close

Вход

Забыли?

вход по аккаунту

?

Organoisocyanates: The conformational stability determination by infrared, raman and microwave spectroscopy and ab initio calculations

код для вставкиСкачать
ORGANOISOCYANATES: THE CONFORMATIONAL STABILITY DETERMINATION
BY INFRARED, RAMAN AND MICROWAVE SPECTROSCOPY
AND AB INITIO CALCULATIONS
A DISSERTATION IN
Chemistry
and
Geosciences
Presented to the Faculty of the University
of Missouri-Kansas City in partial fulfillment of
the requirements of the degree
DOCTOR OF PHILOSOPHY
by
XIAO HUA ZHOU
B.A., University of Missouri-Kansas City, 2008
Kansas City, Missouri
2012
i
UMI Number: 3556121
All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.
UMI 3556121
Published by ProQuest LLC (2013). Copyright in the Dissertation held by the Author.
Microform Edition © ProQuest LLC.
All rights reserved. This work is protected against
unauthorized copying under Title 17, United States Code
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, MI 48106 - 1346
@ 2012
XIAO HUA ZHOU
ALL RIGHTS RESERVED
ii
ORGANOISOCYANATES: THE CONFORMATIONAL STABILITY DETERMINATION
BY INFRARED, RAMAN AND MICROWAVE SPECTROSCOPY
AND AB INITIO CALCULATIONS
Xiao Hua Zhou, Candidate for the Doctor of Philosophy Degree
University of Missouri−Kansas City, 2012
ABSTRACT
Conformational stability studies have been carried out for several quasilinear
organoisocyanates and organoisothiocyanates. The infrared and Raman spectra in the
gaseous, liquid and solid phases have been recorded as well as the variable temperature
infrared spectra of the sample compound dissolved in rare gas solution. The complete
vibrational assignments have been proposed for all the energetically stable conformations.
Microwave spectra for several of the ring molecules have been investigated from 10,000 to
21,000 MHz with transitions for the most stable conformer assigned.
For methylisocyanate, the barrier to internal rotation and linearity has been
determined to be 46 cm-1 and 975 cm-1, respectively.
For the Ethylisocyanate, the cis
conformation is determined to be more stable than the trans form by 100 ± 4 cm-1.
dimethylsilylisocyanate was determined to have an essentially linear SiNCO moiety. The
spectral analyses of Isopropylisothiocyanate indicated one stable conformer (trans) in the
annealed solid but in the fluid phases most of the molecules have energies above the barriers
iii
of the two predicted bound vibrational states i.e., trans and gauche forms. For
dimethylsilylisocyanate, the low wavenumber Raman spectrum of the gas with a significant
number of Q-branches for the SiNC(O) bend is consistent with an essentially linear SiNCO
moiety. For cyclopropylisocyanate, the enthalpy difference has been determined to be 77 ± 8
cm-1 with the trans form more stable than the cis conformer with 59 ± 2% of the trans form
present at ambient temperature. For cyclobutylisocyanate, an enthalpy difference of 131 ± 13
cm-1 was obtained with the equatorial-trans conformer the more stable form.
cyclohexylisocyanate, an enthalpy difference of 397
For
40 cm-1 was obtained from seven
conformer pairs with the equatorial-trans form more stable than the axial-trans conformer.
To support the analysis of the vibrational assignment, ab initio and hybrid density
functional theory calculations have been carried out. For each molecule, a completed
vibrational analysis and potential energy distributions have been proposed. The adjusted r0
structural parameters have been determined for all molecules.
iv
The faculty listed below, appointed by the Dean of the School of Graduate Studies,
have examined a dissertation titled “Organoisocyanates: The Conformational Stability
Determination by Infrared, Raman and Microwave Spectroscopy and ab initio Calculation”
presented by Xiaohua (Sarah) Zhou, candidate for the Doctor of Philosophy degree, and
hereby certify that in their opinion it is worthy of acceptance.
Supervisory Committee
James R. Durig, Ph.D., D.Sc., Chairperson
Curators’ Professor of Chemistry and Geosciences
Zhonghua Peng, Ph.D.
Professor of Chemistry
Nathan A. Oyler Ph.D.
Assistant Professor of Chemistry
Jejung Lee, Ph.D.
Associate Professor of Geosciences
James B. Murowchick, Ph.D.
Associate Professor of Geosciences
v
TABLE OF CONTENTS
ABSTRACT ............................................................................................................
ii
LIST OF TABLES .....................................................................................................
viii
LIST OF ILLUSTRATIONS .....................................................................................
xiii
CHAPTER
1
INTRODUCTION .............................................................................
1
2
THEORECTICAL BACKGROUND ................................................
8
3
EXPERIMENTAL .............................................................................
27
4
VIBRATIONAL SPECTRA AND STRUCTURAL
PARAMETERS OF SOME XNCO AND XOCN (X=H, F, CL,
BR) MOLECULES ............................................................................
31
THE r0 STRUCTURAL PARAMETERS, VIBRATIONAL
SPECTRA, AB INTIO CALCULATIONS AND BARRIERS TO
INTERNAL ROTATION AND LINEARITY OF
METHYLISOCYANATE .................................................................
68
CONFORMATIONAL STABILITY, STRUCTURAL
PARAMETERS AND VIBRATIONAL ASSIGNMENT FROM
VARIABLE TEMPERATURE INFRARED SPECTRA OF
XENON AND KRYPTON SOLUTIONS AND AB INITIO
CALCULATIONS OF ETHYLISOCYANATE ...............................
96
MICROWAVE, RAMAN, AND INFRARED SPECTRA, r0
STRUCTURAL PARAMETERS, CONFORMATIONAL
STABILITY AND AB INITIO CALCULATIONS OF
ISOPROPYLISOCYANATE ............................................................
128
RAMAN AND INFRARED SPECTRA, CONFORMATIONAL
STABILITY, AB INITIO CALCULATIONS AND
VIBRATIONAL ASSIGNMENT OF
DIMETHYLSILYLISOCYANATE ..................................................
165
CONFORMATIONAL STABILITY FROM VARIABLE
TEMPERATURE INFRARED SPECTRA OF XENON
SOLUTIONS, r0 STRUCTURAL PARAMETERS, AND AB
INITIO CALCULATIONS OF CYCLOPROPYLISOCYANATE ..
182
5
6
7
8
9
vi
10
MICROWAVE, RAMAN, AND INFRARED SPECTRA, r0
STRUCTURAL PARAMETERS, CONFORMATIONAL
STABILITY AND AB INITIO CALCULATIONS OF
CYCLOBUTYLISOCYANATE .......................................................
217
MICROWAVE, INFRARED, AND RAMAN SPECTRA, r0
STRUCTURAL PARAMETERS, CONFORMATIONAL
STABILITY AND AB INITIO CALCULATIONS OF
CYCLOHEXYLISOCYANATE .......................................................
255
REFERENCE LIST ...................................................................................................
305
VITA
316
11
............................................................................................................
vii
LIST OF TABLES
TABLE
Calculated frequencies (cm-1) and potential energy distributions
(PEDs) for XNCO (X = H, D, F, Br, Cl) from the MP2(full)/ and
B3LYP/6-31G(d) basis set .................................................................
49
Calculated frequencies (cm-1) and potential energy distributions
(PEDs) for XNCO (X = H, D, F, Br, Cl) from the MP2(full)/ and
B3LYP/6-31G(d) basis set .................................................................
51
Structural parameters, rotational constants, dipole moments and
energies for XNCO (X = H, F, Br, Cl) ..............................................
53
Rotational and centrifugal distortion constants for isotopomers of
HNCO and BrNCO in the ground vibrational state ...........................
55
Force constants (mdyn/Å) experimentally determined and
predicted for BrNCO..........................................................................
57
Structural parameters, rotational constants, dipole moments and
energies for XOCN (X = H, F, Br, Cl) ..............................................
58
Calculated barriers to NCO moiety linearity (cm-1) and barriers
(Hartree) to molecular linearity (cm-1) of XNCO (X = H, F, Cl and
Br) molecules .....................................................................................
60
Calculated barriers to OCN moiety linearitya (cm-1) and barriers
(Hartree) to molecular linearity (cm-1) of XOCN (X = H, F, Cl and
Br) molecules .....................................................................................
61
Structural parameters, rotational constants, dipole moments and
energies for CH3NCO from the 6-311+G(d,p) ..................................
84
10
Observed and calculated frequencies (cm-1) for methyl isocyanate ..
85
11
Calculated energies (hartree) and barriers (cm-1) to methyl torsion,
NCO linearity and molecular linearity of the heavy atoms of
CH3NCO ............................................................................................
86
Coriolis fine structure of pseudodegenerate CH3 antisymmetric
stretching and deformation bands ......................................................
87
1
2
3
4
5
6
7
8
9
12
viii
13
Rotational (MHz) and centrifugal distortion constants (kHz) for
CH3NCO............................................................................................
88
14
Observed and calculated frequencies (cm-1) for cis ethylisocyanate .
113
15
Observed and calculated frequencies (cm-1) from the MP2/631G(d) level for trans ethylisocyanate ..............................................
114
Calculated frequencies (cm-1) at the MP2(full)/6-31G(d) level for
gauche ethylisocyanate ......................................................................
115
Calculated electronic energies (hartree) and energy differences
(cm-1) for the gauche, cis, trans, and the skew conformers of
ethylisocyanate ...................................................................................
110
Temperature and intensity ratios of the conformational bands of
ethylisocyanate from the infrared spectra of the liquid krypton of
the gas ................................................................................................
117
Structural parameters (Å and degree), rotational constants (MHz)
and dipole moments (debye) for the cis and trans conformers of
ethylisocyanate from ab initio calculations (6-311+G(d,p) and
experimental data ...............................................................................
118
Rotational (MHz) and centrifugal distortion (kHz) constants for cis
ethylisocyanate ...................................................................................
119
Calculated and experimental Rotational Constants (MHz) of
ethylisocyanate ...................................................................................
120
Rotational
transitional
frequencies
(MHz)
for
trans
isopropylisocyanate in the ground vibrational state...........................
146
Rotational (MHz), centrifugal distortion (kHz) and quadrupole
coupling (MHz) constants for the trans and gauche conformers of
isopropylisocyanate............................................................................
147
Calculated (MP2(full)/6-31G(d)) and observed vibrational
frequencies (cm-1) of isopropylisocyanate for the trans conformer ...
148
Calculated (MP2(full)/6-31G(d)) and observed vibrational
frequencies (cm-1) of isopropylisocyanate for the gauche
conformer ...........................................................................................
150
16
17
18
19
20
21
22
23
24
25
ix
Calculated electronic energies (hartree) and energy differencesa
(cm-1) for the trans, gauche, skew and cis conformers of
isopropylisocyanate............................................................................
152
Structural parameters (Å and degree), rotational constants (MHz)
and dipole moment (debye) for trans isopropylisocyanate from the
6-311+G(d,p) basis set and adjusted r0 parameters............................
153
28
Symmetry coordinates for isopropylisocyanate .................................
154
29
Temperature and intensity ratios of the conformer bands of
isoproylisocyanate from Raman and infrared spectra of the liquid
xenon solution ....................................................................................
155
Transitions not utilized for the determination of the rotational
constants for isopropylisocyanate ......................................................
156
Calculated and observed vibrational wavenumbers (cm-1) of
dimethysilylisocyanate.at the MP2(full)/6-31G(d) level ...................
175
Structural parameters (Å and degree), rotational constants (MHz)
and dipole moments (debye) for dimethylsilylisocyanate at the
MP2(full)/6-311+G(d,p) level............................................................
177
33
Symmetry coordinates for dimethylsilylisocyanate ...........................
178
34
Calculated energiesa,b (hartree) for the possible conformers of
dimethylsilylisocyanate......................................................................
179
Calculated and Observed Vibrational Frequencies (cm-1) of
Cyclopropylisocyanate for the Trans Conformer at the
MP2(full)/6-31G(d) level ...................................................................
199
Calculated and Observed Vibrational Frequencies (cm-1) of
Cyclopropylisocyanate for the Cis conformer at the MP2(full)/631G(d) level .......................................................................................
201
Calculated Electronic Energies (hartree) and Energy Differencesa
(cm-1) for the Trans, Cis and the Skew conformers of
Cyclopropylisocyanate .......................................................................
203
Structural parameters (å and degree), rotational constants (mhz)
and dipole moments (debye) for the trans and cis
cyclopropylisocyanate from ab initio calculations (6-311+g(d,p)
basis set) and experimental data ........................................................
204
26
27
30
31
32
35
36
37
38
x
39
Symmetry coordinates for cyclopropylisocyanate .............................
205
40
Temperature and Intensity Ratios of the Conformer Bands of
Cyclopropylisocyanate from the Infrared Spectra of the Liquid
Xenon Solution ..................................................................................
206
Comparison of rotational constants (mhz) obtained from modified
ab initio MP2(full)/6-311+g(d,p) predictions, microwave spectra,
and the adjusted structural parameters for cyclopropylisocyanate ....
207
Rotational (mhz), centrifugal distortion (khz) and quadrupole
coupling constants for the trans
and cis conformers of
cyclopropylisocyanate........................................................................
208
A comparison of cyclopropylisocyanate structural parameters (å
and degree), rotational constants (Mhz) and dipole moments
(debye) for the different conformers from MP2/6-31g(d,p)
calculation ..........................................................................................
209
Rotational transitional frequencies (MHz) for equatorial-trans
cyclobutylisocyanate in the ground vibrational state .........................
235
Rotational (MHz), centrifugal distortion (kHz) and quadrupole
coupling (MHz) constants for the equatorial-trans and equatorialgauche conformers of cyclobutylisocyanate at 6-311+G(d,p) basis
set .......................................................................................................
236
Calculated (MP2(full)/6-31G(d)) and observed vibrational
frequencies (cm-1) of cyclobutylisocyanate for the equatorial-trans
conformer ...........................................................................................
237
Calculated (MP2(full)/6-31G(d) and observed vibrational
frequencies (cm-1) of equatorial-gauche cyclobutylisocyanate ........
239
Calculated (MP2(full)/6-31G(d) and observed vibrational
frequencies (cm-1) of axial-trans cyclobutylisocyanate .....................
241
Calculated electronic energiesa (hartree) and energy differencesb
(cm-1) for the equatorial-trans, equatorial-cis, axial-trans and
axial-cis conformers of cyclobutylisocyanate ...................................
243
Structural parameters (Å and degree), rotational constants (MHz)
and dipole moments (debye) for the equatorial-trans, equatorialgauche and axial-trans conformers of cyclobutylisocyanate from
MP2(full) and B3LYP/6-311+G(d,p) ................................................
244
41
42
43
44
45
46
47
48
49
50
xi
51
Symmetry Coordinates for Cyclobutylisocyanate .............................
246
52
Temperature and intensity ratios of the conformer bands of
cyclobutylisocyanate from the infrared spectra of the liquid xenon
solution...............................................................................................
247
Comparison of MP2/6-311+G(d,p) calculated structural parameters
(Å and degree) with adjust r0 values for c-C4H7-X (X = NCO, OH,
F, Cl, Br) ............................................................................................
248
Rotational transitional frequencies (MHz) for equatorial-trans
cyclohexylisocyanate in the ground vibrational state ........................
274
Rotational (MHz), centrifugal distortion (kHz) constants for the
different conformers of cyclohexylisocyanate ...................................
275
Calculated
and
observed
frequencies
(cm-1)
for
cyclohexylisocyanate equatorial-trans (Cs) form ..............................
276
Calculated
and
observed
frequencies
(cm-1)
for
cyclohexylisocyanate axial-trans (Cs) form .....................................
279
Calculated electronic energies (hartree) and energy differences
(cm-1) for the axial-trans, equatorial-trans, axial-cis and
equatorial-cis conformers of cyclohexylisocyanate ..........................
282
Structural parameters (Å and degree), rotational constants (MHz)
and dipole moment (debye) for equatorial-trans and axial-trans
cyclohexylisocyanate from the 6-311+G(d,p) basis set .....................
283
60
Symmetry coordinates for cyclohexylisocyanate ..............................
285
61
Temperature and intensity ratios of the conformer bands of
cyclohexylisocyanate from the infrared spectra of the liquid xenon
solution...............................................................................................
287
Calculated electronic energies (hartree) and energy differencesa
(cm-1) for the axial and equatorial conformers of several
substituted cyclohexanes ....................................................................
288
Structural parameters (Å and degree), rotational constants (MHz)
for equatorial and axial c-C6H11XYZ (XYZ = NC, CN, CCH)
from the MP2(full) and B3LYP/6-311+G(d,p) basis set ...................
289
Heavy atom r0 structural parameters of cyclohexane and some
monosubstituted chair-equatorial cyclohexanes ................................
291
53
54
55
56
57
58
59
62
63
64
xii
LIST OF ILLUSTRATIONS
FIGURE
1
2
3
4
5
6
7
Vibrational spectra of FNCO: (A) experimental infrared spectrum
from argon matrix obtained by UV photolysis of FC(O)N3; †
spectrum of precursor; * the original assignment of 6, later
reassign to 5 band; simulated spectra from MP2/6-31G(d)
calculation (B) infrared and (C) Raman.............................................
62
Vibrational spectra of ClNCO: (A) simulated infrared spectrum
from MP2/6-31G(d) calculation; (B) experimental infrared
spectrum of the gas; (B΄) an expansion of the weaker bands by five
times; * indicating the observed fundamentals; (C) simulated
Raman spectrum from MP2/6-31G(d) calculation ...........................
63
Vibrational spectra of BrNCO: (A) simulated infrared spectra from
MP2/6-31G(d) calculation; experimental (B) infrared spectrum of
the gas (resolution: 1 cm-1), (C) infrared spectrum in neon matrix
(resolution: 0.5 cm-1), (D) Raman spectrum of the liquid
(resolution: 4 cm-1); and (E) simulated Raman spectra from
MP2/6-31G(d) calculation .................................................................
64
Vibrational spectra of HNCO and DNCO: (A) experimental
infrared spectrum of HNCO in nitrogen matrix; and (A΄) HOCN
spectrum generated from HNCO; (B) simulated infrared and (C)
Raman spectra of HNCO from MP2/6-31G(d) calculation; (D)
simulated infrared and (E) Raman spectra of DNCO from MP2/631G(d) calculation. Note scale changes ............................................
65
Vibrational spectra of HOCN and DOCN: (A) experimental
infrared spectrum of HOCN in nitrogen matrix: (
) background,
(-----) after photolysis of HNCO; (B) simulated infrared and (C)
Raman spectra of HOCN from MP2/6-31G(d) calculation; (D)
simulated infrared and (E) Raman spectra of DOCN from MP2/631G(d) calculation. Note scale changes .............................................
66
Simulated vibrational spectra from MP2/6-31G(d) calculation:
FOCN (A) infrared and (B) Raman; ClOCN (C) infrared and (D)
Raman; BrOCN (E) infrared and (F) Raman ....................................
67
Infrared spectra of CH3NCO: experimental infrared spectrum of
the gas (A) and of the solid (B); simulated spectra from scaled
MP2/6-31G(d) values (C) and B3LYP/6-311+G(d,p) calculation
(D) ......................................................................................................
89
xiii
8
Observed fine structures on the CH3 antisymmetric stretch in the
infrared spectra of the gas ..................................................................
90
Observed fine structures on the CH3 antisymmetric stretch in the
infrared spectra of the gas ..................................................................
91
Observed fine structures on the CH3 antisymmetric deformation in
the infrared spectra of the gas ............................................................
92
11
Observed spectra of solid CH3NCO: infrared (A), Raman (B) .........
93
12
Low frequency spectra of CH3NCO: infrared of gas (A), infrared
of solid (B), Raman of solid (C) ........................................................
94
Raman spectrum of the gas for the CNC bend with “hot band”
transitions ...........................................................................................
95
Newman projections for the four different possible conformers of
ethylisocyanate ...................................................................................
121
Far-infrared spectra of ethylisocyanate: (A) gas; (B) crystalline
solid ....................................................................................................
122
Predicted (MP2(full)/6-311+G(d,p)) and observed infrared spectra
of ethylisocyanate: (A) observed spectrum the sample dissolved in
liquid krypton at 155 C; (B) predicted spectrum of the mixture of
cis and trans conformers with H = 100 cm-1; (C) predicted
spectrum of pure trans conformer; and (D) predicted spectrum of
pure cis conformer .............................................................................
123
Infrared spectra of ethylisocyanate: (A) gas; (B) xenon solution at
-85 C; (C) krypton solution at -155 C ..............................................
124
Low-frequency Raman spectra of ethylisocyanate: (A) liquid; (B)
amorphous solid; (C) annealed solid..................................................
125
Temperature dependence of the 983 cm-1 (cis) and the 1002 cm-1
(trans) infrared bands of ethylisocyanate dissolved in liquid
krypton ...............................................................................................
126
Potential energy curve of ethylisocyanate as a function of the
dihedral angle C=N C C2 obtained from MP2(full)/6-31G(d,p)
calculations (dashed line) and MP2(full)/6-311+G(2d,2p)
calculations (solid line) ......................................................................
127
9
10
13
14
15
16
17
18
19
20
xiv
21
Atomic numbering of isopropylisocyanate ........................................
158
22
Observed and predicted (MP2(ful1)/6-311+G(d,p)) infrared spectra
of isoproylisocyanate: (A) Gas; Simulated spectra of (B) mixture
of gauche and trans conformers with H = 115 cm-1, (C) pure
gauche conformer, and (D) pure trans conformer .............................
159
Infrared spectra of isoproylisocyanate: (A) Gas; (B) Xenon
solution at -100 C ..............................................................................
160
Observed and predicted (MP2(ful1)/6-311+G(d,p)) Raman spectra
of isoproylisocyanate: (A) Xenon solution at 100 C; Simulated
spectra of (B) mixture of gauche and trans conformers with H =
115 cm-1, (C) pure gauche conformer, and (D) pure trans
conformer ...........................................................................................
161
25
Low resolution microwave spectrum of isopropylisocyanate ...........
162
26
Temperature dependence of the 902 cm-1 (trans) and the 943 cm-1
(gauche) Raman bands of isopropylisocyanate dissolved in liquid
xenon ..................................................................................................
163
Potential energy curve of isoproylisocyanate as a function of the
dihedral angle C=N C H5 obtained from MP2(full)/6311G(2d,2p) calculations (dash line) and MP2(full)/aug-cc-PVDZ
calculations (solid line) ......................................................................
164
Raman spectra of dimethylisilylisocyanate: (A) gas; (B) liquid; (C)
solid; (D) simulated spectrum from scaled MP2/6-31G(d)
calculation ..........................................................................................
180
Infrared spectra dimethylisilylisocyanate: (A) gas; (B) amorphous;
(C) annealed solid; (D) simulated spectrum from scaled MP2/631G(d) calculation .............................................................................
181
Possible conformations of cyclopropylisocyanate, δ indicates the
dihedral angle C=N–C–H4 .................................................................
210
Infrared spectra of cyclopropylisocyanate: (A) gas; (B) xenon
solution at -100 C. The asterisk indicates the spectral region at
~600 cm-1 has nearly zero energy due to the absorption from the
silicon windows .................................................................................
211
23
24
27
28
29
30
31
xv
32
33
34
35
36
37
38
39
40
Observed (room temperature) and predicted (MP2(ful1)/6-31G(d)
at 25°C) Raman spectra of cyclopropylisocyanate: (A) liquid; (B)
predicted spectrum of the mixture of cis and trans conformers with
H = 77 cm-1; (C) predicted spectrum of the pure cis conformer;
(D) predicted spectrum of the pure trans conformer; (E) solid.
Asterisk indicates that the intensity has been reduced by half...........
212
Observed (room temperature) and predicted (MP2(ful1)/6-31G(d)
at 25°C) Raman spectra of cyclopropylisocyanate: (A) gas; (B)
predicted spectrum of the mixture of cis and trans conformers
with H = 77 cm-1; (C) predicted spectrum of the pure cis
conformer; and (D) predicted spectrum of the pure trans conformer
213
Observed and predicted (MP2(ful1)/6-31G(d)) infrared spectra of
cyclopropylisocyanate: (A) xenon solution at 100 C; (B)
predicted spectrum of the mixture of cis and trans conformers with
H = 77 cm-1; (C) predicted spectrum of the pure cis conformer;
and (D) predicted spectrum of the pure trans conformer ..................
214
van’t Hoff plot of the eight infrared conformer bands utilized in the
determination
of
enthalpy
difference
value
for
cyclopropylisocyanate dissolved in liquid xenon solution ................
215
Potential energy curve of cyclopropylisocyanate as a function of
the dihedral angle C=N C1 H4 obtained from MP2(full)/631G(d,p) calculations (dashed line) and MP2(full)/6-311G(2d,2p)
calculations (solid line) ......................................................................
216
Atomic numbering of cyclobutylisocyanate with the equatorial
form shown. The relative orientation of the –NCO moiety is
indicated by the τC4=N2 Cα−H3 angle: Trans (τ=180°); Cis (τ=0°);
Gauche (τ=73.5°) ...............................................................................
249
Infrared spectra of cyclobutylisocyanate: (A) Gas; (B) Xenon
solution at -100 C ..............................................................................
250
Observed and predicted (MP2(ful1)/6-311+G(d,p)) infrared spectra
of cyclobutylisocyanate: (A) Gas; (B) Predicted spectrum of the
mixture of Eqt-t and Eqt-g conformers with H = 131 cm-1; (C)
Predicted spectrum of the pure Eqt-g conformer; (D) Predicted
spectrum of the pure Eqt-t conformer ................................................
251
Infrared spectra (500-300 cm-1) of cyclobutylisocyanate: (A) Gas;
(B) Amorphous solid; (C) First annealed solid; (D) Second
annealed solid.....................................................................................
252
xvi
41
42
43
44
45
46
47
48
49
Infrared spectra (800-500 cm-1) of cyclobutylisocyanate: (A) Gas;
(B) Amorphous solid; (C) First annealed solid; (D) Second
annealed solid.....................................................................................
253
Observed and predicted (MP2(ful1)/6-311+G(d,p)) Raman spectra
of cyclobutylisocyanate: (A) Liquid; (B) Predicted spectrum of the
mixture of Eqt-t and Eqt-g conformers with H = 131 cm-1; (C)
Predicted spectrum of the pure Eqt-g conformer; (D) Predicted
spectrum of the pure Eqt-t conformer ................................................
254
Atomic numbering of cyclohexylisocyanate with the equatorial
form shown. The relative orientation of the –NCO moiety to the
alpha H is indicated by the τ(C=N Cα−H5) angle: Trans (τ=180°);
Cis (τ=0°) ...........................................................................................
292
Observed and predicted (MP2(ful1)/6-311+G(d,p)) infrared spectra
of cyclohexylisocyanate: (A) Gas; Simulated spectrum of (B)
mixture of Eqt-t and Axl-t conformers with H = 397 cm-1, (C)
pure Eqt-t conformer, and (D) pure Axl-t conformer .........................
293
Infrared spectra (950-400 cm-1) of cyclohexylisocyanate: (A)
Amorphous solid; (B) annealed solid. The asterisk indicates that
the band disappeared upon achieving crystalline solid ......................
294
Variable temperature infrared spectra of cyclohexylisocyanate
dissolved in liquid xenon at 1322 cm-1 (Axl-t) and the 839 cm-1
(Eqt-t) .................................................................................................
295
Observed and predicted (MP2(ful1)/6-311+G(d,p)) Raman spectra
of cyclohexylisocyanate: (A) Liquid; Simulated spectrum of (B)
mixture of Eqt-t and Axl-t conformers with H = 397 cm-1, (C)
pure Eqt-t conformer, and (D) pure Axl-t conformer .........................
296
Calculated potential function of equatorial cyclohexylisocyanate at the
MP2(full)/6-311+G(d,p) and B3LYP/6-311+G(d,p) level during the
internal rotation of the NCO group as defined by the dihedral angle
τ(C=N Cα−H5) ...................................................................................
297
Calculated potential function of axial cyclohexylisocyanate at the
MP2(full)/6-311+G(d,p) and B3LYP/6-311+G(d,p) level during the
internal rotation of the NCO group as defined by the dihedral angle
τ(C=N Cα−H5) ..........................................................................................
298
xvii
ACKNOWLEDGEMENTS
First and foremost, I would like to express my most heart-felt thanks to my research
advisor, Curators’ Professor James R. Durig, for the amazing support that he has provided
throughout my academic journey. Starting out as an undergraduate, Dr. Durig gave me the
opportunity to work in his lab, which then piqued my interest in the field of chemistry. If Dr.
Durig hadn’t given me this opportunity, it is likely that I would not have pursued a Doctorate
degree in Chemistry. Because of Dr. Durig’s support, encouragement, and professional
approach I have been able to accomplish more than I ever expected to. Dr. Durig’s tireless
dedication to the field of vibrational and rotational spectroscopy has enabled me, by
extension, to obtain vast amounts of knowledge in relatively short period of time.
I would also like to extend very special thanks to Dr. Gamil A. Guirgis of the College
of Charleston who has skillfully synthesized and isolated some of the target compounds
utilized in my research. I am also indebted to Dr. Todor K. Gounev who was instrumental in
obtaining the variable temperature study of the infrared spectra. I would also like to thank
Professors Charles J. Wurrey who has provided much needed guidance during my Ph.D. I am
very grateful to Dr. Peter Groner for providing accessibility to some of profound computer
programs necessary for my research. I am also honored and grateful that Drs. Nathan Oyler,
Zhonghua Peng, Jejung Lee and James Murowshick have served as my committee members.
Throughout the entirety of my time in the doctoral program, I have worked daily with Dr.
Savitha S. Panikar, who has proved invaluable as a colleague and friend.
I would also like to acknowledge the financial support of my education by the Gates
Millennium Scholars scholarship fund, Asian and Pacific Islander American Scholarship
Fund and the UMKC Women’s Council Graduate Assistant Funding. It would not have been
xviii
possible for me to complete my research without the generous contributions from these
donors.
Finally, I would like to end by thanking those who have provided their love and
support throughout this journey, including my family and my fiancé, Ken Kieffer.
xix
Dedicated to
My dear mother
Yan Ping Liu,
for her unconditional love and sacrifice
My five beautiful sisters
my beloved fiancé
Kenneth M. Kieffer, Jr.
for his genuine friendship and love
xx
CHAPTER 1
INTRODUCTION
One of the most classic examples utilized in the demonstration of conformational analysis
is ethane [1] (CH3-CH3). Given the sp3 hybridization about the carbon atoms, there is an
internal rotation about the C-C single bond; the resultant potential function is a threefold
barrier that alternates between the most stable staggered form and the lesser-favored eclipsed
conformation. This phenomena of hindered internal rotational has considerable importance
in both chemistry [2, 3] and biological applications [4]. Conformational studies in the early
1900s have essentially ignored this internal motion as a free or non-hindering rotation but it
was well established by the 1930s [5-10] that this internal rotation is not free but rather that it
is restricted by a potential energy barrier. Since these early investigations, structural chemists
have been fascinated with molecular structures that permit internal rotational motion at
various degree of complexity and many studies have been carried out on the effect of the
internal rotation on the conformational stability.
The potential energy function governing the ethane molecule is a relatively simple
one. However, for more complex molecules, the potential energy function needs to take into
account other factors that are present that influence the relative stability of the possible
energy minima present in the potential function. Especially for molecules with internal
rotation, these different forces are what make the shape of these potential unique. Some of
the main forces affecting the conformation stability are the influence of resonance effects of
double bonds [11, 12], the effect of delocalization of an electron, hydrogen bonding [13]
steric repulsion [14]. Some minor interactions are also expected from interatomic forces in
1
condensed phase such as van der Waal’s interactions, London dispersion forces or temporary
dipole-dipole interactions.
Conformational data that are of interest for studying internal rotation includes
identifying the stable conformation and its structure, energy difference between different
isomers, and the potential barrier to internal rotation. A large number of experimental
methods are available for conformational analysis and they include the classical
thermodynamic method (calorimeter, equilibration, pressure effect) [15, 16], relaxation
techniques [17] (NMR, ultrasound, and dielectric relaxation), diffraction techniques [18] (xray, neutron and electron diffraction) and spectroscopic techniques (NMR, microwave,
infrared and Raman). Much of the earlier studies were carried out by NMR in liquid or in
solution and by microwave where the structure of the molecule can be determined utilizing
the rotational constants. In recent decades, along with improved computational methods,
infrared and Raman has became the go to method for obtaining information on molecular
structure, especially for those molecules with some elements of symmetry [19, 20].
Diffraction techniques refer to X-ray diffraction, neutron diffraction and electron
diffraction and they are useful for obtaining structural information on the molecule. These
experimental methods rely on diffraction of energy rich particles by the sample. X-ray
crystallography has only limited use for conformational analysis but it is a rapid and accurate
method of determining the conformation in a crystal lattice. However, in the polycrystalline
solid phase, only one conformation is present and, thus, it is not possible to determine
conformation energy difference between different conformations. Another disadvantage with
X-ray Diffraction is strong intermoleculear association and the conformer identified is based
on crystal packing effect and may not be the most stable conformer. Neutron Diffraction is
2
similar to X-ray Diffraction and is best suited for detecting intramolecular and intermolecular
hydrogen bond that may exist in solid. One of the limitations is the requirement of having a
regular repeating unit in order for a diffraction pattern to be acquired. The more widely used
diffraction method is Electron Diffraction, which is a lot more reliable technique for
molecules that have stable conformations with relatively small enthalpy difference.
Determination of conformation stability by this technique is based on the deconvolution of
the radial distribution pattern that represents an average of all interatomic distances,
molecular structures and orientations. As a result, this method is not as a reliable for the
determination of hydrogen-containing bond distances, bonds with similar bond-length and
conformational energy difference. Electron Diffraction works very well for conformers that
are separated by a large barrier [18]. However, the interpretation of the result allows for too
many assumed bond force constants and the potential for misinterpretation is high.
Spectroscopic techniques (NMR, microwave, infrared and Raman) have the largest
set of restrictions on which molecules can be investigated but they also provide the most
reliable conformational data for molecules that satisfy the requirements for obtaining a
measureable spectrum. Nuclear Magnetic Resonance (NMR) was very popular in
determining the conformational barrier and a large number of experiments are carried in
liquid phase or in solution. It is widely used for determining the spatial molecular structure of
protein in a quantitative way. A lot of the earlier data reported on the internal dynamics of
organic molecules, inversion of configuration, valence tautomerism and proton exchange are
based on NMR. In order to obtain useful NMR data, molecules must have stable molecular
conformer that have higher-energy barrier (~40 kJ/mol) so that the lifetime of an individual
isomer is longer than the relaxation time. However, a large number of molecules can not
3
satisfy this requirement since the rotation along single bond lead to enthalpy difference in
magnitude as low a 0.4 kJ/mol.
NMR is only useful for molecules with very high
conformational interconversion barrier [21].
Microwave spectroscopy is one of more reliable techniques for conformational
analysis albeit not a very convenient one. Experimental results often produce reliable
structural information on individual energetic state due to precise frequency measurements
and high resolutions. It can also provide structural information on each conformer if more
than one conformer is present. However, this method is only practical in gas phase samples,
thus, only small molecules with relatively large vapor pressure can be studied [21]. In
addition, rotational spectra can only be observed for molecules with a non-zero permanent
electric dipole moment. Spectral interpretation can also be very consuming and tedious due
to weak intensity bands, complication due to observed transitions originating from different
rovibrational levels.
Large quadruple, inversion, pseudorotation and large-amplitude
vibrations can also add complexity to the spectra. Also, in order to obtain an accurate and
complete structure of a molecule, a large number of isotope substituted species are needed
which is in itself difficult to prepare and to separate afterward.
Vibrational spectroscopies consist of infrared and Raman techniques with both
operating in the infrared region of the electromagnetic spectrum in the study of vibrational
motions. They are one of the more practical and important tools for the determination of
molecular symmetry and conformational stability. They both measure the vibration energies
of molecules but each operates on a different set of selection rules and restriction. However,
infrared and Raman spectroscopy are actually complementary techniques that when
evaluated together, a more complete set of structural information can be obtained. There are
4
two types molecular vibrations, which are stretching and bending. The number of
fundamental vibrations a molecule has depends on the number of atoms it consists of. A
molecule with n number of atoms has a total of 3n degrees of freedom corresponding to the
Cartesian coordinate of each atom. For nonlinear molecule, 3n degree of freedoms are made
up of 3 rotational, 3 translational vibrations and rest are fundamental vibrations equating to
3n-6 (3n-5 for linear molecule).
One of the biggest advantage with vibrational spectroscopy is the fact that samples
can be studied under all states of aggregations. Consequently, it is possible to carry out
variable temperature study of the infrared spectra to obtain enthalpy difference (ΔH) between
the different conformers. Variable temperature IR and Raman spectroscopy with samples
dissolved in dilute liquid noble gas presents a reliable and accurate alternative method for
conformational enthalpy difference determinations. In cryogenic solvents such as liquefied
xenon, krypton, and argon, the vibrational frequencies of solute molecules are substantially
closer to the vapor phase values than those observed in the corresponding solid matrices.
Therefore, these solutions have been described as a pseudo gas phase [22]. An implication of
the pseudo gas phase hypothesis is that in such solutions the thermodynamics of a
conformational equilibrium should be similar to that in the vapor.
Since liquefied noble gases are mostly inert solvents, only small interactions are
expected to occur between the dissolved molecules and the surrounding noble gas atoms [2225], and as a result, the “pseudo gas phase” spectrum shows only small frequency shifts
compared with the spectrum of the gas. Additionally, variable temperature studies of
conformational stability carried out in noble gas solutions are free of absorption bands due to
solvent in both infrared and Raman spectra, Spectral bands recorded by this method are much
5
narrower and sharper due to the lack of rotational fine structure and low temperature of the
measurements. This is a significant advantage since the conformer bands are better resolved
in comparison with those in the spectrum of the gas and it is possible to identify and
deconvolute closely positioned bands arising from different conformers. Isomers with very
small conformational enthalpy differences can be measured very accurately in noble gas
solutions, often with very small statistical uncertainties. These benefits far outweigh some of
the major technical problems that are associated with this method.
Some of these
disadvantages are associated with the construction of the cell, the windows and the pressure
manifold. For larger and highly polar molecules, solubility also presents a problem. It is
especially hard to eliminate all traces of water even in high purity noble gas, particularly
xenon, bottles which may hydrolyze sensitive samples or form hydrogen bonded species with
certain samples. Regardless of these drawbacks, this method is widely utilized in the study of
conformational stabilities.
The research presented herein is a series of experimental and theoretical investigations
of the conformational stability, vibrational spectra, structural parameters and barriers to
internal rotation of some organoisocyanates. The intended purpose of these studies is to
further examine the conformational landscape of a molecule undergoing a hindered internal
rotation. Due to the cumulene feature of the –NCO group as well as the conformational
complexity resulting from the rotation around the single bond, organoisocyanates are ideal
for this purpose. Several acyclic and cyclic isocyanates have been studied and the research
on each molecule includes these following components:
1. Infrared and Raman spectra recorded in the gas, liquid and/or liquid phase.
Vibrational assignments are made for the most stable conformer.
6
2. Ab initio calculations with full electron correlation by the perturbation method to
the second order and hybrid density functional theory calculations by the B3LYP
method utilizing various basis sets starting from 6-31G(d) are carried out. The
ground state electronic energy, equilibrium structure, equilibrium rotational
constants, dipole moments, harmonic vibrational frequencies, infrared intensities,
Raman activities, Raman depolarization ratios, as well as force constants, dipole
derivative and polarizability derivative matrices of each individual stable
conformation are predicted. Normal coordinate analysis [26, 27] and potential
energy distributions (P.E.D.s) [28] are carried out.
3. Variable temperature studies of the IR/Raman are carried out with samples
dissolved in a noble gas solution to obtain the enthalpy difference value.
4. A set of “adjusted” ground state r0 structural parameters [29] has been determined
for each stable conformer by using an A&M program. The program systematically
fits MP2(full)/6-311+G(d,p) predicted equilibrium re structural parameters to all the
experimentally obtained A, B and C microwave rotational constants. As a result, the
geometrical parameters obtained in this manner are expected to be more accurate
than those from pure microwave / electron diffraction data.
5. Barrier to internal rotation and the potential functions governing the conformational
interchange are carried out experimentally based on torsional frequencies and
theoretically by ab initio calculations.
7
CHAPTER 2
THEORETICAL BACKGROUND
The internal motion of molecules consists of vibrations and rotations. To a good
approximation, these two different motions can be treated separately.
A vibration
corresponds to change in the distance between two nuclei and it can be mathematically based
on the simple harmonic oscillator. Rotation refers to the change in the spatial orientation of
“bonds” joining the nuclei and its mechanic can be understood based on the rigid rotor model
Vibrational Spectroscopy
At the core of vibration spectroscopy is the quantum mechanical treatment of the
one-dimension harmonic oscillator [30] which forms the fundamental basis for the
calculation of the vibrational energies of the molecules. Shrödinger wave function,
-
h 2 d 2y (x)
8p 2m dx 2
+V (x)y (x) = Ey (x)
when applied to the one-dimension harmonic system becomes the quadratic potential energy
as V (x) = 2p 2mv 2 x 2 :
d 2y 8p 2m
+ 2 [E - 2p 2mv 2 x 2 ]y = 0
2
dx
h
The resulting wave function for the 1-D system becomes
1
éæ ö 12
ù2 - x 2
a
1
yn (x) = êç ÷ n ú e 2 H n (x )
êè p ø 2 n!ú
ë
û
:a=
:x
4p 2mv
h
1
=a2x
n x2
: H n (x ) = (-1) e
8
d n e -x
dx n
2
The H n (x ) term is called the Hermite polynomials. Therefore the energy that corresponds to
the permitted nth vibrational level can be determined from:
1
En = (n + )hv , n = 0, 1, 2, 3…
2
The transition selection rule is Dv = ±1 with the v denoting the vibrational frequency.
However, the vibrational motions of real molecules are anharmonic especially at
higher vibrational levels and subsequently the potential energy deviates substantially from
that of an harmonic oscillator. As n increases the nuclei spend more time in regions far from
their equilibrium separation and at that point the vibrational energy is large enough to
dissociate molecule into unbounded atoms. A more accurate representation for the molecular
vibrational energy that allows for anharmonicity is the Morse [31] potential function:
: V (r) = De (1- e-a
EMorse (r) = hcV (r)
( r-re )
)2
m (2p vm )2
a=
:
2hcDe
De is the equilibrium dissociation energy and α is a measure of the curvature at the bottom of
the well. The resulting energy eigenvalues of the system gives reasonably accurate values for
the different energy level for most molecules:
1
1
1
Ev,J = (v + )hve - (v + )2 c e hve + J (J +1)hBe + J 2 (J +1)2 hDe - (v + )J (J +1)hae
2
2
2
1
a æ 2De ö 2
ve =
ç
÷
2p è m ø
Vibrational energy of a harmonic oscillator
9
h
Be =
8p 2 I e
Anharmonicity correction to vibrational energy with coefficient χe.
ce =
hve
4De
Energy level of a rigid rotor
De = -
h3
128p 6m 3ve2 re6
The correction to the non-rigidity with De centrifugal distortion constant.
æ 1
1 ö÷
ç
ae = 16p 2m re2 De çè a re a 2 re2 ÷ø
3h 2 ve
Vibration-rotation interaction with coupling constant αe.
As result of anharmonicity, ΔE gets smaller, and the Δν=±1 selection rule can be broken
leading to overtone bands (Δν=±2,3..) at higher frequencies
and combination bands.
Although individual frequency might change, usually they are not as intense as the
fundamental vibration.
One of the complications encountered in vibrational spectroscopy is the phenomenum
called Fermi resonance [32] which results as a consequence of quantum mechanical mixing.
Fermi resonance can be observed as the shifting of the energy and intensity of absorption
band in an infrared or Raman spectrum. Fermin resonance only occurs when two vibrations
have the same symmetries and the two corresponding wave functions are very similar in
10
energy. When this happens, the high energy mode shifts to higher energy and the low energy
mode shifts even lower and the more intense band becomes weaker and vice versa.
Infrared Spectroscopy
Infrared spectroscopy is based on the interaction of the infrared region of the
electromagnetic radiation with matter and IR active bands can only be observed if there is a
dipole change in the molecular system. The absorption of infrared light induces transitions
between vibrational energy levels. Quantum mechanical treatment shows [30] that infrared band
intensity can be expressed as
ò k (v)dv =
where
Np
(m x2 + m 2y + m z2 )
å
3c
ò k (v) is the absorption coefficient and N is the total number of molecules per unit
volume. The selection rule for infrared spectroscopy is determined by the integral:
(m )vv '= ò yv* (Q)dt
The intensity of an infrared absorption band is proportional to the square of the change in the
molecular electric dipole moment caused by a normal coordinate, Q.
Raman Scattering
Raman spectroscopy is an inelastic scattering phenomenon as the Raman effect
(Raman scattering). This effect occurs when monochromatic light is scattered by a molecule,
and the scattered light has been weakly modulated by the characteristic frequencies of the
molecules. Raman spectroscopy measures the difference between the wavelength of the
incident radiation and the scatter radiation. The incident radiation excites “virtual states” that
persist for the short timescale of the scattering process. Polarization changes are necessary to
11
form the virtual states. Raman selection rules are based on the polarizability of the molecules.
The selection rule for Raman spectroscopy is determined by the intergral:
(a )vv'= ò yv* (Q)ayv ' (Q)dt
The intensity of Raman scattering is proportional to the square of the change in the molecular
polarizability α resulting from a normal mode. Polarizability is expressed as tensor for an
anisotropic molecule [32]:
m X = a XX e X + a XY eY + a XZeZ
mY = aYX e X + aYY eY + aYZeZ
mZ = a ZX e X + aZY eY + a ZZeZ
where μ are components of induced electric dipole moments, ε terms are electric field strength.
Raman depolarization ratio ( r ) is defined as the intensity ratio between the
perpendicular component and the parallel component of the Raman scattered light. When plane
polarized incident light is utilized, the observation is made in a direction perpendicular to the
electric vector, the depolarization ratio (0 to 0.75) is determined by
r^ =
IT (obs. ^) - IT (obs. ||)
3b 2
=
I|| (obs. ^)
45a 2 + 4b 2
If the indicent light is unpolarized, the ratio (0 to 6/7) is determined by:
rn =
IT (obs. ^) - IT (obs. ||) + 12 IT (obs. ||)
I|| (obs. ^) + 12 IT (obs. ||)
=
6b 2
45a 2 + 7b 2
The depolarization value depends on the symmetry of the molecule and the normal vibration
mode. Totally symmetric mode (polarized band) has r < 0.75 and the other modes (depolarized
band) have r = 0.75.
12
Rotational Spectroscopy
The rotation of molecules can be treated quantum mechanically by assuming that the
molecule acts as a rigid rotor and can thus be treated as a particle moving in a spherical
surface. The Shrödinger equation in the spherical polar coordinate system with moment of
inertia I is defined as :
h2 é 1 ¶
¶
1 ¶2y ù
ê
(sin q ) + 2 2 ú + Ey = 0
¶q sin ¶f úû
8p 2 I êë sin q ¶q
The rotational energy eigenvalues are
EJ =
h2
8p 2 I
J (J +1) with J = 0, 1, 2, 3….
The transition selection rule is ΔJ = ±J. The rotation transiton frequency corresponding to
the J +1¬ J transition is 2B(J+1), where the rotational constant B =
h
8p 2 I
.
Structural Parameter
A very extensive study [33] has been shown that ab initio MP2(full)/6-311+G(d,p)
calculations can predict the r0 structural parameters for more than fifty carbon-hydrogen
distances for substituted hydrocarbons to at least 0.002 Å accuracy compared to the
experimentally determined [34] values from “isolated” CH stretching frequencies which were
compared to previously determined values from microwave studies. It has been determined
[29] that good structural parameters can be determined by adjusting the structural parameters
obtained from the ab initio MP2/6311+G(d,p) calculations to fit the rotational constants
obtained from microwave experimental data using a computer program A&M (Ab initio and
Microwave) [29] developed in our laboratory. This program combines the information from
13
microwave and ab initio calculations and gives structural parameters that fit the rotational
constants with the structural parameters remaining close to the ab initio values.
In order to reduce the number of independent variables, the structural parameters are
separated into several sets according to their type. For example, three CH bond lengths may
form one set and three CCH angles may form another set. Each set uses only one
independent parameter in the optimization, and all structural parameters in one set is adjusted
by the same parameter, i.e., an adjustment factor. The information on the differences between
the similar parameters from ab initio calculations can be retained in the final results by the
following method: bond lengths in the same set keep their relative ratio and bond angles and
torsional angles in the same set keep their differences in degrees. If one CH bond is 1%
longer than another CH bond by ab initio calculation, it is still 1% longer after the
optimization. If a CCH angle is 1° larger than another CCH angle from the ab initio
calculation, it is still 1° larger in the final result. This assumption is based on the fact that the
errors from ab initio calculations are systematic, which is now commonly accepted, and this
technique is currently used in our normal coordinate analysis. The program searches the
minima of the function F(k1,k2,...):
F(k1,k2…) = å (100Ki)2 +
i
å (20Kj)2 + å (0.1Kl)2 + å (0.02Km)2
j
l
The adjustment factors K in the formula above are defined as
Ki =
Kj =
Rci - Rai
Rai
Lci - Laj
Laj
Kl = Acl - Aal
14
m
Km = Tcm -Tam
where R, L, A, and T represent the rotational constants, bond lengths, bond angles, and
torsional angles, respectively. The lower case letters c, o, and a indicate calculated
(calculated here means calculated by the AandM program, not by the ab initio program),
observed (rotational constants), and ab initio (bond lengths) bond angles and torsional angles
in degrees. Thus, the symbols Tc and Ta indicate the torsional angles calculated from the
program and the torsional angles obtained from the ab initio predictions, respectively. The
subscript runs over all structural parameters in the optimization. The real meaning of these
formulas is the following: 1% error of any rotational constant contributes 1.0 to the function
F, which is equivalent to a 5% shift of a bond length from its ab initio value, a 10° shift of
bond angle from its ab initio value, or a 50° shift of a torsional angle from its ab initio value.
Only real torsional angles around single bonds are considered as torsional angles in the
calculations, whereas other angles defined by the third internal coordinate in the ab initio
input data are treated as bond angles because they are not as flexible as real torsion angles
around single bonds. This program adds ab initio structural parameters, although with much
smaller weights, to the rotational constants so that the number of observables is always larger
than the number of unknown parameters. As a result, unique results can be obtained without
any arbitrary assumptions for the structural parameters values [29].
Theoretical Calculations
The determination of conformational stability is also carried out by ab initio quantum
mechanical modeling methods such as ab initio and density functional theory (DFT)
calculations by the Gaussian 03 program packages. Several conformations might correspond
15
to energy minima but the main goal is to determine the global minimum structure on the
potential energy surface determined by geometry optimization. Compared with molecularmechanic calculations and semi-empirical, ab initio and DFT can provide higher quality
quantitative predictions for a broad range of systems. Currently, DFT and MP2 are the only
two Hartree-Fock methods capable of calculating analytic frequencies and Raman
polarization. Therefore, theoretical calucations are carried out with these two methods in this
dissertation research utilizing a variety of basis set from 6-31G(d) and larger basis sets.
Information on the relative conformational energies, structural parameters, force fields,
infrared intensities, Raman activities and depolarization ratios are calculated without any
experimental data.
All of these calculations are based on the molecular orbital theory [35]. The central
starting point for molecular quantum chemistry method is the Hartree-Fock (HF) equation
[36]
F̂(1)fi (1) = eifi (1)
where εi is the orbital energy and F̂ is the Hartree-Fock operator.
F̂(1) = Ĥ
core
n/2
(1) + å éë2 Ĵ j (1) - K̂ j (1)ùû
j=1
The first term Ĥ on the HF operator represents the kinetic energy of one electron.
Z
1
Ĥ core (1) = - Ñ12 - å a The first term
2
r
a 1a
16
The second term, Ĵ j , represents the Coulomb operator for the potential energy of the
2
interaction between electron 1 and a smeared-out electron with electronic density f j (2) .
Ĵ j (1) f (1) = f (1) ò f j (2)
2
1
du
r12 2
The third term, K̂ j , is an exchange operator which is defined as:
K̂ j (1) f (1) = f j (1) ò
f *j (2) f (2)
r12
du 2
The Hartree-Fock equation can be solved by an iterative process since the operator F̂
depends on its own eigenfunctions.
The MOs fi can be represented almost exactly if the spatial orbital, fi , is expanded
as a linear combination of a set of infinite number of one-electron basis function c s as
proposed by Roothaan [37]:
b
fi = å csi c s ; b= 1, 2, 3…
s=1
By substituting the linear expression into the HF equation, the HF becomes the Roothaan-HF
(RHF) equation:
åcsi F̂ c s = ei åcsi c s
s
s
The Roothaan-HF equation is most efficiently solved by using a matrix method. By
*
multiplying The Roothaan-HF equation with c r and integrating, the Roothaan-HF becomes:
17
b
å csi (Frs - ei Srs ) = 0 ; r =1, 2, 3…b
s=1
Frs º c r F̂ c s and Srs º c r c s
The intergo-differential equation can become a matrix equation:
b
b
s=1
s=1
å Frscsi = å Srscsiei : r = 1, 2…b
The matrix form of Roothaan-HF equation can be written as FC = SCe .
The Hartree–Fock method is typically used to solve the time-independent
Schrödinger equation for a many-body molecule as described in the Born-Oppheimer
approximation. Since solution for multi-electron molecule cannot be solved exactly, the
problem is solved numerically. Due to the nonlinearity introduced by the Hartree–Fock
approximation, the equations are solved using a nonlinear method such as iteration, which
gives rise to the "self-consistent field method." One of the simplifications that the HartreeFock method makes the exclusion of the electron correlation which can lead to large
deviation from experimental results.
All approaches developed to compensate for this
inadequateness are known as post-Hartree-Fock method.
One of these methods is the
implementation of the Moller-Plesset (MP) perturbation theory [38] which improves the HF
method by adding electron correlation as a perturbation of the Fock operator in which the
unperturbed wave function is the HF function. The MP method is an application of the
Rayleight-Schrödinger perturbation theory, in which the Hamiltonian is divided into two
parts:
Ĥ = Ĥ0 + lVˆ
18
The equation defines that a small perturbation, Vˆ , is added to the unperturbed Hamiltonian
operator, Ĥ 0 . The perturbed wave function and the energy can be solved as a power series in
terms of λ [39]:
n
Y = lim å l Y
i
(i)
n
and
n®¥ i=o
E = lim å l i E (i)
n®¥ i=o
The perturbed wavefunction and energy are then substituted into the time-independent
Schrödinger equation:
æn
ö æn
öæ n
ö
( Ĥ0 + lVˆ ) ç å l i Y (i) ÷ = ç å l i E (i) ÷ç å l i Y (i) ÷
ç
÷ ç
֍
÷
è i=0
ø è i=0
øè i=0
ø
By equating the coefficient of λk, k= 1, 2, 3..n, this equation leads to a successively higher kth
order of perturbation.
In the Møller-Plesset perturbation theory [38], H0 is defined as the sum of the one-electron
Fock operator:
H0 = å F i
i
Since H0 is the sum of the Fock operator, E(0) is the sum of the orbital energy:
E (0) = Y (0) H0 Y (0) = åei
i
The first-order energy
E (1) = Y (0) V Y (0)
The Hartree-Fock energy is the sum of the E(0) and E(1). The first order wave function Ψ(1)
can be expressed as:
19
Y
(1)
æ Y V Y (0)
ç t
= åç
(0)
- Et
t ç E
è
ö
÷
÷÷
ø
where Ψt is an arbitrary substituted function. The second-order energy E(2) is expressed as:
E (2) = å
Y (0) V Y t
t
2
Et - E (0)
The first perturbation to the Hartree-Fock energy is the value of E(2), which is always
negative abut not variational [39].
An alternative to the Hartree-Fock method is the density functional theory (DFT), in
which the molecular electron energy is obtained from the calculated molecular electron
probability density (ρ) rather than the molecular wave function. One of the advantages with
DFT is the inherent inclusion of the electron correlation effects in the calculation via general
functionals. DFT is based on the Hohenberg-Kohn theorem [40], which demonstrated the
existence of unique functions which determined the ground state energy and density exactly.
Following the work of Kohn and Sham [41], the approximate functions employed by
currently DFT methods partition the electronic energy into several terms which are computed
separately [39]. These different components are the kinetic energy (ET), the potential energy
associated with electron-nuclear interaction (EV), the Coulomb repulsion (EJ), and a
exchange term (EX) and the correlation (EC) term.
E = ET + EV + E J + E X + EC
The EX and EC terms are usually approximated [39] as integrals involving only the electron
density, ρ, and their gradients, r . The local exchange function is almost always defined as:
20
1
3 æ 3 ö3
X
E LDA
=- ç ÷
2 è 4p ø
4
ò r 3 d 3r
The major advance in DFT was the introduction of gradient-corrected exchange functionals.
One popular function is known as BLYP, named in homage to Becke’s contribution to the
exchange functionals, and Lee-Yang-Parr’s functional on correlation. Becke’s exchange
functionals [42] gave more accurate dissociation energy:
4
X
X
EBecke88
= E LDA
-g ò
4
r 3 ( r 3 Ñr )2
4
3
1+ 6g sinh -1 ( r Ñr )
d 3r ; γ=0.0042 Hartrees
γ is the parameter chosen to fit the known exchange energies of the inert gas atoms.
The correlation functional developed by Lee, Yang, Parr for a closed-shell system is defined
as:
C
E LYP
1ü
ì
5
ï
-1 é
1
1 2 ù -cr - 3 ï
3
3
= -a ò
í r + br êCF r - 2tW + ( tW + Ñ r )ú e
ý dr
-1
9
18
ë
û
ï
ïþ
3
1+ d r î
1
a = 0.04918
b = 0.132
c = 0.2533
d = 0.349
CF =
2
3
(3p 2 ) 3
10
2
1 Ñr (r)
1
tW (r) =
- Ñ2 r
8 r (r)
8
21
Even more widely used than the BLYP functional is the B3LYP, which is a hybrid
functional theory that combines density functional theory with the exact energy from
Hartree-Fock Theory. The B3LYP functional [39] defines EXC as:
XC
X
XC
Ehybrid
= cHF EHF
+ cDFT EDFT
where c’s are constants.
To predict the energy differences among the most likely conformers, LCAO-MO-SCF
calculations were performed with the Gaussian-03 program [43] by using Gaussian-type
basis functions. The energy minima with respect to nuclear coordinates were obtained by
simultaneous relaxation of all geometric parameters consistent with symmetry restrictions
using the gradient method of Pulay [44]. A number of basis sets starting from 6-31G(d), and
increasing to 6-311G(3df,3pd), were employed at the level of Møller-Plesset perturbation
theory [38] to the second order (MP2), as well as hybrid density functional theory by the
B3LYP method. At all levels of calculations carried out, with and without diffuse functions,
predicted conformational stabilities varied extensively with the size of the basis set.
To aid in making the vibrational assignment a normal coordinate analysis has been
carried out by utilizing the force field obtained from the Gaussian-03 program at the
MP2(full)/6-31G(d) level. The internal coordinates used to calculate the G and B matrices for
cyclohexylisocyanate are listed along with the structural parameters in Table 6. By using the
B matrix, the force field in Cartesian coordinates was converted to a force field in internal
coordinates [45]. Subsequently, scaling factors of 0.88 for the CH stretches and 0.90 for all
other modes except for the coordinates concerning the heavy atom C -NCO angles, along
with the geometric average of scaling factors for interaction force constants, to obtain the
22
fixed scaled force fields and the resultant wavenumbers. A set of symmetry coordinates was
used to determine the corresponding potential energy distributions (P.E.D.s).
In order to identify the fundamental vibrations, simulation of the the infrared and
Raman spectra have been made from the scaled ab initio MP2(full)/6-31G(d) results.
Infrared intensities were calculated based on the dipole moment derivatives with respect to
Cartesian coordinates. The derivatives were transformed into normal coordinate derivatives
by
é ¶m ù
é ¶m ù
ê u ú = åê u úLij
êë ¶Q j úû j êë ¶X j úû
where Qi is the ith normal coordinate, Xj is the jth Cartesian displacement coordinate, and Lij
is the transformation matrix between the Cartesian displacement coordinates and the normal
coordinates. The infrared intensities were then calculated by
Np
Ij = 2
3c
2
éé
ù
êê ¶m x ú +
êë ¶Q û
i
ë
2
é ¶m ù é ¶m ù2 ù
y
ê
ú +ê z ú ú
êë ¶Qi úû ë ¶Qi û ú
û
The evaluation of Raman activity by using the analytical gradient method has been
developed [46, 47]. The activity Sj can be expressed as:
Sj = gj (45
2
j
+7
2
j
where gj is the degeneracy of the vibrational mode j,
polarizability, and
sections,
j/
j
),
j
is the derivative of the isotropic
is that of the anisotropic polarizability. The Raman scattering cross
, which are proportional to Raman activities, can be calculated from the
scattering activities as well as the predicted wavenumbers for each normal mode, by using
the following relationship [48, 49]:
23
¶s j
¶W
where
0
=
(2p) 4
45
(n0 - n j ) 4
h
S
é -hcn ù 8p 2cn j
j
j
ú
1- exp ê
êë kT úû
is the excitation wavenumber,
j
is the vibrational wavenumber of the jth
normal mode, and Sj is the corresponding Raman scattering activity. To obtain the polarized
Raman scattering cross sections, the polarizabilities are incorporated into Sj by multiplying Sj
by (1
j)/(1+ j),
where
j
is the depolarization ratio of the jth normal mode. The Raman
scattering cross sections and calculated wavenumbers obtained from the scaled ab initio force
fields were used together with a Lorentzian function to obtain the simulated Raman spectra.
Enthalpy Difference Determination
The energy difference (ΔE) between two conformers is very difficult to obtain,
however, it is feasible to proximate that value with enthalpy difference. The ΔH value can be
experimentally [50, 51] determined based on the thermodynamic relationship between the
temperature and the Gibbs energy change of the conformational equilibrium:
DG = -RT ln K where equilibrium constant K =
I *High Energy Conformer
I Low Energy Conformer
By substituting ΔH-TΔS in for Gibbs energy (ΔG) and with some rearrangement of the
equation, it follows that the equation takes the form of van’t Hoff isochore:
ln K = -
DH DS
+
RT R
For molecules that have more than one conformer in a mixture, the infrared absorption and
Raman intensity is directly proportional to the population of the conformer in the sample.
Then the relationship follows the Beer-Lambert Law that the intensity/absorbance (I) for
conformer A and B are based on its corresponding concentration (C):
24
I A = C Ae A Lg A and I B = CBe B Lg B
where ε is the molar absorptivity, L is the path length of the sample cell and g is the
degeneracy of the conformer. By substitution IA and IB into the equilibrium constant equation,
K can express as
K=
I A eB gB
I B e Ag A
and it follows that:
ln K = ln
IA
e
g
DH DS
- ln A - ln A = +
IB
eB
gB
RT R
which can be re-arranged to be
ln
IA
e
g
DH DS
=+
+ ln A + ln A
IB
RT R
eB
gB
where absorptivity IA and IB can be experimentally measured as the integrated area
underneath the bands and the corresponding absolute temperature, T. By using a least
squares fit, –ΔH/R can be determined from the slope of the function ln(IA/IB) versus 1/T. The
functions ΔG, ΔS and K actually cannot be directly evaluated by this method since the
extinction coefficients of the proper bands are not known. It was assumed that S and the
ratio of the molar absorption coefficients εA/εB are not a function of temperature in the range
studied. Regardless, the ΔG will be equal to ΔH and the relative abundance can be
determined [52] if the ΔS between the conformers is neglected.
Determination the enthalpy difference by this method requires that the bands of
conformers A and B both have reasonably high intensities, are situated on a relatively even
baseline and do not overlap other bands in the spectra. Particularly important is the
requirement that the bands employed in the van’t Hoff plots must be "pure", meaning that
25
their intensities must be due to one conformer only, with no contribution from the other
conformer(s). With the large number of combination bands or overtones possible, some of
which might be enhanced by Fermi resonance, it is very difficult to decide if the bands are
pure. If a band completely vanishes in the crystal it is probably a pure band of the highenergy conformer. The bands remaining in the crystal, however, can easily be overlap bands
of both conformers. Thus, it is necessary that more than one pair of bands should be
employed and independent calculations of ∆H should be made for each band pair [52].
26
CHAPTER 3
EXPERIMENTAL
In order to obtain a comprehensive set of spectroscopic data, various spectroscopic
techniques have been employed. Although not all these technique were applied to every
molecules mentioned herein, all the mentioned techniques were vital in forming a definite
conclusion. These experimental techniques included recording the mid- and far- infrared
spectra in the gas and solid phase and as well was the sample dissolved in rare gas solution of
either xenon or krypton. In regards to the Raman technique, it was mostly carried out in the
liquid but for isopropylisocyanate, a variable temperature study was carried out with the
sample dissolved in the xenon solution.
Microwave spectroscopy is also immensely
important in obtaining good structural parameters.
The mid-infrared spectra (3500 to 300 cm-1) of the gas and solid were recorded from on
a Perkin-Elmer model 2000 Fourier transform spectrometer equipped with a nichrome wire
source, Ge/CsI beamsplitter and DTGS detector. Atmospheric water vapor was removed from
the spectrometer housing by purging with dry nitrogen. The spectrum of the gas was obtained
with the samples contained in 12 cm cells equipped with CsI windows. Interferograms
obtained after 128 scans for the gas sample and the background reference were transformed
by using a boxcar apodization function with theoretical resolutions of 0.5 cm -1 for the
gaseous sample. For the spectrum of the solid, the spectera were recoreded by depositing a
solid sample film ontot a CsI substrate cooled by boling liquid nitrogen and housed in a
vacuum cell fitted with CsI windows. The sample was annealed multiple until no further
changes were observed for the spectrum of the solid. A theoretical resolution of 2 cm-1 was
used with 128 interferograms added and truncated for the reference and resulting spectrum.
27
The far infrared spectra (710 to 30 cm-1) of gases and annealed solids were obtained
with the Perkin-Elmer model 2000 Fourier transform spectrometer equipped with a far
infrared grid beam splitter and DTGS detector. The spectra of the solids were obtained by
condensing the sample on to a silicon plate held in a cell equipped with polyethylene
windows and cooled with boiling liquid nitrogen at an effective resolution of 1.0 cm-1. The
samples were annealed until no further change was observed in the spectrum. The spectra of
gases were obtained from the sample contained in a 10 cm path 39 cell equipped with
polyethylene windows with an effective resolution of 0.5 cm-1. Typically, 256 scans were
used for both the sample and reference to give a satisfactory signal-to-noise ratio.
The far infrared spectrum of the amorphous and the annealed solid from 650 to 50 cm -1
were recorded with a Perkin-Elmer model 2000 Fourier transform interferometer equipped
with a far infrared grid beamsplitter and a DTGS detector. The spectra were obtained by
condensing the sample onto a silicon plate held in a cell equipped with polyethylene
windows and cooled with boiling liquid nitrogen. The sample was annealed until no further
changes were observed in the spectrum.
For the determination enthalpy difference, variable temperature study of the midinfrared spectra are carried out with the sample dissolved in liquefied rare gas solution of
either xenon ( 100 to 55 C) or krypton ( 155 to 100 C). The spectra were recorded on
a Bruker model IFS-66 Fourier interferometer equipped with a Globar source, Ge/KBr
beamsplitter and DTGS detector. The interferograms were recorded at variable temperatures
ranging from
155 to
100 C with 100 scans and transformed by a Blackman-Harris
apodization function with a theoretical resolution of 1.0 cm-1. The temperature studies were
carried out in a specially designed cryostat cell, which is composed of a copper cell with a 4
28
cm path length and wedged silicon windows sealed to the cell with indium gaskets. The
temperature was monitored by two platinum thermoresistors and the cell was cooled by the
vapors from boiling liquid nitrogen.
The Raman spectra of the liquid were recorded on a Spex model 1403
spectrophotometer equipped with a Spectra-Physics model 2017 argon ion laser operating on
the 514.5 nm line. The laser power used was 0.5 W with a spectral bandpass of 3 cm -1. The
spectrum of the liquid was recorded with the sample sealed in a Pyrex glass capillary.
The microwave spectra of cyclohexylisocyanate were recorded with a “mini-cavity”
Fourier-transform microwave spectrometer [53, 54] at Kent State University. The FabryPerot resonant cavity was established by two 7.5-inch diameter diamond-tip finished
aluminum mirrors with a 30.5-cm spherical radius. The sample was entrained in 70:30 NeHe carrier gas mixture at 2 atm and expanded into the cavity with a reservoir nozzle [54]
made from a modified Series-9 General Valve. The sample was irradiated by microwave
radiation generated by an Agilent Technologies E8247C PSG CW synthesizer; details of the
irradiation and its heterodyne detection circuitry can be found in Ref. [53].
Labview
software controls the timing of the gas and irradiation pulses, as well as the detection of any
free induction decay signal. The software performs signal averaging and can scan the
spectrometer by stepping both the frequency source and the cavity. Microwave circuit
elements allow for a spectral range from 10.5 to 26 GHz. The digital frequency resolution,
governed by the sampling rate and the length of the free induction decay record, is 2.5 kHz.
Rotational transitions are split into Doppler doublets centered at the transition frequency due
to the coaxial orientation of the gas expansion to the cavity axis and the FWHM of each
Doppler component is typically 13 kHz. The vacuum system can accommodate pulse
29
repetition rates of up to 15 s-1 while maintaining a pressure below 10-4 torr, and the
instrument can scan 450 MHz in 6 hours while averaging 100 shots per scan segment.
30
CHAPTER 4
VIBRATIONAL SPECTRA AND STRUCTURAL PARAMETERS OF SOME XNCO
AND XOCN (X=H, F, Cl, Br) MOLECULES
Introduction
Recently we initiated some spectroscopy studies supported by ab initio and density
functional theory calculations of some YN3 [55, 56] and YNCS [57-59, 60] molecules where
Y was an organic, silyl, germyl or halogen moiety. Since the YNC angle is relatively large
or in some cases linear, the NCS moiety has nearly free or free rotation which significantly
effects the vibrational and rotational spectra. For example, the barrier to internal rotation for
methylisothiocyanate (CH3NCS) is ~3 cm-1 which results in essentially free rotation of the
methyl group with the degeneracy of the two NCS bends [59]. In the infrared spectrum of
the gas phase, two of the antisymmetric (pseudodegenerate) vibrations of the CH3 groups
have resolvable fine structure where the spacing is 9.8 cm-1 for the stretch and 13.8 cm-1 for
the deformation where the spacing is determined by the zeta values for these normal modes
for the CH3NCS molecule [59]. Also the two NCS bends give a very strong broad infrared
band in the gas and liquid along with a much weaker lower frequency band. In the infrared
spectrum of the solid the very broad band essentially disappears leaving the single lower
frequency band which may indicate a linear CNCS. This very low C-NCS torsional barrier
for the methyl compound is also found for the corresponding ethyl compound [60] which
results in a single stable cis conformer (CH3 group cis to the NCS moiety) for this molecule
which is at variance from the predictions form ab initio calculations up to TZVP [61].
Because of these large amplitude vibrations for these type of molecules, it has frequently
31
been difficult to assign their microwave spectra from which rotational constants can be
obtained for determining the r0 or rs structural parameters. However, it is frequently possible
to combine a limited number of experimentally determined rotational constants with ab initio
predicted structural parameters to obtain r0 structural parameters that have significantly
smaller uncertainties than those obtain from the microwave spectral data [29]. Such results
have been obtained for HN3 [55], CH3N3 [56], HNCS [57] and GeH3NCS [59].
Since the
XNC (X = H, CH3) for the isocyanate is significantly smaller [62, 63]
than the corresponding angles for the XNCS [57, 59] molecules, it is expected that the barrier
to linearity will be much larger for the corresponding isocyanates. Also the N=C distances
may be significantly different between isocyanate and the corresponding isothiocyanate
molecules. Therefore, as a continuation of these spectroscopic and theoretical investigations
we have carried out similar studies of XNCO and XOCN where X = H, F, Cl, and Br for
comparison to the corresponding isothiocyanate (NCS) molecules. The results of these
theoretical studies along with comparisons to the experimental data when appropriate are
reported herein.
Theoretical Calculations
In order to provide vibrational frequencies with both infrared and Raman intensities and
optimized geometries, ab initio calculations were carried out by using the Gaussian-03
program [43] at both the restricted Hartree-Fock (RHF) level and by the perturbation method
to second order (MP2) [44] with full electron correlation. Three basis sets, 6-31G(d), 6311+G(d) and 6-311+G(2d) have been utilized. Hybrid density functional theory (DFT)
calculations have also been carried out by the B3LYP method utilizing the 6-311+G(d) basis
set. Frequencies for the fundamentals have been predicted for the XNCO (X = H, F, Cl, Br)
32
molecules, along with the predicted infrared and Raman activities and these data are listed in
Table 1. Similar data have also been obtained for the corresponding XOCN (X = H, F, Cl,
Br) molecules which are listed in Table 2. The predicted values are compared to the
experimental values when they are available.
In order to obtain a complete description of the molecular motion involved in the
normal modes, the force field in Cartesian coordinates was calculated with the 6-31G(d) and
6-311+G(d) basis sets at the MP2 level as well as with 6-311+G(d) basis set from the hybrid
DFT calculations by B3LYP method. The internal coordinates were the X N, N=C, and
C=O distances, the XNC and NCO angle bends, and the out-of-plane angle bend for the
isocyanates and X O, O C and C N distances, the XOC and OCN bends, and the out-ofplane bend for the cyanates. The symmetry coordinates were these internal coordinates
individually except we also combined the C=O and N=C coordinates for antisymmetric and
symmetric stretches for the isocyanates. The B matrix was used to convert the ab initio force
field in Cartesian coordinates to a force field in internal coordinates [45]. The frequencies
from the MP2/6-31G(d) calculation were also calculated by utilizing a set of scaling factors
of 0.88 for the N H (N D) stretches, 0.9 for HNC (DNC) bends, and 1.0 for all other
coordinates with the geometric average for the off-diagonal terms for HNCO. Except for the
acid the potential energy distributions (P.E.D.s) are expressed in terms of the symmetry
coordinates where S1 is NCO antisymmetric stretch, S2 is the NCO symmetric stretch, S3 is
the X N stretch, S4 is the NCO bend, S5 is the XNC bend and S6 is out-of-plane bend; these
potential energy distributions are listed in Table 1. For the acid, S1 is N H (N D) stretch, S2
is NCO antisymmetric stretch, S3 is the NCO symmetric stretch, S4 is HNC (DNC) bend, S5
is NCO bend and S6 is the out-of-plane bend. The pure ab initio frequencies, infrared
33
intensities, Raman scattering activities, along with the B3LYP/6-311+G(d) calculation results
are also given in Table 1.
In order to show the differences in the predicted and observed spectra for the fluoro-,
chloro- and bromoisocyanates as well as the difference in the infrared and Raman spectra for
the hydrogen and halocyanates, we calculated the theoretical infrared and Raman spectra.
The calculated frequencies, infrared intensities, and Raman scattering activities were
obtained from both the ab initio and hybrid DFT calculations.
In Figs. 1 3, the predicted infrared spectra from the MP2(full)/6-31G(d) calculations
are shown for each XNCO (X = F, Cl, Br) molecule. For comparison the experimental
infrared spectra of the gas or in a matrix are also shown. The observed ones show some
slight differences from the predicted ones for these molecules. For the other molecules
(HNCO and DNCO Fig.4) mainly the predicted infrared and Raman spectra are shown as
well as for the cyanates XOCN ( X = H, D, F, Cl, and Br) since intensity data are not
available except for a small portion of the infrared spectra of HNCO and HOCN. The
predicted Raman spectrum for the individual molecules is shown below the infrared spectrum
but only the experimental one for BrNCO is conveniently available for comparison.
Vibrational Spectra And Structural Parameters
The predicted vibrational spectra for the XNCO (X = H, D, F, Cl, Br) molecules are
shown in Figs. 1 4 and it should be noted that in general there are significant differences in
the predicted intensities of the lower frequency bending modes of the infrared spectra and
those in the Raman spectra.
For example, with BrNCO, the ν4 fundamental is barely
observable in the infrared spectrum but it is the strongest Raman band. Similarly the ν6 mode
is extremely weak in the Raman spectrum but the third strongest band in the infrared
34
spectrum. Some similar differences are also predicted for the other XNCO molecules (X = F
and Cl). Thus, these data could be of significant importance for any future Raman studies of
the other halocyanates. However, infrared and Raman data as well as experimental structural
data are available for isocyanic acid.
Isocyanic acid (HNCO and DNCO)
The predicted infrared spectrum of HNCO from scaled MP2/6-31G(d) calculations is
shown in Fig. 4B and that for DNCO in 4D, and the corresponding Raman spectra are shown
in 4C and 4E, respectively. The predicted intensities for
2
in the Raman spectra makes them
almost unobservable compared to the other Raman line intensities whereas the corresponding
symmetric NCO stretch (ν3) is the most intense Raman line with the corresponding infrared
band extremely weak as might be expected from a comparison of these corresponding modes
of CO2. Because of this exceedingly small intensity, this mode was misassigned in the early
infrared investigation [64] with a frequency of 1527 cm-1. The question concerning the
correct assignment for ν3 should not have occurred since this mode was clearly observed by
Goubeau [65] in the Raman spectrum. The predicted Raman intensity of this fundamental is
the second highest (Table 1) and is only exceeded by the predicted intensity of the NH stretch
and the only other fundamental with significantly predicted Raman intensity is the HNC bend.
There was also some controversy concerning the assignment for the HNC bend (ν4) as well
as the out-of-plane NCO bend (ν6, A˝). For example, ν6 was assigned at 777.1 cm-1 [66]
which was later supported by ab initio SCF calculations [67] that was in agreement with
earlier CNDO/2 calculations [68]. However, the higher level calculations (Table 1) clearly
show that the out-of-plane mode is the band at 656.3 cm-1 but again the intensity is extremely
small which undoubtly contributed to the earlier misassignment.
35
The potential energy distributions for
4
and
5
for HNCO and DNCO are rather
interesting where for the deuterium compound, there is 64% contribution of the NCO bend
(S5) to the higher frequency band at 767 cm-1 whereas for the hydrogen compound the
corresponding band at 777 cm-1 is 76% HNC bend (S4). The halocyanate NCO bend is
heavily mixed with the XNC bend for the fluoro- and chloro- compounds as well as with the
X-N stretches. However, this mixing is relatively small for the bromide. Therefore the
simple descriptions of the three bands below the symmetric NCO stretch for the
haloisocyanates do not give a very good indication of the atom displacements for these
modes.
The structural parameters for HNCO have been obtained from two microwave studies
[62, 69] with relatively low uncertainties for the distances but with larger uncertainties for
the HNC and NCO angles of 123.9
1.7 and 172.6
2.7 , respectively (Table 3). In the
initial microwave study [69], NCO was assumed to be linear with a relatively large
HNC of
128.0 and a very short N=C distance of 1.209 Å. Later the microwave data were reanalyzed
by using a modified substitution method along with a nonlinear NCO which resulted in more
reasonable angles but with relatively large uncertainties. We have combined the ab initio
predicted parameters from the MP2/6-311+G(d,p) calculations along with the previously
reported rotational constants [69] from the following isotopes 15N, 13C, 18O, and D as well as
with the normal species to obtain the five structural parameters. The values we obtained for
the HNC angle is 126.1
0.5 , and 172.6
0.5 for the NCO angle, both of which are nearly
in agreement with the values previously reported but with smaller uncertainties. Also it
should be noted that the values obtained from the ab initio MP2/6-311+G(d,p) calculations as
well as those from the B3LYP/6-311+G(d,p) calculations are in reasonable agreement with
36
the experimental values but different from the theoretical values previously reported [70]
with the smaller 6-31G(d,p) basis set.
The centrifugal distortion constants for HNCO and four of the isotopomers were
experimentally determined [62]. The value for DK of 6,065(19) MHz is very large but the
predicted value from the B3LYP/6-311+G(d,p) calculations is only about one-half this value
at 3389 MHz (Table 4). However, the value from the MP2/6-311+G(d,p) calculation is about
one-third the experimental value. The DJ and d1 constants are predicted very well with the
DJK also satisfactorily predicted but the d2 constant is very poorly determined. The same
trend is found for the values of all of these constants for the other four isotopomers. It
appears that the different size of the basis sets does not significantly improve the quality of
the predicted values but in some of the cases the B3LYP method seems to give slightly better
predicted values.
Fluorine Isocyanate (FNCO)
The initial infrared spectrum of FNCO was reported by Gholivand et al. [71] from an
argon matrix where the sample was generated by UV-photolysis of FC(O)N3 and five of the
six fundamentals were assigned (Fig 1). The NCO symmetric stretch (ν2) was not assigned
but as can be seen from the ab initio predicted spectrum this fundamental is predicted to be
very weak (Fig. 1B) as might be expected from the similar mass for the oxygen and nitrogen
atoms and their electronegativities. The predicted frequency is expected to be the lower limit
since this mode is predicted too low for all of the other XNCO molecules investigate herein.
Initially there were two modes (ν5, ν6) which were misassigned with the weak band at 646
cm-1 (Fig.1A marked with an asterisk) assigned as ν6 but it is obvious that this band is not a
fundamental. The band at 529 cm-1 which was previously assigned [71] as ν5 is in fact ν6, the
37
out-of-plane NCO mode, and ν5 is predicted to have a frequency of 196 cm-1 which was
beyond the range of their spectroscopic investigation at the time. Later in a reinvestigation
[72] of the infrared spectrum of FNCO including
15
N,
13
C, and
18
O isotopic species by
utilizing a neon matrix, ν5 was observed at 203.5 cm-1 and ν6 reassigned to the band at 533.8
cm-1 (Table 1).
The predicted intensities for the four observed fundamentals shown in Fig. 1 agree
very well with the observed values. It is probable that the predicted intensities for the Raman
lines would also agree well with the Raman data if the spectrum were recorded. Finally the
extensive mixing of two of the A' modes should be noted. Therefore, to refer to one of them
as the FNC bend (Table 1) is an over simplification since this mode is extensively mixed
with the NCO bend. A similar problem is also found for the in-plane NCO bend of the
corresponding chlorine molecule.
No structural studies have been reported for FNCO so it was not possible to obtain
adjusted r0 parameters from the ab initio predicted values but we have some estimated r0
parameters (Table 3). To determine the expected quality of the NF distance we have carried
out some ab initio calculations on some other NF containing molecules: F2C=NF, HNF2, and
CH3NF2 where the NF distances have been experimentally determined [73-75], respectively.
The predicted NF distances were compared to those experimentally determined by the same
basis sets and level of calculations used to predict the parameters for FNCO. For F2C=NF
the MP2(full)/6-311+G(d) calculations predicted the NF distance within 0.002 Å which was
the listed uncertainty of the experimental determination.
For the HNF2 molecule the
predicted distance was too short by 0.009 Å but the B3LYP calculation with the same basis
38
set gave a predicted NF distance of 0.003 Å. A similar results was obtained for the CH3NF2
molecule.
Therefore, we believe the NF distance for FNCO should be somewhere between the
predicted value of 1.403 Å for FNCO from the MP2 calculations (Table 3) and 1.409 Å from
the B3LYP calculations, i.e. 1.406 Å.
3.3 Chlorine Isocyanate (ClNCO)
The infrared spectrum of gaseous ClNCO has been reported [76] and five of the
fundamentals were assigned from observed bands with the low frequency ClNC bend (ν5)
estimated at ~230 cm-1 from combination and difference bands (Table 1). However as can be
seen from the predicted frequency of 166 cm-1 for this fundamental, this estimated frequency
is entirely too high. Based on the predicted frequencies for the FNC and BrNC bending
modes compared to the observed values it is expected that the ClNC bend should be within
the estimated value of 166 ± 4 cm-1. The predicted frequencies for the other fundamentals
have errors similar to those found for the corresponding vibrations for the FNCO molecule.
The predicted intensities of the infrared bands agrees very well with the
experimentally observed values (Fig. 2) where the ν4 mode is predicted to be quite weak and
ν1 very, very strong. The ν6 fundamental is predicted to be quite weak in the Raman spectrum
but there is no experimental data for comparison. These predicted intensities for the infrared
and Raman spectra do not differ significantly among the three different calculations (Table 1).
There have been two microwave studies of ClNCO where in the first investigation
[77] only three isotopic species (35Cl,
37
Cl,
18
O) were studied so only a partial substitution
structure was obtained which necessitated the use of one of the principal moments. This
resulted in a rather large variation in the bond lengths and angles depending on the moment
39
used to obtain the distance so the accuracy of the determined parameters was quite low
(Table 3). However in the second microwave study [78], three more isotopic species were
investigated which included
15
N with
35
Cl and
37
Cl and
13
C with
35
Cl so sufficient spectral
data were obtained with all of the atoms substituted. A complete substitution structure was
obtained and the determined parameters are listed in Table 3. The listed uncertainties are
relatively small so these results were used to evaluate the quality of the adjusted r0
parameters obtained from the ab initio predicted parameters adjusted to fit the microwave
rotational constants. These data show the use of the ab initio predicted values coupled with
the microwave rotational constants provide excellent values for the structural parameters.
Bromine Isocyanate (BrNCO)
The infrared spectrum for BrNCO was initially reported [79, 80] with only about onehalf of the observed bands assigned. A more extensive study of the infrared spectrum of the
gas was reported [76] with frequencies given for four of the fundamentals and six other
bands assigned with four of them as combination bands and two as difference modes. This
study was followed by an in-depth vibrational investigation [81] which included Raman
spectrum of the liquid, infrared spectrum of the gas and matrix-isolated (Ne and Ar) of the
normal species and Ar matrices for
15
N,
13
C, and
18
O enriched species. Some of these
spectral data are shown in Fig. 3 along with the ab initio predicted infrared and Raman
spectra. Remarkably good agreement is found between the observed and predicted spectra of
both the infrared and Raman data.
confidently assigned (Table 1).
From these data the final two fundamentals were
There were some very large differences between the
frequencies in the neon matrix and those from the liquid, i.e. 137 g (150 l); 506 g (490 l); 572
40
g (560 l); 2196 g (2168 l). These differences indicate significant association in the liquid
phase.
The force constants were determined from the large amount of spectral data and the
reported values are given in Table 5 along with those predicted from the ab initio calculations.
The predicted force constant for the N=C stretch is considerably larger (Table 5) than
the value previously reported [81].
Also the interaction force constant for the C=O
stretch/NCO bend is predicted with a value four times (1.255 Å) the value suggested [81]
earlier. Also the C=O stretch/BrNC bending interaction constant is relatively large whereas
the N=C stretch/NCO bend interaction constant is quite small compared to the values
previously given. With the exception of these constants there is reasonable agreement with
the values obtained [81] from the corrected frequencies of the different isotopic species.
There have been two microwave studies of BrNCO where in the first one only two
isotopomers were investigated (79Br and
BNC and
81
Br) and three parameters were varied r(Br-N),
NCO with the other parameters assumed to be the same as the corresponding
parameters of the ClNCO molecules [82, 83]. In the later microwave study [84] the
isotopomers (79Br and
18
O
81
Br) were investigated and with the rotational constants of the four
isotopomers both r0 and rz structural parameters were determine (Table 3). Both of these
results gave a rather short NC distance compared to the value obtained in this study by
combining the ab initio predicted parameters with the fit to the microwave rotational
constants (Table 3). The ab initio predicted parameters for the NCO moiety are nearly the
same for the chlorine and bromine cyanate which is a strong indication that the relatively
large difference suggested from the microwave data alone are probably in error.
41
Three centrifugal distortion constants were determined but ΔK and δK were not
obtained. These two constants are expected to be relatively large (see predicted values in
Table 4). The predicted values for the three experimentally determined constants are in
reasonable agreement for the ΔJK but the ΔJ value, differ by ~15% and the δJ even more by
about 25%. It should be noted that the experimental determination of only three of the
constants could lead to a significant difference in the determination of these parameters,
particularly since the ones not determined are predicted to be quite large.
Cyanic Acid (HOCN and DOCN)
The vibrational spectra of HOCN and DOCN have only been recorded [85] in
matrices since they are relatively unstable. The samples were generated by photolysis of
HNCO and DNCO trapped in Ar and N2 matrices. The frequencies of the observed bands
with their assignments are listed in Table 2 along with the predicted value.
There is
remarkably good agreement with the predicted and observed frequencies (Figs. 4 and 5). For
example, in Fig. 5A in the region from 1050 to 1250 cm-1 the background from HNCO in a
nitrogen matrix is shown before (solid curve) and after photolysis, the new bands at 1241 and
1098 cm-1 are undoubtedly the ν3 (HOCN bend) and ν4 (OC stretch) fundamentals. Also in
Fig. 4A the spectral region below 500 cm-1 has a well defined band at 460 cm-1 which was
previously assigned [85] as the OCN bend. However, these investigators also reported
another band at 438 cm-1 below the published spectra which must be the in-plane OCN bend
(ν5). Thus, the 460 cm-1 band must be ν6 the out-of-plane bend (Table 2). These data clearly
verify that the sample obtained from photolysis of HNCO is in fact the HOCN tautermer and
not some other species. The previous assignment [85] of a band at 437 cm-1 as a fundamental
for the DOCN molecule appears to be in error since ν6 is not predicted to shift with
42
deuteration. Also it can not be ν5 since it is predicted to shift about 20 cm-1 from the 438 cm1
frequency for this mode for the HOCN molecule. Therefore at this time the two OCN bend
have not been identified for the DOCN molecule. Since the published [85] spectral data are
not continuous it is not possible to evaluated the quality of the predicted infrared intensities
but the predicted variations are relatively small except for ν5 which is predicted to be quite
weak (Fig. 5B). The P.E.D.s for HOCN indicated very little mixing of the normal modes but
for the DOCN molecule there is some mixing of the OC stretch with the DOC bend.
There has been no structural predictions made for HOCN but the C≡N distance is
reasonably constant with a value of 1.158(3) Å irrespective of the substitution on the carbon
atom. There is also little variation for the -C≡C- bond distance of substituted acetylenes.
The microwave spectrum has been reported [86] for the methyl analogue of HOCN and by
using a non-linear OCN the C≡N distance was determined to be 1.162(5) Å whereas the
linear OCN gave a value of 1.151(8) Å for this distance. From a later microwave study [87]
which included the 13CH3OCN and CH318OCN isotopomers, rs parameters were reported for
the heavy atoms and the reported C≡N distance of 1.146(3) Å for this parameter is an
unrealistically small value.
From the earlier study [86] the O−C distance varied from
1.319(10) Å (planar model) to 1.283(8) Å (bent model) for the CH3OCN molecule. These
values are quite consistent with the values for the corresponding predicted r0 parameters
given in Table 6 for the HOCN molecule.
Halocyanides (FOCN, ClOCN, BrOCN)
Since there are no vibrational data for any of the haloisocyanates, it is not possible to
evaluate the quality of the predicted infrared and Raman data (Table 2) for any of them. It
should be noted that there is little difference in the predicted frequencies between MP2/6-
43
31G(d) and MP2/6-311+G(d) levels of calculations (Table 2). Also, there is only a small
difference in the predicted infrared intensities or Raman activities between these two levels
of calculation (Fig. 6). Since there is little difference between the predicted spectral data
from the MP2/6-31G(d) and MP2/6-311+G(d) calculations, there is little reason to use the
larger basis sets which consumes much more computing resources, particularly for much
larger molecules. However, it should be noted that the B3LYP/6-31G(d) calculation has
some differences (frequencies and intensities) for the C≡N stretch since there is a significant
difference in the bond length for this moiety compared to the value from the ab initio
calculations.
We have also calculated the structural parameters for these haloisocyanates (Table 6)
and also estimated the r0 parameters by adjusting the predicted values of the C≡N and O-C
distances transferring the adjustments made on the predictions for these parameters for the
isocyanic acid compared to those from the reported microwave data for CH3OCN [86]. The
angles were kept to the values predicted from the MP2/6-311+G(d) calculations. Thus, fairly
confident parameters can be given for all of the parameters except those for the X-O bond
distance (X= F, Cl, Br).
There are very few experimentally determined O-X distances from which the ab initio
predicted values can be compared. There is one set, i.e. HOX (X= F, Cl, Br) which we
initially used for comparison. For HOF the O-F distance was reported to be 1.442(1) Å [88]
but the predicted values from the MP2/6-311+G(d) calculation was 1.430 Å whereas the
B3LYP/6-311+G(d) predicted value was 1.435 Å which is in closer agreement with the
experimental value. For the FOCN molecule the predicted values were much longer at 1.469
(MP2) and 1.492 (B3LYP) Å where the ab initio and DFT calculated values have much
44
greater difference. From these data it is difficult to make a reasonable estimate for the F-O
distance for FOCN. Thus, the MP2 predicted value has been used as the adjusted r0 value
(Table 6).
The situation is even more difficult since the experimental determined value for the OCl distance from HOCl studies has a very large uncertainty (1.695(27) Å) [89] with predicted
values of 1.714 Å (MP2) and 1.732 Å (B3LYP). Again we have simply given the MP2
predicted values of 1.749 Å as the estimated value. Similarly for HOBr the experimentally
determined value is 1.834(1) Å [90] with predicted values 1.859 Å (MP2) and 1.871 Å
(B3LYP). These predicted values are so much larger than the experimentally determined
value, it is expected that a similar problem exist for the BrOCN molecule where BrO distance
is predicted to be 1.939 Å (B3LYP) but the RHF calculation gives a more reasonable value
of 1.833 Å (Table 6)! Nevertheless we have listed the unreasonably long value of 1.897 Å
from the MP2 calculation as the BrO distance for the adjusted r0 value although it is expected
to be much shorter, i.e. range of 1.85 to 1.87 Å.
Discussion
It is clear that the NCO moiety is bent in the ground vibrational state from both the
microwave data and the ab initio predicted parameters. We have calculated the linearity of
the NCO moiety for all of the XNCO (X = H, F, Cl, Br) molecules (Table 7) and the values
are very similar for all four molecules. The predicted values for the HNCO molecule range
from a low 222 cm-1 (RHF/6-311+G(d,p)) value to the high value of 396 cm-1 (MP2(full)/6311+G(d,p)) with an average value of 335 cm-1 from the five different calculations. A
similar average value is found for the BrNCO molecule with the other two molecules having
values of 390 cm-1. These values on an average are ~130 cm-1 larger than the barriers to
45
linearity of the NCS moiety for the similar molecules [57]. The barriers to molecular
linearity of the XNCS (X = H, F, Cl, Br) molecules are somewhat small (~1000 cm -1
average) than the corresponding ones for the XNCO (X = H, F, Cl, Br) molecules. Also it
should be noted that the barrier to molecular linearity of FNCO is about five times larger than
the value for HNCO and about three times larger than those for the other two halides (Table
7). The larger barriers to molecular linearity for the XNCO molecules are undoubtably due
to the significantly small
XNC angle of the cyanates compare to the similar angle of the
isothiocyanantes.
We have also calculated the barriers to linearity of the OCN moiety for the XOCN
molecules as well as the barriers to molecular linearity of these molecules (Table 8). As
expected the OCN barriers are quite small (one-tenth) compared to the NCO barriers but
significantly larger (~ five times) for the molecular linearity except for the FOCN molecule
which is about three times larger. A comparison of the molecular linearity barriers of the
XSCN to those for XOCN shows the sulfur analogues to have values to about a factor of
three larger except for the fluorine which have similar barrier [57]. The barriers for FOCN
are very similar to those for FSCN.
The most interesting information of the vibrational data was for the HOCN and DOCN
molecules where previously only one of the two OCN bends had been assigned [85].
However, it is clear that the OCN in-plane mode had been observed earlier at 438 cm-1 but
not assigned; this band has now been assigned as the in-plane mode whereas the earlier
assigned OCN bend must be the out-of-plane mode. However, the 437 cm-1 band assigned as
an OCN bend for the DOCN molecule must be due to an impurity since the out-of-plane
OCN bend observed at 460 cm-1 for the HOCN molecule will not shift with deuteration
46
whereas the in-plane motion is expected to shift about 20 cm-1. It is interesting that the OCN
out-of-plane bend for the CH3OCN molecule is predicted at 484 cm-1 from similar ab initio
calculation [91]. However, the in-plane OCN bend for the CH3OCN molecule is predicted at
601 cm-1 due to mixing with the COC bend.
The predicted values for the centrifugal distortion constants for BrNCO are of
reasonable magnitude for ΔJ even though the isotopomers have ΔK values 104 larger than ΔJ
values. However, because of the magnitude of ΔK it was not experimentally determined.
Also the predicted ΔJK values are very large (~500 kHz) and positive whereas the reported
experimental values are about one-third this value but negative. Similarly the experimentally
determined δJ value is about twice the predicted value but with a different sign. It is believed
that the experimental determination of only three of the five centrifugal distortion constants
has lead to the very poor agreement with the predicted values. As an indication of the
probable agreement of the predicted and experimental values, one should note those for
HNCO where all five of the centrifugal distortion were experimentally determined.
The CO distances for the XNCO molecules do not appear to vary significantly with
large changes with the electronegativity of the X group based on the distances of the
parameters in HNCO and F-, Cl- and BrNCO. Also the NC distance seems to have a
relatively small variation so it should be possible to determine the structural parameters of
the CH3NCO, SiH3NCO, and GeH3NCO molecules which should provide information on the
effect of the Si atom on the NC bond length. Additionally, the knowledge of the bond
distance of NCO should help to address the problem in the differences in the interpretation of
the microwave data as arising from the trans conformer of ethylisocyanate and the predicted
gauche conformer as the most stable form obtained from ab initio calculation [61, 92]. It
47
should also aid the determination of conformational stabilities and structural parameters for
other RNCO (R = organic moiety) molecules.
48
Table 1.
Calculateda frequencies (cm-1) and potential energy distributions (PEDs) for XNCO (X = H, D, F, Br, Cl) from the
MP2(full)/ and B3LYP/6-31G(d) basis set.
IR Intensityb
Description
FNCO
A'
1
2
3
4
5
NCO asym stretch
NCO sym stretch
N-F stretch
NCO bend
FNC bend
NCO bend
49
A''
6
ClNCO
A'
1 NCO asym stretch
2 NCO sym stretch
3 N-Cl stretch
4 NCO bend
5 ClNC bend
A''
6 NCO bend
BrNCO
A'
1 NCO asym stretch
2 NCO sym stretch
3 NCO bend
4 N-Br stretch
5 BrNC bend
A''
6 NCO bend
Tables continues
Raman
Activityc
MP2 B3LYP Scaled
Obs
Gasd
MP2
B3LYP
MP2
B3LYP
2257
1287
911
699
196
517
2253
1308
875
712
214
540
344
0.4
46.7
7.4
10.7
15.6
567.8
2.4
79.8
9.2
13.6
20.7
11.3
8.6
13.9
2.5
2.1
0.7
17.2
7.3
18.6
3.5
2.0
0.3
2141
1221
864
699
196
517
2311
1347
702
611
166
540
2266
1325
694
617
153
564
635.5
4.9
30.2
2.3
12.1
20.0
918.8
2.5
32.8
0.8
10.5
23.7
14.8
13.4
20.3
11.0
2.9
0.5
19.3
8.1
11.4
22.9
3.2
0.1
2192
1278
702
611
166
540
2212.2
1306.6
707.7
607.7
~230.0
559.0
2298
1319
676
509
135
550
2271
1348
680
491
136
577
703.8
11.9
17.3
0.3
6.5
17.2
1035.3
6.6
22.6
0.5
6.9
22.4
22.2
14.8
7.8
23.2
3.0
0.3
25.2
8.6
8.3
28.3
3.1
0.002
2180
1251
676
509
135
550
2198.0
1294.5
687.7
—
—
569.9
Matrixe
2174.8
1245.0
860.8
701.5
203.5*
533.8*
P.E.D.sf
98S1
80S2, 18S3
79S3, 15S2
56S4, 36S5
67S5, 35S4
100S6
99S1
80S2, 15S3
78S3, 18S2
54S4, 35S5
67S5, 36S4
1006
2196.0
1290.8
691.1
506.0
137.4
572.2
99S1
91S2
79S3, 10S4, 10S5
88S4, 10S3
91S5, 11S3
100S6
Table 1. Continued
IR Intensityb
Description
MP2
B3LYP
MP2
B3LYP
Raman
Activityc
MP2 B3LYP Scaled
50
HNCO
A'
3732
3733
169.6
170.5
86.2
1 N-H stretch
2376
2316
506.3
782.1
0.1
2 NCO asym stretch
NCO
sym
stretch
1316
1293
0.4
0.5
33.5
3
779
811
266.3
199.0
5.9
4 HNC bend
567
577
81.9
97.5
0.8
5 NCO bend
A''
NCO
bend
618
633
0.8
3.5
0.7
6
DNCO
A'
2762
2726
205.7
227.8
37.9
1 N-D stretch
NCO
asym
stretch
2346
2312
453.6
710.4
0.9
2
1297
1318
0.8
0.7
34.2
3 NCO sym stretch
694
703
82.4
64.6
3.1
4 DNC bend
NCO
bend
471
457
108.2
99.1
0.7
5
A''
602
621
8.8
12.9
1.0
6 NCO bend
a For HNCS and DNCS the basis sets included p orbitals for the H and D atoms.
b Calculated infrared intensities in km/mol.
c Raman activities in Å4/u.
d Frequencies for HNCO Ref. [70], DNCO [47], ClNCO and BrNCO [76].
e Frequencies for FNCO Ref. [72], BrNCO Ref. [81].
f Calculated with MP2/6-31G(d) and contributions of less than 10% are omitted.
* Indication of misassignment in an earlier investigation [71].
Obs
Gasd
Matrixe
P.E.D.sf
83.5
2.1
27.4
6.2
0.7
0.4
3540 3538.3
2254 2268.9
1316 1322.6
779 776.6
567 577.4
618 656.3
99S1
99S2
100S3
76S4, 24S5
76S5, 24S4
100S6
36.4
3.1
28.4
3.4
1.1
0.4
2620 2634.9
2226 2235.0
1297 1310.0
694 766.8
471 460.0
602 602.9
91S1
92S2
100S3
64S5, 37S4
64S4, 36S5
100S6
Table 2.
Calculateda frequencies (cm-1) and potential energy distributions (PEDs) for XOCN (X = H, F, Br, Cl) from the
MP2(full)/ and B3LYP/6-31G(d) basis set.
Description
51
FOCN
A' 1 C≡N stretch
2 O-C stretch
3 O-F stretch
4 FOC bend
5 OCN bend
A'' 6 OCN bend
ClOCN
A' 1 C≡N stretch
2 O-C stretch
3 O-Cl stretch
4 OCN bend
5 ClOC bend
A'' 6 OCN bend
BrOCN
A' 1 C≡N stretch
2 O-C stretch
3 OCN bend
4 OBr stretch
5 BrOC bend
A'' 6 OCN bend
Table continues
IR Intensityb
MP2 B3LYP
Raman Activityc
MP2 B3LYP
Scaled
Matrixd
Ar
N2
P.E.D.e
MP2
B3LYP
2212
1035
839
587
232
443
2256
1083
729
569
235
498
1.1
2.4
15.6
8.4
7.0
6.1
2.0
0.1
60.6
37.8
7.7
9.8
48.7
2.5
13.7
2.6
4.2
2.7
139.1
5.5
53.8
28.1
6.9
2.1
2098
982
796
587
232
443
91s1
88s2
91s1
52s4, 35s5
60s5, 40s4
100s6
2286
1143
709
491
168
398
2287
1358
707
615
155
539
36.1
39.6
7.6
0.6
5.6
8.6
740.6
2.3
32.3
1.9
10.2
23.6
140.6
32.5
38.4
15.8
5.7
3.2
10.1
7.0
17.6
11.8
2.6
0.7
2169
1084
709
491
168
398
87s1
85s2
68s3
31s4, 38s5, 23s3
54s5, 48s4
100s6
2205
1062
634
451
173
457
2272
1094
632
403
171
512
24.1
36.0
0.9
1.3
3.6
6.4
16.1
8.9
3.4
6.6
2.7
9.0
198.2
35.5
22.1
26.5
8.1
3.3
596.8
95.6
37.0
81.1
18.9
2.9
2092
1008
634
451
173
457
93s1
90s2
32s3, 37s4, 31s5
60s4, 29s3
61s5, 37s3
100s6
Table 2 continues
Description
MP2
B3LYP
IR Intensityb
MP2 B3LYP
Raman Activityc
MP2 B3LYP
52
HOCN
A' 1 O-H stretch
3726
3698 158.0
120.0
78.4
94.1
C≡N
stretch
2273
2389
63.6
106.2
19.0
25.6
2
1268
1255 119.7
110.5
4.9
7.2
3 HOC bend
1080
1103
59.9
63.2
6.0
5.7
4 O-C stretch
OCN
bend
428
432
16.7
14.9
3.7
4.6
5
A'' 6 OCN bend
471
488
3.8
3.2
2.5
2.6
DOCN
A' 1 O-D stretch
2713
2693
89.9
68.5
41.0
49.3
2270
2387
67.2
110.3
18.6
25.0
2 C≡N stretch
1078
1100
69.2
58.4
5.7
6.1
3 O-C stretch
DOC
bend
969
961
43.8
50.1
2.5
3.2
4
406
410
17.0
15.4
3.7
4.6
5 OCN bend
A'' 6 OCN bend
470
486
6.7
6.5
2.7
2.9
a For HOCN and DOCN the basis sets included p orbitals for the H and D atoms.
b Calculated infrared intensities in km/mol.
c Raman activities in Å4/u.
d .Frequencies from Ref [85]
e The frequency is misassigned
f Calculated with MP2/6-31G(d) and contributions of less than 10% are omitted
* Observed but not assigned.
Scaled
Matrixd
Ar
N2
P.E.D.e
3535
2273
1268
1080
428
471
3572
2288
1228
1080
—
—
3506
2294
1241
1098
438*
460
100s1
93s2
97s3
94s4
97s5
100s6
2574
2270
1078
969
406
470
2635
2285
1078
949
—
—
2590
2292
1093
957
—
(437)d
100s1
93s2
91s3
91s4
93s5
100s6
Table 3.
Parameter
HNCO
r(H N)
r(N=C)
r(C=O)
HNC
NCO
A
B
C
|μa|
|μb|
|μc|
|μt|
-(E+167)
FNCO
r(F N)
r(N=C)
r(C=O)
FNC
NCO
A
B
C
|μa|
|μb|
|μc|
|μt|
-(E+266)
Structural parameters,a,b rotational constants, dipole moments and energies for
XNCO (X = H, F, Br, Cl).
MP2/631G(d)
MP2/6MP2/6- B3LYP/6311+G(d) 311+G(2d) 311+G(d)
1.009
1.224
1.184
125.8
171.1
884,244
10,816
10,686
1.857
1.767
0.0
2.563
1.232429
1.007
1.224
1.172
123.2
171.6
827,323
10,951
10,808
1.700
1.767
0.0
2.452
1.368556
1.002
1.219
1.172
124.2
172.0
858,760
10,986
10,847
1.709
1.578
0.0
2.326
1.411163
1.006
1.212
1.166
126.2
172.8
903,406
11,078
10,944
1.690
1.523
0.0
2.275
1.738737
1.418
1.262
1.176
110.7
168.9
58268
4901
4520
0.740
0.412
0.0
0.847
1.144187
1.403
1.258
1.166
112.0
169.1
61291
4917
4552
0.876
0.411
0.0
0.968
1.350977
1.412
1.255
1.164
111.3
169.6
59285
4950
4568
0.859
0.356
0.0
0.930
1.416678
1.409
1.246
1.157
112.6
169.8
61592
4944
4577
0.806
0.344
0.0
0.876
1.898154
53
Microwave
[69]
[62]d
0.986 0.995(6)
1.209 1.214(3)
1.166 1.166(1)
128.0 123.9(17)
(180)c 172.6(27)
912,712
918,504
11,071
11,071
10,910
10,910
1.575(5)
1.35(10)
0.0
2.07(10)
Adjusted
r0
0.995(3)
1.216(3)
1.165(3)
126.1(5)
172.6(5)
918,423
11,057
10,925
1.406(5)
1.248(5)
1.158(5)
112.6(5)
169.8(5)
61716
4943
4576
Table 3 continues
Parameter
ClNCO
r(Cl N)
r(N=C)
r(C=O)
ClNC
NCO
A
B
C
|μa|
|μb|
|μc|
|μt|
-(E+626)
BrNCO
r(Br N)
r(N=C)
r(C=O)
BrNC
NCO
A
B
C
|μa|
|μb|
|μc|
|μt|
-(E+2739)
MP2/631G(d)
MP2/6311+G(d)
MP2/6- B3LYP/6311+G(2d) 311+G(d)
1.709
1.241
1.181
121.9
169.6
57763
2978
2832
0.807
0.845
0.0
1.168
1.211139
1.708
1.236
1.170
121.1
170.3
55781
3027
2871
0.799
0.775
0.0
1.113
1.408280
1.714
1.233
1.169
120.7
170.8
54329
3039
2878
0.715
0.593
0.0
0.929
1.481774
1.730
1.227
1.162
120.8
171.4
53570
3035
2872
0.603
0.631
0.0
0.873
2.299173
1.880
1.244
1.180
117.9
170.4
41984
2096
1996
1.221
0.796
0.000
1.458
1.369745
1.866
1.238
1.170
118.3
171.2
42550
2120
2020
1.268
0.767
0.000
1.482
1.614637
1.855
1.232
1.170
120.3
171.5
45656
2102
2009
1.238
0.582
0.000
1.368
1.672012
1.883
1.226
1.163
119.9
172.2
43977
2093
1998
1.118
0.650
0.000
1.293
3.233645
a
Microwave
rse
1.705(5)
1.226(5)
1.162(5)
118.8(3)
170.9(3)
51576
3131
2945
rz f,g
1.856(1)
1.222(2)
1.165(1)
118.0(1)
173.1(2)
r0 f,g
1.862 (2)
1.217 (4)
1.169 (1)
117.4 (4)
172.3 (4)
41189
2176
2063
Adjusted
r0
1.706(5)
1.225(5)
1.162(5)
118.9(5)
171.0(5)
51571
3127
2948
1.857(5)
1.228(5)
1.161(5)
117.5(5)
172.3(5)
41188
2174
2065
For HNCO the basis sets included p orbitals for the H atoms.
Bond distances in Å, bond angles in degrees, rotational constants in MHz, dipole moments
in Debye and energies in Hartree
c
Assumed.
d
Modified substitution method
e
Ref. [78].
f
Ref. [84].
g
Initial microwave studies [82] show that the parameters has been deduced from limited
isotopic data, only rBr-N, BrNC and NCO were varied.
b
54
Table 4.
Rotational and centrifugal distortion constants* for isotopomers of HNCO and BrNCO in the ground vibrational state.
A (MHz)
B (MHz)
C (MHz)
DJ(kHz)
DJK (kHz)
DK (MHz)
d1 (kHz)
d2 (kHz)
55
A (MHz)
B (MHz)
C (MHz)
DJ (kHz)
DJK (kHz)
DK (MHz)
d1 (kHz)
d2 (kHz)
A (MHz)
B (MHz)
C (MHz)
DJ (kHz)
DJK (kHz)
DK (MHz)
d1 (kHz)
d2 (kHz)
Table Continues
MP2/6- MP2/6B3LYP /6MWa
31G(d) 311+G(d,p) 311+G(d,p)
HNCO
884184
827323
903178
918504
10816
10952
11078
11071
10686
10808
10944
10911
3.193
3.370
3.330
3.486(3)
627.80
723.13
639.46
931.7(6)
2877.36
2313.00 3389.12
6065(19)
-0.0523
-0.0613 -0.0613
-0.07299(8)
-0.0094
-0.0124 -0.0114
-0.036(3)
HN 13CO
881136
8246788
901156
916294
10817
10952
11079
11071
10686
10808
10944
10911
3.193
3.369310
3.330
3.503(7)
620.8
717.83
636.17
933(3)
2855.61
2294.65
3361.71
6065.573b
-0.0526
-0.0615
-0.0614
-0.0732(2)
-0.0095
-0.0124
-0.0114
-0.041(9)
15
H NCO
874728
818721
894134
908968
10493
10622
10746
10738
10368
10485
10619
10585
3.007
3.171
3.132
3.31(2)
683.09
771.63
715.45
1020(6)
2804.30
2253.46
3300.91
6065.573b
-0.0449
-0.0525
-0.0522
-0.0632(2)
-0.0086
-0.0112
-0.0103
-0.03647b
MP2/631G(d)
513461
10056
9863
2.949
-237.400
1100.64
-0.1557
-0.0201
884106
10229
10112
2.861
548.97
2877.93
-0.0443
-0.0076
MP2/6B3LYP /6MWa
311+G(d,p) 311+G(d,p)
DNCO
478170
517541
512472
10201
10295
10314
9988
10094
10079
3.181
3.162
3.259(1)
-151.88
-378.051 -243.2(1)
886.32
1272.95
1,522(6)
-0.1858
-0.1914 -0.2039(2)
-0.0269
-0.0244 -0.051(1)
HNC 18O
827253
903169
918417
10358
10478
10471
10230
10358
10327
3.018
2.983
3.143(2)
634.63
556.67
821.9(9)
2313.45
3389.21 6065.573b
-0.0518
-0.0516 -0.06141(7)
-0.0100
-0.0091
-0.03647b
Table 4 continues
56
MP2/6MP2/6B3LYP /6MP2/6MP2/6B3LYP /6MWa
31G(d) 311+G(d,p) 311+G(d,p)
31G(d)
311+G(d,p) 311+G(d,p)
79
79
BrNCO
BrNC18O
A (MHz)
41892
42785
44008
41190
41023
41893
43059
B (MHz)
2097
2118
2092
2177
1970
1990
1966
C (MHz)
1997
2018
1997
2063
1879
1900
1880
∆J (kHz)
0.9205
0.9685
0.9115
1.137(2)
0.8243
0.8666
0.8167
∆K (kHz)
9015.24
9356.08 10501.45
—
8731.13
9051.97
10139.03
∆JK (kHz)
-153.54
-161.44
-162.49
-174.0(1)
-144.08
-151.28
-152.45
δJ (kHz)
0.1355
0.1399
0.1295
0.1737(3)
0.1170
0.1207
0.1120
δK (kHz)
5.5183
5.7839
6.0302
—
4.9138
5.1423
5.3311
81
16
81
18
BrNC O
BrNC O
A (MHz)
41843
42735
43958
41142
40974
41842
43009
B (MHz)
2082
2102
2077
2160
1955
1975
1951
C (MHz)
1983
2004
1983
2048
1866
1886
1867
∆J (kHz)
0.9078
0.9552
0.8988
1.128(2)
0.8127
0.8544
0.8051
∆K (kHz)
8992.11
9332.19 10475.14
—
8708.29
9028.40
10113.12
∆JK (kHz)
-152.29
-160.13
-161.16
-172.2(1)
-142.89
-150.03
-151.18
δJ (kHz)
0.1328
0.1372
0.1269
0.1700(3)
0.1146
0.1182
0.1097
δK (kHz)
5.4566
5.7195
5.9649
—
4.8559
5.0819
5.2707
* Please note the different distortion constant between HNCO and BrNCO.
a
Microwave values for HNCO Ref. [62] and for BrNCO Ref. [84].
b
Fixed at the value for the parent species or zero
MWa
40308
2044
1942
1.044(3)
—
-162.2(1)
0.153(3)
—
40280
2028
1928
1.026(5)
—
-160.0(2)
0.148(3)
—
Table 5.
Force constants (mdyn/Å)a experimentally determinedb and predictedc for BrNCO.
A"
A'
frBr-N
2.862 (2.868)
frN=C
0.178
frC=O
f
BrNC
f
NCO
(0.128)
0.129
(0.185)
0.343
(0.342)
0.180
(0.180)
13.501 (10.727)
1.023
(0.355)
0.348
(0.033)
0.388
(-0.098)
-0.114 (-0.581)
0.371
(1.255)
0.062
(0.033)
0.717
(0.734)
12.981 (13.994)
0.378
(0.189)
NCO
57
0.016 (0.012)
a
All bending coordinates weighted by 1 Å.
Ref. [78]
c
Values in parentheses are from MP2(full)/6-31G(d) calculations.
b
Table 6.
Structural parameters,a rotational constants, dipole moments and energies for
XOCN (X = H, F, Br, Cl).
RHF/6Parameter 311G(d)
HOCN
r(H-O)
0.9463
r(O-C)
1.2823
r(C≡N)
1.1289
110.8
HOC
178.4
OCN
A
701374
B
11007
C
10837
|μa|
3.627
|μb|
1.737
|μc|
0.0
|μt|
4.021
-(E+167) 0.772632
FOCN
1.360
r(F O)
1.292
r(O C)
1.127
r(C N)
107.5
FOC
177.1
OCN
A
47629
B
5693
C
5085
|μa|
1.671
|μb|
0.665
|μc|
0.0
|μt|
1.798
-(E+266) 0.490156
Tables continues
MP2/631G(d)
MP2/6311+G(d)
MP2/6B3LYP/6311+G(2d) 311+G(d)
0.976
1.311
1.182
109.4
177.2
650965
10318
10157
3.595
1.846
0.0
4.041
1.191790
0.965
1.302
1.175
108.7
176.4
661139
10458
10295
3.653
1.787
0.0
4.067
1.330047
0.963
1.304
1.170
109.7
176.4
674728
10474
10314
3.742
1.670
0.0
4.098
1.371551
0.9679
1.297
1.156
110.8
176.5
678706
10638
10474
3.598
1.699
0.0
3.979
1.693533
0.964(5)
1.302(3)
1.158(3)
109.7(10)
178.4(6)
665749
10591
10426
1.474
1.312
1.185
104.2
175.9
39291
5393
4742
1.563
0.755
0.0
1.736
1.074699
1.469
1.298
1.178
105.2
176.0
40467
5408
4771
1.459
0.841
0.0
1.684
1.275920
1.468
1.301
1.174
104.8
175.7
40286
5434
4788
1.586
0.877
0.0
1.815
1.343902
1.492
1.279
1.162
106.9
175.2
41882
5355
4748
0.979
0.739
0.0
1.227
1.822109
(1.469)b
1.302(3)
1.158(3)
105.2(5)
176.0(5)
40618
5451
4806
58
Estimated
r0
Table 6 continues
RHF/6311G(d)
MP2/631G(d)
MP2/6311+G(d)
MP2/6B3LYP/6311+G(2d) 311+G(d)
Estimated
r0
Parameter
ClOCN
1.687
1.748
1.749
1.750
1.795
(1.749)b
r(Cl O)
1.283
1.312
1.300
1.304
1.283
1.302(3)
r(O C)
r(C≡N)
1.129
1.185
1.178
1.174
1.160
1.158(3)
114.2
111.0
111.2
110.2
113.2
111.2(5)
ClOC
177.7
175.7
175.6
175.4
175.1
175.6(2)
OCN
A
37358
50467
33327
32506
34858
33546
B
3552
3123
3432
3467
3321
3454
C
3244
2941
3112
3133
3032
3131
|μa|
3.223
3.577
3.388
3.375
2.738
|μb|
0.190
0.017
0.263
0.421
0.312
|μc|
0.0
0.0
0.0
0.0
0.0
|μt|
3.229
3.577
3.398
3.401
2.756
-(E+626) 0.601627
1.149629
1.346058 1.420202
2.235930
BrOCN
1.833
1.899
1.897
1.881
1.939
(1.897)b
r(Br O)
1.278
1.309
1.297
1.301
1.278
1.302(3)
r(O C)
1.131
1.186
1.179
1.175
1.162
1.160(3)
r(C N)
115.7
111.5
111.7
111.3
114.5
111.5(5)
BrOC
178.0
175.5
175.5
175.5
175.5
175.5(5)
OCN
A
33333
28696
29099
29042
31324
29123
B
2409
2332
2346
2378
2251
2361
C
2246
2156
2171
2198
2100
2184
|μa|
4.079
4.008
4.194
4.097
3.410
|μb|
0.344
0.262
0.303
0.471
0.340
|μc|
0.0
0.0
0.0
0.0
0.0
|μt|
4.093
4.017
4.205
4.124
3.427
-(E+2739) 0.488622
2.577739
1.557897 1.615586
3.176526
a
Bond distances in Å, bond angles in degrees, rotational constants in MHz, dipole moments in
debye and energies in hartree.
b
There is limited experimental value available to make a confident estimate.
59
Table 7.
Calculated barriers to NCO moiety linearitya (cm-1) and barriers (Hartree) to molecular
linearityb (cm-1) of XNCO (X = H, F, Cl and Br) molecules
NCO
linearity
molecular
linearity
RHF/6-311+G(d,p)c
-167.814554
222
1332
-266.577300
220
8292
MP2/6-31G(d)
-168.232430
385
1643
-267.144186
450
8553
MP2/6-311+G(d,p)c
-168.368556
396
1725
-267.350977
446
8373
MP2/6-311+G(2d,2p)c -168.41116
373
1969
-267.416678
417
8083
B3LYP/6-311+G(d,p)c -168.738737
302
1422
-267.898154
421
7890
ClNCO
Ground State
NCO
linearity
molecular
linearity
RHF/6-311+G(d,p)c
-626.672232
204
2769
-2739.547696
174
2361
MP2/6-31G(d)
-627.211139
502
3033
-2737.630255
431
3205
MP2/6-311+G(d,p)c
-627.408280
456
3128
-2740.614637
388
3106
MP2/6-311+G(2d,2p)c -627.481774
421
2758
-2740.672012
376
2355
B3LYP/6-311+G(d,p)c -628.299173
366
2933
-2742.233645
313
2763
60
a
NCO
FNCO
Ground State linearity
HNCO
Ground State
BrNCO
NCO
Ground State linearity
molecular
linearity
molecular
linearity
NCO moiety linearity is defined with NCO group assuming C v point group, and with XNCO molecules
assuming Cs point group.
b
Molecular linearity is defined with XNCO molecules assuming C v point group.
c
The second designation in parentheses, p-orbital polarization functions, only apply to the hydrogen atom in
HNCO.
Table 8.
Calculated barriers to OCN moiety linearitya (cm-1) and barriers (Hartree) to molecular
linearityb (cm-1) of XOCN (X = H, F, Cl and Br) molecules.
OCN
FOCN
Ground State linearity
OCN
linearity
molecular
linearity
molecular
linearity
RHF/6-311+G(d,p)c
-167.776735
15
9391
-266.495476
26
39008
MP2/6-31G(d)
-168.191790
24
9858
-267.074699
47
27115
MP2/6-311+G(d,p)c
-168.330047
36
9468
-267.275920
40
26982
MP2/6-311+G(2d,2p)c -168.371551
40
9503
-267.343902
49
26781
B3LYP/6-311+G(d,p)c -168.693533
40
8302
-267.822109
63
25101
ClOCN
Ground State
OCN
linearity
molecular
linearity
RHF/6-311+G(d,p)c
-626.608118
21
12573
-2739.488622
18
10076
MP2/6-31G(d)
-627.153573
48
14839
-2737.577739
62
12969
MP2/6-311+G(d,p)c
-627.346058
54
14270
-2740.557897
56
12460
MP2/6-311+G(2d,2p)c -627.420202
59
14236
-2740.615586
59
12233
B3LYP/6-311+G(d,p)c -628.235930
74
13488
-2742.176526
78
11633
61
HOCN
Ground State
a
BrOCN
OCN
Ground State linearity
molecular
linearity
OCN moiety linearity is defined with OCN group assuming C v point group, and with XOCN molecules
assuming Cs point group.
b
Molecular linearity is defined with XOCN molecules assuming C v point group.
c
The second designation in parentheses, p-orbital polarization functions, only apply to the hydrogen atom in
HOCN.
Fig. 1 Vibrational spectra of FNCO: (A) experimental infrared spectrum (Ref. [71]) from
argon matrix obtained by UV photolysis of FC(O)N3; † spectrum of precursor; * the
original assignment of 6, later reassign to 5 band; simulated spectra from MP2/631G(d) calculation (B) infrared and (C) Raman.
62
Fig. 2 Vibrational spectra of ClNCO: (A) simulated infrared spectrum from MP2/6-31G(d)
calculation; (B) experimental infrared spectrum of the gas (Ref. [76]); (B΄) an
expansion of the weaker bands by five times; * indicating the observed fundamentals;
(C) simulated Raman spectrum from MP2/6-31G(d) calculation..
63
Fig. 3 Vibrational spectra of BrNCO: (A) simulated infrared spectra from MP2/6-31G(d)
calculation; experimental (Ref. [81]) (B) infrared spectrum of the gas (resolution: 1
cm-1), (C) infrared spectrum in neon matrix (resolution: 0.5 cm-1), (D) Raman
spectrum of the liquid (resolution: 4 cm-1); and (E) simulated Raman spectra from
MP2/6-31G(d) calculation.
64
Fig. 4 Vibrational spectra of HNCO and DNCO: (A) experimental (Ref. [85]) infrared
spectrum of HNCO in nitrogen matrix; and (A΄) HOCN spectrum generated from
HNCO; (B) simulated infrared and (C) Raman spectra of HNCO from MP2/6-31G(d)
calculation; (D) simulated infrared and (E) Raman spectra of DNCO from MP2/631G(d) calculation. Note scale changes.
65
Fig. 5 Vibrational spectra of HOCN and DOCN: (A) experimental (Ref. [85]) infrared
spectrum of HOCN in nitrogen matrix: (
) background, (------) after photolysis of
HNCO; (B) simulated infrared and (C) Raman spectra of HOCN from MP2/6-31G(d)
calculation; (D) simulated infrared and (E) Raman spectra of DOCN from MP2/6-31G(d)
calculation. Note scale changes.
66
Fig. 6 Simulated vibrational spectra from MP2/6-31G(d) calculation: FOCN (A) infrared
and (B) Raman; ClOCN (C) infrared and (D) Raman; BrOCN (E) infrared and (F)
Raman.
67
CHAPTER 5
THE r0 STRUCTURAL PARAMETERS, VIBRATIONAL SPECTRA, AB INTIO
CALCULATIONS AND BARRIERS TO INTERNAL ROTATION AND LINEARITY OF
METHYLISOCYANATE
Introduction
A recent vibrational and structural analysis of methylisothiocyanate [59] has
prompted us to similarly investigate another corresponding methyl-pseudo halogen
compound, methylisocyanate. There have been four microwave investigations of CH3NCO
with the initial one by Curl et al. [93] who reported an internal rotational barrier of 17 ± 1
cm-1 (49 ± 3 cal/mol) along with μa = 2.81 ± 0.06 D and the
14
N quadrupole coupling
constant. This study was followed by a more in depth higher resolution investigation with
the barrier to internal rotation reported by Lett and Flygare [63] to be 29 ± 5 cm-1 (83 ± 15
cal/mol) from the splitting of the K = 0, m = ±3 line and the CNC angle of 140° ± 1°, C-N
distance of 1.44 ± 1 Å and a CH3 tilt of 3° towards the lone pair of electrons. The
uncertainties were based on two different structural models where structure I had assumed
N=C=O distance from a millimeterwave study [94] of HNCO (Table 9) whereas the values
for structure II parameters were taken from those of an electron diffraction study [95]. These
investigators [63] also recorded the infrared spectrum of the liquid and solid and reported a
force field and made some slight changes in the vibrational assignments previously reported
[96, 97] as well as questioning the assignment of the methyl torsional mode at 143 cm-1 in the
solid [98] .
68
Since these earlier structural and vibrational investigations there have been one
electron diffraction study [99] and two microwave studies [100, 101] with the determined
structural parameters listed in Table 9. In the more recent microwave study [101], two
different structures were proposed where for structure I the linear NCO proposed by Duckett
et al. [102] for the microwave study of SiH3NCO was utilized and for structure II the bent rs
NCO structure determined by Yamada [62] from HNCO was utilized. These two structures
have CNC angles of 140.0° for structure I and 135.6° for the other one. There is also very
different values for the r(N=C) of 1.199 Å and r(C=O) of 1.174 Å for structure I verse
r(N=C) of 1.214 Å and r(C=O) of 1.166 Å for structure II. The values are quite different and
it is believed that combining the ab initio predicted parameters with the three determined
rotational constants it should be possible to determine these two parameters as well as the
C=N distance and CNC angle with relatively small uncertainties. Additionally both of the
reported CH distance of 1.09984(90) and 1.10106(36) Å from the two structure seem too
long and one of the CH distance should be different from the other two.
Therefore we initiated a structural investigation of methylisocyanate to obtain the r0
parameters more accurately than the currently known values.
Additionally we also
investigated the infrared and/or Raman spectra of the gas, liquid and solid with particular
attention to obtain higher resolution of the CH3 stretches, deformations and rocks to see if the
rotational fine structure of the pseudodegenerate E modes could be observed and assigned
from which the Coriolis constants could be obtained since these data have not been
previously reported in the earlier vibrational studies [93, 96, 97, 103]. Also of interest was
the predicted barrier to linearity as well as the barrier to internal rotation for comparison to
69
values previously reported [93]. The results of these structural, theoretical and spectroscopic
investigations are reported herein.
Experimental
The sample of methylisocyanate was purchased from Chem Service, West Chester, PA.
Purification was initially carried out by trap-to-trap distillation. The sample was further
purified with a low-pressure, low-temperature fractionation column and the purity of the
sample was checked by the infrared spectrum. The purified sample was kept in the dark at
low temperature until it was used.
The mid-infrared spectrum of gas and solid (Fig. 7) were obtained from 3500 to 230 cm1
on a Perkin-Elmer model 2000 Fourier transform spectrometer equipped with a Ge/CsI
beamsplitter and a DTGS detector.
Atmospheric water vapor was removed from the
spectrometer housing by purging with dry nitrogen. The theoretical resolution used to obtain
the spectrum of the gas was 0.5 cm-1. One hundred twenty-four interferograms were added and
transformed with a boxcar truncation function.
The Raman spectra (Fig. 8) were recorded on a Spex model 1403 spectrophotometer
equipped with a Spectra-Physics model 2017 argon ion laser operating on the 514.5 nm line.
The laser power used was 1.5 W with a spectral bandpass of 3 cm-1. The spectrum of the
liquid was recorded with the sample sealed in a Pyrex glass capillary. Depolarization
measurements were obtained for the liquid sample using a standard Ednalite 35 mm camera
polarizer with 38 mm of free aperture affixed to the Spex instrument. Depolarization ratio
measurements were checked by measuring the state of polarization of the Raman bands of
CCl4 immediately before depolarization measurements were made on the liquid sample. The
70
measurements of the Raman frequencies are expected to be accurate to 2 cm-1. All of the
observed fundamentals in both the Raman spectra of the liquid and solid along with their
proposed assignments and depolarization values are listed in Table 10 as well as those for the
infrared spectra of the gas and solid.
The infrared spectra were predicted from the MP2(full)/6-31G(d) and DFT/6-311+G(d,
p) calculations. The predicted scaled frequencies from the MP2 calculations along with
unscaled frequencies from the DFT calculation were used together with a Lorentzian function
to obtain the calculated spectra. Infrared intensities determined from these calculations were
obtained based on the dipole moment derivatives with respect to Cartesian coordinates. In
Fig. 7, a comparison of experimental and simulated infrared spectra of methylisocyanate is
shown. Infrared spectra of the gas and solid along with the two predicted spectra are shown
in Fig. 7 (A-D), respectively.
The predicted spectrum is in good agreement with the
experimental spectrum which indicates the utility of the scaled predicted frequencies and
predicted intensities for supporting the vibrational assignment.
Also to evaluate the quality of the predicted Raman data we have simulated the
Raman spectra from the ab initio MP2(full)/6-31G(d) and DFT/6-311+G(d,p) results. The
evaluation of Raman activity by using the analytical gradient methods has been developed
[46, 47]. Comparison of experimental Raman spectra of the liquid and solid and the predicted
Raman spectra from both calculations are shown in Figure 8 (A-D). There might be some
difference between the predicted and that obtained experimentally due to some association in
the liquid but more probably due to nearly free rotation of the methyl moiety. Nevertheless,
reasonable agreement with the experimental one is observed when one takes into account that
71
the two NCO bends appear as a single Raman line with similar pseudodegeneracies for the
CH3 rocks, antisymmetric deformations and stretches.
Ab Initio Calculations
The LCAO-MO-SCF restricted Hartree-Fock calculations were performed with the
Gaussian-03 program [43] using Gaussian-type basis functions. The energy minima with
respect to nuclear coordinates were obtained by the simultaneous relaxation of all geometric
parameters using the gradient method of Pulay [44] . A variety of basis sets 6-311G(2df,2pd)
as well as the corresponding ones with diffuse functions were employed with the MøllerPlesset perturbation method [38] to the second order (MP2(full)) as well as with the density
functional theory by the B3LYP method.
The total energies for the form with one of the hydrogen atoms of the methyl group
eclipsing the non-bonding electron pair on the nitrogen atom (trans to NCO) is the most
stable (Table 11) with the NCO moiety bend towards the lone pair. The form with a
hydrogen atom eclipsing the NCO group (cis to the NCO) is a transition state with a negative
frequency and an energy difference of 49 ± 25 cm-1 (Table 9) based on the MP2 calculations.
This energy difference is the predicted barrier to internal rotation of the methyl group.
In order to obtain a complete description of the molecular motions involved in the
fundamental modes of methylisocyanate, a normal coordinate analysis has been carried out.
The force field in Cartesian coordinates was obtained with the Gaussian 03 program at the
MP2(full) level with the 6-31G(d) basis set as well as by density functional theory with the 6311+G(d, p) basis set The internal coordinates used to calculate the G and B matrices are
given in Table 9. By using the B matrix [45] , the force field in Cartesian coordinates was
converted to a force field in internal coordinates. Subsequently, scaling factors of 0.88 for
72
CH, and 0.9 for other coordinates were applied, along with the geometric average of scaling
factors for interaction force constants, to obtain the fixed scaled force field and resultant
wavenumbers from the MP2 calculations. A set of symmetry coordinates was used to
determine the corresponding potential energy distributions (P.E.D.s). A comparison between
the observed and calculated wavenumbers, along with the calculated infrared intensities,
Raman activities, depolarization ratios and potential energy distributions are listed in Table
10 from both the MP2 and DFT calculations.
r0 Structural Parameters
We have found that good structural parameters can be determined by adjusting the
structural parameters obtained from the ab initio MP2/6311+G(d,p) calculations to fit the
rotational constants obtained from microwave experimental data using a computer program
A&M (Ab initio and Microwave) [29] developed in our laboratory. We [33] have recently
shown that ab initio MP2/6-311+G(d,p) calculations predict the r0 structural parameters for
more than fifty carbon-hydrogen distances better than 0.002 Å compared to the
experimentally determined values from isolated CH stretching frequencies which were
compared [104] to previously determined values from earlier microwave studies. Therefore,
the three carbon-hydrogen distances can be taken from the MP2/6-311+G(d,p) predicted
values for methylisocyanate.
The earlier microwave study [63] provides rotational constant values for only a single
isotopmer of CH3NCO so only three rotational constants are available but the C=O distance
shows a very small variation with different substituents on the nitrogen atom. Therefore the
three rotational constant were used to determine the C-N, N=C and
CNC r0 structural
parameters. The resulting adjusted parameters obtained are listed in Table 9, where the heavy
73
atom distances should be accurate to ±0.003Å, the C-H distances should be accurate to
±0.002Å, and the angles should be within ±0.5 degree. First it should be noted that the two
different CH bonds have difference of 0.004 Å where the uncertainties for this distance is
0.001 Å although the individual distances have an uncertainty of 0.002 Å. From the latest
microwave study [101] it was assumed that the three CH distances were the same with a
value of 1.1011(4) Å which clearly is too long and the uncertainty is not meaningful relative
to the actual distance. However the four assumed values for the heavy atom parameters that
were taken from the HNCO studies [62] are in excellent agreement with the values we have
obtained. For structure I the values for the assumed parameters were taken from those
obtained earlier [102] for SiH3NCO and they are not in agreement with the values we have
obtained.
In the microwave study [100] preceding the latest one [101] the assumed parameters
were taken from Lett and Flygare’s structure II, except for the C-N distance, which used
previously determined distances [105] of HNCO. In this study [100] the structure I included
a tilt of the methyl group by 1.54 ± 0.23° whereas for structure II the tilt was fixed at zero.
The inclusion of the tilt had minimum effect on the determined C-N distance as well as the
barrier to planarity but significantly effected the V3 term. However the determined CNC
angle is significantly effected by requiring the NCO moiety be linear which results in all
three microwave structural studies [63, 100, 101] having the CNC angle at 140°. Finally it
should be noted that the two determined distances of the NCO moiety from the electron
diffraction study [99] are interchanged with the N=C bond reported as 1.168(5) Å which is
essentially the determined value of 1.166(3) Å for the C=O bond distances and not the N=C
distance. The other distances of 1.202(5) Å attributed to the C=O distance is about 0.010 Å
74
too short for the N=C distance but the value could be effected somewhat by the fixed
linearity of NCO which undoubtedly resulted in the 140.3(4)° value for the CNC angle.
Vibrational Spectra
The previous and most extensive vibrational study [103] of methylisocyanate was
carried out by not only recording the infrared spectra of the gas and solid but also from
matrix isolation studies by using argon and nitrogen. Additionally, Raman spectra of the
liquid and solid were also investigated with the spectroscopic studies supported by ab initio
calculations.
Nevertheless in the current studies the infrared spectrum of the gas was
recorded with higher resolution with a better signal-to-noise ratio which has made it possible
to observe fine structure on the methyl stretch and deformation modes which is the result of
nearly barrier-free rotation of the methyl moiety.
This fine structure on the CH3
antisymmetric stretch is shown in Fig. 9 and for the CH3 antisymmetric deformation in Fig.
10. As can be seen from this fine structure it shows the strong-weak-weak-strong-weakweak sub-bands as observed in perpendicular bands of symmetric-top molecule with CH3
groups such as the methyl halide [105]. For the symmetric top molecule these vibrations are
exactly degenerate and through Coriolis interaction they couple with the overall rotation of
the molecule. The Q-branches of the different sub-bands fit the formula
A 1be shown
B K to arise
A from
B
A following
B K2
ν0sub = c00 ± cA1m1 + c22mB2 which2can
the
formula:
2
2
ν0sub = ν00 + [A'(1
A 1 −2ζ )B − B'] 2± A2[1A'(1 −B ζ K) − B']K
A +B[(A' −
A B')B − (A"
K 2 − B")]K
where ζ is the Coriolis coupling constant and the factor (1- ζ)2 is used [62, 98] rather than (12ζ ) as suggested by Herzberg [106] and Nyguist [107]. For slightly non-symmetric top
molecules such as the 1-halo-2-butynes [108] with sub-band structure like that for the methyl
halides the equation was modified:
75
__
__
__
__
2
2
2
A 1 −2ζ) B− B ] ±
2 [2
A 1A1'(1 B
K 2 k + ( B' − B" )K
ν0sub = ν00 + [A'(1
− ζK) k − 2AB' ]B K +A[(A'B− A")
where A is the rotation constant of the internal rotating group, k is the quantum number for
__
the angular momentum of the rotating group, and B is replaced by B = (B +C)/2. However
for methylisocyanate or the 1-halo-2-butynes the sub-band structure is due to the interaction
of the vibrational angular momentum of the pseudo-degenerate vibration with the internal
rotation angular momentum of the nearly freely rotating internal top (methyl group) and not
the overall rotation of the molecule. Application of the above equation is not correct since
__
the B constant are not appropriate, thus a more correct coupling equation is:
2
2
1 − ζ2 ) B
2 A− 1ζ ) m +
B (F'
K − F")A m B
ν0sub = [ν0 0 + AF(1
] ± 2F (1
A
B
K2
where F" = h/(8π2cIi), the reciprocal reduced moment of inertia for the internal rotation of the
methyl top in the slightly asymmetric top molecule in the grounded vibrational state.
To fit the frequencies of the sub-bands, the value for the kinetic constant F" is
required and it has been obtained from the ab initio structural calculations. For each of the
bands the quadratic coefficient is small since it is the change of the internal rotational
constant between the ground and excited states. By taking F" = 6.0653 cm -1 for the CH3
nearly degenerate stretching fundamental,
1/ν11,
ζ is determined to be 0.211 from the
coefficient of the linear term, 2F(1- ζ) and from ν0 + F(1 − ζ )2 the band center ν0 is
determined to be 2968.2 cm-1 (Table 12). In the same manner for the CH3 deformational
fundamental,
4/ν12,
the ζ = -0.105 from the linear term and ν0 has a value of 1431.0 cm-1.
These values are consistent with those for the corresponding zeta for the methyl halide
molecules [105] .
76
There is also “fine” structure on the CH3 rocking modes but it is not as pronounced as
that observed for the stretches and deformation. Therefore, we did not attempt to assign the
m values to the individual bands but in general, the approximate separations of the bands are
consistent with the spacing expected with the Coriolis interaction of the CH3 rocks.
With the assignment of the two CH3 antisymmetric stretches and two antisymmetric
deformations to pseudodegenerate vibrational modes, the assignments for the fundamentals
of methylisocyanate are significantly different from the one most recently reported [103]
which was from the Raman spectrum of the solid. Thus, vibrational assignments are provided
for the gas, liquid, and solid particularly since the two NCO bends appear as one rather broad
band. The frequencies for the fifteen fundamentals are listed in Table 10. It should be noted
that there are several fundamentals which have significant differences in frequencies with the
changes of phase. These differences are quite noticeable for all three of the CH3 stretches
where the symmetric mode is centered at 2961.4 cm-1 with a well defined A/B contour which
is observed at 2955 cm-1 in the liquid which shifts to 2970 cm-1 in the Raman spectrum. For
the pseudodegenerate CH3 antisymmetric stretches the band center is at 2968.2 cm-1 in the
gas which shifts to ~3007 cm-1 for the liquid and splits to two well-defined fundamentals at
3032 and 3017 cm-1 in the Raman spectrum with comparable bands in the infrared spectrum.
For the CH3 antisymmetric deformation the band center is at 1431.0 cm-1 which is only ~7
cm-1 higher than the symmetric motion in the gas phase, but in the solid it is observed, as
expected, as a doublet at 1485 and 1472 cm-1 in the Raman spectrum but in the infrared
spectrum the doublet is not separated and there is an additional relatively strong band
between the two antisymmetric deformation and the symmetric deformation (Fig. 11).
77
The CH3 rocking modes also have distinctive differences between the Raman and
infrared bands where the A” CH3 rock has nearly zero (0.03 km/mol) predicted intensity so
only the A’ mode is observed in the infrared spectrum whereas both are observed in the
Raman spectrum of the solid. Finally it should be noted that the two NCO bends appear as a
single band in the Raman spectrum of the liquid and the infrared bands of the gas really have
no distinct contours. However, in the infrared spectrum of the solid the two vibrations appear
individually as very broad bands whereas in the Raman spectrum the out-of-plane bend gives
rise to a clear doublet whereas the in-plane mode is relatively strong with a weak shoulder on
the high frequency. Therefore the earlier conclusion [103] that the frequencies in the infrared
and Raman spectra of the solids were approximately the same may have been the result of
limited annealing of the samples which resulted in a more “glassy” material without
significant splitting.
Barriers to Internal Rotation and Linearity
With a maximum in the band contour of the CH3 torsional mode at 50 cm-1 (Fig. 12),
we have calculated the barrier to internal rotation. By neglecting the higher order terms in
the potential governing the internal rotation of the symmetric three-fold rotor, the harmonic
barrier can be obtained from the frequency of the one-dimensional oscillator:
=
(1/2 c)(kIr)1/2 where k = 4 2c2Ir. By definition, the force constant is the second derivative of
the potential: k = (1/2) h2Vn which in wavenumbers becomes: Vn(cm-1) = 8 2Irc 2/nh = 2/n2F
where F is the inverse of the reduced moment of inertia of the top. With F = 6.0653 cm-1 and
the torsional frequency of 50 cm-1 the three-fold barrier is calculated to be 46 cm-1.
This
value is in excellent agreement with the ab initio predicted average value of 49 cm-1 which
has a very large uncertainty (Table 11). Further support for a barrier in this range is the value
78
of 29 ± 5 cm-1 obtain from the splitting of K= 0, m = ±3 microwave lines [63] as well as the
value of 20.7(1) cm-1 obtained from the recent microwave studies [101] . In the earlier
calculation of the barrier from the torsional frequency the F value utilized was 8.559 cm -1
that gives a value of 31 cm-1 which appears more consistent with other values experimentally
determined. However, this F value seems quite large and would result in obtaining a positive
zeta value for the CH3 deformation mode which would be at variance with values obtained
for the corresponding parameter for several other molecules.
The barrier to molecular linearity was determined to be 928.2(47) for structure I and
923.0(39) for structure II from the latest microwave study [101] . We have predicted the
barrier from both MP2 and B3LYP calculations with average values of 801 ± 141 and 647 ±
117 cm-1, respectively, with the value from the ab initio MP2 calculations in agreement with
the value from the microwave study. However, we can obtain a relatively good value from
the Raman spectrum of the gas where either five or six transitions of the CNC bend have
been observed (Fig. 13). The transitions fall to higher frequency and start at 165 cm -1 for the
1← 0 transition with the second one at 174 cm-1 at peak height or 176 cm-1 at one-half width
and the third one at 180 cm-1 with separations of 5 cm-1 for the next two. Assuming that the
fifth transition is at the top of the barrier gives a barrier value of 975 cm-1 which is in
excellent agreement with the value from the microwave study and the ab initio predictions.
The fact that the separation is relatively constant in going to the higher frequency indicates a
significant quadratic term in the potential and also supports the assignment of the lowest
frequency as the fundamental frequency.
The Raman spectrum of the gas is ideally suited for obtaining excited state transitions
since mainly only the Q-branches (Fig. 13) have significant intensity for symmetric and they
79
are relatively sharp lines. Thus the fairly close transition can be easily identified and their
frequencies obtained.
Discussion
The vibrational assignment for the fluid phases of methylisocyanate is quite different
from the one previous [103] given for the solid. For the CH3 deformation, a single band
(1458 cm-1) is observed in the Raman spectrum of the liquid. However, in the Raman
spectrum of the gas the single band in the liquid is clearly split into two Raman line at 1463
and 1443 cm-1 with the higher frequency one more intense [103] . This observed intensity is
consistent with the prediction from the ab initio calculations. In fact the ab initio predicted
intensities of most of the Raman band were reasonably good and significantly better than the
density functional theory prediction where several of the relative intensities of many of the
band were incorrect (Fig. 8). However, for the predictions of the intensities of the infrared
bands there was little variation between the predictions from the ab initio and density
functional theory results (Fig. 7). We also measured the depolarization values of several of
the Raman lines and satisfactory agreement was found for many of the Raman bands between
the predicted and the experimental values.
The determined methyl torsional barrier of 46 cm-1 from the observed methyl torsional
barrier is nearly one-half the value of 86 cm-1 previously reported. It is possible then an
anharmonic model was used to obtain such a high barrier 52cm-1 frequency but the value is
much higher than consistent with any of the experimental results. Also it should be noted
that the V3' value from the microwave study [101] which include the angle gave a value of
30 cm-1 which seem more consistent with most of the results than the lower value of 20.7
cm-1. In the earlier microwave study [63] the authors questioned the assignment of the
80
methyl torsional mode at 128 cm-1 in the solid. However, as can be seen from the infrared
spectrum of the slid there is a very strong band at 143 cm-1 which is clearly due to the methyl
torsion with the CNC shifting from 165 to 198 cm-1 in the spectrum of the solid. The shift
from 50 to 143 cm-1 is not unexpected concerning the large change in vibrational frequencies
in the solid state compared to the corresponding one in the gas. There are three additional
weak infrared bands at undoubtably lattice modes at 103, 95, and 84 cm-1 (Fig. 12) none of
which could be due to the methyl torsional mode. In the Raman spectrum of the solid the
CNC bend is observed as a doublet at 210 and 192 cm-1 with the methyl torsion at 154 cm-1
(Fig. 12). Additionally there are six lattice modes at 108, 103, 84, 68, 58 and 36 cm -1 with
four of them previously reported [103] . Therefore, the previously reported lower value of
128 cm-1 in the infrared spectrum of the solid for the methyl torsion is probably due to the
sample being more amorphous than crystalline.
In addition to the barrier to linearity we also obtained the predicted barrier to linearity
of the NCO moiety which had a predicted value of 286 ± 22 cm-1 from the MP2 calculations
and 231 ± 21 cm-1 from the B3LYP calculations (Table 11). The value from the MP2
calculations is about 100 cm-1 smaller than the predicted value for linearity for NCO of the
HNCO molecule [109] . Also the predicted barrier to molecular linearity is nearly twice as
large for HNCO compared to this value for methylisocyanate.
The only significant coupling is predicted for the symmetric NCO stretch with the C-N
stretch where each has about 30% of the other. This is about 5% higher than previously
reported by Sullivan et al. [103] but drastically different from the predicted coupling in the
earlier vibrational study [63] which was included with the microwave investigation.
81
The determined structural parameters in this study are believed to be as accurate as
can be obtained in the gas phase for this molecule, particularly because of the large amplitude,
low frequency CNC bending mode, as well as the very low barrier to internal rotation. This
problem is clearly shown by the infrared band contours for the CNC bond at 180 cm-1 and the
methyl torsion at 50 cm-1 which are shown in Fig. 12. The bending mode is an A' mode with
a predicted contour of 87% A-type and 13% B-type, but the actual band is essentially
nondescript with the corresponding Raman band with several Q branches. The methyl
torsion is an A" fundamental which should give rise to a pure C-type band with a pronounced
Q branch but the observed band appears very similar to the band for the CNC bend. To
improve on the determination of the band centers on many of the fundamentals requires
much higher resolution of the infrared spectrum with the assignments of the fine structures
which could be a daunting task.
Because of the two low-frequency anharmonic fundamentals it will be difficult to
significantly improve on the structural parameter determinations but an FT-microwave study
with very low temperature could improve on the experimental value of the rotational
constants and with rotational constants from molecules with isotopic substitution of the
heavy atoms these data should significantly reduce the uncertainties in the angles. Such
studies could also provide some information on the centrifugal distortion constants. We have
predicted their values from both the MP2/6-311+G(d,p) calculations as well as those from the
corresponding B3LYP calculations (Table 13). The predicted ∆J, ∆JK and δJ values all have
normal values with similar values from the MP2 and B3LYP calculations. However, ∆K is
predicted to be very large with the value from the B3LYP calculations being nearly twice as
large. As can be seen from these data the K= 0 transitions should have provided a reasonably
82
opportunity as it did to obtain the barrier to internal rotation of the methyl group. Also, it
should be noted that with reliable heavy atom distances and NCO angle the fitting of the
observed transitions for the five parameter model used earlier [101] could significantly
improve the determined values of the barrier to internal rotation and molecular linearity.
Alternatively, the use of the model utilizing quantum monodromy for quasi-linear molecules
[110] might give a good fit to the observed transitions.
Clearly there are still some
significant questions to be answered before a complete understanding of the vibrational and
microwave spectra of methyl isocyanate is forthcoming.
The results on the determination of the structural parameters and very low barrier to
internal rotation should aid in the determination of the most stable conformation of
ethylisocyanate where the microwave results [61] indicate the most stable conformer is the
trans form where as the ab initio predictions have indicate the gauche form as the most stable
conformer [60].
83
Table 9.
Structural parameters,a rotational constants, dipole moments and energies for CH3NCO from the 6-311+G(d,p).
Parameter
MP2
B3LYP
E.D.
[95]
E.D.
[99]
M.W. [63]
b
84
I
1.4371
1.207
1.171
180.0
139.98
1.091
1.091
109.56
109.56
73849
4392
4256
M.W. [100]
c
d
M.W. [101]
e
II
I
II
85
1.44 1.4342(8) 1.4266(10)
1.190
1.207
1.207
1.180
1.171
1.171
180.0
180.0
180.0
140.12 140.18(2) 140.99(1)
1.09 1.091
1.091
1.09 1.091
1.091
f
g
I
II
1.4420(2) 1.456(1)
1.199
1.214
1.174
1.166
180.0
172.6
140.0
135.6
1.0998(9) 1.1011(4)
1.0998(9) 1.1011(4)
108.75
110.38
Adjusted
r0
r(C-N)
1.445
1.442
1.47
1.450(4)
1.447(3)
r(N=C)
1.213
1.198 1.19(3)
1.168(5)
1.215(3)
r(C=O)
1.180
1.174 1.18(3)
1.202(5)
1.166(3)
NCO
172.6
173.9
180.0
180.0
172.6(5)
CNC
135.6
140.7 125(5)
140.26(40)
135.8(5)
r(H6-C)
1.089
1.089
1.09
1.096(10)
1.089(2)
r(H7,8-C)
1.093
1.093
1.09
1.096(10)
1.093(2)
H6CN
108.6
108.9
111.4(14)
108.6(5)
H7,8CN 110.8
110.9
110.8(5)
A
73377 82455
73850
B
4378
4320
4392
C
4242
4215
4258
|μa|
3.126
3.067
|μb|
1.048
0.799
|μc|
0.000
0.000
|μt|
3.297
3.169
a.
Bond distances in Å, bond angles in degrees, rotational constants in MHz, dipole moments in Debye, energies in Hartree, and
parameters with an asterisk are assumed.
b.
Model based on the assumed NCO distance from a millimeterwave studies of HNCO
c.
Parameters are based on electron diffraction study of Eyster et al.
d.
Parameters were taken from Lett and Flygare’s structure II except that of C-N distance.
e.
Same assumed parameters as structure I except without methyl tilt.
f.
Structure based on a linear NCO moiety of SiH3NCO from a MW study by Duckett et al.
g.
A bent rs NCO structure determined by Yamada for HNCO was utilized.
Table 10.
Vib.
No.
A
1
2
3
4
5
6
7
8
9
10
85
A
11
12
13
14
15
a
Observed and calculated frequencies (cm-1) for methyl isocyanate.
Fundamental
Ab initioa
Scaledb
MP2 B3LYP
IR Raman dp
int.c
actd
ratio
CH3 antisym. str.
CH3 sym. str.
NCO antisym. str.
CH3 antisym. def.
CH3 sym. def.
NCO sym. str.
CH3 rock
C N stretch
NCO bend
CNC bend
3229
3116
2413
1562
1517
1491
1190
898
621
181
3121
3029
2377
1514
1484
1455
1153
871
636
163
3029
2923
2290
1482
1440
1415
1129
851
621
181
11.3
40.8
777.4
5.0
26.7
23.1
19.2
27.5
25.3
19.0
74.8
131.4
0.6
14.3
11.5
34.9
3.8
15.3
0.5
1.8
CH3 antisym. str.
CH3 antisym. def.
CH3 rock
NCO bend
CH3 torsion
3202 3089
1568 1506
1167 1129
572 596
72
44
3004
1487
1107
572
72
15.5
5.6
0.03
21.4
0.03
2.7
62.6
21.2
3.4
0.5
1.4
IR
Gas
Solid
Raman
Liquid
Solid
P.E.D.e
A B
0.69
2968
3030
0.01
2961
2968
0.12 ~2300
~2300
0.75
1431
1474
0.68
1424 1453/1436
0.29 ~1416 1412/1402
0.62
1150
1136
0.22
868
854
0.57
624
615
0.70
165
Liquid
~300
3032
97S1 34 66
29557
2970
97S2 87 13
2282
2265
97S3 98 2
1460
1485
84S4 48 52
1460 1435/1443
90S5 94 6
1416
1404
60S6, 28S8 74 26
1130 1131/1149
91S7 97 3
854
856
69S8, 31S6 55 45
596
614
91S9 41 59
178
192/210
95S10 87 13
0.75
0.75
0.75
0.75
0.75
~300
3017
14607
1472
1130
1115
596 584/572
154
2967
1432
1150
592
50
3014
1477
574/560
143
100S11
94S12
94S13
99S14
99S15
MP2(full)/6-31G(d) and B3LYP/6-311+G(d,p) predicted values.
MP2(full)/6-31G(d) fixed scaled frequencies with factors of 0.88 for CH stretches, 0.90 for heavy atom stretches, and CH bendsand
1.0 for all other modes.
c
Scaled infrared intensities in km/mol from MP2(full)/6-31G(d) .
d
Scaled Raman activities in Å4/u from MP2(full)/6-31G(d).
e
Calculated with scaled MP2(full)/6-31G(d) predictions and contributions of less than 10% are omitted.
b
Table 11.
Calculated energies (hartree) and barriers (cm-1)a to methyl torsion, NCO
linearity and molecular linearity of the heavy atoms of CH3NCO.
Staggered
Eclipsed (TS)
NCO
linearity
Molecular
linearity
RHF/6-31G(d)
-206.791283
11
179
270
RHF/6-31+G(d)
-206.796507
9
142
159
RHF/6-311G(d,p)
-206.846044
14
169
333
MP2/6-31G(d)
-207.396251
63
315
776
MP2/6-31+G(d)
-207.408740
86
300
721
MP2/6-311G(d,p)
-207.563259
59
303
971
MP2/6-311+G(d,p)
-207.571643
65
299
927
MP2/6-311G(2d,2p)
-207.617814
14
303
984
MP2/6-311+G(2d,2p)
-207.625663
26
277
867
MP2/6-311G(2df,2pd)
-207.690170
22
268
710
MP2/6-311+G(2df,2pd)
-207.696964
35
252
632
MP2/6-aug-cc-pVTZ
-207.702821
68
261
621
49 ± 25
286 ± 22
801 ± 141
Method / Basis set
Averageb
B3LYP/6-31G(d)
-207.988073
25
266
574
B3LYP/6-311G(d,p)
-208.049717
29
226
623
B3LYP/6-311+G(d,p)
-208.055099
30
211
513
B3LYP/6-311G(2d,2p)
-208.055378
1
236
808
B3LYP/6-311+G(2d,2p)
-208.060934
8
218
718
19 ± 13
231 ± 21
647 ± 117
Average
a
b
All energy differences are relative to the energy of the stagger conformer.
Average of MP2 predicted values.
86
Table 12.
Coriolis fine structure of pseudodegenerate CH3 antisymmetric stretching
and deformation bands.
CH3 antisymmetric stretch
m
obs
obs –
calc
CH3 antisymmetric deformation
obs
obs –
calc
12
3082.4
0.52
11
3072.9
-0.20
10
3063.1
-1.11
1566.7
-0.75
9
3056.2
0.87
1555.8
0.80
8
3046.3
-0.04
1542.7
0.26
7
3037.3
0.01
1529.7
-0.09
6
3027.7
-0.76
1517.3
0.26
5
3019.7
0.73
1504.1
-0.08
4
3010.1
0.39
1491.0
-0.23
3
3000.3
-0.08
1477.4
-0.78
2
2990.7
-0.29
1465.4
0.38
1452.0
0.23
1
Standard deviation
0.68
0.562
c0
2971.99
± 1.04
1438.41
± 0.66
c1
9.565
± 0.333
13.404
± 0.276
c2
-0.034
± 0.023
-0.050
± 0.024
ζ
0.211
-0.105
2968.2
1431.0
0
87
Table 13.
Rotational (MHz) and centrifugal distortion constants (kHz) for CH3NCO.
MP2/6-311+G(d,p)
B3LYP /6-311+G(d,p)
MWa
A
73377.29
82454.90
73849.20
B
4377.64
4320.44
4392.22
C
4241.99
4214.68
4256.66
∆J
1.959
1.946
∆K
18596.55
31184.18
∆JK
-205.94
-183.58
δJ
0.327
0.318
δK
-142.2
-875.7
a
Ref. [63]
88
Figure 7
Infrared spectra of CH3NCO: experimental infrared spectrum of the gas (A) and
of the solid (B); simulated spectra from scaled MP2/6-31G(d) values (C) and
B3LYP/6-311+G(d,p) calculation (D).
89
Figure 8
Raman spectra of CH3NCO: experimental infrared spectrum of the gas (A) and
of the solid (B); simulated spectra from scaled MP2/6-31G(d) values (C) and
B3LYP/6-311+G(d,p) calculation (D).
90
Figure 9
Observed fine structures on the CH3 antisymmetric stretch in the infrared spectra
of the gas.
91
92
Figure 10
Observed fine structures on the CH3 antisymmetric deformation in the infrared spectra of the gas.
93
Figure 11
Observed spectra of solid CH3NCO: infrared (A), Raman (B).
Figure 12
Low frequency spectra of CH3NCO: infrared of gas (A), infrared of solid (B),
Raman of solid (C).
94
Figure 13
Raman spectrum of the gas for the CNC bend with “hot band” transitions.
95
CHAPTER 6
CONFORMATIONAL STABILITY, STRUCTURAL PARAMETERS AND
VIBRATIONAL ASSIGNMENT FROM VARIABLE TEMPERATURE INFRARED
SPECTRA OF XENON AND KRYPTON SOLUTIONS AND AB INITIO
CALCULATIONS OF ETHYLISOCYANATE
Introduction
Organoisocyanates and organoisothiocyanates provide interesting challenges for the
experimental determination of many of their structures because of the large CNC(X) angles
which give rise to many of the molecules having very anharmonic low frequency large
amplitude bending vibrational modes. Additionally, the barrier to internal rotation of the
NC(X) group may be relatively small which can result in nearly free internal rotation. This
can result in a very large number of excited vibrational states populated even at Dry Ice
temperature which makes the assignment of the microwave spectra quite difficult particularly
when the centrifugal distortion constants may have relatively large values. For this reason
there have been several determinations of the most stable conformation of these molecules by
microwave spectral data or electron diffraction studies which seem at variance with
predictions from ab initio calculations or some of the early vibrational investigations of these
molecules.
Therefore, we initiated a more in-depth investigation of several
organoisocyanates and organoisothiocyanates by utilizing variable temperature infrared
spectra of rare gas solutions along with ab initio calculations with significantly larger basis
sets.
Additionally, by combining the ab initio predicted structural parameters with
96
experimentally determined rotational constants it has been possible to obtain values of the
structural parameters for several of these molecules.
We recently [111] investigated the vibrational spectra of methylisocyanate and
carried out an extensive ab initio calculation from which r0 structural parameters were
obtained by utilizing the previously reported rotational constants. Also, the fine structures
were shown to be the results of nearly free internal rotation of the methyl group which was
one of the suggested possibilities for the bands from an earlier study [97]. As a continuation
of this study we have reinvestigated the vibrational spectrum of ethylisocyanate to determine
the relative conformational stabilities.
From an earlier theoretical study [112] the
equilibrium conformation was determined from MP2/6-31G(d,p) ab initio calculations from
which it was concluded the gauche conformer was the most stable form with a second stable
trans conformer.
However, from such a small basis set one must be skeptical of the
conclusion when the energy difference is very small. Since the frequency calculations of the
cis conformer gave one negative value which indicates that the cis form does not correspond
to a minium-energy structure, this form was not considered to be a possible stable conformer.
Also, the argument that the experimentally determined rotational constants could not be used
to indicate the conformer present since a weighted average must be used is somewhat
difficult to comprehend [112]. However, the determined rotational constants are obtained
from the “ground” vibrational state and the predicted rotational constants from reasonable
structural parameters can usually be used to indicate the conformer that is giving rise to the
rotational transitions. Therefore, a reinvestigation of the experimental spectroscopic and
theoretical data is warranted which should result with all the scientific data agreeing with the
determined information.
97
We have recorded the variable temperature ( 55 to 150°C) of xenon and krypton
solutions of ethylisocyanate. We have also re-evaluated the microwave data. To support the
spectroscopic studies we have carried out more extensive ab initio calculations by utilizing a
variety of basis sets and obtained r0 structural parameters by combining the previously
reported rotational constants with the predicted parameters. We have considered the four
most probable forms (Fig. 14) for the calculations of the conformational energy differences.
The results of these new experimental and theoretical studies are reported herein.
Experimental
The sample of ethylisocyanate was obtained from Alfa Aesar with a stated purity of
98%. The sample was further purified by trap-to-trap sublimation and the purity of the
sample was verified by a comparison with the previously reported spectra. The sample was
kept in the dark at low temperature until it was used.
The infrared spectra of the gaseous, amorphous and annealed solid ethylisocyanate
were recorded on a Perkin-Elmer model 2000 Fourier transform spectrometer equipped with
a DTGS detector. Atmospheric water vapor was removed from the spectrometer housing by
purging with dry nitrogen gas. A Ge/CsI beamsplitter was used to collect the mid-infrared
spectra of the gas and solid. The mid-infrared spectrum of the gas was obtained with the
sample contained in a 10 cm cell fitted with CsI windows. Usually 128 scans were collected
with a theoretical resolution of 0.5 cm-1. The spectrum of the solid was obtained at the
temperature of boiling liquid nitrogen by depositing the sample on a CsI plate fitted with CsI
windows. The spectrum of the amorphous solid sample was collected and when no change
was observed in the recorded spectra upon further annealing it was assumed to be a
polycrystalline solid. Usually, 64 scans were collected with a theoretical resolution of 2.0
98
cm-1. A grid beamsplitter was used to collect the far infrared spectra of the solid. A cryostat
cell with polyethylene windows was used to record the far infrared spectrum of the solid (Fig.
15B) with the sample deposited on a silicon substrate at 77 K, multiple annealings were
performed in order to obtain a good polycrystalline solid.
Typically, 128 scans were
collected with a theoretical resolution of 2.0 cm-1. The interferograms were averaged and
then transformed with a boxcar truncation function.
The frequencies for the observed
fundamentals are listed in Tables 14 and 15 for the cis and trans conformers respectively, and
the predicted values for the gauche form are listed in Table 16.
The temperature-dependent spectra of the sample dissolved in liquid xenon from 55 to
100 C and krypton from 105 to 155 C were conducted on a Bruker model IFS 66 Fourier
transform spectrometer equipped with a Globar source, a Ge/KBr beamsplitter and a DTGS
detector. The temperature studies were performed in a specially designed cryostat cell which
consisted of a copper cell with a path length of 4 cm and wedged silicon windows sealed to
the cell with indium gaskets. The temperature was monitored with two Pt thermoresistors.
The complete cell was connected to a pressure manifold to allow for the filling and
evacuation of the cell. After the cell was cooled to the designated temperature, a small
amount of sample was condensed into the cell. Next, the manifold and the cell were
pressurized with xenon or krypton, which immediately started condensing in the cell,
allowing the compound to dissolve. A typical spectrum is shown in Fig. 16A.
Ab Initio Calculations
The LCAO-MO-SCF restricted Hartree-Fock calculations were performed with the
Gaussian-03 program [43] using Gaussian-type basis functions. Calculations were also
carried out by the Møller-Plesset perturbation method [38] to second order with valence and
99
core electron correlation up to the 6-311+G(3df,3pd) basis set and cc-PVQZ. The density
functional theory (DFT) calculations were restricted to the B3LYP method. [113, 114] The
results of these calculations are listed in Table 17.
These data show that the gauche
conformer is more stable than the cis conformer from MP2 calculations but with the larger
basis sets the gauche well maybe so shallow that it will not contain a bound vibrational state.
In order to obtain a complete description of the molecular motions involved in the
fundamental modes of the conformers of ethylisocyanate, normal coordinate analyses have
been carried out. The force fields in Cartesian coordinates were obtained with the Gaussian
03 program [43] from the MP2(full)/6-31G(d) calculation. The B-matrix elements [45] were
used to convert the ab initio force field from Cartesian coordinates into the force field in
designated internal coordinates. These force constants were used to reproduce the ab initio
predicted vibrational frequencies which are listed in tabular form for the two most probable
conformers, cis (Table 14) and trans (Table 15). The diagonal elements of the force field in
internal coordinates were then multiplied by scaling factors of 0.88 for the CH stretches and
CH deformations, 0.90 for heavy atom stretches and CH bends and 1.0 for heavy atom bends.
The geometrical average of the scaling factors was used for the off-diagonal force constants.
The calculation was repeated to obtain the fixed scaled force field, scaled vibrational
frequencies and potential energy distributions (P.E.D.s) which are given for each molecule in
the vibrational tables.
The infrared spectra were predicted from the MP2(full)/6-31G(d) calculations. Infrared
intensities were calculated based on the dipole moment derivatives with respect to the
Cartesian coordinates. The infrared spectrum of the krypton solution (-155°C) and the
predicted infrared spectra for the pure cis and trans conformers, as well as the mixture of the
100
two conformers with relative concentrations calculated for the equilibrium mixture at 25°C
by using experimentally determined enthalpy difference are shown in Figs. 16D, 16C and
16B, respectively. The predicted spectrum is in very poor agreement with the experimental
spectrum compared to what is usually obtained. The scaled predicted frequencies are in good
agreement with the observed bands but the predicted intensities are in very poor agreement
with the observed values so they will be of little value for making the vibrational assignment.
Vibrational Assignment
There is little difference between the ab initio predicted frequencies for the
fundamentals for the cis and trans conformers except for the NCO “in-” and “out-of-plane”
bends as well as the CNC bend which is predicted to be 15 cm-1 higher for the trans
conformer compared to the frequency for the corresponding mode of the cis form. Also, the
CCN symmetric stretch is predicted to be 33 cm-1 higher for the trans conformer than the
corresponding mode of the cis form. However, there is only a 5 cm-1 difference predicted for
the antisymmetric stretch but there is 37 cm-1 difference predicted for the CCN bend. The
NCO torsion is predicted to have a frequency of 31 cm-1 whereas for the cis conformer it is
imaginary since the cis form is predicted to be a transition state (Table 14). For most of the
remaining fundamentals the predicted differences in frequencies are only a few wavenumbers
which for the most part they could not be detected even with the sharp band in the rare gas
solutions.
The frequencies chosen for the fundamental vibrations are nearly the same as those
previously proposed [115] with a few exceptions. The 1140 cm-1 band was previously
assigned as an A′ fundamental whereas we have assigned it as the CH2 rock (A″). We also
reassigned the band at 983 cm-1 which was previously assigned as an A″ mode since it was
101
depolarized to the A′ block where the predicted depolarization value is 0.74. We have also
assigned one of the carbon-hydrogen stretches at a higher frequency than the previous
assignment. Thus, the current assignment is not significantly different from the previously
reported one but the P.E.D.s differ since the force constants differ and the approximate
descriptions may also differ.
Once the assignments were made for the fundamentals of the cis form a search was
made to obtain bands which could be assigned to the second conformer. The heavy atom
vibrations are expected to be the most sensitive to the different conformers and they can
usually be found by a comparison of the spectrum of the gas to that obtained from the rare
gas solution (Fig. 17). Clearly there is a pronounced shoulder on the band at 983 cm-1 in both
the xenon and krypton solution at 1002 cm-1 which also appears as shoulder on ν11 of the gas.
This band is assigned as arising from the trans conformer. The 800 cm-1 band also was a
doublet in the spectrum of the krypton solution but there is a very weak band predicted near
ν12. Thus, a considerable amount of the intensity of the second band must be due to the
fundamentals of the second conformer. Other evidence for the second conformer was found
in the spectrum of the gas for the CCN bends where the one for the cis conformer was
observed at 418 cm-1 and that for the trans form at 385 cm-1. However, the two NCO bends
clearly show bands for a second conformer but well defined band maxima were not observed.
However, in the Raman spectrum of the amorphous solid (Fig. 18) all three of these
fundamentals have well separated bands for the second conformers, which disappear with
annealing the sample. Therefore, these spectral data can only be attributed to the presence of
a second form for ethylisocyanate.
Conformational Stability
102
In order to determine the enthalpy difference between the two conformers, variable
temperature studies in liquid krypton and xenon were carried out. Because only small
interactions are expected to occur between the dissolved molecules and the surrounding
noble gas atoms [22-25], the “pseudo gas phase” spectrum shows only small frequency shifts
compared with the spectrum of the gas. A significant advantage of this type of cryogenic
spectroscopic study is that the conformer bands are better resolved in comparison with those
in the spectrum of the gas. This is particularly important since most of the conformer bands
for this molecule are expected to be observed within a very few wavenumbers of each other.
The sample was dissolved in the liquefied noble gases and the spectra were recorded at
varying temperatures as indicated earlier. The spectral changes in liquid krypton of the
conformer pair at 983 (cis) and 1002 (trans) cm-1 are shown in Fig. 19. This pair was
initially chosen for the conformational enthalpy determination due to their relatively even
baseline, satisfactory band separations and relatively clear band intensities. The relative
intensities were measured as a function of temperature and their ratios were determined. Ten
sets of spectral data were obtained for the pair and by application of the van’t Hoff equation,
lnK = H/(RT)
S/R, and the enthalpy difference was determined from a plot of lnK
versus 1/T, where H/R is the slope of the line and K is substituted with the appropriate
intensity ratios, i.e. Icis/Itrans.
It was assumed that
H,
S and the ratio of the molar
absorption coefficients εtrans/εcis are not a function of temperature in the temperature range
studied.
The conformational enthalpy difference (Table 18) was determined to be 98
(1.17
8 cm-1
0.10 kJ/mol) from the krypton solution for this pair. Another band at 796 cm -1 (cis)
was combined with the 1002 trans band from which an additional H value of 103
103
2 cm-1
(1.23
0.02 kJ/mol) was obtained. An average value was obtained by utilizing all the data as
a single set which gave a H value of 100
more stable conformer.
4 cm-1 (1.20
0.05 kJ/mol) with the cis form the
These error limits were derived from the statistical standard
deviation of one sigma of the measured intensity data where the data from the two pairs were
taken as a single set but the values for each set are also given. These error limits do not take
into account small association with the liquid krypton or the interference of overtones and
combination bands in near coincidence with the measured fundamentals. The abundances of
the less stable trans form present at ambient temperature is 27% if there were only two
conformers present. We also attempted to determine a H value from the xenon solution but
the bands were not as well resolved as obtained from the krypton solution.
Structural Parameters
Using the A&M program, the “adjusted r0” parameters for ethylisocyanate was
obtained by utilizing the microwave rotational constants but the results will be somewhat
limited since only three rotational constants have been determined [116]. In Table 19 the
predicted parameters from the MP2(ful)/6-311+G(d,p) and similar B3LYP calculations are
listed along with those previously obtained from both the microwave [116] and electron
diffraction [117] studies. Of particular note is the very large difference of the predicted
values for the C=N distance from the two different methods with a rather short value for the
C=N bond from the B3LYP calculation. There is also a relatively large
CNC angle
predicted from the B3LYP calculations. In an early investigation [57] of HNCO where the
heavy atom distances were well determined, we found the average value between the two
predicted C=N distances agreed with the experimental value but a different result has been
reported for this parameter for the CH3NCO molecule [111]. Nevertheless, we initially fixed
104
the C=N and C=O distances at the values obtained for these parameters for the CH 3NCO
molecule. Only very small adjustment were necessary for the distances except for the C-C
bond with the final results providing a good fit of the B and C constants to 0.8 MHz. The A
constant has a larger variation but it was determined from only A-type transition which have
very small dependence on the value of the A rotational constant. The angles changed very
little from the predicted values.
Therefore, it is believed that the C-H distances have
maximum error values of 0.002 Å with the heavy atoms determined within 0.005 Å and all
angles within 0.5°.
The reported structural parameters from the microwave study [118] were taken from
the mono-substituted ethanes [116] for the ethyl group and those for the isocyanato group
from methylisocyanate but no reference was given. Thus, only the C-N distances and the
CCN and CNC angles values were determined from the microwave rotational constants.
The values for the two angles differ significantly from the values obtained for them from our
study. From the electron diffraction study [117], both the C=N and C=O reported distance
are significantly longer than the values reported herein whereas the CNC angle is at least 6°
smaller than the value obtain from the adjusted parameters. Therefore, it is believe that the
parameters reported in the current study could only be obtained more accurately by obtaining
more rotational constants from two or three more isotopic species as well as more accurate
values of the A rotational constants. Nevertheless, we believe the estimated error limits are
realistic values.
Discussion
The initial vibrational assignment [97, 119] for ethylisocyanate was made by utilizing
only infrared spectral data since the earlier Raman data [119] were very incomplete. Since
105
the predominant conformer was clearly shown from the microwave data to have a plane of
symmetry, and the Raman depolarization data were extensively used for distinguishing the
A″ vibrations from the A′ modes. The 24 normal modes span the representation of 15 A′ and
9 A″ with the A′ modes having A-, B-, or A/B-hybrid-type infrared band contours, whereas
the A″ modes will have C-type infrared bands with P–R separations of 20 cm-1 and should
have a relatively strong Q branch. However, as can be seen from the infrared spectrum of
the gas (Fig. 17A) there is little indication of any strong Q-branches. Therefore, the band
contours were of little value for making the vibrational assignment. However, in the more
recent vibrational study [115] extensive Raman data was reported which even included the
Raman spectrum of the gas. These data are quite valuable since the Q-branches are quite
strong with very broad RS and OP branches which make it possible to clearly identify the
band centers. Additionally, usually only the A′ modes are observed which makes it possible
to confidently assign them.
With only three scaling factors utilized for the predictions from the MP2/5-31G(d)
calculations, the fundamentals in the A′ block have a frequency difference of 6.8 cm-1 which
is an error of 0.4%. In the A″ symmetry block the frequency difference is 12 cm -1 which is
an error of 0.8%. Utilization of a very few scaling factors makes it possible to readily access
the quality of the ab initio predicted frequencies compared to the use of a large number of
scaling factors to fit all of the fundamental frequencies.
In general the P.E.D.s are relatively “pure” for a molecule with only a plane of
symmetry except for the mixing of the CH3 rocks with the C–C stretch of the ethyl moiety.
For example, the band at 1091 cm-1 (
10)
is designated as the CH3 rock but it has 17% of S14,
13% S11 and 12% S12 all of which are CCN modes. This is typical of the mixing in the ethyl
106
group. Also the CCN antisymmetric stretch at 801 cm-1 has only 27% of this mode with
significant contributions from four other symmetries with one from the CH3 rock and the
other three from heavy atom modes. Nevertheless, the mixing is relatively small with the
major contributions from the approximate description of the indicated vibrational mode.
As can be seen from the information provided in Table 15 most of the carbon-hydrogen
modes for the trans conformer are within a very few wavenumbers from the corresponding
modes of the cis form.
Therefore, these trans modes are considered to be essentially
degenerate with the corresponding modes of the cis conformer. However, for several of the
heavy atom modes the frequencies for the trans form are significantly different from those of
the corresponding vibrations of the cis form. The CCN bend
14
is predicted 37 cm-1 higher
for the cis form than the corresponding vibration for the trans vibration. These bands are
observed clearly at 419 cm-1 (cis) and 385 cm-1 (trans) which is a separation of 34 cm-1.
Similarly for
11
(CCN symmetric stretch), this mode is predicted to have the higher
frequency for the trans conformer (observed at 1002 cm-1) than the corresponding mode for
the cis form (observed at 983 cm-1) but the 19 cm-1 difference is smaller than the predicted
value. Similar accurate predicted differences were also observed for both the in- and out-ofplane NCO bends of the cis and trans conformers (Fig. 18). Therefore, we believe the
proposed assignment for the trans conformer is strongly supported by the ab initio predicted
frequencies even though the potential well for this form may be very shallow.
There is also clear evidence that the barrier to nearly free rotation is very small and a
significant number of the molecules at ambient temperature are in this state. For example,
from the infrared spectrum of the gas (Fig. 17) there is extensive increase of the band width
which is believed to be due to the rotation of the NCO moiety. This breadth is not observed
107
in the spectra of the rare gas solutions. Also for the heavy atom vibrations where the
fundamentals for both the cis and trans conformers are observed with increasing temperature
results in the area between them to be fill up by absorption. These observations are best
explained by nearly free rotation of the NCO moiety.
An estimate can be made for this barrier from the predicted energy for the skew form
relative to the energy of the most stable form which gives a value of 100 cm -1. This low
value is consistent with the expected value based on the methyl barrier for methylisocyanate.
From the observed microwave splitting of the observed lines a barrier of 29 ± 5 cm-1 was
determined [63] which was in agreement with the value of 46 cm-1 from the torsional
frequency and the ab initio value of 49 cm-1 [111]. Thus, the barrier to rotation of the NCO
group is expected to be quite small. However, the barrier to methyl rotation has been
determined to be between 1414 cm-1 (16.92 kJ/mol) by utilizing only the fundamental
torsional frequency or a higher value of 1546.2 ± 14.2 cm-1 (18.50 ± 0.17 kJ/mol) when the
frequency of the first “hot band” is included in the calculation. This large value is the result
of a rather unrealistically large V6 term (45.8 ± 6.9 cm-1). Nevertheless, the predicted barrier
from the ab initio calculations have similar values of 1520 cm-1 (MP2(full)/6-31G(d,p)) and
1482 cm-1 (MP2(full)/6-311+G(d,p)) for the cis conformer where as slightly smaller values
were predicted for the trans conformer.
The reported r0 values (Table 19) are probably as accurate as can be determined for this
molecule because of the large vibrational amplitude.
The parameter with the greatest
uncertainty is the C-C distance which depends significantly on the value of the A rotational
constant which probably has significantly larger uncertainty than the listed value obtained
from the assigned microwave transitions. Also, it would be of interest to obtain values for
108
several other C=N distances for molecules which have no large amplitude vibrations and
compare them to the ab initio predicted values. The other r0 parameters for ethylisocyanate
all seem normal when compared to the corresponding parameters for similar molecules.
Since the reported [118] microwave rotational constants for ethylisocyanate could not
be used to fit any reasonable parameters for the gauche conformer which was predicted from
the ab initio calculations [61, 92] as the stable form for this molecule, it was concluded that
“the microwave analysis, based on first order centrifugal distortion, must be inadequate for
this molecule”.
Therefore, we have calculated the centrifugal distortion constants for
ethylisocyanate for comparison to the experimentally determined [116] values for DJ and DJK.
Since these quantities depend on the force constants which only slowly vary with increased
basis set size, we initially predicted the distortion constants from MP2(full)/6-31G(d)
calculations. The values are listed in Table 20 along with the experimentally determined
values.
Since only two of the normally determined five constants were obtained it is
expected that the errors on the experimental values maybe over optimistic. Therefore, we redetermined the centrifugal distortion constants by utilizing the previously reported [118]
frequencies for the thirty-five assigned transitions. Since the values for K were relatively
limited, we used the ab initio predicted values for DK and d2 and obtained the values for DJ,
DJK and d1. The predicted values for the cis conformer was used for the DK and d2 values and
the differences for these two possibilities have little effect on the values of the other two
experimentally determined values. It should be noted that the ab initio predicted values for
DJ and DJK are in satisfactory agreement with the experimentally determined values.
In the recent theoretical investigation on the structures of azido, isocyanato, and
isothiocyanato derivatives of methane and its derivatives [61, 92], calculations at the MP2
109
level with basis set TZVP for ethylisothiocyanate was reported to have the gauche conformer
to be the stable form with a CCNC dihedral angle of 61.3 and both the cis and trans forms
as saddle points. From the graphical information provided [61, 92], the trans barrier is
predicted as 190 cm-1 and the cis barrier to be 40 cm-1. This potential function is very similar
to the one predicted from MP2(full)/6-311G(d,p) calculations except the cis barrier is 12 cm1
which increases to 52 cm-1 with addition of diffuse functions. Thus, the depth of the
gauche well is very dependent on the basis set. However, with basis sets 6-311G(2d,2p) and
higher the gauche well disappears and the cis saddle point becomes the potential minimum
and the only stable conformer contrary to what was suggested could not happen [61], but, in
fact, it does happen! Therefore in the earlier study of ethylisothiocyanate the authors [61] did
not use sufficiently sized basis sets to obtain the potential consistent with the experimentally
determined results. Therefore, calculations with much larger basis sets were carried out for
ethylisocyanate to see if the prediction of the gauche conformer would eventually disappear.
In Fig. 20 the predicted potential for the rotation of the NCO group is shown from
MP2(full)/6-31G(d,p) and MP2(full)/6-311+G(2d,2p) calculations and the barrier at the cis
position is only 11 cm-1. Thus, such a shallow gauche well could not accommodate a
vibrational state for a fundamental with a frequency of 25 cm-1 or higher. The predicted
frequency for the NCO torsion is 34 cm-1 from the larger basis set which shows that there
would be no bound vibrational energy level with gauche well.
Therefore, the ground
vibrational level for the rotational of the conformer with the lowest energy would be the cis
form and the observation of three torsional excited states is consistent with the cis form. A
plot of the reported bound rotational constant values verses vibrational quantum number
indicates a small barrier since they do not give a straight line. However, the trans well with a
110
depth of 24 cm-1 could have a ground state level for the predicted fundamental of 31 cm-1.
Therefore it should be possible to observe and assign the microwave spectrum of the trans
form by FT-microwave studies. Also more accurate determination of the rotational constants
should make it possible to verify the small barrier at the cis position.
In a study [112] where the equilibrium conformation of ethylisocyanate was revisited it
was concluded that the experimentally determined rotational constants could not be used to
determine the conformer based on ab initio predicted value or those estimated from predicted
structural parameters. However, this is probably true from MP2/6-31G(d) calculations but
predictions from MP2(full)/6-311+G(d,p) which have been shown to give relatively good
structural parameters for substituted hydrocarbons it is probably not true. The data given in
Table 21 indicated clearly that the reported microwave [118] data are for the cis form.
Irrespective of whether this method can be used to distinguish among conformers, there was
more compelling data which showed the heavy atoms lay on a plane of symmetry. For the
planar molecule, there will be only hydrogen atom off the plane so I = Ia + Ib – Ic will be
equal to 8mHC 2H where mH is the mass of the hydrogen and C 2H is the distance from the plane.
The experimental value was determined to be 6.593 amu Å2 where the predicted value was
6.344 amu Å2.
If the molecule was not planar then the heavy atoms should have a
pronounced effect to indicate a non-planar molecule. Also, reasonable structural parameters
could not come close to give values for the rotational constants of the gauche form. Finally,
the experimentally determined dipole moment components also indicate a planar molecule.
Thus, the suggestion that microwave data can be understood by averaging is difficult to
phantom.
111
Based on the results of the study reported herein, we believe that the suggestion [61,
92] that the A rotational constant observed from the microwave study [118] of
ethylisocyanate may not be correct since first order centrifugal distortion was used is
probably not the problem.
We suspect that the actual problem is with the ab initio
calculations which are not predicting the correct cis conformer that which is giving rise to the
microwave data. For example, the larger basis set sizes could significantly reduce the cis
barrier so the effective ground state is the cis conformer with a possible gauche well so
shallow that no ground vibrational state could be accommodated. Thus, further ab initio
calculations are probably needed for these type of molecules when the microwave results
(experimental data) do not agree with the theoretical predictions for this class of compounds.
112
Table 14. Observed and calculated frequencies (cm-1) for cis ethylisocyanate.
Vib.
Approx. Description
No.
ab
fixed
initioa scaledb
IR
int.c
Raman dp
act.d ratio
gas
Xee
IR
Kre
solid
2985
2942
2927
2282
1477
1473
1437
1381
1349
1090
983
804
2988
2946
2912
2301
—
—
—
1376
1347
1092
984
793
P.E.D.f
A*
B*
100S1
92
99S2
55
99S3
38
98S4
87
88S5
78
82S6
100
60S7, 18S14
90
88S8
33
84S9
100
51S10, 17S14, 13S11, 12S12 70
38S11, 28S12, 15S10, 11S7 88
27S12, 22S7, 20S11, 17S14, 74
14S10
83S13, 13S14
14
32S14, 24S11, 23S12
52
91S15
84
8
45
62
13
22
0
10
67
0
30
12
26
A
1
2
3
4
5
6
7
8
9
10
11
113
12
13
14
15
CH3 antisymmetric stretch
CH2 symmetric stretch
CH3 symmetric stretch
NCO antisymmetric stretch
CH2 deformation
CH3 antisymmetric deformation
NCO symmetric stretch
CH3 symmetric deformation
CH2 wag
CH3 rock
CCN symmetric stretch
CCN antisymmetric stretch
NCO in-plane bend
CCN bend
CNC bend
3208
3125
3114
2408
1579
1563
1488
1470
1425
1147
1027
835
3009
2932
2922
2286
1481
1466
1422
1395
1352
1095
974
798
17.8
33.4
8.7
796.3
0.5
0.9
6.4
14.4
37.7
18.3
14.5
19.6
71.6
110.5
111.9
1.1
7.8
17.1
26.5
9.8
9.5
4.6
8.0
10.1
0.68
0.11
0.02
0.04
0.63
0.70
0.29
0.48
0.49
0.32
0.74
0.12
2993
2982
2941
2939
2913
2928
2283
2282
~1481 (1473)
1473 1473
1437
1437
1384
1380
1351
1348
1091
1089
983
983
801
801
631
432
119
631
419
119
29.1
8.1
4.8
0.6
0.6
2.5
0.51
0.53
0.68
638
418
122
638
422
—
638
422
—
609
427
—
3216
3171
1555
1347
1202
826
558
298
—
3017
2975
1459
1278
1141
783
558
282
—
19.6
9.3
8.1
0.1
1.9
1.0
15.6
1.3
4.6
16.5
106.4
17.7
11.3
0.7
0.4
0.3
0.1
1.1
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
2993
2980
1451
—
1140
—
585
265
34
2982
2982
1472
1282
1135
797
~600
—
—
2985
2952
1472
1284
1137
796
580
—
—
2988
2974
1470
1278
1139
786
574
264
—
86
48
16
A
16
17
18
19
20
21
22
23
24
CH3 antisymmetric stretch
CH2 antisymmetric stretch
CH3 antisymmetric deformation
CH2 twist
CH2 rock
CH3 rock
NCO out-of-plane bend
CH3 torsion
NCO torsion
95S16
94S17
93S18
83S19, 15S21
50S20, 36S21
41S21, 48S20
100S22
100S23
100S24
aMP2/6-31G(d) predicted values. bMP2/6-31G(d) fixed scaled frequencies with factors of 0.88 for CH stretches, 1.0 for heavy atom bends and CN torsion and 0.90 for all other modes. cScaled infrared intensities in km/mol from
dScaled Raman activities in Å4/u from MP2/6-31G(d).eFrequencies from mid-IR spectrum of Xe solution at -85 C and Kr solution at -150 C. fCalculated with MP2/6-31G(d) and contributions of less
MP2/6-31G(d) .
than 10% are omitted. *A, B values in the last three column are percentage infrared band contours, entries with bars are symmetry forbidden., referring to 100% C
Table 15. Observed and calculated frequencies (cm-1) from the MP2/6-31G(d) level for trans ethylisocyanate.
A
114
A
Vib. Approx. Description
No.
1 CH3 antisymmetric stretch
2 CH2 symmetric stretch
3 CH3 symmetric stretch
4 NCO antisymmetric stretch
5 CH2 deformation
6 CH3 antisymmetric deformation
7 NCO symmetric stretch
8 CH3 symmetric deformation
9 CH2 wag
10 CH3 rock
11 CCN symmetric stretch
12 CCN antisymmetric stretch
13 NCO in-plane bend
14 CCN bend
15 CNC bend
16 CH3 antisymmetric stretch
17 CH2 antisym. stretch
18 CH3 antisym. deformation
19 CH2 twist
20 CH2 rock
21 CH3 rock
22 NCO out-of-plane bend
23 CH3 torsion
24 NCO torsion
ab
initio
3212
3107
3120
2404
1559
1571
1484
1473
1421
1155
1060
830
619
395
134
3224
3160
1553
1347
1191
824
571
280
31
fixed
IR
scaleda
int.b
3013
18.7
2915
28.7
2927
12.4
2280 843.0
1479
6.4
1491
5.2
1407
59.6
1397
5.7
1347
66.2
1107
7.9
1009
12.9
803
11.1
619
23.1
379
20.4
134
6.3
3024
17.7
2965
14.9
1473
6.4
1278
0.006
1129
2.6
782
0.8
571
20.0
265
1.0
31
1.4
Raman
act.c
68.6
102.1
102.0
1.0
28.1
3.2
20.4
17.6
3.1
5.5
12.5
14.2
0.6
2.1
2.3
37.8
75.9
20.1
11.2
1.8
0.03
0.5
0.1
3.1
dp
ratio
0.74
0.07
0.01
0.09
0.53
0.51
0.19
0.39
0.61
0.19
0.62
0.26
0.75
0.41
0.73
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
Obs.
2300
1393
1002
806
617
385
P.E.D.d
A*
B*
99S1
34
100S2
20
99S3
86
98S4
100
60S5, 37S6
100
53S6, 39S5
50
33S7, 33S8, 20S12, 10S9 97
56S8, 31S7
4
81S9
99
52S10, 27S14
45
42S11, 27S12
13
42S12, 23S10, 12S11
71
89S13
55
18S14, 35S12, 29S11
99
88S15
100
98S16
97S17
93S18
78S19, 18S21
48S20, 34S21, 15S19
40S21, 51S20
99S22
98S23
100S24
66
80
14
0
0
50
3
96
1
55
87
29
45
1
0
C*
100
100
100
100
100
100
100
100
100
aFixed scaled frequencies with factors of 0.88 for CH stretches, 1.0 for heavy atom bends and CN torsion and 0.90 for all other modes. bScaled infrared
intensities in km/mol. cScaled Raman activities in Å4/u. d Contributions of less than 10% are omitted.*A, B and C values in the last three column are percentage
infrared band contours, entries with bars are symmetry forbidden.
Table 16. Calculated frequencies (cm-1) at the MP2(full)/6-31G(d) level for gauche ethylisocyanate.
115
Vib.
ab
fixed
IR Raman dp
Approx. Description
P.E.D.d
A* B * C *
No.
initio scaleda
int.b
act.c ratio
CH3 antisymmetric stretch
3207 3008
22.2
62.5 0.65 90S1
81 19
0
1
CH
symmetric
stretch
3122
2929
33.5
98.2
0.12
91S
37
36
27
2
2
2
CH3 symmetric stretch
3115 2922
10.2 109.3 0.02 99S3
45 55
0
3
NCO
antisymmetric
stretch
2399
2276
798.6
0.9
0.03
98S
89
10
1
4
4
CH2 deformation
1580 1498
2.9
6.1 0.60 86S5, 11S6
68 15 17
5
CH3 antisymmetric deformation
1563 1482
1.6
22.4 0.73 80S6, 13S5
95
2
3
6
NCO
symmetric
stretch
1484
1416
13.5
21.0
0.22
60S
,
14S
,
11S
84
1
15
7
14
12
7
CH
symmetric
deformation
1470
1395
13.9
7.1
0.51
87S
53 47
0
3
8
8
CH2 wag
1426 1353
36.1
8.6 0.49 83S9
98
0
2
9
CH
rock
1145
1094
12.5
5.9
0.17
51S
,
19S
,
12S
62
30
8
3
10
14
11
10
1032 979
15.1
8.2 0.74 40S11, 29S12, 12S10, 11S7
87
6
7
11 CCN symmetric stretch
839 802
12.3
11.0 0.15 20S12, 20S14, 19S7, 18S11, 15S10
62 18 20
12 CCN antisymmetric stretch
NCO
in-plane
bend
627
627
24.1
0.5
0.57
84S
,
11S
21 61 18
13
14
13
428 414
5.1
0.5 0.34 32S14, 27S12, 25S11
6
91
3
14 CCN bend
146 145
9.8
1.7 0.67 87S15
83
6 11
15 CNC bend
CH
antisymmetric
stretch
3220
3020
18.5
27.5
0.72
85S
1
3 96
3
16
16
3182 2985
7.1 100.9 0.69 82S17
15 26 59
17 CH2 antisymmetric stretch
1554 1474
8.2
17.8 0.75 92S18
14 10 76
18 CH3 antisymmetric deformation
CH
twist
1344
1275
6.1
11.6
0.75
78S
,
16S
86 11
3
2
19
21
19
1203 1142
16.2
1.8 0.73 49S20, 35S21, 10S19
71 15 14
20 CH2 rock
CH
rock
827
786
4.7
0.7
0.31
38S
,
44S
62
1 37
3
21
20
21
573 573
21.0
0.4 0.71 98S22
8 37 55
22 NCO out-of-plane bend
296 282
3.5
0.2 0.73 93S23
73
1 26
23 CH3 torsion
NCO
torsion
50
50
1.6
2.4
0.72
99S
7
15
78
24
24
a Scaling factors: 0.88 for CH stretches, 1.0 for heavy atom bends and CN torsion and 0.90 for all other modes. b Scaled infrared intensities in
km/mol. c Scaled Raman activities in Å4/u. d Contributions of less than 10% are omitted. *A, B and C values in the last three column are
percentage infrared band contours.
Table 17. Calculated electronic energies (hartree) and energy differencesa (cm-1) for the
gauche, cis, trans, and the skew conformers of ethylisocyanate.
a
Method / Basis set
gauche
cis
trans
skew
MP2(full)/6-31G(d)
-246.570816
69
126
147
MP2(full)/cc-PVDZ
-246.612124
95
111
125
MP2(full)/6-31+G(d)
-246.584958
121
117
148
MP2(full)/TZVP
-246.792112
101
159
167
MP2(full)/6-311G(d,p)
-246.781351
112
129
139
MP2(full)/6-311+G(d,p)
-246.790243
124
144
163
MP2(full)/aug-cc-PVDZ
-246.661072
50
78
135
MP2(full)/6-311G(2d,2p)
-246.847291
25
92
96
MP2(full)/6-311+G(2d,2p)
-246.855279
11
63
87
MP2(full)/cc-PVTZ
-246.9177573
12
102
124
MP2(full)/6-311G(2df,2pd)
-246.935790
23
113
115
MP2(full)/6-311+G(2df,2pd)
-246.942755
15
97
112
MP2(full)/6-311G(3df,3pd)
-246.956261
31
69
89
MP2(full)/6-311+G(3df,3pd)
-246.962012
21
110
129
MP2(full)/aug-cc-PVTZ
-246.950887
27
43
139
MP2(full)/cc-PVQZ
-247.067289
19
105
―
B3LYP/6-31G(d)
-247.305324
-26
25
32
B3LYP/6-31+G(d)
-247.315950
1
-11
11
B3LYP/6-311G(d,p)
-247.377623
36
5
17
B3LYP/6-311+G(d,p)
-247.382854
16
-9
11
B3LYP/6-311G(2d,2p)
-247.385024
33
24
23
B3LYP/6-311+G(2d,2p)
-247.390390
-2
5
10
B3LYP/6-311G(2df,2pd)
-247.393844
27
15
16
B3LYP/6-311+G(2df,2pd)
-247.398812
-5
0
11
B3LYP/6-311G(3df,3pd)
-247.397065
17
-24
4
B3LYP/6-311+G(3df,3pd)
-247.401458
-11
Energy difference are relative to the gauche conformer.
-6
15
116
Table 18. Temperature and intensity ratios of the conformational bands of ethylisocyanate
from the infrared spectra of the liquid krypton of the gas.
Liquid krypton
T( C)
1/T ( 10-3 K-1)
I983 / I1002
I796/ I1002
110.0
5.950
-1.2722
-0.4862
115.0
6.129
-1.2931
-0.5198
120.0
6.323
-1.3503
-0.5537
125.0
6.529
-1.4006
-0.5799
130.0
6.750
-1.4472
-0.6221
135.0
6.986
-1.4814
-0.6624
140.0
7.238
-1.5153
-0.6945
145.0
7.510
-1.5434
-0.7382
150.0
7.803
-1.5661
-0.7755
-155.0
8.120
-1.5864
-0.8408
a
98
a
8 cm-1
103
2
4 cm-1
(1.20
0.05 kJ/mol) with the cis conformer more stable.
117
Table 19. Structural parameters (Å and degree), rotational constants (MHz) and dipole moments (debye) for the cis and trans
conformers of ethylisocyanate from ab initio calculations (6-311+G(d,p) and experimental data.
118
cis
trans
Parameter
MP2(full)
B3LYP
MP2(full)
B3LYP
MW [118]
r(C–C)
1.522
1.527
1.519
1.524
1.520
r(C–N)
1.449
1.448
1.453
1.453
1.461a
r(C=N)
1.213
1.198
1.214
1.200
1.207
r(C=O)
1.181
1.175
1.180
1.174
1.171
r(C–H) methylene
1.093
1.094
1.095
1.095
1.091
r(C–H) methyl o.p.
1.093
1.093
1.092
1.092
1.091
r(C–H) methyl i.p.
1.093
1.093
1.093
1.093
1.091
112.9
113.7
109.9
110.9
109.14a
CCN
136.5
141.4
135.1
140.2
142.11a
CNC
172.9
173.9
172.8
174.0
180.0
NCO
110.6
110.5
110.3
110.2
109.6
CCH methylene
107.6
107.5
109.2
108.9
NCH methylene
107.4
106.7
108.0
107.6
108.8
HCH methylene
110.5
111.0
110.4
110.8
109.6
CCH methyl o.p.
110.1
109.9
109.9
109.9
109.6
CCH methyl i.p.
108.5
108.4
108.6
108.3
108.8
Hi.pCHo.p. methyl
108.6
108.2
108.7
108.5
108.8
Ho.pCHo.p. methyl
A
13929.5
14753.7
36054.1
35885.1
14110(3)
B
3070.8
2937.5
2249.4
2237.6
3067.01(2)
C
2597.6
2526.6
2174.9
2163.1
2605.01(2)
3.402
3.276
1.932
1.596
2.81(2)
| a|
0.859
0.785
2.707
2.788
0.03
| b|
| c|
3.508
3.369
3.326
3.213
| tot|
a
Parameters determined from three rotational constants with all other parameters assumed.
ED [117]
1.518(11)
1.455(9)
1.221(5)
1.179(4)
1.090(7)
1.090(7)
1.090(7)
113.4(17)
129.2(21)
197.0(34)
110
110
110
13780
Adjusted r0
1.516(5)
1.448(5)
1.211(5)
1.167(5)
1.093(2)
1.093(2)
1.093(2)
112.6(5)
137.5(5)
172.9 (5)
110.6(5)
107.6(5)
107.4(5)
110.5(5)
110.1(5)
108.5(5)
108.6(5)
14180.8
3067.9
2604.1
Table 20. Rotational (MHz) and centrifugal distortion (kHz) constants for cis
ethylisocyanate.
MP2/6-31G(d)
MP2/6-311+G(d,p)
MW [118]
A
14315.3
13930
14110.00(3)
B
2992.4
3071
3067.01(2)
C
2553.7
2598
2605.01(2)
DJ
5.022
3.76
4.17(22)
DJK
-85.86
-53.80
-39.1(5)
DK
872.06
445.82
d1
-1.716
-1.217
d2
-0.144
-0.097
119
Table 21.
a
b
Calculateda and experimentalb Rotational Constants (MHz) of ethylisocyanate.
A
B
C
Experimental
14110
3067
2605
cis
13929
181
3071
4
2598
7
gauche
15018
908
2944
156
2554
59
trans
36054
1944
2249
818
2175
460
Obtained from MP2(full)/6-311+G(d,p) calculations.
Ref. [118]
120
Figure 14.
Newman projections for the four different possible conformers of ethylisocyanate.
121
Figure 15.
Far-infrared spectra of ethylisocyanate: (A) gas; (B) crystalline solid.
122
Figure 16.
Predicted (MP2(full)/6-311+G(d,p)) and observed infrared spectra of
ethylisocyanate: (A) observed spectrum the sample dissolved in liquid krypton
at 155 C; (B) predicted spectrum of the mixture of cis and trans conformers
with H = 100 cm-1; (C) predicted spectrum of pure trans conformer; and (D)
predicted spectrum of pure cis conformer.
123
124
Figure 17.
Infrared spectra of ethylisocyanate: (A) gas; (B) xenon solution at -85 C; (C) krypton solution at -155 C.
Figure 18.
Low-frequency Raman spectra of ethylisocyanate: (A) liquid; (B) amorphous
solid; (C) annealed solid.
125
Figure 19.
Temperature dependence of the 983 cm-1 (cis) and the 1002 cm-1 (trans)
infrared bands of ethylisocyanate dissolved in liquid krypton.
126
127
Figure 20.
Potential energy curve of ethylisocyanate as a function of the dihedral angle C=N C C2 obtained from
MP2(full)/6-31G(d,p) calculations (dashed line) and MP2(full)/6-311+G(2d,2p) calculations (solid line).
CHAPTER 7
MICROWAVE, RAMAN, AND INFRARED SPECTRA, R0 STRUCTURAL
PARAMETERS, CONFORMATIONAL STABILITY AND AB INITIO CALCULATIONS
OF ISOPROPYLISOCYANATE
Introduction
The substitution of isocyanate (NCO) or isothiocyanate (NCS) for the hydrogen atom
on simple hydrocarbons has resulted in a number of small molecules where there has been
considerable controversy regarding the most stable conformation or conformations. There
have been frequent discrepancies observed among the experimental conformational
determinations from microwave spectra, vibrational spectra, and electron diffraction studies
[120-122]. More recently, predicted structures have been reported [92, 123] from ab initio
calculations, which do not agree in some cases with the experimentally determined structures.
These originate from several factors with the most significant one from the large CNC angles
which may become linear with very few vibrations or the potential well may be so shallow
that it cannot accommodate a single vibrational state.
Also, the low frequency large
amplitude vibrations can result where a large number of molecules are linear in the
vibrational excited states at ambient temperature which makes it difficult to determine the
structure of the ground vibrational state. Sometime ago, we initiated the study of several
small hydrocarbons with monosubsituted NCO and NCS groups to determine their
conformational stabilities, the enthalpy differences between or among the conformers, and
their structural parameters. For example, the most recent molecule we investigated was
cyclopropylisocyanate, c-C3H5NCO, where the two stable conformers were clearly
128
determined to be the trans and cis conformers with the trans form the more stable conformer
with an enthalpy difference of 77
8 cm-1 (0.92
0.10 kJ/mol). This result showed
conclusively that the earlier ab initio prediction4 that the stable forms were two gauche
conformers with dihedral C=NC C angles of 144.9° and 324.3° relative to the two carbon
atoms in the ring was undoubtedly wrong. Similarly, from the fact that the centrifugal
distortion constants could not be obtained from the simple quadratic equation it was possibly
taken [124] to indicate that the second conformer may not be the cis form. However, this
failure is due to the large amplitude vibration which must be governed by a potential function
frequently referred to as a “champagne bottle” potential function [110] where the origin of
the coordinate system is at the critical point on top of the potential hump .
As a continuation of these studies we have investigated the conformational stabilities
of isopropylisocyanate and recorded the FT-microwave spectra to determine the structural
parameters. In an earlier microwave study [125] of isopropylisocyanate the individual lines
could not be assigned but the low resolution spectrum provided the value for the sum of the
B + C rotational constants. By using reasonable structural parameters for their values a good
agreement for the skew form was obtained which was significantly different from the
predicted B + C values for the trans, gauche, and cis forms.
Similar conformational
conclusion for this molecule was also obtained from an electron diffraction study [126] with
the conclusion from both structural studies that there was only one form present. Also from
the vibrational study, it was concluded that there was primarily one conformer present since
none of the skeletal modes which should be the most sensitive for conformer determination
exhibited any doubling of frequencies in the fluid phases. Also, the lack of any bands
observed in the infrared spectrum of the gas which disappeared in the infrared spectrum of
129
the solid was further taken as an indication that there was predominant evidence for mainly
only one conformer present.
From our more recent studies of organoisocyanates molecules this single conformer
conclusion is expected to be in error. Therefore, the microwave spectrum has been reinvestigated by a FT-microwave instrument at 4º K at low frequencies so individual lines
could be assigned as well as a search could be made for a second conformer. In addition to
the microwave studies we have investigated the variable temperature Raman and far infrared
spectra of xenon solutions with the expectation of identifying vibrations of a second
conformer. To support these experimental studies we have carried out MP2(full) ab initio
and density functional theory (DFT) calculations by the B3LYP method by utilizing a variety
of basis sets.
The results of these spectroscopic, structural, and theoretical studies of
isopropylisocyanate are reported herein.
Experimental Methods
The isopropylisocyanate sample was purchased from Aldrich, St. Louis, MO with a
stated purity of 99% and it was utilized without further purification.
The microwave spectra of isopropylisocyanate were recorded using a “mini-cavity”
Fourier-transform microwave spectrometer [53, 54] at Kent State University.
The
isopropylisocyanate sample was entrained in 70:30 Ne-He carrier gas mixture at 2 atm and
expanded into the cavity using a reservoir nozzle [54] made from a modified Series-9
General Valve. The reservoir nozzle is mounted in a recessed region of the mirror flange,
external to the vacuum chamber, and the expansion passes through a 0.182-inch diameter
hole into the resonant cavity. The center of the expansion is offset from the center of the
mirror by 1 inch.
130
The sample was irradiated by microwave radiation generated by an Agilent
Technologies E8247C PSG CW synthesizer and details of the irradiation and heterodyne
detection circuitry can be found in Ref. [53]. The vacuum system can accommodate pulse
repetition rates of up to 15 s-1 while maintaining a pressure below 10-4 torr, and the
instrument can scan 450 MHz in 6 hours while averaging 100 shots per scan segment. The
frequencies for the measured transitions in the region of 12,000 to 19,000 MHz for the trans
conformer (Fig. 21) of isopropylisocyanate are listed in Table 22 along with their
assignments. Also listed are the frequency differences between the measured values and the
values obtained from the determined rotational constants and the centrifugal distortion
constants (Table 23).
The mid-infrared spectra from 3500 to 400 cm-1 of the gas (Figs. 22A and 23A) was
recorded on a Perkin-Elmer model 2000 Fourier transform spectrometer equipped with a
nichrome wire source, Ge/CsI beamsplitter and DTGS detector. The spectrum of the gas was
obtained with the sample contained in 12 cm cell equipped with CsI windows. Atmospheric
water vapor was removed from the spectrometer chamber by purging with dry nitrogen.
Interferograms were obtained after 128 scans for the gas sample and the background
reference were transformed by using a boxcar apodization function with theoretical
resolutions of 0.5 cm-1.
The mid-infrared spectra of isopropylisocyanate dissolved in liquefied xenon (Fig. 23B)
were recorded on a Bruker model IFS-66 Fourier interferometer equipped with a Globar
source, Ge/KBr beamsplitter and DTGS detector. The interferograms were recorded at
variable temperatures ranging from 55 to 100 C with 100 scans and transformed by a
Blackman-Harris apodization function with a theoretical resolution of 1.0 cm-1.
131
The
temperature studies in liquefied xenon were carried out in a specially designed cryostat cell,
which is composed of a copper cell with a 4 cm path length and wedged silicon windows
sealed to the cell with indium gaskets. The temperature was monitored by two platinum
thermoresistors and the cell was cooled by the vapors from boiling liquid nitrogen. All of the
observed fundamental modes for the trans and gauche conformers in the infrared and Raman
spectra are listed in Tables 24 and 25, respectively.
The low frequency and far infrared spectra (60 to 600 cm-1) of the sample dissolved in
liquid xenon were recorded on a Bruker model IFS-66 v/S Fourier transform
spectrophotometer equipped with a Globar source, a 6.0 μm Mylar beamsplitter, and a liquid
helium cooled Si bolometer. The sample was contained in a 7 cm cell fitted with Si
Windows and the sample added as described for the mid-infrared studies. For all spectra,
250 interferograms were collected at 0.5 cm-1 resolution, averaged and transformed with a
Blackman-Harris three term function.
Raman spectra were recorded using a Trivista 557 spectrometer consisting of a double f
= 50 cm monochromator equipped with a 2000 lines mm−1 grating, a f = 70 cm spectrograph
equipped with a 2400 lines mm−1 grating, and a back-illuminated LN2-cooled PI Acton Spec10:2 kB/LN 2048 × 512 pixel CCD detector. For all experiments, the 514.5 nm line of a
2017-Ar S/N 1665 Spectra-Physics argon ion laser was used for Raman excitation, with the
power set to 0.8 Watt. Signals related to the plasma lines were removed by using an
interference filter. The frequencies were calibrated by using Neon emission lines, and
depending on the set-up used, are expected to be accurate within 0.4 cm−1. The experimental
set-up [127, 128] used to investigate the solutions has been described. A home-built liquid
132
cell equipped with four quartz windows at right angles was used to record the spectra (Fig.
24).
Ab Initio Calculations
The LCAO-MO-SCF calculations were performed with the Gaussian-03 program [43]
by using Gaussian-type basis functions.
The energy minima with respect to nuclear
coordinates were obtained by simultaneous relaxation of all geometric parameters consistent
with symmetry restrictions using the gradient method of Pulay [44]. A number of basis sets
starting from 6-31G(d), and increasing to 6-311+G(3df,3pd), were employed at the level of
Møller-Plesset perturbation theory [38] to second order (MP2), as well as hybrid density
functional theory by the B3LYP method, to obtain energy differences (Table 26) among the
four most likely conformers of isopropylisocyanate.
To aid in making the vibrational assignment (Tables 24 and 25), we have carried out a
normal coordinate analysis by utilizing the force fields obtained from the Gaussian-03
program at the MP2(full)/6-31G(d) level. The internal coordinates used to calculate the G
and B matrices for isopropylisocyanate are listed along with the structural parameters in
Table 27. By using the B matrix, the force field in Cartesian coordinates was converted to a
force field in internal coordinates.[45] Subsequently, scaling factors of 0.88 for the CH
stretches and 0.90 for all other modes were used, along with the geometric average of scaling
factors for interaction force constants, to obtain the fixed scaled force fields and the resultant
wavenumbers (Tables 24 and 25). A set of symmetry coordinates (Table 28) was used to
determine the corresponding potential energy distributions (P.E.D.s). The observed and
133
calculated wavenumbers of isopropylisocyanate along with the calculated infrared intensities,
Raman activities, depolarization ratios, and P.E.D.s are given in Tables 24 and 25.
To aid in the identification of the fundamental vibrations for possible conformers of
isopropylisocyanate, the infrared and Raman spectrum have been simulated (Figs. 22 and 24,
respectively) from the scaled ab initio MP2(full)/6-31G(d) calculation. Infrared intensities
were calculated based on the dipole moment derivatives with respect to Cartesian coordinates
and transformed into normal coordinate [129]. The evaluation of Raman activity by using the
analytical gradient method [46, 47] has been developed.
Results
Since the previous spectroscopic studies indicated that at ambient temperature there
was predominately a single conformers present for isopropylisocyanate, it was important to
determined which of the four possible conformers was the one giving rise to the observed
microwave spectra. With this determination, it should be possible to provide a confident
vibrational assignment that should provide sufficient information to identity the second
conformer if it was present. Thus, the first task was to assign the microwave spectrum.
Microwave Spectra
To assign the microwave spectrum, preliminary rotational constants were predicted
for the trans, gauche, skew and cis conformer from MP2(full)/6-311G(d,p) calculations and
the A, B and C values are listed in Table 23 for the two most stable conformers. The
previously reported [125] B+C MHz average value was 4031 MHz which is smaller than the
ab initio predicted value of 4254 MHz for the trans form but closer to the value of 3978 MHz
predicted for the skew form. However, there was a much weaker series neglected that had a
spacing of approximately ~4300 MHz which was closer to the predicted value for the trans
134
form (Fig. 25). Assignments were then made for several transitions from which A, B and C
constants were obtained. The dipole moment components were predicted to be | a| = 3.430 D,
| b| = 0, and | c| = 0.297 D so the A-type transitions were expected to dominate with the Ctype transitions difficult to identify. The bands were quite difficult to assign and only three
“J” transitions with the quadrupole splitting were assigned which could be fitted without the
centrifugal distortion values. However, this is expected since a potential function consistent
with the “champagne bottle” model is needed to fit transitions with higher frequencies
particularly those with larger K values.
By utilizing the assigned transitions listed in Table 22, the rotational constant listed in
Table 23 were obtained with uncertainties to the hundredth MHz for the values B and C
constants and to a tenth MHz for the A constant. The quadrupole coupling constants were
also obtained from these data which are compared to those obtained from the MP2 and
B3LYP calculations (Table 23).
Structural Parameters
In the earlier microwave study, the estimated B+C values for the four possible
conformers was obtained by taking the structural parameters for the isopropyl moiety from
those reported [130] for isopropylchloride.
For the isocyanate group, the structural
parameters were estimated from an electron diffraction study [95] of methylisocyanate with
an elongation of the C -N distance by 0.02 Å due to replacement of the two hydrogen atoms
with methyl groups (Table 27). The C-N=C angle was varied from 138 to 147 for the
various conformers. In the present study, some parameters have to be assumed since there
are only three values of the rotational constants.
Since [33] all of the carbon-hydrogen
distances can be taken from the MP2(full)/6-311+G(d,p) predicted values for both the trans
135
and gauche conformers of isopropylisocyanate. It has been shown that triple bond distances
are nearly constant irrespective of the substitution on it [131] and the C=O of the NCO group
behaves similarly to a triple bond. Additionally, there have been several determinations of
the N=C(O) bond for a variety of molecules and it has been found to be approximately the
average of the predicted value from the MP2 and B3LYP calculations with the 6-311+G(d,p)
basis set and this value was used as the predicted value to start the adjustment. Finally, we
fixed the angle N=C=O at the predicted value of 172.6º from the MP2(full)/6-311+G(d,p)
calculation. We then adjusted the five remaining heavy atom parameters by utilizing the
predicted values from the MP2(full)/6-311+G(d,p) calculation and then minimized the
difference between the rotational constants obtained from the adjusted r0 parameters and the
experimental determined values. The determined parameters are listed in Table 27 along
with estimated values for the gauche conformer where the same difference from the adjusted
parameters of the trans conformer were applied to the gauche form. The rotational constants
from the adjusted r0 parameters of the trans conformer have differences of 0.06, 0.89, and
0.10 MHz for the A, B, and C constants, respectively.
Vibrational Assignment
In order to obtain the enthalpy difference between the two conformers it is necessary
to obtain confident vibrational assignments for the most stable conformer as well as many of
the fundamental modes of the second conformer particularly in the “finger print” spectral
region where the conformer pairs are best suit for the enthalpy determination.
The
conclusion that there was only one form present was supported by the electron diffraction
results [126] and the low resolution microwave studies [125] which were interpreted to
indicate the skew form.
Nevertheless, there were some evidences for a second conformer
136
since two of the vibrational modes at 767 and 650 cm-1 in the spectrum of the gas were
reported to have low frequency shoulders but none of the depolarized Raman lines had
corresponding bands in the Raman spectrum of the gas so the earlier reported vibrational
assignment [125] was made on the basis of the molecule having a “pseudo” plane of
symmetry. It was assumed that this “local” Cs symmetry was due to the C-N=C angle
approaching linearity or forming a very low barrier for the asymmetric torsion so the
assignment was made for 19 A + 14 A modes.
Therefore, it is expected that this
assignment will be very similar to the one which we have made for the most stable trans
form.
The assignment for most of the fundamentals for the isopropyl moiety could be made
from well known “group frequencies” except for some of the lower frequency CH3 bending
modes which are extensively mixed with the heavy atom motions. Similarly, the frequencies
for the NCO fundamentals are well known but because of the large amplitude motions of
NCO group these vibrations give rise to extensively broad bands without any contours which
can be used to distinguish the B-contours of the out-of-plane vibrations.
Therefore,
considerable reliance was placed on the ab initio predicted fundamental frequencies for
distinguishing between the two conformers. Additionally, the predicted intensities of the
infrared bands was quite helpful along with some of the predicted depolarization values for
the Raman data as well as the activity. However, the most useful information for assigning
the fundamentals for the second conformer were the data (Fig. 23B) from the xenon solution.
By using all of these data the vibrational assignments listed in Tables 24 and 25 for the trans
and gauche conformers, respectively, were made.
Conformational Stability
137
To determine the enthalpy difference between the trans and the gauche conformers,
variable-temperature studies in liquid xenon were carried out. Because only small
interactions are expected to occur between the dissolved molecules and the surrounding
noble gas atoms, [57, 117-119] the “pseudo gas phase” spectrum shows only small frequency
shifts compared with the spectrum of the gas. A significant advantage of this type of
cryogenic spectroscopic study is that the conformer bands are better resolved in comparison
with those in the spectrum of the gas. This is particularly important because most of the
conformer bands for this molecule were expected to be observed within a few wavenumbers
of each other. The different conformer bands that are expected to be clearly identified are
those in the spectral region below 1000 cm-1 since they have the lowest probability of having
intensity contributions from overtones or combination bands.
The sample was dissolved in liquefied xenon solution and the Raman spectra were
recorded at six different temperatures.
Of the several fundamental modes exhibiting
conformer doublets, the 902 cm-1 (trans) and 943 cm-1 (gauche) pair (Fig. 26) was initially
chosen for the conformational enthalpy determination due to their relatively even baseline,
satisfactory band separations, and relatively strong band intensities. The relative intensities
of this conformer pair were measured as a function of temperature and their ratios were
determined. Six sets of spectral data were obtained for the pair and by application of the
van’t Hoff equation, lnK = H/(RT)
S/R, the enthalpy difference was determined from a
plot of lnK versus 1/T, where H/R is the slope of the line and K is substituted with the
appropriate intensity ratios, i.e. Itrans/Igauche. It was assumed that H, S and the ratio of the
molar absorption coefficients εtrans/εgauche are not a function of temperature in the range
studied. The conformational enthalpy difference was determined to be 131 ± 15 cm -1 (1.57
138
0.18 kJ/mol) from the xenon solution for this pair. One other pair of fundamentals was also
utilized (755 trans/943 gauche), from which a second
H value was obtained. The two
individual values are listed in Table 29 and the statistical average was obtained by treating
the data as a single set which gave a value of 115
13 cm-1 (1.37
0.15 kJ/mol).
We also carried out a variable temperature study in xenon solution by utilizing infrared
spectra in the lower frequency region. The two lines at 407 and 349 cm-1 of the gas are the
only well resolved conformer doublet where the relative intensities could be measured as a
function of temperature in the xenon.
Five sets of spectral data were obtained at
temperatures varying from 60 to 100°C (Table 29). From this conformer pair, a H value
of 109 ± 13 cm-1 (1.30
0.15 kJ/mol) was determined. Additionally, one other conformer
pair from the mid-infrared spectra was obtained for ΔH determination.
For this conformer
pair, intensity ratio of I757t/I742g were collected from 9 different temperatures at 5° interval
starting at 60 °C, and an enthalpy difference of 121 ± 9 cm-1 (1.45
obtained.
0.11 kJ/mol) was
Unfortunately, most of the conformer bands were separated by very few
wavenumbers so it was not possible to obtain another conformer pair from the infrared
spectra.
The four individual ∆H values are listed in Table 29 and the statistical average with
standard deviation of one sigma was obtained by treating all the data as a single set which
gives a value of 115
7 cm-1 (1.38 ± 0.08 kJ/mol). Although the statistical uncertainty is
relatively small, it does not take into account possible contribution from combination or
overtone bands from the other conformer contributing to the measured fundamental band
intensities. The variations of ∆H values are undoubtedly due to these types of interferences.
139
Therefore, a more probable uncertainty is 10% which gives 115 ± 11 cm-1 (1.38
0.13
kJ/mol) for ∆H.
Discussion
From the earlier vibrational study it was suggested [125] that there is only one stable
skew conformer and that there were little indications of a second conformer.
However, it
was noted that the vibrational data indicates that the isopropylisocyanate has a plane of
symmetry. In the current study, the most stable conformer has been identified as the trans
form by both vibrational and rotational data with a second stable gauche form. Since the
previous vibrational assignment has been made for a single conformer (skew), it is expected
to be similar to the assignments made for the trans conformer. However, some of these
fundamentals have been re-assigned for the gauche conformer in the infrared spectrum of the
solid i.e. peaks at 1355, 1344, 1325, and 456 cm-1. The band at 1355 cm-1 was previously
assigned as a combination band (ν18 + ν27) of 257 and 1134 cm-1, respectively, is now
assigned to the gauche form as ν15, CH out-of-plane bend.
The previously listed but
unassigned band at 459 cm-1, which corresponds with the 456 cm-1 in the current study, is
clearly a gauche fundamental for the CC2 wag. The CC2 wag fundamental was predicted to
be separated by 22 cm-1 between the trans (478 cm-1) and gauche (456 cm-1) conformer and,
as predicted, two bands are observed at 477 and 456 cm-1. Some of the bands currently
observed in the infrared spectrum of the solid for the gauche form that were not utilized at all
in the previous assignment included bands at 1419, 1130, 1119, 907 and 417 cm-1. The 1419
cm-1 (ν13, CH3 symmetric deformation) and 1119 cm-1 (ν19, CC2 antisymmetric stretch) was
clearly observed in the infrared spectra of the solid unlike the other three bands which
appeared as shoulders on the much stronger corresponding trans fundamentals. The band at
140
1130 and 907 cm-1 have similar spectral characteristics like that of the 417 cm-1 band. In the
earlier assignment, only one band was listed at 413 cm-1 (CC2 twist) in the infrared spectrum
of the solid, however, this band was observed as a very strong band in the current study at
411 cm-1 (trans, CC2 twist) with a prominent shoulder band at 417 cm-1. It is with this band
that there is band contour information which supported the conclusion that the trans
conformer is more stable. The CC2 twist for both the trans and gauche form are predicted at
~419 cm-1 (CC2 twist) with trans conformer having a B-type band and the gauche form an Atype band. In the infrared spectrum of the gas, a very prominent B-type band contour is
observed at 408 cm-1. The temperature study clearly showed the 408 cm-1 band arose as the
most stable conformer where it gives additional support that the trans form is the more stable
conformer.
One of the problems that was consistently encountered for isocyanate substituted
molecules is the low barrier to nearly free internal rotation. The quasi-linearity of the NCO
group to the isopropyl group has a pronounced effect on the infrared spectra in the gas phase.
Many of the bands are relatively non-descript with large bandwidth that often resolved into
several fundamentals in the spectra of the xenon solution and solid. This effect is clearly
observed in several spectral regions i.e. ~1350, ~1150, ~900, ~760 cm-1. The ~900 cm-1
band is a single broad peak but four fundamentals (943, 930, 907, 896 cm-1) are observed in
this region in the infrared spectrum of the solid. Similarly, the C-N stretching fundamental
(~760 cm-1) in the infrared spectrum of the gas is very broad with a pronounce shoulder band
at ~741 cm-1 (gauche fundamental).
As expected, similar breadths are observed for
vibrational modes associated with the NCO group. The NCO in- and out-of-plane bends are
predict at 636 and 574 cm-1, respectively, for the trans conformer but the band that is
141
observed spanned from 500 to 700 cm-1 with two non-descript peaks at 645 and 611 cm-1.
These bands were barely resolved in the infrared spectra of the xenon solution as well as the
solid. This is not surprising since the NCO fundamentals are often most affected by the low
barrier to nearly free internal rotation. A simplified estimate can be made for barrier to
internal rotation of the NCO group from the predicted energy differences (Table 26) between
the skew form and the most stable trans form and an average value of 128 cm-1 is obtained.
This very low barrier can result in a very large number of excited vibrational states populated
even at Dry Ice temperature.
Besides giving rise to a lower barrier to internal rotation, the NCO group seems to be
affecting the barrier to internal rotation of the methyl rotor as well. It has been shown [132136] that ab initio calculations with a relatively small basis set can be used to obtain the
barrier to internal rotation of the CH3 rotor. For the CH3 rotor of the isopropyl group in this
study, the predicted barriers from the ab initio calculations with the following basis sets are:
MP2(full)/6-31G(d), ΔE=1584 cm-1; MP2/6-311G(d,p), ΔE =1545 cm-1; and MP2/6311+G(d,p), ΔE =1496 cm-1. These values are consistent with the predicted barrier to methyl
rotation in ethylisocyanate [129] from the following basis sets: MP2(full)/6-31G(d),
ΔE=1520 cm-1 and MP2/6-311+G(d,p), ΔE =1482 cm-1. These predicted values are similar to
the experimentally determined barrier of 1546.2 ± 14.2 cm-1 for ethylisocyanate[129] obtained
by utilizing the fundamental torsional frequency. It is interesting to point out that the NCO
group appears to increase the methyl barrier value for the isopropyl group compare to those
substituted with H, Cl, Br, and I i.e. (CH3)2CH2 [137] = 1165 cm-1 (CH3)2CHCl [130] =1231
cm-1, (CH3)2CHBr [138]= 1490 cm-1, (CH3)2CHI [138] = 1444 cm-1. A similar trend for the
methyl barrier is observed for the ethyl compounds where the NCO (1546 cm -1) of
142
ethylisocyanate is substituted with H [139] = 1028 cm-1, F [98] = 1165 cm-1, Cl [140] = 1287
cm-1 and Br [141] = 1287 cm-1.
The low barrier to internal rotation of the NCO group makes the assignment of the
microwave spectra quite difficult particularly when the centrifugal distortion constants are
expected to have relatively large values. This probably contributed to the previous study
[125] of isopropylisocyanate in which it was incorrectly concluded that the skew conformer
was the most stable form based on the determined B+C value. The B+C value of 4031 MHz
fitted more favorably with the calculated B+C value of 4024 MHz for the skew form than the
calculated values of 3727 MHz for the gauche form or 4236 MHz for the trans conformer. A
possible reason the rotational constants reported previously did not give the correct trans
conformer as the predominate conformer was due to the strongest microwave absorption
band arising from the molecules in the excited state where the NCO moiety is freely rotating.
The weaker band at 21000 MHz (Fig. 25) gives a value of ~4270 MHz for the B+C value.
The weakness of the band is due to the small number of molecules in the bound state. By
using the B+C value of 4270 MHz, the trans conformer is more consistent as the more stable
conformer and not the skew form.
In addition to the eighteen transitions reported in Table 22, it was possible to assign
many of the Ka = 2 for J΄ = 3 but they could not be used to obtain the centrifugal distortion
constants.
This failure of the centrifugal distortion analysis is similar to the problem
observed in the study [124] of the microwave spectrum of cis cyclopropylisocyanate. For
this molecule, the frequencies of the Ka = 0, 1, and 2 transitions corresponded within the
standard error with the prediction of the spectrum including centrifugal distortion effects.
However, the frequencies of the lines with Ka > 2 showed considerable deviations from the
143
predicted values and increased significantly with increased Ka value [124] which was
observed for the transitions measured for isopropylisocyanate. There has been a detailed
analysis of the distortion constants for small molecules which have large amplitude, low
frequency vibrations [110] where the authors have shown that the energy maps of such
molecules follow closely the predictions from the mathematical concept of non-trivial
monodromy. It is expected that the large number of microwave lines (Table 30) that were
assigned by their quadrupole hyperfine splittings would have similar energy maps. From the
splitting observed for the transitions, quadrupole coupling constants were obtained (Table 23).
Additionally, the quadrupole coupling constants were predicted from the ab initio and density
functional theory calculations and these values are also listed in Table 23 for comparison to
the experimentally determined values.
Those obtained from the B3LYP/6-311+G(d,p)
calculations agree the best with the experimental values but the agreement is not as good as
frequently obtained from this level of calculation. The centrifugal distortion constants were
also predicted and a very large difference for ΔJK and ΔK was obtained between the values
from the MP2/6-311+G(d,p) and B3LYP/6-311+G(d,p) predicted values. Because of the
large amplitude bending motion of the NCO group the centrifugal distortion constants were
not expected to have meaningful values.
With the assigned transitions (Table 30) for the Ka = 3 and 4 it was hoped that a few
of the transitions were for the predicted gauche conformer. The Ka = 3 lines were predicted
in the 10000 to 11866 MHz but none of them could be identified. However, the predicted
potential function (Fig. 27) from the MP2(full)/aug-cc-PVDZ calculation indicates a well
with a depth of ~50 cm-1 whereas the predicted frequencies of 34 cm-1 for the C-N torsion
cast doubt whether a second energy level would exist in the well for the gauche conformer.
144
This potential also shows why many vibrational modes in the gas phase have very broad nondescript band contours (Fig. 23) with particular note of the vibrational modes of the NCO
bends.
The determined heavy atom structural parameters have relatively small uncertainties
where small changes of 0.001 Å have a pronounced effect on the fit of the rotational
constants. Thus, the estimated uncertainties for three of the distances are ± 0.003 Å with the
other two ± 0.005 Å and ± 0.5 º for the angles. For a comparison of these parameters, the
structural parameters from the electron diffraction study [126] are available where it was
assumed that there was a single skew conformer present. With this assumption, it required
the C-NCO angle to have a very large value of 159.0 ± 3.0 º which clearly is too large. Two
other heavy atom parameters from the electron diffraction study differs significantly from the
adjusted r0 parameters with the C4N=C angle 132.6 ± 1.0 º and C=O distance 1.184 ± 0.004
Å but the other parameters are essentially in agreement with the values obtained in this study.
It is believed that the poor structural parameters obtained from the electron diffraction study
are due to the large number of the molecules at ambient temperature which have “nearly free”
internal rotation of the NCO moiety. Also, it is believed that the one conformer described as
the skew structure is approximately the average of the trans and gauche forms which resulted
from the “nearly free” rotation of the NCO group.
145
Table 22.
Rotational transitional frequencies (MHz) for trans isopropylisocyanate in
the ground vibrational state.
Transition
F′ ← F
30,3
2
1
12609.961
0.210
2
2
12608.683
0.207
3
2
12609.822
0.212
3
3
12610.643
0.213
4
3
12609.800
0.210
1
0
13483.498
0.047
1
1
13481.564
0.039
2
1
13482.321
0.041
2
2
13483.094
0.044
3
2
13482.608
0.043
3
2
16744.483
-0.150
3
3
16743.348
-0.150
4
3
16744.416
-0.147
4
4
16745.258
-0.145
5
4
16744.406
-0.150
2
1
18165.799
-0.075
3
2
18165.499
-0.066
4
3
18165.606
-0.064
21,1
40,4
31,2
20,2
10,1
30,3
20,2
(obs)
146
(obs-calc)
Table 23. Rotational (MHz), centrifugal distortion (kHz) and quadrupole coupling (MHz) constants for the trans
and gauche conformers of isopropylisocyanate.
MP2/631G(d)
MP2/6311+G(d,p)
trans
B3LYP /6311+G(d,p)
6596
2271
1983
0.937
27.68
-20.68
0.0188
8.81
6727
2197
1904
0.974
87.09
-77.65
0.0144
15.65
6693.235(149)
2263.103(34)
1960.050(24)
K
6693
2226
1939
0.874
37.97
-30.33
0.0197
12.45
8221
2008
1695
0.671
33.70
-27.24
0.1035
16.835
8216
2054
1728
0.607
12.60
-4.24
0.081
6.796
8122
2011
1692
0.932
12.16
-4.52
0.134
4.929
χaa
χbb
χcc
2.6679
-1.8676
-0.8003
2.8161
-1.9331
-0.8830
2.8863
-1.6959
-1.1904
2.487(38)
-1.343(59)
-1.144(46)
2.8253
-1.3091
-1.5162
2.8809
-1.2713
-1.6096
2.9105
-1.4687
-1.4418
A
B
C
J
JK
147
K
J
MW
MP2/631G(d)
gauche
MP2/6311+G(d,p)
B3LYP /6311+G(d,p)
Table 24. Calculated (MP2(full)/6-31G(d)) and observed vibrational frequencies (cm-1) of isopropylisocyanate for the trans conformer.
148
A¢
A¢¢
A¢
A¢¢
A¢
A¢
A¢¢
A¢
A¢
A¢
A¢¢
A¢¢
A¢
A¢
A¢¢
A¢
A¢¢
A¢
A¢¢
A¢
Vib.
No
Approx. description
MP2
MP2 IR Raman
scaleda intb actc
n1
n20
n2
n21
n3
n4
n22
n5
n6
n7
n23
n24
n8
n9
n25
n10
n26
n11
n27
n12
CH3 antisym. stretch
CH3 antisym. stretch
CH3 antisym. stretch
CH3 antisym. stretch
CH stretch
CH3 symmetric stretch
CH3 symmetric stretch
NCO antisym. stretch
CH3 antisym. def.
CH3 antisym. def.
CH3 antisym. def.
CH3 antisym. def.
NCO sym. stretch
CH3 symmetric def.
CH3 symmetric def.
CH in-plane bend
CH out-of-plane bend
CH3 rock
CC2 antisym. stretch
CH3 rock
3213
3212
3202
3196
3133
3108
3106
2386
1570
1559
1549
1548
1479
1471
1457
1415
1412
1236
1198
1157
3014
3013
3004
2998
2939
2915
2913
2263
1473
1462
1453
1452
1404
1396
1383
1345
1341
1179
1143
1104
Table continues
16.0
11.6
42.8
0.3
15.1
8.7
9.9
808.8
4.8
9.0
0.0
2.3
4.0
17.0
12.4
37.2
2.5
3.8
6.7
45.0
75.9
42.9
91.3
6.0
91.2
195.6
2.0
1.0
2.2
23.5
25.5
4.3
6.3
14.3
2.4
15.7
6.9
1.4
3.3
9.3
Gas
2990
2990
2979
¾
2935
2924
¾
~2272
1476
1462
¾
¾
1424
1392
1375
~1332
~1332
1167
1134
1103
Infrared
Raman
Xe
3017
3017
2976
Xe
~2979
2982
~2979
¾
2943
2922
2909
2262
1470
1456
~1447
¾
1426
1387
1372
~1329
~1329
1164
1133
1101
Solid Gas
2291
2291
2291
¾
2982 ~2975
¾
¾
2932 2942
2939
2929
2925
2908
¾
¾
2270
2265 2265
1470 1474
1477
1456 1462
1462
¾
¾
1447
¾
1425 1431
1433
1387 1387
1389
1370 1372
¾
1335
1335 1337
¾
1165 1168
1166
1133 1134
¾
1106
1103 1102
PEDd
Solid
2989 70S1, 30S2
69S20, 31S21
2982 70S2, 28S1
69S21, 31S20
2940 95S3
2924 98S4
99S22
2261 98S5
1472 84S6
1454 87S7
1449 88S23
85S24
1430 31S8, 46S9
1387 50S9, 30S8
1368 90S25
1328 53S10, 20S16
50S26, 24S31, 10S29
1168 63S11, 12S17, 10S13
1135 50S27, 27S28, 18S31
1101 59S12, 17S10, 10S8
A
C
23 77
79 21
32 67
48 52
91 9
59 41
21 79
57 43
50 50
93
7
0 100
98
2
Table 24 Continues
149
A¢¢
A¢¢
A¢
A¢
A¢
A¢¢
A¢
A¢¢
A¢
A¢
A¢¢
A¢
A¢¢
a
Vib.
No
Approx. description
MP2
n28
n29
n13
n14
n15
n30
n16
n31
n17
n18
n32
n19
n33
CH3 rock
CH3 rock
CC2 symmetric stretch
C-N stretch
NCO in-plane bend
NCO out-of-plane bend
CC2 wag
CC2 twist
CC2 deformation
CH3 torsion
CH3 torsion
CNC bend
C-N torsion
990
968
944
788
636
574
493
422
366
301
268
122
45
MP2 IR Raman
scaleda intb actc
939
919
894
749
633
574
478
419
363
286
254
122
45
0.0 4.2
1.1 0.6
6.5 7.9
15.3 11.3
24.6 0.8
20.8 0.3
3.9 0.9
4.7 0.3
0.4 0.2
0.4 0.0
0.0 0.0
4.9 2.3
0.4 2.9
Infrared
Raman
Gas
Xe
Solid
¾
901
~760
645
587
478
408
355
276
925
903
757
638
~580
475
408
930
896
~745
~621
575
477
411
363
¾
125
53
Force constant scaling factors: 0.88 for CH stretches; 0.9 for all other modes.
Å4/u. d PEDs from MP2 scaled calculation.
140
¾
b
Gas
¾
¾
Xe
¾
925
902
755
639
582
~469
407
¾
271
¾
126
56
767
648
¾
¾
PEDd
Solid
931
896
¾
623
578
476
413
350
60S28, 29S27
81S29
56S13, 17S14, 13S11
45S14, 23S13, 15S8
70S15, 12S19, 10S16
96S30
47S16, 13S15
53S31, 36S26
77S17
97S18
99S32
89S19
100S33
Infrared intensities in km/mol.
c
A
C
77 23
52 48
39 61
69 31
32 68
81 19
90 10
Raman activities in
Table 25. Calculated (MP2(full)/6-31G(d)) and observed vibrational frequencies (cm-1) of isopropylisocyanate for the gauche
conformer.
150
Vib.
No
Approx. description
n1
n2
n3
n4
n5
n6
n7
n8
n9
n10
n11
n12
n13
n14
n15
n16
n17
n18
n19
n20
CH3 antisym. stretch
CH3 antisym. stretch
CH3 antisym. stretch
CH3 antisym. stretch
CH3 symmetric stretch
CH3 symmetric stretch
CH stretch
NCO antisym. stretch
CH3 antisym. def.
CH3 antisym. def.
CH3 antisym. def.
CH3 antisym. def.
CH3 symmetric def.
CH3 symmetric def.
CH out-of-plane bend
CH in-plane bend
NCO symmetric stretch
CH3 rock
CC2 antisym. stretch
CH3 rock
Table Continues
MP2
MP2
a
scaled
3215
3210
3206
3201
3113
3108
3103
2392
1572
1559
1549
1547
1478
1469
1458
1410
1397
1236
1201
1167
IR Raman
b
c
int
act
3016 16.2 58.3
3011 28.7 79.6
3007 12.3 40.9
3003
9.7 33.9
2921 10.4 127.8
2915 10.5 54.6
2911 20.9 97.4
2270 849.5
1.0
1475
6.2
3.3
1462
6.0 27.4
1453
0.4 19.0
1451
2.0 10.2
1403 23.3
3.1
1393 34.3 24.2
1383 15.6 10.2
1340 15.8
5.0
1326 50.2
8.1
1180 13.4
1.9
1145 11.7
4.7
1113 10.7
4.3
Infrared
Gas
Xenon
Raman
Xenon
Solid (-100°C)
2925
~1424 ~1425
1419
~1426
1357
~1332
~1332
1350
1328
1355
1344
1325
1128 ~1133
~1112 1116
1130
1119
1352
1122
PED
d
56S1, 34S3
48S2, 31S1, 19S3
46S3, 30S2, 18S4
75S4, 18S2
71S5, 29S6
71S6, 29S5
98S7
98S8
84S9
87S10
94S11
91S12
88S13
51S14, 20S24, 13S15
74S15, 13S14
43S16, 20S28
57S17, 19S27
61S18, 13S29, 11S23
47S19, 26S21, 20S28
55S20, 22S17
A B C
9
14
98
90
78
1
14
99
3
1
56
10
92
92
2
84
98
92
47
56
43
14
0
10
9
98
0
1
6
0
41
90
1
0
97
16
2
2
53
9
48
72
2
0
13
1
86
0
91
99
3
0
6
8
1
0
0
6
0
35
Table 25 Continues
Approx. description
n21
n22
n23
n24
n25
n26
n27
n28
n29
n30
n31
n32
n33
CH3 rock
CH3 rock
CC2 symmetric stretch
C-N stretch
NCO in-plane bend
NCO out-of-plane bend
CC2 wag
CC2 twist
CC2 def.
CH3 torsion
CH3 torsion
CNC bend
C-N torsion
151
Vib.
No
a
MP2 IR Raman
MP2 scaleda intb
actc
995
967
955
784
621
571
471
423
354
294
256
120
34
944
917
905
747
619
571
456
420
351
279
242
120
34
0.3
1.3
7.9
10.1
21.0
19.6
4.9
14.6
2.3
0.5
0.0
4.8
0.7
3.6
0.5
6.7
13.2
0.9
0.4
0.8
1.1
0.7
0.1
0.0
2.2
2.9
Infrared
Gas
Xe
944
~905
~741
~611
580
456
~415
903
742
580
456
349
Force constant scaling factors: 0.88 for CH stretches; 0.9 for all other modes.
Å4/u. d PEDs from MP2 scaled calculation.
Raman
Xenon
PEDd
Solid (-100°C)
943
943 58S21, 30S19
928 81S22
907
54S23, 14S24, 12S18
742 41S24, 20S23, 12S14, 10S25
~745
~621
79S25
575
98S26
456
~460 50S27, 13S29, 10S17
417
42S28, 27S16
351 61S29, 14S27
94S30
96S31
89S32
100S33
b
Infrared intensities in km/mol.
c
A B C
0
8
70
5
54
1
69
84
99
81
89
99
7
88
91
15
70
35
35
17
15
1
5
9
0
22
12
1
15
25
11
64
15
1
0
14
2
1
71
Raman activities in
Table 26. Calculated electronic energies (hartree) and energy differencesa (cm-1) for the trans,
gauche, skew and cis conformers of isopropylisocyanate.
Method / Basis set
MP2(full)/6-31G(d)
MP2(full)/cc-PVDZ
Basis
Functions
Trans
104
-285.747424
119
-285.801139
Gauche
131
120
Skew
162
177
Cis
186
147
MP2(full)/6-31+G(d)
MP2(full)/TZVP
128
156
-285.763357
-286.012456
60
124
139
162
152
183
MP2(full)/6-311G(d,p)
MP2(full)/6-311+G(d,p)
MP2(full)/aug-cc-PVDZ
MP2(full)/6-311G(2d,2p)
150
174
201
201
-286.001377
-286.010755
-285.858003
-286.079008
111
85
80
88
190
164
127
98
132
184
180
129
MP2(full)/6-311+G(2d,2p)
MP2(full)/cc-PVTZ
MP2(full)/6-311G(2df,2pd)
MP2(full)/6-311+G(2df,2pd)
225
278
278
302
-286.087200
-286.161888
-286.183627
-286.190707
61
137
113
79
71
127
109
79
124
227
157
153
MP2(full)/6-311G(3df,3pd)
MP2(full)/6-311+G(3df,3pd)
329
353
-286.207187
-286.213089
70
110
83
108
122
182
MP2(full)/aug-cc-PVTZ
MP2(full)/cc-PVQZ
437
540
-286.201499
-286.336201
51
87
-
195
168
B3LYP/6-31G(d)
104
-286.622796
71
82
103
B3LYP/6-31+G(d)
B3LYP/6-311G(d,p)
B3LYP/6-311+G(d,p)
B3LYP/6-311G(2d,2p)
128
150
174
201
-286.634178
-286.705382
-286.710621
-286.714378
20
16
13
12
51
69
55
48
45
20
35
7
B3LYP/6-311+G(2d,2p)
225
-286.719732
18
36
30
B3LYP/6-311G(2df,2pd)
B3LYP/6-311+G(2df,2pd)
278
302
-286.724158
-286.72909
24
19
51
35
20
38
B3LYP/6-311G(3df,3pd)
329
-286.727557
B3LYP/6-311+G(3df,3pd)
353
-286.731983
a
Energy difference are relative to the trans conformer.
-11
24
32
51
6
54
152
Table 27. Structural parameters (Å and degree), rotational constants (MHz) and dipole moment (debye) for trans isopropylisocyanate
from the 6-311+G(d,p) basis set and adjusted r0 parameters
153
Structural
Parameters
r C4-N
r N=C
r C=O
r C4-C6,7
r C4-H5
r C6,7-H10,13
r C6,7-H8,11
r C6,7-H9,12
Ð N=C=O
Ð C4N=C
Ð NC4C6,7
Ð C6C4C7
Ð NC4H5
Ð C6,7C4H5
Ð C4C6,7H8,11
Ð C4C6,7H9,12
Ð C4C6,7H10,13
t C=N2C4H5
A
B
C
|µa|
|µb|
|µc|
|µt|
a
Int.
Coord.
R1
R3
R2
R4, R5
r1
r6, r7
r4, r5
r2, r3
π
s
h1, h2
j
g
b1, b2
b7, b6
b3, b4
b5, b6
Trans
MP2
B3LYP
1.457
1.460
1.217
1.201
1.181
1.175
1.524
1.531
1.094
1.093
1.095
1.094
1.094
1.093
1.092
1.092
172.6
173.7
133.2
138.9
110.3
110.7
112.2
112.6
105.7
105.5
109.1
108.5
110.1
110.1
110.7
110.9
110.1
110.7
180.0
180.0
6596
6727
2271
2197
1983
1904
3.430
3.345
0.297
0.125
3.4428
3.347
Gauche
MP2
B3LYP
1.458
1.462
1.216
1.200
1.180
1.174
1.521/1.524
1.528/1.531
1.096
1.096
1.093/1.094
1.096/1.093
1.094/1.094
1.093/1.093
1.092/1.093
1.092/1.093
172.8
173.9
134.5
139.3
108.3/110.5
108.8/110.7
112.0
112.3
107.6
107.3
109.0/109.2
108.6/108.9
110.0/110.1
110.2/110.1
110.6/111.0
110.8/111.2
110.0/109.9
110.5/110.5
71.3
58.5
8216
8128
2054
2002
1728
1685
3.218
3.151
0.862
0.596
0.679
0.655
3.400
3.273
a,b
MW
1.471
1.171
1.207
1.522
1.091
1.091
1.091
1.091
139.0
112.7
109.9
109.9
109.9
109.9
6693.235
2263.103
1960.050
Ref. [125] assumed structural parameters, rotational study from current study. * fixed
b
E.D.
1.460(8)
1.214(6)
1.184(4)
1.534(5)
1.090*
1.114(5)
1.114(5)
1.114(5)
159.0(30)
132.6(10)
110.0(5)
114.7(9)
105.0*
107*
107*
107*
Adj.r0
Trans
1.453(3)
1.208(5)
1.165(3)
1.524(3)
1.094(2)
1.095(2)
1.094(2)
1.092(2)
172.6(5)
137.0(5)
109.4(5)
113.1(5)
105.7(5)
109.4(5)
110.1(5)
110.7(5)
110.1(5)
180.0
6693.17
2263.99
1960.15
3.427
0.270
3.438
Rotational constants from this study.
estimated
Gauche
1.454(3)
1.207(5)
1.164(3)
1.521/1.524(3)
1.096(2)
1.093/1.094(2)
1.094/1.094(2)
1.092/1.093(2)
172.8(5)
138.32(5)
107.4/109.6(5)
113.0(5)
107.6(5)
109.3/109.6(5)
110.0/110.1(5)
110.6/111.0(5)
110.0/109.9(5)
71.3
8133
2061
1732
3.206
0.833
0.671
3.380
Table 28. Symmetry coordinates for isopropylisocyanate
Symmetry coordinate definitionsa
Approximate description
A¢
CH3 antisymmetric stretch
CH3 antisymmetric stretch
CH stretch
CH3 symmetric stretch
NCS antisymmetric stretch
CH3 antisymmetric deformation
CH3 antisymmetric deformation
CH3 symmetric deformation
CH in-plane bend
CH3 rock
CH3 rock
C-N stretch
CC2 symmetric stretch
NCS symmetric stretch
NCS in-plane bend
CC2 wag
CC2 deformation
CH3 torsion
CNC bend
A²
CH3 antisymmetric stretch
CH3 antisymmetric stretch
CH3 symmetric stretch
CH3 antisymmetric deformation
CH3 antisymmetric deformation
CH3 symmetric deformation
CH out-of-plane bend
CC2 antisymmetric stretch
CH3 rock
CH3 rock
NCS out-of-plane bend
CC2 twist
CH3 torsion
C-N torsion
Not normalized.
S1 =
S2 =
S3 =
S4 =
S5 =
S6 =
S7 =
S8 =
S9 =
S10 =
S11 =
S12 =
S13 =
S14 =
S15 =
S16 =
S17 =
S18 =
S19 =
r2 - r 6 + r 3 - r 7
2r4 - r2 - r6 + 2r5 - r3 - r7
r1
r4 + r 2 + r 6 + r 5 + r 3 + r 7
S20 =
S21 =
S22 =
S23 =
S24 =
S25 =
S26 =
S27 =
S28 =
S29 =
S30 =
S31 =
S32 =
S33 =
r2 - r 6 - r3 + r 7
2r4 - r2 - r6 - 2r5 + r3 + r7
r4 + r 2 + r 6 - r5 - r 3 - r7
2a1 - a3 - a5 - 2a2 + a4 + a6
a3 - a5 - a 4 + a 6
a1 + a3 + a5 - b7 - b3 - b5 - a2 - a4 - a6 + b8 + b4 + b6
h1 - h2 + b1 - b2
R4 - R5
2b7 - b3 - b5 - 2b8 + b4 + b6
b3 - b5 - b4 + b6
t4
h1 - h2 - b1 + b2
t2 + t3
t1
R2 - R3
2a1 - a3 - a5 + 2a2 - a4 - a6
a3 - a5 + a 4 - a 6
a1 + a3 + a5 - b7 - b3 - b5 + a2 + a4 + a6 - b8 - b4 - b6
g
b3 - b5 + b4 - b6
2b7 - b3 - b5 + 2b8 - b4 - b6
R1
R4 + R 5
R2 + R 3
p
h1 + h2
4j - h1 - h2 - b1 - b2
t2 - t3
s
a
154
Table 29. Temperature and intensity ratios of the conformer bands of isoproylisocyanate from Raman and infrared spectra of the
liquid xenon solution.
Raman
Liquid xenon
155
a
Infrared
T(°C)
1/T (´10-3 K-1)
I755t / I943g
I902t / I943g
I408t/ I349g
-50.0
-60.0
-65.0
-70.0
-75.0
-80.0
-85.0
-90.0
-95.0
-100.0
4.4813
4.6915
4.8042
4.9225
5.0467
5.1773
5.3149
5.4600
5.6132
5.7753
2.1905
2.3294
10.0556
10.6739
9.6034
I757t/ I742g
10.4091
10.7727
2.4651
11.3191
9.8189
11.0000
10.9130
2.5000
12.0521
10.1679
11.1207
11.7467
10.5109
2.6614
12.4490
12.0870
12.2979
2.6196
12.8000
11.4429
12.5532
a
98 ± 19
131 ± 15
109 ± 13
121 ± 9
DH
-1
-1
115 ± 13 cm ( 1.37 ± 0.15 kJ/mol)
116 ± 7 cm ( 1.39 ± 0.09 kJ/mol)
-1
Average value obtained by utilizing all data as a single set gives DH = 115 ± 7 cm ( 1.38 ± 0.08 kJ/mol) with the trans
conformer more stable.
Table 30.
Transitions not utilized for the determination of the rotational constants for
isopropylisocyanate.
Transition
F′ ← F
31,3 ¬ 21,2
2¬1
4¬3
3¬2
3¬3
30,3 ¬ 20,2
32,2 ¬ 22,1
2¬2
4¬3
3¬2
2¬1
3¬3
2¬1
4¬3
3¬2
32,1 ¬ 22,0
32,2 ¬ 31,2
2¬1
4¬3
3¬2
3¬3
4¬4
2¬2
31,2 ¬ 21,1
4¬3
2¬1
3¬2
21,1 ¬ 10,1
1¬1
2¬1
n(obs)
12203.637
Transition
F′ ← F
21,1 ¬ 10,1
50,5 ¬ 41,3
1¬0
5¬4
12204.206
12608.683
12609.800
41,4 ¬ 31,3
6¬5
5¬4
12609.822
12609.961
12610.643
40,4 ¬ 30,3
12203.861
12203.966
4¬3
3¬2
12663.800
12664.261
12665.090
12723.833
12724.294
42,3 ¬ 32,2
12725.134
12831.783
12831.807
43,2 ¬ 33,1
12831.939
43,1 ¬ 33,0
13114.526
13114.731
13481.564
3¬2
2¬2
13483.094
4¬4
3¬2
5¬4
4¬3
3¬2
5¬4
4¬3
3¬2
5¬4
4¬3
13114.505
13482.321
13482.608
3¬2
4¬3
3¬3
5¬4
42,2 ¬ 32,1
3¬2
5¬4
4¬3
41,3 ¬ 31,2
Table Continues
156
n(obs)
13483.498
14546.221
14546.259
16255.003
16255.032
16255.099
16743.348
16744.406
16744.416
16744.483
16745.258
16874.065
16874.160
16874.510
16904.906
16905.204
16905.975
16907.699
16907.997
16908.766
17022.634
17022.731
3¬3
17023.092
17466.389
5¬4
17467.316
Tables 30 Continue
Transition
F′ ← F
41,3
31,2
31,2
20,2
51,5
41,4
3
4
4
3
4
2
4
6
2
3
4
2
3
1
4
5
4
5
5
4
5
6
4
5
4
6
3
4
5
4
4
5
3
5
3
5
20293.480
5
4
6
5
6
4
5
4
3
5
4
5
3
4
21074.183
21365.971
21365.994
21366.188
21802.574
21802.607
21802.630
50,5
40,4
52,4
42,3
52,3
42,2
51,4
41,3
(obs)
17467.359
17467.418
17468.131
18165.499
18165.606
18165.799
20292.643
20293.454
20293.509
20294.172
20823.910
20824.935
20824.935
20824.986
20825.786
21073.983
21074.007
157
Figure 21.
Atomic numbering of isoproylisocyanate.
158
Figure 22.
Observed and predicted (MP2(ful1)/6-311+G(d,p)) infrared spectra of
isoproylisocyanate: (A) Gas; Simulated spectra of (B) mixture of gauche and
trans conformers with H = 115 cm-1, (C) pure gauche conformer, and (D) pure
trans conformer.
159
160
Figure 23.
Infrared spectra of isoproylisocyanate: (A) Gas; (B) Xenon solution at -100 C
Figure 24. Observed and predicted (MP2(ful1)/6-311+G(d,p)) Raman spectra of
isoproylisocyanate: (A) Xenon solution at 100 C; Simulated spectra of (B)
mixture of gauche and trans conformers with H = 115 cm-1, (C) pure gauche
conformer, and (D) pure trans conformer.
161
162
Figure 25.
Low resolution microwave spectrum of isopropylisocyanate.
163
Figure 26.
Temperature dependence of the 902 cm-1 (trans) and the 943 cm-1 (gauche) Raman bands of isopropylisocyanate
dissolved in liquid xenon.
Figure 27.
Potential energy curve of isoproylisocyanate as a function of the dihedral angle
C=N C H5 obtained from MP2(full)/6-311G(2d,2p) calculations (dash line)
and MP2(full)/aug-cc-PVDZ calculations (solid line).
164
CHAPTER 8
Raman and Infrared Spectra, Conformational Stability, Ab initio Calculations and
Vibrational Assignment of Dimethylsilylisocyanate
Introduction
The substitution of the pseudohalogen NCO and NCS to a hydrocarbon (R) results in
a very large CNC(X) angle [59, 111] which may result in a very low wavenumber CNC bend
with only a few bound vibrational states with the remaining molecules having essentially free
internal rotation which has been found for CH3NCO1 and CH3NCS [59]. However, when
these pseudohalogens are substituted on a silane the question arises whether there is more
than one bound state for the SiNC(X) bend or simply a small “hump” in the potential
function governing the SiNC bending mode as the effective structure is a linear heavy atom
skeleton. There have been extensive structural studies of SiH3NC(X) molecules to determine
whether the SiNC(X) moiety is linear or bent in the ground state since the symmetry should
be drastically different for C3V symmetry for the linear molecule in comparison with a CS
symmetry for the bent molecule. Thus, microwave, [102, 142-144] infrared, [145-148] and
Raman spectra [145-147, 149] along with electron diffraction [150, 151] data were utilized to
determine the structure of SiH3NCO and SiH3NCS as well as for the corresponding
trimethylsilyl isocyanate and trimethylsilyl isothiocyanate molecules.
The most definitive information for the Si-N=C bend for silylisocyanate was
obtained7 from the Raman spectra of gaseous SiH3NCO and SiD3NCO where 18 transitions
were assigned for both isotopomers with the determined anharmonic potential having a
barrier of 45 cm-1 at the linear skeletal configuration with only two vibrational levels below
165
the barrier. This result is consistent with the microwave result [102] where 10 isotopic
species of silylisocyanate were measure and it was concluded that the heavy atom of
SiH3NCO results in a quasi-linear molecule with a potential hump of 31.5 cm-1.
The
determined structural parameters [102] are consistent with those obtained from the electron
diffraction data [150].
For trimethylsilylisocyanate the SiNC angle was reported to be quite large with a
value of 157(3)° from electron diffraction [151] data which contradicts the results from the
microwave study [144]. A wavenumber of 64
15 cm-1 for the SiNC bend was reported
[144] from vibrational satellites of this lowest bending mode which was later observed [147]
in the Raman spectrum of the gas at 37 cm-1 with an overtone at 84 cm-1. However, from ab
initio and DFT calculations a very low barrier of 4-15 cm-1 has been predicted [146] which
would indicate only one energy level below the barrier. Nevertheless the observed vibrational
spectra could be clearly assigned on the basis of a molecule with C3V symmetry except for
the NCO bend where there are clearly two Raman lines at 612 and 603 cm-1 in the spectrum
of the gas. This observed splitting is consistent with the quasi-linearity of the SiNCO moiety
and further indicates that the Raman spectrum of the gas can be used to obtain structural
information of these large amplitude anharmonic vibrations of these substituted silanes which
contain the NC(X) groups.
Experimental
The sample of dimethylsilylisocyanate was prepared by the reaction of
chlorodimethylsilane with solid sliver cyanate at room temperature for three hours. The
sample was purified by trap-to-trap distillation and the final purification was obtained by
166
using a low-temperature, low-pressure sublimation column. The sample identity was verified
by NMR and infrared spectroscopic data.
The Raman spectra (3200 to 30 cm-1), shown in Fig. 28 were recorded on a Cary
model 82 spectrophotometer equipped with a Spectra-Physics model 171 argon ion laser
operating on the 5145 Å line. The spectrum of the gas was recorded by using a standard
Cary multipass accessory and laser power at the sample of 1.3 W. Reported wavenumbers
for sharp resolvable lines are expected to be accurate to at least ±2 cm-1. The spectrum of the
liquid was recorded from the sample seal in a Pyrex capillary. The depolarization ratios were
obtained for the gas and liquid samples with the standard Cary accessories. The Raman
spectrum of the solid was recorded by condensing the sample onto a blackened copper block
positioned at an angle of 15° from the normal and maintained at 77 K with liquid nitrogen.
To obtain good spectra for the solid phase, the sample was allowed to warm (not melt) and
cool repeatedly. This annealing process was performed until the molecules gained sufficient
thermal energies to re-orient themselves into a final consistently-packed lattice in the
crystalline film.
The mid-IR spectra from 3500 to 400 cm-1 of the gas and solid which are shown in
Fig. 29 were recorded with a Digilab model FTS-14C Fourier Transform interferometer
equipped with a Ge/KBr beamsplitter, nichrome wire source, and a triglycine sulfate (TGS)
detector. A KBr plate cooled with boiling liquid nitrogen was used as the sample substrate to
obtain the spectrum of the solid.
Ab Initio Calculations
The LCAO-MO-SCF calculations were performed with the Gaussian-03 program [43]
by using Gaussian-type basis functions.
The energy minima with respect to nuclear
167
coordinates were obtained by simultaneous relaxation of all geometric parameters consistent
with symmetry restrictions using the gradient method of Pulay [44]. A number of basis sets
starting from 6-31G(d), and increasing to 6-311+G(2df,2pd), were employed at the level of
Møller-Plesset perturbation theory [38] to the second order (MP2) as well as hybrid density
functional theory by the B3LYP method, to obtain energy differences among the likely
conformers of dimethylsilylisothiocyanate. From all levels of calculation carried out, with
and without diffuse function, predicted conformational stability varies extensively with the
size of the basis set.
To aid in making the vibrational assignment (Table 31), we have carried out normal
coordinate analysis by utilizing the force fields obtained from the Gaussian-03 program at the
MP2(full)/6-31G(d) level. The internal coordinates used to calculate the G and B matrices
for dimethylsilylisocyanate are listed along with the structural parameters in Tables 32. By
using the B matrix, the force fields in Cartesian coordinates was converted to force fields in
internal coordinates [45]. Subsequently, scaling factors of 0.88 for the CH stretches and 0.90
for all other modes, except all heavy atom bends and stretches, were used, along with the
geometric average of scaling factors for interaction force constants, to obtain the fixed scaled
force fields and the resultant wavenumbers (Tables 31). Symmetry coordinates listed in
Tables 33 for dimethylsilylisocyanate were used to determine the corresponding potential
energy
distributions
(P.E.D.s).
The
observed
and
calculated
wavenumbers
of
dimethylsilylisocyanate along with the calculated infrared intensities, Raman activities,
depolarization ratios, and P.E.D.s are given in Table 31.
In order to identify the fundamental vibrations for possible conformers of
dimethylsilylisocyanate, we have simulated the Raman spectrum (Fig. 28) from the scaled ab
168
initio MP2(full)/6-31G(d) results. The evaluation of Raman activity by using the analytical
gradient methods [46, 47] has been developed.
To further support the vibrational
assignments, the infrared spectrum (Fig. 29) was predicted by using scaled wavenumbers
from MP2(full)/6-31G(d) results. Infrared intensities were calculated based on the dipole
moment derivatives with respect to Cartesian coordinates. There are some major differences
between the observed spectrum and those of the predicted spectra which will be discussed
more fully in the next section.
Conformatinal Stability
As indicated earlier the Raman spectrum of gaseous silylisocyanate provided the
spectral data to show that the potential barrier was quite low with a value of 45 ± 3 cm -1
which gave a very broad Raman band which extended from 100 to 135 cm-1 with fifteen
transitions assigned. Therefore, we began the conformation analysis of (CH3)2SiHNCO with
the investigation of the spectral region below 150 cm-1 of the Raman spectrum of the gas.
There are seven well defined Q-branches whereas there are only two fundamentals predicted
at 47 and 38 cm-1. If the SiNCO skeleton is bent, the SiNC bend and the asymmetric torsion
with both having significant predicted activities should give rise to only one bend, whereas if
the SiNC angle is linear there will be two SiNC bends. The SiNC bends give rise to a large
number of excited transitions starting at 86 cm-1 in the Raman spectrum of the gas which are
similar to the corresponding Raman lines observed for the SiH3NCO molecule. [145] The
large number of observed bands are undoubtedly due to the SiNC bends of a quasilinear
molecule with large amplitude vibrational motion and a relatively low barrier at the linear
position. In fact, if the ground state vibration energy of the SiNC bend is above the barrier,
the heavy atom skeleton is effectively linear with essentially two SiNC bends with large
169
amplitude motions.
The remaining Raman spectrum of the gas is consistent with this
interpretation where the Si-N stretch is a relatively broad Raman line whereas the two SiC2
stretches (symmetric and antisymmetric) coalesces into a very broad single band with a
maximum at 676 cm-1 (Fig. 28). These two fundamentals are clearly observed in the Raman
spectrum of the liquid and quite pronounced in the Raman spectrum of the solid. Similar
breadth of two SiC2 bends are observed in both the Raman spectra of the gas and liquid, but
however the SiC2 rock (A") is observed in the Raman spectrum of the liquid. Therefore, we
concluded from the Raman spectrum of the gas that the skeletal SiNCO moiety is effectively
linear with a sufficiently low barrier that the ground state energy levels for the fundamental
SiNC bends are probably above the barrier.
Vibrational Assignment
Since the ab initio calculations with relatively small basis sets predict the gauche
conformer as the more stable form whereas the cis form has a negative wavenumber but
determine to be the only stable form from the electron diffraction studies we predicted the
wavenumbers for the fundamental vibrations for these two conformers as well as for the
linear form. With the exception of the asymmetric torsion all of the predicted wavenumbers
for all three conformers are within 1 cm-1. Therefore, the predicted wavenumbers given in
Table 31 can be considered for all of these conformers.
The vibrations of major interest are those associated mainly with the heavy atoms.
The three SiC2 bends are predicted at 261 (observed 277), 242 and 196 (observed 218) cm-1
with the SiC2 rock (A") not observed nor expected to be observed in the Raman spectrum of
the gas. However, all three modes are clearly observed in the Raman spectra of the liquid
and solid with the SiC2 rock observed at 259 cm-1 in the liquid. It should be noted that all
170
three of these bends are observed at significantly high wavenumbers than the predicted
values. As indicated earlier the SiN and two SiC2 stretches along with the two NCO bends
are easily identified between 540 and 717 cm-1 in the Raman spectrum of the solid with the
two NCO bends appearing mainly as one band at 637 cm-1 in the spectrum of the liquid with
breadth on the low wavenumber side of the band.
There is some significant shift of
wavenumbers where the SiN stretch is observed at 560, 549, and 540 cm-1 in going from the
gas, liquid and then solid whereas the opposite shift is observed for the SiC 2 symmetric
stretch with wavenumbers of 671, 680 and 688 cm-1 from the gas, liquid and solid phases,
respectively.
The assignments for the remaining fundamentals can be made routinely based on their
predicted wavenumbers, infrared band intensities, Raman activities and depolarization values
along with the well known group wavenumbers. The assignments for the fundamentals are
listed in Table 31 for all three physical states when observed and, in general, the
wavenumbers are not significantly different in the different physical states.
Structural Parameters
Since the structural parameters of (CH3)2SiHNCO have been previously reported
[152] from an electron diffraction study we have obtained the predicted r0 parameters from
ab initio calculations with some “off sets” for the heavy atoms which should provide
rotational constants that could significantly aid the microwave assignments for the molecule.
We have [33] shown that ab initio MP2(full)6-311+G(d,p) calculations predict the r0
structural parameters for more than fifty carbon-hydrogen distances better than 0.002 Å
compared to the experimentally determined values from isolated CH stretching wavenumbers
which were compared [153] to previously determined values from earlier microwave studies.
171
Thus, all of the carbon-hydrogen parameters can be taken from the MP2(full)/6-311+G(d,p)
predicted values for (CH3)2SiHNCO. It has also been shown that SiH distances can be
obtained from the wavenumbers of the isolated SiH stretching wavenumbers. [104]
Therefore, we have obtained the values of 1.485 Å for the SiH distances (Table 32) which is
0.009 Å longer than the value for the corresponding distance from the ab initio predicted
parameter. This longer distance is the same difference found for many SiH distances in other
organosilanes. [154, 155] Thus, with the SiH and CH distances determined within 0.002 Å
and the corresponding bond angles to an expected uncertainty of 0.5°, there are only four
heavy atom distances and two heavy atom angles to be determined. The N=C=O distances
have been determined for a number of compounds where they have also been predicted from
ab initio MP2(full)/6-311+G(d,p) calculations as well as with density functional theory
calculations so they should be predictable from offset adjustments. Similar data are also
available for the SiC distance which leaves the SiN distances and the two heavy atom angles
to be predicted. The values for the predicted r0 parameters are given in Table 32 along with
ab initio predicted parameters and those obtained from the electron diffraction study.
Discussion
The Raman data, particular for the gas, provided the most convincing information on
the quasi-linear nature of the heavy atom skeleton of the dimethylsilylisocyanate molecule.
A series of transition beginning at 158 cm-1 with succeeding low wavenumber transitions at
141, 124, 118, 108, 94 and 86 cm-1 (Fig. 28) are undoubtedly the excited state transitions of
the SiNC bends. This spectral information is very similar to the corresponding bands that
were observed for silylisocyanate where their proposed assignments gave a potential function
with a barrier of 45 cm-1 and with only one excited state below the barrier. [145] Similar
172
quasi-linearity was found for the SiNCO skeleton of the trimethylisocyanate molecule from a
microwave investigation [144] which is consistent with the vibrational data. [147, 148]
Therefore, the conclusion for the dimethylsilylisocyanate that the SiNC bend is linear or near
linear is consistent with the structural determination for the corresponding linkage for similar
molecules. Further support for the quasi-linearity for the SiNCO moiety is provided from the
MP2 ab initio predictions (Table 34) from the calculation with relatively large basis set such
as the 6-311G(2df,2pd) both with and without diffuse function, in which case the
optimization of the gauche and cis conformers lead to a linear form with the SiNCO skeleton
180º. With the smaller basis sets the cis conformer was predicted to have a lower energy
than the gauche form but subsequent calculations utilizing higher basis sets gave a negative
wavenumber for the lowest vibrational mode.
Even though the gauche conformer is
predicted to have a lower energy than the cis form, the energy difference is almost
insignificant. With the larger basis sets only the linear form is a stable conformer. Therefore
the ab initio calculations clearly indicate a linear SiNC skeletal form.
We have placed the structural parameters obtained for the cis conformer from the
MP2/6-311+G(d,p) calculation in Table 32 but it should be noted that these parameters were
simply listed to indicate that the structural parameters are almost identical to the linear form
even though the cis form is clearly not the stable conformer.
The conformational stability for dimethylsilylisocyanate is in contrast to the
experimentally determined
[125]
form
for
the
corresponding
carbon
analogue,
(CH3)2C(H)NCO, where the stable conformer was found to be the skew form where one of
the methyl groups was eclipsing the NCO moiety.
This conclusion was obtained from the
microwave spectra where the experimentally determined B+C value of 4031 MHz was in
173
excellent agreement with the predicted value of 4024 MHz for the skew form compared to
the predicted value of 4236 MHz for the gauche forms. These values were obtained for a CN=C angle of ca. 139°. Also it should be noted this value was in relatively good agreement
with the angle of 132.6(10)° which was found three years later from an electron diffraction
study. [126] This determined conformation is consistent with the expected structure for the
carbon-nitrogen bond of the isocyanate moiety being mainly a double bond. Nevertheless, an
ab initio prediction of the conformation stability and a more extensive microwave
investigation with the determinations of the A, B and C rotational constants would be of
interest for this molecule.
Also of interest would be a conformational determination of the corresponding
germane compound, (CH3)2Ge(H)NCO, since the germanium atom is expected to have
bonding with the isopropyl moiety more similar to carbon than silicon.
For example, the
GeH3NCO molecule has been shown to be bent with a GeNC angle approximately 142° from
microwave studies. [156, 157] Similarly, from a vibrational study [158] utilizing infrared
and Raman spectral data it was shown that the (CH3)3GeNCO molecules was bent in all
phases. Later from an evaluation [159] of the force constants for the GeNC moiety of the
trimethylisocyanate from the previous vibrational study, the determined force constants were
essentially the same as those for the corresponding values of GeH3NCO. From this result, it
was concluded that the GeNC angle was about 141° for (CH3)3GeNCO; thus a more
definitive experimental determination is desirable.
174
Table 31. Calculated and observed vibrational wavenumbers (cm-1) of dimethysilylisocyanate.at the MP2(full)/6-31G(d) level.
175
Vib.
No
A¢
n1
n2
n3
n4
n5
n6
n7
n8
n9
n10
n11
n12
n13
n14
n15
n16
n17
n18
n19
Approx. description
CH3 antisymmetric stretch
CH3 antisymmetric stretch
CH3 symmetric stretch
NCO antisymmetric stretch
SiH stretch
CH3 antisymmetric def
CH3 antisymmetric def
NCO symmetric stretch
CH3 symmetric def
CH3 rock
CH3 rock
SiH in-plane bend
SiC2 symmetric stretch
NCO in-plane bend
Si-N stretch
SiC2 wag
SiC2 deformation
CH3 torsion
SiNC bend
Table Continues
MP2
3209
3201
3107
2452
2306
1529
1522
1483
1377
953
915
789
694
605
552
269
197
174
47
MP2
scaleda
3011
3003
2915
2326
2164
1450
1444
1407
1307
907
872
750
658
605
552
261
196
165
47
IR Raman Raman
intb
actc
gas
6.6
13.3
1.4
1285.1
196.7
7.9
3.6
43.4
22.8
336.5
68.3
46.6
26.1
23.9
71.7
7.5
0.4
0.1
0.5
90.7
145.2
214.5
7.2
119.4
0.6
28.8
60.9
0.9
6.6
3.6
7.2
12.2
0.4
8.2
1.1
1.5
0.1
3.3
2983*
2979*
¾
2278
2170
1451
1451
1415
1266
900
843
752
676
620
560
277
215
158
¾
IR
gas
Raman
liquid
Raman
solid
2979
2972
2915
2290
2169
1448
1448
2970
2970
2910
2278
2179
1445
1445
1415
1267
900
845
755
¾
637
549
279
218
¾
¾
2974
2974
2913
2271
2172
1455
1455
1415
1262
891
859
755
¾
620
540
272
224
¾
45
1269
902
844
776
690
624
557
¾
¾
¾
¾
IR
solid
2979
2969
¾
2367
2170
1445
1445
1398
1257
900
850
749
681
609
540
¾
¾
¾
¾
PEDd
97S1
96S2
100S3
98S4
100S5
81S6, 14S7
82S7, 14S6
81S8, 17S15
98S9
44S10, 27S12, 23S16
78S11, 10S17
60S12, 35S10
79S13, 10S15
100S14
68S15, 14S8, 12S13
68S16, 10S12
81S17
99S18
10019
Table 31 Continues
176
Vib.
No
A¢¢
n20
n21
n22
n23
n24
n25
n26
n27
n28
n29
n30
n31
n32
n33
a
Approx. description
CH3 antisymmetric stretch
CH3 antisymmetric stretch
CH3 symmetric stretch
CH3 antisymmetric def
CH3 antisymmetric def
CH3 symmetric def
SiH bend
CH3 rock
SiC2 antisymmetric stretch
CH3 rock
NCO out-of-plane bend
SiC2 rock
CH3 torsion
SiNC bend
MP2
MP2
scaleda
3210
3199
3107
1516
1514
1372
960
828
739
659
604
243
157
38
3011
3001
2915
1438
1437
1302
911
789
701
625
604
242
149
38
IR Raman Raman
intb
actc
gas
2.8
0.2
0.4
0.6
1.4
50.5
163.2
46.3
3.8
12.7
23.2
7.2
0.0
0.1
41.5
15.0
0.6
14.5
13.2
1.0
9.1
2.2
5.3
1.7
0.3
0.1
0.0
3.8
¾
¾
¾
¾
¾
¾
¾
¾
¾
¾
600
¾
¾
¾
IR
gas
Raman
liquid
Raman
solid
IR
solid
2983
2968
¾
1438
1438
1262
898
781
¾
624
602
¾
¾
¾
2970
2970
¾
1445
1445
1267
900
783
710
637
¾
259
¾
¾
2983
2968
2880
1440
1429
1262
896
789
717
645
620
263
¾
¾
2986
2962
¾
1432
1426
1251
900
781
¾
635/663
598
¾
¾
¾
PEDd
95S20
95S21
100S22
94S23
93S24
97S25
40S26, 31S29, 21S31
61S27, 27S28
64S28, 26S27
58S29, 26S26, 11S31
96S30
61S31, 32S26
99S32
98S33
Force constant scaling factors: 0.88 for CH stretches; 0.9 for all other modes except for all heavy atom bend and stretches.
Infrared intensities in km/mol.
c
4
Raman activities in Å /u.
d
PEDs from MP2 scaled calculation.
b
Table 32. Structural parameters (Å and degree), rotational constants (MHz) and dipole
moments (debye) for dimethylsilylisocyanate at the MP2(full)/6-311+G(d,p) level .
Structural
Parameters
r C= O
r N= C
r Si–N
r Si–C6,7
r Si–H5
r C6,7–H8,11
r C6,7–H9,12
r C6,7–H10,13
Ð NCO
Ð SiNC(O)
Ð C6,7SiN
Ð C6SiC7
Ð NSiH5
Ð C6,7 SiH5
Ð SiC6,7H8,11
Ð SiC6,7H9,12
Ð SiC6,7H10,13
Ð H8,11C6,7H9,12
Ð H8,11C6,7H10,13
Ð H9,12C6,7H10,13
t N3Si4C6H8
t N3Si4C7H11
t C2N3Si4H5
t CNCS
A
B
C
|µa|
|µb|
|µc|
|µt|
a
Internal
Coordinates
R1
R2
R3
R4,R5
r1
r2, r3
r4, r5
r6, r7
p
s
h1, h2
j
γ
b1,b2
b3, b4
b5, b6
b7, b8
a1, a2
a3, a4
a5, a6
t1
t2
t3
t4
MP2
linear
1.181
1.203
1.733
1.863
1.477
1.095
1.093
1.094
180.0
180.0
108.3
112.2
106.7
110.6
110.7
111.1
110.8
MP2
cis
1.181
1.203
1.733
1.862
1.477
1.095
1.093
1.094
179.6
177.7
108.2
112.2
106.7
110.6
110.7
111.1
110.8
5274
1484
1252
3.350
0.000
0.127
3.352
5298
1481
1249
2.678
1.596
0.242
3.127
Ref. [152], the asterisks indicate that these values are fixed.
177
a
E. D
cis
1.155(4)
1.218(4)
1.719(5)
1.858(3)
1.50*
1.131(7)
1.131(7)
1.131(7)
180*
153.5(13)
111.2(25)
113.3(50)
105*
~109.7
109*
109*
109*
estimated r0
1.165(5)
1.215(5)
1.715(5)
1.855(5)
1.485(3)
1.095(2)
1.093(2)
1.094(2)
180.0
180.0
108.5(5)
112.0(5)
106.5(5)
110.0(5)
110.5(5)
111.0(5)
110.5(5)
5332
1494
1262
Table 33. Symmetry coordinates for dimethylsilylisocyanate.
Symmetry coordinate definitionsa
Approximate description
A¢ CH3 antisymmetric stretch
CH3 antisymmetric stretch
CH3 symmetric stretch
NCO antisymmetric stretch
SiH stretch
CH3 antisymmetric deformation
CH3 antisymmetric deformation
S1 =
S2 =
S3 =
S4 =
S5 =
S6 =
S7 =
CH3 symmetric deformation
S8 =
NCO symmetric stretch
CH3 rock
CH3 rock
SiH in-plane bend
SiC2 symmetric stretch
NCO bend
Si-N stretch
SiC2 wag
SiC2 deformation
CH3 torsion
SiNC bend
A² CH3 antisymmetric stretch
CH3 antisymmetric stretch
CH3 symmetric stretch
CH3 antisymmetric deformation
CH3 antisymmetric deformation
S9 =
S10 =
S11 =
S12 =
S13 =
S14 =
S15 =
S16 =
S17 =
S18 =
S19 =
S20 =
S21 =
S22 =
S23 =
S24 =
CH3 symmetric deformation
S25 =
CH3 rock
CH3 rock
SiH bend
SiC2 antisymmetric stretch
NCO bend
SiC2 twist
CH3 torsion
SiNC bend
S28 =
S29 =
S26 =
S27 =
S30 =
S31 =
S32 =
S33 =
a
Not normalized.
178
r4 - r6 + r5 - r7
2r2 - r4 - r6 + 2r3 - r5 - r7
r2 + r4 + r6 + r3 + r5 + r7
R1 - R2
r1
2a5 - a1 - a3 + 2a6 - a2 - a4
a1 - a3 + a2 - a4
a5 + a1 + a3 - b3 - b5 - b7 + a2 + a4 + a6
- b 4 - b6 - b8
R1 + R2
2b3 - b5 - b7 + 2b4 - b6 - b8
b5 - b7 + b6 - b8
γ
R4 + R5
p1
R3
h1 + h2 - b1 - b2
4j - h1 - h2 - b1 - b2
t1 - t2
s1
r4 - r6 - r5 + r7
2r2 - r4 - r6 - 2r3 + r5 + r7
r2 + r4 + r6 - r3 - r5 - r7
2a5 - a1 - a3 - 2a6 + a2 + a4
a1 - a3 - a2 + a4
a5 + a1 + a3 - b3 - b5 - b7 - a6 - a2 - a4
+ b4 + b6 + b8
2b3 - b5 - b7 - 2b4 + b6 + b8
b5 - b7 - b6 + b8
h1 - h2 + b1 - b2
R4 - R5
p2
h1 - h2 - b1 + b2
t1 + t2
s2
Table 34. Calculated energiesa,b (hartree) for the possible conformers of
dimethylsilylisocyanate.
Method / Basis set
linear
gauche
cis
MP2/6-31G(d)
0.813953
0.813955
0.813958
MP2/6-31+G(d)
0.832667
—
0.832667
MP2/6-311G(d,p)
1.183869
1.183920
1.183916*
MP2/6-311+G(d,p)
1.192917
1.192975
1.192922*
MP2/6-311G(2d,2p)
1.275081
1.275165
—
MP2/6-311+G(2d,2p)
1.283438
1.283458
1.283450
MP2/6-311G(2df,2pd)
1.382655
—
—
MP2/6-311+G(2df,2pd)
1.389371
—
—
a
Energies is given as – (E+536) H.
The dash lines indicate optimization leads to a linear SiNC angle and
* The asterisks indicate a negative frequency was predicted.
b
179
Figure 28. Raman spectra of dimethylisilylisocyanate: (A) gas; (B) liquid; (C) solid; (D)
simulated spectrum from scaled MP2/6-31G(d) calculation.
180
Figure 29. Infrared spectra dimethylisilylisocyanate: (A) gas; (B) amorphous; (C) annealed
solid; (D) simulated spectrum from scaled MP2/6-31G(d) calculation.
181
CHAPTER 9
CONFORMATIONAL STABILITY FROM VARIABLE TEMPERATURE INFRARED
SPECTRA OF XENON SOLUTIONS, R0 STRUCTURAL PARAMETERS, AND AB
INITIO CALCULATIONS OF CYCLOPROPYLISOCYANATE
Introduction
Organoisocyanates and organoisothiocyanates provide interesting challenges to structural
scientists for the experimental determination of their conformational stabilities and/or
structural parameters because of the large CNC(X) angle, which gives rise to many of the
molecules having low-frequency, large amplitude bending vibrations which may be very
anharmonic. Additionally, the barrier to internal rotation about the bond to the C–N=C(=X)
group may be relatively small, which can result in nearly free internal rotation of the NCX
moiety. Also, because of the very low frequency of the bending mode there may be a very
large number of excited vibrational states populated even at Dry Ice temperature, which is
the temperature usually used to record the normal microwave rotational spectra of such
molecules.
It is also difficult to obtain good structural parameters for many of these
molecules from electron diffraction studies. The nearly free rotation of the NCX moiety for
more than 50% of the molecules at ambient temperature, along with a large number of
anharmonic bending vibrations of very low frequency in many excited states, result in a large
number of undetermined parameters, making it difficult to determine conformational
stabilities.
Further confusion about the structure of several of these molecules has arisen from ab
initio predictions of the structures and conformational stabilities of organo-NCX compounds
182
from calculations with relatively small basis sets which often do not agree with the most
stable conformer or the stable conformers determined experimentally. Also, in some cases
the potential wells are so shallow that they cannot accommodate a vibrational state, hence the
microwave determined rotational constants will be different from the conformation which
has the re minimum. Therefore, we have been re-investigating the conformational structural
stabilities of a number of these molecules, particularly by recording variable temperature
vibrational spectra of the gas or of liquid rare gas solutions. Additionally, we have obtained
the heavy atom structural parameters as well as ab initio predictions of the conformational
stabilities by utilizing much larger basis sets than those that were previously used and which
predicted the most stable conformer incorrectly.
We initiated these recent studies by investigations of methylisothiocyanate (CH3NCS),
[59] and methylisocyanate (CH3NCO), [111] where free internal rotation was observed for
the methyl rotors for both molecules and the r0 structural parameters were determined. These
studies have been followed with spectroscopic and theoretical studies of ethylisothiocyanate
(CH3CH2NCS) [60] and ethylisocyanate (CH3CH2NCO) [129]. For the thio- compound, it
was concluded from the microwave spectra [116] that the most stable conformer was the
trans form, whereas from ab initio calculations [61] at the MP2 level with various basis sets
up to TZVP, it was concluded that the gauche form was the most stable form. However, from
a more recent spectroscopic and theoretical study [60] using larger basis sets it was
conclusively shown that the most stable form was the cis conformer. A similar study of
ethylisocyanate with large basis sets predicted the depth of the cis well to range from 11 cm-1
to a high value of 31 cm-1, and with the largest basis set (425 for the MP2(full)cc-PVQZ
calculation) predicting a value of 19 cm-1. This is a strong indication that the gauche wells
183
could not contain a bound vibrational energy level. Also, the spectroscopic data was only
consistent with the most stable conformer being the cis form.
As a continuation of these studies of pseudo-halogen organic compounds we have
again investigated the conformational stability and determined the structural parameters of
cyclopropylisocyanate. In an earlier microwave, infrared and Raman spectral study [160] it
was concluded that there were two conformers present at ambient temperature, which were
the most stable trans form and the cis form (Fig. 30). However, from a theoretical study
[123] seven years later the authors claimed that the trans and cis forms previously reported
are actually two gauche forms with dihedral [CNC(1)C(2)] angles of 144.9° and 324.3°
based on MP2/6-31G(d) calculations. There was little information on how the calculations
were performed and a figure in the article implied that this dihedral angle should be 180° for
the cis form. There was also a later microwave study [124]. Since in the earlier study [160]
the value of the A rotational constants for both conformers had significant uncertainties, a
much larger number of transitions were assigned and measured for both conformers. For the
cis conformer, it was not possible to fit the five usually determined centrifugal distortion
constants by the standard equations, which lead the authors to question the conformation for
the form assigned as the cis conformer. Thus, there is still a question concerning the stable
conformers for cyclopropylisocyanate. Additionally, in the earlier study [160] the enthalpy
difference was determined for the liquid and only estimates were made for the r(C–N),
CNC and
CCN parameters.
Therefore, we have carried out a variable temperature
infrared study of a xenon solution of cyclopropylisocyanate which should give an enthalpy
value near that of the gas. Also, with three well-determined rotational constants it should be
possible to obtain reliable structural parameters, and with ab initio calculations using much
184
larger basis sets the predicted conformational stability is expected to be more reliable. The
results of this spectroscopic and theoretical study are reported herein.
Experiment and Theoretical Calculations
The sample of cyclopropylisocyanate was prepared via a two step procedure
according to the method of Kricheldorf and Regel [161, 162].
In the first step, an
intermediate product of diphenyl diazido silane was formed from dichloro diphenyl silane
and sodium azide. The final product is achieved through the Curtius rearrangement of the
intermediate formed from reaction with cyclopropyl carbonyl chloride.
The sample was
purified by trap-to-trap distillation and the final purification was obtained by using a lowtemperature, low-pressure sublimation column. The sample identity was verified by NMR
and infrared spectroscopic data.
The mid-infrared spectra from 3500 to 400 cm-1 of the gas shown in Fig. 31A was
recorded on a Perkin-Elmer model 2000 Fourier transform spectrometer equipped with a
nichrome wire source, Ge/CsI beamsplitter and DTGS detector. The spectrum of the gas was
obtained with the samples contained in 12 cm cells equipped with CsI windows.
Atmospheric water vapor was removed from the spectrometer chamber by purging with dry
nitrogen. Interferograms obtained after 128 scans for the gas sample and the background
reference were transformed by using a boxcar apodization function with theoretical
resolutions of 0.5 cm-1 for the gaseous sample.
The mid-infrared spectra of cyclopropylisocyanate dissolved in liquefied xenon (Fig.
31B) were recorded on a Bruker model IFS-66 Fourier interferometer equipped with a
Globar source, Ge/KBr beamsplitter and DTGS detector. The interferograms were recorded
at variable temperatures ranging from 55 to 100 C with 100 scans and transformed by a
185
Blackman-Harris apodization function with a theoretical resolution of 1.0 cm-1.
The
temperature studies in liquefied xenon were carried out in a specially designed cryostat cell,
which is composed of a copper cell with a 4 cm path length and wedged silicon windows
sealed to the cell with indium gaskets. The temperature was monitored by two platinum
thermoresistors and the cell was cooled by the vapors from boiling liquid nitrogen. All of the
observed fundamental modes for the trans and cis conformers in the infrared and Raman
spectra are listed in Tables 35 and 36, respectively.
The Raman spectra of the liquid (Fig. 32) was recorded on a Spex model 1403
spectrophotometer equipped with a Spectra-Physics model 2017 argon ion laser operating on
the 514.5 nm line. The laser power used was 0.5 W with a spectral bandpass of 3 cm -1. The
spectrum of the liquid was recorded with the sample sealed in a Pyrex glass capillary. The
Raman spectrum of the gas (Fig. 33A) at ambient vapor pressure was recorded by using the
standard Cary multipass cell with 1-W laser power at the sample.
The LCAO-MO-SCF calculations were performed with the Gaussian-03 program [43]
by using Gaussian-type basis functions.
The energy minima with respect to nuclear
coordinates were obtained by simultaneous relaxation of all geometric parameters consistent
with symmetry restrictions using the gradient method of Pulay [44]. A number of basis sets
starting from 6-31G(d), and increasing to 6-311+G(3df,3pd), were employed at the level of
Møller-Plesset perturbation theory [38] to the second order (MP2), as well as hybrid density
functional theory by the B3LYP method, to obtain energy differences (Table 37) among the
three most likely conformers of cyclopropylisocyanate. At all levels of calculation carried
out, with and without diffuse functions, predicted conformational stabilities varied
extensively with the size of the basis set.
186
To aid in making the vibrational assignment (Tables 35 and 36), we have carried out a
normal coordinate analysis by utilizing the force fields obtained from the Gaussian-03
program at the MP2(full)/6-31G(d) level. The internal coordinates used to calculate the G
and B matrices for cyclopropylisocyanate are listed along with the structural parameters in
Table 38. By using the B matrix, the force field in Cartesian coordinates was converted to a
force field in internal coordinates [45]. Subsequently, scaling factors of 0.88 for the CH
stretches and 0.90 for all other modes were used, along with the geometric average of scaling
factors for interaction force constants, to obtain the fixed scaled force fields and the resultant
wavenumbers (Table 35). A set of symmetry coordinates (Table 39) was used to determine
the corresponding potential energy distributions (P.E.D.s). The observed and calculated
wavenumbers of cyclopropylisocyanate along with the calculated infrared intensities, Raman
activities, depolarization ratios, and P.E.D.s are given in Tables 35 and 36.
In order to identify the fundamental vibrations for possible conformers of
cyclopropylisocyanate, we have simulated the Raman spectrum (Figs. 32 and 33) from the
scaled ab initio MP2(full)/6-31G(d) results. The evaluation of Raman activity by using the
analytical gradient method has been developed [46, 47]. The activity Sj can be expressed as:
Sj = gj (45
2
j
+ 7
2
j
), where gj is the degeneracy of the vibrational mode j,
derivative of the isotropic polarizability, and
Raman scattering cross sections,
j/
j
j
is the
is that of the anisotropic polarizability. The
, which are proportional to Raman activities, can be
calculated from the scattering activities as well as the predicted wavenumbers for each
normal mode, by using the relationship [48, 49]:
(1−exp( −hc
j
/ kT))] [h / (8π2c j)] Sj, where
0
j/
= [(2π)4/45] [(
0
−
is the excitation wavenumber,
j
4
j)
/
is the
vibrational wavenumber of the jth normal mode, and Sj is the corresponding Raman scattering
187
activity. To obtain the polarized Raman scattering cross sections, the polarizabilities are
incorporated into Sj by multiplying Sj by (1
j)/(1+ j),
where
j
is the depolarization ratio of
the jth normal mode. The Raman scattering cross sections and calculated wavenumbers
obtained from the scaled ab initio force fields were used together with a Lorentzian function
to obtain the simulated Raman spectra.
To further support the vibrational assignments, the infrared spectrum (Fig. 34) was
predicted by using scaled wavenumbers from MP2(full)/6-31G(d) results. Infrared intensities
were calculated based on the dipole moment derivatives with respect to Cartesian coordinates.
The derivatives were transformed into normal coordinate derivatives by (
(
u/
u/
Qi) =
j
Xj) Lij, where Qi is the ith normal coordinate, Xj is the jth Cartesian displacement
coordinate, and Lij is the transformation matrix between the Cartesian displacement
coordinates and the normal coordinates. The infrared intensities were then calculated by
[(N )/(3c2)] [(
x/
Qi)2 + (
y/
Qi)2 + (
z/
Qi)2].
Vibrational Assignment
To determine the enthalpy difference between the two conformers it is necessary to
provide vibrational assignments for the fundamentals of each conformer in the region where
the conformer pairs will be chosen for the enthalpy determination. Since the lower frequency
region will have the fewest number of possible overtones or combination bands, the
assignments for the fundamentals in this region will be very important to make before
selecting pairs of conformer bands. However, because of the large amplitude motion of the
NCO and R-NCO bends many of the bands below 700 cm-1 are very broad in the xenon
solution, which makes it difficult to confidently assign them to the correct conformer.
However, in the region between 700 to 1400 cm-1 most of the fundamentals can be
188
indentified for each conformer; but in several cases the similar vibrations for the two
conformers are observed within a very few wavenumbers from each other as predicted from
the ab initio calculations. To make the assignments, considerable reliance was placed on the
predicted fundamental frequencies and intensities from the ab initio calculations as well as
the infrared band contours in the gas.
From these experimental data most of the
fundamentals in the 700 to 1400 cm-1 could be assigned with confidence.
The assignments for the fundamentals for the trans form are shown in Table 35 and
for the cis conformer in Table 36. For some of the fundamentals with B-type band contours
it was very difficult to determine the band center, but for some cases the infrared spectrum of
the xenon solution could aid significantly in the band center determination. Nevertheless,
there is a broad non-descript band centered at 1106 cm-1 where two fundamentals for each
conformer are predicted. The assignments of the fundamentals are somewhat arbitrary since
there is little band contour or intensity information to aid in the identification of the band
centers. A similar problem arises in the region 1130 to 1220 cm-1 which makes it difficult to
confidently assign the four fundamentals predicted in this range at 1207 (ν8, trans), 1173 (ν20,
trans), 1199 (ν8, cis) and 1172 (ν20, cis). All except the 1199 cm-1 band are predicted to have
very low intensities and the band contours in this region are very ill defined. Therefore, the
infrared spectrum of the xenon solution and Raman spectrum of the liquid were needed to
determine the band centers. Only two bands (1198 and 1167 cm-1) can be readily observed in
the xenon solution while only one very strong band (1200 cm-1) is observed in the Raman
spectrum of the liquid. The band in the xenon solution at 1198 cm-1 is assigned as the ν8 (cis)
mode and the 1167 cm-1 bands is assigned as the ν20 (cis) mode. With these assignments,
corresponding band centers were identified in the infrared spectrum of the gas at 1197 and
189
1175 cm-1, respectively. The corresponding Raman band at 1200 cm-1 in the spectrum of the
liquid supports these assignments and the weaker Raman line at 1172 cm-1 is consistent with
the lower frequency assignment.
Enthalpy Difference
The initial
H determination [160] was made from a single conformer pair from
Raman data of the liquid. Unfortunately, the band for the cis conformer was due to stray
light so its intensity did not decrease when the temperature was lowered. Since it did not
decrease, the H was determined from essentially the increase of the fundamental due to the
trans form when the temperature was lowered so the reported value is expected to be slightly
smaller than the actual value. Therefore we have determined the enthalpy difference by a
variable temperature study of the xenon solution.
For the best results the lowest frequency fundamentals should be used for the
conformer pairs since they have the lowest probability of having intensity contributions from
overtones or combination bands. For the cis conformer the bands at 726 and 787 cm-1 have
been confidently assigned to this form and for the trans form the fundamentals at 810, 915,
928, and 1026 cm-1 were utilized (Fig. 31B). These bands were selected since they are well
resolved and well separated, with relatively flat baselines, and they are without predicted
underlying bands from the other conformer. From these fundamentals for each conformer,
eight conformer pairs were selected for the conformational stability determination. The
intensities of the infrared bands were measured as a function of temperature and their ratios
were determined. By application of the van’t Hoff equation, lnK = H/(RT)
S/R, H
was determined from a plot of lnK versus 1/T (Fig. 35), where H/R is the slope of the line
190
and K is substituted with the appropriate intensity ratios, i.e. Itrans / Icis. It is assumed that H
and ΔS are not functions of temperature in the temperature range studied.
The resulting values with statistical uncertainties of one sigma for each set are listed in
Table 40. The average of the eight values is 77
3 cm-1 (0.92
0.04 kJ/mol)) where the
error limit is derived from the statistical standard deviation of one sigma of the measured
intensity data, where the data are taken as a single set. Although the statistical uncertainties
are relatively small, they do not take into account possible contribution from combination or
overtone bands from other conformers contributing to the measured fundamental band
intensities. The variations of ΔH values are undoubtedly due to these types of interferences
but by taking eight conformer pairs it is expected that these effects are nearly cancelled.
However, the nature of the technique utilized for enthalpy determination is not expected to
provide data better than 10%. Therefore, a more realistic ΔH value is 77
8 cm-1 (0.92
0.10 kJ/mol)). From this ΔH value, a realistic estimation of the abundance of the less stable
cis conformer present at ambient temperature is 41 ± 2 %.
Structural Parameters
In the initial microwave study [160] only the r(C1 N),
C2,3C1N and
C N=C
were determined for both conformers from the three rotational constants for each form with
the other parameters estimated from related molecules. The larger C-N distance and the
smaller CCN angle of the trans conformer compared to the corresponding parameters of the
cis form are consistent with the expected difference of these parameters. However, the two
A rotational constants had rather large uncertainties which could have affected the values of
the three determined parameters. Additionally, the values estimated from those obtained
from other molecules may not be comparable to the values of similar parameters for
191
cyclopropylisocyanate. Since bond lengths in the same set keep their relative ratio which
results in only four heavy atom distances for cyclopropylisocyanate, bond angles and
torsional angles in the same set keep their difference in degrees. Additionally, we have also
shown that the difference in predicted distances and angles from the ab initio calculations for
different conformers of the same molecule can usually be used as one parameter, namely the
ab initio predicted difference except for some dihedral angles. However, for the trans and cis
conformers these dihedral angles are zero and 180° so they are not variable parameters. Thus,
there are six heavy atom parameters to be adjusted so it should be possible to obtain
“adjusted r0” structural parameters from the A&M program [29] for cyclopropylisocyanate
by utilizing the six determined microwave rotational constants from the two conformers since
the sets are the same for the two forms.
The determined adjusted r0 parameters are listed in Table 38 and the final fit of the
rotational constants is shown in Table 41. The difference are quite small with all six
rotational constants fit to better than 1.0 MHz. The bond distances that changed the most
during the fitting are: r(C2–C3), r(C1–N) and r(C=O), these adjust r0 parameters differ by at
least 0.010 Å from their predicted values for both of the conformers. The largest adjustment
for the bond angle is associated with C–N=C angle where the adjust r0 is at least 2° larger
than the MP2(full)/6-311+G(d, p) predicted value.
The predicted structural parameters from the MP2(full)6-311+G(d,p) calculations for
both the trans and cis conformers are listed in Table 38 along with those from B3LYP/6311+G(d,p) calculations. These are compared to those previously reported for the two
conformers indicated as gauche-1 and gauche-2 which were obtained [123] from ab initio
predictions from MP2/6-31G(d,p) calculations. By using the dihedral C=N C C2 angle as
192
an indication of the conformation, the gauche-1 form reported differs by only 0.7° from this
angle for the trans form and only 0.1° difference for the assigned gauche-2 form for this
angle for the cis conformer. Further clarification concerning the incorrect dihedral angle that
was used for determining the gauche forms versus cis or trans will be provided in the
discussion
Discussion
The vibrational assignments given earlier [160] were primarily based on infrared
band contours, Raman depolarization data and group frequencies. For those presented in the
current study there is significantly more information, which includes the predicted
frequencies for the fundamentals, the infrared intensities and Raman activities, with probably
the most important data being the infrared spectra of variable temperature xenon solution
from which the bands for the two different conformers could be indentified. Therefore, a few
of the previously assigned bands as fundamentals, such as the 1005 cm-1 band due to stray
light, as well as the bands at 823 and 351 cm-1 have been reassigned. There are some
differences since the band centers have different frequencies in some cases.
A major problem arises due to the significant breadth of several of the fundamentals
of the trans conformer, particularly those arising mainly from motions of the NCO group.
Most of the fundamentals for the cis conformer are very broad, which is probably due to only
two energy levels in the potential well for this conformer since transitions for only one
excited state were observed in the microwave spectrum. For the trans conformer four
excited states of the low frequency NCO bend (torsion) were obtained. Therefore many of
the fundamentals for the cis conformer have bands with nondescript contours.
193
The ab initio predicted frequencies for the fundamentals of the trans form with two
scaling factors of 0.88 for the CH stretches, 0.90 for the CH bends and heavy atom stretches
and no scaling for the other modes agreed with the modes of A' symmetry with an average
frequency difference of 8.0 cm-1, which is 0.6% error. For the A" modes the average
difference is 9.1 cm-1 (0.8%) which indicates the value of using the relatively small basis set
for aiding the assignment.
The descriptions for the vibrational motions usually have only one or two symmetry
coordinate contributions with four modes having contributions from three symmetry
coordinates and only one (ν24) with four contributors with the largest one contributing only
32%. This latter mode has nearly equal contributions from the CH2 rock and the ring
deformation. Also, it should be noted that ν13 is described as the C−N stretch but only 23%
of S23 contributes to the 730 cm-1 band assigned to this motion, which also has a significant
contribution from the NCO symmetric stretch, and lesser amounts from ν11 and ν12.
Nevertheless, for most of the vibrations the approximate description gives the major
contribution to that vibration.
The enthalpy value obtained from the xenon solution is expected to be close to the
value for the gas since only small interactions are expected to occur between the dissolved
molecules and the surrounding noble gas atoms [22-25]. Further, the “pseudo gas phase”
spectrum shows only small frequency shifts compared to the spectrum of the gas.
A
significant advantage of this type of cryogenic spectroscopic study is that the conformer
bands are better resolved in comparison with those in the spectrum of the gas. This is
particularly important since most of the conformer bands for this molecule are expected to be
observed within a few wavenumbers of each other. The different conformer bands which can
194
be clearly identified in the spectral region between 1100 and 600 cm-1 of the xenon solution
are shown in Figure 30 and the ones indicated are due to the individual conformers. The
extremely small uncertainty in ΔH of ± 3 cm-1 (0.04 kJ/mol) is probably smaller than the
technique justifies but it is the statistical value.
Nevertheless, the determined value is
certainly more accurate than the value that could have been obtained for the gas.
There is support for the very small enthalpy value from the ab initio predicted values
where the MP2(full)/aug-cc-PVTZ calculation with the largest basis set of 391
wavefunctions predicts an energy difference of only 20 cm-1.
It is expected that this
estimated value is the most reliable one because of the large basis set [163]. Although large
basis sets are necessary for the determination of the energy difference between conformers,
much smaller ones can be utilized for the determination of quadrupole coupling constants
and usually for the centrifugal distortion constant as well. Satisfactory quadrupole coupling
constant values have been obtained for both the trans and cis conformer (Table 42) and
centrifugal distortion constant values fro the trans form.
The structural parameters obtained from the earlier microwave study [160] were
obtained from the three rotational constants for each conformer. The parameters for the
hydrogen atoms were fixed at the value reported from these parameters for cyclopropyl
chloride [164] and the structure for the NCO group from the reported parameters of vinyl
isocyanate [165, 166] and isocyanic acid [62]. The r(C1–C2) value was estimated from the
planar moment [167] and the C2–C3 distances of 1.515 Å for this molecule were taken from
those of cyclopropylisothiocyanate (electron diffraction [168]). With these parameters fixed,
the r(C–N),
CNC, and
CCN were obtained for both conformers by diagnostic least-
squares adjustment [169]. The previously assumed and determined parameters are listed in
195
Table 38 and there are some significant differences from the parameters obtained in this
study. For example, the assumed r(C1–C2,3) and r(C=O) seem entirely too long whereas
r(C1–H) is too short. Also, the determined difference of 0.010 Å between the two values of
r(C1-N) is too large since the predicted ab initio value is only 0.001 Å! Although the ab initio
predicted value for the distance for one of the conformers may differ significantly from the
experimental value, approximately the same difference is expected for the other conformer so
the predicted difference is usually quite accurately predicted. However, the listed uncertainty
for the r(C1–N) distance could be interpreted as being essentially the same distance. The
other parameter where there is a significant difference between the estimated value and the
determined distance in this study is r(C=O) which is significantly shorter from this study.
However, the previously determined
C2,3C1N and
C–N=C values are in good agreement
with the value obtained for the corresponding parameters in the current study.
In the second microwave study [124], the scientists were mainly interested in the
value of the centrifugal distortion constants. The experimentally determined values of the
quadratic distortion constants for the trans form are listed in Table 42 along with the value
obtained from the force constants obtained from the ab initio and the density functional
theory calculations. Often the prediction from B3LYP calculations are in better agreement
than those from the MP2 prediction but in this case MP2/6-311+G(d,p) values were quite
superior - particularly for DJK and DK. However, for the cis form reasonable centrifugal
distortion values could not be obtained even by using the fourth order fit. In fact, these
investigators [124] initially suggested in the introduction that the conformational behavior of
cyclopropylisocyanate raised some doubts about the existence of the cis form. Also, the
authors [124] attempted to explain the failure of the centrifugal distortion analysis for this
196
conformer by using the basic assumptions of the conventional distortion theory for a rigid
rotor Hamiltonian. For this Hamiltonian, the centrifugal distortion effects are described by a
harmonic potential function and the theory allows only the small displacements of the
internal coordinates from the equilibrium values to be considers. They explain that “These
conditions are obviously not fulfilled for “cis” cyclopropylisocyanate.
Because of the
rigidity of the cyclopropyl frame and of the NCO group especially the torsion around the
Cframe–N bond allows for flexibility of the molecule.
If we assume a very flat and
anharmonic potential near the “cis” conformation with a low barrier to the trans
conformation, the failure of the centrifugal distortion analysis is plausible [124].”
Nevertheless the problem was not the question of cis being the correct conformer, it is due to
the need for a potential function that has been described [110] as like an “champagne bottle.”
This type of potential simply arises from the large amplitude bending motion of the C-NCO
linkage near the top of the potential well. It would be of interest to use such a potential to
assign high J-level values, particularly values larger than J=2. It was noted [124] earlier that
for K-1 = 3 the transitions showed a positive deviation whereas those with K- >3 were
characterized by negative deviation.
The authors of the theoretical paper [123] seemingly did not take into consideration
the information from the microwave studies. [124, 160] The dipole moment components
determined [160] from the Stark effects clearly indicated that the trans form had a plane of
symmetry with |μb| = 0 and similarly for the cis conformer with |μc| = 0. The optimization of
the two possible gauche equilibria [123] from the MP2/6-31G(d,p) calculation gives the
trans and cis conformations unless the gauche conformers are fixed as the structures.
Apparently, from the earlier theoretical studies [123] the investigators really did obtain the
197
trans and cis conformers. In fact, the authors [123] even acknowledged the similarity of the
two “gauche equilibrium structures” with the trans and cis forms of the previous study [160].
For the actual gauche and skew forms, the dipole moments obtained would have all three
components unequal to zero (Table 43). A more appropriate choice of the dihedral angle used
would be C=N–C–H4 instead of C=N–C–C2. When C=N–C–H4 is used as the dihedral angle,
the equilibrium structure for the trans and cis conformers would have a plane of symmetry
and it would also clearly show how the NCO group is oriented relative to the ring if the
molecule is in the two “gauche equilibrium structures.”
We have carried out the MP2(full)/6-31G(d,p) calculation utilized in the earlier
theoretical study [123]. In Fig. 36, the potential functions of MP2(full)/6-31G(d,p) and
MP2(full)/6-311G(2d,2p) are shown, and the smaller basis set gives a much broader and
shallower potential well for the trans conformer then the MP2(full)/6-311G(2d,2p)
calculation. The depth of the well for the cis form is approximately the same from both
calculations. Due to the breadth of the of this potential function from the smaller basis set, it
is possible that the method utilized would make it difficult to determine the minimum since
only two points were used between 90º and 180º. Nevertheless, the MP2(full)6-31G(d, p)
predicted the minimum at the trans and cis position. If a more appropriate dihedral angle
was used for indicating the conformers, the MP2(full)6-31G(d,p) from the theoretical study
[123] would have given the same result as this current study.
198
Table 35.
Calculated and Observed Vibrational Frequencies (cm-1) of Cyclopropylisocyanate for the Trans Conformer at the
MP2(full)/6-31G(d) level.
199
Vib.
Approx. description
No
A¢
n1 CH2 antisym stretch
n2 CH stretch
n3 CH2 symmetric stretch
n4 NCO antisym stretch
n5 CH2 deformation
n6 NCO symmetric stretch
n7 CH bend
n8 ring breathing
n9 CH2 twist
n10 CH2 wag
n11 ring deformation
n12 CH2 rock
n13 ringC-N stretch
n14 NCO bend
n15 CC-N bend
n16 CNC bend
Table Continues
MP2
MP2
IR Raman dp
Infrared
a
b
c
scaled int.
act. ratio Obs.
Xe
Raman
d
gas liquid
3320
3229
3218
2410
1568
1515
1402
1272
1176
1111
979
826
761
616
371
130
3114
3029
3019
2286
1488
1437
1330
1207
1115
1054
929
784
724
614
371
130
3107 3095
3034 3025
−
−
2289 2281
1482 1472
1449 1445
−
~1350
1201 1200
−
−
1034
−
920
913
797
−
734
730
635
632
367
376
−
−
5.2 34.9 0.73 3109
11.0 125.6 0.12 3036
0.9 98.3 0.06
−
938.4
3.7 0.16 2281
32.8 29.2 0.34 1480
18.1 56.1 0.23 1446
53.0
8.3 0.74 1344
0.3 16.9 0.15 1205
4.2
4.1 0.62 1116
11.6
0.3 0.75 1032
7.4
9.1 0.60 918
1.4
3.0 0.28 ~800
7.9 11.7 0.42 733
23.1
2.0 0.65 630
12.4
3.5 0.31 369
6.7
2.6 0.72 126d
3097
3034
3018
2277
1476
1448
1342
1201
1115
1026
915
~800
735
−
−
−
PED
e
Contour
A
C
99S1
43 57
89S2, 10S3
35
65
88S3, 11S2
59
41
98S4
99
0
62S5, 18S8
100
0
39S6, 26S5, 25S13
80
20
64S7, 20S6
99
1
65S8
96
4
38S9, 21S12, 16S7
30
69
92S10
72
28
67S11, 13S13
10
90
67S12, 13S13
0 100
23S13, 34S9, 16S11
2 98
82S14
52
48
77S15
93
7
88S16
100
0
Table 35 Continues
Vib.
Approx. description
No
A¢¢
n17 CH2 antisym stretch
n18 CH2 symmetric stretch
n19 CH2 deformation
n20 CH2 twist
n21 CH bend
n22 CH2 wag
200
a
MP2
MP2
IR Raman dp
Infrared
a
b
c
scaled int.
act. ratio Obs.
Xe
3309
3215
1515
1236
1158
1110
3104
3016
1437
1173
1099
1053
0.4
6.6
3.2
0.2
1.0
3.6
n23 ring deformation
990
939
11.4
11.7 0.75
936
n24 CH2 rock
854
810
8.7
5.2 0.75
811
810
n25 NCO bend
n26 R-NCO bend
n27 ringC-N bend
543
413
12
543
413
12
15.6
2.9
0.5
0.7 0.75
0.4 0.75
3.0 0.75
566
416
−
566
418
−
76.5
27.6
10.0
7.8
1.9
0.2
−
0.75
−
0.75
0.75
−
0.75
−
0.75
−
0.75 1050
−
−
−
1167
1091
1045
~930
Force constant scaling factors: 0.88 for CH stretches; 0.9 for all other modes.
Infrared intensities in km/mol.
c
Raman activities in Å4/u.
d
Ref. [160]
e
PEDs from MP2 scaled calculation: values less than 10% are omitted.
b
Raman
gasd liquid
−
−
−
−
−
−
−
−
−
−
−
PED
e
3095 99S17
3021 100S18
1445 100S19
1172 52S20, 42S24
−
69S21, 22S20
−
94S22
932 64S23, 14S26,
12S24
819 32S24, 30S23,
20S21, 16S20
570 100S25
−
87S26, 10S24
−
100S27
Contour
A
C
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
Table 36. Calculated and Observed Vibrational Frequencies (cm-1) of Cyclopropylisocyanate for the Cis conformer at the
MP2(full)/6-31G(d) level.
201
Vib.
No
A¢
n1
n2
n3
n4
n5
n6
n7
n8
Infrared
Obs.
Xe
Raman
d
gas liquid
CH2 antisymmetric stretch
CH stretch
CH2 symmetric stretch
NCO antisymmetric stretch
CH2 deformation
NCO symmetric stretch
CH bend
ring breathing
3313
3259
3214
2406
1564
1515
1439
1264
3108
7.3 32.1
3057
9.7 100.2
3015
0.9 135.4
2282 797.6
2.2
1484
9.2 17.3
1437 15.7 48.5
1366
9.8
7.4
1199
3.4 16.0
0.50
0.26
0.06
0.09
0.52
0.17
0.53
0.19
3102
3053
−
−
1465
1446
−
1197
3097
3047
−
−
1472
1448
−
1198
3107
3054
−
−
1465
−
−
−
n9
CH2 twist
1154
1095
27.3
9.9
0.75 1106
1101
n10 CH2 wag
1107
1050
19.5
0.3
920
0.9
11.9
0.73 1048
−
0.51
1045
−
n11 ring deformation
971
n12 CH2 rock
822
780
8.7
1.6
0.24
788
787
n13 ringC-N stretch
763
725
14.7
9.7
0.32
725
726
n14 NCO bend
n15 CC-N bend
n16 CNC bend
624
424
107
620
424
107
24.1
15.6
3.6
1.0
1.4
3.2
0.74
0.49
0.58
615
−
−
613
−
−
Approx. description
Table Continues
MP2
MP2
IR Raman dp
a
b
c
scaled int.
act. ratio
3095
3053
−
−
1470
1445
−
1200
PED
e
99S1
98S2
99S3
98S4
77S5, 15S8
53S6, 33S13
79S7
63S8, 11S9
36S9, 22S12,
−
1105
14S7, 14S6
−
−
94S10
65S11, 17S13,
920
−
11S6
66S12, 12S13,
792 792
10S9
26S13, 26S9,
−
726
23S11
623
−
78S14, 11S16
−
−
72S15
107
−
88S16, 11S15
Contour
A B
89
59
57
88
23
93
97
42
11
41
43
12
77
7
3
58
97
3
9 91
0 100
81 19
33 67
31 69
64 36
92 8
Table 36 Continues
202
a
Vib.
Approx. description
No
A¢¢
n17 CH2 antisymmetric stretch
n18 CH2 symmetric stretch
n19 CH2 deformation
3302
3210
1520
3098
3011
1442
0.5
6.8
3.5
73.2
21.0
8.2
n20 CH2 twist
1235
1172
0.5
7.9
n21 CH bend
n22 CH2 wag
1169
1108
1109
1051
0.7
3.7
n23 ring deformation
991
940
n24 CH2 rock
856
n25 NCO bend
n26 R-NCO bend
n27 ringC-N bend
551
415
46
MP2
MP2
IR Raman dp
a
scaled int.b act.c ratio
Infrared
Obs.
Xe
−
−
1448
−
−
−
0.75 1175
1167
−
1.2
0.1
0.75 1112
0.75 1031
1112
1026
−
−
10.7
8.2
0.75
930
928
−
812
8.6
5.9
0.75
805
805
−
551
315
46
18.2
2.4
1.1
0.5
0.6
2.4
0.75
0.75
0.75
571
408
~40d
566
405
−
−
−
−
0.75
0.75
0.75
−
−
−
Force constant scaling factors: 0.88 for CH stretches; 0.9 for all other modes.
Infrared intensities in km/mol.
c
Raman activities in Å4/u.
d
Ref. [160]
e
PEDs from MP2 scaled calculation: values less than 10% are omitted.
b
Raman
gasd liquid
PED
e
3095 100S17
−
100S18
1445 100S19
45S20, 44S24,
1172
10S21
−
68S21, 27S20
−
94S22
63S23, 14S26,
−
12S24
33S24, 31S23,
−
17S20, 16S21
−
100S25
−
88S26
−
100S27
Contour
A B
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
Table 37.
Calculated Electronic Energies (hartree) and Energy Differencesa (cm-1) for the
Trans, Cis and the Skew conformers of Cyclopropylisocyanate.
Method / Basis set
Basis Functions
Trans
Cis
Skew
MP2(full)/6-31G(d)
100
-284.531149
52
105
MP2(full)/6-31+G(d)
124
-284.548210
95
173
MP2(full)/6-31G(d,p)
115
-284.572075
57
112
MP2(full)/6-31+G(d,p)
139
-284.588575
97
184
MP2(full)/6-311G(d,p)
138
-284.767042
36
90
MP2(full)/6-311+G(d,p)
162
-284.776779
7
124
MP2(full)/aug-cc-PVDZ
183
-284.623931
19
84
MP2(full)/6-311G(2d,2p)
183
-284.839084
75
122
MP2(full)/6-311+G(2d,2p)
207
-284.847659
54
121
MP2(full)/6-311G(2df,2pd)
250
-284.943928
39
97
MP2(full)/6-311+G(2df,2pd)
274
-284.951234
16
96
MP2(full)/6-311G(3df,3pd)
295
-284.968773
72
129
MP2(full)/aug-cc-PVTZ
391
-285.960505
53
133
B3LYP/6-31G(d)
100
-285.372666
203
237
B3LYP/6-31+G(d)
124
-285.384722
250
284
B3LYP/6-311G(d,p)
138
-285.453174
257
279
B3LYP/6-311+G(d,p)
162
-285.458582
232
264
B3LYP/6-311G(2d,2p)
183
-285.461373
271
285
B3LYP/6-311+G(2d,2p)
207
-285.466843
226
250
B3LYP/6-311G(2df,2pd)
250
-285.472328
243
268
B3LYP/6-311+G(2df,2pd)
274
-285.477358
209
241
B3LYP/6-311G(3df,3pd)
295
-285.476357
256
285
B3LYP/6-311+G(3df,3pd)
319
-285.480813
198
233
a
b
All energy differences are relative to the energy of the trans conformer.
Transition state.
203
b
Table 38.
a
Parameter
204
r(C1–C2,3)
r(C2–C3)
r(C1–N)
r(N=C)
r(C=O)
r(C1–H4)
r(C2,3–H8,10)
r(C2,3–H9,11)
ÐC2C1C3
ÐC2,3C1N
ÐC-N=C
ÐN=C=O
ÐNC1H4
ÐC2,3C1H
C=N-C1-C2
A
B
C
|ma|
|mb|
|mc|
|mtot|
a
b
Structural parameters (å and degree), rotational constants (mhz) and dipole moments (debye) for the trans and cis
cyclopropylisocyanate from ab initio calculations (6-311+g(d,p) basis set) and experimental data.
Int.
Coord
R1, R2
R3
R4
R5
R6
r5
r1, r 3
r2, r 4
π1
e1, e2
ψ
q
z
s1, s2
Trans
MP2 B3LYP
1.504
1.512
1.425
1.216
1.179
1.085
1.083
1.083
60.4
117.2
134.3
172.9
115.4
117.9
145.6
17016
1769
1706
2.997
¾
1.061
3.179
1.507
1.510
1.425
1.200
1.174
1.084
1.082
1.083
60.1
118.1
139.0
174.1
114.5
117.7
145.4
17190
1758
1693
2.966
¾
0.867
3.090
Gauche-1
a
1.499
1.504
1.421
1.215
1.190
1.081
1.078/1.079
1.079
60.2
117.4
137.4
172.7
115.2
117.9
144.9
17218
1753
1691
2.91
Cis
MP2 B3LYP
1.505
1.516
1.423
1.215
1.180
1.082
1.084
1.083
60.4
120.2
135.3
172.9
112.0
117.6
324.4
9997
2208
2114
3.288
0.079
¾
3.289
1.508
1.513
1.421
1.198
1.175
1.082
1.083
1.083
60.2
121.1
142.1
174.2
111.5
117.1
324.1
10706
2085
2029
3.217
0.076
¾
3.218
Ref. [123], Optimized geometry determined from MP2/6-31G(d).
Structural parameter from Ref. [160]
MWb
Gauche-2
a
Trans
Cis
1.501
1.506
1.419
1.216
1.191
1.079
1.079
1.079
60.2
120.6
137.9
172.7
1.520
1.515
1.417(10)
1.210
1.170
1.079
1.086
1.082
1.520
1.515
1.407(14)
1.210
1.170
1.079
1.086
1.082
117.5
324.3
10346
2139
2063
116.1
2.99
117.9(11)
136.9(38)
172.6
16691(14)
1784.303(3)
1716.133(3)
2.56(2)
¾
0.71(3)
2.65(2)
Adjusted r0c
Trans
1.509(3)
1.523(3)
1.412(3)
1.214(3)
1.163(3)
1.085(2)
1.083(2)
1.083(2)
60.6(5)
120.3(24)
116.7(5)
138.6(13)
136.3(5)
172.6
172.2(5)
115.4(5)
116.1
117.9(5)
145.6(5)
10161(13)
16941.7
2186.797(10) 1783.4
2106.242(10) 1717.0
2.720(4)
0.17(0.01)
¾
2.726(1)
Cis
1.509(3)
1.521(3)
1.411(3)
1.212(3)
1.164(3)
1.082(2)
1.084(2)
1.083(2)
60.5(5)
120.1(5)
137.6(5)
173.0(5)
112.0(5)
117.7(5)
324.4(5)
10214.9
2186.6
2106.5
Table 39.
Symmetry coordinates for cyclopropylisocyanate.
a
Species
Symmetry Coordinate
A¢
n1 CH2 antisym str
r1 - r2 + r3 - r4
n2 CH stretch
r5
n3 CH2 sym stretch
r1 + r 2 + r 3 + r 4
n4 NCO antisym str
R5-R6
CH
deformation
n5
4g - a1 - a2 - b1 - b2 + 4g¢ - a3 - a4 - b3 - b4
2
R5+R6
n6 NCO sym str
n7 CH bend
2z - s1 - s2
n8 ring breathing
R1 + R 2 + R 3
n9 CH2 twist
a1 - a2 - b1 + b2 + a3 - a4 - b3 + b4
CH
wag
n10
a1 + a2 - b1 - b2 + a3 + a4 - b3 - b4
2
n11 ring deformation
2R3 - R1 - R2
n12 CH2 rock
a1 - a2 + b1 - b2 + a3 - a4 + b3 - b4
n13 C-N stretch
R4
n14 NCO bend
q
R-NCO
bend
n15
e1 + e2
ψ
n16 CNC bend
A¢¢
n17 CH2 antisym str
r1 - r 2 - r 3 + r 4
n18 CH2 sym str
r1 + r2 - r3 - r4
n19 CH2 deformation
4g - a1 - a2 - b1 - b2 - 4g¢ + a3 + a4 + b3 + b4
n20 CH bend
s1 - s2
CH
twist
n21
a1 - a2 - b1 + b2 - a3 + a4 + b3 - b4
2
n22 CH2 wag
a1 + a2 - b1 - b2 - a3 - a4 + b3 + b4
n23 ring deformation
R1 - R2
n24 CH2 rock
a1 - a2 + b1 - b2 - a3 + a4 - b3 + b4
φ
n25 NCO bend
R-NCO
bend
n26
e1 - e2
χ
n27 CN bend
a
symmetry species for the trans conformer.
205
Table 40. Temperature and Intensity Ratios of the Conformer Bands of Cyclopropylisocyanate from the Infrared Spectra of
the Liquid Xenon Solution.
T(°C)
-3
-1
1/T (´10 K )
I810 / I726
206
-60.0
4.6915
2.6141
-65.0
4.8042
2.6281
-70.0
4.9225
2.6721
-75.0
5.0467
2.6770
5.1773
2.7893
-80.0
-85.0
5.3149
2.8239
-90.0
5.4600
2.8493
-95.0
5.6132
2.9066
-100.0
5.7753
2.9042
DHa
77 ± 7
a
Average value obtained by utilizing all
trans conformer more stable.
I810/ I787
I928 / I726
12.7273
1.9835
13.1950
2.0410
13.4683
2.0661
13.1596
2.0891
13.6409
2.0722
13.9234
2.0954
14.0000
2.1005
14.5783
2.1923
14.6962
2.2556
87± 9
66 ± 10
data as a single set gives
I928/ I787
4.9618
4.7980
4.9505
5.0292
5.1135
5.2184
5.2857
5.3826
5.5291
82 ± 9
DH = 77 ±
I1026 / I726
I1026/ I787
I1342 / I726
I1342/ I787
8.0745
36.1111
5.3114
23.3764
8.0116
37.5000
5.3696
23.8948
8.0674
38.5366
5.3331
23.9598
8.2652
39.0986
5.4850
24.0773
8.5417
38.8182
5.5966
24.2691
8.7174
39.8649
5.6052
25.1982
8.6462
40.7333
5.6632
25.6448
8.8091
41.7391
5.7683
26.3534
8.8769
41.7722
5.8333
26.6806
73 ± 9
88 ± 9
63 ± 5
87 ± 7
-1
3 cm (0.92 ± 0.03 kJ/mol) with the
Table 41.
Isotopomer
Trans
Cis
a
Comparison of rotational constants (mhz) obtained from modified ab initio
MP2(full)/6-311+g(d,p) predictions, microwave spectra, and the adjusted
structural parameters for cyclopropylisocyanate.
Rotational
constant
MP2(full)/6- Microwavea
311+G(d,p)
Adjusted r0
|D|
A
17018.0
16941.889(8)
16941.4
0.5
B
1769.3
1784.3121(6)
1783.4
0.9
C
1706.2
1716.1278(6)
1717.0
0.9
A
10006.4
10215(6)
10214.9
0.1
B
2206.2
2186.859(6)
2186.6
0.3
C
2113.2
2106.172(6)
2106.5
0.3
Rotational constants obtained from a fourth order centrifugal distortion fit, Ref. [124].
207
Table 42. Rotational (mhz), centrifugal distortion (khz) and quadrupole coupling constants for the trans and cis
conformers of cyclopropylisocyanate.
Trans
MP2/631G(d)
208
a
MP2/6- B3LYP /6311+G(d,p) 311+G(d,p)
Cis
MWa
MP2/631G(d)
MP2/6B3LYP /6311+G(d,p) 311+G(d,p)
MWa
A
B
C
DJ
DJK
DK
D1
D2
17221
1753
1691
0.115
23.670
-2.843
-0.002
-0.002
17016
1769
1706
0.114
47.713
-22.405
-0.001
-0.011
17191
1758
1693
0.160
2.319
19.965
-0.002
0.017
16941.889(8)
1784.3121(6)
1716.1278(6)
0.153(2)
52.28(3)
-6(3)
-0.027(1)
10359
2137
2062
1.504
3.140
57.49
-0.22
0.07
9997
2208
2114
1.448
-5.393
51.83
-0.23
0.03
10705
2085
2029
1.550
7.270
72.69
-0.22
0.07
10215(6)
2186.859(6)
2106.172(6)
2.01(1)
29.5(3)
n.d
-0.31(2)
χaa
χbb
χcc
2.9179
-1.7909
-1.1270
2.9715
-1.8545
-1.1170
2.9504
-1.5362
-1.4142
2.6306(26)
-1.3839(31)
-1.2467(31)
2.6501
-0.7920
-1.8581
2.8092
-0.8843
-1.9249
2.8517
-1.2248
-1.6269
2.5647(49)
-1.0624(75)
-1.5022(75)
Ref. [124]; n.d.- not determined
Table 43.
Parametera
A comparison of cyclopropylisocyanate structural parameters (å and degree), rotational constants (Mhz) and
dipole moments (debye) for the different conformers from MP2/6-31g(d,p) calculation.
Int.
Coord.
R1, R2
R3
R4
R5
R6
r5
r1, r 3
r2, r 4
209
r(C1–C2,3)
r(C2–C3)
r(C1–N)
r(N=C)
r(C=O)
r(C1–H4)
r(C2,3–Htrans)
r(C2,3–Hcis)
π1
ÐC2C1C3
ÐC2,3C1N
e1, e2
ψ
ÐC-N=C
ÐN=C=O
q
ÐNC1H4
z
C=N-C-C2
C=N-C-H4 (δ)
A
B
C
|ma|
|mb|
|mc|
|mtot|
-(E+284)
a
Trans
Gauche-1b
1.501
1.504
1.424
1.218
1.191
1.082
1.079
1.080
60.2
117.4
136.3
172.5
115.2
145.6
0.0
17216
1753
1691
3.036
0.000
1.032
3.207
0.544698
1.499/1.499
1.504
1.421
1.215
1.190
1.081
1.078/1.079
1.079
60.2
117.4
137.4
172.7
144.9
0.832
17218
1753
1691
0.5446979
Cis
Gauche-2b
1.502
1.501/1.501
1.507
1.506
1.422
1.419
1.218
1.216
1.192
1.191
1.080
1.079
1.081
1.079/1.079
1.080
1.079
60.2
60.2
120.7
120.6
137.4
137.9
172.5
172.7
111.6
324.3
324.3
180
179.96
10286
10346
2146
2139
2067
2063
3.298
0.084
0.000
3.300
0.5444754 0.5444755
Gauche
Skew
1.503/1.497 1.502/1.502
1.506
1.507
1.425
1.423
60.3
285.7
72.8
14331
1884
1783
3.028
0.441
0.823
3.169
0.5443916
60.2
0.0
146.0
10788
2107
2011
3.257
0.210
0.337
3.281
0.5442258
Adjusted r0d
Trans
1.509(3)
1.523(3)
1.412(3)
1.214(3)
1.163(3)
1.085(2)
1.083(2)
1.083(2)
60.6(5)
116.7(5)
136.3(5)
172.2(5)
115.4(5)
145.6(5)
0.0
16941.7
1783.4
1717.0
Cis
1.509(3)
1.521(3)
1.411(3)
1.212(3)
1.164(3)
1.082(2)
1.084(2)
1.083(2)
60.5(5)
120.1(5)
137.6(5)
173.0(5)
112.0(5)
324.4(5)
180.0
10214.9
2186.6
2106.5
Cis and trans refers to the orientation of the hydrogens relative H4. b Ref. [123], Optimized geometry determined from MP2/631G(d). c Structural parameter from Ref. [160]. d Rotational constants utilized for the fitting are from Ref. [124]
Figure 30. Possible conformations of cyclopropylisocyanate, δ indicates the dihedral angle
C=N–C–H4.
210
Figure 31. Infrared spectra of cyclopropylisocyanate: (A) gas; (B) xenon solution at -100 C.
The asterisk indicates the spectral region at ~600 cm-1 has nearly zero energy
due to the absorption from the silicon windows.
211
Figure 32.
Observed (room temperature) and predicted (MP2(ful1)/6-31G(d) at 25°C)
Raman spectra of cyclopropylisocyanate: (A) liquid; (B) predicted spectrum of
the mixture of cis and trans conformers with H = 77 cm-1; (C) predicted
spectrum of the pure cis conformer; (D) predicted spectrum of the pure trans
conformer; (E) solid. Asterisk indicates that the intensity has been reduced by
half.
212
Figure 33.
Observed (room temperature) and predicted (MP2(ful1)/6-31G(d) at 25°C)
Raman spectra of cyclopropylisocyanate: (A) gas; (B) predicted spectrum of the
mixture of cis and trans conformers with H = 77 cm-1; (C) predicted
spectrum of the pure cis conformer; and (D) predicted spectrum of the pure
trans conformer.
213
Figure 34.
Observed and predicted (MP2(ful1)/6-31G(d)) infrared spectra of
cyclopropylisocyanate: (A) xenon solution at 100 C; (B) predicted spectrum of
the mixture of cis and trans conformers with H = 77 cm-1; (C) predicted
spectrum of the pure cis conformer; and (D) predicted spectrum of the pure
trans conformer.
214
-0.5
-1.0
ln(Itrans/Icis)
-1.5
810/726
810/787
928/726
928/787
1026/726
1026/787
1342/726
1342/787
-2.0
-2.5
-3.0
-3.5
-4.0
4.5
4.7
4.9
5.1
5.3
-3
5.5
5.7
5.9
-1
1/T (x10 K )
Figure 35. van’t Hoff plot of the eight infrared conformer bands utilized in the determination
of enthalpy difference value for cyclopropylisocyanate dissolved in liquid xenon
solution.
215
200
180
160
MP2(full)/6-311G(2d,2p)
V( ), cm
-1
140
120
100
80
216
60
40
20
MP2(full)/6-31G(d,p)
0
-180
-120
-60
0
60
120
180
DIHEDRAL ANGLE, ( )
Figure 36.
Potential energy curve of cyclopropylisocyanate as a function of the dihedral angle C=N C1 H4 obtained from
MP2(full)/6-31G(d,p) calculations (dashed line) and MP2(full)/6-311G(2d,2p) calculations (solid line).
CHAPTER 10
MICROWAVE, RAMAN, AND INFRARED SPECTRA, R0 STRUCTURAL
PARAMETERS, CONFORMATIONAL STABILITY AND AB INITIO CALCULATIONS
OF CYCLOBUTYLISOCYANATE
Introduction
Initial structural interest in cyclobutane, c-C4H8, was whether the ring was planar or
bent and after it was shown [170] to be bent there was extensive scientific investigations to
determine whether monosubstituted cyclobutanes had equatorial and/or axial conformers at
ambient temperature. Some of the earliest molecules which were investigated for conformers
were the halocyclobutanes (X = Br, Cl, F) where vibrational [171-173] and microwave [174,
175] studies were carried out. From the early vibrational studies it was concluded that both
axial and equatorial conformers were present for all three molecules but only the equatorial
conformer was found in the original conventional microwave study [175] of chloro- and
fluorocyclobutane. Structural parameters were reported for the chloride where in addition to
the rotational constants for the 35Cl species rotational constant from eight of its isotopes were
utilized to obtain the r0 parameters for this molecule. However, the CH distance ranged from
1.090 to 1.110 Å which is a very large difference. An attempt was made to obtain more
realistic parameters by combining ab initio predicted values and the experimentally
determined rotational constants to determine the r0 structural parameters more accurately by
utilization of diagnostic least square [176]. However, many of the A rotational constants of
the isotopomers could not be fitted but from a later microwave study [177] the spectrum of
both the axial and equatorial conformers were assigned and the A rotational constants for the
217
35
Cl and 37Cl isotopomers were significantly different from those previously reported [175].
Therefore, by using these new values of the rotational constants and by combining them with
MP2(full)/6-311+G(d,p) ab initio calculations adjusted r0 structural parameters were obtained.
From the two FT-microwave studies [174, 178] the spectra of both the axial and
equatorial conformers of fluoro- and bromocyclobutane were assigned from which rotational
constants were determined and used to obtain the r0 structural parameters for both of these
molecules.
As a continuation of these conformational and structural studies of these
halocyclobutanes we initiated a spectroscopy investigation of the pseudohalogen, -NCO,
attached to the cyclobutane ring, c-C4H7NCO. Since the CNC(O) angle of CH3NCO and
CH3CH2NCO are relatively large there is the question of whether the axial well would have
the depth to accommodate a ground vibrational state.
In addition to the spectroscopic
investigation, ab initio calculations have been carried out to predict the stability of the
conformations and aid in the vibrational assignment. The results of these spectroscopic and
theoretical studies are reported herein.
Experiment and Theoretical Calculations
The
sample
of
cyclobutylisocyanate
was
prepared
by
the
reaction
of
cyclobutanecarbonyl chloride with diazidodiphenylsilane (PH2Si(N3)2) by using the Curtius
arrangement as previously reported by Kricheldorf and Regel [161, 162]. The sample was
initially purified by trap-to-trap distillation and the final purification was obtained by using a
low-temperature, low-pressure sublimation column. The sample identity was verified by
NMR and infrared spectroscopic data.
The microwave spectra of cyclobutylisocyanate were recorded with a “mini-cavity”
Fourier-transform microwave spectrometer [53, 54] at Kent State University. The Fabry-
218
Perot resonant cavity is established by two 7.5-inch diameter diamond-tip finished aluminum
mirrors with a 30.5-cm spherical radius. The Fabry-Perot cavity resides inside a vacuum
chamber formed by a 6-way cross and a 15-inch long, 8-inch diameter extension tube. One
of the cavity mirrors is formed on an 8-inch diameter vacuum flange and mounted on the 6way cross. The second mirror is mounted on 0.75-inch diameter steel rails that pass through
ball bearing brackets mounted inside the extension arm; a motorized micrometer is used to
position the movable mirror over a two-inch range of travel. The two cavity mirrors are
nominally separated by 30 cm.
The vacuum chamber is pumped by a Varian VHS-6
diffusion pump (2400 L s-1) backed by a two-stage Edwards E2M30 rotary pump.
The cyclobutylisocyanate sample was entrained in a 70:30 Ne-He carrier gas mixture at
2 atm and expanded into the cavity with a reservoir nozzle [54] made from a modified Series9 General Valve. The reservoir nozzle is mounted in a recessed region of the mirror flange,
external to the vacuum chamber, and the expansion passes through a 0.182-inch diameter
hole into the resonant cavity. The center of the expansion is offset from the center of the
mirror by 1 inch.
The sample is irradiated by microwave radiation generated by an Agilent Technologies
E8247C PSG CW synthesizer; details of the irradiation and heterodyne detection circuitry
can be found in Ref. [53]. Labview software controls the timing of the gas and irradiation
pulses, as well as the detection of any free induction decay signal. The software performs
signal averaging and can scan the spectrometer by stepping both the frequency source and the
cavity. Microwave circuit elements allow for a spectral range from 10.5 to 26 GHz. The
digital frequency resolution, governed by the sampling rate and the length of the free
induction decay record, is 2.5 kHz. Rotational transitions are split into Doppler doublets
219
centered at the transition frequency due to the coaxial orientation of the gas expansion to the
cavity axis and the FWHM of each Doppler component is typically 13 kHz. The vacuum
system can accommodate pulse repetition rates of up to 15 s-1 while maintaining a pressure
below 10-4 torr, and the instrument can scan 450 MHz in 6 hours while averaging 100 shots
per scan segment. The frequencies for the measured transitions in the region of 12,000 to
19,000 MHz for the equatorial-trans conformers (Fig. 37) of cyclobutylisocyanate are listed
in Table 44 along with their assignments. Also listed are the frequency differences between
the measured values and the values obtained from the determined rotational constants and the
centrifugal distortion constants (Table 45).
The mid-infrared spectra of the gas (Figs. 38A and 39A) and solid (Figs. 40 and 41)
were recorded from 3500 to 300 cm-1 on a Perkin-Elmer model 2000 Fourier transform
spectrometer equipped with a nichrome wire source, Ge/CsI beamsplitter and DTGS
detector. Atmospheric water vapor was removed from the spectrometer housing by purging
with dry nitrogen. The spectrum of the gas was obtained with the samples contained in 12 cm
cells equipped with CsI windows. Interferograms obtained after 128 scans for the gas sample
and the background reference were transformed by using a boxcar apodization function with
theoretical resolutions of 0.5 cm-1 for the gaseous sample. For the spectrum of the solid,
theoretical resolution of 2 cm-1 was used with 128 interferograms added and truncated. Multiple
annealings were required to obtain satisfactory spectra of the solid.
The mid-infrared spectra of cyclobutylisocyanate dissolved in liquefied xenon (Fig.
38B) were recorded on a Bruker model IFS-66 Fourier interferometer equipped with a
Globar source, Ge/KBr beamsplitter and DTGS detector. The interferograms were recorded
at variable temperatures ranging from 55 to 100 C with 100 scans and transformed by a
220
Blackman-Harris apodization function with a theoretical resolution of 1.0 cm-1.
The
temperature studies in liquefied xenon were carried out in a specially designed cryostat cell,
which is composed of a copper cell with a 4 cm path length and wedged silicon windows
sealed to the cell with indium gaskets. The temperature was monitored by two platinum
thermoresistors and the cell was cooled by the vapors from boiling liquid nitrogen. All of the
observed fundamental modes for the equatorial-trans (Eqt-t), equatorial-gauche (Eqt-g), and
axial-trans (Ax-t) conformers in the infrared and Raman spectra are listed in Tables 46, 47
and 48, respectively.
The Raman spectra of the liquid (Fig. 42) was recorded on a Spex model 1403
spectrophotometer equipped with a Spectra-Physics model 2017 argon ion laser operating on
the 514.5 nm line. The laser power used was 0.5 W with a spectral bandpass of 3 cm -1. The
spectrum of the liquid was recorded with the sample sealed in a Pyrex glass capillary.
The LCAO-MO-SCF calculations were performed with the Gaussian-03 program [43]
by using Gaussian-type basis functions.
The energy minima with respect to nuclear
coordinates were obtained by simultaneous relaxation of all geometric parameters consistent
with symmetry restrictions using the gradient method of Pulay [44]. A number of basis sets
starting from 6-31G(d), and increasing to 6-311G(3df,3pd), were employed at the level of
Møller-Plesset perturbation theory [38] to the second order (MP2), as well as hybrid density
functional theory by the B3LYP method, to obtain energy differences (Table 49) among the
three most likely conformers of cyclobutylisocyanate.
At all levels of calculation carried
out, with and without diffuse functions, predicted conformational stabilities varied
extensively with the size of the basis set.
221
To aid in making the vibrational assignment (Tables 46, 47 and 48), we have carried
out a normal coordinate analysis by utilizing the force fields obtained from the Gaussian-03
program at the MP2(full)/6-31G(d) level. The internal coordinates used to calculate the G
and B matrices for cyclobutylisocyanate are listed along with the structural parameters in
Table 50. By using the B matrix, the force field in Cartesian coordinates was converted to a
force field in internal coordinates.[45] Subsequently, scaling factors of 0.88 for the CH
stretches and 0.90 for all other modes were used, along with the geometric average of scaling
factors for interaction force constants, to obtain the fixed scaled force fields and the resultant
wavenumbers (Tables 46, 47 and 48). A set of symmetry coordinates (Table 51) was used to
determine the corresponding potential energy distributions (P.E.D.s). The observed and
calculated wavenumbers of cyclobutylisocyanate along with the calculated infrared
intensities, Raman activities, depolarization ratios, and P.E.D.s are given in Tables 46, 47 and
48.
To further support the vibrational assignments, the infrared spectrum (Fig. 39) was
predicted by using scaled wavenumbers from MP2(full)/6-31G(d) results. Infrared intensities
were calculated based on the dipole moment derivatives with respect to Cartesian coordinates.
The derivatives were transformed into normal coordinate derivatives by (
(
u/
u/
Qi) =
j
Xj) Lij, where Qi is the ith normal coordinate, Xj is the jth Cartesian displacement
coordinate, and Lij is the transformation matrix between the Cartesian displacement
coordinates and the normal coordinates. The infrared intensities were then calculated by
[(N )/(3c2)] [(
x/
Qi)2 + (
y/
Qi)2 + (
z/
Qi)2].
In order to identify the fundamental vibrations for possible conformers of
cyclobutylisocyanate, we have simulated the Raman spectrum (Fig. 42) from the scaled ab
222
initio MP2(full)/6-31G(d) results. The evaluation of Raman activity by using the analytical
gradient method [46, 47] has been developed. The activity Sj can be expressed as: Sj = gj (45
2
j
2
j
+7
), where gj is the degeneracy of the vibrational mode j,
isotropic polarizability, and
cross sections,
j/
j
j
is the derivative of the
is that of the anisotropic polarizability. The Raman scattering
, which are proportional to Raman activities, can be calculated from the
scattering activities as well as the predicted wavenumbers for each normal mode, by using
the relationship [48, 49]:
Sj, where
0
j/
= [(2π)4/45] [(
is the excitation wavenumber,
j
0−
4
j) /
(1−exp( −hc j / kT))] [h / (8π2c j)]
is the vibrational wavenumber of the jth normal
mode, and Sj is the corresponding Raman scattering activity. To obtain the polarized Raman
scattering cross sections, the polarizabilities are incorporated into Sj by multiplying Sj by
(1
j)/(1+ j),
where
j
is the depolarization ratio of the jth normal mode. The Raman
scattering cross sections and calculated wavenumbers obtained from the scaled ab initio force
fields were used together with a Lorentzian function to obtain the simulated Raman spectra.
Microwave spectra
The microwave spectra of Eqt-t and Eqt-g cyclobutylisocyanate were predicted with
the rotational constants of the estimated structures optimized by MP2(full)/6-311+G(d,p)
calculation. The estimated structures were determined by adjusting the heavy atom bond
lengths and angles from similar four-membered ring molecules to closely reproduce the
experimentally determined rotational constants. These adjusted structural parameters were
used to calculate rotational constants from MP2(full)/6-311+G(d,p) calculations and the
predicted A, B and C values are list in Table 45. Assignments were then made for several
transitions from which A, B and C constants were obtained. Eleven transitions for
cyclobutylisocyanate have been assigned (Table 44) for the most stable conformer Eqt-t.
223
From the data in Table 45, it can be seen that the experimentally determined constants for B
and C were well determined with an uncertainties of 0.03 and 0.02 MHz , respectively,
whereas the A rotational constant has a large uncertainties of 302 MHz.
The dipole moment components were predicted from the MP2(full)/6-311+G(d,p)
calculation to be | a| = 3.432 D, | b| = 0, and | c| = 0.066 D and only A-type transitions were
expected. Nevertheless, the bands were quite difficult to assign and only four “J” transitions
were obtained which could be fitted without the centrifugal distortion constants.
The
centrifugal distortion values were fairly well determined for ΔJ and δJ. However, ΔJK, Δk and
δJ were not determined because the number of higher Ka of the measured transitions was
insufficient to allow reliable determination of all these quartic centrifugal distortion constants.
Vibrational Assignment
Although only one conformer was identified in the microwave spectrum, the infrared
spectra [Fig. 40 and 41] of the amorphous and annealed solid show the presence of two
additional conformers. In order to determine the enthalpy difference between the Eqt-t and
Eqt-g conformers, a relatively complete vibrational assignment for the fundamentals of each
conformer has been carried out. Of particular interest are the frequencies for the heavy atom
ring modes and those for the NCO moiety. In general, five of the six ring modes are usually
bunched over a relatively small frequency range from about 800 to 1100 cm-1 with three of
them giving rise to relatively pronounced infrared bands (Fig. 38). For the Eqt-t conformer,
these fundamental are assigned at 1032 cm-1 for the ring breathing mode and at 922 and 896
cm-1 for the ring deformations. The other two ring deformations are assigned at ~1035 and
906 cm-1 with the latter one giving rise to a relatively strong Raman band. Since there is also
the possibility that there may be one of the CH2 bending modes in this spectral region there is
224
expected mixing with these bending motions so the description may be different for the two
conformers even though the bands are clearly the corresponding vibrations for the
conformers. The corresponding fundamentals for the Eqt-g form are observed at 1043,
~1035, 927, 906 and ~896 cm-1 with only two of these fundamentals having different
frequencies from those of the corresponding modes for the Eqt-t form. Also it should be
noted that the higher frequency 1043 cm-1 band is the ring breathing mode as noted from the
predicted infrared intensity. The assignments for the ring modes was possible based on the
infrared spectrum of the xenon solution along with the determination of the conformer
remained in the infrared spectra of the annealed solid.
The other vibrations of considerable interest are those arising from the NCO moiety.
These nine vibrations are best described as three stretches and six bending modes. The bands
associated with the –NCO modes in the infrared spectrum of the gas are rather nondescription and broad, a spectral characteristic commonly observed for quasilinear molecules.
Thus, the infrared spectrum of xenon solution was very significant in the determination of the
band center and, also, for determining the relative intensities of the bands. It has been well
established where the NCO stretches are with the antisymmetric mode predicted at 2270 cm -1
for the Eqt-t form and it is observed at 2272 cm-1 in the infrared spectrum of the solid. The
NCO symmetric stretch for the Eqt-t is predicted at 1414 cm-1 but it is assigned at a higher
frequency around ~1445 cm-1 in the infrared spectrum of the gas. It is very difficult to
confidently assign this band in the spectrum of the gas since the band width is very broad
with the band center at approximately 1445 cm-1. However, in the infrared spectrum of the
xenon solution a doublet is observed with band centers at 1443 and 1437 cm-1, and they are
assigned to the Eqt-t and Eqt-g forms, respectively. It is interesting to note that the NCO
225
symmetric stretch for both conformers were significantly mixed with Cα-N stretch (Eqt-t,
26S19; Eqt-g 28S31) at 558 and 556 cm-1 in the infrared spectrum of the solid for Eqt-t and
Eqt-g form, respectively. Also, it should be noted that the C N stretches are extensively
mixed with the ring deformations (46S16 for Eqt-t, 39S26 Eqt-g) so the description for this
fundamental is more for bookkeeping than conveying a molecular motion.
The ring-NC in-plane and out-of-plane bends are readily assigned at 448 and 360
cm-1, respectively, for the Eqt-t conformer and the 415 cm-1 band is the in-plane bend for the
Eqt-g form. The out-of-plane bend for the Eqt-g conformer had the same frequency of 360
cm-1 as obtained for this fundamental for the Eqt-t form. The four C N=C bends have not
been assigned but their predicted frequencies are significantly different so it should be
relatively simple to assign them when this spectral region is investigated. Both of these
vibrations for both conformers have significant predicted Raman band intensities but they
were not observed in the Raman spectra. In the low frequency Raman spectrum very, very
broad non-descript scattering was observed.
The Raman spectrum of low temperature
Raman spectrum of xenon solution would be of interest for obtaining information of these
low frequency vibrations.
Conformational Stability
To determine the enthalpy difference between the two conformers, the mid-infrared
spectra of cyclobutylisocyanate dissolved in liquefied xenon as a function of temperature,
from -55 to -100 ºC, were recorded. Only small interactions are expected to occur between
the dissolved molecules and the surrounding noble gas atoms [57, 117-119], and as a result,
the “pseudo gas phase” spectrum shows only small frequency shifts compared with that of
the gas. A significant advantage of this type of cryogenic spectroscopic study is that the
226
conformer bands are better resolved in comparison with those in the spectrum of the gas.
This is particularly important because most of the conformer bands for this molecule were
expected to be observed within a few wavenumbers of each other. The different conformer
bands that are expected to be clearly identified are those in the spectral region below 1000
cm-1 since they have the lowest probability of having intensity contributions from overtones
or combination bands.
The intensities of four well-isolated and well-shaped conformational bands confidently
assigned in the previous section were measured as a function of temperature (at 5.0 C
intervals between 55 and 100 C) and their ratios were determined as listed in Table 52. By
application of the van’t Hoff equation, lnK =
H/(RT)
S/R, enthalpy difference was
determined from a plot of lnK versus 1/T, where H/R is the slope of the line and K is
substituted with the appropriate intensity ratios, i.e. IEqt-t/IEqt-g. It was assumed that H, S
and the ratio of the molar absorption coefficients εEqt-t/εEqt-g are not a function of temperature
in the range studied.
The conformational enthalpy difference (Table 52) was determined to be 131 ± 5 cm-1
(1.57
0.06 kJ/mol) from the xenon solution with the Eqt-t conformer more stable. The
error limit of ± 5 cm-1 is derived from the statistical standard deviation of one sigma of the
measured intensity data, where the data are taken as a single set. Although the statistical
uncertainties are quite small, they do not take into account small associations with the liquid
xenon or the interference of overtones and combination bands in near coincidence with the
measured fundamentals. The variations are undoubtedly due to these types of interferences
but by taking four pairs it was hoped the effect of such interferences would be minimized.
227
Thus, a more realistic error limit is probable 10% and a more appropriate enthalpy difference
value is 131 ± 13 cm-1 (1.57
0.16 kJ/mol).
Structural Parameters
With only two determined rotational constants, it might seem that little structural
parameters could be obtained but with the predicted structural parameters from the Gaussian
MP2(full)/6-311+G(d,p) calculations, adjusted r0 values can be obtained with relatively small
uncertainties. First, we [33] have also shown that ab initio MP2/6-311+G(d,p) calculations
predict the r0 structural parameters for more than fifty carbon-hydrogen distances to better
than 0.002 Å compared to the experimentally determined values from isolated CH stretching
frequencies which were compared to previously determined [34] values from earlier
microwave studies. Therefore, all of the carbon-hydrogen distances were taken from the
MP2/6-311+G(d,p) predicted values for cyclopropylisocyanate. A complete structure (Table
50) has been determined for each of the two conformers.
With the C H distances having uncertainties of 0.002 Å the predictions from the
MP2(full)/6-311+G(d,p) calculations of the heavy atoms were adjusted taking into account
how the corresponding parameters of similar molecules were adjusted to agree with the
experiment rotational constants.
For the four-membered cyclocarbon ring moiety of c-
C4H7NCO, the adjustments made for cyclobutanol [179] provided excellent predicted
distance since the 13C atoms as well as
18
O isotopomer were used to obtain the r0 parameter
for the alcohol. The adjustments from the MP2 predictions for the alcohol were then
transferred to the ring of the isocyanate molecule.
The distances for the N=C=O moiety
were predicted by a comparison of several N=C=O species and the C=O distance is very
similar since it behaves like a triple bond which is relatively constant irrespective of what
228
species on which it was attached. The N=C has been found to be significantly smaller than
the value predicted by the MP2 predicted value but significantly longer than the value from
the density function theory prediction by the B3LYP method with the basis set 6-311+G(d,p).
Therefore, the average of the two values is the adjusted distance for this parameters. The C αN distance and the CCN angle were obtained by the adjusting them until the rotational
constants from the adjusted parameters agreed with the experimentally determined values of
the B and C rotational constants. A change in the heavy atom distance by 0.003 Å changes
the value of the rotational constants so they no longer agree with the values of the
experimentally determined ones. Thus, it is believed that the uncertainties in these adjusted
r0 distances are at least as good as 0.005 Å and the angles within ± 0.5 º.
Discussion
In general the carbon-hydrogen vibrations are nearly the same for the two conformers
except for the C H stretch and in-plane bend. The C-H stretch for the Eqt-g form has the
lowest predicted (2923 cm-1) frequency and the C-H in-plane bend has a frequency of 1337
cm-1 which is lower by 34 cm-1 than the corresponding mode for the Eqt-t conformer.
Otherwise, the remaining carbon-hydrogen modes varied little between the corresponding
modes for the two conformers.
In making the vibrational assignment, the bands in the xenon solution made it
possible to identify many of the vibrations for the two conformers which were separated by
only a few wavenumbers.
The predicted frequencies from the MP2(full)/6-31G(d)
calculations with scaling factors of 0.88 for the CH stretches and 0.90 for the remaining
vibrations except for the heavy atom bends which were not scaled provided excellent
predicted values within 6.5 cm-1 (0.4% error) for the A' modes of the Eqt-t conformer. For
229
the A" fundamentals, the average difference was 7.46 cm-1 (0.6% error) and for the Eqt-g
conformer the difference was 8.41 cm-1 which is 0.9% error. These values clearly show that
these predicted values are very adequate for aiding the vibrational assignment and multiple
scaling factors are not needed for making the assignments. The predicted intensities of the
infrared bands as well as the Raman activities also helped in distinguishing the fundamentals
for the two conformers.
However, the band contours, which are frequently useful for
vibrational assignments of conformers, were usually of little value since many of the bands
were relatively broad and not distinguishable. The major contributing factor is the excited
vibrational states which are significantly populated as well as the large ring-NC angle which
becomes nearly linear. Nevertheless, it is believed that reasonable assignments were made
for both conformers.
For the Eqt-t form, there are three vibrations where there is extensive mixing with the
fundamental at 564 cm-1 which is assigned as the Cring-N stretch with only 12% of S19.
However, there are six fundamentals that have 10% or more contribution with the largest
amount of 26% for the NCO symmetric stretch. The total of the six contributions is 87 %
and excluding the largest amount the other five are 15%, 14%, 12%, 10% and 10% so the one
with 12% is the reasonable description used. The other two vibrations are the γ-CH2 twist
(28% S29) at 1167 cm-1 and the ring deformation at 1035 cm-1 with 24% S30. Both of these
vibrations have larger contributions from other modes where the twist has 42% of the β-CH2
wag (S28) and it is mixed with 31% S29. One could indicate both of these contributions as the
approximate descriptions but for simplicity we have one for each vibration with the listed
P.E.D.s given which clearly indicates the extensive mixing. Similar mixing is also shown for
the ring deformation where the mixing is with another ring deformation. Nevertheless, there
230
are only five vibrations out of the thirty-six that have contributions from four symmetry
coordinates which is a relatively small number for such a large molecule and only one plane
of symmetry. Finally, it should be noted that the Eqt-g conformer has similar mixing for the
corresponding vibrations.
The cyclobutylisocyanate molecule is a linear symmetric prolate rotor and the
predicted value of |μa| to be quite small it was not possible to obtain assignments for a few of
the transition which had a small contribution of the A rotational constant. Therefore, the
large uncertainty on the A rotational constant undoubtedly had a significant effect on not
being able to identify the desired transitions. However, it was still possible to obtain some
significant structural information from the values of the B and C rotational constant.
The structural parameters obtained (Table 50) for this molecule are believed to be as
good as can be determined without the data from several isotopic species. Since there are no
other rotational constants available, many of the parameters are adjusted based on similar
molecules. For the cyclocarbon ring parameters, there will be little change of the structural
parameters by substitution and, therefore, it is appropriate to use cyclobutanol [179], along
with fluoro-[180], chloro- [181], and bromocyclobutane[182] to assist in adjusting the
parameter. In comparison (Table 53) with other halocyclobutane and cyclobutanol, the C αCβ distance closes to the substitution is shorter than the Cβ-Cγ bond length by at least 0.01 Å
whereas the MP2(full)/6-311+G(d,p) calculations predict Cα-Cβ distance shorter by at least
0.005 Å. For the Eqt-t conformer, Cα-Cβ and Cβ-Cγ are adjusted accordingly to 1.548(5) and
1.557(5) Å, respectively. For the NCO parameters, the C=N and C=O parameters are
consistently predicted too long from MP2(full)/6-311+G(d,p) calculations by at least 0.008 Å
for C=N bond length and by ~0.02 for the C=O distances. By comparing the corresponding
231
parameters with ethylisocyanate [129] and cyclopropylisocyanate [183], the C=N and C=O
parameters are adjusted to be 1.208(5) and 1.161(5) Å, repectively. The two parameters that
are the most sensitive to adjustment and significantly affect the fit to the rotational constants
are the Cα-N distance and ring-NC angle which should not be too surprising since the
isoelectronic NCO group would cause some σ and π electron withdrawing and donation
between the ring.
The fit favors a structure with a larger CNC angle but based on
corresponding parameters of similar molecules the 135.2(5)º angle is a reasonable value. For
the Cα-N distance, it was adjusted to 1.433(5) Å, and any changes to this value by 0.003 Å is
sufficient to affect the fit.
The ab initio calculations predicted that there are four possible stable conformers
which consisted of Eqt-t, Eqt-g, Ax-t and Ax-g forms, whereas the cis orientation of both the
equatorial and axial forms are predicted to be transition states. The calculations predicted
the Eqt-t and Eqt-g forms to be the two most stable forms, with transition energies predicted
from MP2(full)/6-31G(d), 6-31+G(d), 6-311G(d,p) and 6-311+G(d,p) to be 143, 152, 143
and 167 cm-1, respectively. The Eqt-t and Eqt-g forms were most easily distinguished in the
infrared spectra of the xenon solution and solid. The presents of the Ax-t form was identified
only by five conformer bands at 1260, 1221, 943, 902, 552 and 402 cm-1 in the infrared
spectra of the amorphous solid, which upon the final annealing the bands disappeared. The
significant number of MP2 calculations gave an average energy difference of 137 ± 36 cm-1
(1.64 ± 0.43 kJ/mol) (Table 49) with the Eqt-t the more stable form which is in agreement
with the experimentally determined ΔH value of 131 ± 13 cm-1 (1.57 ± 0.16 kJ/mol).
However, it is worth noting that the prominent Eqt-t band at 450 cm-1 is gone upon annealing
and only Eqt-g bands are left in the crystalline solid. It should not be surprising that the
232
preferred conformation adopted in the crystalline solid is different than that observed in the
xenon solution since the conformation preferred in the crystalline solid depends upon the
specific interaction of the crystal lattice and the temperature of crystallization as well as the
ΔG° at the temperature for the two conformers. The packing factor in the crystalline solid is
further influenced by the volume difference between the two conformers [184, 185].
It should be noted that DFT calculations by the B3LYP method with higher basis sets
(6-311G(2d,2p), 6-311G(2df,2pd) and 6-311G(3df,3pd)) without diffuse functions predicted
the Eqt-g conformer as the more stable form by 15 cm-1. In addition, the calculations by the
B3LYP method (excluding those that predicted a more stable Eqt-g form) gave a much
smaller energy difference of 39 cm-1. Upon closer analysis of the DFT calculations, the Eqtg conformer converges into the Eqt-c form with higher basis sets which can be seen by the
τC4=N2 Cα-H3 angle predicted from MP2 and B3LYP at the 6-311+G(d,p) basis set to be
73.5° and 7.5° (Table 50), respectively. However, both the MP2 and B3LYP methods
predicted the axial form to be the less stable conformer by 569 cm-1 and 690 cm-1,
respectively. Although it was possible to determine the enthalpy difference between the
equatorial form and the axial form for the fluoro-, chloro- and bromocyclobutane when the
predicted energy different was about ~600 cm-1, it was very difficult to identify any Ax-t
conformer band in the infrared spectrum of the xenon solution. One explanation for why the
bands are not well separated is the expected low barrier to internal rotation of the
quasilinearity created by the NCO group.
For the cyclobutylisocyanate, the equatorial form which is more stable than the axial
form is consistent with the other substituted cyclobutanes [179-182]. To determine how the
NCO group behaves when attached to a ring, we compared cyclobutylisocyanate with
233
cyclopropylisocyanate [183] and cyclohexylisocyanate. The prefer position for the NCO
appears to be equatorial to the ring and trans relative to the alpha hydrogen. It should be
noted that the ring-N distance for the 3-membered ring NCO is about 0.02 Å shorter than the
corresponding distance for the 4- and 6-membered ring NCO. If the double bond on the N=C
acts like two separated bond eclipsing the Cα-Cβ and Cα-Cβ' bond, the shorter C-N distance
might explain why it is sterically favorable to have a structure where the NCO group is away
from the ring.
The isoelectronic characteristic of isocyanate does not determine the
conformational preference of these ring molecules and a trend that cannot be determined as it
for simpler systems such as the halides. Currently, we are studying the conformational
stability for the 5- and 6-membered ring isocycanate molecule to elucidate the factor or
factors that determines stabilities.
234
Table 44.
Rotational transitional frequencies (MHz) for equatorial-trans
cyclobutylisocyanate in the ground vibrational state.
Transition
(obs)
41,4
31,3
11876.62
0.24
40,4
30,3
11939.88
0.37
41,3
31,2
12005.09
0.19
51,5
41,4
14845.38
0.16
50,5
40,4
14923.63
0.31
51,4
41,3
15005.95
0.08
61,6
51,5
17813.80
-0.08
60,6
50,5
17906.60
0.17
71,7
61,6
20781.92
-0.41
70,7
60,6
20888.57
-0.12
71,6
61,6
21006.80
-0.40
235
Table 45.
Rotational (MHz), centrifugal distortion (kHz) and quadrupole coupling (MHz)
constants for the equatorial-trans and equatorial-gauche conformers of
cyclobutylisocyanate at 6-311+G(d,p) basis set.
Equatorial-Trans
MP2
B3LYP
K
6985
1509
1482
0.555
2.316
20.167
-0.0619
-48.953
7174
1464
1424
0.570
-1.790
33.483
-0.061
-17.707
χaa
χbb
χcc
-1.4085
2.1004
-0.8234
-1.2224
2.1387
-0.9163
A
B
C
J
JK
K
J
Equatorial-Gauche
MW
6891(302)
1508.682(26)
1476.551(23)
0.416(42)
-0.078(17)
236
MP2
B3LYP
9200
1382
1259
0.003
0.118
-0.094
0.001
0.066
10157
1289
1197
0.106
5.957
13.920
0.015
-12.549
2.7839
-1.2072
-1.5767
2.8968
-1.4970
-1.3998
Table 46. Calculated (MP2(full)/6-31G(d)) and observed vibrational frequencies (cm-1) of cyclobutylisocyanate for the equatorialtrans conformer.
237
A¢
A¢¢
A¢
A¢
A¢
A¢
A¢¢
A¢
A¢
A¢
A¢¢
A¢
A¢
A¢¢
A¢
A¢¢
A¢¢
A¢
A¢¢
A¢
n13 β-CH2 rock
1184 1124
9.9
2.2 1125
1122
1124
1122
A¢¢
A¢
n30 Ring deformation
n14 Ring breathing
1085 1029
1081 1025
3.2
20.6
0.2 ~1035
14.1 1032
1034
1029
1033
1026
¾
1029
Approx. description
γ-CH2 antisymmetric str.
β-CH2 antisymmetric str.
β-CH2 antisymmetric str.
CH str.
γ-CH2 symmetric str.
β-CH2 symmetric str.
β-CH2 symmetric str.
NCO antisymmetric str.
γ-CH2 scissors
β-CH2 scissors
β-CH2 scissors
NCO symmetric str.
CH in-plane bend
CH out-of-plane bend
β-CH2 wag
γ-CH2 wag
β-CH2 wag
β-CH2 twist
γ-CH2 twist
Table Continues
MP2
3211
3200
3196
3149
3138
3130
3128
2393
1574
1550
1538
1490
1435
1336
1312
1296
1284
1276
1228
MP2
a
scaled
3012
3002
2988
2954
2944
2936
2934
2270
1493
1471
1460
1414
1361
1268
1245
1229
1218
1211
1165
Infrared
Raman
IR Raman
b
c
int
act
gas
Xe solid liquid
36.9 53.6 3000 2993 2992 ~2988
19.5 66.7 2993
¾ 2992 ~2985
8.5 77.2 2986 2982 2968 2981
16.3 69.3 2965 2962 2968 2963
20.6 103.2 2950 2955 2953 2955
2.1 154.7
¾
¾ 2944 2940
27.0
0.6
¾ 2918 2920
¾
868.8
1.3 ~2272 ~2270 2272 2264
3.5
5.3 1488 1486 1491
¾
5.2 21.7
¾
¾ 1477
¾
1.9
5.3 ~1445 1464 1465
¾
1441
1447
21.7 35.4 ~1445 1443
37.9 21.1 ~1360 1361 1361 1361
0.2
9.9
¾ 1268 1269 ~1262
5.0
0.5 1244 1246 1246 ~1248
0.9
0.3
¾ 1230 1232 1233
0.0
2.0
¾
¾ 1215 1217
0.4 10.2
¾ 1214 1213
0.2
8.7 1167 1168 1173 1173
Vib.
No
n1
n23
n2
n3
n4
n5
n24
n6
n7
n8
n25
n9
n10
n26
n11
n27
n28
n12
n29
d
Contour
A C
16 84
PED
52S1, 46S3
97S23
47S2, 46S1
100
91S3
41
95S4
95
91S5
91
97S24
98S6
85
57S7, 39S8
9
60S8, 40S7
32
99S25
62S9, 26S19
93
61S10, 19S20
96
52S26, 26S27
78S11, 10S15
35
46S27, 27S26, 21S30
53S28, 31S29
80S12, 10S17
16
28S29, 42S28, 11S30
46S13, 14S17, 12S21,
83
10S14
24S30, 30S32, 31S31
67S14, 13S9
95
0
59
5
9
15
91
68
7
4
65
84
17
5
Table 46 Continues
A¢¢
n31 β-CH2 twist
981
931
3.7
0.7
942
937
939
939
A¢
n15 Ring deformation
968
918
13.9
5.8
922
920
915
920
A¢¢
n32 Ring deformation
963
914
0.5
13.4
906
907
906
911
A¢
n16 Ring deformation
940
891
21.3
4.7
896
892
890
889
A¢¢
n33 β-CH2 rock
825
782
0.6
1.1
¾
785
782
785
A¢
n17 γ-CH2 rock
774
735
4.1
4.4
739
738
740
742
A¢
A¢¢
A¢
A¢
n18
n34
n19
n20
651
570
576
470
643
570
550
450
28.8
19.2
3.1
5.9
1.5 643
0.3 ~584
2.1 ~564
1.2 448
650
576
¾
447
~640
575
558
450*
¾
¾
561
441
Contour
d
PED
A C
31S31, 24S30, 17S26,
17S29
48S15, 15S19, 13S11,
100
0
10S9
61S32, 20S30
41S16, 17S17, 14S13,
99
1
10S19
78S33, 13S29
40S17, 22S15, 13S12,
98
2
10S19
73S18
9 91
98S34
12S19, 46S16, 18S13
83 17
41S20, 14S19, 13S18
33 67
A¢¢
n35
370
351
3.7
0.6
360
¾
371
366
86S35, 11S30
A¢
A¢
n21
n22
222
93
213
92
5.0
2.1
0.4
3.0
¾
¾
¾
¾
¾
¾
¾
¾
68S21, 17S22
74S22, 15S20
A¢¢
n36
39
39
0.3
2.7
¾
¾
¾
¾
100S36
Vib.
No
238
a
Approx. description
NCO in-plane bend
NCO out-of-plane bend
Cα-N str.
Ring-NC in-plane bend
Ring-NC out-of-plane
bend
Ring Puckering
Cα–N=C in-plane bend
Cα–N=C out-of-plane
bend
Infrared
MP2 IR Raman
a
b
c
MP2 scaled int
act
gas
Xe solid
Force constant scaling factors: 0.88 for CH stretches; 0.9 for all other modes.
Infrared intensities in km/mol.
c
Raman activities in Å4/u.
d
PEDs from MP2 scaled calculation: values less than 10% are omitted.
* This band disappears upon annealing.
b
Raman
liquid
64
98
36
2
Table 47.
Calculated (MP2(full)/6-31G(d) and observed vibrational frequencies (cm-1) of equatorial-gauche cyclobutylisocyanate.
239
Vib.
Approx. description
No
n1 γ-CH2 antisymmetric str.
n2 β-CH2 antisymmetric str.
n3 β-CH2 antisymmetric str.
n4 γ-CH2 symmetric str.
n5 β-CH2 symmetric str.
n6 β-CH2 symmetric str.
n7 CH str.
n8 NCO antisymmetric str.
n9 β-CH2 scissors
n10 γ-CH2 scissors
n11 β-CH2 scissors
n12 NCO symmetric str.
n13 CH in-plane bend
n14 CH out-of-plane bend
n15 β-CH2 wag
n16 γ-CH2 wag
n17 β-CH2 wag
n18 β-CH2 twist
n19 γ-CH2 twist
n20 β-CH2 rock
n21 Ring breathing
3211
3201
3195
3138
3136
3131
3116
2397
1577
1551
1539
1486
1409
1333
1316
1293
1286
1277
1231
1184
1101
n22 Ring deformation
1087 1031
Table Continues
MP2
MP2
a
scaled
3013
3003
2997
2944
2942
2937
2923
2274
1496
1472
1460
1410
1337
1265
1248
1227
1220
1212
1168
1123
1044
IR Raman
Infrared
Raman
b
c
int
act
gas
Xe solid liquid
34.5 54.7 3000 2993 2992
16.6 65.2
2993 2992
8.4 72.1
2993 2992
19.4 158.9 2940 2947 2943 2944
14.6 75.2 2940 2947 2943 2940
25.7 27.1
2943
14.3 81.6
2918 2917
949.4
2.1 2272 2270 2272 2264
7.1
5.3 1488 1485 1491
4.2 22.2 1470
1478
2.2
6.7
1460 1465
64.1 49.6 ~1445 1437 1436
62.3
8.9 1344 1345 1347 1348
0.2
7.2
1264 1265 ~1262
2.3
1.1 1246
1242 1248
0.9
0.9
1237 1232
0.2
3.3
1221
0.3 11.2
1217 1218 1217
0.5
9.5
0.9
2.4
1115
22.2
4.6 1043 1039 1044 1032
5.1
3.6 ~1035 ~1034
1035
Contour
d
PED
A B C
48S1, 49S3
0 2 98
91S2
16 81
3
47S3, 41S1
71 22
7
84S4, 11S5
97 0
3
64S5, 24S6, 11S4
14 67 19
75S6, 24S5
11 86
3
97S7
9 1 90
98S8
97 3
0
58S9, 36S10
90 1
9
62S10, 37S10
0 1 99
99S11
11 89
0
56S12, 28S31
94 3
3
57S13, 17S32, 12S12 100 0
0
44S14, 32S16
27 72
1
71S15, 13S24
52 27 21
41S16, 32S14, 22S23 11 89
0
52S17, 30S19
79 13
8
74S18, 11S28
95 0
5
29S19, 40S17, 10S22 48 52
0
45S20, 14S28, 12S34 25 17 58
45S21, 10S12, 10S25 97 0
3
18S22, 23S23,
0 96
4
21S25, 18S21
Table 47 Continues
Vib.
No
Approx. description
MP2
MP2
IR Raman
a
b
c
scaled int
act
gas
Infrared
Xe solid
Raman
liquid
240
n23 β-CH2 twist
983
933
5.3
3.8
~942
937
945
939
n24 Ring deformation
n25 Ring deformation
973
964
923
915
8.0
1.1
8.2
11.9
927
906
920
907
920
910
920
911
n26 Ring deformation
946
897
8.7
3.5
~896
892
888
n27
n28
n29
n30
823
776
644
574
780
737
636
568
0.8
6.8
20.9
19.1
0.9
3.7
0.7
0.4
780
777
746
638
585
n31 Cα-N str.
561
542
1.5
3.5
560
556
n32 Ring-NC in-plane bend
432
410
6.2
1.6
406
414
371
353
15.5
2.4
211
104
200
104
4.1
2.9
25
25
0.1
β-CH2 rock
γ-CH2 rock
NCO in-plane bend
NCO out-of-plane bend
Ring-NC out-of-plane
n33 bend
n34 Ring Puckering
n35 Cα–N=C in-plane bend
C –N=C out-of-plane
n36 α
bend
a
Contour
A B C
66 20
14
47 29
4 95
24
1
86
6
8
58
2
37
6
42
19
44
39
0
79
19
55
4 94
2
73 10
17
71S33, 10S22
81 19
0
0.8
2.5
66S34, 18S32
78S35
75
98
2
0
23
2
3.3
100S36
18 26
56
630
415
Force constant scaling factors: 0.88 for CH stretches; 0.9 for all other modes.
Infrared intensities in km/mol.
c
Raman activities in Å4/u.
d
Values less than 10% are omitted.
b
d
PED
24S23, 14S24,
13S14, 12S19, 12S22
29S24, 12S31, 11S23
54S25, 27S22
39S26, 18S28,
15S20, 11S24
78S27, 13S19
41S28, 20S24, 14S18
76S29, 11S26
92S30
12S31, 39S26,
16S20, 12S29
40S32, 13S31,
12S34, 10S13
375
415
Table 48.
241
Vib.
No
A¢ n1
n2
n3
n4
n5
n6
n7
n8
n9
n10
n11
n12
n13
n14
n15
n16
n17
n18
n19
n20
n21
n22
Calculated (MP2(full)/6-31G(d) and observed vibrational frequencies (cm-1) of axial-trans cyclobutylisocyanate.
Approx. description
MP2
β-CH2 antisymmetric str.
γ-CH2 antisymmetric str.
CH str.
γ-CH2 symmetric str.
β-CH2 symmetric str.
NCO antisymmetric str.
γ-CH2 scissors
β-CH2 scissors
NCO symmetric str.
CH in-plane-bend
β-CH2 wag
β-CH2 twist
β-CH2 rock
Ring breathing
Ring deformation
Ring deformation
Cα-N str.
γ-CH2 rock
NCO in-plane bend
Ring-N in-plane bend
Ring Puckering
Cα–N=C in-plane bend
3209
3192
3168
3139
3131
2391
1572
1548
1483
1417
1326
1265
1158
1066
960
913
809
707
619
420
208
93
Table Continues
MP2
IR Raman dp
a
b
c
scaled int
act
ratio
3010 37.6 51.8 0.67
2995
5.4 72.8 0.66
2972 30.3 128.2 0.16
2945 18.9 93.2 0.10
2937
6.1 148.9 0.21
2268 771.6
0.9 0.00
1491
2.8
6.4 0.63
1468
2.4 15.9 0.70
1407 13.0 26.2 0.21
1345 23.9 11.0 0.44
1258 10.2
1.1 0.37
1200
2.7 15.6 0.71
1098 25.8
7.1 0.23
1011
4.8 16.3 0.20
910
0.3
1.8 0.32
866
4.0
3.4 0.26
768 24.9
8.7 0.12
674 12.9
0.7 0.19
614 12.7
0.5 0.73
400
5.0
1.1 0.31
199
3.8
1.1 0.44
92
1.8
2.2 0.64
IR
solid
1260
402
PED
d
50S1, 50S2
47S2, 49S1
96S3
96S4
93S5
98S6
67S7, 31S8
66S8, 32S7
61S9, 25S17, 10S10
61S10, 17S20
83S11, 13S15
66S12, 15S18
41S13, 13S18, 11S10, 10S21
83S14
68S15, 15S11
49S16, 24S18, 14S13
48S17, 18S9
32S18, 25S16, 11S19, 10S13
71S19
39S20, 11S16, 10S13
67S21, 19S22
71S22, 16S20
Contour
A B
99 1
1 99
16 84
5 95
68 32
79 21
0 100
44 56
99 1
92 8
3 97
94 6
78 22
78 22
84 16
0 100
91 9
43 57
4 96
64 36
58 42
93 7
Table 48 Continues
242
Vib.
No
A¢¢ n23
n24
n25
n26
n27
n28
n29
n30
n31
n32
n33
n34
n35
n36
a
Approx. description
β-CH2 antisymmetric stretch
β-CH2 symmetric stretch
β-CH2 scissors
γ-CH2 wag
CH out-of-plane bend
β-CH2 wag
γ-CH2 twist
β-CH2 twist
Ring deformation
Ring deformation
β-CH2 rock
NCO out-of-plane bend
Ring-NC out-of-plane bend
Cα-N=C out-of-plane bend
MP2
3200
3129
1531
1327
1295
1288
1206
1107
982
955
787
565
403
47
MP2
scaleda
3002
2935
1452
1259
1229
1222
1143
1050
932
906
748
565
381
47
IR Raman dp
intb
actc
ratio
8.3 60.1 0.75
29.8
4.8 0.75
4.7
8.2 0.75
0.7
1.0 0.75
0.0
6.3 0.75
0.0
5.9 0.75
2.1
5.2 0.75
1.7
2.9 0.75
0.1 13.5 0.75
4.7
0.5 0.75
0.1
0.3 0.75
17.1
0.3 0.75
3.3
0.4 0.75
0.6
2.6 0.75
Force constant scaling factors: 0.88 for CH stretches; 0.9 for all other modes.
Infrared intensities in km/mol.
c
Raman activities in Å4/u.
d
Values less than 10% are omitted.
b
IR
solid
1221
943
902
552
PEDd
98S23
98S24
100S25
50S26, 25S27, 11S32
40S27, 25S26, 16S28, 11S29
48S28, 32S29
23S29, 22S28, 16S31, 15S27
58S30, 13S33, 11S29
72S31
54S32, 12S30, 11S33
61S33, 21S32, 12S29
98S34
90S35
100S36
Contour
A B
Table 49.
Calculated electronic energiesa (hartree) and energy differencesb (cm-1) for the
equatorial-trans, equatorial-cis, axial-trans and axial-cis conformers of
cyclobutylisocyanate.
Equatorial
Axial
Method / Basis set
basis
functions
Trans
Gauche
Cis
MP2(full)/6-31G(d)
119
0.711590
123
MP2(full)/6-31+G(d)
147
0.729366
MP2(full)/6-311G(d,p)
168
MP2(full)/6-311+G(d,p)
Trans
Gauche
Cisc
150
583
619
640
83
105
625
614
626
0.991272
88
116
568
589
603
196
1.001398
133
185
654
680
709
MP2(full)/6-311G(2d,2p)
224
1.075177
108
111
511
617
617
MP2(full)/6-311+G(2d,2p)
252
1.084078
135
141
579
683
683
MP2(full)/6-311G(2df,2pd)
308
1.194852
136
136
522
670
670
MP2(full)/6-311+G(2df,2pd)
336
1.202511
179
167
587
770
738
MP2(full)/6-311G(3df,3pd)
364
1.223016
148
116
529
656
627
MP2(full)/aug-cc-PVTZ
483
1.215660
¾
134
559
¾
660
MP2(full)/cc-PVQZ
595
1.367471
¾
173
580
¾
745
B3LYP/6-31G(d)
119
1.694688
70
68
708
693
695
B3LYP/6-31+G(d)
147
1.706899
35
12
708
627
625
B3LYP/6-311G(d,p)
168
1.783952
2
-2
679
589
575
B3LYP/6-311+G(d,p)
196
1.789530
37
35
709
638
626
B3LYP/6-311G(2d,2p)
224
1.793679
-24
-26
660
565
560
B3LYP/6-311+G(2d,2p)
252
1.799342
30
27
692
637
627
B3LYP/6-311G(2df,2pd)
308
1.804687
-3
-7
666
599
595
B3LYP/6-311+G(2df,2pd)
336
1.809882
42
39
699
656
650
B3LYP/6-311G(3df,3pd)
364
1.808828
-18
-22
683
591
585
B3LYP/6-311+G(3df,3pd)
392
1.813473
57
55
692
661
662
a
c
Energy of Eqt-t conformer is given as –(E+323) hartree.
b
All energy differences are relative to the energy of the Eqt-t conformer.
c
Transition state.
243
Table 50.
Structural parameters (Å and degree), rotational constants (MHz) and dipole moments (debye) for the
equatorial-trans, equatorial-gauche and axial-trans conformers of cyclobutylisocyanate from MP2(full) and
B3LYP/6-311+G(d,p).
Parameter
244
rC=O
rN=C
rCα-N
rCα-Cβ,β"
rCγ-Cβ,β"
rCα-H
rC β,β"-H9,11
rC β,β"-H10,12
rCγ-H14
rCγ-H13
ÐNCO
ÐCαNC
ÐC β,β"CαN
ÐCβCαCβ"
ÐCγC β,β"Cα
ÐCβCγCβ"
Table Continues
Int.
Coord
R1
R2
R3
δ1, δ2
δ3, δ4
r
r1,r3
r2,r4
r5
r6
π
θ
η2, η3
Δ3
Δ1 , Δ2
Δ4
Eqt-Trans
MP2 B3LYP
1.180 1.175
1.216 1.199
1.438 1.438
1.543 1.552
1.549 1.554
1.093 1.091
1.094 1.093
1.092 1.091
1.091 1.090
1.093 1.092
172.7 173.9
133.9 140.7
119.8 119.7
88.3
88.8
86.9
87.7
87.9
88.7
Eqt-Gauche
MP2
B3LYP
1.180
1.174
1.216
1.200
1.441
1.443
1.536/1.545 1.549/1.551
1.551/1.547 1.554/1.554
1.096
1.094
1.093/1.094 1.092/1.092
1.092/1.092 1.091/1.091
1.091
1.090
1.093
1.092
172.8
174.0
134.3
139.9
118.2/118.6 117.6/117.7
88.3
88.6
87.1/87.0
88.1/88.1
87.7
88.3
Axial-Trans
MP2
B3LYP
1.181
1.175
1.215
1.199
1.444
1.442
1.550
1.560
1.549
1.554
1.090
1.089
1.092
1.091
1.093
1.092
1.093
1.092
1.091
1.091
172.7
173.9
135.1
141.6
112.8
114.7
87.8
88.6
88.4
89.5
87.9
89.1
Adj. r0
Eqt-T
1.161(5)
1.208(5)
1.433(5)
1.548(5)
1.557(5)
1.093(2)
1.094(2)
1.092(2)
1.091(2)
1.093(2)
172.0(5)
135.2(5)
119.8(5)
88.9(5)
87.2(5)
88.3(5)
Est. r0
Eqt-G
1.161(5)
1.208(5)
1.436(5)
1.541/1.550(2)
1.559/1.555(2)
1.096(2)
1.093/1.094(2)
1.092/1.092(2)
1.091(2)
1.093(2)
172.1(5)
135.6(5)
118.0/118.3(5)
88.9(5)
87.4/87.3(5)
88.1(5)
Table 50 Continues
Parameter
245
Ð HCαC β,β"
Ð HCαN
Ð H9,11C β,β"Cα
Ð H9,11C β,β"Cγ
Ð H10,12C β,β"Cα
Ð H10,12C β,β"Cγ
Ð H9,11Cβ,β'H10,12
Ð H14CγC β,β"
Ð H13CγC β,β"
Ð H14CγH13
Ð C4=N2-Cα-H3
Ð CγCβCβ'Cα
A(MHz)
B(MHz)
C(MHz)
|ma|
|mb|
|mc|
|mt|
Int.
Coord
φ1, σ1
η1
α1, α3
ε1, ε3
α2, α4
ε2, ε4
λ1, λ2
φ2, σ2
φ3, σ3
λ3
t
t1
Eqt-Trans
MP2 B3LYP
110.0 110.2
107.7 107.2
109.8 110.8
110.5 111.4
118.2 117.4
119.5 118.8
110.0 109.2
117.8 117.2
111.2 112.0
109.4 108.7
180.0 180.0
146.4 151.7
6985
7174
1509
1464
1482
1424
3.432 3.352
0.0
0.0
0.066 0.218
3.433 3.359
Eqt-Gauche
MP2
B3LYP
110.5/110.0 110.9/110.8
109.7
109.8
109.5/109.5 110.2/110.2
110.6/110.8 111.5/111.5
118.3/118.0 117.4/117.4
119.4/119.5 118.8/118.8
110.1
109.3
117.8/117.9 117.2/117.2
111.6/111.0 112.2/112.1
109.4/110.0 108.7/109.3
73.5
7.5
146.5
152.1
9200
10157
1382
1289
1259
1197
3.279
3.114
0.417
0.045
0.690
0.732
3.377
3.199
Axial-Trans
MP2
B3LYP
117.5
115.7
107.5
107.0
116.8
115.9
119.0
117.8
110.2
111.1
111.2
112.7
109.7
108.8
111.8
112.6
117.6
116.7
108.8
108.3
0.0
0.0
-150.7 -160.4
5365
5684
1888
1685
1784
1641
3.474
3.359
0.227
0.419
0.00
0.000
3.481
3.385
Adj. r0
Est. r0
Eqt-T
Eqt-G
110.2(5) 110.7/110.2(5)
107.7(5)
109.7(5)
109.8(5) 109.5/109.5(5)
108.3(5) 108.4/108.6(5)
118.2(5) 118.3/118.0(5)
121.3(5) 121.2/121.3(5)
110.0(5)
110.1(5)
118.4(5) 118.4/118.5(5)
110.1(5) 110.5/109.9(5)
109.4(5) 109.4/110.0(5)
7097
1508
1476
9568
1358
1244
Table 51.
Symmetry Coordinates for Cyclobutylisocyanate.
n1 β-CH2 antisymmetric str.
n2 γ-CH2 antisymmetric str.
n3 CH symmetric str.
n4 γ-CH2 symmetric str.
n5 β-CH2 symmetric str.
n6 NCO antisymmetric str.
n7 γ-CH2 scissors
n8 β-CH2 scissors
n9 NCO symmetric str.
n10 β-CH2 wag
n11 CH in-plane-bend
n12 β-CH2 twist
n13 β-CH2 rock
n14 Ring breathing
n15 γ-CH2 rock
n16 Ring deformation
n17 C-NCO stretch
n18 Ring deformation
n19 NCO in-plane-bend
n20 Ring-NCO in-plane-bend
n21 C-NCO in-plane-bend
n22 Ring puckering
A¢¢ n23 β-CH2 antisymmetric str.
n24 β-CH2 symmetric str.
n25 β-CH2 scissors
n26 γ-CH2 wag
n27 CH out-of-plane bend
n28 γ-CH2 twist
n29 β-CH2 wag
n30 Ring deformation
n31 Ring deformation
n32 β-CH2 twist
n33 β-CH2 rock
n34 NCO out-of-plane-bend
n35 Ring-NC out-of-plane bend
n36 CαN out-of-plane bend
A¢
Symmetry Coordinate
r1 – r2 + r3 – r4
r5 – r6
r
r5 + r6
r1 + r2 + r3 + r4
R1 – R2
4λ3 – φ2 – σ2 – φ3 – σ3
4λ1 – α1 – α2 – ε1 –ε2 + 4λ2 – α3 – α4– ε3 – ε4
R1 + R2
α1 + α2 – ε1 – ε2 + α 3 + α4– ε3 – ε4
4η1 – η2 – η3 – φ1 – σ1
α1 – α2 – ε1 – ε2 + α 3 – α4 – ε3 + ε4
α1 – α2 + ε1 – ε2 + α 3 – α4 + ε3 – ε4
δ1 + δ2 + δ3 + δ4
φ2 – φ3 + σ2 – σ3
δ1 + δ2 – δ3 – δ4
R3
Δ1 + Δ2 – Δ3 – Δ4
π
η2 + η3 – φ1 – σ1
θ
Δ1 + Δ2 + Δ3 + Δ4
r1 – r2 – r3 + r4
r1 + r2 – r3 – r4
4λ1 – α1 – α2 – ε1 –ε2 – 4λ2 + α3 + α4+ ε3 + ε4
φ2 + φ3 – σ2 – σ3
φ1 – σ1
φ2 – φ3 – σ2 + σ3
α1 + α2 – ε1 – ε2 – α 3 – α4 + ε3 + ε4
δ1 – δ2 – δ3 + δ4
δ1 – δ2 + δ3 – δ4
α1 – α2 – ε1 + ε2 – α 3 + α4 + ε3 – ε4
α1 – α2 + ε1 – ε2 – α 3 + α4 – ε3 + ε4
Φ
η2 – η3
Σ
246
Table 52.
Temperature and intensity ratios of the conformer bands of cyclobutylisocyanate from the infrared spectra of the liquid
xenon solution.
247
T( C)
Liquid xenon
55.0
60.0
65.0
70.0
75.0
80.0
85.0
90.0
95.0
100.0
Ha
a
trans conformer more stable.
1/T ( 10-3 K-1)
4.5840
4.6915
4.8042
4.9225
5.0467
5.1773
5.3149
5.4600
5.6132
5.7753
I576 / I560
1.804
1.850
1.868
1.907
1.935
1.963
2.080
2.143
2.211
2.271
137
7
I650/ I630
I738 / I780
I738 / I1264
1.930
1.944
1.993
2.048
2.097
2.137
2.177
2.250
2.282
2.333
9.364
8.917
9.826
9.667
10.348
10.040
11.042
10.769
11.480
11.615
3.552
3.627
3.767
3.867
3.839
3.922
4.077
4.242
4.284
4.441
116
4
145
17
5 cm-1 (1.57
126
7
0.06 kJ/mol) with the equatorial-
Table 53.
Comparison of MP2/6-311+G(d,p) calculated structural parameters (Å and degree) with adjust r 0 values for
c-C4H7-X (X = NCO, OH, F, Cl, Br).
X=
Parametera
rN=C
rC=O
rCα- X
rCα-Cβ,β"
rCγ-Cβ,β"
248
ÐNCO
ÐCαNC
ÐC β,β"CαX
ÐCβCαCβ"
ÐCγC β,β"Cα
ÐCβCγCβ"
Ð HCαC β,β"
Ð HCαX
A(MHz)
B(MHz)
C(MHz)
NCO
Eqt-t
MP2
Adj. r0
1.216 1.208(5)
1.180 1.161(5)
1.438 1.433(5)
1.543 1.548(5)
1.549 1.557(5)
172.7 172.0(5)
133.9 135.2(5)
119.8 119.8(5)
88.3
88.9(5)
86.9
87.2(5)
87.9
88.3(5)
110.0 110.2(5)
107.7 107.7(5)
6985
7128
1509
1503
1482
1470
Eqt-g
OH[28]
Eqt-trans
MP2 Adj. r0
F[29]
Equatorial
MP2 Adj. r0
MP2
Est. r0
1.216
1.208(5)
1.180
1.161(5)
1.441
1.438(5)
1.408 1.412(3) 1.386
1.536/1.545 1.541/1.550(5) 1.542 1.547(3) 1.530
1.551/1.547 1.559/1.555(5) 1.548 1.556(3) 1.551
172.8
172.1(5)
134.3
135.6(5)
118.2/118.6 118.0/118.3(5) 120.6 120.2(5) 117.3
88.3
88.9(5)
88.3 88.9(5)
89.4
87.1/87.0
87.4/87.3(5)
86.8 87.1(5)
86.4
87.7
88.1(5)
87.9 88.3(5)
87.8
110.5/110.0 110.7/110.2(5) 110.2 110.4(5) 112.1
109.7
109.7
105.9 105.9(5) 107.7
9200
9157
10247
10120 10392
1382
1397
4290
4283 4271
1259
1271
3439
3421 3411
Cl[30]
Equatorial
MP2 Adj. r0
1.383(3) 1.777 1.783(5)
1.543(3) 1.534 1.539(3)
1.554(3) 1.551 1.558(3)
117.4(5)
89.3(5)
85.0(5)
88.6(5)
112.1(5)
107.7(5)
10250
4275
3403
Br[31]
Equatorial
MP2 Adj. r0
1.938 1.942(3)
1.535 1.541(3)
1.551 1.552(3)
118.7 118.1(5) 118.9 118.4(5)
89.0 89.7(5)
89.1 89.7(5)
86.6 86.9(5)
86.4 86.8(5)
87.8 88.3(5)
87.9 88.9(5)
111.3 111.6(5) 111.4 111.8(5)
107.1 107.1(5) 106.4 106.4(5)
10274
10095 10208
10003
2521
2522 1624
1630
2200
2196 1485
1488
Figure 37.
Atomic numbering of cyclobutylisocyanate with the equatorial form shown.
The relative orientation of the –NCO moiety is indicated by the τC4=N2 Cα−H3
angle: Trans (τ=180°); Cis (τ=0°); Gauche (τ=73.5°).
249
Figure 38. Infrared spectra of cyclobutylisocyanate: (A) Gas; (B) Xenon solution at -100 C.
250
Figure 39.
Observed and predicted (MP2(ful1)/6-311+G(d,p)) infrared spectra of
cyclobutylisocyanate: (A) Gas; (B) Predicted spectrum of the mixture of Eqt-t
and Eqt-g conformers with H = 131 cm-1; (C) Predicted spectrum of the pure
Eqt-g conformer; (D) Predicted spectrum of the pure Eqt-t conformer.
251
Figure 40.
Infrared spectra (500-300 cm-1) of cyclobutylisocyanate: (A) Gas; (B)
Amorphous solid; (C) First annealed solid; (D) Second annealed solid.
252
Figure 41.
Infrared spectra (800-500 cm-1) of cyclobutylisocyanate: (A) Gas; (B)
Amorphous solid; (C) First annealed solid; (D) Second annealed solid.
253
Figure 42.
Observed and predicted (MP2(ful1)/6-311+G(d,p)) Raman spectra of
cyclobutylisocyanate: (A) Liquid; (B) Predicted spectrum of the mixture of
Eqt-t and Eqt-g conformers with H = 131 cm-1; (C) Predicted spectrum of the
pure Eqt-g conformer; (D) Predicted spectrum of the pure Eqt-t conformer.
254
CHAPTER 11
MICROWAVE, INFRARED, AND RAMAN SPECTRA, r0 STRUCTURAL
PARAMETERS, CONFORMATIONAL STABILITY AND AB INITIO CALCULATIONS
OF CYCLOHEXYLISOCYANATE
Introduction
The conformational analysis of cyclohexane and its derivatives have been of constant
interest to molecular spectroscopists since the work of Hassel [186, 187]. We began our
investigation of 6-membered rings with non-substituted cyclohexane by studying the relative
intensities of the Raman active fundamentals and the potential function governing the
conformational interchange between the chair and twisted boat form [188]. Since the boat
form is so much higher in energy compared to the chair form, the amount of it present at
ambient temperature is insignificant. To further understand the conformational stability of
the chair form, studies were initiated to spectroscopically investigate monosubstituted
chloro-[189], bromo-[190] and fluorocyclohexane [191] since there are only two conformers,
equatorial and axial forms, which exist at ambient temperature. Due to the repulsive nonbonded nature of the interaction between the axial and the equatorial substituent, the
equatorial conformation is generally favored [186, 192] and these monosubstituted
cyclohexanes exist predominantly in the equatorial form. A significant number of studies
have been carried out on the enthalpy differences for these halocyclohexanes by using
vibrational spectroscopy and theoretical calculations such as infrared, Raman, nuclear
magnetic resonance, and microwave spectroscopy. However, the enthalpy values varied over
a wide range and for several cases the uncertainties were large i.e. 140
105 cm-1 for
fluorocyclohexane [191]. Additionally, several pseudohalogen substituted cyclohexanes i.e.
255
cyanocyclohexane [193] and ethynylcyclohexane [194] have been studied and the lowest
energy forms have been reported. However, the most stable conformation for
cyanocyclohexane is an axial form [184, 193] compared to the lowest energy equatorial form
determined for all other substituted cyclohexanes in our investigations. Therefore, a study on
another pseudohalogen, cyclohexylisocyanate, has been initiated to obtain additional
information on the effect of the substituent on the ring. This molecule is expected to provide
an additional level of complexity to the conformational landscape of the conformation of sixmembered ring molecules.
Unlike the mono-halogenated and the linear polyatomic
substitution, the isocyanate group has a higher degree of conformational flexibility due to the
internal rotation angle of the isocyanate group. This angle makes it possible for more than
one conformational minima to exist.
There have been three studies [195-197] carried out on cyclohexylisocyanate that
appeared to be inconclusive regarding its conformation, where especially lacking is any
reliable data regarding the relative position of the NCO moiety. The earliest one [195] was a
nuclear magnetic resonance study where the investigators reported that the preferential
conformation is the equatorial conformer with a predicted equatorial:axial ratio of ~4:1. A
later vibrational study [196] was aimed to determine the conformational equilibria in the
liquid state and in the crystal. The result of the study indicated that both the axial and
equatorial form existed in the liquid phase as well as in solution.
However, at low
temperature cyclohexylisocyanate crystallizes in the equatorial form but when pressurized to
20 kbar at ambient temperature, the axial form is the preferred conformer. The third study
[197] was carried out by utilizing low resolution microwave spectroscopy and seven
transitions were assigned in the frequency range of 39 to 28 GHz. The author attempted to
256
use the experimental B+C value of 1702.9 ± 0.2 MHz to determine the most stable conformer
by comparing it to the calculated B+C values of the four possible conformers: equatorialtrans (Eqt-t), equatorial-cis (Eqt-c), axial-trans (Axl-t) and axial-cis (Axl-c). The B+C
values were calculated by using structural parameters from cyclohexyl fluoride [198, 199]
with the exception of the
C -N=C, which is adopted from CH3NCO [93]. The calculated
B+C value for Eqt-c form (1582 MHz) and Eqt-t form (1820 MHz) was the closest to the
experimental value of 1703 MHz. The second part of the name refers to the orientation of the
NCO moiety relative to the alpha H (Fig. 43). The structural adjustments needed to bring the
experimental and calculated B+C values into agreement would severely skew the
conformation. The author [197] concluded that due to the low barrier to internal rotation, the
molecule is freely rotating around the C -N bond and, therefore, the effective B+C value
does not correspond to any specific orientation of the NCO group.
Although NCO
substituted molecules tend to have a smaller barrier to internal rotation, it has been shown
[111] that it is possible to determine decent rotational constants for molecules with similar
low barrier.
As a continuation of our conformational analysis of cyclohexanes, extensive
rotational and vibrational studies have been carried out on cyclohexylisocyanate. Variable
temperature study of the infrared spectra has been recorded in liquid xenon and the relative
stability of the Eqt-t and Axl-t conformer has been determined. In addition, r0 structural
parameters have been determined for the two most stable conformers by systematically
adjusting the ab initio MP2(full)/6-311+G(d,p) optimized structure to fit the determined
rotational constants obtained in this study. The results of these spectroscopic, structural and
theoretical studies are reported herein.
257
Experiment and Theoretical Calculations
The cyclohexylisocyanate sample was purchased from Sigma Aldrich with a stated
purity of 98%. The sample was further purified by using a low-temperature, low-pressure
sublimation column and the purity of the sample was verified by infrared spectroscopic data.
The sample was kept in the dark at low temperature until it was used.
The microwave spectra of cyclohexylisocyanate were recorded with a “mini-cavity”
Fourier-transform microwave spectrometer [53, 54] at Kent State University. The FabryPerot resonant cavity was established by two 7.5-inch diameter diamond-tip finished
aluminum mirrors with a 30.5-cm spherical radius. The cyclohexylisocyanate sample was
entrained in 70:30 Ne-He carrier gas mixture at 2 atm and expanded into the cavity with a
reservoir nozzle [54] made from a modified Series-9 General Valve.
The sample was
irradiated by microwave radiation generated by an Agilent Technologies E8247C PSG CW
synthesizer; details of the irradiation and its heterodyne detection circuitry can be found in
Ref. [53]. Labview software controls the timing of the gas and irradiation pulses, as well as
the detection of any free induction decay signal. The software performs signal averaging and
can scan the spectrometer by stepping both the frequency source and the cavity. Microwave
circuit elements allow for a spectral range from 10.5 to 26 GHz. The digital frequency
resolution, governed by the sampling rate and the length of the free induction decay record, is
2.5 kHz. Rotational transitions are split into Doppler doublets centered at the transition
frequency due to the coaxial orientation of the gas expansion to the cavity axis and the
FWHM of each Doppler component is typically 13 kHz. The vacuum system can
accommodate pulse repetition rates of up to 15 s-1 while maintaining a pressure below 10-4
torr, and the instrument can scan 450 MHz in 6 hours while averaging 100 shots per scan
258
segment. The frequencies for the measured transitions in the region of 12,000 to 19,000 MHz
for the Eqt-t conformers of cyclohexylisocyanate are listed in Table 54 along with their
assignments. Also listed are the frequency differences between the measured values and
those obtained from the determined rotational constants and the centrifugal distortion
constants (Table 55).
The mid-infrared spectra of the gas (Fig. 44) and solid (Fig. 45) were recorded from
3500 to 300 cm-1 on a Perkin-Elmer model 2000 Fourier transform spectrometer equipped
with a nichrome wire source, Ge/CsI beamsplitter and DTGS detector. Atmospheric water
vapor was removed from the spectrometer housing by purging with dry nitrogen. The spectrum
of the gas was obtained with the samples contained in 12 cm cells equipped with CsI
windows. Interferograms obtained after 128 scans for the gas sample and the background
reference were transformed by using a boxcar apodization function with theoretical
resolutions of 0.5 cm-1 for the gaseous sample. For the spectrum of the solid, a theoretical
resolution of 2 cm-1 was used with 128 interferograms added and truncated. Multiple annealings
were required to obtain satisfactory spectra of the solid.
The mid-infrared spectra of cyclohexylisocyanate dissolved in liquefied xenon (Fig. 46)
were recorded on a Bruker model IFS-66 Fourier interferometer equipped with a Globar
source, Ge/KBr beamsplitter and DTGS detector. The interferograms were recorded at
variable temperatures ranging from 100 to 55 C with 100 scans and transformed by a
Blackman-Harris apodization function with a theoretical resolution of 1.0 cm-1.
The
temperature studies in liquefied xenon were carried out in a specially designed cryostat cell,
which is composed of a copper cell with a 4 cm path length and wedged silicon windows
sealed to the cell with indium gaskets. The temperature was monitored by two platinum
259
thermoresistors and the cell was cooled by the vapors from boiling liquid nitrogen. All of the
observed fundamental modes for the Eqt-t and Axl-t conformers in the infrared and Raman
spectra are listed in Tables 56 and 57, respectively.
The Raman spectrum of the liquid (Fig. 47) was recorded on a Spex model 1403
spectrophotometer equipped with a Spectra-Physics model 2017 argon ion laser operating on
the 514.5 nm line. The laser power used was 0.5 W with a spectral bandpass of 3 cm-1. The
spectrum of the liquid was recorded with the sample sealed in a Pyrex glass capillary.
To predict the energy differences (Table 58) among the four most likely conformers of
cyclohexylisocyanate, LCAO-MO-SCF calculations were performed with the Gaussian-03
program [43] by using Gaussian-type basis functions. The energy minima with respect to
nuclear coordinates were obtained by simultaneous relaxation of all geometric parameters
consistent with symmetry restrictions using the gradient method of Pulay [190]. A number
of basis sets starting from 6-31G(d), and increasing to 6-311G(3df,3pd), were employed at
the level of Møller-Plesset perturbation theory [38] to the second order (MP2), as well as
hybrid density functional theory by the B3LYP method. At all levels of calculations carried
out, with and without diffuse functions, predicted conformational stabilities varied
extensively with the size of the basis set. The predicted inversion pathway of the isocyanate
group calculated at the B3LYP/6-311+G(d,p) and MP2(full)/ 6-311+G(d,p) level is shown
in Figures 48 and 49.
To aid in making the vibrational assignment for the Eqt-t and Axl-t conformers (Tables
56 and 57, respectively), a normal coordinate analysis has been carried out by utilizing the
force field obtained from the Gaussian-03 program at the MP2(full)/6-31G(d) level. The
internal coordinates used to calculate the G and B matrices for cyclohexylisocyanate are
260
listed along with the structural parameters in Table 59. By using the B matrix, the force field
in Cartesian coordinates was converted to a force field in internal coordinates [45].
Subsequently, scaling factors of 0.88 for the CH stretches and 0.90 for all other modes except
for the coordinates concerning the heavy atom C -NCO angles, along with the geometric
average of scaling factors for interaction force constants, to obtain the fixed scaled force
fields and the resultant wavenumbers (Tables 56 and 57). A set of symmetry coordinates
(Table 60) was used to determine the corresponding potential energy distributions (P.E.D.s).
The observed and calculated wavenumbers of cyclohexylisocyanate along with the calculated
infrared intensities, Raman activities, depolarization ratios, and P.E.D.s are given in Tables
56 and 57.
In order to identify the fundamental vibrations for the different conformers of
cyclohexylisocyanate, the infrared (Fig. 44) and Raman (Fig. 47) spectra have been
simulated from the scaled ab initio MP2(full)/6-31G(d) results. Infrared intensities were
calculated based on the dipole moment derivatives with respect to Cartesian coordinates.
The derivatives were transformed into normal coordinate derivatives by (
(
u/
u/
Qi) =
j
Xj) Lij, where Qi is the ith normal coordinate, Xj is the jth Cartesian displacement
coordinate, and Lij is the transformation matrix between the Cartesian displacement
coordinates and the normal coordinates. The infrared intensities were then calculated by
[(N )/(3c2)] [(
x/
Qi)2 + (
y/
Qi)2 + (
z/
Qi)2].
The evaluation of Raman activity by
using the analytical gradient method has been developed [46, 47]. The activity Sj can be
expressed as: Sj = gj (45
2
j
+7
2
j
), where gj is the degeneracy of the vibrational mode j,
is the derivative of the isotropic polarizability, and
261
j
j
is that of the anisotropic polarizability.
The Raman scattering cross sections,
j/
, which are proportional to Raman activities, can
be calculated from the scattering activities as well as the predicted wavenumbers for each
normal mode, by using the following relationship [48, 49]:
(1−exp( −hc
j
/ kT))] [h / (8π2c j)] Sj, where
0
j/
= [(2π)4/45] [(
is the excitation wavenumber,
0
−
j
4
j)
/
is the
vibrational wavenumber of the jth normal mode, and Sj is the corresponding Raman scattering
activity. To obtain the polarized Raman scattering cross sections, the polarizabilities are
incorporated into Sj by multiplying Sj by (1
j)/(1+ j),
where
j
is the depolarization ratio of
the jth normal mode. The Raman scattering cross sections and calculated wavenumbers
obtained from the scaled ab initio force fields were used together with a Lorentzian function
to obtain the simulated Raman spectra.
Results
From the previous low resolution microwave studies [197], the data indicated that
there was predominately a single conformer present and, thus, it was important to determine
which conformation was the one giving rise to the observed microwave spectra. With this
determination it was possible to provide a confident vibrational assignment which then will
give sufficient information to identify the second conformer assuming that it is present. The
determined [197] B+C value of 1703 MHz indicated it either belonged to the Eqt-t form or
the second most likely Eqt-c conformer. By utilizing the ab initio MP2(full)/6-311+G(d,p)
calculations, B+C values were calculated for comparison to the four conformational forms:
Eqt-c, 1578; Axl-c, 1908; Eqt-t 1793; Axl-t, 2612 MHz.
Microwave Spectra
To assign the microwave spectrum, the preliminary rotational constants were
predicted from MP2(full)/6-311+G(d,p) calculations with adjustments made to the three
262
ringC-C distances consistent with those made for the
adjusted r0 values of the
halocyclohexanes molecules [189-191, 193, 194]. The parameters for the NCO-moiety were
also taken from adjusted r0 values of molecules recently studied. Based on the rotational
constants reported earlier, the structural adjustment resulted with two different angles of
136.0 and 139.8° for the ring-NCO angle. These two values were chosen to fit the previously
determined B+C value [197]. Initially, the predictions from the larger angle were used to
assign the spectrum but this value provided no reasonable fits. Predictions obtained with the
smaller angle were then utilized and several predicted transition in the 12000 MHz region
were observed as predicted. The transitions observed are the 717 ← 616, 707 ← 606, and 716 ←
615 transitions which were predicted at 12035, 12263, and 12695 MHz and the respective
observed frequencies differ by 30, 20, and 33 MHz. Once these three lines were assigned,
additional A-type lines J = 7←6 transition and eventually 8←7 and 9←8 transitions were
measured with a total of 33 transitions assigned (Table 54). The dipole moment components
were predicted to be | a| = 3.698 D and | c| = 0.285 D so the A-type transitions were expected
to dominate with the C-type transitions difficult to identify. Nevertheless, a search for Ctype transitions was carried out and 11 transitions were assigned (Table 54).
By utilizing the assigned transitions listed in Table 54, the rotational constant (Table
55) were obtained with uncertainties of 0.01 MHz for the B and C constants and 0.3 MHz for
the A constant.
The centrifugal distortion constants were also determined but only the DJ
and DJK constants had values (0.26
8 and 10.9 ± 0.5 kHz, respectively) with reasonable
uncertainties. This result is not unexpected because of the large amplitude motion associated
with most quasilinear molecules.
Vibrational Assignments
263
From the previous vibrational study [196], a relatively complete list of the bands
observed in the infrared and Raman spectra of the liquid at ambient temperature and from
low temperature (-175°C) solid were reported as well as those from the infrared spectrum of
the high pressure (~20 kbar) solid. Also from the infrared spectra of the liquid at 150, 30 and
-20°C, it was suggested that several band pairs should be suitable for determining an
enthalpy value: (Eqt-t/Axl-t) 1010/1020, 893/863, 881/863, 839/815, 780/759 cm-1. However,
an enthalpy determination was not carried out. In order to investigate these suggested band
pairs and to obtain a value of the enthalpy difference, a temperature dependent (-100 to 55°C) study of the infrared spectra was carried out.
However, to obtain these data it is
important to assign the spectra for the Eqt-t and Axl-t forms to correctly identify the band
pairs mentioned in the previously study [196]. Thus, it was necessary to record the Raman
spectrum of the liquid and infrared spectra of the gas and solid to provide additional spectral
data to confidently make the assignment for the two conformers.
The infrared spectrum of the gas was not very useful due to the band contours of most
of the fundamentals being relatively non-descript and, additionally, many of them have major
overlaps which obscured the contour. However, the infrared spectrum of the solid provided a
significant amount of information for assigning many of the fundamental bands for the
second conformer since they disappeared from the spectrum of the amorphous solid with
annealing (Fig. 45) to the crystalline solid. These fundamentals in the fingerprint spectral
region included the following bands (Table 57): 1399, 1326, 1321, 1234, 865, 820, 758, 599,
and 481 cm-1 which can be confidently assigned to the Axl-t conformer.
The assignment
thereafter was relatively simple since the assignment of the modes of the six-membered ring
moiety can be made rather straightforwardly based on those already assigned previously for
264
cyclohexane [188] as well as those for the monohalocyclohexanes [189-191].
The
frequencies of the observed fundamentals in the spectra of the gas, xenon solution, solid and
liquid for the most stable Eqt-t conformer are listed in Table 56 as well as those for the
second conformer (Axl-t) in Table 57.
The predicted infrared intensities and Raman activities as well as the predicted
depolarization values contributed significantly to the assignment. As has been observed
previously [188] the predicted Raman activities of certain ring modes are very poor,
particularly for
24,
which the ab initio calculations have consistently under-predicted for 6-
membered rings. It should be noted that, as with other cyclohexane molecules, there are
significant mixing for the vibrational modes, particularly those below 1000 cm-1. For the Eqtt conformer, there are five vibrations for which the maximum contribution of any symmetry
coordinate to the P.E.D. is less than 30%, and for
45, five
symmetry coordinates contribute
greater than 10% to the vibration, with the ring stretch to which it is assigned contributing
only 15% of the total P.E.D. This is consistent with previous results for ring motions in
cyclohexanes; therefore the descriptions listed are approximate and are for bookkeeping
purposes, rather than indicators of the primary molecular motions.
Conformational Stability
To determine the enthalpy difference between the Eqt-t and Axl-t form, the midinfrared spectra of cyclohexylisocyanate dissolved in liquefied xenon as a function of
temperature from -55 to -100 ºC were recorded. Only small interactions are expected to
occur between the dissolved molecules and the surrounding noble gas atoms [22-25], and as
a result, the “pseudo gas phase” spectrum shows only small frequency shifts compared with
the spectrum of the gas. A significant advantage of this type of cryogenic spectroscopic study
265
is that the conformer bands are better resolved in comparison with those in the spectrum of
the gas. This is particularly important because most of the conformer bands for this molecule
were expected to be observed within a few wavenumbers of each other. The different
conformer bands that are expected to be clearly identified are those in the spectral region
below 1000 cm-1 since they have the lowest probability of having intensity contributions
from overtones or combination bands.
The intensities of seven well-isolated and well-shaped conformational bands (five Eqt-t
and two Axl-t bands) confidently assigned in the previous section were measured as a
function of temperature and their ratios were determined as listed in Table 61. Only the
spectral data from -70 to -100°C provided reproducible results and, thus, the spectrum from 55 to -65°C were not utilized in the enthalpy determination. By application of the van’t Hoff
equation, lnK = H0/(RT)
S0/R, the enthalpy differences were determined from a plot of
lnK versus 1/T, where H/R is the slope of the line and K is substituted with the appropriate
intensity ratios, i.e. IEqt/IAxial.
It was assumed that
H,
S and the ratio of the molar
absorption coefficients εEqt/εAxial are not a function of temperature in the range studied.
The ten individually determined enthalpy differences are listed in Table 61. The
statistical average for each one of these values from the two bands of the Axl-t conformer
where five of the values were taken as a single set. Also, the average of all ten values is given
in Table 61 where all of the values were taken as a single set.
By this process, the
conformational enthalpy difference (Table 61) was determined to be 397
8 cm-1 (4.75
0.10 kJ/mol) from the xenon solution with the Eqt-t conformer more stable. The error limit
of ± 8 cm-1 is derived from the statistical standard deviation of one sigma of the measured
intensity data. Although the statistical uncertainties are quite small, they do not take into
266
account small associations with the liquid xenon or the interference of overtones and
combination bands in near coincidence with the measured fundamentals. The variations are
undoubtedly due to these types of interferences but by taking ten pairs it was hoped the effect
of such interferences would be minimized. Nevertheless, a more realistic error limit is
probably 10% and an appropriate enthalpy difference value is 397
40 cm-1 (4.75
0.47
kJ/mol).
Structural Parameters
In the initial microwave study [197], the nine assumed structural parameters utilized
(Table 59) were reported for an arbitrary conformer with a much shorter C-N and ringC-C
distances and a significantly larger Cα−N=C angle. In addition, there was no attempt to
determine the parameters from the resulting experimental B+C values [197]. In fact, the
values utilized were those reported from the microwave study of fluorocyclohexane [198,
199] and methylisocyanate [93]. However, structural parameters from the fluorocyclohexane
were not determined from those studies but from the values of isopropane. Therefore, the
parameters used earlier were quite different from those predicted from ab intio calculation.
We have found that good structural parameters for hydrocarbons and many substituted ones
can be determined by adjusting the ab initio MP2(full)/6-311+G(d,p) optimized structural
values to fit the experimentally determined rotational constants by using a computer program
“A&M” (Ab initio and Microwave) developed [29] in our laboratory.
It [33] has been shown that ab initio MP2(full)/6-311+G(d,p) calculations predict the
r0 structural parameters for more than fifty carbon-hydrogen distances for substituted
hydrocarbons to at least 0.002 Å compared to the experimentally determined values from
isolated CH stretching frequencies which were compared [34] to previously determined
267
values from microwave studies. Therefore, all of the carbon-hydrogen distances can be taken
from the MP2(full)/6-311+G(d,p) predicted values for both the Eqt-t and Axl-t conformers of
cyclohexylisocyanate. Also, it has been shown that triple bond distances are nearly constant
irrespective of the substitution on them [131] and the C=O of the NCO group behaves
similarly to a triple bond.
Additionally, there have been several determinations of the
N=C(O) bond for a variety of molecules and it has been found to be approximately the
average of the predicted values from the MP2 and B3LYP calculations with the 6311+G(d,p) basis set and this value was used as the predicted value to start the adjustment.
Finally, the N=C=O angle was fixed at the predicted value of 172.6º from the MP2(full)/6311+G(d,p) calculation.
This leaves seven remaining heavy atom parameters to be
determined. Even though only three rotational constants are available, reasonable adjusted r0
values can be obtained since it is possible to reduce the number of independent variables. In
order to reduce the number, the structural parameters are separated into sets according to
their types.
Bond lengths in the same set keep their relative ratio which then leave C α-N
and C-C heavy atom bond distances for cyclohexylisocyanate. Also, bond angles and
torsional angles in the same set keep their difference in degrees where this assumption is
based on the fact that the errors from ab initio calculations are systematic. Additionally, we
have also shown that the difference in predicted distances and angles from the ab initio
calculations for different conformers of the same molecule can usually be used as one
parameter, namely the ab initio predicted difference except for some dihedral angles.
However, for the Eqt-t and Axl-t conformers these dihedral angles are zero and 180° so they
are not variable parameters. Thus, it should be possible to obtain “adjusted r0” structural
parameters for cyclohexylisocyanate by utilizing the three determined rotational constants
268
since there are only three independent parameters by this process to be adjusted which are:
the C -N, C-C and N-Cα-Cβ,β'.
The determined parameters are listed in Table 59 along with estimated values for the
Axl-t conformer, where the same differences from the predicted parameters of the Eqt-t
conformer were applied to the Axl-t form. In comparing the structural parameters from the
previous study, some of the parameters utilized in the B+C prediction undoubtedly
contributed to the huge discrepancy between the predicted and estimated B+C value. Thus,
from our A&M fitting it was observed that the C -N bond distance and the C -N=C angle
have a huge effect on the rotational constants. Some other parameters that were used earlier
that are different from the adjusted r0 parameters included C=O, C -N, Cα Cβ,β', and a linear
NCO.
Discussion
Although the predicted fundamentals for the two conformers were obtained with a
relatively small basis set of MP2(full)/6-31G(d) with two scaling factors, the predicted
frequencies were in excellent agreement with the observed bands. For example, for the A'
modes of the Eqt-t conformer the average difference was 13 cm-1 which represents a
percentage error of 0.86 %. For the A" modes the average difference between the predicted
and observed frequencies were 13 cm-1, an error of 1.0 %. Similarly for the Axl-t conformer,
the percent error was under 1.0 % for both the A' and A" modes.
Therefore, it is not
necessary to utilize larger basis sets to adequately predict useful data. Also, the use of two
scaling factors does not create a significant problem in comparison to that obtained from
multiple scaling factors. Thus, the data obtained from the MP2(full)\6-31G(d) can be very
useful for the predictions of much of the data listed in Tables 56 and 57.
269
By utilizing the ab initio predicted structural parameters with slight adjustments, a set
of predicted transitions aided in the determination of the three rotational constants. Only Atype transitions were expected to be detected, but eleven C-type transitions were observed.
In all, forty-four transitions have been obtained but a significant number were from very
small Ka values, i.e. 1, 2, or 3. It was expected that the determination of the rotational
constants for the axial conformer would be very difficult due to the low abundance estimated
by using the ΔH value of 397
40 cm-1 (4.75
0.47 kJ/mol). At a temperature of 298 K, the
abundance of the axial form present should be 12.8% but the microwave measurement is
performed at 4 K, which brings the abundance of axial form to nearly zero.
From the observed transitions, only two reliable distortion constants were obtained.
The ab initio predicted value from the MP2(full)/6-311+G(d,p) calculation gives a value of
0.9265 where as the value from the B3LYP calculations with the same basis set is -6.9566
compare to the experimental value of 300.3(556). This is a meaningless result which is
mainly due to the small values of Ka required for the assignment of the microwave spectra.
There was difficulty in obtaining higher Ka values due to the large amplitude low frequency
vibration. However, the experimental value was obtained for DJ and DJK although there is
not a consistent agreement between the determined and the predicted values. Nevertheless,
the values with the statistical uncertainties provide results which are quite reasonable (Table
55).
Although cyclohexylisocyanate was previously studied by using both vibrational
[196] and microwave [197] spectroscopy but the results merely confirmed that the equatorial
form is energetically favored for the six-membered ring. These studies did not provide
essential information on the relative orientation of the NCO group but from the previous
270
vibrational study [196] the investigators determined that the molecule exists in both the
equatorial and axial conformers in the fluid phase. Under ca. 20 kbar at ambient temperature,
the compound crystallizes in the axial form, but under low temperature the compound
crystallizes in the equatorial conformation [196]. It is well known that conformational
volume greatly influences packing under high pressure for six-member rings [184, 196, 200204] and, therefore, it is not surprising that the axial form packs more favorably.
In the
current study, the conformation of both the ring and the subsitutent has been confidently
determined. From the infrared spectra of the solid [Fig. 45] the disappearance of at least six
Axl-t fundamental bands are clearly shown with annealing of the solid at low temperature,
and many of them had also been observed by Klaeboe and coworkers [196]. Thus, by using
determination were confidently identified. However, with the decrease in temperature band
intensities of the axial bands observed in the spectrum of the solid were not as apparent in the
xenon solution.
Only two axial bands showed sufficient change in band intensity to give
reproducible results. The 1322 cm-1 axial band was ideal for ΔH determination since the
decrease in intensity is very obvious and it has little interference from other bands. The
resulting ΔH = 397
40 cm-1 from 10 band pairs is in good agreement with the energy
difference predicted by the B3LYP calculations, which gave an average value of 374 cm-1.
Ab initio MP2 calculations have been very reliable in predicting conformational
stabilities with predicted energy difference usually in good agreement with the experimental
results.
However, for some of the substituted cyclohexane molecules B3LYP calculations
are producing an accurate enthalpy difference regarding conformation compared to those
from the MP2 calculations.
This occurrence [Table 62] has been observed for other
271
isoelectronic cyclohexane molecules, i.e. cyano-[193], isocyano- and ethynylcyclohexane
[194], with predicted energy differences from MP2 calculations favoring the axial form
(Table 62) and B3LYP calculations give the equatorial form. Only cyanocyclohexane has
experimentally been shown to be more stable in the axial form. The difference in the
predictions by method is probably due in part to the difference in the predicted structures
from MP2 and B3LYP (Table 63) calculations. The B3LYP calculations yield structural
parameters, particularly for the heavy atoms, that are much closer to the determined r0 values
which results in more accurate predictions of the vibrations of the ring and steric interactions
between the ring and the substituent.
Excellent structural parameters for the Eqt-t conformer have been obtained by
utilizing the three experimental rotational constants. The parameters for the ring and the
NCO moiety compare accordingly to other six-membered rings and other isocyanate alkanes,
respectively.
It should be noted that the C -N distance appears to be elongating with
increasing ring size, possibly due to the steric interaction of the ring and the NCO moiety.
Parameters for the Axl-t conformer were predicted in accordance with the adjustments made
for the Eqt-t structure and the known structure of other monosubstituted cyclohexanes. It
should be noted that the structural model utilized [197] in the initial microwave study was
based on the CH3NCO structure with a linear NCO moiety. In a recent structural study of
CH3NCO molecule [111], the effect of having a linear NCO on the resulting parameter was
shown for several microwave study on this molecule with the one utilized for Kitchin’s
microwave [197] being the initial one. In the study by Lett and Flygare [100], two structures
were proposed with Structure I including a tilt of the methyl group by 1.54 ± 0.23° whereas
for Structure II the tilt was fixed at zero. The inclusion of the tilt had minimum effect on the
272
determined C-N distance as well as the barrier to planarity but significantly affected the V3
term. In addition, the determined C -N=C angle is significantly affected by requiring the
NCO moiety to be linear which results in all three microwave studies [63, 100, 101] having
the C -N=C angle at 140°.
From our studies of monosubstituted cyclohexane molecules (Table 64) the more
stable conformer has been the equatorial form except for cyanocyclohexane [193]. However,
it should be noted that the enthalpy difference was obtained from a single conformer pair
which could lead to an incorrect determination of the conformer stability. This could be
possible if one of the bands chosen for the determination had a large amount of the other
conformer at approximately the same frequency.
The previous conformational stability
determination was made based on the stability at high pressure and at low temperature.
Several molecules were studied and at high pressure the more stable conformer was the axial
form and the equatorial form was found to be the more stable conformer at low temperatures.
However, for the cyanocyclohexane, the axial form is more stable in the both higher pressure
and at low temperature studies. Thus, the determination of the conformer stability was based
only on the low temperature effect. Thus this is the only molecule that contradicts the
predictions of the B3LYP predictions. However, a similar case has been reported for the
corresponding five membered ring so it is suggested that these two molecules should have
their conformation stability again determined to clearly verify that the cyano-cyclopentane
and -cyclohexane have significantly different conformer stabilities which are predicted from
the MP2(full)/6-311+G(d,p) calculation.
273
Table 54.
Rotational transitional frequencies (MHz) for equatorial-trans cyclohexylisocyanate in the ground
vibrational state.
Transition
274
61,6
60,6
62,5
64,2
62,4
90,9
61,5
22,0
22,1
71,7
51,4
70,7
72,6
74,3
74,4
73,5
73,4
72,5
71,6
32,1
32,2
81,8
51,5
50,5
52,4
54,1
52,3
81,7
51,4
11,0
11,1
61,6
40,4
60,6
62,5
64,2
64,3
63,4
63,3
62,4
61,5
21,1
21,2
71,7
(obs)
10339.34
10559.78
10638.04
10657.65
10729.36
10887.87
10918.90
11486.50
11581.00
12053.77
12283.03
12283.03
12404.31
12436.50
12436.52
12443.54
12449.23
12546.74
12728.03
13175.20
13452.95
13764.69
Transition
-0.04
-0.08
-0.19
0.03
0.33
0.03
-0.30
-0.07
0.02
-0.01
-0.28
-0.05
-0.18
-0.01
0.08
-0.10
-0.09
-0.29
0.53
0.17
0.00
0.03
80,8
82,7
84,5
84,4
83,6
83,5
82,6
81,7
42,2
42,3
91,9
90,9
92,8
94,5
94,6
93,7
93,6
92,7
91,8
52,3
71,6
52,4
70,7
72,6
74,4
74,3
73,5
73,4
72,5
71,6
31,2
31,3
81,8
80,8
82,7
84,4
84,5
83,6
83,5
82,6
81,7
41,3
60,6
41,4
(obs)
13992.56
14167.46
14216.62
14216.62
14225.41
14236.72
14376.01
14530.35
14830.38
15372.83
15471.99
15689.21
15927.12
15998.07
15998.11
16008.18
16028.80
16212.96
16324.54
16463.12
16547.34
17340.90
0.03
-0.18
0.17
-0.04
-0.10
-0.10
0.44
0.54
0.05
-0.10
0.10
0.13
-0.15
-0.20
0.35
-0.09
-0.07
-0.32
-0.29
0.11
0.29
-0.18
Table 55.
Rotational (MHz), centrifugal distortion (kHz) constants for the different conformers of
cyclohexylisocyanate.
MP2/631G(d)
275
A
B
C
DJ
DJK
DK
D1
D2
A
B
C
3573
927
833
0.1342
2.2709
0.5456
0.0023
0.0066
4229
833
733
Equatorial-Trans
MP2/6- B3LYP /6311+G(d,p) 311+G(d,p)
3511
943
850
0.139
1.9470
0.9265
0.0038
0.0075
3598
911
815
0.0869
10.3665
-6.9566
0.0018
-0.0192
Equatorial-Cis
4192
4199
839
828
739
726
MW
3546.87(28)
936.12(1)
839.06(1)
0.2637(804)
10.95(45)
300.3(556)
MP2/631G(d)
2318
1315
1269
0.5280
0.6826
-0.7388
-0.0697
0.0309
3163
995
917
Axial-Trans
MP2/6B3LYP /6311+G(d,p) 311+G(d,p)
2307
1331
1281
0.5434
0.6582
-0.7322
-0.0733
0.0336
Axial-Cis
3195
994
913
2358
1231
1209
0.8102
2.5406
-2.5976
-0.0828
0.1204
3176
974
894
Table 56.
Vib.
No.
276
A'
n1
n2
n3
n4
n5
n6
n7
n8
n9
n10
n11
n12
n13
n14
n15
n16
n17
n18
n19
n20
Calculateda and observed frequencies (cm-1) for cyclohexylisocyanate equatorial-trans (Cs) form.
Approx. description
ab fixed IR
b
initio scaled int.
(CH2)4 antisymmetric str. 3160
(CH2)4 antisymmetric str. 3152
CH2 antisym str.
3145
CH stretch
3109
CH2 symmetric str.
3092
(CH2)4 symmetric str.
3088
(CH2)4 symmetric str.
3088
NCO antisymmetric str. 2385
(CH2)4 deformation
1568
(CH2)4 deformation
1556
CH2 deformation
1547
NCO symmetric stretch 1475
(CH2)4 wag
1441
(CH2)4 wag
1427
CH in-plane bend
1381
(CH2)4 twist
1325
(CH2)4 twist
1309
(CH2)4 rock
1208
Ring stretch
1077
(CH2)4 rock
1054
Table Continues
Raman dp
act. ratio
IR
Gas
Xe
Raman
Solid liquid dp
c
P.E.D.
2964 53.4 34.4 0.15 2944 2946 2944 2945 0.19 74S1, 14S2
2957 17.3 102.4 0.56 ~2940 2937 2939 ~2945
54S2, 37S3
50S3, 19S2, 16S1
2950 28.5 153.3 0.31
2926 2926 2926
82S4, 13S7
2916 34.2 33.7 0.53 2917 2914 2912
2900 26.8 17.3 0.75
2898 2899 2902 0.26 42S5, 42S1, 13S6
2897
1.6 314.0 0.09
¾
2862 2861 0.07 67S6, 13S2, 12S5
2897 18.6
8.3 0.50 2858 2858 2859 ¾
41S7, 38S5
0.29
98S8
2263 874.8
0.9 0.05 2268 ~2261 ~2260 2257
72S9, 27S111
1488
2.3 1.7 0.49 1464 1465 1461 1466
1476 16.1
3.8 0.56 1452 1453 1451 1449 0.57 91S10
1468
1.2 16.1 0.73 1436 1435 1433 ~1430
67S11, 23S9
1399 10.6 24.6 0.19 1387 1384 1384 ~1400
61S12, 21S24, 16S15
0.62
35S13, 20S15, 13S14, 10S27
1367 18.8
9.5 0.54 1368 ~1362 1369 1364
1354
3.7 3.2 0.74 1348 1349 1347 1350 0.75 44S14, 27S13
1310
7.2 3.0 0.14 1308 1310 1306 1310 0.68 34S15, 17S16, 12S17, 11S13
1257
1.7 16.2 0.72 1265 1258 1261 ~1266
49S16, 24S17
1242
0.9 11.6 0.72 1247 1245 1243 1247 0.61 34S17, 17S16, 12S13, 10S15
1146
5.9 2.6 0.19 1148 1146 1143 1147 0.16 53S18, 16S22
1022
0.4 9.2 0.75 1036 1032 1036 ~1031 0.75 38S19, 25S14, 19S24
1000 22.7
7.0 0.43 1010 1010 1011 1012 0.39 31S20, 27S26, 10S30
Band
Contour
A C
1
39
89
55
7
81
99
81
12
25
58
92
99
99
97
97
28
60
90
96
99
61
11
45
93
19
1
19
88
75
42
8
1
1
3
3
72
40
10
4
Table 56 Continues
Vib.
No.
277
A'
n21
n22
n23
n24
n25
n26
n27
n28
n29
n30
n31
A"
n32
n33
n34
n35
n36
n37
n38
n39
n40
Approx. description
ringC-N stretch
CH2 rock
Ring stretch
Ring stretch
NCO in-plane bend
Ring puckering
Ring-NCO in-plane bend
Ring bending
Ring puckering
Ring bending
C-NC in-plane bend
ab fixed IR
b
initio scaled int.
Raman dp
act. ratio
IR
Gas
Xe
908
891
839
780
~620
522
~455
¾
¾
¾
¾
950
922
879
814
653
538
478
393
351
188
81
901
875
834
772
646
513
456
373
333
181
79
14.8
8.9
6.0
7.6
26.1
0.5
2.8
0.4
2.0
4.6
1.3
2.6
2.0
4.8
11.3
1.7
1.1
1.1
0.8
0.5
0.8
2.6
0.67
0.45
0.20
0.07
0.05
0.03
0.02
0.35
0.11
0.48
0.66
908
898
840
782
¾
525
¾
375
(CH2)4 antisymmetric str. 3156
(CH2)4 antisymmetric str. 3147
(CH2)4 symmetric stretch 3091
(CH2)4 symmetric stretch 3087
(CH2)4 deformation
1549
(CH2)4 deformation
1542
CH2 wag
1429
CH out-of-plane bend
1421
(CH2)4 wag
1407
2960
2952
2899
2896
1470
1463
1355
1349
1335
11.5
36.7
3.0
19.0
5.1
0.4
0.1
0.9
0.1
73.2
32.4
35.5
0.4
5.0
23.1
6.4
5.0
2.9
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
2946
Table Continues
Raman
Solid liquid dp
909
893
839
781
621
521
453
382
340
¾
895
841
783
¾
526
451
385
339
181
2861
2866
¾
1447 1447 1447 1443
~1424 1419 ~1426
¾
¾
1340 1342 1341 1339
¾
¾ 1325
0.57
0.06
0.04
0.05
0.22
0.13
0.17
0.44
c
P.E.D.
24S21, 40S23, 13S12
26S22, 23S20, 16S17, 10S21
34S23, 19S19, 17S22, 16S24
38S24, 26S18
75S25
22S26, 19S20, 14S30, 11S22
27S 27, 17S28, 10S30
67S28
66S29, 15S30
36S30, 27S26, 28S31
62S31, 15S27, 11S26
71S32, 22S33
64S33, 26S32
94S34
84S35
86S36
85S37, 13S36
38S38, 16S41, 10S45
44S39, 24S43
39S40, 28S38, 12S42
Band
Contour
A C
91
86
90
94
10
13
49
51
67
60
99
9
14
10
6
90
87
51
49
33
40
1
Table 56 Continues
278
Vib.
No.
Approx. description
A"
n41
n42
n43
n44
n45
n46
n47
n48
n49
n50
n51
n52
n53
n54
(CH2)4 wag
CH2 twist
(CH2)4 twist
Ring stretch
Ring stretch
(CH2)4 twist
Ring stretch
(CH2)4 rock
(CH2)4 rock
NCO out-of-plane bend
Ring twisting
Ring-NC out-plane bend
Ring twisting
C-NC out-of-plane bend
a
ab fixed IR
initio scaledb int.
1372
1325
1248
1156
1122
1099
968
930
821
571
460
349
243
34
1302
1257
1184
1097
1065
1043
919
882
779
571
436
331
231
34
0.0
2.3
0.1
0.8
0.5
0.5
0.0
7.8
0.3
18.2
0.1
5.1
0.0
0.0
Raman dp
act. ratio
5.5
6.3
3.7
0.1
2.7
4.7
0.7
0.4
0.9
0.2
0.8
0.4
0.0
3.2
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
IR
Gas
Xe
1287
1255
1193
1094
~1075
1052
¾
1288
1190
1091
1075
1051
¾
787
Raman
Solid liquid dp
1292
1257
1191
1092
1073
1054
¾
882
791
580
446
332
P.E.D.c
Band
Contour
A C
52S41, 17S46, 15S39
~1266
45S42, 16S43, 13S48
0.82
20S43, 18S45, 18S40
1192
58S44, 26S47, 11S43
¾
~1078 0.79 15S45, 19S46, 16S42, 16S40, 11S41
1054 0.74 47S46, 26S45, 13S38, 12S39
¾
27S47, 32S48, 11S44, 10S42
48S48, 22S47, 10S44
84S49, 11S45
97S50
64S51, 17S52
71S52, 22S51
92S53
100S54
MP2(full)/6-31G(d) ab initio calculated frequencies, infrared intensities (km/mol), Raman activities (Å4/u), depolarization ratios (dp)
and potential energy distributions (P.E.D.s).
b
Scaled frequencies with scaling factors of 0.88 for CH stretches and 0.90 for all other modes except C -NCO angles.
c
Symmetry coordinates with P.E.D. contribution less than 10% are omitted.
Table 57.
Vib.
No.
279
A'
n1
n2
n3
n4
n5
n6
n7
n8
n9
n10
n11
n12
n13
n14
n15
n16
n17
n18
n19
n20
Calculateda and observed frequencies (cm-1) for cyclohexylisocyanate axial-trans (Cs) form.
Approximate
description
(CH2)4 antisymmetric str.
(CH2)4 antisymmetric str.
CH2 antisymmetric str.
CH stretch
(CH2)4 symmetric str.
(CH2)4 symmetric str.
CH2 symmetric str.
NCO antisymmetric str.
(CH2)4 deformation
(CH2)4 deformation
CH2 deformation
NCO symmetric stretch
(CH2)4 wag
(CH2)4 wag
CH in-plane bend
(CH2)4 twist
(CH2)4 twist
(CH2)4 rock
Ring stretch
(CH2)4 rock
Table Continues
ab fixed IR Raman dp
b
initio scaled int. act. ratio
3160
3152
3143
3127
3100
3094
3084
2391
1569
1556
1542
1479
1446
1427
1404
1332
1296
1174
1080
1059
IR
Gas
Xe
2964 59.3 27.5 0.63
2957 12.4 116.9 0.55
2948 24.2 124 0.30
2934 39.2 135.6 0.19
2908 11.9 108.1 0.10
2902 19.5 140.9 0.15
2894 20.7 75.2 0.16
2269 725.8 0.5 0.07
1489
4 3.5 0.50
1476 13.4 5.8 0.54
1463
2.9 15.6 0.72
1404
8.8 17.2 0.17 1399
1372 12.5 2.8 0.37 1363 ~1362
1354
2 0.6 0.74
1332 16.5 8.5 0.48
1264
0.8 25.9 0.74 1269 1268
1229
3.5 6.4 0.41
1235
1114 14.7 1.8 0.67 ~1119 ~1120
1025
1.2 11.3 0.75 1023 1021
1005
3.2 0.1 0.54
Raman
Solid liquid dp
1457
1399*
1363
1351
1326*
1267 1269
1234*
1117 ~1121
1021 ~1031
1008
P.E.D.
c
87S1
74S2, 20S3
63S3, 13S2, 10S1
95S4
51S5, 44S6
45S6, 44S5
89S7
98S8
67S9, 16S10, 16S11
54S10, 44S11
38S11, 31S9, 28S10
60S12, 21S24, 15S15
37S13, 20S15, 12S16
67S14, 11S17
33S15, 21S13
76S16
51S17, 10S22
48S18, 13S22, 11S15
42S19, 28S13, 21S23
37S20, 33S27, 18S30
Band
Contour
A B
98
41
15
5
38
46
82
57
6
64
1
96
96
44
99
27
79
59
80
82
2
59
85
95
62
54
18
43
94
36
99
4
4
56
1
73
21
41
20
18
Table 57 Continues
Vib.
No.
Approximate
description
ab fixed IR Raman dp
b
initio scaled int. act. ratio
IR
Gas
Xe
Raman
Solid liquid dp
c
P.E.D.
Band
Contour
A B
A'
n21 Ring stretch
n22 CH2 rock
n23 Ring stretch
n24 ringC-N stretch
280
n25 Ring-NCO in-plane bend
n26 NCO in-plane bend
n27 Ring puckering
n28
n29
n30
n31
A"
n32
n33
n34
n35
n36
n37
n38
Ring bending
Ring puckering
Ring bending
C-NC in-plane bend
(CH2)4 antisymmetric str.
(CH2)4 antisymmetric str.
(CH2)4 symmetric stretch
(CH2)4 symmetric stretch
(CH2)4 deformation
(CH2)4 deformation
(CH2)4 wag
Table Continues
936
887
18.7
2.7 0.40
903
855
856
811
3.6
3.7
0.3 0.74
9.3 0.27
795
689
591
755
664
591
499
403
325
184
89
474
383
308
177
89
1.1
0.1
2.5
4
1.1
0.2
0.3
0.6
1.3
2.1
0.74
0.70
0.61
0.65
0.67
3157
3149
3101
3091
1556
1536
1434
2961
2954
2909
2899
1476
1457
1361
6.6
35.6
18.7
9.8
7.2
3.5
0
79.5
14.8
33.9
16.3
8.7
19.9
0.6
0.75
0.75
0.75
0.75
0.75
0.75
0.75
816
844
820*
15.2 11.6 0.01 757 761
19.3 3.5 0.02
5.8 0.6 0.36 ~591 ~598
758*
646
599*
1441
¾
40S21, 19S24, 15S17,
883 0.21 10S12
32S22, 25S21, 19S20,
10S14
817 0.10 43S23, 17S19, 15S18
21S24, 16S19, 12S23,
762 0.02 11S22
10S25, 34S26
54S26, 11S24, 10S20
~600
18S 27, 16S29, 16S20,
12S25, 10S30, 10S28
76S28
~313
50S29, 22S25, 15S30
36S30, 30S31, 26S27
62S31, 14S25, 11S30
82S32, 13S33
78S33, 17S32
56S34, 43S35
48S35, 42S34
65S36, 34S37
65S37, 33S36
42S38, 29S41, 10S39
99
1
100 0
79 22
89
1
11
11
99
89
1 99
100 0
24 76
33 67
95 5
Table 57 Continues
Vib.
No.
A"
n39
n40
n41
n42
Approximate
description
CH2 wag
(CH2)4 wag
CH out-of-plane bend
CH2 twist
1429
1391
1383
1331
1356
1320
1312
1263
1208 1146
5.3
n44 (CH2)4 twist
1169 1109
0
1134 1076
1083 1028
969 919
907 860
813 771
565 565
499 474
397 376
239 226
44
44
0
0.5
4.2
2.6
0.5
18.3
1.7
1.3
1.2
0.1
281
n46
n47
n48
n49
n50
n51
n52
n53
n54
Ring stretch
(CH2)4 rock
Ring stretch
(CH2)4 rock
NCO out-of-plane bend
Ring twisting
Ring-NC out-plane bend
Ring twisting
C-NC out-of-plane bend
IR
Gas
Xe
Raman
Solid liquid dp
0.75
0.75 1320 ~1322
0.75
~1310
0.75
¾
1321*
1312
1265
1266
2.8 0.75 1152 1152
1157
1157
0 2.4
0.2 0.1
0
3
1.2 15.2
n43 (CH2)4 twist
n45 Ring stretch
a
ab fixed IR Raman dp
initio scaledb int. act. ratio
0 0.75
1.4
6.9
1.2
0
0.9
0.1
0.2
0.6
0.1
2.8
¾
¾
0.75
0.75 1028 1024
0.75 928 927
0.75 864 863
0.75
0.75
0.75
¾ 482
0.75
0.75
0.75
¾
¾
1029 ~1031
928
927
865*
¾
481*
484
P.E.D.
c
40S39, 17S44, 11S40
67S40, 10S39
33S41, 21S44, 12S39
48S42, 18S43
36S43, 21S45, 13S48,
12S47
33S44, 32S38, 17S41,
10S42
30S45, 26S43, 16S48,
10S39
47S46, 18S40, 10S42
71S47
56S48, 28S45
83S49, 13S46
97S50
51S51, 35S52
42S52, 38S51
82S53, 17S52
100S54
MP2(full)/6-31G(d) ab initio calculated frequencies, infrared intensities (km/mol), Raman activities (Å4/u), depolarization
ratios (dp) and potential energy distributions (P.E.D.s).
b
Scaled frequencies with scaling factors of 0.88 for CH stretches, and 0.90 for all other modes except C -NCO angles.
c
Symmetry coordinates with P.E.D. contribution less than 10% are omitted.
*Indication of the disappearance of a band upon achieving crystalline solid.
Band
Contour
A B
Table 58.
Calculated electronic energies (hartree) and energy differencesa (cm-1) for the
axial-trans, equatorial-trans, axial-cis and equatorial-cis conformers of
cyclohexylisocyanate.
*
Method / Basis set
Axial-trans
Eqt-trans
Axial-cis*
Eqt-cis
MP2(full)/6-31G(d)
-402.0961016
52
63
279
MP2(full)/6-31+G(d)
-402.1164348
-52
36
95
MP2(full)/6-311G(d,p)
-402.4622969
56
-35
216
MP2(full)/6-311+G(d,p)
-402.4734178
-8
93
190
MP2(full)/6-311G(2d,2p)
-402.5725573
138
123
290
MP2(full)/6-311+G(2d,2p)
-402.5816218
79
183
257
MP2(full)/6-311G(2df,2pd)
-402.7234332
146
182
323
MP2(full)/6-311+G(2df,2pd)
-402.731205
82
231
288
MP2(full)/6-311G(3df,3pd)
-402.7583535
217
196
391
MP2(full)/6-311+G(3df,3pd)
-402.7645395
142
277
379
B3LYP/6-31G(d)
-403.3583913
-368
-38
-242
B3LYP/6-31+G(d)
-403.3713085
-433
-92
-368
B3LYP/6-311G(d,p)
-403.4682095
-329
-152
-294
B3LYP/6-311+G(d,p)
-403.4736348
-376
-86
-321
B3LYP/6-311G(2d,2p)
-403.4813968
-339
-38
-324
B3LYP/6-311+G(2d,2p)
-403.4867367
-396
-69
-339
a
Energy difference are relative to the Axial-trans conformer.
* Transition State
282
Table 59.
Structural parameters (Å and degree), rotational constants (MHz) and dipole
moment (debye) for equatorial-trans and axial-trans cyclohexylisocyanate from
the 6-311+G(d,p) basis set.
Equatorial-Trans
Parameter
r(N=C)
r(C=O)
r(Cα–N)
r(Cα–Cβ,β')
r(Cβ,β'–C γ,γ')
r(Cγ,γ'–Cδ)
r(Cα–H)
r(Cβ,β'–H20,19)
r(Cβ,β'–H18,17)
r(Cγ,γ'–H12,11)
r(Cγ,γ'–H14,13)
r(Cδ–H3)
r(Cδ–H2)
ÐN=C=O
ÐCα-N=C
Ð Cβ,β'CαN
ÐCβCαCβ'
ÐCαCβ,β'Cγ,γ'
Ð Cβ,β'Cγ,γ'Cδ
Ð CγCδCγ'
ÐNCαH
ÐCβ,β'CαH
Int.
Coord.
R8
R9
R7
R6, R5
R4, R3
R2, R1
r12
r11, r10
r9, r8
r5, r4
r7, r6
r2
r1
Λ
Λ2
q2, q1
f6
f5, f4
f3, f2
f1
z
ω2, ω1
Axial-Trans
MP2
B3LYP
MP2
1.217
1.181
1.453
1.528
1.530
1.530
1.096
1.095
1.099
1.098
1.095
1.095
1.098
172.4
132.8
111.1
110.9
110.7
111.1
110.8
106.1
108.7
1.200
1.175
1.456
1.537
1.536
1.535
1.095
1.094
1.097
1.098
1.094
1.094
1.097
173.6
139.1
111.5
111.3
111.0
111.6
111.4
105.8
108.3
1.214
1.182
1.453
1.532
1.529
1.530
1.094
1.095
1.098
1.097
1.095
1.095
1.099
172.4
136.8
110.8
111.2
111.2
111.0
111.1
105.7
109.1
Table Continues
283
B3LYP MW
1.197
1.177
1.457
1.541
1.535
1.536
1.092
1.093
1.097
1.097
1.094
1.094
1.098
173.7
142.7
111.4
111.2
112.7
111.5
111.5
105.4
108.6
a
1.207
1.171
1.437
1.526
1.526
1.526
1.096
180.0
140.0
111.7
Adj. r0
EqtTrans
1.208
1.167
1.455
1.534
1.529
1.534
1.096
1.095
1.099
1.098
1.095
1.095
1.098
172.5
135.3
110.9
111.1
110.3
111.1
110.8
106.1
108.8
Est. r0
AxialTrans
1.205
1.169
1.455
1.538
1.528
1.535
1.094
1.095
1.098
1.097
1.095
1.095
1.099
172.4
139.0
110.6
111.3
112.1
111.2
111.1
105.7
109.2
Table 59 Continues
Equatorial-Trans
Parameter
Ð CαCβ,β'H20,19
Ð CαCβ,β'H18,17
Ð Cγ,γ'Cβ,β'H20,19
Ð Cγ,γ'Cβ,β'H18,17
Ð Cβ,β' Cγ,γ'H12,11
Ð Cβ,β' Cγ,γ'H14,13
Ð CδCγ,γ'H12,11
Ð CδCγ,γ'H14,13
Ð Cγ,γ'CδH3
Ð Cγ,γ'CδH2
τ CαCβ Cβ' Cγ'
τ Cβ' Cγ'Cγ Cδ
A
B
C
|ma|
|mb|
|mc|
|mtot|
a
Int.
Coord.
n2, n1
µ2, µ1
σ2, σ1
π2, π1
δ2, δ1
κ2, κ1
ε2, ε1
η2, η1
a2, β2
a1, β1
Axial-Trans
MP2
B3LYP
MP2
109.4
108.2
111.2
109.8
109.2
109.8
109.3
110.6
110.3
109.2
128.5
129.3
109.5
108.6
110.8
109.8
109.3
109.6
109.4
110.4
110.2
109.2
129.6
130.8
109.2
107.5
111.1
109.3
110.1
109.4
109.5
110.4
110.2
109.0
131.9
129.3
109.4
107.2
111.0
109.3
109.9
109.4
109.5
110.4
110.2
109.2
133.3
130.6
2307
1331
1281
3.437
1.069
¾
3.599
2358
1231
1209
3.295
1.160
¾
3.493
3511
943
850
3.698
¾
0.285
3.709
3598
911
815
3.640
¾
0.351
3.657
Ref. [197].
284
B3LYP MW
a
Adj. r0
EqtTrans
109.4
108.3
111.2
110.1
109.4
109.6
109.3
110.6
110.3
109.2
128.0
129.1
3547
936
839
Est. r0
AxialTrans
109.2
107.5
111.2
109.3
110.1
109.3
109.5
110.4
110.2
109.0
132.1
129.2
2321
1304
1265
Table 60.
A¢
n1
n2
n3
n4
n5
n6
n7
n8
n9
n10
n11
n12
n13
n14
n15
n16
n17
n18
n19
n20
n21
n22
n23
n24
n25
n26
n27
n28
n29
Symmetry coordinates for cyclohexylisocyanate.
a
Description
Symmetry Coordinate
(CH2)4 antisymmetric stretch
r4 - r6 + r5 - r7 - r8 + r10 - r9 + r11
(CH2)4 antisymmetric stretch
r4 - r6 + r5 - r7 + r8 - r10 + r9 - r11
CH2 antisymmetric stretch
r1 - r2
CH stretch
r12
CH2 symmetric stretch
r 1 + r2
(CH2)4 symmetric stretch
r4 + r6 + r5 + r7 - r8 - r10 - r9 - r11
(CH2)4 symmetric stretch
r4 + r6 + r5 + r7 + r8 +r10 + r9 + r11
NCO antisymmetric stretch
R8 - R9
(CH2)4 deformation
4c1 - e1 - h1 - d1 - k1 + 4c2 - e2 - h2 - d2
- k2 + 4r1 - m1 - n1 - p1 - s1 + 4r2 - m2 n2 - p2 - s2
(CH2)4 deformation
4c1 - e1 - h1 - d1 - k1 + 4c2 - e2 - h2 - d2
- k2 - 4r1 + m1 + n1 + p1 + s1 - 4r2 + m2 +
n2 + p2 + s2
CH2 deformation
4g - a1 - a2 - b1 - b2
NCO symmetric stretch
R8 + R 9
(CH2)4 wag
e1 - d1 + h1 - k1 + e2 - d2 + h2 - k2 - p1 +
m1 - s1 + n1 - p2 + m2 - s2 + n2
(CH2)4 wag
e1 - d1 + h1 - k1 + e2 - d2 + h2 - k2 + p1 m1 + s1 - n1 + p2 - m2 + s2 - n2
CH in-plane bend
2z - q1 - q2 - w1 - w2
(CH2)4 twist
e1 - d1 - h1 + k1 + e2 - d2 - h2 + k2 + p1 m1 - s1 + n1 + p2 - m2 - s2 + n2
(CH2)4 twist
e1 - d1 - h1 + k1 + e2 - d2 - h2 + k2 - p1 +
m1 + s1 - n1 - p2 + m2 + s2 - n2
(CH2)4 rock
e1 - d1 + h1 - k1 + e2 - d2 + h2 - k2 + p1 s1 + m1 - n1 + p2 - s2 + m2 - n2
Ring stretch
R3 + R 4
(CH2)4 rock
e1 + d1 - h1 - k1 + e2 + d2 - h2 - k2 - p1 s1 + m1 + n1 - p2 - s2 + m2 + n2
C-CN stretch
R7
CH2 rock
a1 - a2 + b1 - b2
Ring stretch
R1 + R 2 - R 5 - R6
Ring stretch
R1 + R 2 + R 5 + R 6
NCO in-plane bend
Λ1
Ring puckering
f2 + f3 - f4 - f5
Ring-NCO in-plane bend
q1 + q2 - w1 - w2
Ring bending
f1 + f6
Ring puckering
f2 + f3 + f4 + f5
Table Continues
285
Table 60 Continues
A¢ n30
n31
A" n32
n33
n34
n35
n36
Description
Ring bending
C-NC in-plane bend
(CH2)4 antisymmetric stretch
(CH2)4 antisymmetric stretch
(CH2)4 symmetric stretch
(CH2)4 symmetric stretch
(CH2)4 deformation
n37 (CH2)4 deformation
n38 CH2 wag
n39 CH out-of-plane bend
n40 (CH2)4 wag
n41 (CH2)4 wag
n42 CH2 twist
n43 (CH2)4 twist
n44 Ring stretch
n45 Ring stretch
n46 (CH2)4 twist
n47 Ring stretch
n48 (CH2)4 rock
n49 (CH2)4 rock
n50
n51
n52
n53
n54
a
NCO out-of-plane bend
Ring twisting
Ring-NC out-of-plane bend
Ring twisting
C-NC out-of-plane bend
a
Symmetry Coordinate
f1 - f6
Λ2
r4 - r6 - r5 + r7 - r8 + r10 + r9 - r11
r4 - r6 - r5 + r7 + r8 - r10 - r9 + r11
r4 + r6 - r5 - r7 - r8 - r10 + r9 + r11
r4 + r6 - r5 - r7 + r8 + r10 - r9 - r11
4c1 - e1 - h1 - d1 - k1 - 4c2 + e2 + h2 + d2
+ k2 + 4r1 - m1 - n1 - p1 - s1 - 4r2 + m2 +
n2 + p2 + s2
4c1 - e1 - h1 - d1 - k1 - 4c2 + e2 + h2 + d2
+ k2 - 4r1 + m1 + n1 + p1 + s1 + 4r2 - m2 n2 - p2 - s2
a1 + a2 - b1 - b2
w1 - w2
e1 - d1 + h1 - k1 - e2 + d2 - h2 + k2 + p1 m1 + s1 - n1 - p2 + m2 - s2 + n2
e1 - d1 + h1 - k1 - e2 + d2 - h2 + k2 - p1 +
m1 - s1 + n1 + p2 - m2 + s2 - n2
a 1 - a 2 - b1 + b 2
e1 - d1 - h1 + k1 - e2 + d2 + h2 - k2 + p1 m1 - s1 + n1 - p2 + m2 - s2 + n2
R1 - R2 + R5 - R6
R1 - R2 - R5 + R6
e1 - d1 - h1 + k1 - e2 + d2 + h2 - k2 - p1 +
m1 + s1 - n1 + p2 - m2 - s2 + n2
R3 - R4
e1 + d1 - h1 - k1 - e2 - d2 + h2 + k2 + p1 +
m1 - s1 - n1 - p2 - m2 + s2 + n2
e1 + d1 - h1 - k1 - e2 - d2 + h2 + k2 - p1 m1 + s1 + n1 + p2 + m2 - s2 - n2
Λ3
f2 - f3 - f4 + f5
q1 - q2
f2 - f3 + f4 - f5
Λ4
Not normalized.
286
Table 61.
Temperature and intensity ratios of the conformer bands of cyclohexylisocyanate from the infrared spectra of the liquid
xenon solution.
T(°C)
Liquid
xenon
-70.0
-75.0
-80.0
-85.0
-90.0
-95.0
-100.0
-3
-1
1/T (´10 K )
4.9225
5.0467
5.1773
5.3149
5.4600
5.6132
5.7753
a
DH
I880 / I1322
1.3830
1.4783
1.5861
1.6720
1.9524
1.9660
2.1026
355 ± 29
0.6493
0.1806
0.1716
0.7138
0.2021
0.1757
0.7772
0.2165
0.1974
0.8458
0.2411
0.2128
0.9131
0.2674
0.2242
1.0004
0.2882
0.2522
1.0925
0.3065
0.2704
418 ± 10
437 ± 23
387 ± 19
-1
392 ± 11 cm (4.69 ± 0.13 kJ/mol)
0.1675
0.1826
0.2074
0.2161
0.2229
0.2490
0.2685
367 ± 30
I839 / I927
I880 / I927
I1191 / I927
287
I839 / I1322
T(°C)
-70.0
-75.0
-80.0
-85.0
-90.0
-95.0
-100.0
a
-3
-1
1/T (´10 K )
4.9225
5.0467
5.1773
5.3149
5.4600
5.6132
5.7753
7.7381
8.2036
8.9125
9.3011
11.2162
11.0345
11.7730
362 ± 40
I1051 / I1322
I1051 / I927
I1091/ I1322
I1091/ I927
I1191 / I1322
3.6234
1.0077
0.9574
0.9346
3.9091
1.1066
0.9623
1.0002
4.3285
1.2058
1.0994
1.1552
4.6919
1.3373
1.1805
1.1987
5.2234
1.5295
1.2828
1.2749
5.5959
1.6123
1.4110
1.3928
6.0729
1.7035
1.5029
1.4926
a
DH
428 ± 15
446 ±30
397 ± 23
377 ± 26
-1
402 ± 13 cm (4.81 ± 0.15 kJ/mol)
Average value obtained by utilizing all data as a single set gives DH = 397 ± 8 cm-1 (4.75 ± 0.10 kJ/mol) with the equatorialtrans conformer more stable.
Table 62.
Calculated electronic energies (hartree) and energy differencesa (cm-1) for the axial and equatorial conformers of several
substituted cyclohexanes.
Cyclohexylisocyanate
Method / Basis set
Cyanocyclohexane
Isocyanocyclohexane
Ethynylcyclohexane
Axial
Eqt
Axial
Eqt
Axial
Eqt
Axial
Eqt
MP2(full)/6-311G(d,p)
-402.462297
56
-327.355443
255
-327.317715
324
-311.262998
224
MP2(full)/6-311+G(d,p)
-402.473418
-8
-327.362623
172
-327.324865
221
-311.268708
135
MP2(full)/6-311G(2d,2p)
-402.572557
138
-327.445684
184
-327.408945
206
-311.350814
122
MP2(full)/6-311+G(2d,2p)
-402.581622
79
-327.451279
120
-327.414206
172
-311.355171
94
Averageb
66 ± 60
183 ± 56
231 ± 65
160 ± 56
288
B3LYP/6-311G(d,p)
-403.468209
-329
-328.206399
-142
-328.173676
-102
-312.102937
-230
B3LYP/6-311+G(d,p)
-403.473635
-376
-328.209717
-180
-328.176765
-174
-312.105129
-289
B3LYP/6-311G(2d,2p)
-403.481397
-339
-328.218097
-146
-328.186063
-152
-312.114852
-255
B3LYP/6-311+G(2d,2p)
-403.486737
-396
-328.221396
-195
-328.188761
-201
-312.117007
-309
Averageb
a
b
-360 ± 31
-166 ± 26
Energy difference are relative to the Axial conformer.
Arithmetic mean of the four highest basis set predictions with standard deviation error.
-157 ± 42
-271 ± 35
Table 63.
Structural parameters (Å and degree), rotational constants (MHz) for equatorial and axial c-C6H11XYZ
(XYZ = NC, CN, CCH) from the MP2(full) and B3LYP/6-311+G(d,p) basis set.
XYZ=
289
Parameter
r(Cδ –Cγ,γ')
r(C γ,γ'– Cβ,β')
r(Cβ,β'–Cα)
r(Cα–X)
r(XºY)
Ð CγCδCγ'
Ð CδCγ,γ' Cβ,β'
Ð Cγ,γ'Cβ,β'Cα
ÐCβCαCβ'
Ð Cβ,β'Cα X
Ð Cα-XºY
tCδCγ,γ'Cβ,β'Cα
A
B
C
Table Continues.
-CºN
Equatorial
MP2 B3LYP Adj. r0
1.529
1.535 1.536
1.529
1.534 1.532
1.537
1.548 1.544
1.464
1.464 1.464
1.175
1.154 1.161
110.8
111.4 110.9
111.3
111.8 110.8
110.5
110.9 110.2
110.0
111.3 110.9
110.9
111.1 110.8
179.0
180.0 179.0
56.2
55.1
56.9
4303
4224 4240
1403
1390 1400
1134
1120 1129
MP2
1.530
1.528
1.539
1.467
1.177
111.4
111.1
111.3
110.6
110.0
179.0
55.8
3021
1785
1585
-NºC
Axial
B3LYP
1.536
1.534
1.545
1.468
1.154
111.6
111.7
112.1
110.8
111.0
180.0
54.5
3036
1726
1514
Adj. r0
1.536
1.532
1.544
1.470
1.162
111.1
111.6
111.4
110.5
110.0
178.8
55.6
3005
1765
1563
MP2
1.529
1.529
1.529
1.428
1.186
110.8
111.1
110.3
111.4
110.0
179.0
56.3
4279
1470
1177
Equatorial
B3LYP
1.535
1.535
1.539
1.434
1.170
111.3
111.7
110.8
111.6
110.5
180.0
55.1
4238
1457
1164
Est. r0
1.536
1.532
1.534
1.440
1.166
110.9
111.6
111.4
111.3
110.0
179.0
55.3
4259
1467
1173
MP2
1.530
1.528
1.532
1.433
1.186
111.2
111.0
111.3
111.3
109.2
179.0
55.4
3035
1869
1640
Axial
B3LYP Est. r0
1.536
1.536
1.534
1.532
1.542
1.537
1.440
1.440
1.170
1.166
111.5
110.9
111.6
110.5
112.3
109.9
111.3
111.2
110.2
109.1
180.0
178.8
54.1
56.9
3070
3021
1808
1874
1571
1640
Table 63 Continues
-CºCH
XYZ=
290
Parameter
r(Cδ –Cγ,γ')
r(C γ,γ'– Cβ,β')
r(Cβ,β'–Cα)
r(Cα–X)
r(XºY)
Ð CγCδCγ'
Ð CδCγ,γ' Cβ,β'
Ð Cγ,γ'Cβ,β'Cα
ÐCβCαCβ'
Ð Cβ,β'Cα X
Ð Cα-XºY
tCδCγ,γ'Cβ,β'Cα
A
B
C
MP2
1.529
1.529
1.538
1.461
1.220
110.8
111.1
111.0
110.5
111.2
180.0
56.4
4278
1384
1122
Equatorial
B3LYP Adj. r0
1.535
1.532
1.534
1.544
1.548
1.541
1.462
1.462
1.204
1.210
111.4
110.8
111.8
111.0
111.4
110.2
110.7
111.1
111.6
110.7
180.0
179.9
55.1
56.7
4234
4248
1377
1386
1112
1122
MP2
1.530
1.528
1.540
1.465
1.221
111.2
111.0
111.4
110.0
110.8
180.0
56.3
2985
1745
1559
Axial
B3LYP
1.536
1.534
1.551
1.466
1.204
110.1
111.6
112.2
110.1
111.7
180.0
54.9
3020
1697
1499
Adj. r0
1.537
1.534
1.545
1.468
1.212
111.2
111.2
112.0
110.5
110.3
179.8
55.3
2995.8
1730.3
1541.0
Table 64.
Heavy atom r0 structural parameters of cyclohexane and some monosubstituted chair-equatorial cyclohexanes.
291
r(CδCγ) (Å)
r(CγCβ)
r(CαCβ)
r(CαX)
ÐCγCδCγ (°)
ÐCδCγCβ
ÐCαCβCγ
ÐCβCαCβ
ÐXCαCβ
tCδCγCβCα
A (MHz)
B (MHz)
C (MHz)
DH (calculated)
DH (observed)
a
Cyano a
1.536(3)
1.532(3)
1.544(3)
1.464(3)
110.9(5)
110.8(5)
110.2(5)
110.9(5)
110.8(5)
56.9(10)
4240.34
1399.67
1128.54
-175(42)
-63(9)h
-46(21)i
Germylb
1.533(3)
1.532(3)
1.540(3)
1.958(3)
111.1(5)
111.2(5)
111.3(5)
110.7(5)
111.1(5)
55.7(10)
4058.83
886.00
768.79
413(22)
453(38)
i
Silylc
1.534(3)
1.530(3)
1.544(3)
1.880(3)
111.0(5)
111.1(5)
111.5(5)
110.3(5)
111.6(5)
56.0(10)
4074.7
1366.6
1105.7
525(10)
i
520(70)
414(19) h
Methyld
1.535(3)
1.535(3)
1.536(3)
1.532(3)
110.8 (5)
111.0(5)
111.9(5)
110.0(5)
111.4(5)
55.7(10)
4201.8
2195.7
1593.6
644(34)
712(71)
i
Cyclohexanee
1.536(3)
1.536(3)
1.536(3)
-111.1(2)
111.1(2)
111.1(2)
111.1(2)
-55.7(2)
4305.90
4305.90
2463.30
Fluorof
1.533(3)
1.543(3)
1.518(3)
1.404(3)
111.0(5)
111.0(5)
110.0(5)
112.4(5)
108.7(5)
55.9(10)
4313.3
2188.7
1591.7
16(8)
Chlorof
1.532(3)
1.536(3)
1.524(3)
1.802(5)
110.6(5)
111.3(5)
109.7(5)
112.6(5)
110.1(5)
56.3(10)
4293.2
1397.6
1127.1
161(18)
Bromog
1.532(3)
1.539(3)
1.524(3)
1.966(5)
110.9(5)
111.3(5)
109.4(5)
112.5(5)
109.6(5)
55.9(10)
4280.9
894.8
775.9
168(22)
48(5)h
132(13) h
239(24) h
[193]
J.R. Durig, R.M. Ward, A.R. Conrad, M.J. Tubergen, G.A. Guirgis, J. Phys. Chem. A 114 (2010) 9289.[205]
c
J.R. Durig, R.M. Ward, A.R. Conrad, M.J. Tubergen, G.A. Guirgis, T.K. Gounev, J. Mol. Struct. 922 (2009) 19. [206]
d
J.R. Durig, R.M. Ward, G.A. Guirgis, T.K. Gounev, J. Raman Spectrosc. 40 (2009) 1919. [207]
e
Ref [188], f Ref [189], g Ref [190]
h
Infrared variable temperature study of xenon solution
i
Raman variable temperature study of the liquid
b
Figure 43.
Atomic numbering of cyclohexylisocyanate with the equatorial form shown.
The relative orientation of the –NCO moiety to the alpha H is indicated by the
τ(C=N Cα−H5) angle: Trans (τ=180°); Cis (τ=0°).
292
Figure 44.
Observed and predicted (MP2(ful1)/6-311+G(d,p)) infrared spectra of
cyclohexylisocyanate: (A) Gas; Simulated spectrum of (B) mixture of Eqt-t and
Axl-t conformers with H = 397 cm-1, (C) pure Eqt-t conformer, and (D) pure
Axl-t conformer.
293
294
Figure 45.
Infrared spectra (950-400 cm-1) of cyclohexylisocyanate: (A) Amorphous solid; (B) annealed solid. The asterisk indicates
that the band disappeared upon achieving crystalline solid.
Figure 46.
Variable temperature infrared spectra of cyclohexylisocyanate dissolved in
liquid xenon at 1322 cm-1 (Axl-t) and the 839 cm-1 (Eqt-t).
295
Figure 47.
Observed and predicted (MP2(ful1)/6-311+G(d,p)) Raman spectra of
cyclohexylisocyanate: (A) Liquid; Simulated spectrum of (B) mixture of Eqt-t
and Axl-t conformers with H = 397 cm-1, (C) pure Eqt-t conformer, and (D)
pure Axl-t conformer.
296
297
Figure 48.
Calculated potential function of equatorial cyclohexylisocyanate at the MP2(full)/6-311+G(d,p) and B3LYP/6311+G(d,p) level during the internal rotation of the NCO group as defined by the dihedral angle τ(C=N Cα−H5).
298
Figure 49.
Calculated potential function of axial cyclohexylisocyanate at the MP2(full)/6-311+G(d,p) and B3LYP/6-311+G(d,p)
level during the internal rotation of the NCO group as defined by the dihedral angle τ(C=N Cα−H5).
CHAPTER 12
RESULTS AND CONCLUSION
For each of the projects carried out, various spectroscopic methods along with
advance computational methods were utilized to get a comprehensive set of data on the
structure and conformationals stability of the molecules under study. In all, the researches
that are documented in this dissertation provide a comprehensive set of data that adequately
support the resulting reported conformational stability. In addition, some of the problems
that have been encountered in these researches pinpointed the direction of future
conformational analysis that might be of substantial interest.
It also demonstrates the
importance of an interdisciplinary approach to drawing a reliable conclusion.
For the XNCO (X = H, Cl, F, Br), adjusted r0 parameters have been obtained. The r0
values for BrNCO are: r(Br N) = 1.857(5); r(N=C) = 1.228(5); r(C=O) = 1.161(5) Å;
BrNC = 117.5 (5) and
NCO = 172.3(5) . For ClNCO the determined r0 parameters are in
excellent agreement with the previously determine rs values, whereas those for HNCO the
HNC angle is larger with a value of 126.3(5) compared to the previous reported value of
123.9(17) . However, considering the relatively large uncertainty in the value given initially
the two results are in near agreement.
Structural parameters are also estimated for FNCO
and XOCN (X = H, F, Cl, Br). The centrifugal distortion constants have been calculated and
are compared to the experimentally (XNCO: X = H, Cl, Br) determined values. Predicted
values for the barriers of linearity are given for both the XNCO (X = H, F, Cl, Br) molecules
and the results were compared to the corresponding isothiocyanate molecules. The predicted
frequencies for the fundamentals of the XNCO molecules compare favorably to the
experimental values but some of the predicted intensities differ significantly from those in
299
the observed spectra.
The two OCN bends for HOCN have been assigned and the
frequencies for the two corresponding fundamentals of DOCN are predicted.
For methylisocyanate, fine structures of the nearly free internal rotation of the methyl
rotor has been observed for the pseudodegenerate CH3 stretch and deformation from which
the band centers and Coriolis coupling constants have been determined. Several differences
are noted between the predicted and experimental values. By combining the three previously
reported rotational constants for CH3NCO with the ab initio MP2/6-311+G(d,p) predicted
structural values, adjusted r0 parameters have been obtained.
The r0 values for the distance
(Å) are: r(C N) = 1.447(3); r(N=C) =1.215(3); r(C=O) = 1.166(3); r(C-Ha) =1.089(2); r(CHs) = 1.093(2), and for the angles (degrees):
= 108.6(5);
CNC = 135.9(5);
NCO = 172.6(5);
NCHa
NCHs =110.8(5).
For ethylisocyanate, the spectroscopic data indicate two conformers in the fluid states
which are the cis and trans forms with a large proportion of molecules in the gas phase at
ambient temperature in the excited states of the NCO torsional mode which has a very low
barrier to conformational interchange. Variable temperature ( 110 to
155 C) studies of
krypton solutions were carried out and by using two conformer pairs, an enthalpy difference
of 100 ± 4 cm-1 (1.20 ± 0.05 kJ/mol) was obtained with the cis conformer the more stable
form. To aid in the analyses of the vibrational and rotational spectra, ab initio calculations
have been carried out by the perturbation method to the second order (MP2) with full
electron correlation using a variety of basis sets up to 6-311+G(2df,2pd) and cc-PVQZ. With
the basis sets 6-311+G(2d,2p) and larger, the barrier at the cis position ranged from a low
value of 11 cm-1 to a high value of 31 cm-1 with a value of 19 cm-1 from the largest basis set
of cc-PVQZ. Thus, the gauche well is probably so shallow that it does not contain a bound
300
vibrational state. This results in the cis conformer as the most stable form which is consistent
with the experimental rotational and vibrational data. The predicted energy difference from
these calculations between the cis conformer and the transitional-state skew form is ~100 cm1
which is consistent with the assigned microwave lines for four excited states of the NCO
torsion.
For isopropylisocyanate, 18 transitions for the more stable trans conformer were
assigned and A=6693.23(15), B=2263.10(3), C=1960.05(2) MHz were obtained.
By
utilizing these rotational constants along with ab initio MP2(full)/6-311+G(d,p) predicted
structural values, adjusted r0 parameters have been obtained for the trans conformer and
estimated values for the gauche conformer.
Variable temperature Raman and infrared
spectra of xenon solutions have been recorded and a ΔH value of 115 11 cm-1 (1.38±0.13
kJ/mol) has been determined. The conformational stabilities have been predicted from ab
initio calculations utilizing several different basis sets up to cc-PVQZ for MP2(full) and 6311+G(3df,3pd) for density functional theory calculations by the B3LYP method. From the
MP2(full)/cc-PVQZ calculations an energy difference of 87 cm-1 (1.04 kJ/mol) is predicted
between the trans and the gauche forms and 168 cm-1 (2.01 kJ/mol) for the cis form which is
a transition state.
For dimethylsilylisocyanate, the low wavenumber Raman spectrum of the gas with a
significant number of Q-branches for the SiNC(O) bend is consistent with an essentially
linear SiNCO moiety. The ab initio calculations supported this conclusion since all possible
orientations of the NCO moiety leads to nearly the same energy. This result is at variance
with the conclusion from the electron diffraction study that the heavy atom skeleton was bent
with an angle of 152(5)° with one stable cis conformer. It is believed that this reported angle
301
difference from 180° is due to the shrinkage effect. The SiH distance of 1.486 Å has been
obtained from the isolated SiH stretching wavenumber.
For cyclopropylisocyanate, the enthalpy difference has been determined to be 77
cm-1 (0.92
8
0.10 kJ/mol) with the trans form more stable than the cis conformer with 59
2 % present at ambient temperature.
By utilizing three rotational constants for each
conformer, combined with structural parameters predicted from MP2(full)/6-311+G(d,p)
calculations, the adjusted r0 parameters have been obtained.
Heavy atom structural
parameters for the trans [cis] conformers are: distances (Å) (C–C2,3) = 1.509(3) [1.509(3)],
(C2–C3) = 1.523(3) [1.521(3)], (C–N) = 1.412(3) [1.411(3)], (N═C) =1.214(3) [1.212(3)],
(C═O) = 1.163(3) [1.164(3)]; angles (°)
CCN = 116.7(5) [120.1(5)],
CNC = 136.3(5)
[137.6(5)].
The microwave spectrum of cyclobutylisocyanate, c-C4H7NCO, has been investigated
from 21,000 to 11,000 MHz and eleven transitions for the more stable equatorial-trans
conformer were assigned. The rotational constants of the ground vibrational state have been
determined and the molecule has been shown to be a near symmetric prolate rotor (К = -0.99).
The B and C rotational constants have been confidently determined to be B=1508.68(3) and
C=1476.55(2) MHz, respectively, whereas the value for the A rotational constant of
6891(302) MHz had a large uncertainty. Spectral data indicated the present of three
conformers in the fluid states which are the equatorial-trans, equatorial-gauche and axialtrans forms. The second part of the conformational name refers to the relative position of the
NCO moiety relative to the alpha hydrogen. By utilizing four conformer pairs, an enthalpy
difference of 131 ± 13 cm-1 (1.57 ± 0.16 kJ/mol) was obtained with the equatorial-trans
302
conformer the more stable form, which is in good agreement with the ab initio predicted
value of 137 ± 36 cm-1 (1.64 ± 0.43 kJ/mol).
The cyclohexylisocyanate has been investigated from 10,000 to 21,000 MHz and
forty-four transitions for the more stable equatorial-trans conformer were assigned. The
rotational constants of the ground vibrational state were determined: A=3546.87(28),
B=936.12(1) and C=839.06(1) MHz.
With these rotational constants and ab initio
MP2(full)/6-311+G(d,p) predicted structural values, adjusted r0 parameters for the
equatorial-trans conformer were obtained. Two conformers (equatorial-trans and axialtrans) were identified in the fluid states. An enthalpy difference of 397 40 cm-1 (4.75 0.47
kJ/mol) was obtained from seven conformer pairs with the equatorial-trans form more stable.
Each one of these projects will show that it was essential to use an interdisciplinary
approach to get a complete picture of the result. It is sometime the case that each technique
will contradict another. For these isocyanate molecules it was necessary to study it under all
phases of aggregation and to observe the changes that take place with changes. Each
technique has its own particular problems but the combination of all techniques provies a
much more reliable support for the concluding result. To limit the conformational analysis to
only one technique would lead to a spurious interpretation.
With the isocyanate molecules it was very apparent the benefits of the low
temperature spectra with the sample dissolved in the noble gas solution.
Isocyanate
molecules tends to have a very low barrier to internal rotation and often time the spectra
observed are very non-descript in the fluid phase and the band center is difficult to assign.
However, the peaks observed in the xenon solutions are well resolved and close to the gas
phase frequency. One of the disadvantages of this technique are problems that are associated
303
with dimers and other aggregated molecular cluster. Fortunately, this was not a problem with
the isocyanate molecules studied. Based on this technique, many conformer bands were
confidently assigned and reasonable enthalpy values obtained.
As can be observed with the ab intio predictions for the ethylisocyanate and the
cyclohexylisocyanate, gaussian basis sets for large molecule are inadequate to correctly
simulate the experimental results. This is an important implication since it is a goal of
computational methods to be able to be applied to much large molecules such as proteins.
304
References
1.
Wilson, E.B., Chem. Soc. Rev., 1 (1972) 293.
2.
Brown, H.C., Horowitz, R.H., J. Am. Chem. Soc., 77 (1955) 1733.
3.
Taft, R.W., J. Am. Chem. Soc., 75 (1953) 4538.
4.
Edsall, J.T., Structural chemistry and moelcular biology, ed. A. Rich and N.
Davidson168, San Francisco: W. H. Freeman and Co.
5.
Kemp, J.D., Pitzer, K.S., J. Am. Chem. Soc., 59 (1937) 276.
6.
Kemp, J.D., Pitzer, K.S., The Journal of Chemical Physics, 4 (1936) 749.
7.
Pitzer, K.S., J. Chem. Phys., 8 (1940) 711.
8.
Pitzer, K.S., Kilpatrick, J.E., Chemical Reviews, 39 (1946) 435.
9.
Teller, E., Topley, B., J. Chem. Soc. (Resumed), (1935) 876.
10.
Eyring, H., J. Am. Chem. Soc., 54 (1932) 3191.
11.
Jones, G.I.L., Owen, N.L., J. Mol. Struct., 18 (1973) 1.
12.
Hirota, E., Sugisaki, R., Nielsen, C.J., Sørensen, G.O., J. Mol. Spectrosc., 49 (1974)
251.
13.
Marstokk, K.M., Møllendal, H., J. Mol. Struct., 18 (1973) 247.
14.
Riveros, J.M., Wilson Jr, E.B., The Journal of Chemical Physics, 46 (1967) 4605.
15.
Mizuchimia, S.I., Structure Of Molcules and Internal Rotation, 1954, New York:
Academic Press.
16.
Lefevre, R.J.W., Advances In Physical Organic Chemistry. Advances In Physical
Organic Chemistry, ed. V. Gold. Vol. 3. 1965, New York: Academic Press.
17.
Emsley, J.W., Feeney, J., Sutcliffe, L.H., High Resolution Nuclear Magnetic
Resonance Spectroscopy, 1965, Oxford: Pergamon Press.
18.
Clark, A.H., Internal Rotation In Molecules, ed. W.J. Orville-Thomas1974, New
York: Wiley-Interscience.
19.
Durig, J.R., Durig, D.T., Durig, B.R., Sullivan, J.F., J. Mol. Struct., 126 (1985) 83.
305
20.
Van Der Veken, B.J., Vibrational Spectra And Structure. Vibrational Spectra And
Structure, ed. J.R. Durig. Vol. 15. 1986, Amsterdam: Elsevier.
21.
Drew, B.R., Conformational Studies Of Some Molecules With Double and Triple
Bonds By Variable Temperature Ft-Ir Studies Of Rare Gas Solutions and Ab Initio
Calculations, 2001, University of Missouri - Kansas City: United States -- Missouri.
p. 373.
22.
Bulanin, M.O., J. Mol. Struct., 19 (1973) 59.
23.
Bulanin, M.O., J. Mol. Struct., 347 (1995) 73.
24.
Van Der Veken, B.J.,De Munck, F.R., J. Chem. Phys., 97 (1992) 3060.
25.
Herrebout, W.A., Van Der Veken, B.J., Wang, A., Durig, J.R., J. Phys. Chem., 99
(1995) 578.
26.
Wilson, J.E.B., J. Chem. Phys., 15 (1947) 736.
27.
Wilson, J.E.B., J. Chem. Phys., 7 (1939) 1047.
28.
Schachtschneider, J.H.,Snyder, R.G., Spectrochim Acta, 19 (1963) 117.
29.
Van Der Veken, B.J., Herrebout, W.A., Durig, D.T., Zhao, W., Durig, J.R., J. Phys.
Chem. A, 103 (1999) 1976.
30.
Pauling, L.,Wilson Jr, E.B., Introduction To Quantum Mechanics, 1935, New York,
NU: McGraw-Hill.
31.
Morse, P.M., Physical Review, 34 (1929) 57.
32.
Wilson Jr, E.B., Decius, J.C., Cross, P.C., Molecular Vibrations - The Theory Of
Infrared And Raman Vibrational Spectra,1955, New York, NY: McGraw-Hill.
33.
Durig, J.R., Kar, W.N., Zheng, C., Shen, S., Struct. Chem., 15 (2004) 149.
34.
McKean, D.C., J. Mol. Struct., 113 (1984) 251.
35.
Hebre, W.J., Radom, L., Schleyer, P.V.R., Pople, J.A., Ab Initio Molecular Orbital
Theory, 1986, New York: John Wiley.
36.
Szabo, A.,Ostlund, N.S., Modern Quantum Chemistry: Introduction To Advanced
Electronic Structure Theory, 1982, New York: Macmillan Publishing Co.
37.
Roothaan, C.C.J., Rev. Mod. Phys., 23 (1951) 69.
306
38.
Møller, C., Plesset, M.S., Phys. Rev., 46 (1934) 618.
39.
Foresman, J.B.,Frisch, M.J., Exploring Chemistry With Electronic Structure Methods.
Second Edition 1996, Pittsburgh, PA: Gaussian Inc.
40.
Hohenberg, P., Kohn, W., Phys. Rev., 136 (1964) B864.
41.
Kohn, W., Sham, L.J., Phys. Rev., 140 (1965) A1133.
42.
Becke, A.D., Phys. Rev. A, 38 (1988) 3098.
43.
Frisch, M.J., Burant, J.C., Millam, J.M., Iyengar, S.S., Tomasi, J., Barone, V.,
Mennucci, B., Cossi, M., Scalmani, G., Rega, N., Petersson, G.A., Trucks, G.W.,
Nakatsuji, H., Hada, M., Ehara, M., Toyota, K., Fukuda, R., Hasegawa, J., Ishida, M.,
Nakajima, T., Honda, Y., Kitao, O., Schlegel, H.B., Nakai, H., Klene, M., Li, X.,
Knox, J.E., Hratchian, H.P., Cross, J.B., Bakken, V., Adamo, C., Jaramillo, J.,
Gomperts, R., Scuseria, G.E., Stratmann, R.E., Yazyev, O., Austin, A.J., Cammi, R.,
Pomelli, C., Ochterski, J.W., Ayala, P.Y., Morokuma, K., Voth, G.A., Salvador, P.,
Robb, M.A., Dannenberg, J.J., Zakrzewski, V.G., Dapprich, S., Daniels, A.D., Strain,
M.C., Farkas, O., Malick, D.K., Rabuck, A.D., Raghavachari, K., Foresman, J.B.,
Cheeseman, J.R., Ortiz, J.V., Cui, Q., Baboul, A.G., Clifford, S., Cioslowski, J.,
Stefanov, B.B., Liu, G., Liashenko, A., Piskorz, P., Komaromi, I., Montgomery Jr,
J.A., Martin, R.L., Fox, D.J., Keith, T., Al-Laham, M.A., Peng, C.Y., Nanayakkara,
A., Challacombe, M., Gill, P.M.W., Johnson, B., Chen, W., Vreven, T., Wong, M.W.,
Gonzalez, C., Pople, J.A., Kudin, K.N., Gaussian 03, Revision B.05, (2004).
44.
Pulay, P., Mol. Phys., 17 (1969) 197.
45.
Guirgis, G.A., Zhu, X., Yu, Z., Durig, J.R., J. Phys. Chem. A, 104 (2000) 4383.
46.
Frisch, M.J., Yamaguchi, Y., Gaw, J.F., Schaefer Iii, H.F., Stephen Binkley, J., J.
Chem. Phys., 84 (1985) 531.
47.
Amos, R.D., Chem. Phys. Lett., 124 (1986) 376.
48.
Polavarapu, P.L., J. Phys. Chem., 94 (1990) 8106.
49.
Chantry, G.W., The Raman Effect, ed. A. Anderson. Vol. 1. 1971, New York, NY:
Marcel Dekker Inc.
50.
Fishman, A.I., Stolov, A.A., Remizov, A.B., Spectrochim. Acta A, 41 (1985) 505.
51.
George, W.O., Goodfield, J.E., Analytical Application Of Ft-Ir To Molecular and
Biological System. Analytical Application Of Ft-IR To Molecular and Biological
System, Ed. J.R. Durig, 1980, Dordrecht, Holland: D. Deidel.
307
52.
Klaeboe, P., Vib. Spectrosc., 9 (1995) 3.
53.
Suenram, R.D., Grabow, J.U., Zuban, A., Leonov, I., Rev. Sci. Instrum., 70 (1999)
2127.
54.
Suenram, R.D., Lovas, F.J., Plusquellic, D.F., Lesarri, A., Kawashima, Y., Jensen,
J.O., Samuels, A.C., J. Mol. Spectrosc., 211 (2002) 110.
55.
Shen, S.,Durig, J.R., J. Mol. Struct., 661-662 (2003) 49.
56.
Durig, D.T., Durig, M.S., Durig, J.R., Spectrochim. Acta A, 61 (2005) 1287.
57.
Durig, J.R., Zheng, C., Deeb, H., J. Mol. Struct., 784 (2006) 78.
58.
Durig, J.R., Zhou, X., El Defrawy, A.M., Guirgis, G.A., Gounev, T.K., Zheng, C., J.
Mol. Struct., 839 (2007) 107.
59.
Zheng, C., Guirgis, G.A., Deeb, H., Durig, J.R., J. Mol. Struct., 829 (2007) 88.
60.
Durig, J.R.,Zheng, C., Spectrochim. Acta A, 68 (2007) 783.
61.
Palmer, M.H.,Nelson, A.D., J. Mol. Struct., 689 (2004) 161.
62.
Yamada, K., J. Mol. Spectrosc., 79 (1980) 323.
63.
Lett, R.G.,Flygare, W.H., J. Chem. Phys., 47 (1967) 4730.
64.
Reid, C., J. Chem. Phys., 18 (1950) 1544.
65.
Goubeau, J., Chem. Ber., 68 (1935) 912.
66.
Dixon, R.N.,Kirby, G.H., Trans. Faraday Soc., 64 (1968) 2002.
67.
Botschwina, P., Nachbaur, E., Rode, B.M., Chem. Phys. Lett., 41 (1976) 486.
68.
Rode, B.M., Kosmus, W., Nachbaur, E., Chem. Phys. Lett., 17 (1972) 186.
69.
Hocking, W.H., Gerry, M.C.L., Winnewisser, G., Can. J. Phys., 53 (1975) 1869.
70.
Brown, S.S., Berghout, H.L., Crim, F.F., J. Chem. Phys., 107 (1997) 8985.
71.
Gholivand, K., Willner, H., Bielefeldt, D., Haas, A., Z. Naturforsch., 39 B (1984)
1211.
72.
Jacobs, J., Jülicher, B., Schatte, G., Willner, H., Mack, H.G., Chem. Ber., 126 (1993)
2167.
308
73.
Christen, D., Oberhammer, H., Hammaker, R.M., Chang, S.C., Desmarteau, D.D., J.
Am. Chem. Soc., 104 (1982) 6186.
74.
Lide Jr, D.R., The Journal of Chemical Physics, 38 (1963) 456.
75.
Pierce, L., Hayes, R.G., Beecher, J.F., The Journal of Chemical Physics, 46 (1967)
4352.
76.
Devore, T.C., J. Mol. Struct., 162 (1987) 287.
77.
Hocking, W.H.,Gerry, M.C.L., J. Mol. Spectrosc., 42 (1972) 547.
78.
Hocking, W.H., Williams, M.L., Gerry, M.C.L., J. Mol. Spectrosc., 58 (1975) 250.
79.
Gottardi, W., Angew. Chem. Int., Ed. Engl., 10 (1971) 416.
80.
Gottardi, W., Darstellung und Eigenschaften von Bromisocyanat, 103 (1972) 1150.
81.
Gerke, M., Schatte, G., Willner, H., J. Mol. Spectrosc., 135 (1989) 359.
82.
Jemson, H.M., Lewis-Bevan, W., Westwood, N.P.C., Gerry, M.C.L., Chem. Phys.
Lett., 108 (1984) 496.
83.
Jemson, H.M., Lewis-Bevan, W., Westwood, N.P.C., Gerry, M.C.L., J. Mol.
Spectrosc., 118 (1986) 481.
84.
Hensel, K.D., Lam, M.E., Gerry, M.C.L., Willner, H., J. Mol. Spectrosc., 151 (1992)
184.
85.
Jacox, M.E.,Milligan, D.E., The Journal of Chemical Physics, 40 (1964) 2457.
86.
Sakaizumi, T., Mure, H., Ohashi, O., Yamaguchi, I., J. Mol. Spectrosc., 140 (1990)
62.
87.
Sakaizumi, T., Namikawa, M., Ohashi, O., J. Mol. Spectrosc., 345 (1995) 185.
88.
Kim, H., Pearson, E.F., Appelman, E.H., J. Chem. Phys., 56 (1992) 1.
89.
Anderson, W.D., Gerry, M.C.L., Davis, R.W., J. Mol. Spectrosc., 115 (1986) 117.
90.
Koga, Y., Takeo, H., Kondo, S., Sugie, M., Matsumura, C., Mcrae, G.A., Cohen, E.A.,
J. Mol. Spectrosc., 138 (1989) 467.
91.
Pasinszki, T.,Westwood, N.P.C., J. Phys. Chem., 99 (1995) 1649.
309
92.
Palmer, M.H.,Nelson, A.D., J. Mol. Struct., 660 (2003) 49.
93.
Curl Jr, R.F., Rao, V.M., Sastry, K.V.L.N., Hodgeson, J.A., J. Chem. Phys., 39
(1963) 3335.
94.
Jokes, L.H., Shoolery, J.N., Shulman, R.G., Yost, D.M., J. Chem. Phys., 18 (1950)
990.
95.
Eyster, E.H., Gillette, R.H., Brockway, L.O., J. Am. Chem. Soc., 62 (1940) 3236.
96.
Hirschmann, R.P., Kniseley, R.N., Fassel, V.A., Spectrochim. Acta, 20 (1964) 809.
97.
Hirschmann, R.P., Kniseley, R.N., Fassel, V.A., Spectrochim. Acta, 21 (1965) 2125.
98.
Fateley, W.G.,Miller, F.A., Spectrochim. Acta, 17 (1961) 857.
99.
Anderson, D.W.W., Rankin, D.W.H., Robertson, A., J. Mol. Struct., 14 (1972) 385.
100.
Koput, J., J. Mol. Spectrosc., 106 (1984) 12.
101.
Koput, J., J. Mol. Spectrosc., 115 (1986) 131.
102.
Duckett, J.A., Robiette, A.G., Gerry, M.C.L., J. Mol. Spectrosc., 90 (1981) 374.
103.
Sullivan, J.F., Heusel, H.L., Zunic, W.M., Durig, J.R., Cradock, S., Spectrochimica
Acta Part A: Molecular Spectroscopy, 50 (1994) 435.
104. Duncan, J.L., Harvie, J.L., Mckean, D.C., Cradock, S., J. Mol. Struct., 145 (1986) 225.
105.
Guirgis, G.A., Bell, S., Durig, J.R., Spectrochim. Acta A, 57 (2001) 1235.
106.
Herzberg, G., Molecular Spectra And Molecular Structure: Ii. Infrared, And Raman
Spectr Of Polyatomic Molecules, 1945, Nostrand, Princeton.
107.
Nyquist, R.A., Spectrochim. Acta, 21 (1965) 1245.
108.
Mclachlan, R.D., Spectrochim. Acta A, 23 (1967) 1793.
109.
Durig, J.R., Zhou, X., Durig, N.E., Nguyen, D., Durig, D.T., J. Mol. Struct., 917
(2009) 37.
110.
Winnewisser, M., Winnewisser, B.P., Medvedev, I.R., De Lucia, F.C., Ross, S.C.,
Bates, L.M., J. Mol. Struct., 798 (2006) 1.
111.
Zhou, S.X., Durig, J.R., J. Mol. Struct., 924-926 (2009) 111.
310
112.
Feher, M., Pasinszki, T., Veszprémi, T., J. Am. Chem. Soc., 115 (1993) 1500.
113.
Lee, C., Yang, W., Parr, R.G., Physical Review B, 37 (1988) 785.
114.
Becke, A.D., The Journal of Chemical Physics, 98 (1993) 5648.
115.
Sullivan, J.F., Durig, D.T., Durig, J.R., Cradock, S., J. Phys. Chem., 91 (1987) 1770.
116.
Sakaizumi, T., Ohashi, O., Yamaguchi, I., Bull. Chem. Soc. Jpn., 49 (1976) 948.
117.
Cradock, S., Durig, J.R., Sullivan, J.F., J. Mol. Struct., 131 (1985) 121.
118.
Sakaizumi, T., Yamada, O., Ushida, K., Ohashi, O., Yamaguchi, I., Bull. Chem. Soc.
Jpn., 49 (1976) 2908.
119.
Kopper, H.,Pongratz, A., Monatshefte für Chemie, 62 (1933) 78.
120.
Palmer, M.H.,Nelson, A.D., J. Mol. Struct., 782 (2006) 94.
121.
Palmer, M.H., J. Mol. Struct., 692 (2004) 43.
122.
Hargittai, I., Paul, I.C., Patai, S., The Chemistry Of Cyanates And Their Thio
Derivatives, ed. S. Patai. Vol. 1. 1977, Great Britain: John Wiley and Sons, Ltd.
123.
Nguyen, M.T., Hajnal, M.R., Vanquickenborne, L.G., Ha, T., Stohner, J., J. Chem.
Soc., Faraday Trans., 89 (1993) 2381.
124.
Heldmann, C.,Dreizler, H., Z. Naturforsch., 45 (1990) 1175.
125.
Durig, J.R., Kanes, K.J., Sullivan, J.F., J. Mol. Struct., 99 (1983) 61.
126.
Cradock, S., Sullivan, J.F., Durig, J.R., J. Mol. Struct., 140 (1986) 199.
127.
Herrebout, W.A., Nagels, N., Van Der Veken, B.J., ChemPhysChem, 10 (2009) 3054.
128.
Michielsen, B., Dom, J.J.J., Veken, B.J.V.D., Hesse, S., Xue, Z., Suhm, M.A.,
Herrebout, W.A., Phys. Chem. Chem. Phys., 12 (2010) 14034.
129.
Durig, J.R., Zhou, S.X., Zheng, C., Durig, D.T., J. Mol. Struct., 971 (2010) 23.
130.
Tobiason, F.L.,Schwendeman, R.H., J. Chem. Phys., 40 (1964) 1014.
131.
Pauling, L., J. Am. Chem. Soc., 53 (1931) 1367.
132.
Durig, J.R., Bell, S., Guirgis, G.A., Spectrochim. Acta A, 52 (1996) 1843.
311
133.
Bell, S., Guirgis, G.A., Li, Y., Durig, J.R., J. Phys. Chem. A, 101 (1997) 5987.
134. Durig, J.R., Guirgis, G.A., Bell, S., Brewer, W.E., J. Phys. Chem. A, 101 (1997) 9240.
135.
Bell, S., Guirgis, G.A., Won Hur, S., Durig, J.R., Spectrochim. Acta A, 55 (1999)
2361.
136.
Bell, S., Groner, P., Guirgis, G.A., Durig, J.R., J. Phys. Chem. A, 104 (2000) 514.
137.
Hirota, E., Matsumura, C., Morino, Y., Bull. Chem. Soc. Jpn., 40 (1967) 1124.
138.
Möller, K.D., De Meo, A.R., Smith, D.R., London, L.H., J. Chem. Phys., 47 (1967)
2609.
139.
Kistiakowsky, G.B., Lacher, J.R., Stitt, F., J. Chem. Phys., 7 (1939) 289.
140.
Schwendeman, R.H.,Jacobs, G.D., J. Chem. Phys., 36 (1962) 1245.
141.
Lide, J.D.R., J. Chem. Phys., 30 (1959) 37.
142.
Duckett, J.A., Robiette, A.G., Mills, I.M., J. Mol. Spectrosc., 62 (1976) 34.
143.
Gerry, M.C.L., Thompson, J.C., Sugden, T.M., Nature, 211 (1966) 646.
144.
Careless, A.J., Green, M.C., Kroto, H.W., Chem. Phys. Lett., 16 (1972) 414.
145.
Durig, J.R., Kalasinsky, K.S., Kalasinsky, V.F., J. Chem. Phys., 69 (1978) 918.
146.
Fernández-Liencres, M.P., Navarro, A., Ben Altabef, A., López-González, J.J.,
Fernández-Gómez, M., Kearley, G.J., Chem. Phys., 330 (2006) 26.
147.
Durig, J.S., Sullivan, J.F., Cox Jr, A.W., Streusand, B.J., Spectrochim. Acta, 34 A
(1977) 719.
148.
Veszprémi, T.J., Nagy, J., Barta, I.A., Zsombok, G., J. Organomet. Chem., 185
(1980) 323.
149.
Durig, J.R.,Jalilian, M., Appl. Spectrosc., 33 (1979) 364.
150.
Glidewell, C., Robiette, A.G., Sheldrick, G.M., Chem. Phys. Lett., 16 (1972) 526.
151.
Cradock, S., Huntley, C.M., Durig, J.R., J. Mol. Struct., 127 (1985) 319.
152.
Cradock, S., Huntley, C.M., Rankin, D.W.H., Robertson, H.E., J. Chem. Soc., Dalton
Trans. (1986) 859.
312
153.
Mckean, D.C.,Torto, I., J. Mol. Struct., 81 (1982) 51.
154.
Durig, J.R., Pan, C., Guirgis, G.A., Spectrochim. Acta A, 59 (2003) 979.
155.
Durig, J.R., Pan, C., Guirgis, G.A., J. Mol. Struct., 688 (2004) 95.
156.
Durig, J.R., Sullivan, J.F., Li, Y.S., Mohamad, A.B., J. Mol. Struct., 79 (1982) 235.
157.
Durig, J.R.,Mohamad, A.B., J. F. Sullivan, J. Mol. Spectrosc, 128 (1988) 68.
158.
Durig, J.R., Sullivan, J.F., Cox Jr, A.W., Turner, J.B., J. Mol. Struct., 44 (1978) 31.
159.
Durig, J.R., Jalilian, M.R., Sullivan, J.F., Turner, J.B., J. Raman Spectrosc., 11 (1981)
459.
160.
Durig, J.R., Berry, R.J., Wurrey, C.J., J. Am. Chem. Soc., 110 (1988) 718.
161.
Kricheldorf, H.R., Chem. Ber, 106 (1973) 3765.
162.
Kricheldorf, H.R., Regel, W., Chem. Ber., 106 (1973) 3753.
163. Galabov, B., Kim, S., Xie, Y., Schaefer Iii, H.F., Leininger, M.L., Durig, J.R., J. Phys.
Chem. A, 112 (2008) 2120.
164.
Schwendeman, R.H., Jacobs, G.D., Krigas, T.M., J. Chem. Phys., 40 (1964) 1022.
165.
Kirby, C., Kroto, H.W., J. Mol. Spectrosc., 70 (1978) 216.
166.
Bouchy, A., Roussy, G., J. Mol. Spectrosc., 68 (1977) 156.
167.
Penn, R.E., Boggs, J.E., J. Chem. Soc., Chem. Commun., 0 (1972) 666.
168.
Durig, J.R., Sullivan, J.F., Berry, R.J., Cradock, S., J. Chem. Phys., 86 (1987) 4313.
169.
Groner, P., Church, J.S., Li, Y.S., Durig, J.R., J. Chem. Phys., 82 (1985) 3894.
170.
Lord, R.C.,Nakagawa, I., J. Chem. Phys., 39 (1963) 2951.
171.
Durig, J.R.,Green, W.H., J. Chem. Phys., 47 (1967) 673.
172.
Durig, J.R.,Morrissey, A.C., J. Chem. Phys., 46 (1967) 4854.
173.
Durig, J.R., Willis, J.J.N., Green, W.H., J. Chem. Phys., 54 (1971) 1547.
174.
Rothschild, W.G.,Dailey, B.P., J. Chem. Phys., 36 (1962) 2931.
313
175.
Hyunyong, K.I.M.,Gwinn, W.D., J. Chem. Phys., 44 (1966) 865.
176.
Curl Jr, R.F., J. Comput. Phys., 6 (1970) 367.
177.
Velino, B., Favero, L.B., Caminati, W., J. Mol. Spectrosc., 179 (1996) 168.
178.
Caminati, W., Favero, L.B., Maris, A., Favero, P.G., J. Mol. Struct., 376 (1996) 25.
179.
Lin, W., Ganguly, A., Minei, A.J., Lindeke, G.L., Pringle, W.C., Novick, S.E., Durig,
J.R., J. Mol. Struct., 922 (2009) 83.
180.
Ganguly, A., Klaassen, J.J., Guirgis, G.A., Gounev, T.K., Durig, J.R., Spectrochim.
Acta A, 79 (2011) 831.
181.
Durig, J.R., Klaassen, J.J., Ganguly, A., Gounev, T.K., Guirgis, G.A., Lin, W., Struct.
Chem., 19 (2008) 935.
182.
Durig, J.R., Klaassen, J.J., Ganguly, A., Gounev, T., Groner, P., J. Mol. Struct., 934
(2009) 66.
183.
Durig, J.R., Zhou, S.X., Guirgis, G.A., Wurrey, C.J., J Phys Chem A, 115 (2011)
2297.
184.
Woldbaek, T., Berkessel, A., Horn, A., Klaeboe, P., Acta Chem. Scand. A, 36 (1982)
719.
185.
Christian, S.D., Grundnes, J., Klaboe, P., J. Am. Chem. Soc., 97 (1975) 3864.
186.
Hassel, O.,Ottar, B., Acta Chem. Scand., 1 (1947) 929.
187.
Hassel, O.,Viervoll, H., Acta Chem. Scand., 1 (1947) 149.
188.
Durig, J.R., Zheng, C., El Defrawy, A.M., Ward, R.M., Gounev, T.K., Ravindranath,
K., Rao, N.R., J. Raman Spectrosc., 40 (2009) 197.
189.
Durig, J.R., El Defrawy, A.M., Ward, R.M., Guirgis, G.A., Gounev, T.K., Struct.
Chem., 19 (2008) 579.
190.
Durig, J.R., El Defrawy, A.M., Ward, R.M., Guirgis, G.A., Gounev, T.K., J. Mol.
Struct., 918 (2009) 26.
191. Durig, J.R., Ward, R.M., Nelson, K.G., Gounev, T.K., J. Mol. Struct., 976 (2010) 150.
192.
Beckett, C.W., Pitzer, K.S., Spitzer, R., J. Am. Chem. Soc., 69 (1947) 2488.
314
193.
Durig, J.R., Ward, R.M., Conrad, A.R., Tubergen, M.J., Nelson, K.G., Groner, P.,
Gounev, T.K., J. Mol. Struct., 967 (2010) 99.
194.
Durig, J.R., Ward, R.M., Conrad, A.R., Tubergen, M.J., Nelson, K.G., Gounev, T.K.,
J. Mol. Struct., 975 (2010) 5.
195.
Jensen, F.R., Bushweller, C.H., Beck, B.H., J. Am. Chem. Soc., 91 (1969) 344.
196.
Horntvedt, H.T.,Klaeboe, P., Acta Chem. Scand., 29 (1975) 528.
197.
Kitchin, R.W., Studies In Microwave Spectroscopy: Conformational Analysis Of 1,4Thioxane, Thiane, 2-Cyclohepten-1-One, Cyclohexyl Cyanide, Cyclohexyl Isocyanate
And Cyclohexyl Isothiocyanat; Qualitative Analysis Of Low Molecular Weigh
Alcohols in Department of Chemistry, 1976, Mississippi State Univeristy.
198.
Pierce, L.,Nelson, R., J. Am. Chem. Soc., 88 (1966) 216.
199.
Pierce, L.,Beecher, J.F., J. Am. Chem. Soc., 88 (1966) 5406.
200.
Woldbaek, T.,Klaeboe, P., J. Mol. Struct., 63 (1980) 195.
201.
Klaeboe, P., Nielsen, C.J., Woldbaek, T., J. Mol. Struct., 60 (1980) 121.
202.
Woldbaek, T., Nielsen, C.J., Klaeboe, P., J. Mol. Struct., 66 (1980) 31.
203.
Gustavsen, J.E., Klaeboe, P., Kvila, H., Acta Chem. Scand., 32 (1978) 25.
204.
Horntvedt, H.T.,Klaeboe, P., Acta Chem. Scand., 25 (1971) 772.
205. Durig, J.R., Ward, R.M., Conrad, A.R., Tubergen, M.J., Guirgis, G.A., J. Phys. Chem.
A, 114 (2010) 9289.
206.
Durig, J.R., Ward, R.M., Conrad, A.R., Tubergen, M.J., Giurgis, G.A., Gounev, T.K.,
J. Mol. Struct., 922 (2009) 19.
207.
Durig, J.R., Ward, R.M., Guirgis, G.A., Gounev, T.K., J. Raman Spectrosc., 40
(2009) 1919.
315
VITA
Xiaohua Zhou was born in Guangzhou, China on June 1, 1987. She spent a large part
of her childhood in Honolulu, Hawaii where she attended Prince David Kawananakoa
Middle School and President William McKinley High School. She relocated to Kansas City,
Missouri during her sophomore year in high school. In 2005, she graduated from Lincoln
College Preparatory Academy. With an all expense paid scholarship awarded by the Gates
Millennium Scholarship and Asian Pacific Islander American Scholarship Fund, she
completed her Bachelor degree in Chemistry at the Univeristy of Missouri-Kansas City in
2008.
During her undergraduate study, she carried out extensive research in the fields of
molecular spectroscopy and conformational analysis under the supervision of Dr. James R.
Durig. With strong encouragement from Dr. Durig, she was convinced to pursue a Ph. D. in
Physical Chemistry under his mentorship. In 2009, she officially started her graduate career
in Dr. Durig’s Spectroscopy Laboratory, which was again fully supported by Gates
Millennium Scholarship and Asian Pacific Islander American Scholarship Fund. As a
graduate student, Xiaohua has received several awards which included the William G.
Fateley Student Award, Coblentz Student Award, 1st Place FACSS Student Poster Award,
Outstanding Merit Recipient of Women’s Council Graduate Assitance Funding, Mr. & Mrs.
Fung Wu Cheng Scholarship, and Dr. and Mrs. Y.-C Jerry Jean New Graduate Scholarship.
Xiaohua has co-authored numerous papers, which are listed below. Under the
sagacious recommendation from Dr. Durig, it was necessary for Xiaohua to adopt an English
name, Sarah. As a result, she has published papers under the following names: Xiaohua
Zhou, Xiaohua Sarah Zhou and Sarah Xiaohua Zhou.
316
1.
J.R. Durig, X. Zhou, A. M. El-Defrawy, G. A. Guirgis, T. K. Gounev, C Zheng,
“Conformational Stability, Structural Parameters, Vibrational Spectra, and ab initio
Calculations of Isopropyl- and Tertiary-butylisothiocyanate,” J. Mol. Struct., 839, 107
(2007).
2.
J. R. Durig, S. Panikar, X. Zhou, A. El-Defrawy, “Vibrational Spectrum,
Conformational Stability, Structural Parameters and ab initio Calculations of
Dimethylaminodifluorophosphine,” Spectrochim. Acta, 69A, 715 (2008).
3.
J. R. Durig, P. Tschudin, K. Cohn, B. R. Durig, X. Zhou, C. Zheng, A. El Defrawy,
“Infrared and Raman Spectra, Conformational Stability, Structural Parameters, Ab
Initio Calculations and Vibrational Assignments of Amino-difluorophosphine,” J.
Mol. Struct., 875, 406 (2008).
4.
G. A. Guirgis, Z. Yu, C. Zheng, X. Zhou, J. R. Durig, “Conformational Stability from
Variable Temperature Ft-IR Spectra of Krypton Solutions, r0 Structural Parameters,
Vibrational Assignment and ab initio Calculations of 4-Fluoro-1-Butene,” J. Phys.
Chem. A, 112, 2268 (2008).
5.
Y. E. Nashed, M. A. Qtaitat, C. Zheng, S. X. Zhou, G. A. Guirgis, J. F. Sullivan, J. R.
Durig, “Conformational Stability from Rare Gas Solutions, r0 Structural Parameters,
Barriers to Internal Rotation, and ab initio Calculations for Vinyl Silyl Fluoride,” J.
Phys. Chem., 113, 1653 (2009).
6.
J. R. Durig, S. X. Zhou, N. E. Durig, D. Nguyen, D. T. Durig, “Vibrational Spectra
and Structural Parameters of Some XNCO and XOCN (X=H, F, Cl, Br) Molecules,” J.
Mol. Struct., 881, 102 (2008).
7
J. R. Durig, S. X. Zhou, A. X. Garner, N. E. Durig, “Structural Parameters,
Centrifugal Distortion Constants, and Vibrational Spectra of F2C=NX (X = H, F, Cl,
Br) Molecules,” J. Mol. Struct. 922, 11 (2009).
8.
S. X. Zhou, J. R. Durig, "The r0 Structural Parameters, Vibrational Spectra, ab initio
Calculations and Barriers to Internal Rotation and Linearity of Methylisocyanate," J.
Mol. Struct., 924, 111 (2009).
9.
J. R. Durig, S. X. Zhou, J. Klassen, A. Ganguly, “The Utilization of Low Frequency
Raman Spectra of Gases for the Study of Molecules with Large Amplitude Vibration,”
Journal of Light Scattering, 21, 201 (2009).
10.
G. A. Guirgis, S. X. Zhou, J. R. Durig, “Raman and Infrared Spectra, Conformational
Stability,
Ab
initio
Calculations
and
Vibrational
Assignment
of
Dimethylsilylisocyanate,” J. Raman Spectrosc., 41, 303 (2009).
317
11.
J. R. Durig, S. X. Zhou, C. X. Zhou, N. E. Durig, “Structural Parameters, Vibrational
Spectra and Centrifugal Distortion Constants of F(CN)C=NX (X=H, F, Cl, Br) and
CH3(Y)C=NH (Y= H, CN)” J. Mol. Struct., 967, 1 (2009).
12.
J. R. Durig, S. X. Zhou, C. Zheng, D. T. Durig, "Conformational Stability, Structural
Parameters and Vibrational Assignment from Variable Temperature Infrared Spectra
of Xenon and Krypton Solutions and Ab Initio Calculations of Ethylisocyanate" J.
Mol. Struct., 971, 33 (2010).
13.
J. R. Durig, S. X. Zhou, G. A. Guirgis, C. J. Wurrey, "Conformational Stability from
Variable Temperature Infrared Spectra of Xenon Solutions, r0 Structural Parameters,
and Ab Initio Calculations of Cyclopropylisocyanate" J. Phys. Chem. 115, 2297
(2011).
14.
S. X. Zhou, G. A. Guirgis, K. K. Gause, A. R. Conrad, M. J. Tubergen, J. R. Durig,
“Microwave, Raman, and Infrared spectra, r0 Structural Parameters, Conformational
Stability and ab initio Calculations of Cyclobutylisocyanate” Accepted by Structural
Chemistry (May 2012)
15.
S. X. Zhou, R. Ward, M. J. Tubergen, R. M. Gurusinghe, J. R. Durig, “Microwave,
Infrared, and Raman spectra, r0 Structural Parameters, Conformational Stability and
ab initio Calculations of Cyclohexylisocyanate” Accepted by to Chemical Physics
(July 2012)
16.
J. R. Durig, S. X. Zhou, A. R. Conrad, M. J. Tubergen, W. Herrebout, J. J. J. Dom, B.
J. van der Veken, T. K. Gounev, “Raman, Infrared, and Microwave Spectra, r0
Structural Parameters, Conformational Stability and ab initio Calculations of
Isopropylisocyanate” Submitted
318
Документ
Категория
Без категории
Просмотров
0
Размер файла
10 909 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа