Midinfrared and quantum-chemical study of the structure, conformation, and isomerization of the unstable CH3CH2OCN molecule

Article abstract of DOI:10.1021/jp0211990

Gaseous ethyl cyanate, CH₃CH₂OCN, has been generated from the gas/solid reaction of O-ethyl thiocarbamate with mercury oxide and characterized in the gas phase by infrared spectroscopy for the first time. Experimental data indicate the presence of two conformers in the gas phase, the gauche (synclinal) and the trans (antiperiplanar) form. The molecular geometries and energetics of the possible conformers are obtained from DFT calculations at the B3LYP level and from ab initio calculations at the MP2, MP3, MP4, QCISD, and CCSD(T) levels of theory. The assignment of the gas-phase infrared spectrum is assisted by normal coordinate calculations based on the scaled computed force field of the two conformers. The kinetic instability of CH₃-CH₂OCN toward isomerization is studied at the B3LYP level, in a vacuum and in solutions. Solvent effects are modeled using the polarized continuum model (PCM). Calculations show that the isomerization is not a unimolecular process at ambient temperatures, and bimolecular processes are responsible for the instability. In polar solvents, the OCN^- anion plays a key role in the isomerization, being an effective catalyst for the cyanate-isocyanate rearrangement.

Full text of DOI:10.1021/jp0211990

1720
J. Phys. Chem. A 2003, 107, 1720-1726
Midinfrared and Quantum-Chemical Study of the Structure, Conformation, and
Isomerization of the Unstable CH3CH2OCN Molecule
,†
†
‡
Tibor Pasinszki,* Bal a´ zs Havasi, and Attila Kov a´ cs
Department of Inorganic Chemistry, Budapest UniVersity of Technology and Economics, H-1521 Budapest,
Gell e´ rt t e´ r 4, Hungary, and Research Group for Technical Analytical Chemistry of the Hungarian Academy of
Sciences at the Institute of General and Analytical Chemistry, Budapest UniVersity of Technology and
Economics, H-1521 Budapest, Gell e´ rt t e´ r 4, Hungary
ReceiVed: May 14, 2002; In Final Form: December 2, 2002
3 2
Gaseous ethyl cyanate, CH CH OCN, has been generated from the gas/solid reaction of O-ethyl thiocarbamate
with mercury oxide and characterized in the gas phase by infrared spectroscopy for the first time. Experimental
data indicate the presence of two conformers in the gas phase, the gauche (synclinal) and the trans
(antiperiplanar) form. The molecular geometries and energetics of the possible conformers are obtained from
DFT calculations at the B3LYP level and from ab initio calculations at the MP2, MP3, MP4, QCISD, and
CCSD(T) levels of theory. The assignment of the gas-phase infrared spectrum is assisted by normal coordinate
calculations based on the scaled computed force field of the two conformers. The kinetic instability of CH
3
-
CH OCN toward isomerization is studied at the B3LYP level, in a vacuum and in solutions. Solvent effects
2
are modeled using the polarized continuum model (PCM). Calculations show that the isomerization is not a
unimolecular process at ambient temperatures, and bimolecular processes are responsible for the instability.
-
In polar solvents, the OCN anion plays a key role in the isomerization, being an effective catalyst for the
cyanate-isocyanate rearrangement.
Introduction
tion on the conformers of CH3CH2OCN. The four possible
conformers for this molecule can be derived by a rotation around
the C(H2)-O single bond: cis (synperiplanar), gauche (syncli-
nal), skew (anticlinal), and trans (antiperiplanar) forms with
CCOC dihedral angles of 0°, ∼60°, ∼120°, and 180°, respec-
tively. The structures of the cis, gauche, and trans conformer
were calculated recently at the HF and MP2 levels, neglecting,
Covalent cyanates (RsOsCtN) are in general unstable
compounds and isomerize into their thermodynamically more
stable isocyanate derivatives. The cyanate structure is stable only
1
,2
in special cases, e.g., if the substituent, R, is an aryl group,
or the substituent is bulky, preventing the isomerization ster-
icly.3,4 Ethyl cyanate, CH3CH2OCN, is no exception. It isomer-
1
1
5
however, the skew conformer. These calculations predicted
the gauche and trans forms to be minima on the potential energy
surface, whereas the cis form turned out to be a transition state.
The calculated energy difference between the gauche and trans
conformers was small, favoring the gauche at the MP2 and the
trans at the HF and MP4//MP2 (single-point MP4 calculation
izes readily at room or higher temperatures, and the process is
believed to be autocatalyzed by the isomer.1 Because of
instability, its application in chemical reactions requires an in
situ generation or prior to use, and it must be stored at low
temperatures.6
Because ethyl cyanate is unstable, little is known experimen-
1
1
tally about its structure and spectroscopy. The IR7,8 and NMR
8
at the MP2 geometry) levels.
spectra were, however, recorded in cold CCl4 solution, and the
Alkyl cyanates are known to isomerize at ambient temper-
9
1,12
only gas-phase investigations are the early mass and recent
atures,
although the mechanism of this process is not yet
1
0
microwave spectroscopic study. In the microwave work, by
comparing the experimental rotational constants to those of
model geometries, the observed spectrum was assigned to the
trans (antiperiplanar) conformer. The search for another con-
former among the weak spectral lines was unsuccessful;
therefore, the trans conformer was concluded to be the most
stable. No geometrical parameters could be derived from the
experimental data, because of insufficient number of isotope
substitutions.
understood. The only work addressing this question so far is
1
that of Martin et al., where the kinetics of the isomerization of
ethyl cyanate was investigated in various solvents and temper-
atures. It has been observed that the speed of isomerization is
influenced by solvents and depends on the concentration. In
nonpolar aprotic solvents (diethyl ether, CH3Cl, CCl4, benzene,
and toluene), CH3CH2OCN is fairly stable and can be stored in
a refrigerator for weeks. The half-life of CH3CH2OCN in toluene
is 654 and 232 min at 79.5 and 91 °C, respectively. During the
rearrangement, the formation of both ethyl isocyanate and
triethylisocyanurate (ethyl isocyanate trimer) is observed. In
polar solvents (nitrobenzene, acetonitrile), the isomerization is
ca. 20-times faster than in toluene and the formation of
ethylisocyanurate is not observed at 73.5 °C over a time of 140
min. In dimethylformamide (DMF), CH3CH2OCN isomerizes
rapidly at room temperature (half-lives at 0 and 9 °C are 441
and 75 min, respectively), and the speed of isomerization is ca.
Because of the lack of detailed experimental work, quantum
chemical calculations become important in providing informa-
*
To whom correspondence should be addressed. E-mail: pasinszki@
mail.bme.hu.
†
Department of Inorganic Chemistry, Budapest University of Technology
and Economics.
‡
Research Group for Technical Analytical Chemistry of the Hungarian
Academy of Sciences at the Institute of General and Analytical Chemistry,
Budapest University of Technology and Economics.
1
0.1021/jp0211990 CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/25/2003

Study of the Unstable CH3CH2OCN Molecule
J. Phys. Chem. A, Vol. 107, No. 11, 2003 1721
SCHEME 1
(2d,2p), MP3(full)/6-311+G(2d,2p), QCISD(full)/6-311+G-
(
d,p), and MP4(full)/6-31G(d,p) levels. These latter geometry
optimizations were extended by harmonic frequency calculations
at the HF, MP2, and QCISD levels. To have a better estimate
of the relative energies of the conformers, single-point calcula-
tions were performed on the optimized B3LYP, MP2, MP4,
and QCISD structures at the MP4(full)/6-311+G(2d,2p) and
CCSD(T)(full)/6-311+G(2d,2p) levels, denoted as MP4//MP4,
CCSD(T)//B3LYP, CCSD(T)//MP2, CCSD(T)//MP4, and CCSD-
3
-4 order of magnitude higher than in nitrobenzene or
acetonitrile. The isomerization in dimethyl sulfoxide (DMSO)
and DMF is fast at room temperature, and the major isolated
product from the reaction mixture is triethylisocyanurate. One
of the key experiments on this subject was the investigation of
the isomerization of an equimolar mixture of C2H5OC N and
n-C4H9OCN in nitrobenzene, which indicated the equimolar
formation of C2H5 NCO and n-C4H9 NCO. On the basis of
this and experiments discussed above, it have been suggested
that the isomerization proceeds via a solvent separated ion pair
(T)//QCISD. For testing the reliability of theoretical levels used,
1
5
calculations were done for the geometry of CH OCN, at the
3
same levels used for CH3CH2OCN, and results were compared
to the available experimental microwave structure.
15
15
1
Unimolecular isomerization and intermolecular reactions of
CH3CH2OCN were modeled in a “vacuum” and in the presence
of a solvent at the B3LYP/6-311+G(2d,2p) level. Five solvents
were selected from the available solvent list in Gaussian 98:
heptane (dielectric constant: ꢀ ) 1.92), dichloroethane (ꢀ )
10.36), acetone (ꢀ ) 20.7), nitromethane (ꢀ ) 38.2), and DMSO
(ꢀ ) 46.7). Solvent effects were calculated using the popular
PCM as incorporated in Gaussian 98. All calculations were
carried out with the Gaussian 98 quantum chemistry package14
implemented on a Silicon Graphics Inc. Origin 200 workstation.
Computational Details of the SQM Analysis. The scaled
+
-
[
C2H5 |OCN ]. It has also been noted that the formation of
+ -
free solvated C2H5 and OCN ions is not expected because of
the weak ionizing power of nitrobenzene.1
The aim of the present study is to generate and characterize
CH3CH2OCN in the gas phase, to derive information about the
structure and energetics of its conformers, and to obtain
information about its stability and isomerization. To this end,
CH3CH2OCN is studied in the gas phase by mid-infrared (IR)
spectroscopy and the experimental work is complemented by
extensive quantum-chemical calculations.
1
5
quantum mechanical (SQM) analysis of ethyl cyanate was
based on the computed harmonic B3LYP/6-311+G(2d,2p) force
field in a nonredundant internal coordinate representation.16 The
Cartesian force field was transformed to internal coordinates
Experimental Section
A mixture of gaseous ethyl cyanate and water was generated
by passing O-ethyl thiocarbamate13 over solid yellow mercury-
1
7
using the program TRA3. For the scaling scheme, Pulay’s
1
8
standard (selective) scaling method was used. The solution
of the secular equation and optimization of the scale factors
(II) oxide (FERAK Laborat GMBH) packed into a small Pyrex
1
0,12
tube (Scheme 1).
The CH3CH2OCN/H2O mixture from the
1
9,20
were done with the program SCALE3.
For characterization
reaction zone was led directly into the IR spectrometer or
condensed in a U-trap inserted between the reaction tube and
the spectrometer.
of the fundamentals, their total energy distribution (TED)2
was used which provides a measure of each internal coordinate’s
contribution to the normal coordinate in terms of energy.
1,22
-
1
-1
Mid-IR spectra (4000-500 cm , 1 cm resolution) were
collected on a Perkin-Elmer System 2000 Fourier transform
spectrometer equipped with a mercury cadmium telluride (MCT)
detector. Measurements were performed at room temperature
using a 240 cm multipass gas cell with KCl windows (PikeTech).
The effluent from the reaction tube or the sample from the
U-trap was continuously pumped through the cell using a rotary
pump while maintaining the pressure between 550 and 600
mTorr. The IR spectra of gaseous O-ethyl thiocarbamate and
CH3CH2NCO (Aldrich) were recorded on a similar manner.
Results and Discussion
Theoretical Calculations. The electronic and geometric
structures of pseudohalides are known to be sensitive to electron
correlation effects, often presenting a serious challenge for
quantum chemical methods. The effect of electron correlation
on the calculated structure depends on the kind of molecule
investigated, hence the methods used for a systematic study of
a certain type of pseudohalide must be tested prior to use. The
simplest method is to compare calculated quantities to experi-
mental data, keeping in mind that the calculated equilibrium
structure (re) is usually compared to a vibrationally averaged
experimental structure (r0, rs, ra, rg, etc.). Experimental informa-
tion about the geometry of covalent cyanates is very scarce;
Computational Methods
Theoretical calculations for the structure of CH3CH2OCN
were carried out first using the density functional theory, DFT,
in the form of B3LYP (Becke’s three-parameter exchange
functional in combination with Lee, Yang, and Parr correlation
functional). The conformers of CH3CH2OCN and the intercon-
necting transition states (TS) with respect to the C(ethyl)-O
single bond were investigated at the B3LYP/6-311+G(2d,2p)
level. Equilibrium molecular geometries were fully optimized,
and the existence of true minima was confirmed by harmonic
frequency analysis (zero imaginary frequency). TSs were
characterized by one imaginary frequency. The potential energy
curve of the torsional motion was obtained by proceeding along
the given reaction coordinate and simultaneously relaxing all
other bond lengths and angles. Geometries of the minimum
energy structures (gauche and trans forms) and transition states
23
24
CH3OCN and F5SOCN are the only derivatives with known
gas-phase structures. We note that, although an ra structure was
2
5
published for F5SeOCN, it is still ambiguous whether the
cyanate, and not the isocyanate, is synthesized and studied. The
performance of various computational levels for selected
covalent cyanates is compared in Table 1, which shows, that
the calculated bond lengths are sensitive to the theoretical level
used, especially for the CtN triple bond. The HF method
underestimates the CtN bond length, MP2 and MP4 over-
estimate it, and a clear oscillation can be seen within the MP
expansion series. The QCISD method also seems to overestimate
the CtN bond length, although to a lesser extent than MP2 or
MP4. The B3LYP geometry is in the best agreement with
experimental MW data. When the calculated structure of CH3-
CH2OCN is compared to that of CH3OCN, the same trends are
(
3
cis and skew forms) were also optimized at the MP2(full)/6-
11+G(2d,2p) level, and structures of the gauche and trans
conformers were additionally calculated at the HF/6-311+G-

1722 J. Phys. Chem. A, Vol. 107, No. 11, 2003
Pasinszki et al.
TABLE 1: Comparison of the Structural Characteristics of
Selected Covalent Cyanates
TABLE 2: Calculated Equilibrium Structures, Total
Energies, Dipole Moments, and Rotational Constants of
Gauche and trans-CH CH
3
2
OCNa
C-O
(Å)
O-C
(Å)
CtN
(Å)
OCN
(deg)
COC
(deg)
molecule/method
trans expt.
b
gauche re
trans re
(r0)
CH
3
OCN (MW r
s
)a
1.455
1.302
1.271
1.146
1.162
178.4
175.3
113.8
b
C1-C2
C2-O
1.511 (1.516)
1.477 (1.461)
1.286 (1.294)
1.156 (1.165)
1.086 (1.091)
1.089 (1.095)
1.091 (1.095)
1.089 (1.094)
1.089 (1.094)
111.7 (111.4)
115.6 (113.9)
178.2 (178.7)
102.8 (103.3)
107.4 (108.0)
109.1 (109.2)
111.0 (110.5)
111.2 (111.1)
180.0 (180.0)
69.5 (66.8)
1.507 (1.513)
1.476 (1.460)
1.287 (1.294)
1.156 (1.164)
1.089 (1.094)
1.089 (1.094)
1.090 (1.095)
1.089 (1.094)
1.089 (1.094)
107.3 (107.0)
115.6 (113.6)
178.2 (178.7)
107.4 (108.0)
107.4 (108.0)
109.1 (109.2)
111.0 (110.6)
111.0 (110.6)
180.0
F
5
SOCN (ED, r )
a
c
3 e
CH OCN (r )
O-C3
B3LYP
HF
MP2
MP3
MP4
1.459
1.436
1.454
1.445
1.457
1.449
1.288
1.272
1.294
1.291
1.307
1.295
1.156
1.129
1.173
1.151
1.185
1.164
178.0
179.4
178.2
178.7
178.3
178.7
115.3
115.7
113.3
113.7
113.4
113.5
C3-N
C2-H1
C2-H2
C1-H3
C1-H4
C1-H5
C1C2O
C2OC3
OC3N
QCISD
CH
3
CH
2
OCN (r
e
)c,d
B3LYP
HF
MP2
MP3
MP4
1.476
1.448
1.466
1.457
1.469
1.460
1.287
1.272
1.293
1.291
1.306
1.294
1.156
1.130
1.174
1.151
1.186
1.164
178.2
179.4
178.3
178.7
178.6
178.7
115.6
115.9
113.4
113.8
113.6
113.6
H1C2O
H2C2O
H3C1C2
H4C1C2
H5C1C2
C2OC3N
C1C2OC3
QCISD
a
b
Ref 23a (from microwave spectroscopy, MW). Ref 24 (from
c
d
electron diffraction, ED). This work. Trans conformer.
180.0
total energy: -247.345617
-247.345934
(-246.781654)
5.13 (5.17)
(
-246.781627)
µc
4.92 (4.60)
4.72
d
A
B
C
12.4348 (11.9157)
3.3949 (3.5213)
2.9007 (2.9593)
31.0545 (30.5102) 30.055
2.5172 (2.5381)
2.3986 (2.4150)
2.54353
2.41967
a
Calculated at the B3LYP/6-311+G(2d,2p) and QCISD(full)/6-
11+G(d,p) (in parentheses) levels. Bond angles are in degrees; bond
3
b
lengths are in Å; total energies are in a.u.; see Figure 1. From
c
d
microwave spectrum (ref 10). Dipole moments in Debye. Rotational
constants in GHz. Isotopes: 12C, H, N, and O.
1
14
16
MP4. There is an oscillation in the HF ) MP1 f MP2 f MP3
f MP4 series regarding the gauche-trans relative energy,
similar to the oscillation in bond lengths. On the basis of the
calculated results summarized in Table 3, especially on that of
the highest CCSD(T)//B3LYP level, we conclude that the major
conformer at room temperature is the trans one. However, the
energy difference between the two minima is small, and because
the interconnecting skew transition state is also rather low in
energy, there must be a thermal equilibrium between the gauche
and trans conformers in the gas phase. Calculations at various
levels in Table 3 predict a gauche/trans ratio of 0.54-1.09/1,
so both conformers may be observable in a sample at room
temperature. The major component trans form has been observed
Figure 1. Structures, numbering of the atoms, the deformation
potential, and relative energies of the conformers of CH CH OCN.
3 2
Ground vibrational levels are estimated from harmonic frequency
calculations.
observed. We note a similar performance of the computational
1
0
methods in the case of nitrile oxides, RsCtN f O, one of the
in the previous microwave work, supported by the present
calculated rotational constants in Table 2. We note that, because
B3LYP provides good equilibrium geometries for pseudohalides
and CCSD(T) provides the best estimate of the total energy,
we believe that the best estimate of the conformer ratio is given
by the CCSD(T)//B3LYP method (0.76/1). Our IR spectroscopic
investigation (see below) suggests the presence of both con-
formers in a room-temperature sample. Figure 1 shows the
relative energies of the various conformers of CH3CH2OCN and
the C-O torsional potential calculated at the B3LYP level. In
this figure, the ν′′ ) 0 level is marked, on the basis of the
calculated (torsional) zero-point vibrational energy. The calcu-
2
6-28
isomers of cyanates.
recently
We have observed and pointed out
2
7-29
that B3LYP is capable of describing the equilib-
rium structure of the nitrile oxides, and DFT methods provide
an efficient alternative for calculating the structure if high level
ab initio methods are computationally not feasible. On the basis
of the analogy between CH3OCN and CH3CH2OCN, we can
assume the same behavior of the tested computational methods
for describing the molecular geometry and electronic structure
of CH3CH2OCN.
The calculated structure of CH3CH2OCN is shown in Figure
1
with calculated bond lengths and angles listed in Table 2.
-
1
According to our calculations, CH3CH2OCN has two stable
conformers, the gauche and the trans forms, whereas the cis
and skew are transition states on the potential energy surface.
Table 3 lists the relative energies of conformers. The trans form
is thermodynamically favored at the HF (not shown in Table
lated harmonic torsional frequency is 61 and 77 cm for the
trans and gauche conformer, respectively, so it is expected that
a hindered torsional motion exists around the gauche and trans
minima in the ground and lowest lying excited states (i.e., ν′′
) 1-3) of this vibration, which may become free only at higher
levels of excitation.
3
), MP3, B3LYP, QCISD, and CCSD(T) levels, and only MP4
and MP2 calculations are in disagreement. However, using a
larger basis set in the MP4 calculation and taking into account
ZPE and thermal corrections (∆G), the trans is also favored at
Identification and Infrared Spectrum of CH3CH2OCN.
An important aspect of this work is to show that pure ethyl
cyanate has been generated for IR investigation and that it is

Study of the Unstable CH3CH2OCN Molecule
J. Phys. Chem. A, Vol. 107, No. 11, 2003 1723
a
b
TABLE 3: Calculated Relative Energies (∆E) and Gibbs Free Energies (∆G) (kJ/mol) of CH
3
CH
2
OCN Conformers
∆
G (∆E)
level of theory
B3LYP
cis TS
gauche
skew TS
trans
gauche/trans ratioc
11.70 (8.67)
1.52 (1.27)
1.41 (0.61)
0.07 (0.00)
0.86 (0.56)
0.00 (0.00)
0.20 (0.00)
0.47 (0.18)
0.50 (0.21)
0.69 (0.44)
1.18 (0.38)
7.17 (3.87)
g
7.95 (5.13)
g
g
g
g
7.82 (4.78)
8.04 (4.74)
g
0.00 (0.00)
0.00 (0.00)
0.00 (0.22)
0.00 (0.00)
0.20 (0.50)
0.00 (0.09)
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
0.54/1
0.57/1
0.97/1
0.71/1
QCISD
MP2
g
10.07 (7.44)
d
MP3
g
g
g
g
d
MP4
1/0.92
d
MP4//MP4
0.92/1
0.83/1
0.82/1
0.76/1
0.62/1
d
d
CCSD(T)//MP4
CCSD(T)//MP2
10.85 (7.99)
11.19 (8.16)
g
e
CCSD(T)//B3LYPf
CCSD(T)//QCISD
a
Zero-point vibrational energy (ZPE) is added to the total energy. b Calculated at temperature of 298.15 K and pressure of 1.0 Atm. Calculated
c
d
e
using ∆G values and the Boltzman equation. ZPE and thermal corrections are calculated at the MP2/6-311+G(2d,2p) level. ZPE and thermal
f
g
corrections are calculated at the B3LYP/6-311+G(2d,2p) level. ZPE and thermal corrections are calculated at the QCISD/6-311+G** level. Not
calculated.
3
1G* level,30,31 but we are not aware of any transferable scale
factors for our B3LYP/6-311+G(2d,2p) level. Because the scale
factors are known to be well transferable between very closely
related derivatives, we have developed the necessary scale
factors on the basis of the known gas-phase IR data of the
1
2
analogous CH3OCN. The optimized scale factors for CH3
stretch, OC stretch, CtN stretch, Cet-O stretch, HCH bend,
and CCH bend are given in the footnote to Table 4. In the lack
of experimental information, no scale factors could be developed
for the low-frequency bending and torsional internal coordinates
and for the C-C stretch; thus, we kept these scale factors at
the preliminary value of 1.000. We note that the optimized scale
factors are close to unity, which reflects the high quality of the
unscaled B3LYP/6-311+G(2d,2p) force field.
The gas-phase IR spectrum of CH3CH2OCN (Figure 2)
concurs in general with that previously obtained in CCl4
8
-1
solution and shows two strong features at 2255 and 1106 cm
3 2
Figure 2. Infrared spectrum of gaseous CH CH OCN. Inset shows
the expanded 1250-750 cm region together with calculated band
positions and relative IR intensities (scaled frequencies are from Table
corresponding to the CtN and O-C(N) stretches, which are
the most characteristic fingerprints of the cyanate group (note
the alternative description of OCN asymmetric and symmetric
stretches, respectively). These bands show B/A- and A/B-type
-
1
4
).
not contaminated with the isocyanate isomer, O-ethyl thiocar-
bamate precursor, or any other side product. To this end, an
authentic sample of the stable CH3CH2NCO isomer and the
precursor O-ethyl thiocarbamate were also investigated by IR
spectroscopy under the same conditions. We note that the
reaction could be monitored visually following the formation
of black HgS from yellow HgO and the reaction was stopped
before the black zone of HgS reached the end of the reaction
tube. Supporting the 100% conversion in the above reaction,
no evidence was found for residual starting material in the IR
spectrum. In a separate experiment, the effluent from the reaction
tube was condensed at -78 °C in a U-trap, and this was then
gradually warmed allowing the detection of reaction products
in the order of their relative volatility. The most volatile side
product, the isocyanate isomer, was observed only in trace
amounts and was pumped off before recording the spectrum of
CH3CH2OCN. CH3CH2OCN had sufficient vapor pressure at 0
-
1
band structure with ca. 14 and 15 cm
PR separation,
respectively, in agreement with the calculated values.32 The
spectrum shows other peaks with medium to small intensity
-
1
corresponding to CH stretches (around 3000 cm ), CH3 and
-
1
CH2 deformations (1500-1100 cm region), and C-C and
-
1
C-O stretches (1000-820 cm region). Assignments of IR
bands are listed in Table 4.
A crucial question is whether both stable conformers, the
gauche and the trans, can be identified in the experimental IR
spectrum. It is not straightforward to decide in this question
because most of the fundamentals of the two conformers lie
close to each other with similar intensities. The most straight-
forward proof of the coexistence of the two conformers is found
-
1
in the 1200-800 cm spectral region (see inset to Figure 2).
The scaled calculated frequency of the O-C(N) stretch of the
trans conformer matches the position of the strong Q branch in
-
1
°C for IR investigation, and no indication of isomerization were
the spectrum at 1106 cm , so we assign this band to the trans
conformer. The partially overlapped bands at 1136, 1008, and
observed at this temperature during the measurement.
The IR spectrum of gaseous CH3CH2OCN is shown in Figure
-
1
825 cm are also assigned to the trans conformer, on the basis
of calculated frequencies, IR intensities, and conformer ratio.
2
with observed and calculated frequencies listed in Table 4.
-
1
For a better assistance in assignment of the spectrum, the
computed harmonic frequencies are corrected using the popular
scaled quantum mechanical (SQM) method of Pulay et al.,
where the computed harmonic force field is improved by a small
The band at 1180 cm with medium intensity and the weak
-
1
bands at 974 and 798 cm clearly show the presence of the
gauche conformer. We note that there are other indications of
the presence of two conformers. The weak bands at 1377 and
1
5
-1
number of properly chosen transferable scale factors. Such
general scale factors have been developed for the B3LYP/6-
1290 cm can only be explained with the presence of the trans
and gauche conformers, respectively. The strong band at 2255

1724 J. Phys. Chem. A, Vol. 107, No. 11, 2003
Pasinszki et al.
TABLE 4: Experimental and Calculated Vibrational Frequencies (cm-1) of CH
3 2
CH OCN
transb
gaucheb
assignment and TEDe
expt
frequency
a
frequencyc intensd
assignment and TEDe
frequencyc intensd
3027 (3150) 14.2
3
021 Q m
007 Q m
997 Q m
989 Q w
ν
1
78% CH
2
3
asst, 13% CH asst
3
2
2
3011 (3134) 24.3
2995 (3117) 14.2
ν
ν
ν
ν
ν
16 65% CH
98% CH
17 65% CH
3
3
2
2
asst, 35% CH
asst
asst, 35% CH
sst
2
3
asst
asst
1
2996 (3119)
8.8
ν
ν
ν
ν
2
3
4
5
87% CH
95% CH
85% CH
99% CH
3
3
2
3
asst
asst
sst, 11% CH
sst
2986 (3108)
2945 (3065) 13.5
2933 (3052) 6.8
1.0
2987 (3110) 11.9
2957 (3078) 16.0
(2963)
2
98% CH
2
asst
2
2
2
2
934 w
262 R
3
100% CH
3
sst
2928 (3047)
5.3
255 Q vs 2259 (2337) 225.9
248 P
ν
4
83% CtN st, 17% OC st
2257 (2334) 201.7
ν
6
83% CtN st, 17% OC st
}
1
506 (1520)
6.2
2.5
7.7
ν
ν
ν
ν
ν
ν
ν
ν
5
95% CH
78% CH
18 89% CH
72% CH
77% CH
19 79% CH
20 54% CH
2
3
3
3
2
2
2
3
sci
asd
asd, 11% CH
sd, 14% CH
w, 20% CH
tw, 17% CH
r, 30% CH
1501 (1516)
1477 (1503)
1462 (1492)
1414 (1426) 15.8
1387 (1400) 2.9
1306 (1317) 13.3
1190 (1203) 43.9
5.6
0.9
7.7
ν
ν
ν
ν
ν
ν
ν
7
8
9
92% CH
77% CH
88% CH
2
3
3
3
2
2
2
sci
asd
asd, 11% CH
sd
w, 12% CH
tw, 11% CH
3
r, 33% CH r
1
1
1
1
1
1
1
1
1
1
1
9
8
480 w
454 w
401 m
377 w
290 w
180 Q m
136 sh
113 R
106 Q vs 1107 (1158) 133.1
098 P
008 m
74 w
25 Q m
1479 (1506)
1460 (1490)
6
3
r
w
sd
3
r
1418 (1430) 24.1
1391 (1403) 12.2
7
2
10 81% CH
11 84% CH
12 84% CH
13 47% CH
8
3
3
sd
3
r
1289 (1300)
1161 (1171)
0.7
4.2
3
r
3
r, 11% CH
2
tw
1137 (1121) 63.6
9
27% CH
r, 25% OC st, 21% CC st
1110 (1105) 22.5
1090 (1124) 115.7
ν
ν
14 41% CH
3
r, 14% CC st, 13% OC st
ν
10 42% OC st, 20% C-O st, 19% CH
3
r
}
15 39% OC st, 24% CC st, 14% C-O st
16 47% CC st, 22% CH r, 12% OC st
1007 (1008) 39.3
ν
11 56% CC st, 15% CH r, 15% CO st
3
978 (983)
827 (828)
24.6
10.9
ν
ν
3
825 (821)
29.6
1.3
ν
ν
12 51% C-O st, 22% CH
3
21 44% CH r, 42% CH r
3
r
17 29% C-O st, 29% CH
3
r, 16% CH
2
r
8
14 (821)
2
7
98 Q m
796 (796)
25.7
4.0
9.9
0.8
1.8
5.0
0.9
ν
ν
ν
ν
ν
ν
ν
18 33% C-O st, 26% CH
2
r, 23% CH
19 46% OCN b, 35% CCO b
20 97% OCN b
3
r
5
5
3
2
1
97 (596)
28 (528)
78 (379)
42 (243)
77 (177)
5.2
8.1
6.1
0.1
4.9
1.8
ν
ν
ν
ν
ν
ν
13 52% OCN b, 30% CCO b, 12% C-O st 622 (622)
22 100% OCN b
14 69% COC b, 10% OCN b
23 97% CC t
15 55% CCO b, 32% OCN b, 12% COC b
24 100% C-O t
532 (532)
395 (396)
270 (271)
188 (188)
77 (77)
21 74% COC b, 13% OCN b
22 73% CC t, 12% CCO b, 11% OCN b
23 45% CCO b, 25% OCN b, 23% CC t
24 100% C-O t
6
1 (61)
a
For the relative intensities see also Figure 3. b Calculated at the B3LYP/6-311+G(2d,2p) level. c Scaled frequencies (see text). Scale factors
st: 0.9231, OC st: 0.9285, CN st: 0.9371, C-O st: 1.0378, CC st: 1.0000, HCH b: 0.9577, CCH b: 0.9846,
COC b: 1.0000, OCN lin. b: 1.0000, CH
t: 1.0000. Unscaled harmonic frequencies are in parentheses. d In km/mol. e Characterization of the
used in the SQM analysis: CH
3
3
fundamentals: total energy distribution (TED). Only contributions above 10% are given. The abbreviations sst, asst, st, sd, asd, b, sci, r, tw, w, t
mean symmetric stretching, asymmetric stretching, stretching, symmetric deformation, asymmetric deformation, bending, scissoring, rocking, twisting,
wagging and torsion, respectively. Scaled force constants were used.
TABLE 5: Calculated Barriers (TS2) of the Bimolecular Reaction of CH
3
CH
2
OCN and Energetics (D2) for the Bimolecular
Formation of OCN Iona
-
b
solvent
vacuum
heptane
dichloroethane
acetone
nitromethane
DMSO
TS2a[∆G (∆E)], kJ/mol
TS2b[∆G (∆E)], kJ/mol
TS2c[∆G (∆E)], kJ/mol
D2[∆G (∆E)] , kJ/mol
172.9 (128.0)
173.6 (131.5)
155.6 (112.9)
151.8 (110.5)
153.2 (112.4)
152.3 (112.8)
169.7 (124.4)
161.1 (128.7)
145.9 (110.4)
140.3 (109.3)
139.1 (108.8)
141.3 (109.7)
c
c
418.1 (425.9)
223.1 (230.5)
62.0 (56.6)
39.4 (36.9)
34.0 (29.8)
32.3 (27.8)
155.2 (109.9)
140.2 (107.7)
145.6 (107.8)
148.4 (110.1)
a
Calculated at the B3LYP/6-311+G(2d,2p) level. Energies (including ZPE) and Gibbs free energies (298 K, 1 atm) are relative to that of the
sum of two CH
formation of OCN ion. Energetics is determined for the CH
calculated.
3
CH
2
OCN molecules; TS2a: barrier to two molecules of CH
3
CH
2
NCO, TS2b: barrier to the dimer, DP, TS2c: barrier to the
-
b
+
-
c
3
2
CH OCN + CH
3
2
CH OCN f CH
3 2
CH OCNCH
2
CH
3
+ OCN reaction. Not
cm- has an unusual contour showing a shoulder at 2278 cm-1.
Although there are several strongly overlapped bands in the CH
stretching region, four Q-branches at 3021, 3007, 2997, and
1
5 and Figures 3 and 4. Calculations predict that the barrier of
the unimolecular isomerization in various solvents is between
∆G ) 218.9 (in a vacuum) and 195.7 kJ/mol (in DMSO),
q
-1
-1
2
989 cm , and a broader feature between 2960 and 2900 cm
which seems to be too high to explain instability at room or
lower temperatures. Search for potential bimolecular reactions
revealed two reaction routes in all solvents, leading to the isomer
(2) or to a dimeric product (DP), and also a third one in polar
solvents, producing cyanate ion. Kinetic energy barriers of these
reactions are 40-50 kJ/mol lower than those of the unimolecular
processes (Table 5). Figure 4 shows that the formation of the
isomer 2, via TS2a, is thermodynamically more favored than
the formation of DP, via TS2b, and the two OCN groups are
practically exchanged between two molecules of 1 in this
reaction. DP may be an intermediate for the formation of the
are clearly observed. These also imply the presence of two
conformers in the vapor. Overall, the band positions and relative
intensities compare favorably with the calculated positions and
intensities. The spectrum confirms the presence of two conform-
ers, but a conformer ratio could not be derived from the spectrum
because of the extensive band overlap.
Isomerization of CH3CH2OCN. The uni- and bimolecular
isomerization of CH3CH2OCN (1) was investigated in a
vacuum” and in solvents using the B3LYP/6-311+G(2d,2p)
method and the PCM model. Results are summarized in Table
“

Study of the Unstable CH3CH2OCN Molecule
J. Phys. Chem. A, Vol. 107, No. 11, 2003 1725
the presence of two conformers, the gauche and the trans form.
Ab initio and DFT calculations provide the equilibrium struc-
tures and the ratio of these two conformers, and together with
normal coordinate calculations, support the assignment of the
spectrum. The isomerization of ethyl cyanate was studied in a
vacuum and in various solvents using DFT and PCM. Calcula-
tions show that ethyl cyanate is unstable in the condensed phase
because of bimolecular reactions. The isomerization is predicted
to be faster and more complicated in a polar solvent than in an
apolar medium because of the formation and catalytic action
of cyanate ions.
Acknowledgment. We thank the Hungarian Scientific
Research Fund and the Hungarian Academy of Sciences (OTKA
Grant F022031, AKP Grant 98-69 2,4) for research grants in
support of this work.
References and Notes
Figure 3. Unimolecular (top) and OCN- ion catalized (bottom)
(
1) Martin, D.; Niclas, H.-J.; Habisch, D. Liebigs Ann. Chem. 1969,
27, 10.
2) The Chemistry of Cyanates and their Thio DeriVatiVes; Patai, S.,
Ed.; Wiley: New York, 1977.
3) Al-Juaid, S. S.; Al-Nasr, A. A. K.; Ayoko, G. A.; Eaborn, C.;
Hitchcock, P. J. Organomet. Chem. 1995, 488, 155.
4) Eaborn, C.; Lickiss, P. D.; Markuina-Chidsey, G.; Thorli, E. Y. J.
Chem. Soc., Chem. Commun. 1982, 1326.
isomerization of CH
relative to that of CH
respectively. Calculated at the B3LYP/6-311+G(2d,2p) level (∆G in
3
CH
2
OCN (in a vacuum). Gibbs free energies are
OCN (1) and CH CH OCN + OCN ion,
2 3 2
7
-
3
CH
(
kJ/mol).
(
(
(
5) Martin, D. Tetrahedron Lett. 1964, 39, 2829.
(6) Martin, D.; Weise, A. Chem. Ber. 1966, 99, 976. Holm, A.; Huge-
Jensen, E. Acta Chem. Scand. B 1974, 28, 705. Fischer, H.; Zeuner, S.;
Ackermann, K.; Schubert, U. J. Organomet. Chem. 1984, 263, 201.
(
(
(
7) Reich, P.; Martin, D. Chem. Ber. 1965, 98, 2063.
8) Groving, N.; Holm, A. Acta Chem. Scand. 1965, 19, 443.
9) Jensen, K. A.; Holm, A.; Wentrup, C. Acta Chem. Scand. 1966,
2
0, 2107.
10) Sakaizumi, T.; Mure, H.; Ohashi, O.; Yamaguchi, I. J. Mol.
Spectrosc. 1989, 138, 375.
11) Urban, J.; Nowek, A.; Venkatraman, R.; Babinec, P.; Leszczynski,
J. Struct. Chem. 1998, 9, 161.
(
(
(
(
(
12) Pasinszki, T.; Westwood, N. P. C. J. Phys. Chem. 1995, 99, 1649.
13) Davies, W.; Maclaren, J. A. J. Chem. Soc. 1951, 1434.
14) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A.
D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi,
M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.;
Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick,
D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.;
Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi,
I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M.
W.; Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon,
M.; Replogle, E. S.; Pople, J. A. Gaussian 98, revision A.7; Gaussian,
Inc.: Pittsburgh, PA, 1998.
Figure 4. Bimolecular reaction of CH
free energies are relative to that of two molecules of CH
1+1). Calculated at the B3LYP/6-311+G(2d,2p) level using PCM (∆G
in kJ/mol).
3
CH
2
OCN in heptane. Gibbs
3
CH OCN
2
(
(
15) Pulay, P.; Fogarasi, G.; Pongor, G.; Boggs, J. E.; Vargha, A. J.
Am. Chem. Soc. 1983, 105, 7037.
16) Fogarasi, G.; Zhou, X.; Taylor, P. W.; Pulay, P. J. Am. Chem. Soc.
trimer, triethylisocyanurate, but the investigation of this is
beyond the scope of this work. It is important to note the third
(
-
reaction route in polar solvents, which produces OCN ion, via
1992, 114, 8191.
-
TS2c. Calculations show that OCN ion can effectively isomer-
(17) Coffin, J. M.; Pulay, P. Program TRA3; Department of Chemistry
and Biochemistry, University of Arkansas: Fayetteville, AR, 1989.
ize CH3CH2OCN, because the kinetic energy barrier for this
(18) Pulay, P. In Applications of Electronic Structure Theory, Modern
q
q
process is small; it is ∆G ) 93.0 kJ/mol (∆E ) 62.5 kJ/mol)
Theoretical Chemistry; Schaefer, H. F., III., Ed.; Plenum: New York, 1977;
Vol. 4, p 153.
(19) Pongor, G.; Fogarasi, G.; Magd o´ , I.; Boggs, J. E.; Keresztury, G.;
Ignatyev, I. S. Spectrochim. Acta 1992, 48A, 111.
in DMSO and decreasing by decreasing the polarity of the
solvent (see Figure 3). The OCN ion is thus expected to be an
-
isomerization catalyst for cyanat-isocyanat rearrangement;
therefore, the isomerization of 1 is expected to be faster in polar
(20) Pongor, G. Program SCALE3; Department of Theoretical Chem-
istry, E o¨ tv o¨ s Lor a´ nd University: Budapest, Hungary, 1993.
(21) Pulay, P.; T o¨ r o¨ k, F. Acta Chim. Hung. 1965, 44, 287.
(22) Keresztury, G.; Jalsovszky, G. J. Mol. Struct. 1971, 10, 304.
-
than in apolar solvents. We note that the OCN ion is known
1
,33
to catalyze the trimerization of the isomerization product, 2.
(23) (a) Sakaizumi, T.; Sekishita, K.; Furuya, K.; Tetsuda, Y.; Kaneko,
K.; Ohashi, O.; Yamaguchi, I. J. Mol. Spectrosc. 1993, 161, 114. (b)
Sakaizumi, T.; Mure, H.; Ohashi, O.; Yamaguchi, I. J. Mol. Spectrosc. 1990,
Conclusion
140, 62.
The unstable ethyl cyanate molecule has been generated by
passing O-ethyl thiocarbamate over yellow mercuric oxide. A
pure sample was obtained by trapping and reevaporizing reaction
products at 0 °C. Ethyl cyanate was characterized in the gas
phase by infrared spectroscopy, and the IR spectrum indicates
(24) Zylka, P.; Mack, H.-G.; Schmuck, A.; Seppelt, K.; Oberhammer,
H. Inorg. Chem. 1991, 30, 59.
25) Seppelt, K.; Oberhammer, H. Inorg. Chem. 1985, 24, 1227.
Oberhammer, H.; Seppelt, K.; Mews, R. J. Mol. Struct. 1983, 101, 325.
26) Koput, J.; Winnewisser, B. P.; Winnewisser, M. Chem. Phys. Lett.
1996, 255, 357.
(
(

1726 J. Phys. Chem. A, Vol. 107, No. 11, 2003
Pasinszki et al.
(
(
27) Pasinszki, T.; Westwood, N. P. C. J. Phys. Chem. 1995, 99, 6401.
28) Pasinszki, T.; Westwood, N. P. C. J. Phys. Chem. A 1998, 102,
(32) The theoretical PR separations for pure A-, B-, and C-type bands
using the computed rotational constants and the equations of Seth-Paul (Seth-
Paul, W. A. J. Mol. Struct. 1969, 3, 403) are A: 15.8 and 14.0; B: 13.1
and 11.7; C: 23.7 and 21.0 cm for the gauche and trans forms,
respectively.
4
1
939.
-
1
(
(
(
29) Pasinszki, T.; Westwood, N. P. C. J. Phys. Chem. 1996, 100, 16856.
30) Rauhut, G.; Pulay, P. J. Phys. Chem. 1995, 99, 3093.
31) Baker, J.; Jarzecki, A. A.; Pulay, P. J. Phys. Chem. A 1998, 102,
(33) Arnold, R. G.; Nelson, J. A.; Verbane, J. J. Chem. ReV. 1957, 57,
47.
412.