close

Вход

Забыли?

вход по аккаунту

?

Microwave spectroscopy and quantum chemical calculations of triple bonded molecules and benzyl compounds

код для вставкиСкачать
Microwave Spectroscopy and Quantum Chemical Calculations
of Triple Bonded Molecules and Benzyl Compounds
Karissa Atticks Utzat, Ph.D.
University of Connecticut, 2004
Triple bonded hvdrocarbons
The rotational spectrum of 5-hexynenitrile and n-butyl acetylene,
respectively, was measured with high-resolution microwave spectroscopy.
Rotational constants and other spectroscopic constants were measured for the
conformations belonging to each molecule. Nuclear quadrupole splittings were
observed and characterized for each assigned conformer of 5-hexynenitrile. The
spectra of the singly substituted
isotopomers were assigned for two of the
four observed conformers of n-butyl acetylene. Also, a Kraitchman analysis was
done to characterize the substitution structures of the parent isotopomers for
these conformations. Ab initio calculations were carried out on the
aforementioned compounds and n-butyl cyanide in order to assess the relative
energy ordering of each compound’s conformations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Karissa Utzat Atticks -University of Connecticut, 2004
Benzvi-X Compounds
We have observed the microwave spectrum of benzyl alcohol and its OD
isotopomers at high resolution in a pulsed-jet Fourier transform microwave
spectrometer. The spectrum is consistent an asymmetric stable conformation
characterized by a C—C—C—O dihedral angle of approximately 60°. Tunneling
interactions strongly perturb the spectrum. Tunneling interactions between two
equivalent conformational minima is manifested by transitions split into doublets.
The observed splittings diminish upon deuterium substitution.
Previous studies on benzyl alcohol have suggested that a weak attraction
between the
k
electrons of the phenyl ring and the substituent —OH group as the
reason for the observed stable conformation. A theoretical analysis of the atomic
charges in benzyl alcohol suggests another possible explanation for the
observed structure. Atomic charges, generated by fits to the electrostatic
potential, indicate a relatively strong dipoie-dipole coupling between the —CH
group in the methylene side chain and the closest —CH group in the phenyl ring,
which results in a nearly planar orientation of the —OH group in the methylene
side chain with the phenyl ring.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Microwave Spectroscopy and Quantum Chemical Calculations
of Triple Bonded Molecules and Benzyl Compounds
Karissa Atticks Utzat
B.A., University of Connecticut, 1997
B .S .,
University of Connecticut, 1997
A Dissertation
Submitted in Partial Fulfillment of the
Requirements for the Degree of
Doctor of Philosophy
at the
University of Connecticut
2004
III
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
U M I N u m b e r: 3 1 6 6 0 2 1
IN F O R M A TIO N TO U S E R S
The quality of this reproduction is dependent upon the quality of the copy
submitted. Broken or indistinct print, colored or poor quality illustrations and
photographs, print bleed-through, substandard margins, and improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if unauthorized
copyright material had to be removed, a note will indicate the deletion.
UMI
UMI Microform 3166021
Copyright 2 005 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 4 8 1 0 6 -1 3 4 6
R eproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPROVAL PAGE
Doctor of Philosophy Dissertation
Microwave Spectroscopy and Quantum Chemical Calculations of
Triple Bonded Molecules and Benzyl Compounds
Presented by
Karissa Utzat
Major Advisor_
Robert K. Bohn
s / ' -x/'"
Associate Advisor
_
_
_
_
_
_
_
_
_
_
_
_
_
_<
r
\
„ - ^
_
_
_
_
_
_
_
_
_
H. Harvey Michels
Associate Advisor
Robert Birge
Associate Advisor_
W illiam Hines
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
for Chris
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table Of Contents
I.
General Introduction........................................................ 1
II.
Rotational Mechanics
III.
Fourier Transform
Microwave Spectroscopy............................
IV.
.8
.....
Tripiy Bonded Molecules.
A. Introduction.
15
...........................
25
.......27
B. Experimental Results and
Quantum Chemical Structure Calculations................................45
1. n-Butyl Acetylene..
..... ..45
a.
Experiment.
..................................45
b.
Results and Discussion..................................... 45
2. 5-Hexynenitrile..........................................
73
C. Ab initio Calculations......................................................
V.
..94
Conformational Analysis of Benzyl Alcohol
and Quantum Chemical Calculations
of Benzyl Alcohol and Benzyl Fluoride.................................... 112
A. Introduction............................
B. Experimental
C. Results
........................................
...........
....112
116
122
D. Theoretical Calculations................................................ 127
E. Discussion...............
...........................................1 3 5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I. General Introduction
Weak molecular forces can greatly influence molecular conformations.
Delicately balanced forces can act to mold the geometry of the molecule into its
most stable configuration. Although individually weak, the interplay of dispersive
interactions, dipole-dipole interactions, hydrogen bonding. Coulomb forces, and
steric interactions can determine a molecule’s three-dimensional structure. This
work reports on studies of molecular structure based on experimental rotational
spectroscopy and accompanying quantum chemical calculations.. Quantum
chemical calculations and experimental studies enjoy a complementary
relationship with each contributing a better understanding of molecular structure.
How experimental analysis aids theoretical development
Some levels of theory In quantum chemical calculations describe
conformational properties of molecules and predict their structures more
accurately than others. As it is the challenge for theorists to bring calculations of
structure into agreement with experiment, comparisons of calculated data with
experimental results can illustrate those methods that provide better numerical
agreement. The goal is to understand the limitations for each computational
method. The following discussion is an example where quantum chemical
calculations differ from results obtained from experiment.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0-Nitmsobis(tnfluoromethyl)-hydroxylamine: (CFsjaAfOWO
The conformational properties of this molecule are governed by the
interactions of lone pair electrons with the antibonding orbitals of vicinal bonds
(anomeric effect). Data from gas-eiectron diffraction and infrared spectroscopy,
studies show a mixture of 20(10)% trans-anti, 80% trans-syn with AG°=G°(transanti)-G°(trans-syn)= 3.3(12) kJ/mol at 298 K [1]. The possible conformations are
shown in Figure 1. Results from ab initio calculations are shown in Table 1.
Figure 1; Conformations of 0-Nitrosobis(trifluoromethyi)hydroxylamine
iN O
F,0W'^
FaC
C
FgC
trans-anti 20(10)%
W
trans-syn
CIS
\
W
F3<^'7
FaC
F 3 C
/>
I
1
I
/ / ^
o
ds-aoti
________ cis-syn__________
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
D
■C
D
—
i
O
o.
c
o
CD
Q.
■C
D
D
C
/)
W
o'
o
o
Table 1: Geometric Parameters and Relative Energies of the
Syn and Anti Conformatlonsof (CFalaNONO from Experiment
CD
O
O
■D
cq'
o
HF/3-21G
o
T|
c
CD
CD
"D
O
Q.
a
O
■D
O
g;
C
D.
Q
§
O
c
■C
D
D
3
(/)
(/)
HF/6-31GfcO
MP2/6-31Gfd)
GED
X-ray
syn
syn
anti
syn
anti
C-F
1.321(3)
1.314(4)
1.337
1.309
1.310
1.38
1.338
N-C
1.426(10)
1.431(4)
1.412
1.431
1.432
1.437
1.439
N1-O 1
1.410(15)
1.376(3)
1.417
1.371
1.371
1.382
1.408
O 1-N 2 (sp2)
1.572(21)
1.669(3)
1.527
1.401
1.402
1.717
1.598
N2~02
1.156(8)
1.120(4)
1.150
1.136
1.139
1.156
1.164
F-C-F
108.7(3)
108.0(2)
108.5
108.6
108.7
108.6
108.6
C-N-C
121.5(14)
118.2(2)
122.1
119.6
119.2
118.4
117.5
C-N 1-O 1
112.1(11)
109.6(2)
111.4
108.6
108.4
108.9
107.2
N 1-O 1-N2
107.6(19)
106.9(2)
111.2
114.4
108.4
103.1
103.7
O i -N2~02
115.743)
110.9(2)
113.0
115.0
109.2
110.1
108.9
AE (kJ/mol)
-3.3 (12)
-75
+2.5
—
-
-13.8
Distances in A, angles in (°), AE = E(syn) - E(anti) from vibrational analysis anc
ab initio calculations
00
The Hartree-Fock calculations do not show the two cis forms to be stable
conformers. The calculation with the larger basis disagrees with experimental
results by indicating that the trans-anti conformer is more stable. The electron
correlated MP2 calculation over-estimates the anomeric effect by predicting the
trans-syn conformer to be more stable than it really is; and, none of the
calculations predict the correct O— N(sp^) bond length. By investigating the
reasons as to why the discrepancies occur for various methods the theoretician
can better establish when a calculation is reliable for a particular system. In this
example, a more extensive study of both basis set and level of theory is
necessary.
We cannot expect theoretical calculations to accurately predict all
experimentally observed properties. Quantum chemical methods calculate
molecular structure in the context of the Born-Oppenheimer approximation.
These methods predict geometry, energies, and electron distribution with the
assumption that the nuclei are at rest. The calculated energy can readily be
corrected for zero-point vibrational motion of the molecule through second
derivative analysis. In addition, the vibration amplitudes are usually small
enough so that the differences between vibrationally averaged parameters, r,
extracted from experimental data and equilibrium geometries, re, calculated from
theory are negligible. Bond lengths that agree to within 0.03
A, bond angles that
agree to within 3°, dihedral angles that agree to within 10°, and energies that
agree to within 2 kJ/mole, with experimental results are generally considered
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
satisfactory [2]. Better agreement can generally be achieved by using a higher
level of theory.
How theoretical calculations aid experimental analysis
By providing geometric models and information on conformational space
(i.e., minima on the potential energy hypersurface), an experimentalist can use
theoretical models to calculate rotational constants which are useful for predicting
and assigning rotational spectra. Furthermore, estimates of dipole moments can
be used to predict the intensities of spectral lines. Quantum chemical
calculations also fill in information that may be missing in the rotational spectrum.
Since rotational spectroscopy measures the average of reciprocal moments of
inertia, <r'^>, over a vibrational state, observed spectra are insensitive to atoms
close to a principal axis. Theoretical calculations provide useful information on
such atoms, complementing experimental information on the structure.
Theoretical calculations also provide information on vibrational amplitudes
from theoretical force fields. If vibrational amplitudes are small, then the rigid
rotor model remains accurate. If the moments of inertia no longer represent a
rigid framework but an average over the vibrational motion, then the molecular
structure determined from the spectrum may differ from the equilibrium
configuration as determined by theory.
In conclusion, although computational chemists and experimentalists
enjoy a mutually beneficial relationship whereby each can look to the other to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
improve his craft, it should not be overlooked that the geometrical parameters
that come out of their respective fields may have different accuracies and
interpretations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Literature Cited
1. Ang, How Ghee; Klapdor, Martin Frank; Kwik, Whei Lu; Lee, Yiew Wang;
Mack, Hans-Georg; Mootz, Dietrich; Poll, Wolfgang; Oberhammer, Heinz.
0-Nitrosobis(trifluoromethyl)hydroxylamine; Unexpected Conformational
Properties and an Unusually Long (CF3 )2 NO-NO Bond. A Combined
Study of the Gaseous and Solid States. J. Am. Chom. Soc. 1993, 115,
6929-6933.
2. Oberhammer, Heinz. Molecular Structures and Conformations:
Experiment and Theory. J. Comput. Chem. 1998 , 19 (2), 123-128.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
II. Rotational Mechanics
Energy transitions in the microwave region of the electromagnetic
spectrum are chiefly manifested from rotational stationary states. Classically the
angular momentum of a rigid collection of particles is given by [1]
P = ! ®csj
(1)
whereI is the moment of inertia tensor and lu is the classicalangular momentum.
The moment of inertia tensor written in vector form is:
I = Ixx / / + ixy i j + l y x j i + \ y y j j + lyz j k
+ Izx k l + Izy k j + Izz k k
(2 )
where,
Ixx
= Im (/+ z ^ ): ly y = Im ( z W ): l^z = Im (x ^ + /):
^3^
Ixy “ lyx ~ "Zmxy; Izx " Ixz “ -Zmxzj lyz = Izy ” -Zmyz,
where m is the mass of a particular particle; x, y, and z are the particle
coordinates in a Cartesian coordinate system with its origin at the center of mass
and the sum is over all particles. The center of mass is chosen as the origin
because then the kinetic energy may be written as a sum of the kinetic
translational motion of the center of mass and the kinetic energy of the motion
relative to the center of mass, thus allowing translational motion to be treated
separately from rotational motion.
A coordinate system is chosen in which the moment products vanish and
the diagonal elements, called the principal moments of inertia, are left. In this
system, x, y, and z are transformed, through a diagonalization procedure, to a, b,
and c, though not necessarily in that order.
8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
^xx
I
ly x
ly y
\
xz
I
y.
yL Izy Izzy
"la
0
0
^0
0
o'
0
(4)
Ic;
Molecules are classified by their moments of inertia around these three
perpendicular axes. For molecules classified as symmetric top rotors, two of the
Principal axes of the prolate rotor:
lb = Ic > la.
c
Principal axes of the oblate rotor:
lb =
Ic <
la .
Figure 1: Shapes of prolate and oblate
symmetric rotors.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
three moments of inertia are equal. A symmetric top rotor may be prolate or
oblate, depending on which moments of inertia are equal. A prolate molecule is
cigar-shaped where the two equivalent axes, b and c, are perpendicular to the
molecular symmetry axis, a and lb = ic > la- See Figure 1. In an oblate rotor, the
geometry of the rotor is of a pancake shape with the moment of inertia around
the symmetry axis larger than that around the other two axes (Figure 1).
In the principal axis system the rotational kinetic energy,
E
r
,
of the
molecule is a sum of the kinetic energy around the three axes.
E
1
=
r
> 2
W
I
W
( la ® a + lb ® b + lc ® c )
2
-j /^d2
4.
2V
p2
p
" b b
' '
(6)
a
For a freely rotating body, Euler’s equations give.
dP,
dP.,
- + 0 3 ,P ^ -f0 ,P ,,= — ^ +
dt
dt
V^b
dPuj,
™
dPuK
— -f 0)_p.- -CO-P„ = — — +
dt
dt
Jc
r jP
=
0
(7a)
0
(7b)
Kj
lay
HP
dt
dt
(7c)
+
V^a
lb y
Multiplying (7a) by Paa, (7b) by Pbb, and (7c) by Pec and adding gives
+ Rbb
bb
!+ P.,
dP.,
= 0,
which, upon integration and multiplication by two gives
+ P^ = a constant = P^
(9a)
ind
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Er =-
" P i . + _bb.
P i + Lcc^
V
= a constant.
(9b)
a
(9a) is the equation of a sphere with radius P; (9b) is the equation of an ellipsoid
with principal semi-axes of (2*la*ER)’"'', (2*ib*ER)’'^", and (2*^c*ER)’'^^ Only if the end
points of P are along the intersection of the sphere and the ellipsoid can Paa, Pbb,
and Pec satisfy both equations. The vector P is fixed in magnitude and direction
(Eqn (9a)), while the energy ellipsoid is constrained to the molecule which means
the molecule must rotate in such a manner that the P continues to
Figure 2: Restriction of motion of a prolate
symmetric top where the allowed values of
P are illustrated in the intersection of the
momentum sphere and the energy
ellipsoid.
V, p
X
1,
X T
w
■ '/
X
— -------- M
Ref. 1, p. 11.
terminate on the surface of the ellipsoid. This is shown for the prolate symmetric
top case, where two of the axes of the energy ellipsoid are equal, Ib = Ic, in
Figure 2. The motion of the prolate symmetric top may be seen by considering
two cones. The cone surrounding the P vector in Figure 3 is fixed in space while
the second cone has Pc running through its center and rolls around the first
cone.
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Ref. 1, p. 11,
I
Figure 3: Classical motion
of a prolate symmetric top.
By setting lb = Ic in Equations 7a to 7c, it is shown that Paa, like P, is a constant of
the motion of a symmetric top:
dPag
= 0; P., = a constant.
dt
(10)
For a prolate symmetric rotor the moment of inertia about the symmetry
axis, la, is elongated, and the moments of inertia around the other two axes are
equal. Equation 9b may then be written as
P"
1
v 2Ia
1
(11)
2ley
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Quantum mechanics requires that the motion of angular momentum can take
only discrete values. Consequently, the square of the angular momentum is also
quantized and equal to
J (j+\y
, where J = 0,1,2,...any positive integer.
Similarly, the component of angular momentum along some axis, in this case the
a-axis is chosen (Figure 2), is also quantized so that
= -----r-, K = 0, ±1,...,±J.
Defining rotational constants. A, B, and C as
the energy may be more succinctly written as
(13)
=2BJ(J + 1 )+ (A -B )K ^
The resulting spectrum consists of lines, equally spaced by 2B, and displaced
from each other by a factor proportional to A-B for a prolate top and C-B for an
oblate top. The selection rules are AJ = ±1, AK = 0.
The projection of angular momentum along the a-axis, or symmetry axis,
has two values, one for clockwise rotation, the other for counterclockwise
rotation, corresponding to quantum numbers Ko and Kp. For a symmetric rotor,
these rotations are degenerate: Ko = Kp = K. This degeneracy is lifted for an
asymmetric rotor such that we speak in terms of the projection of K along the
prolate axis, Kp and the projection of K along the oblate axis, Ko The energy
levels are not as symmetrically spaced as in the symmetric top case. Similarly,
in expressions of the energy, the degeneracy in the variable B is removed. The
energy levels are not as symmetrically spaced as in the symmetric top case.
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Since there is no analytic energy function for an asymmetric rotor, it is
useful to quantify the degree of asymmetry in a molecule. Ray’s asymmetry
parameter,
k
,
is one variable which does this:
2B-A-C
K = -------------A -C
, ^
^ ,
-1 < K < 1
(14)
^ ’
K reduces to -1 in the prolate case, A > B = C; and reduces to +1 in the oblate
case, A = B < C. The most asymmetric molecules have k values close to 0. The
selection rules for an asymmetric rotor are AJ = ±1, AKp = AKo = 0. All of the
molecules discussed in this thesis happen to have
k
values near -1 and are
described as nearly prolate symmetric top molecules.
Reference 1: Gordy, Walter; Cook, Robert L. Theoretical Aspects of Microwave
Rotation. In Microwave Molecular Spectra, 2"*^ edition; Weissberger, Arnold Ed.;
Techniques of Organic Chemistry, Chemical Applications of Spectroscopy Part II;
Interscience: New York, 1970; pp 7-35. This discussion is drawn primarily from
this chapter and my notes which have accumulated over the years.
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ill. Fourier Transform Microwave Spectroscopy
The data obtained for the rotational spectra analyses of the molecules
described in this thesis have been acquired from the cavity pulsed Fourier
transform microwave spectrometer, located in Professor Stewart Novick’s
laboratory at Wesleyan University. This spectrometer is the major facility of the
Southern New England Microwave Consortium, a collaboration between
Wesleyan University and the University of Connecticut. This instrument
combines the properties of a standing wave Fourier transform spectrograph, in a
Fabry-Perot cavity, with a pulsed supersonic molecular beam source. The cavity
is tuned to resonate at a given frequency. A pulse of gas is injected into the
cavity. While the gas pulse traverses the cavity, it is irradiated by a short pulse of
microwave radiation centered at the resonant frequency. If a molecular rotation
lies within the bandwidth of the microwave pulse frequency, radiation is absorbed
by the molecules and the polarized gas emits its resonant frequency, which is
detected in a superheterodyne detector and Fourier transformed to give the
resulting spectrum. The following basic hardware elements of the experiment will
be discussed in the following pages:
1. A resonant Fabry-Perot cavity
2. The pulsed molecular beam
3. Free induction decay detection
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The Fabry-Perot Cavity
The Fabry-Perot cavity uses spherical mirrors to create a resonant
standing wave at a given frequency. The design allows for a high Q-value, which
for the Wesleyan instrument [1] is on the order of 10^ The Wesleyan instrument
places two spherical mirrors facing each other in a high vacuum chamber
Torr). The cavity mirrors are machined aluminum; have a diameter, D, of 35.6
cm; a radius of curvature, b, of 84 cm; and an adjustable separation distance, d,
of 73 to 78 cm.
Tuning the cavity involves adjusting the mirror distance until a standing
wave is set-up for the chosen frequency. In the Wesleyan spectrometer, one
mirror is stationary and the other moves on four stainless steel rods. Powered by
an Oriel Encoder Mike, the mirror is moveable over a range of 50 mm [1]. The
spectrometer operates in the 5.5 to 26.5 GHz range.
The Pulsed Molecular Beam
Molecules at room temperature are found in many different rovibrational
states and can result in hundreds of spectral lines from just one compound. At
room temperature the fraction of molecules occupying a particular rotational state
is quite small leading to weak, complex spectra. By lowering the temperature of
the molecules, fewer states are populated resulting in a larger fraction of
molecules in each state and a simpler spectrum.
In the microwave experiment, a monatomic carrier gas is used to sweep a
small amount of sample through a nozzle and into the Fabry-Perot cavity
chamber. Initially the gas sample is a static reservoir behind the nozzle. The
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
temperature of the gas and the average particle velocity are related by the
8kT
Maxwell distribution: v = J
, where k is the Boltzmann constant, T is
\ n: m
temperature in kelvin, and m is the particle mass. The width of the velocity
distribution is greater at higher temperature.
In a supersonic jet, the formerly random kinetic energy and velocity
distribution are transformed into a flow velocity, v, of the jet and a small fraction
of the gas remains with random velocities within the jet. The flow velocity [2] is
given by:
V =
1 -1
m ( r - i)
Ma,
where y is the heat capacity ratio, 5/3 for a noble gas and m is the mass of the
noble gas particle; T is local temperature; To is the temperature of the source
(i.e., gas before expansion); M is the Mach number; and, a is the speed of
sound. Although the temperature is characterized by the width of the velocity
distribution centered on the flow velocity, the actual definition of temperature is
independent of the value of that flow velocity because, provided the flow rate is
high enough, the flow velocity does not affect the chemistry and physics of the
individual gas particles,
The reduction in gas temperature is described in terms of translational and
rotational cooling due to narrowing of the relative velocity distribution in a
supersonic expansion. Consider a component of velocity perpendicular to the
direction of flow, namely the transverse velocity component. The gas atoms are
separated according to their transverse velocities; that is, if there are no
17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
collisions between atoms, those atoms in the center of expansion (small
transverse velocities) stay in the center of the expansion, while those with large
transverse velocities move away from the center of expansion. Atoms close to
the nozzle, where gas pressure is high, encounter frequent collisions. These
collisions equilibrate the relative velocity components.
A small amount of sample, ~ 1%, to be studied spectroscopically is mixed
with a large amount of monatomic carrier gas. Neon and argon are typically the
carrier gases used in our experiments. As the mixture expands through the
nozzle, it translationally and rotationally cools. Near the nozzle, there are many
collisions of the noble gas particles and the smaller number of sample molecules.
The rate of collisions decreases as the mixture flows downstream from the
nozzle because the average distance between carrier gas atoms and sample
molecules increases. Consequently, most cooling occurs near the nozzle. For a
circular nozzle diameter with D = 1 mm and a pulse time of 3 ms, at 300 K and 1
atmosphere, about 4x10^® particles are released from the nozzle to the cavity [2].
With the sample mixture traveling at 3.8x10'^ cm(s'^), one tenth of this total is
inside the region of the microwave standing wave; and, of that, only ~1 % are
sample molecules [2] Energy exchange between rotational degrees of freedom
to translational motion occurs efficiently. Consequently, the using the
approximation that the rotational temperature is close to the translational
temperature is valid^. The translational temperature within the jet is given by
, where p is pressure in atmospheres and D is nozzle
T = -------- ^
1+ 5896(pDf®
diameter in centimeters. Thus, if To = 300 K, D = 0.1 cm, p = 1 atm, then T « 0.3
18
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
K. in our experiments, the effective temperature is about 2 K. Higher energy
conformers present in the room temperature gas do not relax to lower energy
conformers in the jet if the energy barrier between them exceeds about 4.1
kJ/mole [3]. Such high energy barriers prevent relaxation to other lower energy
conformations of the molecule. If the barrier is lower, collisions may bump the
molecule over that energy barrier to a lower energy form. If the barrier is higher,
the molecules are trapped in the higher energy conformation.
The effective temperature along the center of the jet can be modified by
adjusting the nozzle diameter, or by increasing the backing pressure behind the
nozzle. Both actions allow for more collisions close to the nozzle by increasing
the amount of gas entering into the cavity. Since the supersonic expansion
works in a vacuum, however, these molecules must be removed from the cavity
after each pulse to maintain the vacuum. Therefore, the size of the pumping
system ultimately limits the repetition rate to pulse the supersonic jet. The pulse
rate used in the experiments described in this thesis is 5 Hz.
In our experiments, the jet cools the sample molecules to about 2 to 3 K.
A co-axial nozzle is used to maximize the number of molecules in the high power
regions of the microwave resonance cavity.
19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
standing Wave
Molecular
Beam
Microwave
Pulse
Detector
Aluminum
Mirrors
Figure 2: Cartoon of Microwave Cavity
with Co-axial Nozzle
Molecular Beam and Microwave Pulse
We choose a frequency and adjust the mirrors to achieve resonance. A
puff of the sample gas mixture is injected into the cavity. While the gas pulse
traverses the cavity, a short pulse of microwaves is turned on at the standing
wave frequency. If this frequency resonates with a rotational transition, then the
gas absorbs that radiation polarizing the sample into a non-equilibrium
distribution of energy states. As the excited molecules relax back to the
equilibrium distribution, they radiate. These photons are measured as a free
induction decay which is Fourier transformed into the observed spectrum (Figure
3).
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 3: Free-induction decay signal and Fourier transform of a
Doppler doublet signal
126516.401
+ .S ,
UV
0.0
-
0
.
us
62.400
31.250
100
Free-induction decay signal
1.0
I
0. 0
------r- MHz
1 8 0 5 0 .7 6 0
1 8 0 5 3 .2 6 0
1 8 0 5 5 .7 6 0
Fourier transformed signal
As seen in the Fourier transformed signal, the transition is a doublet. As the
excited molecules decay, they emit radiation isotropically. Half of the relaxing
molecules emit in the direction of the molecular beam and give a signal vi = vo +
Av, where vo is the unshifted frequency and Av is the Doppler shift. Similarly, the
21
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
other half of the molecules emit in the direction going against the molecular beam
and the observed signal is vi = vo - Av. The rest frequency is the average of
these two Doppler peaks:
.
The Eiectronic Components and Fourier transform method
All of the spectrometer operations are controlled by the computer software
written by Dr. Jens-Uwe Grabow of the University of Hannover, Germany. User
controlled variables are the following:
1. Gas pulse width: how long the gas valve is open, ca. milliseconds.
2. Microwave pulse width: typically 1 to 2 ps.
3. Gas-microwave delay: delay between molecular gas pulse and
microwave pulse. This time allows the pulsed gas to expand and cool
before irradiation.
4. Microwave .delay: time between the end of the microwave pulse and data
collection.
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 4: Time Elements of Microwave Spectroscopy Experiment
Nozzle:
opens
closes
Microwave pulse:
starts
ends
- i - O
-
Source width
Expansion width
Delay
Collect Data
Microwave
pulse
Critical electronic components are the following:
1) an oscilloscope used to track the cavity mode positions;
2) a frequency synthesizer to generate the microwave radiation;
3) an oscillator to provide the radio frequency signal, which is then used to
mix microwave frequencies in the detection process;
4) a timing box to ensure all processes occur in the correct order at the
correct time; and
5) a power box to supply stable power to the microwave circuit.
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
References
1
) Munrow, Michaeleen. Fourier Transform Microwave Spectroscopy of
Weakly Bound Complexes, Multi-Conformational Molecules, and
Interstellar Radicals (van der Waals Complexes) Ph.D. Thesis, Wesleyan
University, Middletown, CT, 1998.
2) Balle, I . J.; Flygare, W. H. Fabry-Perot cavity pulsed Fourier transform
microwave spectrometer with a pulsed nozzle particle source. Rev. Sci.
Instrum. 1981, 52(1)33-45.
3) Ruoff, R.S; Klots, T.D.; Emilsson, I . ; Gutowsky, H.S. Relaxation of
conformers and isomers in seeded supersonic jets of inert gases. J.
Cham. Phys. 1990, 93 (5), 3142-3150.
24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IV. Triply Bonded Molecules
A Note on Conformations and Nomenclature
The literature is full of different methods for naming and describing
molecular conformations of the small hydrocarbons that will be discussed in the
subsequent pages. To simplify the situation of several disparate naming
schemes in the literature, a general naming scheme described below, has been
developed for structurally similar molecules.
Consider propane, or propyl, with two substituents, X and Y (Figure 1).
The conformational degrees of freedom that result arise from rotation around two
dihedral angles: 9i (XCCC) and
02
(CCCY). X and Y can be anti (A) or gauche
(G) about each of their respective angles: If X = Y, then the conformational
possibilities are AA, GA, GG-cis, and GG-trans. If X
Y, then AG and GA are
not identical and one more conformational degree of freedom is added to the
present possibilities.
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 1: Triple Bonded Molecules Studied in this thesis:
•
n- butyl acetylene: X = -C=CH; Y = -CH3
^
n-butyl cyanide:
*
5-Hexynenitrile: X = -C=CH; Y = -C=N
X
= -C=N ;
Y
= -CH 3
T
0 1
Possible Conformations
•
AA: X anti, Y anti
•
GA: X gauche, Y anti
•
AG: X anti, Y gauche
•
GG-trans: X and Y gauche but on
0 2
opposite sides of propyl group
•
GG-cis: X and Y gauche on same
side of propyl group
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A. Introduction
Several groups have reported studies of the molecular structure of npropyi acetylene (1-pentyne) and isoelectronic n-propyl cyanide (butyronitrile)
molecules [1-3, 9-10, 12-14], An interesting feature of these compounds is that,
in contrast to their alkane prototypes, the anti and gauche conformers are nearly
equal in energy. Theoretical calculations (that involve electron correlation)
reported for these molecules agree that the gauche forms are slightly more
stable. However, experiments on n-propyl acetylene show the gauche form to
be more stable in the gas phase [9] and the anf/form more stable in both the
solid phase and in noble gas solutions [14]. This gauche form stability has been
attributed to an attractive interaction between the n-electrons of the triple bond
with the hydrogens located on the terminal methyl group. In recent microwave
studies, Bohn et al. [3] have analyzed three conformers of n-butyl cyanide and
predicted that the gauche-anti (GA) and anti-anti (AA) conformers should lie
lower in energy than the anti-gauche (AG) structure, a conclusion consistent with
results from n-propyl cyanide studies [9]. The cis and trans gauche-gauche (GG)
forms of n-butyl cyanide were not observed in these studies.
Gaseous n-butane is 2.80 ±0.39 kJ/mole more stable in the anti
configuration than in the gauche form [4]; but, by replacing the terminal methyl
with an electron-withdrawing substituent, the gauche conformer becomes the
more stable form. This observation is similar to the structural preference,
described by the generalized anomeric effect, for conformations about the bond
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
C—Y, in the system X—C—Y—C, to occupy gauche positions [5], In a
microwave study, it was found that the gauche form of n-propyl fluoride was 1.97
±1.30 kJ/mol more stable than the anti form [6 ]. This result Is consistent with
results of a temperature dependent infrared spectral study where n-propyl
fluoride was dissolved in both krypton and xenon. AH was found to be 1.24
±0.07 kJ/mol in krypton solution and 1.16 ±0.07 kJ/mol in xenon solution, with
the gauche form being the more stable conformer in both noble gas solutions [7].
Similarly, conformational and structural studies on n-propyl chloride and n-propyl
bromide show the gauche form to be more stable than the anf/form: AH= 0.62
±0.06 kJ/mol and 0.86 ±0.08 kJ/mol, respectively in the 123 to 218 K
temperature range [8 ].
In an effort to explain the relative stability of these conformers, ab initio
calculations of n-propyl chloride and n-propyl bromide were conducted at the
Hartree-Fock and Moller Plesset theory levels with a variety of basis sets (Figure
1).
28
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 1: Relative Energy calculations of n-propyl chloride
and n-propyl
bromide
1.40 -
0.90 MP2/6-31G(d)
0.40 Ai
- 0.10 -
m
MP2/6-311G(d,p)
MP2/6-311+G{d,p)
X
MP2/6-311+G(2d,2p}
1 AE=0
HF/6-31G(d)
C -0 .6 0 '
O) 1 .1 0
^
-
•
- 2.10 -
■
O n-prop^ chioride
Un-propyl bromide
-2.60
Calculation
At the Hartree-Fock level of theory and with a small basis set, the anti form of npropyl chloride is predicted to be the more stable conformer. Larger basis sets
that include electron correlation in the calculation method show the gauche
configuration to be the more stable form for both molecules. The MP2/6-31G(d)
calculation is the exception. Here the anf/form is calculated to be more stable
than the gauche form, but this result may be due to the inappropriate basis set
used. Polarization functions on hydrogen are necessary to model this system
correctly. Calculations of AE with electron correlation and large basis sets show
results consistent with the AH experimental data and support the theory that
when an electronegative atom or group is attached to a propyl fragment gauche
is the preferred configuration of that group.
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
When X in propyl-X does not have the electron-withdrawing inductive
strength of a halogen, there is controversy among researchers as to which
conformer is more stable, but the differences are small. For example, an
electron diffraction experiment suggests that the gauche form is more stable if
the electron releasing group -CH3 is replaced with the electron withdrawing group
-C=N [9]. In contrast, authors of a microwave experiment contend that anti is the
more stable configuration [10]. There is a similar controversy among
researchers that have studied the effect of replacing the terminal methyl with
acetylene. These results are summarized in Table 1.
30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 1; Conform ational Energies of some Substituted Propyl Derivatives
Molecule
Butane
Propyl
fluoride
Propyl
chioride
Propyl
bromide
Propyl
cyanide
Propyl
acetylene
'kJ/mol
Stable
Form
]{Anti-Gauche)]'
% Anti (or comment)
anti
AG° = -2.1(9)
anti
AH° = -2.8(4)
54 ±(9)% at 300 K
7 temperatures
ranging from 223 K
to 297 K
gauche
AE = 2.0 ±1.3
gauche
AH° = 0.62(6)
28% at 298 K
gauche
AH° = 0.86(8)
26% at 289 K
anti
AE = 1.1(3)
at 233 and 288 K
gauche
gauche
AG° = 1.0(3)
AG° = 0 2
IR, Raman [13]
Equal
AE = 0
!R (dissolved in
liquid Xe)
[1]
anti
25 ± 6 %
31 ±5% 297 K
Liquid and gas
phase data.
173 to 213 K, 3
conformer pairs
173 to 218 K, 7
conformer pairs
Experiment
G E D [11]''
Gas Infrared
Spectroscopy
[4]
Microwave
Spectroscopy
[6]
Infrared
Spectroscopy
(dissolved in
Kr) [8]
Infrared
Spectroscopy
(dissolved in
Xe) [8 ]
Microwave and
mm Wave
Spectroscopy
[101
GED [9]
GED [12]
AH° = 1.3(3)
AH° = 0.60(7)
Microwave data
gauche
AE = 0.32 ±0.43
.............
[2]
Gas Electron Diffraction
Wodarczyk and Wilson [2] identified two rotamers of n-propyl acetylene
(1-pentyne) in its microwave spectrum. They identified these rotamers as anti, 0
= 180°, and gauche, 0 = 115° (Figure 2). The relative energy difference between
31
R eproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the conformers was determined by making careful intensity measurements at dry
ice temperatures and then taking ratios for specific transitions. The gauche
conformer was found to be 0.32 ±0.43 kJ/mol marginally more stable than the
anti form. The large error is due to the conservative set of error limits for
measuring the area under the bands the authors imposed on their calculation.
As there were large errors associated with the small line half-widths, it was
difficult to determine AE= {Egauche - EantI)The authors suggest that a dipole-dipole interaction stabilizes the gauche
conformer relative to the anf/form. As alkyl groups have an electron releasing
inductive effect, the electropositive terminal methyl is electrostatically attracted to
the 7t system of the triple bond. The authors note that the 5° degree discrepancy
of the measured gauche dihedral angle from the nominal
120
° gauche dihedral
angle is probably due to steric repulsions. The Ca to C5 distance of 3.05 A is
less than the sum of the van der Waals radii: r(CH3 ) =
2 .0
A and r{GsC) =
1
.4 A.
Although, since the acetylene group is more elliptical than spherical, a smaller
radius may be operative in the direction of the terminal methyl.
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 2: n-propyl acetylene:
inductive effect is shown to be
attractive when 9 corresponds to
the gauche conformer.
In order to test their hypothesis that the positive end of the methyl group is
aligning itself toward the negative ix-system, thereby stabilizing the gauche form,
the authors calculated the dipole-dipole interaction energy using bond moments.
Although this calculation is approximate, the interaction energy was found to be
on the order of hundreds of calories, which is about the observed difference in
stability of the two forms.
Variable temperature studies in liquid Xe, carried out by Durig et al. [1]
show that the preferred conformation of n-propyl acetylene is anti. Using only
three conformer pairs, AH° is found to be 1.35 ±0.31 kJ/mol in the temperature
region 173 to 213 K. To measure enthalpy differences, the van’t Hoff equation
was used; - \ r \ K ■
AH
RT
AS
■. In a plot of -ln(K) versus 1/T, the slope is
R
proportional to the enthalpy difference. Because conformer peaks are better
33
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
resolved and intensities are more easily measured in noble gas solutions than in
the gas phase, n-butyl acetylene was diluted in liquid Xe. The authors believe
that as the dipole moments of the two conformers and their respective molecular
volumes are similar, the results obtained in the noble gas solution are easily
extended to the gas phase. The fact that the gauche conformer dominates the
far-infrared region of the spectrum is attributed to this configuration’s larger
dipole derivative and not to the possibility that it is lower in energy. The authors
do not believe that replacing a methyl group (i.e. butane) with an acetylene group
could alter the conformational stability by such a large amount as to switch the
relative energy ordering. They state the following:
However, it should be noted that for the n-butane molecule the trans [anti\
form is more stable by 234 ±33 cm'^ [2.80 ±0.39 kJ/mol] than the gauche
form in the gas phase. Thus, it would be surprising if the acetylenic group
would alter the conformational stability by such a large amount that the
gauche conformer is the more stable form in the gas phase.
[Ref. 1, p. 5994]
The same authors, in a later paper, report on a similar experiment in
which, instead of xenon, a solution of krypton was used [14]. In krypton, the
conformer peaks are better resolved and the area under them is more easily
measured. Ten sets of conformer pairs were used to determine the enthalpy
difference. The results of this temperature study are consistent with the results
observed in the xenon study: the anti conformer was observed to be more stable
by 0.54 ±0.05 kJ/mole in krypton (123 to 168 K) and 0.60 ±0.07 kJ/mol in xenon
(173 to 218 K). It is interesting that Durig comes up with almost the same
34
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
number as Wilson but with an entirely opposite meaning and that these new
enthalpy values differ significantly from the value obtained in the first experiment.
Ab initio caiculations at the MP2 level report results contrary to the
experimental conclusions. Figure 3 is a bar graph of the various calculations
reported for n-propyl acetylene by this group.
Figure 3; Conformers of n-propyl acetylene*
1,5
n
O
E 1.0
MP2 calculations
lo ,=
o
O
0.0
o
6-31G(d)
« -0.5
(Q
O
O
>>
6-31G(d,p)
J
HF calculations
(d)
r
6 -3 |d i d.p)
6-31
i(d,p)
6 -3 1 1 ';- 2d,2p)
311+®2df,2pd)
1
1^.
•
1
I '■
- 1 .0
Ui
m
r ...
[
-
,
1-5
>
JS -2.0
)
0
-2.5
Calculation
’ D u rig , J . R . e t a l.
J. Mol. S truct. 2 0 0 i, 5 6 0 . 2 4 7 - 2 6 9 .
J. Phys. Chem. A 1 9 9 7 . 101 ( 3 4 ) ,
D u rig , J a m e s R . e t a l.
5 9 8 7 -5 9 9 6 .
The results shown in Figure 3 are very informative. At the Hartree-Fock
level, regardless of basis set, the gauche conformer was found to be less stable
than the anf/form. MP2 calculations, regardless of basis set, show the gauche
conformer to be the more stable form. In the n-propyl chloride and n-propyl
bromide report [8 ], Durig et al. comment that the ab initio calculations are
consistent with the experimental results. In this paper however, Durig et al. [14]
explain the discrepancy as being unable to rely on ab initio MP2/6-31G(d)
35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
calculations to provide correct conformer stability when the enthalpy differences
between conformations is less than 1.80 kJ/mol. Regardless of the calculation
method, the 6-31 G(d) basis set is Inadequate for this molecule; Polarization
functions on hydrogen are needed to accurately describe this system and
correctly predict its properties. Durig et al. indicate that the MP2/6-311+G(d,p)
level of theory gives excellent estimates of C— H and C— C structural parameters
for monosubstituted hydrocarbons. They discredit all other M P 2 calculations for
reporting that the C=C bond lengths are too long; however, the authors did not
consider other MP2 calculations with different basis sets. Table 2 is a portion of
Table 3 in Durig’s paper and shows the C=C bond length from different
calculations and experiments. Although r(C=C) is calculated to be too long with
the smallest basis set ( 1 . 2 2 0 A), as the basis sets get larger, the bond length
becomes smaller and approaches the experimental results. At the largest basis
set used, 6-311+G(2df,
2
pd) the triple bond length is calculated to be 0.003 A
longer than that determined by the microwave spectroscopy experiment, a result
that is well within the acceptable ab initio calculation error limits [15].
An electron diffraction study by Traetteberg [9] et al. report similar
computational problems regarding the C=N bond length in gaseous n-propyl
cyanide (butyronitrile). A series of ab initio calculations were carried out to
determine the effect of calculation method and basis set on the G=N bond
length. It was concluded that most geometric parameters were insensitive to the
choice of calculation and basis set except for the C=H bond length, which ranged
from 1 .1 3 to 1 .1 8 A (Table 3).
36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7
J
CD
■—
Di
O
o .
c
o
CD
Q .
■D
CD
(/)
W
o'
o
o
—ty
Table 2: Comparison of calculated”[14] and experimentally [2,12] determined r(C=C) bond lengths in (A)
CD
O
°D
■
‘<
cq'
of n-propyl acetylene
MP2/
6-31 G(d)
6-311+Gfd.Dl
6-311+G(2d.2D)
6-311+G(2df.2Dd)
GED [21
MW [121
o
o
CD
r{C=C)
CD
—
i
CD
—
Oi
■D
o .
c
a
o
o
■—
Di
o
CD
Q .
■CDD
C/)
C/)
03
■-4
anti
gauche
anti
gauche
anti
gauche
anti
gauche
gauche
gauche
1.219
1.220
1.219
1.219
1.214
1.214
1.212
1.213
1.2220(7}
1 .2 1 0
CD
■—
Di
O
o .
c
o
CD
Q .
■CDD
C/)
(/)
Table 3: Comparison of calculated and experimentally determined gauche r(C=N) bond
o
°
length (A) of n-propyl cyanide [9].
^
cq
-
.......
'
r(C=N)
O’
CD
CD
■D
O
Q.
C
a
o
■D
O
CD
Q.
■CDD
C/)
C/)
w
CD
HF/6-31G(d)
B3LYP/6-31G(d)
MP2/6-31G(d)
MP2/6-311G(d,p)
MP2/6-311++G(2d,2p)
GED
1.13
1.16
1.18
1.17
1.17
1.1576(10)
The relative energy of the two rotamers was also calculated at different
levels of theory and basis sets (Figure 4). Again, the HF calculation predicts the
anf/form to be the more stable configuration and the MP2 methods, which
include electron correlation, predict the gaoche conformer to be more stable.
Figu re 4 : C o n fo rm ers
of n -p ro p y l cyanide*
o
M P2 C alculations
E
0.5 -
O
C
o
O
§
(B
6-31 G(d)
6-31G (d)
HF
B3LYP
6-311
-0.5
<3
O
E
U1
0>
>
TO
TO
EC
Calculation
•Tratteberg, M. et al. J. Mol. Struct. 2000, 5S6, 189-196.
The operative interaction of n-propyl acetylene’s gauche conformer,
described by Wodarczyk and Wilson, is used by Traettteberg et al. to explain the
relative gauche stability of gaseous n-propyl cyanide as determined from
electron diffraction data (Figure 5). As in the case of n-propyl acetylene, electron
diffraction studies show that the gauche conformer is clearly more abundant at
297 K, with a reported conformational mixture of 68.6(4.5)% gauche to 31.4%
anti. An MP2/6-31G(d) calculation agrees with the experimental data, reporting
39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
73.0% gauche to 27.0% anti conformationai mixture, but again, this basis set is
inadequate for this system.
(CH,C
I/
I
\
\
‘■"'vx
'" x
-0.07
X c ’
\
-0 .2 2 \
^ +0.07
Figure 5: Gauche conf ormer of n-propyl cyanide. This
projection shows the gi2 ometry of the nearly coplanar
G=N and C'^-H bonds that may be essential for the
stability of this conforme r. Charges are from the MP2/6311++G(2d,2p) calculatic)n. [Ref. 9, p. 193]
Similarly, studies of n-propyl cyanide’s gas electron diffraction pattern
show the gauche conformer to be more abundant than the anti, with a
conformational ratio 75.1 ±6.0% gauche to 24.9% anti at 295 K. This result also
agrees with an MP2/6-31G(d) calculation, which predicts a conformational
mixture of 77% gauche to 23% anti. Bohn et al. studied the rotational spectrum
of n-butyl cyanide and assigned three conformers, AA, GA, and AG.
In summary, when an electron withdrawing halogen terminates a propyl
fragment, then the preferred heavy atom configuration is gauche, not anti.
Butane may be thought of, for purposes of comparison, as propyl methane. The
terminal methane, an electron-releasing group, is more stable in the anti
configuration. Whether an acetylene or a cyano group in propyl acetylene and
propyl cyanide, respectively, prefers the anti or gauche configuration remains
40
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
unclear. What is clear, however, is that the relative energy difference between
the two forms is quite small. Variable temperature studies on n-propyl acetylene
in liquid xenon and in liquid krypton, separately, show the anti form to be more
stable. An electron diffraction study shows the gauche form to be more
abundant. A microwave study also shows the gauche form to be slightly more
stable than the anti form and attributes the greater stability to an attractive
dipole-dipole interaction. The comparisons of the results of the IR and GED
experiments are not exact. The information that comes out of the IR
experiments is a difference in enthalpy between the different conformers over a
specific temperature range. The information extracted from the electron
diffraction data is conformational composition at a particular temperature. The
ratio of conformational abundance may then be used to give a AG value through
-A G °
.
..
2 (%anti)
the following equation: — -------- = exp
RT
%gauche
, where R is the ideal gas
constant, T is temperature in Kelvin, and multiplication by two is to account for
the double degeneracy of the gauche form. As the gauche form is doubly
degenerate, the authors of the IR experiment point out that being more abundant
does not mean greater stability.
There is similar inconsistency between the results of Hartree-Fock and
Moller-Piesset caiculations. When electron correlation is included in the
calculation and the basis set is appropriate to the system for study, the universal
result is that the gauche form is more stable than the anti. This result lends
credence to the argument that there is an attractive interaction between the
methyl group and the triple bond that is operative. These studies, while showing
41
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the different interpretations of experimental results, also serve to show that the
two forms are at the very least almost equally stable and that MP2 calculations
with electron correlation built-in are needed to describe accurately the operative
effect that stabilizes the gauche form.
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Literature Cited
1. Bell, Stephen; Guirgis, Gamil A.; Yin, Li; Durig, James R. Far Infrared
Spectrum, ab initio Calcuiations, and Conformational Analysis of 1 Pentyne. J. Phys. Chem. A, 1997, 101 (34), 5987-5996.
2. Wodarczyk, Francis J.; Wilson, E. Bright. Rotational Isomerism in 1Pentyne from the Microwave Spectrum. J. Chem. Phys. 1972, 56(1), 166176.
3. Bohn, Robert K.; Pardus, Jennifer L.; August, Julie; Brupbacher, Thomas;
Jager, Wolfgang. Microwave studies of the three conformers of butyl
cyanide. J. Mol. Struct. 1997, 413-414, 293-300.
4. Herrebout, W.A.; van der Veken, B.J.; Wang, Aiying; Durig, J.R. Enthalpy
Differences between Conformers of n-Butane and the Potential Function
Governing Conformational Interchange. J. Phys. Chem. A 1995, 99 (2),
578-585.
5. Moss, G.P. Basic Terminology of Stereochemistry. Pure Appl. Chem.
1996, 68(12), 2193-2222.
6
. Hi rota, Eizi. Rotational Isomerism and Microwave Spectroscopy. I. The
Microwave Spectrum of Normal Propyl Fluoride. J. Chem. Phys. 1962, 37
(2), 283-291.
7. Guirgis, G.A.; Zhu, X.; Durig, J.R. Conformational and Structural Studies
of 1-Fluoropropane from Temperature Dependant FT-IR Spectra of Rare
Gas Solutions and Ab Initio Calculations. Struct. Chem. 1999, 10 (6 ), 445461.
43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8. Durig, J.R.; Xhu, X.; Shen, S. Conformational and structural studies of 1chloropropane and 1-bromopropane from temperature-dependant FT-IR
spectra of rare gas solutions and ab initio calculations. J. Mol. Struct.
2001, 570, 1-23.
9. Tfgetteberg, M.; Bakken, P.; Hopf, H. Structure and conformations of
gaseous butyronitrile: C— H --tc interaction? J. Mol. Struct. 2000, 556,
189-196.
1 0
. Wlodarczak, G.; Martinache, L.; Demaison, J.; Marstokk, K.-M.; Mollendal,
Harold. Rotational Spectrum of Butyronitrile: Dipole Moment, Centrifugal
Distortion Constants and Energy Difference between Conformers. J. Mol.
Spectrosc. 1988, 127, 178-185.
11
. Bradford, W.F.; Fitzwater, Susan; Bartell, L.S. Molecular Structure of nButane: Calculation of Vibrational Shrinkages and an Electron Diffraction
Re-investigation. J. Mol. Struct. 1977, 38, 185-194.
12.Traetteberg, M.; Bakken, P.; Hopf, H. Unexpected conformational behavior
of gaseous 1-pentyne. J. Mol. Struct. 1999, 509, 213-220.
13.Crowder, G.A.; Pick, Harold. Vibrational Analysis of 1-Butyne and 1Pentyne. J. Mol. Struct. 1986, 147, 17-27.
14. Durig, J.R.; Drew, B.R. Conformational stability of 1-pentyne from
temperature dependent FT-IR spectra of liquid rare gas solutions and ab
initio calculations. J. Mol. Struct. 2001, 560, 247-259.
15.0berhammer, Heinz. Molecular Structures and Conformations:
Experiment and Theory. J. Comp. Chem. 1998, 19 (2), 123-128.
44
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
B. Experimental Results and Quantum Structure Calculations
1. /7-butyl acetylene
a. Experiment
/7-butyi acetylene was purchased from Aldrich Chemical Company,
Milwaukee, Wl, with a stated purity of 99%. GC/mass spectral analysis did not
reveal any significant impurities and a sample was evaporated into a 4 L
8 8
bulb
and then filled to 4 atm with the carrier gas, Ne (~76%Ne, ~24%He mixture); nbutyl acetylene composed about 1% of the vapor entering the spectrometer. The
bulb was used as a reservoir to provide about
1
atm backing pressure to the
pulsed jet Fourier transform spectrometer of the Southern New England
Microwave Consortium [1]. The carrier gas mixture, at pressures of 1 to 2 atm,
pulsed into the resonant cavity at 5 Hz in ca. 0.3 ms spurts. The nozzle was
mounted in one of the resonant cavity mirrors and rotational transitions were
observed in the region between 5 and 18 GHz. The line widths in the power
spectrum ranged from 10 to 20 kHz. Transitions were observed as Doppler
doublets with splittings ranging from 27 kHz at ca. 5 GHz to 112 kHz at ca. 19
GHz. Uncertainties in the reported frequencies are estimated to be less than 2
kHz.
Most of the transitions of the AA and GA conformers could be seen in a
single pulse, while those of the GG-trans and AG species were much less
intense and required integration of a few hundred pulses.
45
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
b.
Results and Discussion
Spectral assignment
Structural models for each of the four expected conformers were
developed using ab initio calculations |2]. The conformations are labeled as antianti, AA; anti-gauche, AG {gauche at the methyl end); gauche-anti, GA {gauche
at the acetylene end); and two gauche-gauche forms: both methyl and
acetylene gauche on opposite sides of the propyl moiety, GG-trans; and both
methyl and acetylene gauche on the same side of the propyl moiety, GG-cis.
See Figure 1. The coordinates for each model were put into a program that
diagonalizes the interna! coordinate z-matrix into the principal axis system from
which rotational constants were calculated. These rotational constants were
used to predict the rotational spectrum for each conformer, respectively.
Rotational transitions were found near their predicted frequencies and
assignments to quantum labels for observed transitions were made. Transitions
assigned to the AA and GA conformers were intense enough to observe and
assign the singly substituted
isotopomer species, which occur at 1 % natural
abundance. The assigned transitions of the AA parent isotopomer and its singly
substituted '*^0 isotopomers are presented in Tables 1-7. Similar tables for the
GA form and its singly substituted
isotopomers are shown In Tables 8-14;
Tables 15 and 16 summarize the assigned rotational transitions of the AG and
GG-trans rotamers, respectively. The stericaliy encumbered geometry, GG-cis
form was predicted to be highly energetic (> ca. 7 kJ/mol from MP2/6-311+G(d,p)
calculations) and was not expected to be observed in the spectrum.
46
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 1: Conformations of n-butyi acetylene
AA
w
GG4rans
G G - C IS
47
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 1: Rotational frequencies (MHz) of the
AA parent isotopomer of /7-butyl acetylene
J' Kp'
Ko'
J"
Kp"
Ko"
Observed
Obs-Calc
2
0
2
1
0
1
5162.4581
-0.0013
2
1
2
1
1
1
5090.8392
-0.0014
2
1
1
1
1
0
5234.7224
0.0027
3
0
3
2
0
2
7742.9566
-0 . 0 0 0 2
3
1
3
2
1
2
7636.0734
0.0003
3
1
2
2
1
1
7851.8884
0 .0 0 0 2
3
2
2
2
2
1
7744.3047
0.0024
3
2
1
2
2
0
7745.4634
0.0033
4
0
4
3
0
3
10322.5754
-0.0005
4
1
4
3
1
3
10181.0816
0 .0 0 0 2
4
1
3
3
1
2
10468.8253
-0.0007
5
0
5
4
0
4
12901.0238
-0.0007
5
1
5
4
4
12725.7916
-0.0007
5
1
4
4
1
1
3
13085.4542
-0.0006
5
2
4
4
2
3
12906.4853
0.0040
5
2
3
4
2
2
12912.2670
-0.0014
5
3
2
4
3
1
12908.5135
-0.0046
6
0
6
5
0
5
15478.0106
-0 . 0 0 1 1
6
1
6
5
1
5
15270.1339
-0.0005
6
1
5
5
1
4
15701.6946
0 .0 0 0 2
6
2
5
5
2
4
15487.2117
0.0027
7
0
7
6
0
6
18053.2479
-0.0009
7
1
7
6
1
6
17814.0368
-0 . 0 0 1 1
7
1
6
6
1
5
18317.4630
-0 . 0 0 0 1
7
2
5
6
2
4
18083.8198
0.0026
1
1
0
1
0
1
13445.3916
0 .0 0 0 0
2
1
2
1
0
1
18464.2900
-0 . 0 0 2 1
2
1
2
1
0
1
18464.2900
-0 . 0 0 2 1
3
1
2
3
0
3
13626.6012
0.0180
48
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0
13772.8212
-0.0122
Table 2 : Rotational frequencies (MHz) of the
AA
isotopomer of #>butyl acetylene
J* Kp'
Ko*
J"
Kp" Ko"
Observed
Obs-Calc
3
0
3
2
0
2
7556.1104
-0.0001
3
1
3
2
1
2
7453.122
0.0014
3
1
2
2
1
1
7661.0165
-0.0016
4
0
4
3
0
3
10073.5405
0.0014
4
1
4
3
1
3
9937.168
-0.0005
4
1
3
3
1
2
10214.357
0.0003
5
0
5
4
0
4
12589.875
-0.0007
5
1
5
4
1
4
12420.9386
-0.0006
6
1
5
5
1
4
15320.0947
0.0004
Table 3 : Rotational frequencies (MHz) of the
AA
isotopomer of n-butyl acetylene
J'
Kp* Ko*
J"
Kp"
Ko"
Observed
Obs-Calc
3
0
3
2
0
2
7661.7480
0 .0 0 1 1
3
1
3
2
1
2
7557.0438
-0.0009
3
1
2
2
1
1
7768.4150
-0.0006
4
0
4
3
0
3
10214.3531
-0 . 0 0 0 1
4
1
4
3
1
3
10075.7244
-0 . 0 0 0 2
4
1
3
3
1
2
10357.5436
-0.0003
5
0
5
4
0
4
12765.8377
-0 . 0 0 0 2
5
1
5
4
1
4
12594.1201
0.0006
5
1
4
4
1
3
12946.3760
0 .0 0 0 2
6
1
6
5
1
5
15112.1614
0.0006
49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
5
5
4
15534.8349
0.0002
7
7
6
6
17529.7808
-0.0006
Table 4: Rotational frequencies (MHz) of the
AA
isotopomer of n-butyi acetylene
J'
Kp- Ko'
J"
Kp" Ko"
Observed
Obs-Calc
3
0
3
2
0
2
7727.2298
0.0007
3
1
3
2
1
2
7618.1633
-0.0008
3
1
2
2
1
1
7838.4575
-0 . 0 0 1 2
4
0
4
3
0
3
10301.5266
0.0013
4
1
4
3
1
3
10157.1831
0.0003
4
1
3
3
1
2
10450.8998
-0 . 0 0 0 1
5
0
5
4
0
4
12874.5819
-0.0006
5
1
5
4
1
4
12695.8878
0.0003
5
1
4
4
1
3
13063.0143
-0 . 0 0 0 1
6
5
0
5
15446.092
-0.0007
6
6
1
6
5
1
5
15234.203
0.0003
6
1
5
5
1
4
15674.7182
0.0006
7
0
7
6
0
6
18015.7511
0.0004
7
1
7
6
1
6
17772.0543
-0.0005
Table 5: Rotational frequencies (MHz) of the
AA
isotopomer of n-butyl acetylene
J'
Kp'
Ko'
J"
Kp"
Ko"
Observed
Obs-Calc
3
0
3
2
0
2
7739.1761
-0 . 0 0 1 2
3
1
3
2
1
2
7632.1183
-0 . 0 0 0 1
3
1
2
2
1
1
7848.2949
0.0003
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
0
4
3
0
3
10317.5289
-0 .0 0 0 4
4
1
4
3
1
3
10175.8069
0 .0 0 0 4
4
1
3
3
1
2
10464.0323
-0 .0 0 0 3
5
0
5
4
0
4
12894.7048
0 .0 0 01
5
1
5
4
1
4
12719.1958
0 .0 0 0 2
5
1
4
4
1
3
13079.4605
0.0008
6
0
6
5
0
5
15470.4118
0.0008
6
1
6
5
1
5
15262.2142
0.0004
6
2
5
5
2
4
15479.656
-0 . 0 0 1 2
6
2
4
5
2
3
15489.8334
0 .0 0 0 2
7
0
7
6
0
6
18044.3579
-0.0003
7
1
7
6
1
6
17804.7908
-0 . 0 0 0 2
Table 6: Rotational frequencies (MHz) of the
AA
isotopomer of n-butyl acetylene
J'
Kp-
Ko'
J"
Kp"
Ko"
Observed
Obs-Calc
3
0
3
2
0
2
7679.6562
0.0008
3
1
3
2
1
2
7573.5008
-0.0013
3
1
2
2
1
1
7787.8400
-0 . 0 0 1 2
4
0
4
3
0
3
10238.1865
0.0015
4
1
4
3
1
3
10097.6558
-0.0003
4
1
3
3
1
2
10383.4318
-0 . 0 0 1 1
5
0
5
4
0
4
12795.5536
0 .0 0 0 0
5
1
5
4
1
4
12621.5153
0 .0 0 0 1
5
1
4
4
1
3
12978.7183
0.0004
6
0
6
5
0
5
15351.4739
0.0013
6
1
6
5
1
5
15145.0084
0 .0 0 0 0
6
1
5
5
1
4
15573.6170
0 .0 0 0 3
7
0
7
6
0
6
17905.6546
-0 .0 0 1 2
7
1
7
6
1
6
17668.0665
0 .0 0 0 1
51
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 7: Rotational frequencies (MHz) of the
AA
isotopomer of if-butyl acetylene
J'
Kp'
Ko'
J"
Kp"
Ko"
Observed
Obs-Caic
3
0
3
2
0
2
7573.3075
0 .0 0 0 2
3
1
3
2
1
2
7469.8368
-0.0015
3
1
2
2
1
1
7678.7110
0.0004
4
0
4
3
0
3
10096.4558
0.0003
4
1
3
3
1
2
10237.9422
-0.0003
5
0
5
4
0
4
12618.5004
-0.0003
5
1
5
4
1
4
12448.7911
0.0009
52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 8: Rotational frequencies (MHz) of the
GA parent isotopomer of />butyl acetylene
Observed
Obs-Calc
2
13658.5116
-0.0004
2
2
13195.6330
0 .0 0 1 0
3
2
1
13267.4952
-0.0032
5
4
0
4
16358.5122
-0 . 0 0 1 0
1
5
4
1
4
15880.5190
0.0006
5
1
4
4
1
3
17058.4588
0.0006
1
1
1
0
0
0
8969.2220
-0.0047
1
1
0
1
0
1
5905.1543
0 .0 0 1 0
2
1
1
2
0
2
6148.6245
-0.0017
2
1
2
1
0
1
12033.3249
-0 . 0 0 0 1
3
1
2
3
0
3
6527.2585
0 .0 0 1 0
3
1
3
2
0
2
14982.0953
0.0023
4
1
3
4
0
4
7056.6954
0 .0 0 2 0
4
0
4
3
1
3
8018.8914
-0.0005
5
1
4
5
0
5
7756.6382
-0 . 0 0 0 2
J‘ Kp’ Ko’
J"
4
1
3
3
1
4
2
3
3
4
2
2
5
0
5
Kp" Ko"
53
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 9: Rotational frequencies (MHz) of the
GA
isotopomer of o-butyl acetylen e
J'
Kp- Ko'
J"
Kp” Ko"
Observed
Obs-Calc
3
0
3
2
0
2
9662.4666
0.0016
3
1
2
2
1
1
10032.9285
-0.0018
2
0
2
1
0
1
6453.335
-0.0007
4
0
4
3
0
3
12850.8035
0.0009
4
1
3
3
1
2
13368.6286
-0.0008
4
1
4
3
1
3
12444.5514
-0.0001
4
2
3
3
2
2
12915.4025
-0.0003
4
2
2
3
2
1
12985.1415
0.0012
5
0
5
4
0
4
16012.2102
0.0000
5
1
5
4
1
4
15543.3184
-0.0010
5
1
4
4
1
3
16696.5456
0.0006
Table 10: Rotational frequencies (MHz) of the
GA
isotopomer of n-butyl acetylene
J'
Kp'
Ko'
J"
Kp"
Ko"
Observed
Obs-Calc
3
0
3
2
0
2
9773.9367
-0.0004
3
1
3
2
1
2
9450.0890
0.0014
3
1
2
2
1
1
10144.5722
0.0007
4
0
4
3
0
3
12999.8270
0.0011
4
1
4
3
1
3
12592.2712
0.0007
4
2
3
3
2
2
13063.6339
-0.0042
4
2
2
3
2
1
13132.5199
0.0001
5
0
5
4
0
4
16199.0865
-0.0008
5
1
5
4
1
4
15728.1096
0.0003
5
1
4
4
1
3
16882.9402
0.0013
54
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 11: Rotational frequencies (MHz) of the
GA
isotopomer of #i-butyl acetylene
J’ Kp'
Ko'
J"
Kp" Ko"
Observed
Obs-Calc
3
0
3
2
0
2
9829.8952
0.0006
3
1
3
2
1
2
9495.3393
0.0002
3
1
2
2
1
1
10216.0317
-0.0001
4
0
4
3
0
3
13071.1800
0.0000
4
1
4
3
1
3
12651.8224
-0.0007
4
1
3
3
1
2
13611.9886
-0.0003
5
0
5
4
0
4
16283.2567
-0.0001
5
1
5
4
1
4
15801.3572
0.0002
5
2
3
4
2
2
16567.6388
0.0000
5
2
4
4
2
3
16417.1585
0.0001
5
1
4
4
1
3
16999.4397
0.0001
Table 12: Rotational frequencies (MHz) of the
GA
isotopomer of n-butyl acetylene
J'
Kp'
Ko'
J"
Kp"
Ko"
Observed
Obs-Calc
3
0
3
2
0
2
9850.4398
-0.0001
3
1
3
2
1
2
9518.701
0.0012
4
0
4
3
0
3
13099.5052
0.0004
4
1
4
3
1
3
12683.1926
-0.0011
4
1
3
3
1
2
13634.0504
-0.0005
4
2
3
3
2
2
13167.9495
-0.0005
4
2
2
3
2
1
13241.8637
0.0002
5
0
5
4
0
4
16320.0965
0.0007
5
1
5
4
1
4
15840.9088
-0.0005
5
1
4
4
1
3
17027.4436
-0.0003
5
2
3
4
2
2
16597.0196
0.0006
55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 13: Rotational frequencies (MHz) of the
GA
isotopomer of />butyl acetylene
J'
Kp'
Ko'
J"
Kp"
Ko"
Observed
Obs-Calc
3
0
3
2
0
2
9816.947
0.0009
3
1
3
2
1
2
9491.1034
-0.0005
3
1
2
2
1
1
10190.5205
-0.0013
4
0
4
3
0
3
13056.5277
0.0017
4
1
4
3
1
3
12646.8026
-0.0012
4
2
3
3
2
2
13121.6184
-0.0008
4
2
2
3
2
1
13191.8996
0.0016
5
0
5
4
0
4
16268.966
0.0000
5
1
4
4
1
3
16959.0072
-0.0005
Table 14: Rotational frequencies (MHz) of the
GA ’ ^Cs isotopomer of /i-butyl acetylene
J'
Kp'
Ko'
J"
Kp"
Ko"
Observed
Gbs-Caic
3
0
3
2
0
2
9644.9715
-0.0005
3
1
3
2
1
2
9327.1088
0.0001
3
1
2
2
1
1
10007.9100
0.0023
4
0
4
3
0
3
12829.0388
0.0004
4
1
4
3
1
3
12428.5799
0.0001
5
1
5
4
1
4
15523.9290
0.0013
4
2
3
3
2
2
12890.5255
-0.0031
5
0
5
4
0
4
15987.4166
-0.0003
56
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 15: Rotational frequencies (MHz) of the
AG parent isotopomer of n-butyl acetylene
J‘ Kp’ Ko‘
J"
3
0
3
2
0
3
1
3
2
3
1
2
4
0
4
Observed
Obs-Calc
2
8615.9335
0.0007
1
2
8516.2468
-0.0025
2
1
1
8717.8420
0.0020
4
3
0
3
11486.3722
0.0006
1
4
3
1
3
11354.5980
-0.0004
4
1
3
3
1
2
11623.3848
0.0034
4
2
3
3
2
2
11489.4545
0.0027
4
2
2
3
2
1
11492.6927
-0.0049
5
0
5
4
0
4
14355.4928
0.0001
5
1
5
4
1
4
14192.6088
0.0018
5
1
4
4
1
3
14528.5702
-0.0023
5
2
4
4
2
3
14361.3711
0.0010
5
2
3
4
2
2
14367.8572
-0.0031
6
0
6
5
0
5
17222.9700
0.0010
6
2
5
5
2
4
17232.9930
0.0010
6
2
4
5
2
3
17244.3420
-0.0033
6
3
4
5
3
3
17236.3838
0.0100
6
3
3
5
3
2
17236.4210
-0.0077
3
1
3
3
0
3
10232.0985
-0.0008
4
1
4
4
0
4
10100.3269
0.0008
Kp" Ko"
57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 16: Rotational frequencies (MHz) of the
GG-trans parent isotopomer of n-butyl
acetylene
J'
Kp* Ko*
J"
Kp"
Ko"
Observed
Obs-Calc
2
0
2
1
0
1
7985.4830
0.0004
2
1
2
1
1
1
7751.8045
0.0000
2
1
1
1
1
0
8242.8278
0.0000
3
0
3
2
0
2
11948.8603
-0.0007
3
1
3
2
1
2
11620.3851
0.0003
3
1
2
2
1
1
12356.4438
-0.0002
3
2
1
2
2
0
12042.9455
0.0003
4
0
4
3
0
3
15878.0556
-0.0012 ,
4
1
4
3
1
3
15480.6708
0.0010
4
1
3
3
1
2
16460.2176
0.0010
4
2
2
3
2
1
16101.0595
-0.0006
4
2
3
3
2
2
15985.4051
0.0000
1
1
1
0
0
0
7725.6918
0.0003
2
1
2
1
0
1
11478.8436
-0.0002
58
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Ali frequencies were assigned using a Hamiltonian with Watson’s
symmetric reduction In the prolate f representation [3]. Spectroscopic constants
and other features of the spectral fitting are reported in Table 17 for the AA, GA,
AG, and GG-trans conformations.
Conformational Assignment
By examining B -i- C values for each conformer an estimate of where atype, R branch transitions will occur in the microwave spectrum was made.
Table 17 lists the band spacing for each conformer. Rotational transitions for the
species belonging to the 2580, 2871, 3300, and 3998 MHz band spacings were
found and assigned. Transitions for each J occur near these B+C values. For
example, the J = 3 to 2, K = 0 transition of the AA conformer occurs at 3*2580 =
7740 MHz. The 303 to 202 transitions listed in Table I shows this transition was
measured at 7742.9566 MHz. R-branch a-type transitions of a particular J, K ^
0, are split due to disparate values of B and C but occur in a regular pattem
around the central K = 0 transition frequency. By searching for these a-type, R
branch pile-ups and comparing the frequency of occurrence for each conformer
in the spectrum, the conformers were distinguished and assigned.
59
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CD
■—Di
O
o.
c
o
CD
Q.
■CDD
Table 17: Spectroscopic Constants of the Parent Conformers
of n-butyl a c e ty le n e __
C/)
W
o'
o
o
—ty
AA
GA
AG
GG-trans
A/M Hz
14700.0943(25)
7437.1722(7)
11872.33(5)
5848.952(12)
B
1326.6585(6)
1768.25012(3)
1469.8077(5)
2122.1060(3)
"n
0
1254.7182(5)
1532.03742(20)
1402.61047(5)
1876.5610(3)
CD
Paa/uA^
374.67
273.87
330.79
210.53
Pbb
28.11
56.00
29.52
58.78
P cc
6.27
11.94
13.06
27.62
D j/M H z
0.0001439(36)
0.000803(4)
0.0002161(6)
0.00370101(6)
D jk
-0.00782(4)
-0.00969(4)
-0.00272(4)
-0.025158(4)
CD
O
O
■D
cq'
o
o
CD
c
—i
CD
■—Di
O
o.
c
a
o
o
■—Di
o
CD
Q.
-0.138(13)
Dk
■CDD
(/)
(/)
o>
o
di
-0.0000238(38)
-0.000225(3)
a
0.0049
0.0018
-0.001043(5)
0.0041
0.0008
A second test was used to confirm that the AA conformer was correctly
assigned. The second moment, Pec, is the sum over all the atoms of the mass
times the distance from the ab plane squared. This value is zero for a rigid,
planar moiecule; A pair of methylene hydrogens contributes 1.6 uA^ to this
value; Four pairs contribute 4*1.6 uA^. The heavy atom planar configuration of
n-butyl acetylene should have a Pec of -6.4 uA^, consistent with four pairs of
hydrogen atoms out of the ab plane. The observed value is 6.269 uA^, which is
in excellent agreement. The Paa and Pbb values easily discriminate the AG form
from the GA and GG-trans forms; likewise, Pec discriminates the GA rotamer
from the GG-trans form.
Singly Substituted
Isotopomers
Figure 2 shows the labeling scheme for the substituted carbon atoms. For
example, the AA-3 isotopomer refers to the anti-anti conformer with
at atom
position labeled “3”; AA-4 refers to ^^0 at atom position labeled “4”, etc....
61
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 2;
Isotopomer Labeling Scheme"
*a label of AA-3 refers to the AA conformer
singly substituted with
at the atom “3” site
is 1.107% of naturally occurring carbon; consequently, transitions measured
from
species will be -1% as intense as the parent transition. The AA, GA,
and GG-trans conformers had sufficiently intense spectra that singly substituted
isotopes could be assigned and measured.
Tables 18 and 19 are a
summary of the measured spectroscopic constants of the parent and singly
substituted
isotopomers for the AA and GA conformers, respectively. We did
not feel it was necessary to look for the GG-trans ^^0 isotopomer transitions.
For each transition,
isotopes of a particular conformer show a pattern.
All transitions attributed to the ^^0 isotopomer occur at frequencies red-shifted
from the parent. Since the rotational constants are inversely proportional to the
moments of inertia around each respective axis, the increased mass from the
isotope leads to a smaller rotational constant and smaller B + C spacings.
Furthermore, the farther the substituted atom is from the center of mass, the
62
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
more of an effect this mass change has on the isotopomer’s frequency. Figure 3
shows this frequency pattern for the 303-202 transition of the AA parent
conformer and its singly-substituted
isotopomers.
Figure 3; 303-202 Transition Frequencies of the Parent and
13,
C Isotopomers of n-butyl acetylene
10
r.
6
AA-3
AA 8
7556
7573
AA-4
7662
AA-7
7680
AA-5
AA-6
7727
F re q u e n c y (M H z )
substituted at carbon atom position “3” is furthest from the a-axis and its 303202 transition frequency is the most displaced from the transition frequency of
the parent molecule. A substitution nearest to the center of mass, i.e., at position
6, is closest to the parent frequency for this transition. This pattern is the same
for all the transitions for a particular conformer. Measuring and assigning the
63
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
isotopic spectra is necessary if a Kraitchman analysis is to be done and a
substitution structure defined.
64
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CD
■—Di
O
o.
c
o
CD
Q.
■CDD
C/)
W
o'
o
Table 18: Spectroscopic Constants for the AA Parent and
isotopomers of i>butyl acetylene
o
o
■oD
cq
Parent
13#"^
O5
i3r^
'
A/M Hz
14700.0943(25)
14596(4)
14704.7(21)
14476.0(8)
B
1326.6585(6)
1294.18473(21)
1312.37466(6)
1324.79529(7)
C
1254.7182(5)
1224.88363(16)
1241.91574(8)
1251.361752(7)
Paa/UA^
374.67
384.24
378.83
373.72
Pbb
28.11
28.36
28.11
28.64
Pcc
6.27
6.26
6.26
6.26
D j/M H z
0.0001439(36)
0.0001325(29)
0.0001406(9)
0.0001434(10)
D jk
-0.00782(4)
di
-0.0000238(38)
a
0.0049
0.001248
0.000667
0.000779
O’
CD
—i
CD
■—Di
O
o.
c
a
o
Q
■—Di
O
CD
Q.
■CDD
(/)
o'
3
ai
CD
■—Di
O
o.
c
o
CD
Q.
■CDD
C/)
W
o'
o
L/6
o
—ty
CD
O
O
■D
cq
IS/-'
L/7
iSr*
A/MHz
14675.9(5)
14615.7(12)
14602(3)
B
1326.08981(7)
1315.86071(10)
1297.21518(15)
C
1254.029109(8)
1244.41236(12)
1227.58913(14)
Paa/uA^
374.84
377.80
383.33
Pbb
28.17
28.31
28.35
P cc
6.27
6.26
6.26
Dj/MHz
0.0001464(8)
0.0001438(13)
0.000144
'
O’
o
Q
CD
"n
c
O
’
CD
CD
■D
O
Q.
C
a
o
Q
■—Di
O
D jk
-0.007900
di
-0.000024
o
CD
Q.
■CDD
(/)
(/)
O
o>
0.000669
0.001036
0.000945
CD
■—Di
O
o.
c
o
CD
Q.
■CDD
C/)
W
o'
o
Table 19: Spectroscopic Constants for the GA Parent and
C isotopomers of n-butyl acetylene
o
o
o
Parent
13p
L/3
13^
O4
13/-^
L/5
A/MHz
7437.1722(7)
7330.901(5)
7436.68(9)
7308.69(14)
B
1768.25012(3)
1730.71770(19)
1749.3917(3)
1763.56432(13)
C
1532.03742(20)
1499.46995(22)
1517.8376(3)
1523.26633(14)
Paa/uA^
273.86
280.12
276.94
274.63
Pbb
56.00
56.92
56.02
57.17
P cc
11.94
12.01
11.94
11.98
Dj/MHz
0.000803(4)
0.000786(4)
0.000784(7)
0.000767(4)
Djk
-0.00969(4)
■D
O’'
cq
o
O
’
CD
—i
CD
■—Di
O
o.
c
a
o
Q
■—Di
O
CD
Q.
§
O’
o
■CDD
cn
cn
-0.00911 (4)
0.16(4)
Dk
di
-0.000225(3)
a
0.0018
0.0012
0.0020
0.0008
CD
■—Di
O
o.
c
o
CD
Q.
■CDD
C/)
W
o'
o
i3 r'
o
—ty
CD
O
O
■D
cq
A/MHz
7355.26(4)
7408.90(13)
7401.36(15)
B
1765.68437(18)
1757.4769(4)
1725.4463(4)
C
1527.72336(15)
1524.27675(29)
1498.4562(6)
Paa/uA^
274.16
275.45
280.94
Pbb
56.65
56.10
56.33
P cc
12.06
12.1
11.95
Dj/MHz
0.0007936(29)
0.000784(8)
0.0007843(10)
D jk
-0.00951(3)
-0.009401 (7)
a
0.0009
0.0016
'
O’
o
Q
CD
"n
c
O
’
CD
CD
■D
O
Q.
C
a
o
Q
■—Di
O
CD
Q.
■CDD
(/)
(/)
u>
05
0.0021
Molecular Structure and Kraitchman’s Equations
Transitions between rotational states occur in the microwave region.
Rotational constants can be determined from a molecule’s rotational spectrum
and are proportional to the reciprocals of the molecule’s moments of inertia
averaged over the occupied vibrational states. The moments of inertia calculated
from these rotational constants are called the effective moments of inertia. Once
the spectrum is assigned and the rotational constants are determined, the
molecule’s geometry should be established.
Kraitchman’s method for solving exact solutions for molecular coordinates
involves measuring rotational spectra and determining rotational constants for
every possible mono-isotopically substituted species of the parent molecule. The
method’s implicit assumption is that the structural parameters are the same for all
isotopic species of the molecule [4]. As this assumption is only valid for
equilibrium structural parameters, it is necessary to solve a set of self-consistent
equations based on equilibrium moments of inertia. Consequently, the
differences between the calculated and equilibrium parameters are rooted in the
zero-point vibrational effects which are not completely addressed within the BornOppenheimer approximation.
Solving Kraitchman’s equations for Asymmetric Top Molecules
Solving for the structure of a nonplanar asymmetric top molecule is
simplified by deriving equations in terms of planar moments, or second moments,
rather than using moments of inertia. Whereas the moment of inertia describes
69
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
an atom’s position from an axis, the planar moment relates the atom’s position to
a molecular plane. The planar dyadic, P, with respect to the center of mass, is
defined as [4, p. 19]
(Sfn,r,|E/n,r,.)
( 1)
Em,
where my is the mass of the i^^ atom, and r,- is the distance of the i* atom from the
center of mass. The moment equations are
= E m,x,y,. = 0
P ^ = E m,x,z, = 0
F ^ = E m,z,y,. = 0
(2 )
= E m ,xf
Fy = E m,yf
F ^ = E m,zf
where X, Y, and Z signify the principal axes. Likewise, for the substituted
molecule, the elements of P are
P '> a = P x +
F '^ = F ^ + /^ ;
P'yy = Py +
(3)
P'=/ixy
P' ^ = juxz
P'=Mzy,
where the prime indicates an isotopically substituted molecule. The planar
moments for the substituted molecule are the eigenvalues, P’, obtained by
solving the following secular equation [4, pg. 22]:
+ nx^ - F'
m
jjLZX
/ixy
Py+MV^-P'
^ zy
iixz
m
0
(4)
+ /iz ^ - F
70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The roots of the foliowing cubic equation in P’ give the planar moments in the
principal axis frame, namely P’x, P’y,
P ’z -
P '^ - { P \ + P 'y + P 'z ) P '^ + { P \ P 'y+ P 'x
P 'z + P 'y P 'z )
P '=
P ‘x
P ’y P ' z
•
(5 )
By equating the coefficients of the expanded secular equation to those defined in
the cubic polynomial, given in Equation (5), x^, y^, and
can ultimately be
determined. Only the absolute value of the x, y, and z coordinates are obtained
from this analysis, but one can usually deduce the sign of the axis the atom
coordinate from a model.
71
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 4: Substitution Structure of
/>butyl acetylene
1.193 A
1.173 A
1.477 A
1.503 A
1.522 A
1.506 A
1.538 A
1.523 A
1.524 A
k
A A conform er
GA conform er
Relative energies of the conformers: Experiment and Calculations
Dipole moments of the various species were not determined
experimentally; rather, they were calculated from ab initio quantum structure
calculations of the equilibrium geometries. The results of our geometry
optimization calculations are given in Tables 28-37. Large a- and b- dipole
moments are predicted from an MP2/6-311+G(d,p) calculation for the AA
(Pa = -0.75 D, !ib = 0.62 D), GA (pa = 0.62 D, pb = 0.46 D), and GG-trans
(pa = -0.66 0, pb = 0.42 D) conformers. Transitions of the a- and b-dipole
72
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
moments vt/ere assigned for all three conformers. Predicted dipole moment
components for the AG conformer are |ia = 1.0 D, fit, = 0.09 D, and jic = 0.10 D
using the same level of theory. Transitions of the a- and c-type were identified
for this form but they were much less intense than any of the a- or b-types of the
other three conformers. This result suggests that the gas contained considerable
populations of the AA, GA, and GG-trans conformers but much less of the AG
form.
Our HF calculations predict the AA form lies lowest in energy. At the
correlated MP2 level of theory, the GA conformer is predicted to be the lowest
energy structure, independent of the choice of basis set. These observations will
be discussed in a later chapter of this thesis. We conclude from our
experimental and theoretical observations that the AA, GA, and GG-trans
conformations are the more stable conformers and are close in energy, and that
the AG form lies higher in energy.
2. 5-Hexynenitriie
a. Experiment
5-Hexynenitrile was purchased from Aldrich Chemical Company. A proton
nuclear magnetic resonance (NMR) spectrum revealed only trace levels of
impurities and the sample was studied without further purification. The
experiment was done on the pulsed-jet Fourier Transform Spectrometer of the
Southern New England Microwave Consortium located at Wesleyan University in
73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Middletown, CT [1]. The sample was introduced into the cavity by placing a few
drops in a small Cu U-tube, located immediately behind the nozzle, and bubbling
the carrier gas (-75% Ne, -25% He) at 1 atm pressure over the liquid, carrying
the liquid’s vapor into the resonant cavity at -5 Hz in ca. 0.3 ms bursts. 5Hexynenitrile composed 1% of the vapor entering the spectrometer. The nozzle
was mounted in one of the cavity mirrors and rotational transitions were observed
in the region between 6 and 18 GHz. Transitions were observed as Doppler
doublets, consistent with the gas pulses moving at several hundred meters per
second in the microwave cavity. Doppler splittings varied from 38 kHz at 7 GHz
to 76 kHz at 17 GHz. Line widths in the power spectrum ranged from 10-20 kHz.
The reported frequencies were measured from the power spectrum of the free
induction decay and their precision is mainly limited by the line width. The
uncertainties are estimated to be less than 2 kHz for transitions for lower J
transitions with well-resolved
quadrupole splittings and about 5 kHz for the
higher J transitions with partially overlapped quadrupole components.
b. R esults and D iscussio n
Spectral Assignment
Structural models, using parameters from geometrically similar systems,
were constructed for the four expected conformers: anti-anti, AA; gauche-anti
{gauche at the acetylene end), GA; anti-gauche {gauche at the nitrile end), AG;
and gauche-gauche trans (acetylene and nitrile gauche opposite the propyl
moiety), GG-trans. The gauche-gauche cis, GG-cis, conformer was expected to
74
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
be too energetic to be populated in the jet. Individual rotationai constants were
predicted for each species and a rigid rotor spectrum calculated. Only three of
the four expected conformers were observed and assigned: AA, GA, and AG.
The quantum labels for the quadrupole components were assigned and the
splittings were used to determine the quadrupole coupling constants and the
hypothetical unsplit frequencies for each transition. Tables 20-22 show
quadrupole hyperfine components and the derived coupling constants for each
observed conformer.
The hypothetical unsplit frequencies from the quadrupole analyses were
determined using a Hamiltonian with Watson’s asymmetric reduction in the
prolate f representation [3]. The assignments and fits are reported in Table 23
for the AA conformer, Table 24 for the GA conformer, and Table 25 for the AG
conformer.
75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 20: 5-Hexynenitrile AA
Quadrupole component fitting
404-303
P>
4
3
5
414-313
F"
4
3
5
505-404
F"
4
5
6
515-414
F"
5
6
606-515
F"
5
6
7
Fitted unsplit frequency: 7443.3712 MHz
F
3
2
4
r
1
1
1
Obs.(MHz)
7443.3720
7443.3248
7443.3867
Obs-Calc
0.0021
-0.0005
-0.0016
Fitted unsplit frequency: 7329.7747 MHz
p
3
2
4
r
1
1
1
Obs.(MHz)
7329.7229
7329.7581
7329.8178
Obs-Caic
0.0031
0.0100
0.0069
-
Fitted unsplit frequency: 9302.4602 MHz
F'
3
4
5
I'
1
1
1
Obs.(MHz)
9302.4394
9302.4531
9302.4735
Obs-Calc
0.0043
-0.0054
0.0012
Fitted unsplit frequency: 9161.7789 MHz
F'
4
5
Obs.(MHz)
9161.7496
9161.8035
Obs-Calc
-0.0016
0.0016
Fitted unsplit frequency: 11160.3824 MHz
F«
4
5
6
Obs.(MHz)
11160.3643
11160.3784
11160.3963
Obs-Calc
-0.0027
0.0020
0.0046
-
76
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
616-515
F"
5
7
Fitted unspiit frequency: 10993.4805 MHz
i"
1
1
707-606
F"
6
7
8
r
1
1
Obs.(MHz)
10993.4657
10993.5028
Obs-Calc
-0.0063
0.0063
Fitted unsplit frequency: 13016.9200 MHz
1"
1
1
1
717-616
F"
6
7
8
F
4
6
F'
5
6
7
r
1
1
1
Obs.(MHz)
13016.9128
13016.9168
13016.9255
Obs-Calc
0.0029
-0.0009
-0.0020
Fitted unsplit frequency: 12824.8422 MHz
I"
1
1
1
F
5
6
7
r
1
1
1
Obs.(MHz)
12824.8337
12824.8337
12824.8537
Obs-Calc
-0.0016
0.0018
-0.0002
Quadrupole Coupling Constants:
Chi AA = -2.18(9)
Chi BB = 0.59(5)
77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 21: 5-Hexynenitrile GA
Quadrupole component fitting
303-202
F"
3
4
Fitted unsplit frequency: 6862.4858 MHz
r
1
1
312-211
F"
3
4
2
I"
1
1
1
I"
1
1
F r
2 1
3 1
2 1
Obs.(MHz)
7055.1218
7055.4467
7056.4363
Obs-Calc
-0.0011
0.0010
0.0001
P r
2 1
3 1
Obs.(MHz)
6682.7400
6683.0575
Obs-Calc
0.0000
0.0000
Fitted unsplit frequency: 9140.7897 MHz
!'■
1
1
1
1
413-312
F"
4
3
5
Obs-Calc
-0.0001
0.0001
Fitted unsplit frequency: 6682.9641 MHz
404-303
F"
5
4
3
3
Obs.(MHz)
6862.4877
6862.5280
Fitted unsplit frequency: 7055.3470 MHz
313-212
F"
3
4
F l‘
2 1
3 1
F
4
3
2
3
r
1
1
1
1
Obs.(MHz)
9140.8165
9140.7925
9140.7111
9142.3205
Obs-Calc
0.0007
0.0002
0.0000
-0.0009
Fitted unsplit frequency: 9404.7448 MHz
I"
1
1
1
F
3
2
4
r
1
1
1
Obs.(MHz)
9404.6550
9404.7463
9404.7997
Obs-Calc
-0.0007
0,0003
0.0003
78
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
423-322
Fitted unsplit frequency: 9159.0829 MHz
F"
I"
F'
r
3
5
4
1
1
1
2
4
3
1
1
1
422-321
F"
4
5
3
I"
1
1
1
I"
1
1
1
I"
1
1
1
Obs.(MHz)
9178.3611
9178.8557
9178.9825
Obs-Calc
-0.0011
0.0003
0.0008
F'
3
4
5
r
1
1
1
Obs.(MHz)
11411.2439
11411.2914
11411.3064
Obs-Calc
0.0000
-0.0005
0.0004
P
4
3
5
r
1
1
1
Obs.(MHz)
11131.7761
11131.8162
11131.8524
Obs-Calc
-0.0002
0.0000
0.0002
Fitted unsplit frequency: 11752.0380 MHz
I"
1
1
1
524-423
F'
5
6
r
1
1
1
Fitted unsplit frequency: 11131.8204 MHz
514-413
F"
5
4
6
F
3
4
2
Fitted unsplit frequency: 11411.2885 MHz
515-414
F"
5
4
6
Obs-Calc
-0.0003
0.0002
0.0001
Fitted unsplit frequency: 9178.7238 MHz
505-404
F"
4
5
6
Obs.(MHz)
9159.3390
9159.2136
9158.7241
F
4
3
5
r
1
1
1
Obs.(MHz)
11751.9940
11752.0280
11752.0725
Obs-Calc
0.0000
-0.0004
0.0004
Fitted unsplit frequency: 11446.3390 MHz
I"
1
1
F
4
5
r
1
1
Obs.(MHz)
11446.1595
11446.4158
Obs-Calc
-0.0004
0.0004
79
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fitted unsplit frequency: 13672.1900 MHz
606-505
F"
5
6
7
!■■
1
1
1
616-515
F"
6
5
7
1"
1
1
1
I"
1
1
1
F
5
4
6
r
1
1
1
F'
5
4
6
1'
1
1
1
I"
P
r
1
1
5
6
1
1
Obs.(MHz)
13352.9788
13352.9965
13353.0253
Obs-Calc
-0.0001
-0.0001
0.0002
Obs.(MHz)
14096.6040
14096.6180
14096.6521
Obs-Calc
-0.0001
-0.0002
0.0003
Obs.(MHz)
13731.8107
13731.9624
Obs-Calc
-0.0003
0.0003
Fitted unsplit frequency: 13800.2045 MHz
I"
1
1
707-606
F"
6
7
8
Obs-Calc
-0.0001
-0.0005
0.0006
Fitted unsplit frequency: 13731.9130 MHz
624-523
F"
6
7
Obs.(MHz)
13672.1604
13672.1934
13672.2027
Fitted unsplit frequency: 14096.6287 MHz
625-524
F"
6
7
r
1
1
1
Fitted unsplit frequency: 13353.0036 MHz
615-514
F"
6
5
7
F’
4
5
6
F
5
6
I'
1
1
Obs.(MHz)
13800.0986
13800.2553
Obs-Calc
0.0000
0.0000
Fitted unsplit frequency: 15921.9062 MHz
1"
1
1
1
P
5
6
7
r
1
1
1
Obs.(MHz)
15921.8852
15921.9098
15921.9152
Obs-Calc
0.0004
-0.0009
0.0005
80
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
717-616
Fitted unsplit frequency: 15571.5307 MHz
F"
I"
F
r
Obs.(MHz)
Obs-Calc
7
1
6
1
1 5 5 7 1 .5 1 7 9
0 .0 0 2 2
6
8
1
1
5
7
1
1
15571.5230
15571.5444
-0.0006
-0.0016
716-615
F"
7
6
8
I"
1
1
1
726-625
F"
7
8
i"
1
1
312-202
F"
2
4
3
I"
1
1
1
606-515
F"
6
7
r
1
1
Fitted unsplit frequency: 16437.8680 MHz
F
!'
6
5
7
1
1
1
Obs.(MHz)
16437.8527
16437.8598
16437.8836
Obs-Calc
-0.0005
0.0014
-0.0009
Fitted unspilt frequency: 16015.4832 MHz
F'
6
7
r
1
1
Obs.(MHz)
16015.4202
16015.5165
Obs-Caic
0.0004
-0.0004
Fitted unspiit frequency: 13117.7144 MHz
F'
1
3
2
!'
1
1
1
Obs.(MHz)
13117.4008
13117.7022
13117.8814
Obs-Caic
-0.0007
0.0007
-0.0001
Fitted unsplit frequency: 8673.6763 MHz
F
5
6
r
1
1
Obs.(MHz)
8673.5414
8673.7418
Obs-Calc
0.0000
0.0000
Quadrupole Coupling Constants:
Chi AA = -3.5920(15)
Chi BB = 2.0222(15)
81
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 22: 5-Hexynenitrile AG
Quadrupole component fitting
312-211
F"
I"
4
1
3
1
2
1
Fitted unsplit frequency; 6977.6495 MHz
F' l‘
Obs.(MHz)
Obs-Calc
3 1
6977.6112
0.0008
2 1
6977.6784
0.0008
1 1
6977.7500
-0.0016
404-303
Fitted unsplit frequency: 9064.9479 MHz
F"
5
3
F*
4
2
413-312
423-322
505-404
515-414
1"
1
1
Obs.(MHz)
9301.6694
9301.6972
9301.7391
Obs-Caic
0.0018
-0.0005
-0.0013
Fitted unsplit frequency: 9079.1194 MHz
Obs.(MHz)
9079.1019
9079.1674
Obs-Calc
0.0002
0.0002
-
Fitted unsplit frequency: 11319.8480 MHz
F
4
5
3
F"
5
6
4
F"
4
5
!'
1
1
1
F
4
3
F"
5
4
Obs-Calc
0.0000
0.0000
Fitted unsplit frequency: 9301.6901 MHz
F'
4
3
2
F"
5
4
3
Obs.(!VIHz)
9064.9522
9064.9709
Obs.(MHz)
11319.8236
11319.8555
11319.8683
Obs-Calc
0.0007
-0.0005
0.0002
-
Fitted unsplit frequency: 11062.9055 MHz
F‘
3
4
r
1
1
Obs.(MHz)
11062.8836
11062.9052
Obs-Calc
0.0003
-0.0006
82
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
1
514-413
F“
6
5
4
1"
1
1
1
606-505
F"
6
7
5
I"
1
1
1
616-515
F"
5
6
7
I"
1
1
1
615-514
F"
7
6
5
r
1
1
1
625-524
F"
7
6
I"
1
1
624-523
F"
7
6
I"
1
1
5
1
11062.9153
0.0003
Fitted unsplit frequency: 11624.1142 MHz
F
5
4
3
I'
1
1
1
Obs.(MHz)
11624.1005
11624.1147
11624.1450
Obs-Calc
0.0001
0.0006
-0.0007
Fitted unspiit frequency: 113567.2635 MHz
F
5
6
4
!'
1
1
1
Obs.(MHz)
13567.2342
13567.2745
13567.2837
Obs-Calc
0.0005
-0.0001
-0.0004
Fitted unsplit frequency: 13271.4451 MHz
P
4
5
6
r
1
1
1
Obs.(MHz)
13271.4324
13271.4415
13271.4545
Obs-Calc
-0.0009
0.0000
0.0009
Fitted unsplit frequency: 13944.4688 MHz
F'
6
5
4
!'
1
1
1
Obs.(MHz)
13944.4600
13944.4648
13944.4915
Obs-Calc
-0.0004
0.0004
0.0000
Fitted unsplit frequency: 13613.4741 MHz
F
6
5
r
1
1
Obs.(MHz)
13613.4679
13613.4849
Obs-Calc
-0.0013
0.0013
Fitted unsplit frequency: 13666.2214 MHz
F'
6
5
r
1
1
Obs.(MHz)
13666.2045
13666.2598
Obs-Calc
0.0010
-0.0010
83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
707-606
P‘
1"
1
1
1
7
8
6
717-616
F"
7
8
1"
1
1
716-615
F"
7
8
1"
1
1
725-624
F"
6
8
7
I"
1
1
1
111-■000
F"
1
2
1"
1
1
313-202
F"
4
2
1"
1
1
Fitted unsplit frequency: 15805.9138 MHz
F
6
7
5
r
1
1
1
Obs.(MHz)
15805.8791
15805.9274
15805.9361
Obs-Calc
-0.0004
-0.0002
0.0007
Fitted unsplit frequency: 15477.8951 MHz
F
6
7
!'
1
1
Obs.(MHz)
15477.8885
15477.9036
Obs-Calc
-0.0006
0.0006
Fitted unsplit frequency: 16262.2681 MHz
F
6
7
r
1
1
Obs.(MHz)
16262.2606
16262.2636
Obs-Calc
-0.0002
0.0002
Fitted unsplit frequency: 15962.4211 MHz
F
5
7
6
r
1
1
1
Obs.(MHz)
15962.4009
15962.4027
15962.4602
Obs-Calc
-0.0008
-0.0008
0.0016
Fitted unsplit frequency: 8451.6540 MHz
F
1
1
1'
1
1
Obs.(MHz)
8451.0311
8451.7784
Obs-Calc
0.0003
-0.0003
Fitted unsplit frequency: 12711.5882 MHz
F
3
1
I'
1
1
Obs.(MHz)
12711.7784
12712.0784
Obs-Calc
0.0007
-0.0007
84
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
413-404
F"
3
5
4
Fitted unsplit frequency: 6818.0801 MHz
I"
1
1
1
615-606
F"
5
7
3
5
4
!'
1
1
1
Obs.(MHz)
6817.6659
6817.8704
6818.6590
Obs-Calc
-0.0006
0.0008
-0.0002
Fitted unsplit frequency; 7499.5533 MHz
I"
1
1
909-818
F"
10
9
F
F
5
7
r
1
1
Obs.(MHz)
7499.1522
7499.3020
Obs-Calc
-0.0005
0.0005
Fitted unsplit frequency: 15791.0664 MHz
I"
1
1
P
9
8
r
1
1
Obs.(MHz)
15790.9022
15791.4566
Obs-Calc
-0.0011
0.0011
Quadrupole Coupling Constants:
Chi AA=0.502(5)
Chi BB=-2.493(3)
85
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CD
■—Di
O
o.
c
o
CD
Q.
■CDD
C/)
W
o'
o
Table 23: 5-Hexynenitrile Transition Assignments -AA conformer
o
J* Kp'
o
o
■D
cq'
CD
—i
Ko‘
J"
Kp"
Ko"
Observed/MHz
Obs-Calc
4
0
4
3
0
3
7443.3693
0.0016
A/MHz
11719.009(18)
4
1
4
3
1
3
7329.7767
-0.0005
B
959.6571(13)
5
0
5
4
0
4
9302.4601
0.0006
0
901.7685(8)
5
1
5
4
1
4
9161.7777
0.0004
Dj
0.000020(4)
6
0
6
5
0
5
11160.3823
-0.0036
D jk
-0.04155(14)
6
1
6
5
1
5
10993.4846
0.0003
di
0.000043(6)
6
1
5
6
0
6
11410.7326
0.0000
a
0.002239
7
0
7
6
0
6
13016.9182
0.0018
7
1
7
6
1
6
12824.8425
-0.0003
7
1
6
7
0
7
11623.9246
0.0000
CD
■—Di
O
o.
c
a
o
o
■—Di
o
CD
Q.
■CDD
(/)
(/)
w
o>
CD
■—Di
O
o.
c
o
CD
Q.
■CDD
C/)
W
o'
o
Table 24: 5-Hexynenitrile Transition Assignments -GA conformer
o
o
o
■D
cq
'
O
’
CD
—i
CD
J'
Kp'
Ko'
J"
Kp"
Ko"
Observed/MHz
Obs-Calc
3
0
3
2
0
2
6862.4858
0.0004
A/MHz
7019.2238(8)
3
1
3
2
1
2
6682.9641
0.0002
B
1207.13410(11)
3
1
2
2
1
1
7055.3471
-0.0002
0
1082.99483(13)
4
0
4
3
0
3
9140.7896
0.0000
Dj
0.0003291(8)
4
1
4
3
1
3
8908.3382
-0.0012
D jk
-0.005671(13)
4
1
3
3
1
2
9404.7448
0.0006
di
-0.0000691(8)
■—Di
O
o.
c
a
o
Q
■—Di
O
4
2
3
3
2
2
9159.0829
-0.0006
d2
-0.0000029(6)
4
2
2
3
2
1
9178.7238
0.0000
o
0.0010150
5
0
5
4
0
4
11411.2885
0.0000
5
1
5
4
1
4
11131.8204
0.0007
CD
Q.
5
1
4
4
1
3
11752.0380
-0.0001
5
2
4
4
2
3
11446.3390
0.0019
5
2
3
4
2
2
11485.5225
-0.0021
6
0
6
5
0
5
13672.1900
0.0000
6
1
6
5
1
5
13353.0036
0.0011
6
1
5
5
1
4
14096.6287
-0.0004
6
2
5
5
2
4
13731.9130
-0.0014
■CDD
(/)
(/)
w
’- s i
in o 00
o in o o
o o
o o o o o
o o o o o
o
o o
d d 9 9 o
in
o
CM
d
o
CM
CD
O
CD
h-
CM
rv .
o
CO
in
o
00
CO
00
CD
CO
T~
CD
in
in in
ITT-
T-
K
CO
'MCD
T—
CO
CO
CD
LO
CM
O
Lo
CO
^
CM
CO
00
o o
o o
o o
5 <o
T“
h-
to
in K cd
T“
o X— CD
1”
CO
CO
03
LO
CM
m
CM
O
CO
CO
CO CM
m
h-
CO
CO
CD
CM
CM
O
T-
T-
CM
O
CO
f* .
|x .
|N.
|x . CO
CD
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CD
■—Di
O
o.
c
o
CD
Q.
Table 25: 5-Hexynenitrile Transition Assignments -AG conformer
■CDD
C/)
w
o'
o
J'
o
o
o
■D
cq
Kp- Ko'
J"
Kp"
Ko"
Observed/MHz
Obs-Calc
3
0
3
2
0
2
6804.0505
0.0001
A/MHz
7372.7740(6)
3
1
2
2
1
1
6977.6495
-0.0002
B
1191.17918(15)
4
1
3
3
1
2
9301.6901
-0.0007
C
1078.87035(9)
4
2
3
3
2
2
9079.1194
0.0011
Dj
0.00032056(9)
5
0
5
4
0
4
11319.8480
0.0003
D jk
-0.00561 (5)
5
1
5
4
1
4
11062.9055
0.0007
di
-0.000065939(8)
5
1
4
4
1
3
11624.1142
-0.0004
d2
-0.0000032(7)
6
0
6
5
0
5
13567.2635
0.0004
H jk
0.0000012(6)
6
1
6
5
1
5
13271.4451
0.0001
a
0.000968
6
1
5
5
1
4
13944.4688
0.0003
6
2
5
5
2
4
13613.4741
-0.0011
6
2
4
5 ■2
3
13666.2214
-0.0022
7
0
7
6
0
6
15805.9138
0.0009
7
1
7
6
1
6
15477.8951
0.0000
7
1
6
6
1
5
16262.2681
0.0007
7
2
5
6
2
4
15962.4211
0.0009
4
1
3
4
0
4
6818.0801
-0.0002
6
1
5
6
0
6
7499.5533
0.0008
1
1
1
0
0
0
8451.6540
-0.0006
3
1
3
2
0
2
12711.5882
-0.0005
9
0
9
8
1
8
15791.0664
-0.0006
'
O
’
CD
—i
CD
■—Di
O
o.
c
a
o
Q
■—Di
O
CD
Q.
■CDD
(/)
(/)
w
CD
Conformational Assignment
Four of the five possible conformers (GG-cis is the exception) are nearly
prolate symmetric tops that show simple patterns of a-type, R branch transitions
at approximate integral frequency intervals of their respective B + C values. The
patterns displayed at intervals of 1860 MHz were identified as belonging to the
AA conformer. The conformer labeled AA was then rigorously assigned to the
heavy atom planar geometry based on the second moment, Pec- (See Table 26.)
Similar to the analysis of the AA conformer in n-butyl acetylene, three pairs of
hydrogen atoms from the methylene groups are outside the ab-plane. As each
methylene hydrogen pair contributes 1.6 uA^ to the second moment, a value of
3*1.6 uA^ is expected. The observed value of 4.69 uA^ is in excellent agreement.
Although the GA and AG have similar rotational constants, models
consistently predict the B and C rotational constants of the AG form to be about
1% smaller than values of the GA form. The observed values were assigned in
accordance with these observations. The spectral constants and comparisons of
other features of the spectra for the three rotamers are reported in Table 26.
90
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 26; Spectroscopic Constants of the Observed
Conformers of S-Hexynenitriie_________
AA
GA
AG
A /M H z
11719.009(18)
7019.2238(8)
7372.7740(6)
B
959.6571(13)
1207.13410(11)
1191.17918(15)
C
901.7685(8)
1082.99483(13)
1078.87035(9)
Paa/uA^
522.0
406.7
412.1
P bb
38.5
59.0
56.4
P cc
4.7
12.0
12.2
Dj/MHz
0.000020(4)
0.0003291 (8)d2
0.00032056(9)
D jk
-0.04155(14)
-0.005671(13)
-0.00561(5)
di
0.000043(6)
-0.0000691(8)
-0.000065939(8)
-0.0000029(6)
-0.0000032(7)
62
0.0000012(6)
Hjk
X aa
-3.5920(15)
0.502(5)
X bb
2.0222(15)
-2.493(3)
0.0010
0.00097
0
0.0022
91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
quadrupole coupling constants
If the molecule under investigation has a nucleus with a quadrupole
moment, the observed transitions show hyperfine structure. In a molecule like 5hexynenitriie, for example, the electric field gradient at the nitrogen nucleus
provides a torque which couples with the overall rotation of the molecule and
causes energy levels to split. The hyperfine splitting depends on the geometry of
the molecule, i.e., where the nitrogen sits in the inertial framework, and the
quantum numbers of a transition.
The dipole moments predicted for a particular conformer allow one to
predict its quadrupole coupling constants. It was assumed that the electric field
gradient at the N nucleus in n-butyl cyanide [5] is essentially unchanged by
substituting a methyl group for an acetylene group. The quadrupole coupling
constants for each conformer were predicted using the results of Bohn et al.’s
analysis of n-butyl cyanide [5]. These coupling constant values are reported in
Table 27 and are compared with the results of Bohn et al.’s n-butyl cyanide
results.
92
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CD
■—Di
O
o.
c
o
CD
Q.
■CDD
C/)
W
o'
o
o
Tabie27r^W q^S rupole coupling constants of the three conformers of 5Hexynenitrile and compared to those reported for n-butyl cyanide [R ef. 3]
o
o
■D
cq
'
AA
O’
CD
—i
CD
■—Di
O
o.
c
a
o
Q
■D
O
CD
Q.
■CDD
(/)
(/)
CO
GA
AG
n-butyl CN
5-Hexynenitrile
n-butyl CN
5-Hexynenitrile
n-butyl CN
Xaa (MHz )
-2.18(9)
-2.726(2)
-3.5920(15)
-3.645(3)
0.502(5)
-0.042(2)
X bb(MHz )
0.59(5)
0.674(2)
2.0222(15)
1.988(3)
-2.493(3)
-1.935(2)
5-Hexynenitrile
C.
Ab initio Calculations
To assess the stability of the five conformers of n-butyl acetylene, /i-butyl
cyanide, and 5-hexynenitrile, ab initio calculations were carried out with the
Gaussian 98 program [2], on a dual processor Dell 530 workstation operating
with the Linux OS. Geometry for the conformations of all three molecules was
optimized at two levels of theory: Hartree-Fock and MP2. Two different basis
sets, small and large, were used. Basis functions comprised of linear
combinations of Gaussian functions, are called contracted functions [6], Those
Gaussian functions for which the linear combinations are made are called
primitives. For example, the split basis set, 6-31G(d,p) is made up of the
following:
• six primitives for each inner shell orbital {“ 6” }
• two functions for each valence shell orbital
o
a contracted Gaussian composed of three primitives {“3”}
o
a single uncontracted primitive {“ 1”} (i.e. a basis function
consisting of a single Gaussian function)
• one d-type polarization function {“d”} for each heavy atom
• one p-type polarization function (“p”} for each hydrogen
The “split valence” refers to two sets of basis functions for each valence
shell orbital. Likewise, a “triple split valence” basis uses three sets of functions
for each orbital type. Split valence basis sets allow orbitals to change size within
a fixed molecular group, i.e., sigma, pi, etc. Inclusion of polarization functions
94
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
increases an orbital’s flexibility by adding orbitals with angular momentum
beyond that which is required for the minimum basis set (i.e., a basis set that
consists of only the orbitals represented in the molecule). Diffuse functions,
represented by the
in the larger basis set, are diffuse versions of valence size
s and p-type functions. These functions, which allow orbitals to occupy larger
regions of space, are important for systems where electrons are further away
from the nucleus, such as molecules with lone pair electrons, anions, and other
systems with high negative charge. The standard Gaussian 6-31G(d,p) split
valence set, augmented with polarization functions, was used in conjunction with
HF and MP2 calculations. The Gaussian 6-311G+(d,p) triple valence set,
augmented with polarization functions and diffuse functions on both carbon and
nitrogen, was used at the MP2 level.
The results of our ab initio geometry optimization calculations are given in
Tables 28-30 for n-butyl acetylene, Tables 31-33 for n-butyl cyanide and Tables
34-36 for 5-hexynenitrile, Harmonic vibrational frequencies were calculated at
both the HF and MP2 levels of theory to verify that stationary states of the five
conformations had been located. Appropriate scaling factors were used to
correct the calculated zero-point energies [7]. The energy ordering of the
conformers is found to depend mainly on level of theory rather than on the size of
the basis set. For n-butyl cyanide and n-butyl acetylene, HF calculations predict
that the AA form lies lowest in energy. At the correlated MP2 level of theory, the
GA conformer is predicted to be the lowest energy structure, using either of the
basis sets, 6-31G(d,p) or 6-311 ■+-G(d,p). A different energy ordering is observed
95
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
for 5-hexynenitrile. At the HF level, the GA conformer has the lowest energy
while at the correlated MP2 level, the GG-trans conformer Is lowest. These
results suggest that a dynamic correlation effect must be important in
determining the relative stability of these structures. Since we carried out the
quantum chemical calculations after the experiment was completed it became
apparent that we erred in not searching for the GG-trans conformer.
96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 28: Gradient optimized HF/6-31G(d,p) structures for
n-butyl acetylene
Param eter
AA
GA
AG
GG-trans
GG-cis
r(Ci-C 2 )
1.188
1.188
1.188
1.188
1.188
r(C2-C3)
1.471
1.472
1.471
1.471
1.473
r(C3-C4)
1.534
1.531
1.536
1.537
1.540
r(C4-C5)
1.529
1.529
1.532
1.532
1.533
r(C5-C6)
1.528
1.528
1.529
1.529
1.529
r(G i-Hii)
5.282
3.296
5.275
4.785
4.653
r(C2-Hii)
4.137
2.786
4.140
4.050
3.700
Z(C2-C3-C4)
112.84
113.55
112.44
113.51
115.18
Z(C3-C4'C5)
112.43
113.92
113.80
115.32
115.97
Z(C4-C5-C6)
112.82
112.64
114.68
114.46
115.90
Z(C2-C3-C4-C5)
180.0
295.2
176.7
299.5
280.3
Z(C3”C4-C5-C6)
180.0
180.6
65.9
297.0
68.4
Energy+ZPE
(-232+)
-0.8459
-0.8454
-0.8442
-0.8440
-0.8416
0.00
1.31
4.46
4.99
11.29
0.00
5.37
8.36
3.31
9.53
Rel.energy
(kJ/mol)
Gibbs energy
(kJ/mol)
Energies (hartrees), distances (A), angle (°), A (anti), G (gauche); zero-point
energies scaled by 0.9181 [Ref. 21]
97
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 29: Gradient Optimized MP2/6-31G(d,p) structure of
n-butyl acetylene
Parameter
AA
GA
AG
GG-trans
GG-cis
r(Ci-C2)
1.221
1.221
1.221
1.221
1.221
r(C2-C3)
1.464
1.465
1.464
1.465
1.465
r(C3-C4)
1.533
1.534
1.534
1.535
1.538
r(C4-Cs)
1.525
1.524
1.528
1.527
1.529
r(C5-C6)
1.525
1.524
1.526
1.525
1.525
r(C i-H ii)
5.298
3.190
5.284
4.737
4.661
r(C2-Hii)
4.124
2.698
4.125
4.003
3.669
Z^(C2-C3“C4)
113
113
112
113
114
Z{C3-C4-Cs)
112
113
113
114
115
Z{C4-C5~Ce)
112
112
114
114
115
Z(C2-C3-C4"C5)
180.0
298
176
304
282
Z(C3-C4-C5-C6)
Energy+ZPE
(-232+)
Rel.energy
(kJ/mole)
Gibbs energy
(kJ/mole)
180.0
181
63
301
65
-0.6802
-0.6806
-0.6792
-0.6800
-0.6775
1.05
0.00
3.68
1.58
8.14
0.61
0.00
2.46
3.24
9.17
Energies (hartrees), distances (A), angle (°), A (anti), G (gauche); zero-point
energies scaled by 0.9608 [Ref. 21]
98
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 30; Gradient Optimized MP2/6-311+G(d,p) structure of
H"butyl acetylene
Parameter
AA
GA
AG
GG-trans
GG-cis
r(Ci-C2)
1.220
1.220
1.220
1.220
1.220
r(C2-C3)
1.454
1.466
1.464
1.465
1.466
r(C3-C4)
1.537
1.538
1.538
1.540
1.542
r(C4-C5)
1.528
1.528
1.531
1.531
1.532
r(C5-C6)
1.529
1.529
1.530
1.530
1.530
r(C i-H ii)
5.300
3.183
5.279
4.715
4.711
r(C2-Hii)
4.129
2.702
4.127
4.001
3.718
Z(C2"C3“C4)
112.4
112.4
112.0
112.3
114.1
Z(C3-C4"C5)
112.5
113.0
113.1
113.7
114.5
Z(C4"C5-C6)
112.3
112.3
113.6
113.4
114.8
Z(C2-C3-C4-C5)
180.00
298.4
174.9
303.8
280.2
Z(C3-C4-C5-C6)
Energyn-ZPE
(-233+)
Rel.energy
(kJ/mole)
Gibbs energy
(kJ/mole)
180.00
181.7
62.0
301.2
62.5
-0.7645
-0.7651
-0.7636
-0.7645
-0.7622
1.79
0.00
4.07
0.72
7.86
2.86
0.00
3.92
1.43
8.45
Energies (hartrees), distances (A), angle (°), A (anti), G (gauche): zero-point
energies scaled by 0.9766 [Ref. 21],
99
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 31: Gradient optimized HF/6-31G(d,p) structures of
n-butyl cyanide
Param eter
AA
GA
AG
GG-trans
GG-cis
r(Ni-C2)
1.135
1.135
1.135
1.135
1.135
r(C2-C3)
1.472
1.473
1.472
1.473
1.473
r(C3-C4)
1.535
1.536
1.537
1.537
1.540
r(C4“C5)
1.529
1.524
1.532
1.532
1.533
r(C5-C6)
1.528
1.528
1.529
1.529
1.530
r(N i-H ii)
5.211
3.231
5.215
4.716
4.565
r(G2-Hii)
4.128
2.768
4.132
4.030
3.671
Z(C2-C3-C4)
112.2
112.8
112.0
112.8
114.5
Z(C3-C4-C5)
111.9
113.8
113.3
115.2
115.8
Z(C4-C5“C6)
112.6
112.4
114.7
114.5
115.8
Z(C2"C3-C4‘ C5)
180.0
295.8
176.7
299.6
281.9
Z(C3"C4-C5-C6)
Energy+ZPE
(-232+)
Rel.energy
(kJ/mol)
Gibbs energy
(kJ/mol)
180.0
181.0
65.7
296.8
68.4
-0.9179
-0.9176
-0.9163
-0.9162
-0.9141
0.00
0.78
4.28
4.56
9.99
0.52
0
3.36
4.14
9.17
Energies (hartrees), distances (A), angle (°), A (anti), G (gauche), zero-point
energies scaled by 0.9181 [Ref. 21]
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 32: Gradient optimized MP2/6-31G(d,p) structures of
n-buty! cyanide
Parameter
AA
GA
AG
GG-trans
GG-cis
r(Ni-C2)
1.181
1.181
1.181
1.181
1.181
r(C2-C3)
1.465
1.466
1.465
1.466
1.466
r(C3-C4)
1.533
1.534
1.534
1.535
1.537
r(C4*C5)
1.525
1.525
1.528
1.527
1.528
r(C5-C6)
1.524
1.524
1.525
1.525
1.526
r(N i-H ii)
5.249
3.126
5.241
4.663
4.577
r(C2-Hii)
4.115
2.684
4.119
3.981
3.651
Z(C2-C3-C4)
112.2
111.9
111.9
111.9
113.8
Z(C3-C4-C5)
112.0
113.2
112.8
113.9
114.9
Z(C4-C5-C6)
112.4
112.3
113.9
113.7
115.1
Z(C2-C3-C4"C5)
180.0
298.9
175.9
303.4
283.5
Z(C3-C4-C5-C6)
Energy-fZPE
(-249+)
Rel.energy
(kJ/mol)
Gibbs energy
(kJ/mol)
180.0
181.2
62.8
300.0
67.8
-0.7782
-0.7788
-0.7773
-0.7781
-0.7761
1.35
0.00
3.95
1.75
7.00
2.40
0.00
3.73
2.51
7.03
Energies (hartrees), distances (A), angle (°), A (anti), G (gauche); zero-point
energies scaled by 0.9608 [Ref. 21]
101
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 33: Gradient optimized MP2/6-311+G(d,p) structures of
/7-butyl cyanide
Parameter
AA
GA
AG
GG-trans
GG-cis
r(Ni-C2)
1.175
1.175
1.175
1.175
1.175
r(C2-C3)
1.465
1.467
1.465
1.466
1.466
r(C3-C4)
1.536
1.537
1.538
1.539
1.541
r(C4“C5)
1.528
1.528
1.532
1.531
1.532
r(C5-C6)
1.529
1.529
1.530
1.530
1.530
r(N i-H ii)
5.247
3.122
5.231
3.710
4.625
r(C2-Hii)
4.120
2.689
4.120
3.991
3.693
^(C2-C3"C4)
112.1
111.8
111.8
111.7
113.5
Z(C3-C4-C5)
111.9
113.0
112.6
113.7
114.4
Z(C4-C5"C6)
112.2
112.1
113.6
113.4
114.8
Z(C2-C3-C4"C5)
180.0
298.9
175.2
304.0
281.8
Z(C3-C4-C5-C6)
180.0
182.0
162.1
300.8
62.3
-0.8684
-0.8692
-0.8676
-0.8688
-0.8666
1.92
0.00
4.09
0.95
6.87
2.77
0.00
4.01
2.32
7.68
Energy+ZPE
(-249+)
Rel.energy
(kJ/mol)
Gibbs energy
(kJ/mol)
Energies (hartrees), distances (A), angle (°), A (anti), G (gauche); zero-point
energies scaled by 0.9766 [Ref. 21]
102
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 34: Gradient optimized HF/6-31G(d,p)
structures of 5-hexynenitrile
Param eter
AA
GA
AG
GG-trans
GG-cis
r(Ci-C2)
1.187
1.188
1.187
1.188
1.187
r(C2-C3)
1.470
1.472
1.470
1.472
1.470
r(C3-C4)
1.533
1.534
1.534
1.535
1.535
r(C4-C5)
1.534
1.534
1.535
1.535
1.539
rtCs-Cg)
1.472
1.472
1.473
1.473
1.473
r(C6-Ny)
1.135
1.135
1.135
1.135
1.135
r(Ci-Hi2)
5.288
3.226
5.292
4.753
4.096
r(C2-Hi2)
4.145
2.751
4.157
4.036
3.292
r(C6-H9)
4.130
4.631
3.400
4.038
4.284
r(N7-H9)
5.223
5.619
4.174
4.724
5.127
Z(C2-C3-C4)
112.23
113.27
111.94
112.88
114.63
Z(C3-C4-C5)
111.08
112.40
113.03
113.97
115.15
Z(C4-C5-C6)
112.11
111.94
112.87
112.42
114.79
Z(C2-C3-C4-C5)
180.0
297.57
177.70
297.24
295.03
Z(C3-C4-C5"C6)
Energy+ZPE
(-285+)
ReI.energy
(kJ/mol)
Gibbs energy
(kJ/mol)
180.0
181.21
63.12
295.66
82.80
-0.5669
-0.5677
-0.5671
-0.5677
-0.5631
2.13
0.13
1.72
0.00
12.13
3.05
0.00
1.42
0.36
11.12
Energies (hartrees), distances (A), angle (°), A (anti), G (gauche); zero-point
energies scaled by 0.9181 [Ref. 21]
103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 35: Gradient optimized MP2/6-31G(d,p) structures of
5-hexynenitrile
P aram eter
AA
GA
AG
GG-trans
GG-cis
r(Ci-C2)
1.221
1.221
1.221
1.221
1.221
r(C2-C3)
1.463
1.464
1.463
1.464
1.463
r(C3-C4)
1.532
1.533
1.532
1.533
1.533
r(C4-C5)
1.533
1.532
1.533
1.532
1.535
KCs-Ce)
1.465
1.464
1.466
1.466
1.467
r(Ce-N7)
1.181
1.181
1.181
1.181
1.181
r(Ci-Hi2)
5.302
3.106
5.303
4.695
4.127
r(C2-Hi2)
4.128
2.655
4.141
3.987
3.283
r(C6-H9)
4.114
4.630
3.302
3.987
4.212
r(N7-Hg)
5.249
5.657
4.073
4.671
5.063
Z(C2-C3-C4)
111.98
112.33
111.87
112.00
113.82
Z(C3”C4-C5)
111.14
111.60
112.34
112.51
114.12
Z(C4-C5-C6)
111.97
111.93
111.94
111.57
114.22
Z(C2-C3-C4-C5)
180.0
300.49
177.28
300.28
297.08
Z(C3-C4-C5-C6)
Energy+ZPE
(-286+)
ReI.energy
(kJ/mol)
Gibbs energy
(kJ/mo!)
180.0
181.35
59.82
298.97
77.14
-0.5267
-0.5283
-0.5277
-0.5292
-0.5252
6.53
2.27
3.91
0.00
10.57
6.30
1.55
2.98
0.00
9.42
Energies (hartrees), distances (A), angle (°), A (anti), G (gauche); zero-point
energies scaled by 0.9608 [Ref. 21]
104
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 36: Gradient optimized MP2/6-311+G(d,p) structures of
5-hexynenitrile
Param eter
AA
GA
AG
GG-trans
GG-cis
r(Ci-C2)
1.219
1.220
1.220
1.220
1.220
r(C2-C3)
1.463
1.465
1.463
1.465
1.464
r(C3-C4)
1.536
1.537
1.536
1.537
1.537
r(C4-C5)
1.536
1.535
1.537
1.536
1.539
r(C5-C6)
1.465
1.464
1.466
1.466
1.467
r(Ce-N7)
1.175
1.175
1.175
1.175
1.175
r(Ci-Hi2)
5.304
3.110
5.297
4.693
3.981
r(C2-Hi2)
4.134
2.668
4.141
3.991
3.186
r(Ce-H9)
4.121
4.637
3.292
3.989
4.218
r(Ny-H9)
5.249
5.656
4.055
4.662
5.055
Z(C2"C3-C4)
111.75
112.16
111.67
111.81
113.40
Z(C3-C4-Cb)
111.03
111.45
112.20
112.40
113.60
Z(C4-C5-C6)
11 1.78
111.76
111.81
111.41
113.91
Z(C2-C3-C4"C5)
180.0
300.67
122.61
301.52
299.92
Z(C3-C4-C5-Ce)
180.0
182.10
59.71
300.50
80.25
-0.6311
-0.6333
-0.6326
-0.6348
-0.6301
8.41
3.42
5.22
0.00
11.57
7.52
2.52
4.17
0.00
10.22
Energy+ZPE
(-286+)
ReI.energy
(kJ/mol)
Gibbs energy
(kJ/mol)
Energies (hartrees), distances (A), angle (°), A (anti), G (gauche); zero-point
energies scaled by 0.9766 [Ref. 21]
105
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The observation that the energy ordering of the five conformers depends
on the level of theory used in the calculation is an important result. The
calculated geometries for both the HF and MP2 level treatments have been
examined. Calculated atomic distances between the triple bonded groups and
the hydrogen atoms located further down the hydrocarbon chain are also
reported in the tables. At both the HF and MP2 levels there is a net attractive
electrostatic interaction between the rc-electron density in the triple bonded
groups (— 0 = 0 — or —0=N) and the closest hydrogen located on the nextnearest neighbor carbon. This net attraction is largest for the GA and GG-trans
conformer geometries for n-butyl cyanide (and n-butyl acetylene) since the N^H’’"' and 0^-H^^ (and 0^-H” and C^-H^'') bond lengths are shortest for these
structures. See Figure 5 for atom labeling.
106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 5 Atom labels for S-Hexynenitriie and />butyl acetylene
5-Hexynenltrlle
The GG-cis conformers also exhibit this net attraction but any energy
improvement there is offset through crowding of the cyanide and acetylene
groups. The net attraction is strongest at the MP2 level of theory owing to
correlation effects. In the case of 5-hexynenitrile, the GA and GG-trans
conformers are very close at the HF level; the GG-trans conformer is clearly the
lowest energy structure at the MP2 correlated level of theory. Relative to the GA
structure, the net attractive electrostatic interaction is nearly twice as strong for
the GG-trans conformation owing to the presence of triple bonded functional
groups at both ends of 5-hexynenitrile.
107
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
These experimental and computational studies demonstrate that tripiebonded functional groups are as stable or more stable in gauche conformational
arrangements as compared to anti conformations. This is in contrast to most
organic functional groups that are typically more stable in the anti conformations.
These results support the model discussed by Tr®tteberg et al. [8] and first
proposed by Wodarczyk and Wilson [9] in the 1970s. Traetteberg’s analysis of npropyl cyanide shows an energy ordering that favors the gauche conformer. As
n-butyl cyanide is isolectronic and structurally similar to n-butyl acetylene, we
predict that the GA conformer of n-butyl cyanide is the most stable structure, or,
at least as stable as the AA form. The microwave analysis by Bohn et al. [10] of
the relative energies of the GA, AA, and AG conformers of n-butyl cyanide,
based on relative spectral intensities, favors the GA and AA structures. This
observation agrees with our current studies of the theoretical interpretation
presented above.
A note on the zero-point enerqv scaling
Zero-point energies calculated using the Gaussian 98 programs, were
scaled in order to calculate energies that are more accurate. Theoretical zeropoint energies are generally overestimated. Neglect of anharmonicity,
incomplete incorporation of electron correlation, and the use of finite basis sets
tend to overestimate vibrational frequencies and, consequently, of calculated
zero-point energies. For example, improper dissociation obtained from the
Hartree-Fock potential energy curve leads to overestimation of vibrational
108
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
frequencies. Fortunately, this overestimation has been shown to be rather
uniform with regard to the basis set used in the calculation [7], As such, scaling
factors are often applied to the ab initio zero-point energy and the result is a
scaled zero-point energy (labeled ZPE®° in this paper) that is much closer to the
true zero-point energy. Scaling factors were determined from a comparison
between zero-point energy values derived from theoretically determined
harmonic vibrational frequencies, with those values determined, from a large set
of experimental harmonic frequencies of similar molecules. The scaling factors
used in the energy calculation are 0.9181 for HF/6-31G(d,p), 0.9608 for MP2/631 G(d,p), and 0.9766 for MP2/6-31 UG(d,p).
109
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Literature Cited
1. High! Walker, A. R.; Lou, Q.; Bohn, R; K.; Novick, S. E. Rotational spectra
of methyl ethyl and methyl propyl nitrosamines. Conformational
assignment, internal rotation and quadrupole coupling. J. Mol. Struct.
1995, 346, 187-195.
2. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman
JR, Zakrzewski VG, Montgomery JA Jr, Stratmann RE, Burant JO,
Dapprich S, Millam JM, Daniels AD, Kudin KN, Strain MG, Farkas O,
Tomasi J, Barone V, Gossi M, Gammi R, Mennucci B, Pomelli G, Adamo
G, Clifford S, Ochterski J, Petersson GA, Ayala PY, Gui Q, Morokuma K,
Malick DK, Rabuck AD, Raghavachari K, Foresman JB, Gioslowski J,
Ortiz JV, Baboul AG, Stefanov BB, Liu G, Liashenko A, Piskorz P,
Komaromi I, Gomperts R, Martin RL, Fox DJ, Keith T, Al-Laham MA, Peng
GY, Nanayakkara G, Gonzalez G, Ghallacombe M, Gill PMW, Johnson B,
Ghen W, Wong MW, Andres JL, Head-Gordon M, Replogle RS, Pople JA.
Gaussian 98, Revision A. 7. Gaussian, Inc., Pittsburgh PA, 1998.
3. Watson, J. K. G. In Vibrational Spectra and Structure, Vol. 6; Durig, J. R.,
Ed.; Elsevier: Amsterdam, 1977; pp 1-89.
4. Kraitchman, J. Determination of Molecular Structure from Microwave
Spectroscopic Data. American Journal of Physics 1953, 21, 17-24.
5. Bohn, Robert K.; Pardus, Jennifer L.; August, Julie; Brupbacher, Thomas;
Jager, Wolfgang. Microwave studies of the three conformers of butyl
cyanide. J. Mol. Struct. 1997, 413-414, 293-300.
6. Foresman, James B.; Frisch, /Eleen. Exploring Chemistry with Electronic
Structure Methods; Gaussian, Inc.: Pittsburgh, PA, 1996; pp 97-110.
110
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7. Scott, Anthony. P.; Radom, Leo. Harmonic Vibrational Frequencies: An
Evaluation of Hartree-Fock, Moller-Plesset, Quadratic Configuration
Interaction, Density Functional Theory, and Semiemprical Scale Factors.
J. Phys. Chem. A, 1996, 100(41),
1 6 5 0 2 -1 6 5 1 3 .
8. Tr^tteberg, M.; Bakken, P.; Hopf, H. Structure and conformations of
gaseous butyronitrile: 0 — H - n interaction? J. Mol. Struct. 2000, 556,
189-196.
9. Wodarczyk, Francis J.; Wilson, E. Bright. Rotational Isomerism in 1Pentyne from the Microwave Spectrum. J. Chem. Phys. 1972, 56 (1), 166176.
10. Bohn, Robert K.; Pardus, Jennifer L.; August, Julie; Brupbacher, Thomas;
Jager, Wolfgang. Microwave studies of the three conformers of butyl
cyanide. J. Mol. Struct. 1997, 413-414, 293-300.
Ill
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
V. Conformational Analysis of Benzyl Alcohol and
Quantum Chemical Calculations of Benzyl Alcohol and
Benzyl Fluoride
A. introduction
Benzyl alcohol displays another, but different kind of weak hydrogen
bonding interaction which will be described in the following pages. The
conformational properties of benzyl compounds, CeHsCHaX, have been the
subject of several spectroscopic and theoretical studies [1-11]. When X is an
alkyl group, an amine or a halide, the C-C-C-X dihedral angle is 90° and the C-X
bond lies in a plane orthogonal to the benzene ring [12-19] (Figure 1). This
orientation is consistent with a model that minimizes steric repulsion effects
between the X of the -C H 2 X group and the closest hydrogens in the phenyl ring.
If X is ethynyl, -C=CH, or cyanide, -C=N, the C-C-C-X dihedral angle is 0° and
the C-X bond lies in the plane of the benzene ring [20]. These structures suggest
a model where there is weak hydrogen bonding between the rt-electron cloud of
the ligand and the nearest hydrogen in the phenyl ring.
In a third group, where X is -NH2 o r-O H , several experimental studies [111, 21] have attempted to determine the most stable molecular conformation,
Benzylamine has two stable conformers, with C-C-C-N dihedral angles of 90°
and 64° [19]. For benzyl alcohol, the stable conformations, shown in Figure 2,
are characterized by two dihedral angles, t(CCCO) and cp(CCOH).
112
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 1: Conformations of Benzyl -X
orthogonal
' -alkyl
X=
»NH2
' -halide
planar
X=
skew
-C=CH
-CEN
or
gauche
-NH.
OH
113
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 2: Conformations of Benzyl Alcohol
gauche: x « -60°, cj) ^ 60'
trans: x «-60°, A « 180°
114
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Experimental studies have only found evidence for the gauche
conformation of cp (OH group located over the phenyl ring) while several
theoretical calculations indicate a second stable trans conformation about the CO
bond, with the OH group oriented away from the phenyl plane. There has been
little agreement about the value of the dihedral angle x(CCCO) among the
several experimental studies. IR and NMR studies of benzyl alcohol in the
condensed phase suggest values of 0°, 60°, or 90° [1 ,4]. Gas phase electron
diffraction studies suggest a value of 54° [5], and IR-UV spectroscopic studies
report that benzyl alcohol has a stable planar conformer with both i(CCCO) and
(p(CGOH) ~ 0° [3]. Most recently, both UV and IR-UV ion dip spectroscopy
studies [22], and REMPI and photoelectron spectroscopy studies [23] have been
carried out which conclude that the COCO dihedral angle of benzyl alcohol lies in
the range 35° to 60°. The conflicting conclusions described above are not
surprising if the rotational barrier about the dihedral angle t(GGCO) is very small
and, also, sensitive to those studies carried out in the condensed phase. In an
effort to provide more precise estimates of the structure of benzyl alcohol highresolution microwave spectroscopy studies were undertaken.
115
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
B. Experimental
Benzyl alcohol was purchased from Aldrich Chemical Co (> 99% pure).
An NMR spectrum showed no significant impurities and the sample was studied
without further purification. A few drops of the sample was placed in a reservoir
before the nozzle of the pulsed-jet Fourier Transform microwave spectrometer
[24] of the Southern New England Microwave Consortium [29], and its room
temperature vapor swept into the chamber by an atm of first-run Ne (-75% Ne,
-25% He). The nozzle is mounted in one of the resonant cavity mirrors and
rotational transitions were observed in the region between 5 and 18 GHz.
Transitions were observed as Doppler doublets and line widths in the power
spectrum range from 1 0 -2 0 GHz with uncertainties in the measured frequencies
estimated to be less than 2 kHz. Spectra were also observed for the deuterated
OD isotopomers between 5 and 25 GHz. A small amount of benzyl alcohol was
allowed to equilibrate with a several-fold excess of D2 O in the presence of trace
KOH. An NMR spectrum just prior to the experiment showed that the sample
consisted mostly of the deuterated
0 0
sample showed new transitions of the
isomer; and, microwave spectra of this
0 0
isotopomer with remaining traces of
the spectrum of the OH compound. For both compounds, it was found that a low
backing pressure (ca. 0.5 atm) produced the most intense spectrum.
Benzyl alcohol is a prolate near-symmetric top with an asymmetry
parameter value of approximately -0.83. Its spectrum shows characteristic a-
i16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
type R-branch transitions and the most intense Kp = 0 and 1 lines have been
tentatively assigned for both the OH (Table 1a, 1b) and OD (Table 2a, 2b)
isotopomers. The transitions are all split into doublets which we believe is due to
tunneling motions in the compounds. The tunneling splittings range from 5-16
MHz for the J = 3 ^ 2 transitions and steadily increase with J to 16 - 46 MHz for J
= 6<-5. With deuterium substitution, the corresponding splittings decrease to 312 MHz for J = 3<-2 and 4-11 MHz for J = 6<r5 indicating that motion of the OH
(OD) group plays a major role in the tunneling path.
117
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CD
■—
Di
O
o.
c
o
CD
Q.
■CDD
C/)
Table 1a; Tentative Benzyl Alcohol Fit Set 1
22
o'
o
o
o
o
■D
cq'
CD
—
i
CD
■—
Di
O
co.
a
o
o
■—
Di
J'
Kp'
Ko'
J"
Kp"
Ko"
Observed Frequency
MHz
Obs.-Calc.
MHz
Rotational
Constants
2
0
2
1
0
1
5293.4582
-2 . 0
A = 4612.8(8)
3
0
3
2
0
2
7895.3082
-1 . 6
B = 1471.82(16)
4
0
4
3
0
3
10445.7728
-0 . 1
0 = 1185.26(11)
5
0
5
4
0
4
12935.0568
2 .0
6
0
6
5
0
5
15363.2908
2.9
Paa = 330 uA^
3
1
3
2
1
2
7528.4833
-1 . 6
Pbb = 96 uA^
4
1
4
3
1
3
10016.8034
-3.2
Pcc=13uA^
5
1
5
4
1
4
12492.0535
-2.7
B+C=2657 MHz
6
1
6
5
1
5
14953.2236
0 .1
7
1
7
6
1
6
17397.0601
2 .0
4
1
3
3
1
2
11158.3450
-3.3
5
1
4
4
1
3
13911.7530
0 .1
6
1
5
5
1
4
16633.9186
3.4
3
2
2
2
2
1
7968.2471
-3.0
4
2
2
3
2
1
10795.1830
-1 . 0
4
1
3
3
0
3
15387.6854
1 .2
o
CD
Q.
■CDD
(/)
CO
00
k
= -0.83
a = 2.3 MHz
CD
■—
Di
O
o.
c
o
CD
Q.
■CDD
C/)
W
o'
o
2
2
1
1
1
0
15023.0550
-0 . 6
o
o
o
■D
cq'
Table 1b: Tentative Benzyl Alcohol Fit Set 2
CD
—
i
Obs.-Calc.
MHz
2
Observed Frequency
MHz
7878.8977
-3.3
Rotational
Constants
A = 4583(37)
0
3
10421.0008
-2 . 1
B = 1472.17(38)
4
0
4
12900.1236
-0.4
0 = 1180.94(35)
6
5
0
5
15316.7802
1
3
2
1
2
7519.6968
8.9
Pbb = 97 uA^
4
1
4
3
1
3
9992.0274
-1.5
Pec = 13 uA^
5
1
5
4
1
4
12458.4581
-2 . 1
6
1
6
5
1
5
14910.7901
0 .0
3
1
2
2
1
1
8379.5314
-4.1
4
1
3
3
1
2
11151.5310
-2 . 0
5
1
4
4
1
3
13900.3997
0.5
6
1
5
5
1
4
16618.1526
4.5
J'
Kp
Ko'
J"
Kp"
Ko"
3
0
3
2
0
4
0
4
3
5
0
5
6
0
3
CD
■—
Di
O
o.
c
a
o
o
■D
o
CD
Q.
■CDD
(/)
(/)
<0
0 .1
P aa
K
0
= 330 uA^
= -0.83
= 4 MHz
7)
CD
■D
O
Q.
C
o
CD
Q.
■CDD
C/)
W
o'
o
Table 2a: Tentative d-Benzyl Alcohol Fit Set 1
o
o
■oD
cq
'
3CD
J'
Kp-
Ko‘
J1"
Kp"
Ko"
Observed Frequency
MHz
Obs.-Calc.
MHz
Rotational
Constants
3
0
3
2
0
2
7744.3324
-0.707
A = 4777.4(5)
3
1
2
2
1
1
8213.3184
0.330
B = 1439.49(6)
4
1
4
3
1
3
9832.0005
0.674
C = 1163.88(5)
4
1
3
3
1
2
10930.4515
0.365
Paa = 339 uA^
5
1
5
4
1
4
12262.5905
0.206
Pbb= 94 uA^
5
1
4
4
1
3
13627.3211
-0.372
Pcc= 11 uA^
6
1
6
5
1
5
14678.2129
-0.372
B-t-C=2603 MHz
2
1
2
1
0
1
8269.1812
0.019
CD
■D
O
Q.
C
a
o
3
■D
O
k
= -0.85
a = 0.55 MHz
CD
Q.
■CDD
C/)
C/)
ro
o
7
J
CD
■—
Di
O
o.
c
o
CD
Q.
■CDD
(/)
W
o'
o
o
—
ty
Table 2b: Tentative d-Benzyl Alcohol Fit Set 2
J'
Kp'
Ko'
J"
Kp"
Ko"
Observed Frequency
MHz
Obs.-Calc.
MHz
Rotational
Constants
3
0
3
2
0
2
7742.1117
-1.48
A = 4790.3(14)
3
1
2
2
1
1
8212.6004
0.30
B = 1439.50(15)
4
0
4
3
0
3
10253.8260
2.57
C = 1163.39(10)
4
1
4
3
1
3
9828.4706
0.088
Paa = 340 uA^
4
1
3
3
1
2
10929.2964
0 .1 2
Pbb = 94 uA^
5
0
5
4
0
4
12704.5792
1.40
Pec = 11 uA^
5
1
5
4
1
4
12258.1668
-0.53
B+C=2602 MHz
5
1
4
4
1
3
13625.7116
-0.85
6
1
6
5
1
5
14672.9262
-1.224
2
1
2
1
0
1
8280.3790
-0.135
CD
O
O
■D
CD
CD
—
i
CD
■—
Di
O
o.
c
a
o
o
■—
Di
o
CD
Q.
■CDD
C/)
C/)
ro
k
= -0.85
a = 1.38 MHz
C. Results
To date we have been unable to model the spectrum of benzyl alcohol
accurately because the tunneling motion is fundamentally a 2-dimensional
torsion, coupling C-C-C-0 and C-C-O-H torsions, both of which involve
asymmetric potentials and asymmetric internal rotors (Figure 3). Nevertheless,
we have reproduced the essentials of the microwave spectrum by a model with A
= 4612.8(8) MHz, B = 1471.82(16) MHz, and C = 1185.26(11) MHz. The
preliminary geometry was made by plotting out the calculated rotational spectrum
for X = 90° and x = 0°, respectively, and comparison of these spectra were made
with the observed spectrum (Figure 4). The Pec second moment has the value
12-13 uA^, consistent with a -65° C-C-C-0 torsional angle. Figure 5 shows a
plot of Fee values at various CCCO dihedral angles for CCOH torsional angle
equal to 60° and 180°, respectively.
122
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
■■
I
«#—
o
CD
0)
3
O
3
4-» £
CD
o
C <
_a
CO Sx
> N
£
3 0>
—
o
o
wm
UJ
i.
3
O
y.
aa
CO
CD
»c
o
3
CO
0
m
'8L
3
O)
E
p
123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
u
o
o
o
o
N
CD
O
U.
O-
CD
CD
m
XJ
o
o
o
o <
o
“■5 >
■a
C
D
N
c
DL 0
m
T3
c o
as
u E
0) 3
lU
> 4-1
SBBBSBBiSSia
m O
m 0
Cl
n
m
O
Ba «
c
oBSO
0 B
aSg
u
3 «
ra ©
LL ££
CO
o
CO
o
II
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
m
m
U)
c
CM
»<
3
00
T3
CO
C
O
O
O
03
£0
o
CD T -
O
J
T—
00
c
0)
o
CD
o
ffi
m
D
X!
.52
o
o
.9
DL
CO
>
O
0)
3
O
CD
CO
CO
>
o
JZ
o
CD
Cfl
X!
_o
03
<
o
C
CO
S'
c
O
0
DQ
O
O
O
O
c/3
0
_D
CO
>
o
o
o
CM
a.
LO
0
D
D)
k .
o
CD
(,yn) -d
125
R eproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
The splitting pattern of benzyl alcohol’s rotational transitions into doublets
reminded us of an earlier study on benzyl fluoride made in our group. In benzyl
fluoride, the microwave spectrum of benzyl fluoride is qualitatively similar to that
of benzyl alcohol, also displaying tunneling splittings of the rotational transitions
[16]. The splittings are due to the internal rotation between the phenyl and CH2 F
groups. In benzyl fluoride, however, the torsional splittings are 500 to 1000 times
smaller than the splittings observed in benzyl alcohol. Bohn et al. fit the benzyl
fluoride rotational spectrum using a model of distinct torsional ground and excited
states with slightly different rotational energy level manifolds, displaced by the
tunneling splitting and connected by c-type rotation-torsion transitions. The two
manifolds are displaced by the tunneling splitting, 3.426(2) MHz which translates
into a 2 -fold barrier of 0.67 kJ/mol for benzyl fluoride.
Benzyl alcohol C-C-C-0 torsional minima are separated by about 50°
(-65° to -115°) rather than the 180° of benzyl fluoride. Thus, for the same
barrier height, larger tunneling splittings are expected in benzyl alcohol than in
benzyl fluoride; however, the observed splittings are so much larger in benzyl
alcohol that we believe the 0.67 kJ/mol benzyl fluoride barrier value is a
conservative upper limit to the barrier in benzyl alcohol. The potential surface is
described by the CCCO torsion, t, and the CCOH torsion, (j), and the two
dimension potential surface connects four equivalent structures. Rotation
between / and // {Hi and iv) involves two coordinates (Figure 3): flipping the CH2 OH laterally (x) and adjusting the hydroxyl group, (j). The interchange
between / and Hi {11and iv) requires movement through the plane of the benzene
126
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ring. Since transitions are only split into doublets, consistent with a double
minimum problem, we believe that tunneling is only important in one of those
interchanges. Computations (see below) suggest that the / and // {Hi and iv)
interchange is the lower barrier process.
D. Theoretical Calculations
In order to assess the relative stability of the several possible
conformational structures of benzyl fluoride and benzyl alcohol, a series of ab
initio calculations was carried out at the HF and MP2 levels of theory. Our
previous MP2 studies of alkyl chains with triply bonded substituents indicated
good agreement between the relative energies of predicted conformations and
the experimental ordering of their stability [26, 27]. In contrast, the HF and DFT
calculations were not capable of predicting the correct stability ordering of these
substituted alkyl chains that exhibit only a few kilojoules difference in the energy
of the stable conformations. The basis sets used in the present study ranged
from the standard Gaussian 6-31G (d,p) split-valence set the Gaussian 6 311+(3df,2p) triple-zeta valence set augmented with extensive polarization
functions. Optimized geometries were found for the stable conformations at both
the HF and MP2 levels of theory. All calculations were performed using the
Gaussian 98 set of programs [28] implemented on a dual-processor Dell 530
workstation operating with the Linux OS. Harmonic vibrational frequencies were
carried out at both the HF and MP2 levels of theory to verify that stationary states
127
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of the stable conformers had been located. Tight optimization criteria were
employed due to the small energy changes with dihedral angle variation.
Our optimized energy calculations for benzyl fluoride are shown in Table 3
for both the planar (x = 0°) and perpendicular (x = 90°) orientation of the fluorine
atom relative to the phenyl plane. The stable orientation has x = 90°, in
agreement with several previous studies [12,15]. Our calculations indicate that
diffuse functions on the heavy atoms are essential to yield the correct stable
structure for benzyl fluoride. As shown in Table 3, both the HF and MP2
calculations predict the incorrect orientation (x = 0 °) for this molecule in the
absence of diffuse functions in the basis. Rotational constants and the
calculated relative energies of the stable perpendicular orientation and the planar
transition state orientation are shown in Table 4. Our calculated rotational
constants for the stable form of benzyl fluoride agree with the experimental
results [Ref. 15]; the agreement is particularly good in B and C. The splittings
observed in the microwave spectrum are consistent with a small 2 -fold barrier of
0.69 kJ/mol. The MP2 calculated energy difference between the two orientations
of the ligand appears to be converging to the measured experimental value,
albeit slowly, with increase in the basis set. To examine this difference in energy
between the perpendicular and planar forms of benzyl fluoride in more detail,
optimized structure calculations were carried out at the QCISD/6-311+G(d,p) and
CBS-QB3 levels of theory. As indicated in Table 4, both of these methods, which
more fully account for the correlation error than MP2 theory, predict a much
128
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
smaller energy difference that appears to be converging to the measured
experimental value.
129
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 3: Relative Energies of Benzyl Fluoride
(kJ/mol)
t (90°-0°)
A(E®* + Z P E ® °^
(kJ/mol)
x(90°-0°)
Transition
State
3.6
-4.0
90° form
-0.076
0.28
0
° form
0
0.34
0
° form
MP2/6-31G(d,p)
0 .2 2
0.39
9 0 °form
MP2/6-31+G(d,p)
-4.2
-2 . 6
0° form
MP2/6-311+G(d,p)
-3.7
-1 . 2
0
MP2/6-311+G(3d,2p)
-2.4
N/A
N/A
MP2/6-311+G(3df,2p)
-2 . 0
N/A
N/A
QCISD/6-311+G(d,p)
-2.39
N/A
N/A
GBS-QB3
-0.87
N/A
N/A
Calculation
HF/6-31G(d,p)
HF/6-31+G(d,p)
HF/6-311+G(d,p)
Experiment (Microwave)
° form
-0.696(5)
130
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
®
2
_
_
^_
o
3
>«
N
C
o
OOBBX.,
N
CO
CO b - CD CO CD q
CM IX )
m
CD IX ) CM 1—
CM K
05 CD c\i d
CO
CJ)
CM CD CM 00 CO
O)
T-"
■?— ■«5~- T“ T--- T“ T~T—
—
a>
n—
I
T”
CO
X— f —
■?—
T“
T“
05
CD
00
t
T~
CO
05 05 CD T» T~ 05
CD
CO m C^ C\j CD CO
05 CM CO CO CD q
T“
T~
1—
T~
T”
T”
T“
T“
s
iuiiAa
00
T“
<n
CO
(0
TO
S
C
<D
a ..
m
c
m
m
>
’■ *3
a
o
N
c
CQ
o
O
<D CM CO h - CO q
o
o q XT CO CM
CO O CM cq
to
C\j d
K d CM CD CO CO CM ■jL d
CO i r i CO CD d
CS|
CO CO CD CO CO CO LO CO
CM LO CO CO
'S t
'd'5 t
x r "M1— ■
rr1— T - t — 1—
•i— 1“
T— T—
t— T—
1“
CO
CM CD 05
CM
t
K
in
't T—
CM
CO
"St
CM
T“
«
c
o
*«p
T"
S
o
DC
■u
c
cvi d CO d
CM O ) T” (X )
o
o h-
CD CM o
CO CO O )
CM O ) CM
O
h - O)
in
in
O
q
'c f-
h-
in
's f
'cr
q
05 CD ■r* CO q
T -'
C£) CJ)
CD IT— CO ■1— LO CO CO CD
CD CD CD 05 CD 05 CO 05
's f
Xf
'= t
T-
q
'C f
q
q
q
KT
CD
05 - r “
CD 05
1—
'S t
'S t
u:> m 05
CM
CD 05 CO q
"St •St CM
o
CO
CO
'S t
c
CO
<#«•
m
c
Q
O
15
c
o
*3
CO
o
QC
BB
®
n
m
o
H
o
O
o
O
o
0
o Oo o
CD
o>
0
O) O
o
O
o
O
o
O
O
CD O
05
o
O
o
O
o
o
o
O) O
Q.
£?
o
o
qI
0)
x:
h-
*o
>
0
_J
q
CL
S
CD
q
Q .
t i '
0
O
+
+
5
m
d
CD
O
+
m
CO
<h
CO
CM
CO
CO
CO
CM
u
ix
u
Q
3
3
3
2
d
C
D
+
o
o
O
05
05 O
q
■q I
q
Q_
o
o
05
O
05
o
lO
q
d
d
CO
CO
0
0
0
DC
0
+
+
+
m
CO
CO
CM
CO
CM
CO
CD
Q_
CL
a.
Q
W
O
2
2
2
O
CM
o
O
d
•r“
CO
CD
0
o
o o
CD
•4—*
m
O
(h
m
o
C
CD
P
LL
‘ l0
Q .
X
UJ
131
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In contrast, benzyl alcohol exhibits two stable conformations at both the
HF and MP2 level of theory. This molecule, which is isoelectronic with benzyl
fluoride, has stable conformations in both the gauche form (t ~ -60°, (j)« +60°),
with the OH group located above the phenyl plane, and in the trans form (x » 60°, p « 180°), with the OH group located behind the phenyl ring. Structural
models of these two stable forms are shown in Figure 2. Our optimized energy
calculations for benzyl alcohol are shown in Table 5 for both the gauche and
trans conformations. Both structures are shown to be stable at both the HF and
MP2 level of theory for all of the basis sets that were explored. Benzyl alcohol is
predicted to be a Ci structure for both gauche and trans conformations of the OH
group. In contrast to the results for benzyl fluoride, the planar and perpendicular
orientations of X, in C-C-C-X, toward the phenyl plane are found to be transition
states. Calculated rotational constants and relative energies of the two stable
forms for this molecule are shown in Table 5. The calculated dihedral angle
t(CCGO) is sensitive to both the level of theory and the basis set. MP2
calculations predict that the dihedral angles x(GGCO) and (p(GGOH) are
approximately 58° and -52°, respectively, in agreement with the recent study by
Mens et al. [22], Our best estimates for the rotational constants are in
reasonable agreement with our preliminary microwave spectroscopy results, as
discussed in the previous section. The calculated energy difference between
these two stable structures for benzyl alcohol is ca. 5 kj/mol. As shown in Table
5, our MP2 calculations indicate a small barrier
3.3 kJ/mol) to the OH group
132
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
rotation from the gauche form through the transition state with ^ = 0°. Our
experimental studies suggest that the barrier is much smaller. The rotation from
the trans form through cj) = 180° exhibits a very weak barrier since there is no
effective OH---n bonding for this orientations. Presently, only the gauche form
has been positively identified in experimental studies.
133
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CD
■—
Di
O
o.
c
o
CD
Q.
■CDD
C/)
(/)
O
O
■D
cq
'
O’
CD
—
i
CD
■—
Di
O
o.
c
a
o
■—
D
Oi
CD
Q.
■CDD
(/)
(/)
Table 5: Rotational Constants (MHz) and Relative Energies (kJ/mol) of the Conformations
Benzyl Alcohol
Level of theory
HF/6-31G(d,p)
gauche
trans
HF/6-31+G(d,p)
gauche
trans
HF/6-311+G(d,p)
gauche
trans
MP2/6-31(d,p)
gauche
trans
MP2/6-31+G(d,p)
gauche
trans
MP2/6-311+G(d,p)
gauche
trans
MP2/6-311+G(3d,2p)
gauche
trans
MP2/6-311+G(d,p)
transition state, Cs
MP2/6-311+G(d,p)
transition state, Cs
Experiment (this work)
AE®'
+5.1
Dihedral (°)
t
<P
A
8
0
36
-59
15
175
4919.0
5013.9
1489.0
1501.8
1180.5
1167.6
44
4886.2
5007.6
1480.6
1499.0
1183.0
1164.2
+3.5
12
-59
176
+3.2
46
9
-58
176
4890.3
5022.5
1481.3
1501.1
1186.5
1165.3
+6 . 1
35
37
-56
171
4831.2
4864.6
1482.2
1492.7
1170.3
1173.1
+4.4
57
73
-54
178
4704.0
4666.1
1471.1
1478.7
1191.8
1213.1
+4.9
60
70
-52
176
4692.7
4683.5
1468.1
1474.9
1192.3
1205.2
58
77
-52
178
4712.3
4678.1
1465.7
1476.5
1187.2
1214.9
90
0
4607.3
1467.7
1214.7
4643.8
4612.8(8)
1470.9
1471.82(16)
1214.8
1185.26(11)
+5.0
E®'(ts)-E®'(gauche)
+3.3
E®\ts)-E®'(gauche)
+0 . 2 2
gauche
90
180
—
E. Discussion
The experimental microwave study of benzyl fluoride [15] unambiguously
identified the CCCF dihedral angle to be 90° for the equilibrium structure. In
contrast, an electron diffraction study reports benzyl fluoride to be a Ci structure
with a CCCF dihedral angle of -52° [21], The authors of this study considered
four models to fit their data: a single planar conformer, a single 90° minimum, a
single asymmetric conformer, and a free intemally rotating model. Of these, the
model that best fits their data is an asymmetric model with a mean dihedral angle
of 52°. Since the electron diffraction nozzle temperature was 313 K, the
experiment described in [2 1 ] sampled a thermal average of torsional states of the
ground vibrational state plus several appreciably populated torsionally exoited
states. Since kT at 313 K is 2.63 kJ/mol, it is likely that any of the four models
proposed by Trastteberg et al. [21] could fit the electron diffraction data. The
microwave study supports a model with a 90° dihedral angle and a 0.69 kJ/mol
rotational barrier [15] (Figure 6 ).
135
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
|V .
to
LL
E
o
Q)
JD
O
S B
h«
m
o>
Q-
B
m
m
m
B
i
X
■ S o
■ §S
3 o
)
c
JZ
o
0
q=
N
C
Q
C
CD
-4— »
C
CQ
CD
■c
Q)
O
£T
•x
o
XJ
cb9
0
=3 0
_D)“^
LL
jQ
S
O
§
CM
X
LL
X
136
R eproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Previous studies of benzyl alcohol have suggested values for the dihedral
angle x(CCCO) of benzyl alcohol ranging from 0° to 90°. The recent
experimental studies by Mons et al. [22] and Dessent et al. [23] clearly indicate
that the benzyl alcohol species called gauche in this thesis has Ci symmetry,
with estimates of the dihedral angle x(CCCO) ranging from 35° to 60°. The major
uncertainty is the nature of the long-range forces that gives rise to this observed
dihedral angle. The current explanation, given in the most recent studies of
benzyl alcohol, is that of long-range intramolecular bonding between the OH
group and the n -electron density in the phenyl ring. This effect is well
documented in the literature, dating back to the 1960s [2]. This attractive
electrostatic interaction does explain why the gauche form is more stable that the
trans form, where the QH - tc bonding is minimized or absent, but offers no real
explanation for the observed CCCO dihedral angle value, -60°, of the gauche
form.
The electron diffraction study of benzyl alcohol [5] reports a mode! with a
CCCO dihedral angle of -54°. In this study the authors considered a single
structure and found a best fit to the experimental data. They did not consider the
model developed in this work, an equilibrium structure with dihedral angle near
60°, which exhibits a very flat potential energy surface. The electron diffraction
studies were carried out at 385 K (kT = 3.23 kJ/mol) so it is likely that the authors
sampled structures thermally averaged over a number of torsionally excited
states. The electron diffraction data is not very sensitive to the torsion; the
137
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
authors note that; “A planar conformer cannot be excluded at this stage, but the
54° conformer obviously fits the experimental data best” [5].
In an effort to find another explanation for the observed gauche
conformation of benzyl alcohol, atomic point charges were calculated from a fit to
the electrostatic potential. The calculation indicates a dipole coupling between
the -C-H group in the methylene side chain and the closest -C-H group in the
phenyl ring. This dipole coupling results in a nearly planar orientation of the -CH group in the methylene chain with the closest -C-H group in the phenyl ring, as
illustrated in Figure
6
. A planar orientation of these -C-H groups would result in
a dihedral angle x(CCCO) of 60°, close to the latest experimental estimates and
in agreement with the prediction of our best MP2 calculations as shown in Table
5. The Cs'Hb and C 9 -H 1 3 bonds (see Figure 7) lie in the same plane within 2°.
138
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 7: Charge Density* of Gauche
Conformation of Benzyl Alcohol
-
0.16
+0.09
*MP2/6-311+G(3d,2p)
Courtesy of Albeiro Restrepo
139
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Literature Cited
1. Schaefer, Ted; Sebastian, Rudy; Peeling, James; Renner, Glenn H.;
Koh, Kevin. Conformational Properties of benzyl alcohol in dilute
solution. Can. J. Chem 1989, 67, 1015-1021.
2. Campbell, I. D.; Eglinton, G.; Raphael, R. A. An Infrared Study of
Competitive Intramolecular Hydrogen Bonding in a Series of Benzyl
Alcohols. J. Chem. Soc. B 1968, 338-343.
3. Guchhait, Nikhil; Ebata, Takayuki; Mikami, Naohiko. Discrimination of
Rotamers of Aryl Alcohol Homologues by Infrared-Ultraviolet DoubleResonance Spectroscopy in a Supersonic Jet. J. Am. Chem. Soc.
1999, 121 (24), 5705-5711.
4. Visser, T.; van der Maas, J. H. Structural information from OH
stretching vibrations—XVI1. On the different conformers in
benzylalcohols and anthracylmethanols. Spectrochim. Acta 1986, 42A
(5), 599-602.
5. Traetteberg, Marit; Ostensen, Herta; Ragnhild, Seip. The Molecular
Structure and Conformation of Gaseous Benzyl Alcohol by Electron
Diffraction. Acta Chem. Scand. 1980, A 34 (6 ), 449-454.
6
. Im, Hoong-Sun; Bernstein, E. R.; Secor, Henry V.; Seeman, Jeffrey I.
Supersonic Jet Studies of Benzyl Alcohols: Minimum Energy
Conformations and Torsional Motion. J. Am. Chem. Soc. 1991, 113
(12), 4422-4431.
7. Abraham, R. J.; Bakke, J. M. Rotational Isomerism—22 An NMR Study
of O H ...71 Bonding and the Conformations of Benzyl Alcohol and
Derivatives. Tetrahedron, 1978, 34 (19), 2947-2951.
8
. Guchhait, Nikhil; Ebata, Takayuki; Mikami, Naohiko. Structures of
hydrogen-bonded clusters of benzyl alcohol with water investigated by
infrared-ultraviolet double resonance spectroscopy in supersonic jet.
J. Chem. Phys. 1991, 111 {18), 8438-8447.
9. Kowski, Klaus; Luttke, Wolfgang; Rademacher, Paul. Evidence for
intramolecular GH - tu hydrogen bonding in unsaturated alcohols from
UV photoelectron spectroscopy. J. Mol. Struct. 2001, 567-568, 231240.
1 0
. Li, S.; Bernstein,E. R. Benzyl alcohol—water and benzyl alcohol—
ammonia clusters: Ion fragmentation and chemistry. J. Chem. Phys.
1992, 9 7 ( 10), 7383-7391.
140
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11 .Bakke, Jan M.; Lorentzen, Geir B. intramoiecular Hydrogen Bonding in
Aryl Substituted Ailphatic Alcohols. Acta. Chem. Scand. 1974, B 28
(6), 650-656.
12. Schaefer, Ted; Schurko, Robert W.; Sebastian, Rudy; Hruska, Frank
E. Experimental and theoretical assessments of the substituent and
medium dependence of the internal rotational potentials in benzyl
fluoride. 3,5-Difluorobenzyl fluoride and 4-fluorobenzyl fluoride. Can.
J. Chem. 1995, 73, 816-825.
13. Sorenson, Sterling A.; True, Nancy S. Temperature-dependent lowresolution microwave studies of benzyl halides. J. Mol. Struct. 1991,
263, 21-35.
14.Celebre, G.; De Luca, G.; Longeri, M. The barrier to internal rotation in
benzyl fluoride investigated using the N.M.R. spectra of samples
dissolved in liquid crystalline solvents. Mol. Phys. 1989, 67 (2), 239248.
15. Bohn, Robert K.; Sorenson, Sterling A.; True, Nancy S.; Brupbacher,
Thomas; Gerry, Michael C. L.; Jager, Wolfgang. The Conformation and
Internal Rotation Barrier in Benzyl Fluoride from Its Microwave
Spectrum. J. Mol. Spectrosc. 1997, 184, 167-171.
16. Schaefer, Ted; Peeling, James; Sebastian, Rudy. Solvent-induced
conformational changes in benzyl fluoride. Long-range G,F; H,H; H,F
spin-spin couplings. Can. J. Chem. 1985, 63, 3219-3225.
17. Shelley, M. Y.; Dai, H. -L.; Troxler, T. Electronic relaxation and
vibrational predissociation of benzene-acetylene van der Waals
complexes. J. Chem. Phys. 1999, 110 (18), 9081-9090.
18. Ramos, Christopher; Winter, Paul R.; Stearns, Jaime A.; Zweir,
Timothy S. The Spectroscopic Consequences of C-H-- re H-Bonding:
C 6 H6 -(C 4 H2 )n Clusters with n= 1 and 2. J. Phys. Chem. A, 2003, 107
(48), 10280-10287.
19.Melandri, Sonia; Maris, Assimo; Favero, Paolo G.; Caminati, Walther.
Intramolecular Hydrogen Bonds and Conformational Properties of
Benzylamine. Chem. Phys. Chem. 2001, 7(3), 172-177.
20.Liu, Xiao-Zhou; Bohn, Robert K.; Sorenson, Sterling A.; True, Nancy S.
Rotational Spectra of Benzyl Cyanide Assignment of the Planar
Conformer and Evidence of a Low Barrier to Internal Rotation. J. Mol.
Struct. 1991, 243, 325-339.
141
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21 .Abraham, R. J.; Hearmon, R. A.; Tr«tteberg, M.; Bakken, P. The
Molecular Structure of Benzyl Fluoride -A Combined Theoretical and
Experimental Study. J. Mol. Struct. 1979, 57, 149-162.
22. Mons, Michel; Robertson, Evan G.; Simons, John P. Intra- and
Intermolecuiar rc-Type Hydrogen Bonding in Aryl Alcohols: UV and IRUV Ion Dip Spectroscopy. J. Phys. Chem. A, 2000, 104 (7), 14301437.
23. Dessent, Caroline. E. H.; Geppert, Wolf D.; Ulrich, Susanne; MullerDethlefs, Klaus. Ionization-induced conformational changes: REMPl
and ZEKE spectroscopy of salicyl and benzyl alcohol. Chem. Phys.
Lett. 2000, 319, 375-384.
24-Balle, T. J.; Flygare, W. H. Fabry-Perot cavity pulsed Fourier transform
microwave spectrometer with a pulsed nozzle particle source. Rev.
Sci. Instrum. 1981, 52(1), 33-45.
25. Might Walker, A. R.; Lou, Q.; Bohn, R. K.; Novick, S. E. Rotational
spectra of methyl ethyl and methyl propyl nitrosamines. Conformational
assignment, internal rotation and quadrupole coupling. J. Mol. Struct.
1995, 346, 187-195.
26. Atticks, K.; Bohn, R. K.; Michels, H. H. Three Conformers Observed
and Characterized in 1-Hexyne. Int. J. Quantum. Chem. 2001, 85, 514.
27. Atticks, K. A.; Bohn, R. K.; Michels, H. H. Structure and Relative
Energies of the Conformers of n-Butyl Cyanide and 5-Hexynenitrile.
Int. J. Quantum. Chem. 2002, 90, 1440.
28. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M.
A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A.
D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.;
Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford,
S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.;
Malick, 0. K.; Rabuck, A. 0.; Raghavachari, K.; Foresman, J. B.;
Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, 0. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.;
Wong, H. W.; Andres, J. L.; Head-Gordon, M.; Replogle, E. S.; Pople,
J. A. Gaussian 98, Rev. A.6, Gaussian, Inc., Pittsburgh PA, 1998.
142
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29. Scott, Anthony. P.; Radom, Leo. Harmonic Vibrational Frequencies:
An Evaluation of Hartree-Fock, Moller-Plesset, Quadratic Configuration
Interaction, Density Functional Theory, and Semiemplrical Scale
Factors. J. Phys. Chem. A, 1996, 100 (41), 16502-15513.
143
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Документ
Категория
Без категории
Просмотров
0
Размер файла
5 252 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа