figure on the face of the
oscilloscope,. whose frequency of rotation is kept lessvthan one c. p. s.
in order to keep the error in the LFO frequency below 1. c.p. s. The
output of the AM- 1A is then a single signal of some frequency between
.. 20 and.40 mc plus a1 mc reference signal which are-input to the'FM-4A.
This latter instrument is locked on a harmonic of the'AMlA $.10 me.
This signal (between 500 and 1000 mc) .is then sent, by means of a
coaxial cable, to themixer-multiplier crystal where its harmonics
are generated‘and mixed with the microwave signal. The-difference
frequencies (between the harmonics of the 500-1000 mc signal and
.themicrowave frequency) are then sent back along the same cable

through a filter which blocks all frequencies above 500 me.

nopmnocoo evacuees oesonomom HH shaman

14

 

 

 

 

. Aoflswobsw 5 H6995 c.5— podnonpnoov one:
. - headaomo oz H9330
jig mum-Hg 08180.nl00M— LOB §H300m @fimfigg

 

     

     

 

 

‘

I. -

v —

HI I l l I .[ I ' I | L

 

"llll

as
#8302 cogenpnoovo llll...
H9350 nopwaomo ohoHHanHsz 836w
segues as 3.8 flames

cos D310N

 

 

 

 

 

 

 

 

 

«A :63. m Gamma

.2: .n

 

 

 

 

 

A

 

 

 

05.4. 8.3

hopmaomo o - L
18:250-. noun”: Ca mumd
33 023 4..- .3333:

 

 

 

 

 

 

 

Amfloomowquo .HofimeQEoo on: . m . _ _

oflnmm 78ng

 

 

 

 

ll sausage

85 3:3

 

 

 

 

apnoaonm

 

 

 

14

3009306 humouwofim oodohmMom HH 0.3m..."

 

 

“03.30.63 5” 30.30 0.3— $320538

 

 

 

 

 

 

 

. . 90903000
3...... 0000000 oadooaéomh T... 08703
I... l l I II II. .
._l
. 0s
, 0303.02 004403.580 .Illln
Hohboo hopwaomo chowdnfiflunz SOHAON
50:00on .08 o3lom gage .08 .n

 

 

 

 

 

 

 

4H :24 m omega

 

 

 

 

_
_ _
_ _
w .05 33.6w _
Ho .m o o - OE .
“ 08Mwnod~mw08. 4- -. 005.2 96 mumd “mafiamsd _
”—l .60 .3. .3333. - , .o_.—. H _

.os mad

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Amfioomowdwomo hopmhwmaoo 0.3 . Ax _ .. _ _ - . . _
. I 005 —H 3' .IJ .HOflHV-Hflma
.03”an zomzdz Hopwaomo . G05 G05 dado.“ _
2050033 >33 0.3 m 089m .05 H Hmpmhho stoma." .

 

 

 

 

 

gr llllll

 

15

This difference frequency then may be measured on the interpolation
receiver. Each time the klystronvsweeps over the frequency
measured by the system there will be "pip" on the oscilloscope and a
sound will be heard in the speaker of the interpolation receiver.

. The dual beam oscilloscope makes it possible to place an absorp-
tionxline on one trace and measure it with the frequency standard marker
placed on the second trace. If desired the recorder-may be used in
connection with the frequency measuring device. The output of the
phase sensative detector is fed to the recorder. ~Asthek1ystron sweeps
over the frequency which has been set up a sound .is heard. - At this
instant a telegraph key may be closed causing a momentary change in
the voltage which the recorder sees that is duly registered on the

recorder paper.

CHAPTER III

CALIBRATI®N @F THE FREQUENCY METER

The De :Mornay‘Bonardi Frequency meter was calibrated by putting
a known frequency marker generated by thefrequency standard on the
os cilloscoPeYJ and thenxadjus-ting, the frequency meter until the-power dip
inthe‘klystron output caused by the'absorgption’by the-meter and the
frequency marker were superimposed. The. correction term was
obtained by notingthe actual reading. of the scale». and the chart reading
corresponding to the frequency at which the standard was set. The
, correction term must be'added algebraically to the actual frequency
meter reading to givethe chart value corresponding to the true
frequency, or subtracted algebraically from the chart value to obtain

the actual reading .

16

17

CALIBRATION OF THE De MORNAY BONARDI WAVEMETER.
No. 715-2 720
KLYSTRON QK306

 

 

 

~__

Actual Frequency Interpolation True Wavemeter Cor-
. Wavemeter Standard - Receiver Frequency Reading rection
—Reading Base (mc) Should Be Term

7 (From Chart)

234. 8 18050 -40 18010 234.6 -0. 2
.230. 45 +40 18090 230. 2 -0. 25
226.7 18200 -40 18160 226.5 -0.2
222.5 , +40 18240 221.25 -0.25

216.4 18400 -40 18360 216. 25 -0. 15

212.4 +40 18440 212.3 -0.1

206.6 18600 -40 18560 206.5 -0.1

202.8 +40 18640 202.75 -0. 05
197.3 18800 -40 18760 197.2 --0.1

193.8 +40 18840. 193.6 -0. 2

188.6 19000 -40 18960 188.4 -0. 2
185.2 ' +40 19040 185.1 -0.1

182. 2 19150 ----40 19110 182.4 +0.2
'179. 0 +40 19190 179.4 +0.4

172. 3 19400 -40 19360 172.7 +0.4

169.2 +40 19440 169.4 +0. 2

164.7 19600 --40 19560 164.6 -0. 1

161.? +40 ’ 19640 161.6 -0.1

157.4 19800 ‘40 3.9760 157.3 -0. 1

154.5 +40 19840 154. 5 0. 0

150.4 20000 -40 19960 150.4 0.0

147. 7 +40 20040 147.7 0. 0

143. 7 20200 -40 20160 143.7 0. 0

141.2 +40 20240 141.1 -0.1
- 137». 3 20400 -40 20360 137. 2 -0. 1

134.8 +40 20440 134.7 -0. 1

131.1 20600 -40 20560 131.0 -0.0

128.7 +40 20640 128.75 +0. 05

125. 2 20800 -40 20760 125. 3 +0.1

122.9 +40 20840 123. 1 +0.2

119.6 21000 -40 20960 119.65 +0.05

117. 3 +40 21040 117.4 +0. 1

Continued

Klystron QK306 1... Continued

18

 

 

-Frequency Interpolation True

 

Actual Wavemeter. Cora
Wavemeter - Standard ~Receiver -Frequency Reading rection
vReading Base (me) ' . Should Be 'Term
-- (From. Chart)
114.0 21200 :40 - 21160 ‘114.1 +0.1
111.9 +40 21240 112.0 +0.1
108.7 121400 ~40 21360 108.9 +0.2
106.7 +40 21440 106.8 -+0. 1
103.7 21600 ‘940 21560 103.8 +0.1
101.7 +40 21640 101.8 +0.1
98.7 21800 ~40 21760 98.85 +0. 15
96. 7 +40 21840 97. 0 +0. 3
94.0 22000 ~40 21960 94.3 +0.3
92.1 +40 22040 92.4 +0. 3

 

19

KLY ST RON 2K33

 

 

Actual Frequency Interpolation -True Chart Cor-=-
'Wavemeter . Standard Receiver ‘Frequency ' Wavemeter rection
wReading Base (mci , , Reading Term
92.8 22050 ~40 22010 '93.1 +0.3
91.0 ~+4O ' 22090 91.3 +0.3
88.2 22250 «40 22210 88.5 +0.3
86.4 +40 22290 86.7 +0. 3
79,8 22550 +40 22590 80.2 +0.4
77.4 22750 ~40 22710 77.7 +0.3
75.7 +40 122790 76.1 +0.4
73.3 22950 .40 22910 73.6 +0.3
71.7 +40 22990 72.0 +0.3
69 3 241.50 40 2.3110 69.6 +0. 3
67.7 +40 23190 68.1 +0.3
65.5 23350 ~40 23310 65.8 +0.3
3.9 +40 23390 64.3 +0.4
61.7 23550 -40 23510 62.1 +0.4
60.2 +40 23590 60.6 +0.4
58.0 23750 ~40 23710 58.5 +0.5
56.5 +40 23790 57.0 +0.5
54.5 23950 m40 23910 54.9 +0.4
53.1 - +40 23990 53.6 +0. 5
51.0 24150 440 24110 51.5 +0.5
49.6 +40 24190 50.2 +0.6
47.7 24350 ~40 24310 48.2 +0.5
46.4 +40 24390 46.9 +0.5
44.5 24550 440 24510 44.9 +0.4
43.2 +40 24590 43.7 +0.5
41.3 24750 ”40 24710 ' 41.8 +0.5
40.0 +40 24790 40.6 +0.6
37.9 24960 ~40 24920 38.6 +0.7
36.7 +40 25000 37.4 +0.7

 

CHAPTERIV

METHYLCYCLOPROPANE

Introduction _

‘It is known that the chemical properties of cyc10pr0py1 compounds
are similar in some respects to the chemical properties of vinyl
compounds. The barrier to internal rotation of the methyl group in
ethyl derivatives is of the order of 3 kcal/mole, whereas the barrier
to‘rotation of thefmethyl group in vinyl derivatives (is. approximately
2 kcal/mole (15). It was thought that a study' of the potential barrier in
met'hylcyclopropane (would give further insight into the nature of bonding
in the cyclopropyl ring, and specifically the degree of resemblance to
the vinyl system. It was also hoped that a dipole moment could be

obtained for the compound.

Preparation

Methylcy'clopropane was prepared using the method of Shortridge
3:031. (16). This method was first applied to the preparation of methyl.-
cycloprOpane by J. D. Roberts and R. H. Mazur (17). However some
important unreported differences in the preparation of the two compounds
came to light during the actual Operation. 1, 3-Dibromo butane (pur-
chased from Matheson, Coleman, and Bell, highest possible purity) was
added to a gently refluxing mixture of 95% ethanol, distilled H20 and
an excess of zinc dust while vigorous stirring was maintained. Whereas
they had added 1, 3-dibromo-2, Z-dimethyl prOpane to the mixture and
then continued heating and stirring for a period of 24 hours to prepare
the dimethyl compound, it was found that the preparation of methyl-

cyclopropane went much faster. The 1, 3-dibromobutane was added

20

21

over a period of approximately 1 hour. The methyl cyc10propane came
over during the addition of the l, 3~dibromobutane and was captured in
a trap immersed in a drymice acetone bath. It was distilled into a
sample bulb and an infrared Spectrum taken. This Spectrum checked
very well with the Spectrum obtained and reported by F. E. Condon,
andD. E. Smith (18). .

From literature vapor pressure data (18) a graph of log p vs l/T
was made and the vapor pressure atr-22. 8°C of methyl cyclopropane
(the freezing point of CC14) was found to be approximately 269. 2 mm.
A rough determination of the vapor pressure at this temperature was
made on the methyl cycloPropane prepared by the above method and
was found to be, 288 mm.

Later an attempt was made to remove possible alkenes ‘and water
from the methyl cycloproPane by running it through KMnO4, activated
charcoal, and CaClz, but no changes were observed in the infrared
Spectrum. A literature check was made in an] effort to find methods
of purification but the only method given was that of passing the sample

through activated charcoal (19).

Calculations
The moments of inertia and rotational constants, for cyclopropyl

cyanide were calculated using the following structural parameters (20).

‘1 cc 1. 53 X

0
d C--H 1.08 A
x. C-C-H 116°24'

O
4: H-C-leing) 118 12'

o
4 H"C”H(methy1 group) 109 30'

22

The results were as follows:

IA = 33.347
IB = 79.214
IC = 89.598
A = 15154.627
B = 6381.84

9 C = 5642.12

On the basis of the above calculation methyl cyclopr0pane was
expected to Show a-type selection rules and the transitions 110-211,
101-203, and 111-212 were predicted to lie at 24787. 75, 24003. 32, and
23308. 53 mc/second reSpectively.

.. Experimental
-A sweep of the entire range covered by the 2K33 klystron
(22, 000-25, 000 mc/sec) was made in a search for absorption lines.
Only four weak lines were found. The measured frequencies were:
25056.1 mc/second
23870. 1 mc/second
24139. 3 mc/second
24532. 2 mc/second
with the frequencies. considered accurate to :1: . l mc/sec. No com-
bination of any three of these lines are compatible with the expected

assigmnent for any values of the three rotational constants.

23

.1 .H ,_n-- ‘a
l I. f -
D4413 u-w'...’

ion

U)

~When the predicued pattern of lines did not appear it became

apparent that the wiser course to follow would be to defer further

st"

M-

idy of the compound until the capability of the instrument could be

increased in two ways:

0
1 =2 .
o 'J...

nce 2~3 transitions would be predicted in the 35-37, 000

51

c/sec. frequency range it would be desirable to obtain a

'3

klystron for use in this region, and

. the general sensitivity of the instrument should be-increased

in order to bring out any weaker-lines.

CHAPTER V

CY CLOPROPYL CYANIDE

Preliminary Considerations

In the September, 1958 issue of The Journal of Chemical Physics

 

an article was published by Friendand- Bailey entitled "Microwave

» Studies of the Structure of Cyc10pr0py1 Derivatives" (21). In this
paper data were presented to lend support to the theory of Coulson
and Moffitt (22), which-states essentially that the C-C internuclear
distance in the cyclopropyl system is shorter than a normal C-C bond,
the H—-C-H angle is equal to 1160 d: 20, and the ring has a greater
ability to conjugate with halogens and multiple bonds than the corres-
ponding unstrained hydrocarbons (21).

Friend and Bailey assigned transitions involving the component
of the dipole moment along the a-axis, and, due to the insensitivity
of the frequencies of these transitions to the value of the moment of
inertia along the a-axis, the uncertainty in the a-moment is stated
to be 0. 4% (compared to O. 001% in the other two moments).
Furthermore, the reported structure based upon the Spectra of the
parent compound and cis and trans C3H419CN predicts a-moments of
inertia which differ by 3. 7%, 3. 1%, and 5. 1% re8pectively from the
measured values. Such large deviations certainly detract from the
validity of their reported structure.

Calculations based upon the structure given by Friend and-Dailey)
assuming the dipole moment to lie entirely along the CN bond)suggested
that transitions involving a component of the dipole moment along the
c-axis should be visible. These calculations werezlater verified by
a measurement of the dipole moment along the c-axis as described

below.
24

25

Materials

The cyclopropylcyanide used forthis work was obtained from the
Columbia Organic Chemical Company, Inc. ,1. andwas of "the highest
possible" purity. It was placed inva suitable sample bulb for use on

the vacuum line and was used without further purification.

TABLE I

_ ROTATIONAL CONSTANTS AND STRUCTURAL PARAMETERS
0F CYCLOPROPYL CYANIDE (21)

 

 

d c—ching) 1.5131 4 0.00118 d (SN 1.157440.0005 .8

d C‘CmN) 1.4720 4 0.00118 :7. H-C-H 114°36' 41‘.’

d C—H . 1.107 4 0.002 R #- C(CN)-C-H 119°34' 4 30'
Calculated (mc) Observed (mc)

A 15339.4 15917 4 58

B 3461.4 3465.06 4 0.02

c 3289.7 3286.22 4 0.02

bp -0.00718 -0.00713 4 0.00003

 

OBSERVED FREQUENCIES FOR CYCLOPROPYL CYANIDE (21)

TABLE II

Frequency (mc)

26

 

2,, 11...... 3,. 20,521.10
2., 4..-... 73.. 20, 261.17
2... -..—.-—> 3.. 20, 253. 82
20.4“...» 3.3 ‘ 20,246.27
213.444.4941 .313 19, 984. 59
313-4444 4.3 27,358.91
3.. 1--.... 4... 27,022.79
539 --~-~---»»-> 431 27,008. 78
~":31 ‘7 “"9 432

33. "-4-; 4... 27,003.61
3,. "-4-44 4..., 26, 985.78
3.3 ..-.._._....-.;.. 414 26, 643. 54

 

Attempted Assignment of CuTyp-e Transitions

The 2-3 transitions observed by Bailey were remeasured and

found to be good to 4 ’. 1 me. . Using-Bailey's values of B andC and

assuming first his variationlin'A (4 58-mc/second)and then 4 100
mc/second the following.C»=-type transitions were predicted to lie in

the regions of the QK306 klystron indicated:

27

TABLE III

PREDICTED LINES AND FREQUENCIES FOR
‘C—TYPE TRANSITIONS

 

 

 

>Re1ative Frequency Range Frequency Range

Line Intensity in mc/second in mc/second
Assuming A to Assuming A to
Change by :t 58 Change by Approxi-
mc/sec. mately :E 100 mc/sec.

103.-945 1.05 19750-20600 19475-20875

144,1...13.“8 1. 5 17, 900-18, 750 17, 500-19100

725—817 1. 31 19125—19490 19000-19600

827-919 1. 56 18850-19210 18725-19325

4,...»505 2. 0 20200-20300 20160-20360

 

These ranges were scanned, many rough'frequencies and some

exact frequencies were obtained, and an attempt made to fit lines
B+C

2 .
were used to calculate the rotational constants. These constants were

observed to a reasonable value of A - Also various lines

 

then used to predict other-lines. No consistent as signmentwas found

although many lines were found that could be C-type transitions.

Lines Measured

The following unreported lines were measured during the course

of the work:

28

21737. 1 mc/second

17984. 0 mc/second 20341. 7 mc/second
18100.6 20547. 2 21742. 1
. 18937.6 20555.2 21832.5 ‘
19185.0 20587.2 21907.3
19349.4 20830.6 22309.5
19411.8 21449.5 22338.5
19808. 5 21709. 3
All these frequencies are considered accurate to :2: . 1 mc/second.

Determination of Dipole Moment

 

From the study of the M = O Stark component of the 211-312
(20, 521.10 m./.o-ono, 2.3.303 (20246.27 mc/second) and 2.2-313
(19984. 59 Inc/second} lines a plot of the square of the voltage vs.
frequency {see Tables IV, V, and VI,- and'Figures'III, IV, and V)
gives experimental slopes of ~.1746x10'4, ~..1905x10-‘, and
—1. 234x10"’4 mc/voltsa resPectively. Using calculations based on the

theory of Ciriapter I

”A: 3.88 D

110: 0.86 D

PLtotal : 3° 97 D
TABLE IV

(See Figure III)

TRANSITION 2,.«312 (20521.1mo)

 

 

 

Voltage (Volta_gre)2x10-4 Frequency
300 9. 00 20519.70
400 16. 00 20518. 12
500 25.00 20516.56.
600 36. 00 20514.46
696 48.44 20512. 36
706 49.84 20512.46
800 64. 00 20509. 70
900 81. 00 20507. 18

 

29

TABLE V
(See Figure IV)
TRANSITION 212-313 (19984. 6 mC)

 

 

 

 

Voltage (Voltage)zx10"4 Frequency
406 16.48 19981. 08
508 25. 81 19979. 00
602 36. 24 19977. 50
706 49. 84 19974. 77
800 64. 00 19972. 00
898 80.64 19969. 69
904 81.72 ‘ 19970.00
994 98. 80 19966. 75
TABLE VI

(See Figure‘V)
TRANSITION 202-303 (20246. 27 me)

 

 

 

Voltage (Voltage)“"x10"4 Frequency
150 2.25 20243.71
200 4.0 3.104 20241.43
*250 6.25 4 20238.625
300 9.03::104 20235.20
400 16.0 3.10 20227.00

 

30

See Table VII for a comparison of dipole moment values for ethyl
cyanide, cycloprOpyl cyanide, and acrylonitrile in solution (benzene at

250C) and as obtained in the vapor phase using microwave techniques.

TABLE VII

DIPOLE MOMENTS OF ETHYL CYANIDE, CYCLOPROPYL CYANIDE
.AND ACRYLONITRILE IN SOLUTION AND USING
MICROWAVE TECHNIQUES:

 

 

 

Solution Mic rowave
Ethyl Cyanide 3. 56 D (23) 4.02 .D (25)
CycloPropyl Cyanide 3. 75 D (24) 3. 97 D
.5.
Acrylonitrile Not available 3.89 D (26)

 

If the structure of Friend and Dailey is used the ~C-C3N
configuration (21) is calculated to make. anlangle of 180 with the a-axis.
Since the sign of the dipole moment is undetermined from the Stark
effect measurements, there are two possible orientations of the dipole
moment. One forms an angle of IV 310 with the CN bond while the
other forms an angle of Iv 50. Because of the polar nature of the CN

bond, the smaller is the most probable.

 

3::
The dipole moments of these compounds using microwave techniques
are in excellent agreement with those determined using dielectric
measurements (27, 28).

31

The uncertainty in the slope of thelines is estimated to be 2%)
compared with 1% uncertainty in the very sharp OCS'lines, leading
to 1% uncertainty in the total dipole moment. The 2% uncertainty
in the sloPe may be attributed to a) broadness of the line; b) the
fact that frequency standard markers are not sharply detectable,

c) the variationin line voltage which causes a variation in the Stark

field seen by the molecule, and (1) meter inaccuracy.

32

 

 

3203863 0min... 222.5sz
.38 room snow 269.. Cs: >u2msomm... as... 88... Soon
- _ Til L. P h - pi . .L w . r. b o
1 O.
.. on
.A/
1 w
L
“a
. 9
8323\91 #13 x “JP—olflWQOlaw My».
. X
Ir
0-
.7
3

 

 

2

on

163

 

 

MN. .mmaOE

33

02 o.rwoo:m.m Jaw 20:62.45.

 

 

 

_ 5mg . IFS. .058. p “we? ._ .Auzw xozwaemmu- I. dose. 1.5:! m?!
w , i . - . - ~ . g . — b o

l O.

0 n

.on.
o . 1A;

. .6

9

.G

X
3.45%.... 312 x no! _. l u machm 0 1 W.»
-8

o

.. 2.

oo .8
1.2.
o 4‘ ,

NH HEDGE. ,

34

GE 5.2.8010 Man 2286sz
ammo“

mo 2 o >ozm=0wmm

m Snow. . room Know

1‘

endow 8.92..

P

/.,

I b I P

3m

  

a. roars. 4.19 x rmn 4| nmeJm

2-01 x 43.90 .L'IOA)

\T

Q"

 

 

 

 

10.

11.

12.

13.

14.
15.
16.

17.

18.

O‘WfiLflNr—I

35

LITERATURE CITED

. H. Townes, American Scientist, 43, 270, (1952).

. E. Cleeton and N. H. Williams, Phys. Rev., 45, 234 (1934).
. Bleaney and R. P. Penrose, Nature, _1_5__7_, 339 (1946).

. Bleaney and R. P. Penrose,, Phys. Rev., 12, 775L (1946).

. K. Co1€s and w. E. Good, Phys. Rev., 7_0_, 9791. (1946).

oowwoo

. H. Townes, and A. L. Schawlow, Microwave Spectroscopy,
(New York: McsGrawui-Iill Book Company, Inc. , 1955) Preface.

 

. W. Gordy, W. V. Smith, and R. E. Trambarulo, Microwave

 

Spectroscopy, (New York: JohnWiley and Sons, .nc. , 1953).

 

. M. W. P. Strandberg, Microwave Spectroscopy, (New York:

 

John‘Wiley and Sons, Inc. , 1954).

. C. H. Townes and A. L. Schawlow, Micrpwave Spectroscopy,

 

(New York: McGraw-«Hill Book Company, Inc. , 1955).

C. H. Townes and A. L. Schawlow, Microwave Spectroscopy,
(New York; McGraw-Hill Book Comp my, Inc. , 1955), Appendix III.

R. H. Schwendeman, J. Chem. Phys., 21, 9861.. (1957).

R. H. Schwendeman and V._ W. Laurie, "Tables of Line Strengths
for Rotational Transitions of Asymmetric Rotor Molecules, "
(New York: Pergamon Press, 1958).

S. Golden and E. B..Wilson, Jr., J. Chem. Phys., 16, 669,
(1948). ‘—

L. C. Hedrick, Review of Scientific Instruments, _2_(_)_, 781 (1949).

 

E. B. Wilson, Jr., Proc. Nat.-Acad. Sciences, 43, 816 (1957)

R. W. Shortridge, R. A. Craig, K. W. Greenlee, J. M. Derfer,
and C. E. Board, J. Am. Chem. Soc., _'_7_(_)_, 946 (1948).

J. D. Roberts and R. H. Mazur, J. Am. Chem. Soc., 13, 2509

‘ (1951).

F. E. Condon and D..E. Smith, J. Am. Chem. Soc. 29, 965 (1947).

36

. W. A. Lott and W. G. Chr;¥..~3tiansen, J. Am. Pharm. Assoc.,

__2_.7 12.5 (1938).

. P. W. Allen and L. E. Sutton, Acta... Cryst. 3 Part 1, 46, (1950).
. .‘J. P. Friendand B. P. Dailey, J. Chem. Phys., 39, 577 (1958).
. C. A. Coulson and W. E. Moffitt, J. Chem. Phys., 15_, 151

(19471.; Phil. Mag. 40, 15.9149}.

I..- -1

.. P. Trunel, Ann. China. .4) 12, 93 (1979‘...
. M T. Rogers, J. Am. Chem. Soc., 38, 843 (9146).
. V. W. Laurie, to be published.

W. S. 'W.‘..'..3«.:-s:>r. J H. Coldsts'n, and .7. W. Simmons, J. Chem.

-»—. -:I .7-...»-- 1
PET-'1' by 9 5‘6"? bulb \L 7&1470

. E. C. Hardis and C. P. Smy'th, J. Am. Chem. Soc., 65, 89 (1943).
S S

J. Chem. Soc., 1931]., 158 (1937).

”TIT 1111111111311 [[1111 ‘Efiflflflmflfilflfilflfi