Conformational stability of CH3CH2P(Z)F2 (Z = O,S) from temperature dependent FT-IR spectra of rare gas solutions and r(o) structural parameters

Article abstract of DOI:10.1016/S0022-2860(99)00194-5

Variable temperature (from -55 to -150°C) studies of the infrared spectra (3500-400 cm^-1) of ethylphosphonic difluoride, CH₃CH₂P(O)F₂ and ethylphosphonothioic difluoride, CH₃CH₂P(S)F₂ dissolved in liquid xenon or krypton have been recorded. From these data, the enthalpy differences have been determined to be 76 ± 9 cm^-1 (0.91 ± 0.11 kJ/mol), for CH₃CH₂P(O)F₂ with the trans conformer the more stable rotamer and 53 ± 7 cm^-1 (0.63 ± 0.08 kJ/mol) for CH₃CH₂P(S)F₂ but with the gauche conformer the more stable form. Complete vibrational assignments are presented for both molecules, which are consistent with the predicted frequencies obtained from the ab initio MP2/6-31G(d) calculations. The optimized geometries, conformational stabilities, harmonic force fields, infrared intensities, Raman activities, and depolarization ratios have been obtained from RHF/6-31G(d) and/or MP2/6-31G(d) ab initio calculations. These quantities are compared to the corresponding experimental quantities when appropriate as well as with some corresponding results for some similar molecules. The r(o) adjusted structural parameters have been obtained for both molecules from a combination of the microwave rotational constants and ab initio predicted parameters. The corresponding r(o) structural parameters have been obtained for some similar molecules for comparison.

Full text of DOI:10.1016/S0022-2860(99)00194-5

Journal of Molecular Structure 516 (2000) 131–152
www.elsevier.nl/locate/molstruc
Conformational stability of CH3CH2P(Z)F2 (Z ˆ O,S) from
temperature dependent FT-IR spectra of rare gas solutions and ro
structural parameters
1
*
J.R. Durig , J.B. Robb II, J. Xiao , T.K. Gounev
Department of Chemistry, University of Missouri-Kansas City, Kansas City, MO 64110-2499, USA
Received 4 March 1999; accepted 6 April 1999
Abstract
Variable temperature (from Ϫ55 to Ϫ150ЊC) studies of the infrared spectra (3500–400 cmϪ1) of ethylphosphonic difluoride,
CH3CH2P(O)F2 and ethylphosphonothioic difluoride, CH3CH2P(S)F2 dissolved in liquid xenon or krypton have been recorded.
From these data, the enthalpy differences have been determined to be 76 ^ 9 cmϪ1 (0.91 ^ 0.11 kJ/mol), for CH3CH2P(O)F2
with the trans conformer the more stable rotamer and 53 ^ 7 cmϪ1 (0.63 ^ 0.08 kJ/mol) for CH3CH2P(S)F2 but with the gauche
conformer the more stable form. Complete vibrational assignments are presented for both molecules, which are consistent with
the predicted frequencies obtained from the ab initio MP2/6-31G(d) calculations. The optimized geometries, conformational
stabilities, harmonic force fields, infrared intensities, Raman activities, and depolarization ratios have been obtained from RHF/
6-31G(d) and/or MP2/6-31G(d) ab initio calculations. These quantities are compared to the corresponding experimental
quantities when appropriate as well as with some corresponding results for some similar molecules. The ro adjusted structural
parameters have been obtained for both molecules from a combination of the microwave rotational constants and ab initio
predicted parameters. The corresponding ro structural parameters have been obtained for some similar molecules for compar-
ison. ᭧ 2000 Elsevier Science B.V. All rights reserved.
Keywords: Conformational stability; FT-IR spectra; Rare gas solutions; Ethylphosphonic difluoride; Ethylphosphonothioic difluoride
1. Introduction
that the trans conformer (methyl group trans to the
nonbonded electron pair on the phosphorous atom) is
more stable than the gauche form by 80 ^ 7 cmϪ1
(0.96 ^ 0.24 kJ/mol). The value agrees with the
earlier reported [2] value of 56 ^ 22 cmϪ1 (0.67 ^
0.26 kJ/mol) with the trans conformer more stable
from a Raman and infrared investigation of the gas.
This result indicates that the conclusion from the
microwave study [3] that the gauche form is more
stable in the vapor is in error. As a continuation of
these studies, we were interested in what effect the
substitution for the non-bonded lone pair by an
oxygen or sulfur atom would have on the
We [1] recently carried out a temperature depen-
dent infrared study of ethyldifluorophosphine
dissolved in liquid krypton to determine the confor-
mational stability. From this investigation, it was clear
* Corresponding author. Tel.: ϩ 1-8816-235-1136; fax: ϩ 1-
816-235-5191.
E-mail address: durigj@umkc.edu (J.R. Durig).
1 Taken in-part from the dissertation of Jinpeng Xiao, which will
be submitted to the Department of Chemistry of the University of
Missouri-Kansas City, Kansas City, MO in partial fulfillment of the
PhD degree.
0022-2860/00/$ - see front matter ᭧ 2000 Elsevier Science B.V. All rights reserved.
PII: S0022-2860(99)00194-5

132
J.R. Durig et al. / Journal of Molecular Structure 516 (2000) 131–152
values when appropriate. The calculations have been
carried out with several basis sets at both the restricted
Hartree–Fock (RHF) and with full electron correla-
tion by the perturbation method (Moller–Plesset) to
second order [7]. Additionally, the ro structural para-
meters have been determined for both molecules from
the combined microwave data and ab initio calcula-
tions. The results of this experimental and theoretical
study are reported herein.
2. Experimental
The samples of ethylphosphonic difluoride,
CH3CH2P(O)F2, and ethylphosphonothioic difluoride,
CH3CH2P(S)F2 were prepared by the fluorination of
ethylphosphonic dichloride (Aldrich Chemical Co,
Milwaukee, WI) and ethylphosphonothioic dichloride
(Strem Chemicals Inc., Newburyport, MA), respec-
tively, with freshly sublimed antimony trifluoride
(Aldrich), dispersed in “glass wool”. The products
were separated by trap-to-trap distillation and purified
on a low-temperature, low-pressure fractionation
column. The purified samples were contained in
sample tubes equipped with greaseless stopcocks
and stored under vacuum in a refrigerator at Ϫ5ЊC.
The mid-infrared spectra of the samples dissolved
in liquified xenon or krypton as a function of tempera-
ture (Figs. 1(a) and 2(a)) were recorded on a Bruker
IFS-66 Fourier transform interferometer equipped
with a globar source, a Ge/KBr beamsplitter, and a
DTGS detector. The temperature studies ranged from
Ϫ55 to Ϫ100ЊC or Ϫ105 to Ϫ150ЊC and were
performed in a specially designed cryostat cell
consisting of a 4 cm pathlength copper cell with
wedged silicon windows sealed to the cell with
indium gaskets. The complete system is attached to
a pressure manifold to allow for the filling and
evacuation of the cell. The cell is cooled by boiling
liquid nitrogen and the temperature is monitored by
two Pt thermoresistors. Once the cell is cooled to the
desired temperature, a small amount of sample is
condensed into the cell. The system is then pressur-
ized with xenon or krypton gas, which immediately
starts condensing in the cell, allowing the compound
to dissolve. For each temperature investigated, 100
interferograms were recorded at 1.0 cmϪ1 resolution,
averaged, and transformed with a boxcar truncation
Fig. 1. Comparison of experimental and calculated infrared spectra
of ethylphosphonic difluoride: (a) infrared spectrum of the sample
dissolved in liquid krypton; (b) mixture of the calculated trans and
gauche conformers with the experimentally determined DH of
76 cmϪ1; (c) pure gauche; and (d) pure trans. The bands marked
with an asterisk are due to an impurity.
conformational stability. Therefore, we have carried
out variable temperature studies of the infrared
spectra
of
ethylphosphonic
difluoride,
CH3CH2P(O)F2, and ethylphosphonothioic difluoride,
CH3CH2P(S)F2 dissolved in liquid xenon or krypton to
determine the conformational stabilities and the
enthalpy differences between the conformers.
Although, the conformational stabilities of these
molecules [4–6] have been confidently determined
for the liquids and solids, there are still questions on
their stabilities in the vapor state.
To support the spectroscopic studies, we have
carried out ab initio calculations to predict the confor-
mational stabilities and structural parameters. Addi-
tionally, we have calculated the vibrational
frequencies and the infrared and Raman intensities
along with the Raman depolarization values. These
quantities have been compared to the experimental

J.R. Durig et al. / Journal of Molecular Structure 516 (2000) 131–152
133
the simultaneous relaxation of all of the geometric
parameters using the gradient method of Pulay [9].
The structural optimizations for both the trans and
gauche conformers were carried out with initial para-
meters taken from those previously reported from the
microwave studies [5,6] of CH3CH2P(O)F2 and
CH3CH2P(S)F2. The 6-31G(d) and 6-311 ϩ G(d,p)
or 6-311ϩϩG(d,p) basis sets were employed at the
level of RHF and/or Moller–Plesset (MP2) to second
order [7]. The predicted structural parameters for
CH3CH2P(O)F2, and CH3CH2P(S)F2, are listed in
Tables 3 and 4, respectively.
In order to obtain a more complete description of
the molecular motions involved in the normal modes
of these molecules we have carried out a normal coor-
dinate analysis. The force fields in Cartesian coordi-
nates were calculated by the Gaussian-94 program
[8] with the MP2/6-31G(d) basis set. Internal coordi-
nates (Fig. 3) were used to calculate the G and B
matrices using the structural parameters given in
Tables 3 and 4. Using the B matrix, the force field
in Cartesian coordinates was then converted to a force
field in internal coordinates, and the pure ab initio
vibrational frequencies were reproduced. The force
constants for the trans and gauche conformers
can be obtained from the authors. Subsequently,
scaling factors of 0.9 for stretchings and bendings
and 1.0 for the torsional coordinates, and the
geometric average of scaling factors for interac-
tion force constants were used to obtain the
fixed scaled force field and resultant wavenum-
bers. A set of symmetry coordinates was used
(Table 5) to determine the corresponding potential
Fig. 2. Comparison of experimental and calculated infrared spectra
of ethylphosphonothioic difluoride: (a) infrared spectrum of the
sample dissolved in liquid xenon; (b) mixture of the calculated
trans and gauche conformers with the experimentally determined
DH of 53 cmϪ1; (c) pure trans; and (d) pure gauche.
energy distributions (PEDs).
A
comparison
between the observed and calculated frequencies
for the two molecules along with the calculated
infrared intensities, Raman activities, depolariza-
tion ratios and PEDs are given in Tables 1 and 2.
Theoretical infrared (Figs. 1 and 2) and Raman
(Figs. 4 and 5) spectra were predicted using fixed
scaled wavenumbers and infrared intensities deter-
mined from the MP2/6-31G(d) calculations and
Raman scattering activities and depolarization values
determined from the RHF/6-31G(d) basis set. Infrared
intensities were calculated based on the dipole
moment derivatives with respect to the Cartesian
coordinates. The derivatives were taken from the ab
initio calculations and transformed to normal
function. The observed fundamental bands for the
trans and gauche conformers are listed in Tables 1
and 2 for ethylphosphonic difluoride and ethylpho-
sphonothioic difluoride, respectively.
3. Ab initio calculations
The LCAO-MO-SCF RHF calculations were
performed with the Gaussian-94 program [8] using
Gaussian-type basis functions. The energy minima
with respect to nuclear coordinates were obtained by

Table 1
Observed and calculated frequencies (cmϪ1) and PED for ethylphosphonic difluoride
Species Vib. Fundamental
Noa
Trans
Gauche
Ab
Fixed
IR
Raman Calc.
act.c dep.c
Obs.
gasd
Xenon PED
soln.
Ab
Fixed
IR
Raman Calc. Obs.
act.c dep.c gasd
Xenon PED
soln.
initio scaledb int.
initio scaledb int.
A0
n1
CH3 antisymmetric stretch
CH2 symmetric stretch
CH3 symmetric stretch
3220 3055
3131 2971
3129 2968
9.4
76.53 0.70
2965 2971
2941 2928
100S1
3219 3053
3133 2972
3126 2965
1571 1490
1522 1444
1479 1403
1382 1312
1370 1300
1103 1047
10.0
5.3
9.1
10.7
8.8
2.5
124.8
55.0
15.3
4.0
188.4
18.5
45.2
11.7
0.7
58.56 0.68
2963
–
–
–
–
–
–
689S1,29S17
95S2
95S3
91S4
98S5
98S6
64S8,16S7
58S7,20S8
46S9,40S10
46S10,22S9,21S7
72S11,12S12
56S12,18S11
27S14,23S13,18S16,18S24
21S13,38S15,12S24
33S15,27S26,16S14
45S16,22S15,18S24,12S14
65S17,32S1
93S18
n2
10.5 112.93 0.10
75S2,24S3
76S3,24S2
92S4
99S5
98S6
39S7,42S8
48S8,35S7
45S9,43S10
41S10,23S9,22S7
68S11,11S12
57S12,23S11
74S13,20S16
40S14,38S15,11S12
53S15,23S16,12S14
42S16,36S14,13S13
96S17
99.32 0.13
92.60 0.02
11.37 0.74
14.12 0.75
1.73 0.75
3.24 0.23
1.24 0.27
3.90 0.29
7.28 0.51
0.83 0.04
9.76 0.04
1.90 0.75
0.55 0.67
0.26 0.75
0.08 1.66
29.92 0.73
–
–
–
–
–
n3
5.5
10.0
11.6
1.0
123.9
74.8
11.8
5.9
165.9
16.7
47.4
28.0
5.8
78.18 0.02
12.19 0.74
13.58 0.75
1.79 0.74
3.33 0.32
3.08 0.21
3.85 0.45
7.95 0.45
0.78 0.30
9.86 0.04
1.31 0.74
3.42 0.42
1.56 0.48
0.12 0.64
21.42 0.75
98.30 0.75
2906 2893
1470 1461
1415 1408
1395 1389
(1342) 1355
1281 1274
1049 1045
n4
CH3 antisymmetric deformation 1572 1491
n5
CH2 deformation
CH3 symmetric deformation
PyO stretch
CH2 wag
CH3 rock
CC stretch
PF2 symmetric stretch
CP stretch
PF2 wag
CPO bond
PF2 deformation
CCP bend
CH3 antisymmetric stretch
CH2 antisymmetric stretch
CH3 antisymmetric deformation 1566 1486
CH2 twist
CH3 rock
PF2 antisymmetric stretch
CH2 rock
PF2 rock
PF2 twist
1520 1442
1480 1404
1403 1331
1347 1278
1108 1051
n6
n7
1366 1349
1287 1282
1046 1038
980
–
720
493
389
(298)
(180)
2996
–
–
1232
1024 1015
897
760
n8
n9
n10
n11
n12
n13
n14
n15
n16
n17
n18
n19
n20
n21
n22
n23
n24
n25
n26
n27
1034
908
729
491
414
329
187
981
861
692
466
393
312
177
983
888
713
496
426
983
882
712
490
425
–
1030
906
740
486
398
298
173
978
859
702
461
379
288
165
–
876
719
490
–
–
–
–
–
333
1.0
7.9
0.0
8.9
(185)
–
1.6
7.0
A00
3228 3062
3186 3022
(2992) 2990
2956 2951
(1460) 1461
1242 1237
1031 1024
3225 3059
3191 3027
1565 1484
1310 1243
1082 1027
96S18
92S19
0.4 107.56 0.73
8.9
6.0
19.0
0.75
18.48 0.75
7.01 0.73
2.59 0.62
2.06 0.68
0.80 0.36
2.58 0.50
1.44 0.43
0.17 0.75
0.07 0.75
–
-
90S19
1315 1248
1087 1032
1.8
6.47 0.75
0.92 0.75
1.46 0.75
0.20 0.75
1.33 0.75
0.18 0.75
0.09 0.75
0.00 0.75
61S20,29S21
36S21,35S20,20S23
87S22
62S23,21S22
67S24,22S25
46S25,48S26
39S26,30S25,19S24,12S23
97S27
54S20,29S21
34S21,35S20,19S23
90S22
59S23,24S21
69S24,35S13
50S25,15S15,23S16
50S26,21S14
92S27
21.2
165.4
1.6
19.0
1.3
8.5
932
777
405
286
218
79
884
738
386
279
211
79
896
752
407
(292)
(201)
(76)
–
749
405
–
–
–
938
792
421
322
214
73
890
753
400
306
209
73
171.4
10.2
30.3
8.6
0.4
0.5
904
754
425
–
–
–
426
324
(192)
(68)
CH3 torsion
Ethyl torsion
0.2
1.0
a For Convenience, the vibrations are ordered according to the trans conformer.
b Scaled ab initio calculations with factor 0.9 for all coordinates except the torsions at the MP2/631G(d) level.
c Obtained from the RHF/6-31G(d) basis set.
d Wavenumbers in parentheses taken from Ref. [4].

Table 2
Observed and calculated frequencies (cmϪ1) and PED for ethylphosphonothioic difluoride
Species Vib. Fundamental
Noa
Trans
Gauche
Ab
Fixedb
IR
int.c
Raman Calc.
act.d dep.c
Obs.
gasd
Xenon PED
soln.
Ab
Fixed
IR
Raman Calc. Obs.
Xenon PED
soln.
initio scaled
initio scaledb int.
act.c
dep.c gasd
A0
n1
CH3 antisymmetric stretch
CH2 symmetric stretch
CH3 symmetric stretch
3217 3052
3123 2963
3127 2966
9.4
5.5
71.97 0.67
76.06 0.04
2993 2988
89S1,10S17
79S2,21S3
78S3,21S2
91S4
100S5
3217 3052
3128 2968
3126 2966
1571 1490
1517 1439
1476 1401
1371 1300
1371 1300
8.3 90.94
18.1 91.84
0.4 99.45
8.8 12.79
10.9 13.18
0.70
0.13
0.03
0.74
0.71
2993 2988
99S1
n2
2902 2890
2932 2922
1466 1458
1414 1406
(1390) 1385
1281 1274
2890
2887
1458
1406
70S2,30S3
70S3,30S2
92S4
100S5
98S6
72S7,11S8
72S7,11S8
37S9,28S7,25S8
57S13,21S15,16S12
43S11,30S12
28S12,31S11,22S10
57S13,21S15,16S12
76S14,15S16
36S15,25S13,17S16,12S14
58S16,27S15
97S17
n3
8.3 128.70 0.10
9.0
7.1
4.4
7.3
13.0
5.8
270.8
135.8
3.5
24.1
2.3
9.2
n4
CH3 antisymmetric deformation 1569 1488
10.32 0.74
14.75 0.75
1.84 0.72
0.47 0.49
3.17 0.33
6.42 0.54
2.05 0.47
1.75 0.54
15.99 0.09
2.17 0.60
2.69 0.74
2.77 0.74
1.47 0.66
23.24 0.67
n5
CH2 deformation
CH3 symmetric deformation
CH2 wag
CH2 rock
C–C stretch
PF2 stretch
C–P stretch
PyS stretch
PF2 symmetric wag
PF2 deformation
CCP bend
CPS bend
CH3 antisymmetric stretch
CH2 antisymmetric stretch
CH3 antisymmetric deformation 1564 1484
CH2 twist
CH3 rock
PF2 antisymmetric stretch
CH2 rock
1517 1440
1476 1401
1364 1294
1102 1045
n6
98S6
70S7,12S8
2.2
7.7
7.7
4.7
24.6
133.3
2.12
0.09
0.09
7.93
5.31
1.75
0.75
0.16
0.16
0.44
0.33
0.29
0.08
0.33
0.44
0.54
0.72
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
(1390) 1385
1366 1274
1287 1274
n7
n8
1044
984
(885)
807
621
376
104
982
893
804
619
–
044S8,42S9
39S9,28S7,23S8
61S10,23S12
40S11,18S12,17S10
45S12,30S11,14S10
61S13,17S14
25S14,44S26,11S16
32S15,18S14,18S24,17S16
30S16,37S15,16S24,13S25
85S17,11S1
94S18
n9
1033
927
835
640
376
286
429
165
980
879
792
607
358
278
407
156
1032
434
820
662
434
371
295
165
979
412
778
628
412
352
280
157
1046
429
788
647
429
982
429
785
646
429
–
n10
n11
n12
n13
n14
n15
n16
n17
n18
n19
n20
n21
n22
n23
n24
n25
n26
n27
8.3 12.49
24.6
5.31
4.59
4.45
1.59
(277)
429
(170)
–
429
–
14.9
11.7
0.2
367
(288)
(170)
–
–
0.7
7.0
A00
3222 3057
3186 3022
2998 2988
2963 2949
2963 2949
1241 1236
1030 1019
3229 3063
3183 3019
1563 1483
1314 1247
1082 1027
7.1 24.49
0.0 85.64
9.5 20.88
2998 2988
2998 2988
0.4 109.91 0.74
10.8
97S17
92S19
18.61 0.74
6.10 0.74
1.74 0.75
1.30 0.25
1.30 0.28
3.72 0.48
2.02 0.70
0.77 0.73
94S19
–
1458
1313 1246
1087 1032
4.5
53.4
134.6
2.5
14.7
2.4
71S20,29S21
33S21,38S20
90S22
74S23,21S21
25S24,30S14,19S13
72S25,15S24
37S26,25S16
91S27
0.5
7.7
131.8
0.7
1.9
1.6
6.04
0.79
1.47
0.15
2.98
1.73
0.19
1232 1236
1022 1019
73S20,28S21
38S21,41S20,11S23
90S22
81S23,22S21
62S24,33S26
74S25,19S26
41S26,24S24,23S25
98S27
910
784
349
270
206
863
745
332
257
200
879
743
357
873
741
–
906
775
320
262
216
859
736
309
251
210
879
760
(313)
(262)
(187)
870
731
–
–
–
PF2 rock
PF2 twist
CH3 torsion
Ethyl torsion
(262)
(183)
–
–
0.1
0.2
0.2
77
77
0.34
0.75
(71)
–
75
75
0.4
0.14
(73)
–
a For convenience, in comparison with the CH3CH2P(O)F2 molecule the vibrations are ordered according to the trans conformer.
b Scaled ab initio calculations with factor 0.9 for all coordinates except the torsions at the MP2/6-31G(d) level.
c Obtained from the RHF/6-31G(d) basis set.
d Wavenumber in parentheses taken from Ref. [6].

Table 3
˚
Structural parameters (bond distance in A, bond angles in degrees, rotational constnats in MHz, and dipole moments in Debye), rotational constants, dipole moments and energy for
CH3CH2P(O)F2
Parameter
RHF/6-31G(d)
MP2/6-31G(d)
MP2/6-311 ϩ G(d,p)
Trans Gauche
1.786
Microwavea (ro)
This studyb (roadj
)
Trans
Gauche
1.791
Trans
Gauche
1.787
Trans
Gauche
Trans
Gauche
1.776
r(P–C1)
r(C1–C2)
r(C2–H1)
r(C2–H2)
r(C2–H3)
r(C1–H4)
r(C1–H5)
r(PyO)
r(P–F1)
r(P–F2)
ՄPC1C2
ՄH1C2C1
ՄH2C2C1
ՄH3C2C1
ՄH1C2H2
ՄH1C2H3
ՄH2C2H3
ՄH4C1C2
ՄH5C1C2
ՄH4C1H5
ՄPC1H4
ՄPC1H5
ՄOPC1
ՄF1PC1
ՄF2PC1
ՄF1PF2
ՄOPF1
ՄOPF2
t(H1C2C1P)
t(OPC1C2)
t(F1PC1O)
t(F2PC1O)
|ma|
1.792
1.533
1.083
1.083
1.083
1.085
1.085
1.438
1.550
1.550
115.0
109.6
111.3
111.3
108.1
108.1
108.3
111.1
111.1
106.4
106.3
106.3
118.1
104.1
104.1
98.60
114.8
114.8
180.0
180.0
128.6
Ϫ128.6
4.29
1.789
1.528
1.092
1.091
1.091
1.094
1.094
1.471
1.583
1.583
114.2
109.9
111.0
111.0
108.2
108.2
108.3
111.3
111.3
106.6
106.4
106.4
118.7
103.2
103.2
98.42
115.2
115.2
180.0
180.0
128.9
Ϫ128.9
4.38
1.785
1.532
1.091
1.092
1.092
1.094
1.093
1.463
1.582
1.580
1.802 ^ 0.012
1.795 ^ 0.013
1.522
1.777
1.531 ء
1.092 ء
1.091 ء
1.091 ء
1.094 ء
1.094 ء
1.452
1.535
1.083
1.084
1.083
1.085
1.085
1.438
1.551
1.549
1.530
1.091
1.092
1.091
1.094
1.093
1.471
1.583
1.582
1.531
1.092
1.091
1.091
1.094
1.094
1.462
1.582
1.582
1.522
1.091
1.091
1.091
1.093
1.093
1.440
1.549 ^ 0.013
1.549 ^ 0.013
113.6 ^ 1.0
110.0
111.0
111.0
108.4
108.4
108.0
111.3
1.532 ء
1.091 ء
1.092 ء
1.092 ء
1.094 ء
1.093 ء
1.452
1.549
1.548
111.8 ء
109.6 ء
111.1 ء
110.9 ء
108.1 ء
108.5 ء
108.4 ء
111.0 ء
111.6 ء
107.7 ء
106.4 ء
108.0 ء
118.0
1.091
1.091
1.091
1.093
1.093
1.440
1.546 ^ 0.008
1.546 ^ 0.008
111.8 ^ 1.2
110.0
111.0
111.0
108.4
108.4
108.0
111.3
1.549
1.549
112.5
111.4
114.7
111.8
114.7 ء
109.6 ء
111.1 ء
111.1 ء
108.2 ء
108.2 ء
108.4 ء
111.5 ء
111.5 ء
106.8 ء
106.0 ء
106.0 ء
118.5
109.7
111.2
111.0
108.2
108.4
108.4
110.6
111.2
107.3
106.8
108.2
118.6
103.9
103.7
98.75
114.6
114.9
178.7
302.6
128.5
Ϫ128.7
2.95
110.0
110.9
110.7
108.2
108.5
108.4
110.9
111.6
107.5
106.9
108.3
118.9
103.1
103.4
98.67
115.0
115.4
178.4
302.9
128.4
Ϫ129.2
3.04
109.6
111.1
111.1
108.2
108.2
108.4
111.5
111.5
106.8
106.0
106.0
119.2
103.4
103.4
98.6
109.6
111.1
110.9
108.1
108.5
108.4
111.0
111.6
107.7
106.4
108.0
119.0
103.3
103.5
98.8
111.3
107.6
106.3
106.3
111.3
107.6
107.3
107.3
118.6
118.6
104.4 ^ 0.4
104.4 ^ 0.4
98.8
114.2
114.2
180.0
180.0
128.4
Ϫ128.4
103.6 ^ 0.8
103.6 ^ 0.8
99.3
114.7
114.7
178.4
302.0
128.4
Ϫ128.4
2.649(55)
2.476(61)
0.000
3.6262(32)
4278
104.4
104.4
98.3
114.6
104.5
104.2
98.3
114.6
114.8
114.8
180.0
180.0
128.8
Ϫ128.8
114.6
115.0
178.9
303.0
128.3
Ϫ129.0
114.6
114.7
180.0 ء
180.0 ء
(128.4)
(Ϫ128.4)
178.9 ء
303.0 ء
(128.7)
(Ϫ128.6)
4.68
3.46
|mb|
|mc|
|mt|
A
B
C
0.14
0.00
4.29
4329
2336
2277
2.90
0.33
4.15
4279
2276
2236
0.39
0.0
4.40
4329
2336
2277
2.86
0.78
4.25
4118
2269
2221
0.62
0.00
4.72
2.89
0.74
4.57
4215
4142
4336
2344
2288
4336
2345
2289
4277
2294
2249
2312
2266
2262
2220
2292
2252
Ϫ (E ϩ 693)
0.217840
29
0.217970
1.146185
12
1.146238
1.588594
5
1.588616
DE (cmϪ1
)
a From Ref. [5].
b Parameters marked with asterisks ( ء ) are kept fixed at the ab initio MP2/6-311 ϩ G(d,p) values. Values in parentheses are dependent parameters.

Table 4
˚
Structural parameters (Bond distance in A, bond angles in degrees, rotational constnats in MHz, and dipole moments in Debye), dipole moments and total energies for ethylpho-
sphonothioic difluoride
Parameter
RHF/6-31G(d)
MP2/6-31G(d)
MP2/6-311 ϩ G(d,p)
Trans Gauche
1.794
1.530
1.092
1.091
1.091
1.094
1.094
1.888
1.592
1.592
Microwavea (ro)
This studyb (roadj
)
Trans
Gauche
1.801
Trans
Gauche
1.796
Trans
Gauche
Trans
Gauche
1.797
r(P–C1)
r(C1–C2)
r(C2–H1)
r(C2–H2)
r(C2–H3)
r(C1–H1)
r(C1–H2)
r( PyS)
r(P–F1)
r(P–F2)
ՄPC1C2
ՄC1C2H1
ՄC1C2H2
ՄC1C2H3
ՄH1C2H2
ՄH1C2H3
ՄH2C2H3
ՄC2C1H1
ՄC2C1H2
ՄH1C1H2
ՄPC1H2
ՄC1PS
ՄC1PF1
ՄC1PF2
ՄF1PF2
ՄSPF1
1.802
1.534
1.084
1.083
1.083
1.085
1.085
1.899
1.558
1.558
114.6
109.3
111.4
111.4
108.1
108.1
108.4
111.1
111.1
106.4
106.4
119.5
102.9
102.9
98.14
115.3
115.3
180.0
180.0
50.20
Ϫ50.80
4.166
1.409
0.000
4.398
3860
1.797
1.528
1.092
1.091
1.091
1.095
1.095
1.892
1.593
1.593
114.2
109.7
111.0
111.0
108.3
108.3
108.4
111.3
111.3
106.4
106.4
119.7
102.1
102.1
97.80
116.0
116.0
180.0
180.0
50.40
Ϫ50.40
4.212
1.266
0.000
4.198
3778
1.793
1.531
1.092
1.092
1.092
1.094
1.094
1.889
1.591
1.592
1.814 ^ 0.011
1.532 ^ 0.005
1.093
1.093
1.093
1.800 ^ 0.007
1.532 ^ 0.006
1.093 ^ 0.003
1.093 ^ 0.003
1.093 ^ 0.003
1.095 ^ 0.008
1.095 ^ 0.008
1.880 ^ 0.003
1.555 ^ 0.005
1.555 ^ 0.005
112.6 ^ 0.3
110.0 ^ 0.2
110.0 ^ 0.2
110.0 ^ 0.2
108.99
1.798
1.534
1.092
1.092
1.092
1.094
1.094
1.877
1.562
1.562
114.3
111.3
110.6
110.6
108.4
108.4
107.2
111.6
111.6
105.2
105.2
119.5
102.5
102.5
97.5
1.533
1.083
1.084
1.083
1.085
1.085
1.899
1.557
1.558
1.529
1.092
1.092
1.092
1.094
1.095
1.892
1.592
1.594
1.534
1.091
1.093
1.093
1.094
1.094
1.878
1.561
1.562
1.091
1.091
1.861 ^ 0.007
1.563 ^ 0.008
1.563 ^ 0.008
114.9 ^ 0.2
110.50
110.50
110.50
108.42
108.42
108.42
110.48
110.48
105.78
113.6
112.2
114.5
112.5
111.9
109.4
111.2
111.3
108.1
108.3
108.5
110.9
111.0
106.9
106.7
120.5
102.0
102.4
98.32
115.2
115.1
180.8
Ϫ61.16
176.4
53.98
4.075
0.790
1.012
4.273
2747
109.9
110.8
111.0
108.2
108.4
108.4
111.2
111.3
107.1
107.1
120.2
101.9
101.5
97.94
116.1
115.8
180.7
Ϫ60.08
176.1
55.75
3.920
0.461
1.134
4.106
2672
109.5
111.2
111.2
108.2
108.2
108.4
111.4
111.4
106.0
106.0
120.5
1020
102.0
97.67
115.6
115.6
180.0
180.0
50.4
109.6
111.1
111.0
108.2
108.4
108.4
111.1
111.4
106.5
107.0
120.7
101.9
101.7
98.01
115.6
115.4
180.2
59.8
109.0
110.6
110.5
108.1
108.3
108.2
110.3
110.6
106.4
106.9
119.9
102.3
102.2
97.8
115.8
115.6
181.6
Ϫ57.0
(173.1)
(72.2)
108.99
108.99
110.5 ^ 0.7
110.5 ^ 0.7
106.84
105.78
106.84
119.0 ^ 0.7
101.1 ^ 0.5
101.1 ^ 0.5
97.40
117.31
117.31
180.0
180.0
49.96
Ϫ49.96
119.4 ^ 0.5
102.0 ^ 0.2
102.0 ^ 0.2
98.43
115.94
115.94
180.0
56.94 ^ 0.20
173.78
115.8
115.8
180.0
180.0
(50.4)
(Ϫ50.4)
ՄSPF2
t(PC1C2H1)
t(SPC1C2)
t(F1PC1C2)
t(F2PC1C2)
|ma|
|mb|
|mc|
|mt|
A
170.4
69.4
3.873
0.056
1.232
4.064
Ϫ50.4
3.568
2.080
0.000
4.130
3806
72.34
2687
3874
1664
1564
2712
2118
1625
3874
1664
1564
2712
2118
1625
B
C
1648
1549
2049
1600
1642
1551
2073
1604
1639
1552
2064
1605
Ϫ (E ϩ 1015)
0.851122
34
0.851278
1.726620
92
1.727042
2.256117
91
2.25630
DE (cmϪ1
)
a From Ref. [6].
b Values in parentheses are dependent parameters.

138
J.R. Durig et al. / Journal of Molecular Structure 516 (2000) 131–152
Table 5
Symmetry coordinates for CH3CH2P(O)F2 and CH3CH2P(S)F2
Species Descriptiona
Symmetry coordinateb
A0
CH3 antisymmetric stretch S1 ˆ 2r1Ϫ r2Ϫr3
CH3 symmetric stretch
CH2 symmetric stretch
CH3 antisymmetric
deformation
S2 ˆ r1 ϩ r2 ϩ r3
S3 ˆ r4 ϩ r5
S4 ˆ 2a1Ϫa2Ϫa3
CH3 symmetric
deformation
S5 ˆ a1 ϩ a2
ϩ
a3Ϫb1Ϫb2Ϫb3
CH2 deformation
CH2 wag
C–C stretch
CH3 rock
PF2 stretch
CP stretch
PyO (S) stretch
PF2 wag
PF2 deformation
CCP bend
CPO (S) bend
S6 ˆ 4dϪp1Ϫp2Ϫe1Ϫe2
S7 ˆ p1 ϩ p2Ϫe1Ϫe2
S8 ˆ R
S9 ˆ 2b1Ϫb2Ϫb3
S10 ˆ x1 ϩ x2
S11 ˆ T
Fig. 3. Internal coordinates of ethylposphonic difluoride and ethyl-
phosphonothioic difluoride.
S12 ˆ Q
S13 ˆ f1 ϩ f2Ϫm1Ϫm2
S14 ˆ 4hϪf1Ϫf2Ϫm1Ϫm2
S15 ˆ u
coordinates by:
!
ꢀ
ꢁ
X
2mu
2mu
S16 ˆ g
ˆ
Lij;
A00
CH3 antisymmetric stretch S17 ˆ r2Ϫr3
CH2 antisymmetric stretch S18 ˆ r4Ϫr5
2Qi
2Xj
j
CH3 antisymmetric
deformation
CH2 twist
S19 ˆ a2Ϫa3
where Qi is the ith normal coordinate, Xj the jth Carte-
sian displacement coordinate, and Lij the transforma-
tion matrix between the Cartesian displacement
coordinates and normal coordinates. The infrared
intensities were then calculated by:
S20 ˆ p2Ϫp1Ϫe2 ϩ e1
S21 ˆ b2Ϫb3
CH3 rock
PF2 antisymmetric stretch S22 ˆ x1Ϫx2
CH2 rock
S23 ˆ p2Ϫp1Ϫe2 ϩ e1
"
#
PF2 twist
PF2 rock
CH3 torsion
Asymmetric torsion
S24 ˆ f1Ϫf2Ϫm1 ϩ m2
S25 ˆ f1 ϩ f2 ϩ m1Ϫm2
S26 ˆ t1
ꢀ
ꢁ
ꢀ
ꢁ
ꢀ
ꢁ
2
2
2
2my
Np
3c2
2mx
2mz
Ii ˆ
ϩ
ϩ
:
2Qi
2Qi
2Qi
S27 ˆ t2
The predicted infrared spectra of the trans and
gauche conformers of CH3CH2P(O)F2 and
CH3CH2P(S)F2 are shown in Figs. 1(c) and (d) and
2(c) and (d), respectively, with the mixtures of the
two conformers shown in Figs. 1(b) and 2(b). The
calculated spectra are in reasonably good agreement
with the mid-infrared spectra of the samples dissolved
in liquified krypton (Fig. 1(a)) and xenon (Fig. 2(a)),
which demonstrates the utility of the predicted spectra
for identifying the fundamental modes for the two
different conformers.
a Symmetry coordinates are listed in order for the CH3CH2P(S)F2
molecule.
b Not normalized.
scattering cross sections, dsj=sV; which are propor-
tional to the Raman intensities, can be calculated from
the scattering activities and the predicted wavenumbers
for each normal mode using the relationship [12,13]
0
1
!
!
4
24p4
h
B
B
B
@
C
C
C
A
The evaluation of the Raman activity by using the
analytical gradient methods has been developed
[10,11]. The activity Sj can be expressed as
2sj
sV
ꢀn0Ϫnj†
1Ϫexp
ꢂ
ꢃ
ˆ
Sj;
8p2cnj
Ϫhcnj
45
kT
Sj ˆ gjꢀ45a2j ϩ 7bj2†;
where gj is the degeneracy of the vibrational mode j,
aj the derivative of the isotropic polarizability and bj
that of the anisotropic polarizability. The Raman
where n0 is the exciting wavenumber, nj the vibra-
tional wavenumber of the jth normal mode, and Sj
the corresponding Raman scattering activity. To
obtain the polarized Raman scattering cross sections,

J.R. Durig et al. / Journal of Molecular Structure 516 (2000) 131–152
139
Fig. 4. Raman spectra of ethylphosphonic difluoride: (a) experi-
mental spectrum of the liquid; (b) calculated spectrum of the
mixture of both conformers with a DH of 76 cmϪ1; (c) pure trans;
and (d) pure gauche.
the polarizabilities are incorporated into Sj by
Sj‰ꢀ1Ϫrj†=ꢀ1 ϩ rj†Š where rj is the depolarization
ratio of the jth normal mode. The Raman scattering
cross sections and calculated wavenumbers from the
scaled force constants obtained from the standard
Gaussian program [8] were used together with a
Lorentzian function to obtain the calculated spectra.
The predicted Raman spectra of the pure trans
and gauche conformers of CH3CH2P(O)F2 and
CH3CH2P(S)F2 are shown in Figs. 4(c) and (d) and
5(c) and (d), respectively. In Figs. 4(b) and 5(b) the
mixtures of the two conformers are shown with the
experimentally determined DH values of 76 cmϪ1
with the trans conformer the more stable rotamer
for CH3CH2P(O)F2, and 53 cmϪ1 with the gauche
Fig. 5. Raman spectra of ethylphosphonothioic difluoride: (a)
experimental spectrum of the liquid; (b) calculated spectrum of
the mixture of both conformers with a DH of 53 cmϪ1; (c) pure
trans; and (d) pure gauche.
conformer
the
more
stable rotamer
for
CH3CH2P(S)F2. The experimental Raman spectra of
the liquids are shown in Figs. 4(a) and 5(a) for

140
J.R. Durig et al. / Journal of Molecular Structure 516 (2000) 131–152
comparative purposes and the agreement is quite
good.
In order to gain information about the enthalpy
difference between the two conformers for both of
these molecules, variable temperature studies in liqui-
fied xenon and krypton were carried out. The samples
were dissolved in liquified xenon or krypton and the
spectra were recorded at different temperatures
varying from Ϫ55 to Ϫ150ЊC. Only small interactions
are expected to occur between the dissolved sample
and the surrounding xenon or krypton atoms and,
consequently, only small frequency shifts are antici-
pated when passing from the gas phase to the liquified
noble gas solutions [14–18]. Thus, the resulting
spectra (Figs. 1(a) and 2(a)) should be representative
of the infrared spectra of the vapor phases. A signifi-
cant advantage of this type of temperature study is that
the conformer bands are better resolved in comparison
with those in the infrared spectrum of the gas. This is
particularly important as most of the conformer bands
for these molecules are predicted to be observed
within a few wavenumbers of each other.
The best separated bands in the infrared spectra of
the xenon and krypton solutions for both molecules
are those assigned to the P–C stretch (Figs. 6 and 7).
The higher frequency band in each doublet for each of
these molecules has been confidently assigned to the
gauche rotamer. Ten sets of spectral data were
obtained for the 720/712 cmϪ1 conformer pair of the
CH3CH2P(O)F2 molecule as well as for the CH3 rock
at 1038/1045 cmϪ1. The intensities of the infrared
bands were measured as a function of temperature
and their ratios were determined. By applying the
van’t Hoff equation Ϫln K ˆ (DH/RT)Ϫ (DS/R),
where DS is the entropy change, we have determined
DH from a plot of Ϫln K versus 1/T, where DH/R is
the slope of the line and K is the appropriate intensity
ratio. It is assumed that DH is not a function of
temperature.
4. Conformational stability
There are several fundamentals for both of these
molecules which show conformer doublets in the
infrared and Raman spectra of the fluid phases. For
ethylphosphonic difluoride, the P–C stretches are
observed at 713 (trans) and 719 cmϪ1 (gauche) and
are sufficiently separated to obtain the enthalpy differ-
ence between the conformers. Fundamentals are
predicted for the trans conformer at 491, 414 and
405 cmϪ1 with similar modes for the gauche
conformer predicted at 486, 421 and 398 cmϪ1
,
respectively. However, as only three bands are
observed at 490, 425 and 405 cmϪ1 in the infrared
spectra of the xenon and krypton solutions, it is
concluded that the individual bands are not resolved
for these conformer doublets. The CH2 rocks are also
expected to produce a doublet with the gauche
conformer having the higher frequency. The funda-
mental at 749 cmϪ1 has a high frequency shoulder
and the wavenumber for this mode for the trans
conformer in the infrared spectrum of the solid is
750 cmϪ1. Therefore, we assign the band to the
trans conformer and the shoulder to the gauche
rotamer. The only other region where doublets are
observed is the PyO stretching and CH2 wag, but
these modes have a greater chance of interference
from combination bands.
For ethylphosphonothioic difluoride, the ab initio
calculations predict the separation between the
conformational doublets of the P–C stretch and PyS
stretch to be 15 and 22 cmϪ1, respectively. The P–C
stretch is observed at 785 (trans) and 804 cmϪ1
(gauche) whereas the PyS stretch is observed at 646
(trans) and 619 cmϪ1 (gauche), giving separations of
19 and 27 cmϪ1, respectively. Other fundamentals are
predicted for the trans conformer at 736 and 434 cmϪ1
with corresponding wavenumbers of 745 and
429 cmϪ1 for the gauche modes. However, the
observed bands at 731 and 741 cmϪ1 are weak and
unresolved whereas only one band is observed at
429 cmϪ1. The remaining fundamentals of the two
conformers are near coincident and, thus, cannot be
used for the conformational analysis.
From the plot of the natural logarithm of the ratio of
intensities as a function of the reciprocal of the abso-
lute temperature (Table 6), an average DH of 76 ^
9 cmϪ1 (913 ^ 109 J/mol) with the trans conformer
the more stable form was determined for ethylpho-
sphonic difluoride. Similarly, the pair of conforma-
tional bands at 785 (trans) and 804 (gauche) cmϪ1
belonging to the P–C stretch of ethylphosphonothioic
difluoride (Table 7) gives a DH value of 51 ^ 8 cmϪ1
(610 ^ 96 J/mol), but the gauche conformer is the
more stable form of this molecule. An additional

J.R. Durig et al. / Journal of Molecular Structure 516 (2000) 131–152
141
Fig. 6. Temperature dependent infrared spectra of ethylphosphonic difluoride in the PC stretching region.
Fig. 7. Temperature dependent infrared spectra of ethylphosphonothioic difluoride in the PC stretching region.

142
J.R. Durig et al. / Journal of Molecular Structure 516 (2000) 131–152
Table 6
Table 7
Temperature and intensity ratios for the conformational study of
ethylphosphonic difluoride, CH3CH2P(O)F2 in liquid xenon
Temperature and intensity ratios for the conformational study of
ethylphosphonothioic difluoride, CH3CH2P(S)F2 in liquid xenon
T
(ЊC)
T
1000/T I719/I712 ln k
I1039/I1045 ln k
T
(ЊC)
T
1000/T I785/I804 Ϫln
I619/I646
Ϫln
I619/I646
(K) (K)
(K) (K) 785/I804
I
Ϫ55
Ϫ60
Ϫ65
Ϫ70
Ϫ75
Ϫ80
Ϫ85
Ϫ90
218 4.587 1.592 0.465
213 4.695 1.606 0.474
208 4.808 1.600 0.470
203 4.926 1.570 0.451
198 5.051 1.553 0.440 1.287
193 5.181 1.537 0.430 1.367
188 5.319 1.521 0.420 1.289
183 5.464 1.486 0.396 1.286
178 5.618 1.442 0.366 1.252
173 5.780 1.398 0.335 1.225
78 ^ 7
Ϫ55
Ϫ60
Ϫ65
Ϫ70
Ϫ75
Ϫ80
Ϫ85
Ϫ90
218 4.587 0.395 0.928
213 4.695 0.393 0.933 1.661
208 4.808 0.386 0.952 1.611
203 4.926 0.388 0.946
198 5.051 0.388 0.946 1.592
193 5.181 0.381 0.965
188 5.319 0.383 0.959
183 5.464 0.376 0.978 1.582
178 5.618 0.372 0.990
173 5.780 0.354 1.038 1.496
51 ^ 8
Ϫ0.508
Ϫ0.477
0.252
0.310
0.254
0.251
0.224
0.203
70 ^ 30
Ϫ0.465
Ϫ0.459
Ϫ95
Ϫ95
Ϫ100
Ϫ100
Ϫ0.403
55 ^ 13
DHa (cmϪ1
)
DHa (cmϪ1
)
a Average value is 76 ^ 9 cmϪ1 (913 ^ 109 J/mol) with the trans
conformer the more stable species.
a Average value is 53 ^ 7 cmϪ1 (634 ^ 84 J/mol) with the
gauche conformer the more stable species.
pair of bands (619/646 cmϪ1) which had a much
poorer signal-to-noise ratio was used in the enthalpy
determination of CH3CH2P(S)F2 which gave a value
of 55 ^ 13 cmϪ1 (658 ^ 156 J/mol) with the gauche
conformer more stable. The average DH value for
these two determinations is 53 ^ 7 cmϪ1 (151 ^
19 cal/mol) with the gauche conformer the more
stable rotamer. We also attempted to conduct
temperature studies in liquified krypton, but the
sample began to crystallize at temperatures below
Ϫ115ЊC. Also, there were problems with the decom-
position of the ethylphosphonic difluoride sample in
the liquid xenon and krypton solutions, which lead to
interference with some of the conformer doublets.
Therefore, these conformer pairs such as the ones at
1282/1274 cmϪ1 were omitted.
temperatures and sharpness of the lines attained in the
spectra of xenon and krypton solutions make it easier
to observe the splitting of fundamentals between the
two conformers. Therefore, we will only discuss the
changes in the assignments that result from the obser-
vation of additional fundamentals in the noble gas
solutions and any reassignments of the fundamentals
due to our normal coordinate calculations, which
utilize ab initio force constants from a higher level
of calculation. Complete assignments of the observed
vibrational fundamentals of CH3CH2P(O)F2 and
CH3CH2P(S)F2 are presented in Tables 1 and 2,
respectively, along with the PEDs for the trans and
gauche conformers.
The initial investigation of ethylphosphonic
difluoride utilized RHF ab initio calculations using
the 3-21G(d) and 6-31G(d) basis sets with scaling
factors of 0.9 for the stretching, 0.8 for the bending,
and 1.0 for the torsional modes [4]. The current study
employed the 6-31G(d) basis set with full electron
correlation at the MP2 level [7] and scaling factors
of 0.9 for the stretching and bending coordinates and
1.0 for the torsional modes. In the investigations of
CH3CH2P(O)F2, significant mixing was found for the
normal modes of both the trans and gauche confor-
mers, especially for the low frequency vibrations. As
these modes are strongly mixed, the assignment order
of several fundamentals depends upon the PEDs
resulting from the different levels of the ab initio
5. Vibrational assignment
Complete vibrational assignments were provided in
earlier studies of ethylphosphonic difluoride [4,5] and
ethylphosphonothioic difluoride [6]. However, the
normal coordinate calculations of CH3CH2P(S)F2
were based upon the transfer of force constants from
the methylphosphonothioic difluoride and ethyldi-
fluorophosphine molecules, whereas those for
CH3CH2P(O)F2 utilized ab initio (RHF/3-21G(d))
calculated force constants. Additionally, the reduced

J.R. Durig et al. / Journal of Molecular Structure 516 (2000) 131–152
143
calculation. Thus, PED obtained from MP2/6-31G(d)
calculations indicate a reversal in the assignment of
three pairs of fundamentals compared to the
previously reported ones [4]; the PyO stretch with
the CH3 symmetric deformation, the CC stretch with
the CH3 rock, and the PF2 deformation with that of the
CPO bend. In addition, four bands at 1349, 1038, 972,
and 904 cmϪ1 belonging to the PyO stretch, CH3 rock,
CC stretch, and PF2 antisymmetric stretch of the
gauche conformer, and one band at 1274 cmϪ1
belonging to the CH2 wag of the trans rotamer were
assigned as a result of the improved resolution of the
bands from the infrared spectrum of the sample
dissolved in the liquified noble gas solutions.
For ethylphosphonothioic difluoride, three pairs of
fundamentals were reassigned according to the PED
obtained from MP2/6-31G(d) ab initio calculations. A
reversal in the assignments from that previously
reported [6] is suggested for the CH3 rock with the
CC stretch, the CCP bend with the CPS bend, and the
PF2 rock with the PF2 twist. As a result of the reas-
signments and calculated vibrational frequencies, the
band at 1022 cmϪ1 previously assigned [6] to the CC
stretch of the gauche conformer is now assigned to the
CH3 rock (n22) of this conformer. The PF2 antisym-
metric stretch previously assigned to a single band at
879 cmϪ1 is now assigned in the noble gas solution to
the bands at 873 and 870 cmϪ1 for the trans and
gauche conformers, respectively. Additionally, the
higher resolution of the infrared spectrum of the gas
phase allowed the CH3 rock (n21) of the trans
C1–C2 bonds are in the synperiplanar and anticlinal
positions with respect to the PyO and PyS bonds, the
energies were maximized with respect to the OPC1C2
and SPC1C2 dihedral angels and minimized with
respect to all other structural parameters. In this
way, the gauche to gauche and gauche to trans poten-
tial barriers of 661 cmϪ1 (7.86 kJ/mol) and 863 cmϪ1
(10.27 kJ/mol), respectively, were obtained for the
CH3CH2P(O)F2 molecule with the MP2/6-
311ϩϩG(d,p) basis set (Table 8). Similar barriers
of 631 cmϪ1 (7.51 kJ/mol) and 1068 cmϪ1 (12.71 kJ/
mol) were obtained with the MP2/6-31G(d) basis for
the gauche to gauche and gauche to trans potential
barriers of the CH3CH2P(S)F2 molecule (Table 8). The
potential surfaces obtained with both the MP2/6-
31G(d) and MP2/6-311ϩϩG(d,p) basis sets are
shown in Figs. 8 and 9 for ethylphosphonic difluoride
and ethylphosphonothioic difluoride, respectively.
In order to determine the potential parameters for
the function governing the internal rotation around the
C–P bond experimentally, we fitted the far infrared
transitions of the P(O)F2 and P(S)F2 torsions to a
potential function of the form:
n
X
Vꢀf† ˆ
ꢀVi=2†ꢀ1Ϫcosif†:
iˆ1
The internal rotation constant, F(f), was also repre-
sented as a Fourier series in terms of the torsional
dihedral angle, f:
n
X
Fꢀf† ˆ Fo ϩ
Fi cosif;
conformer to be observed as
a Q-branch at
iˆ1
1030 cmϪ1. The remaining vibrational assignments
are consistent with those previously reported for
CH3CH2P(O)F2 [4,5] and CH3CH2P(S)F2 [6].
where the coefficients, Fi, were determined using the
optimized structural parameters for the trans and
gauche conformers using the ab initio MP2/6-
311G(d) level of calculation for each molecule. The
relaxation of the structural parameters B(a) during the
internal rotation is incorporated into the above equa-
tion by assuming them to be small periodic functions
of the torsional angle of the general type:
6. Discussion
The potential surfaces governing internal rotation
about the P–C bonds of ethylphosphonic difluoride
and ethylphosphonothioic difluoride were obtained
by calculating the molecular energies for a number
of values of the OPC1C2 and SPC1C2 dihedral angles
while keeping all other structural parameters fixed at
the optimized values for the trans conformers of each
molecule, respectively. At the maxima of the potential
surfaces, corresponding to the structures in which the
Bꢀa† ˆ a ϩ b cos a ϩ c sin a:
Values are calculated at 15Њ intervals and the
computer program employed, was developed in our
laboratory.
Four torsional transitions of the trans conformers and
three torsional transitions of the gauche conformers

Table 8
Potential function coefficient v for the asymmetric torsion of ethylphosphonic difluoride and ethylphosphonothioic difluoride
Potential
constants
Ethylphosphonothioic difluoride
Ethylphosphonic difluoride
Ref. [6]
MP2/6-
31G(d)
Fita
MP2/6-
311ϩϩG(d,p)
Fitb
Ref. [4]
MP2/6-
31G(d)
Fita
MP2/6-
311ϩϩG(d,p)
Fitb
V1
V2
V3
V4
Ϫ53 ^ 6
138 ^ 5
744 ^ 26
Ϫ68
Ϫ49
903
Ϫ57 ^ 2
Ϫ21 ^ 3
858 ^ 3
Ϫ7 ^ 1
Ϫ16 ^ 1
Ϫ53 ^ 8
833
Ϫ357
99
886
Ϫ31 ^ 1
Ϫ54 ^ 2
860 ^ 2
15
Ϫ16 ^ 1
Ϫ53 ^ 7
823
Ϫ35 ^ 6
203 ^ 4
661 ^ 1
Ϫ96
93
797
Ϫ7
Ϫ74 ^ 3
164 ^ 4
683 ^ 1
16 ^ 1
Ϫ127
122
796
20
Ϫ70 ^ 3
158 ^ 5
679 ^ 1
17 ^ 1
Ϫ6
121
V6
Ϫ165
Ϫ93
845
Ϫ297
Ϫ103
965
DH (cmϪ1
)
63 ^ 37
819
120 ^ 11
808
Ϫ12
837
72 ^ 8
800
9
872
72 ^ 9
793
c
Trans/gauche
barrier
Gauche/
612
927
854
631
882
506
712
533
661
533
gauche barrier
Gauche/trans
barrier
756
938
886
1068
60.96
876
688
849
728
863
721
Dihedral angle
56.94
60.08
60.10
61.00
57.35
57.10
57.09
57.25
57.23
a Calculated using the assignments in Table 9 and Fourier coefficients (Fi) calculated from the MP2/6-31G(d) geometry.
b Calculated using the assignments in Table 9 and Fourier coefficients (Fi) calculated from the MP2/6-311 ϩ G(d,p) geometry.
c A negative value indicates that the gauche form was calculated to be more stable.

J.R. Durig et al. / Journal of Molecular Structure 516 (2000) 131–152
145
Fig. 8. Potential function of the asymmetric torsion of ethylphosphonic difluoride as determined by the MP2/6-31G(d) basis set (given by the
dashed line) and the MP2/6-311ϩϩG(d,p) basis set (given by the solid line).
were used in the least squares fit (Table 9) for each of
the molecules studied to obtain the first four Vi coeffi-
cients for ethylphosphonic difluoride and the first four
Vi coefficients and the sixth coefficient of ethylpho-
sphonothioic difluoride. Additionally, the enthalpy
differences between the trans and gauche conformers
obtained from the infrared spectra of the samples
dissolved in liquified xenon and the torsional dihedral
angles of the gauche conformers obtained from the ab
initio calculations were also included in the fit. The
results obtained for the potential coefficients and
barriers to internal rotation are summarized in Table
8 and are compared to those obtained from the ab
initio calculations.
Comparisons between the potential functions of the
asymmetric torsions in this study and those previously
reported [4,6] are also of some interest as seen in
Table 8, the calculated potential barriers of
CH3CH2P(O)F2 are very similar to those obtained
previously [4], but the V1 and V2 potential constants
Fig. 9. Potential function of the asymmetric torsion of ethylphosphonothioic difluoride as determined by the MP2/6-31G(d) basis set (given by
the dashed line) and the MP2/6-311ϩϩG(d,p) basis set (given by the solid line).

146
J.R. Durig et al. / Journal of Molecular Structure 516 (2000) 131–152
Table 9
Observed (far infrared data from Ref. [4] for CH3CH2P(O)F2 and Ref. [6] for CH3CH2P(S)F2) and calculated transitions and splittings for the
asymmetric internal rotation in ethylphosphonic difluoride and ethylphosphonothioic difluoride
Conformer
Ethylphosphonic difluoride
Ethylphosphonothioic difluoride
Transition
Observed
Obs.Ϫ Calc.a
Obs.Ϫ Calcb
Transition
Observed
Obs.Ϫ Calc.a
Obs.Ϫ Calcb
Trans
1 ← 0
2 ← 1
3 ← 2
4 ← 3
1 ← 0
2 ← 1
3 ← 2
75.72
73.62
71.64
69.56
67.48
65.18
63.20
Ϫ0.07
Ϫ0.13
0.03
0.18
0.02
Ϫ0.11
Ϫ0.14
0.04
0.22
0.01
1 ← 0
2 ← 1
3 ← 2
4 ← 3
1 ← 0
2 ← 1
3 ← 2
73.25
71.72
69.93
68.15
73.62
72.24
70.82
0.05
0.08
0.00
0.06
Ϫ0.06
Ϫ0.08
Ϫ0.05
Ϫ0.05
0.07
Ϫ0.04
Ϫ0.03
Ϫ0.02
Ϫ0.02
0.04
Gauche
Ϫ0.15
0.12
Ϫ0.15
0.13
a Calculated using the fitted potential constants listed in Table 8 for the MP2/6-31G(d) geometry.
b Calculated using the fitted potential constants listed in Table 8 for the MP2/6-311ϩϩG(d,p) geometry.
are quite different. This is because the ab initio calcu-
lated dihedral angle and the experimental DH values
were used in the earlier fit and they are highly corre-
lated. The enthalpy difference determined from the
noble gas infrared study is considerably lower than
that determined previously [4], and the use of the
calculated dihedral angle essentially fixes the V2 term.
From the potential coefficients of CH3CH2P(S)F2
given in Table 8, it is evident that the V2 and V3
terms from our calculations are at variance with
those previously reported [6]. One notable difference
is that the dihedral angles calculated at the MP2/6-
31G(d) and MP2/6-311ϩϩG(d,p) levels are 3–4Њ
larger and this difference affects the V2 term. In the
previous investigation of the potential function the
authors [6] calculated a large positive V2 term,
whereas in our calculations, a negative V2 term is
obtained. Such a large difference affects the value of
the V3 term and results in the gauche to gauche and
gauche to trans potential barriers being larger than the
previously reported ones [6], and also larger than the
respective barriers determined for the CH3CH2P(O)F2
molecule. Although the electronegativity of the
oxygen atom is higher than that of sulfur, oxygen
also has a smaller van der Waal’s radius. On the
basis of steric interactions, there is an increase in
the steric hindrance as the methyl group eclipses the
sulfur atom when going from one gauche conformer
to the other as opposed to that encountered from the
oxygen atom.
gauche conformers above 1400 cmϪ1 to be relatively
pure modes. However, as predicted from the MP2/6-
31G(d) calculations, many of the vibrational modes
below 1400 cmϪ1 have significant mixing. With the
exception of A00CH3 rocking motion, each funda-
mental of the trans conformer is calculated to have
a contribution of approximately 40% or more from its
descriptive assignment. Many of the fundamental
modes of the gauche conformer, however, are calcu-
lated to consist of complex mixtures with contribu-
tions from several other low wavenumber modes.
The calculated PED of ethylphosphonothioic
difluoride is similar to CH3CH2P(O)F2 in that the
fundamentals above 1200 cmϪ1 are relatively pure
modes, but the mixing of the modes below
1200 cmϪ1 for the trans conformer consist of complex
mixtures like those of the gauche conformer. It is
interesting to note that the PF2 rock and PF2 deforma-
tion of the gauche conformer have a greater percen-
tage of contribution from other vibrations than of their
own descriptive assignment whereas these modes for
the trans conformer are in different species and are
relatively pure. A similar phenomenon occurs in the
gauche conformer of ethylphosphonic difluoride, but
the mixing consists of approximately equal contribu-
tions between the PF2 rocking and PF2 wagging
modes.
The predicted Raman and infrared wavenumbers,
intensities and activities of ethylphosphonic difluoride
are in good agreement with a few exceptions. The
calculated Raman intensities of the CP stretching
and CH2 waging vibrations are lower in intensity
The calculated PED of ethylphosphonic difluoride
indicates all of the vibrational modes of the trans and

J.R. Durig et al. / Journal of Molecular Structure 516 (2000) 131–152
147
with the trans conformer more stable than the gauche
form by 76 ^ 9 cmϪ1 (0.91 ^ 0.11 kJ/mol) whereas
ethylphosphonothioic difluoride has the gauche
conformer more stable in the fluid phase by 53 ^
7 cmϪ1 (0.63 ^ 0.08 kJ/mol) and the trans conformer
the form that predominates in the solid. The ab initio
calculations support the findings for the
CH3CH2P(S)F2 molecule, but the conformational
stability of the CH3CH2P(O)F2 molecule is reversed
in favor of the gauche form in the lower level calcula-
tions (RHF/6-31G(d) and MP2/6-31G(d)), and is not
predicted correctly until the MP2/6-311ϩϩG(d,p)
level is used.
As the predicted energy difference of ethylpho-
sphonic difluoride is small, the correct determination
of the enthalpy difference is necessary for the calcula-
tion of the theoretical spectra (Figs. 1, 2, 4 and 5). In
Fig. 10 the comparison of the experimental spectrum
versus the calculated spectra are shown at two
different values of DH (both having the trans
conformer the more stable form) for the P–C
stretching region of CH3CH2P(O)F2. It is evident
that the calculated spectrum with the previously deter-
mined [4] DH value of 126 cmϪ1 (Fig. 10(c)) reverses
the intensities of the bands, and thus does not corre-
spond to the experimental spectrum.
Fig. 10. Comparison of the effect of the DH value of ethylpho-
sphonic difluoride on the calculation of the theoretical spectra in the
region of the PC Stretch: (a) infrared spectrum of the sample
dissolved in liquid xenon at Ϫ75ЊC; (b) calculated spectrum with
DH value of 76 cmϪ1; (c) calculated spectrum with DH value of
126 cmϪ1. Both the calculated spectra have the trans conformer the
more stable rotamer.
To better understand the conformational stabilities
of these molecules, it is useful to compare their struc-
tural parameters (Tables 3 and 4). The structural para-
meters calculated for the two different conformers of
CH3CH2P(O)F2 are very similar with major differ-
ences being the PCC, PCH, and HC1H angles. For
the trans conformer, the PCC angle is approximately
2.7Њ larger, and the PCH and HC1H angles are 1–2Њ
smaller than the corresponding gauche values. These
differences are also obtained from the microwave data
[5] (Table 3) except for the HC1H angle which was a
fixed parameter. As for the structural differences
between the two conformers of CH3CH2P(S)F2
(Table 4), the ab initio calculations indicate an
opening of the PCC angle by 1–2Њ when going from
the gauche to the trans conformer and smaller differ-
ences are calculated for the PCH and HC1H angels
(0.5–l.1Њ). Similar differences are reported for the
experimental data [6] (Table 4) where these para-
meters vary by 2.5, 1.1, and 0.25Њ for the PCC,
PCH, and HC1H angles, respectively.
than observed in the spectrum of the liquid (Fig. 4).
Also, there are some problems in the CH stretching
region of the calculated spectra where a number of
observed bands in the infrared spectrum are more
intense than predicted and the calculated activities
of the predicted Raman spectrum are significantly
weaker than what was observed.
For ethylphosphonothioic difluoride, the predicted
frequencies and intensities of the infrared spectrum
agree well with those obtained experimentally (Fig.
2). However, the P–C stretch is predicted too weak
when compared to the Raman spectrum of the liquid.
Although there are some differences in the calculated
versus experimental intensities, particularly for the
CH stretching region, these data demonstrate the
utility of ab initio calculations in predicting spectra
for conformer identification as well as for analytical
purposes.
From the experimental data, the ethylphosphonic
difluoridemoleculeexhibitsconformationalequilibrium
The structure of some small molecules containing

Table 10
Structural parameters for F3PO and F3PS
c
d
Parameters
MWa
EDb
RHF/6-
31G(d)
MP2/6-
31G(d)
MP2/6-
311G(d)
MP2/6-
311ϩG(d)
(roadj
)
MP2/6-
311ϩG(2d)
MP2/6-
311ϩG(2df)
MP2/6-
311ϩG(3df)
(roadj
)
F3PO
r(PyO)
r(P–F)
ՄFPF
F3PS
1.437(4)
1.522(4)
101.14(10)
1.436(6)
1.524(3)
101.3(2)
1.425
1.526
100.7
1.458
1.558
100.3
1.446
1.550
100.5
1.449
1.554
100.6
1.441
1.445
1.539
100.6
1.442
1.533
100.8
1.438
1.531
100.8
1.431
1.521e
1.524e
101.3
100.9
r( PyS)
r(P–F)
ՄFPF
1.849(3)
1.538(10)
98.5(20)
1.866(5)
1.538(3)
99.6(3)
1.874
1.535
99.6
1.870
1.569
99.1
1.864
1.560
99.3
1.867
1.565
99.3
1.865
1.881
1.550
99.1
1.868
1.544
99.4
1.860
1.542
99.3
1.861
1.534c
1.535e
99.8
99.5
a From Ref. [19] for F3PO and Ref. [21] for F3PS.
b From Ref. [20] for F3PO and Ref. [22] for F3PS.
c Initial parameters taken from the MP2/6-311 ϩ G(d) calculation.
d Initial parameters taken from the MP2/6-311 ϩ G(3df) calculation.
e Initial values from the RHF/6-31G(d) calculation.

J.R. Durig et al. / Journal of Molecular Structure 516 (2000) 131–152
149
the P(O)F2 and P(S)F2 moieties have already been
determined quite reliably by microwave or electron
diffraction methods [19–29]. It is feasible to compare
the experimentally determined structural parameters
of such compounds with the calculated parameters
with different basis sets and levels of ab initio calcula-
tions. This allows the selection of the most appro-
priate basis sets to reliably predict the geometry of
the larger chemical analogues or to obtain systematic
corrections for specific structural parameters. In Table
10, we have listed the geometric parameters of F3PO
and F3PS as obtained from microwave [19,21] and
electron diffraction [20,22] data together with the
predictions from a variety of basis sets up to MP2/6-
311 ϩ G(3df). The PyO bond lengths are reasonably
predicted at the MP2 level if the triple-split-valence
(6-311) basis sets are used. However, the P–F
distances are calculated too long at the MP2 level
unless an expensive addition of diffuse orbitals
(3df) is used. Actually, the RHF/6-31G(d) calcula-
tion gives a much more reasonable value for the
P–F distance, and we recommend it for predic-
tions of this parameter in P-containing compounds.
The values of the FPF angles are quite good at all
levels of calculation.
Resulting from a limited number of possible
isotopic substitutions in F3PO and F3PS and the C3v
symmetry of the molecules, only the B rotational
constant can be obtained reliably from the microwave
data. This fact leads to a significant correlation
between the structural parameters in the fit of the ro
structure. It is possible, however, to incorporate the
predicted differences among the structural parameters
obtained from ab initio calculations in the determina-
tion of the molecular geometry assuming that the
errors associated with ab initio calculations are mainly
systematic. A computer program [30] developed in
our laboratory consists of an iterative procedure
which starts from the ab initio structures and modifies
them in order to fit the observed rotational constants.
The program restricts the changes of all or a selected
number of equivalent geometric parameters in a way
that keeps the differences among them unchanged as
predicted by the ab initio calculations. In addition to
the fitting of the observed rotational constants, the
program also minimizes the differences between the
initial (ab initio) structural parameters and their
current values during the iterations. However, these
differences are given only a small weight in the
overall fit.
In that sense, the initial values of the structural
parameters influence the outcome of the structural
fit. For example, in Table 10 we have listed ro struc-
tures obtained by the application of the described
method and the microwave rotational constants
[21,27–29] using initial geometries taken from the
MP2/6-311 ϩ G(d) and MP2/6-311 ϩ G(3df) calcula-
tions with the exception of the P–F distance which
was taken from the RHF/6-31G(d) basis set. The final
˚
results show 0.01 and 0.003 A differences in the PyO
and P–F distances, respectively. Thus, the two
obtained geometries can no longer be described as
pure ro structures, but rather as modified or adjusted
ro structures, and to acknowledge this, we have
labeled them with roadj. For the F3PS molecule the
two roadj structures are consistent with each other,
but different from the previously reported [21] micro-
wave structures where the PyS distance is much
shorter. Our result, however, is in agreement with
the electron diffraction data, which should be more
reliable for estimation of the PyS distance. Again,
the importance of good initial parameters cannot be
underestimated. This is especially true for larger
molecules, where the experimentally determined rota-
tional constants are less than the number of geometric
parameters of the molecule. For example, the mole-
cules CH3P(O)F2 and CH3P(S)F2 have 3 and 6 experi-
mentally determined rotational constants, respectively
[23,25,26], and 18 structural parameters. Obviously,
not all geometric parameters can be allowed to change
in the structural fit, and in the previous studies [23,25]
most of the parameters were kept fixed at some
assumed values.
Using the above mentioned method and the
previously reported [23,25,26] rotational constants,
we obtained the roadj structures of CH3P(O)F2 and
CH3P(S)F2 listed in Tables 11 and 12, respectively.
The starting geometries were taken from the MP2/6-
311 ϩ G(d,p) calculations corrected for the shorter P–
F distances from the RHF/6-31G(d) basis set. The
heavy atom parameters are in good agreement with
the previously reported [23,25] ones with the excep-
tion of the P–C bond distances which are 0.025 and
˚
0.013 A shorter for CH3P(O)F2 and CH3P(S)F2,
respectively, in this study. However, with the help
of electron diffraction data [24], it has been shown

150
J.R. Durig et al. / Journal of Molecular Structure 516 (2000) 131–152
Table 11
Structural parameters for CH3P(O)F2
Parameters
MWa (ro)
MWϩEdb (rz)
RHF/6-31G(d)
MP2/6-31G(d)
MP2/6-311ϩG(d,p)
This study (roadj
)
r(P–C)
r(PyO)
r(P–F)
ՄOPC
ՄOPF
ՄFPC
ՄFPF
1.795(19)
1.442(9)
1.544(11)
118.2(15)
117.5(20)
100.5(23)
99.2(10)
1.770(5)
1.444(3)
1.545(2)
117.8(8)
115.0(8)
103.7(8)
99.2(2)
1.786
1.437
1.548
118.4
114.9
103.6
98.8
1.784
1.470
1.581
118.7
115.4
103.0
98.6
1.779
1.461
1.579
119.1
114.9
103.2
98.8
1.770
1.456
1.545
119.2
115.3
102.7
99.1
a From Ref. [23].
b From Ref. [24].
Table 12
Structural parameters for CH3P(S)F2
Parameters
MWa (ro)
RHF/6-31G(d)
MP2/6-31G(d)
MP2/6-311ϩG(d,p)
This study ꢀroadj
†
r(P–C)
r(PyS)
r(P–F)
ՄSPC
ՄSPF
ՄFPC
ՄFPF
1.799(9)
1.877(4)
1.550(2)
119.1(9)
116.2(8)
101.9(4)
98.5(2)
1.794
1.897
1.556
120.0
115.4
102.3
98.3
1.790
1.890
1.591
120.0
116.1
101.6
97.9
1.786
1.886
1.590
120.6
115.6
101.7
98.1
1.786
1.880
1.555
119.9
116.0
102.2
98.1
a From Ref. [25].
Table 13
Rotational constants (MHz) of some RP(O)F2 and RP(S)F2 molecules (R ˆ F and CH3)
Molecule
Rotational
constants
Obs.a
Calc.b
Obs.Ϫ calc.
Ref.
F3P16O
F3P16O
A
B
A
B
B
B
B
B
B
A
B
C
A
B
C
A
B
C
4811.756
4594.263
4811.758
4594.262
4395.287
2657.653
2617.757
2617.757
2579.790
4495.91
4271.86
4125.90
4366.24
2596.38
2496.86
4366.02
2522.84
2428.65
4811.663
4593.454
4811.663
4593.454
4394.295
2657.763
2617.874
2617.874
2579.901
4495.68
4271.65
4123.64
4366.58
2595.84
2496.81
4366.21
2522.30
2728.59
0.093
0.809
0.091
0.808
0.992
0.110
0.123
0.123
0.111
0.23
0.21
2.26
0.34
0.54
0.05
0.19
0.54
0.06
[27]
[19]
F3P18O
F3P32S
[28]
[29]
[21]
[21]
[21]
[23]
F3P32S
F3P33S
F3P34S
CH3P(O)F2
CH3P(32S)F2
CH3P(34S)F2
[26]
[25]
a From microwave data.
b From the roadj structures obtained in this study (Tables 10–12).

J.R. Durig et al. / Journal of Molecular Structure 516 (2000) 131–152
151
Table 14
Rotational constants (MHz) of some CH3CH2P(O)F2 and CH3CH2P(S)F2
Molecule
Rotational
constants
Trans
Gauche
Obs.a
Calc.b
Obs.Ϫ cal.
Obs.a
Calc.b
Obs.Ϫ cal.
12CH312CH2P32SF2
12CD312CD2P32SF2
12CD312CH2P32SF2
12CH312CH2P34SF2
12CD312CD2P34SF2
13CH312CH2P32SF2
12CH313CH2P32SF2
13CH312CH2P34SF2
12CH313CH2P34SF2
CH3CH2P(O)F2
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
3873.71
1664.09
1564.37
3578.82
1510.26
1409.92
3779.09
1526.22
1441.36
3860.10
1624.81
1527.39
3565.00
1475.75
1377.69
3865.15
1628.24
1531.33
3834.16
1656.54
1551.45
3850.82
1590.00
1495.23
3821.71
1617.38
1514.81
4335.81c
2343.936c
2287.766c
3873.54
1664.12
1564.38
3578.80
1510.32
1409.92
3779.12
1526.24
1441.31
3859.71
1624.75
1527.38
3565.01
1475.72
1377.67
3865.00
1628.30
1531.35
3835.36
1656.55
1551.48
3850.60
1589.98
1495.24
3821.46
1617.33
1514.84
4335.867
2344.904
2288.638
0.17
0.03
0.01
0.02
0.06
0.00
0.03
0.02
0.05
0.39
0.06
0.01
0.01
0.03
0.02
0.15
0.06
0.02
1.20
0.01
0.03
0.22
0.02
0.01
0.25
0.05
0.03
0.057
0.968
0.872
2711.62
2117.64
1625.27
2565.33
1874.66
1472.77
2633.10
1934.78
1502.54
2669.27
2088.23
1592.82
2520.58
1852.99
1444.74
2699.43
2069.87
1592.71
2705.08
2093.79
1613.38
2654.25
2043.14
1561.28
2663.03
2064.34
1581.09
4277.596c
2292.464c
2251.988c
2711.65
2117.80
1625.24
2565.25
1874.68
1472.78
2633.26
1934.73
1502.61
2669.34
2088.24
1592.82
2520.50
1852.94
1444.77
2699.48
2070.06
1592.76
2704.85
2093.64
1613.25
2654.39
2043.23
1561.35
2662.91
2064.13
1580.99
4276.830
2244.102
2249.360
0.03
0.11
0.03
0.08
0.02
0.01
0.16
0.05
0.07
0.07
0.01
0.00
0.08
0.05
0.03
0.05
0.19
0.05
0.23
0.15
0.13
0.14
0.09
0.07
0.12
0.21
0.10
0.766
1.638
2.628
a From Ref. [6].
b From the roadj structures obtained in this study (Tables 3 and 4).
c From Ref. [5].
that the P–C distance in CH3P(O)F2 should be about
1.770 ^ 0.005 A which is consistent with our ro
structural fit. Therefore, we applied the described
method to determine more complete roadj structures
for the two conformers of CH3CH2P(O)F2 and
CH3CH2P(S)F2 than the previously reported [5,6] ro
parameters. As only three rotational constants were
available for each conformer of the CH3CH2P(O)F2
compound, we were able to vary only 10 of the 27
parameters of the molecule (Table 3). The rest of the
parameters (mainly carbon–hydrogen distances and
angles) were kept fixed at the ab initio (MP2/6-
311 ϩ G(d,p)) values, which were used as starting
geometries. The abundance of experimental data for
the different isotopomers of CH3CH2P(S)F2 molecule
adj
˚
value for this parameter (Table 11). The fitting of
the rotational constants for all molecules mentioned
above are given in Table 13, and are acceptable in all
cases.
The data given in Tables 10–13 demonstrate that
reasonably good roadj structures can be obtained with
the help of ab initio predictions even when the amount
of experimental data is limited. The MP2/6-
311Gϩ(d,p) calculations with a single correction for
the length of the P–F bond distance from the RHF/6-
31G(d) basis set give excellent initial values for the

152
J.R. Durig et al. / Journal of Molecular Structure 516 (2000) 131–152
[4] J.R. Durig, T.J. Hizer, R.J. Harlan, J. Phys. Chem. 96 (1992)
541.
[5] J.R. Durig, T.K. Gounev, O.A. Braathens, J. Mol. Struct. 355
(1995) 159.
[6] J.R. Durig, R.D. Johnson, H. Nanaie, T.J. Hizer, J. Chem.
Phys. 88 (1988) 7317.
[7] C. Moller, M.S. Plesset, Phys. Rev. 46 (1934) 618.
[8] Gaussian-94 (Revision B. 3), M.J. Frisch, G.W. Trucks, H.B.
Schlegel, P.M.W. Gill, B.G. Johnson, M.A. Robb, J.R. Chee-
seman, T.A. Keith, G.A. Petersson, J.A. Montgomery, K.
Raghavachari, M.A. Al-Laham, V.G. Zakrzewski, J.V.
Ortiz, J.B. Foresman, J. Cioslowski, B.B. Stefanov, A.
Nanayakkara, M. Challacombe, C.Y. Peng, P.Y. Ayala, W.
Chen, M.W. Wong, J.L. Andres, E.S. Replogle, R. Gomperts,
R.L. Martin, D.J. Fox, J.S. Binkley, D.J. Defrees, J. Baker, J.P.
Stewart, M. Head-Gordon, C. Gonzalez, J.A. Pople, Gaussian
Inc., Pittsburgh, PA, 1995.
allowed the elucidation of all 27 geometric parameters
of both conformers (Table 4).
The major disagreement between the results from
this study and the previously reported [5] ones (Table
3) is in the P–C distances of the CH3CH2P(O)F2
˚
compound which now are about 0.02 A shorter. Addi-
tionally, two of the fixed parameters, the C–C and
˚
PyO distances, are about 0.01 A longer in the current
results compared to the previously reported values.
However, for the CH3CH2P(S)F2 compound the agree-
ment is much better (Table 4), because the large
number of rotational constants make the initial struc-
tures less important in the fit. Nevertheless, the P–C
˚
distance of the trans conformer is still 0.016 A longer
in the previous study [6], whereas the PyS distance is
0.016 A longer in the current results. These last two
[9] P. Pulay, Mol. Phys. 17 (1969) 197.
[10] M.J. Frisch, Y. Yamaguchi, J.F. Gaw, H.F. Schaefer III, J.S.
Binkley, J. Chem. Phys. 84 (1986) 531.
˚
discrepancies are not seen in the geometry of the
gauche conformer, but it should be noticed that all
utilized basis sets of ab initio calculations predict
nearly identical values for the P–C and the PyS
distances in the two conformers and this is consistent
with our results. Satisfactory fits of the rotational
constants were achieved for both molecules, and
they are shown in Table 14. These latter results clearly
show that reliable structural parameters can be
obtained for these types of phosphorus molecules by
a combination of ab initio results and a limited
number of rotational constants from microwave
studies.
[11] R.D. Amos, Chem. Phys. Lett. 124 (1986) 376.
[12] P.L. Polavarapu, J. Phys. Chem. 94 (1990) 8106.
[13] G.W. Chantry, in: A. Anderson (Ed.), The Raman Effect, 1,
Marcel Dekker, New York, 1971 chap. 2.
[14] M.O. Bulanin, J. Mol. Struct. 19 (1973) 59.
[15] B.J. van der Veken, F.R. De Munck, J. Chem. Phys. 97 (1992)
3060.
[16] M.O. Bulanin, J. Mol. Struct. 347 (1995) 73.
[17] W.A. Herrebout, B.J. van der Veken, A. Wang, J.R. Durig,
J. Phys. Chem. 99 (1995) 578.
[18] W.A. Herrebout, B.J. van der Veken, J. Phys. Chem. 100
(1996) 9671.
[19] R.H. Kagann, I. Ozier, M.C.L. Gerry, J. Mol. Spectrosc. 71
(1978) 281.
[20] T. Moritani, K. Kuchitsu, Y. Morino, Inorg. Chem. 10 (1971)
344.
[21] J.G. Smith, I. Thompson, Mol. Phys. 32 (1976) 1247.
[22] K. Karakida, K. Kuchitsu, Inorg. Chem. Acta 16 (1976) 29.
[23] J.R. Durig, A.E. Stanley, Y.S. Li, J. Mol. Struct. 78 (1982)
247.
[24] S.R. Carlowitz, H. Oberhammer, J. Mol. Struct. 178 (1988)
255.
[25] K.K. Chatterjee, J.R. Durig, J. Mol. Struct. 351 (1995) 25.
[26] J.R. Durig, J.A. Meadows, Y.S. Li, A.E. Stanley, Inorg. Chem.
22 (1983) 4134.
[27] C. Styger, I. Ozier, A. Bauder, J. Mol. Spectrosc. 153 (1992)
101.
Acknowledgements
JRD would like to acknowledge partial support of
these studies by the University of Missouri-Kansas
City Faculty Research Grant program.
References
[1] J.R. Durig, J.B. Robb II, J. Mol. Struct. 413 (1997) 371.
[2] J.R. Durig, J.S. Church, C.M. Whang, R.D. Johnson, B.J.
Streusand, J. Phys. Chem. 91 (1987) 2769.
[28] J.G. Smith, Mol. Phys. 32 (1976) 621.
[29] H. Bu¨rger, U. Goergens, H. Ruland, M. Quack, U. Schmitt,
Mol. Phys. 87 (1996) 469.
[3] P. Groner, J.S. Church, Y.S. Li, J.R. Durig, J. Chem. Phys. 82
(1985) 3894.
[30] B.J. van der Veken, W.A. Herrebout, D.T. Durig, W. Zhao,
J.R. Durig, J. Phys. Chem. 103 (1999) 1976.