Infrared and Raman spectra, conformational stability, ab initio calculations, and vibrational assignments for methylaminothiophosphoryl difluoride

Article abstract of DOI:10.1016/S0022-2860(00)00759-6

Infrared spectra (4000-50 cm^-1) of the vapor, amorphous and crystalline solids and Raman spectra (3600-10 cm^-1) of the liquid with qualitative depolarization data as well as the amorphous and crystalline solids of methylaminothiophosphoryl difluoride, CH₃N(H)P(=S)F₂, and three deuterated species, CD₃N(H)P(=S)F₂, CH₃N(D)P(=S)F₂, and CD₃N(D)P(=S)F₂, have been recorded. The spectra indicate that in the vapor, liquid and amorphous solid a small amount of a second conformer is present, whereas only one conformer remains in the low temperature crystalline phase. The near-infrared spectra of the vapor confirms the existence of two conformers in the gas phase. Asymmetric top contour simulation of the vapor shows that the trans conformer is the predominant vapor phase conformer. From a temperature study of the Raman spectrum of the liquid the enthalpy difference between the trans and near-cis conformers was determined to be 368 ± 15 cm^-1 (4.4l ± 0.2 kJ/mol), with the trans conformer being thermodynamically preferred. Ab Initio calculations with structure optimization using the 6-31G(d) and 6-311+G(d,p) basis sets at the restricted Hartree-Fock (RHF) and/or with full electron correlation by the perturbation method to second order (MP2) support the occurrence of near-trans (～5° from trans) and near-cis (～20° from cis) conformers. From the RHF/6-31G(d) calculation the near-trans conformer is predicted to be the more stable form by 451 cm^-1 .(5.35 kJ/mol) and from the MP2/6-311+G(d,p) calculation by 387 cm^-1 (4.63 kJ/mol). All of the normal modes of the near-trans rotamer have been assigned based on infrared band contours, depolarization values and group frequencies and the assignment is supported by the normal coordinate calculation utilizing harmonic force constants from the MP2/6-31G(d) ab initio calculations.

Full text of DOI:10.1016/S0022-2860(00)00759-6

Journal of Molecular Structure 560 (2001) 57±86
www.elsevier.nl/locate/molstruc
Infrared and Raman spectra, conformational stability, ab initio
calculations, and vibrational assignments for
methylaminothiophosphoryl di¯uorideq
B.J. van der Vekena, R.S. Sandersa, J. Xiaob, J.R. Durigb,
*
aRijksuniversitair Centrum Antwerpen, 171 Groenenborgerlaan, Antwerpen 2020, Belgium
bDepartment of Chemistry, University of Missouri-Kansas City, Kansas City, MO 64110 USA
Received 24 July 2000; accepted 21 August 2000
Abstract
Infrared spectra (4000±50 cm21) of the vapor, amorphous and crystalline solids and Raman spectra (3600±10 cm21) of the
liquid with qualitative depolarization data as well as the amorphous and crystalline solids of methylaminothiophosphoryl
di¯uoride, CH3N(H)P(ˆS)F2, and three deuterated species, CD3N(H)P(ˆS)F2, CH3N(D)P(ˆS)F2, and CD3N(D)P(ˆS)F2,
have been recorded. The spectra indicate that in the vapor, liquid and amorphous solid a small amount of a second conformer
is present, whereas only one conformer remains in the low temperature crystalline phase. The near-infrared spectra of the vapor
con®rms the existence of two conformers in the gas phase. Asymmetric top contour simulation of the vapor shows that the trans
conformer is the predominant vapor phase conformer. From a temperature study of the Raman spectrum of the liquid the
enthalpy difference between the trans and near-cis conformers was determined to be 368 ^ 15 cm21 (4.41 ^ 0.2 kJ/mol), with
the trans conformer being thermodynamically preferred. Ab Initio calculations with structure optimization using the 6-31G(d)
and 6-3111G(d,p) basis sets at the restricted Hartree±Fock (RHF) and/or with full electron correlation by the perturbation
method to second order (MP2) support the occurrence of near-trans (,58 from trans) and near-cis (,208 from cis) conformers.
From the RHF/6-31G(d) calculation the near-trans conformer is predicted to be the more stable form by 451 cm21 (5.35 kJ/mol)
and from the MP2/6-3111G(d,p) calculation by 387 cm21 (4.63 kJ/mol). All of the normal modes of the near-trans rotamer
have been assigned based on infrared band contours, depolarization values and group frequencies and the assignment is
supported by the normal coordinate calculation utilizing harmonic force constants from the MP2/6-31G(d) ab initio calcula-
tions. q 2001 Elsevier Science B.V. All rights reserved.
Keywords: Infrared spectra; Raman spectra; Restricted Hartree±Fock
1. Introduction
q
Taken in part from the thesis of R.S. Sanders which was
For some time we have been investigating the confor-
submitted to the Department of Chemistry, University of Antwerp
mational stability of a number of organophosphorus
molecules [1±15]. For several of these molecules we
have found that the more stable conformer is not the one
that would be predicted from the `gauche effect' [16].
At this time, there does not appear to be any simple
theory, which will correctly predict the more stable
(Belgium), in partial ful®llment of the PhD degree and the thesis of
J. Xiao, which will be submitted to the Department of Chemistry of
the University of Missouri-Kansas City in partial ful®llment of the
PhD degree.
* Corresponding author. Tel.: 11-816-235-1136; fax: 11-816-
235-5502.
E-mail address: durigj@umkc.edu (J.R. Durig).
0022-2860/01/$ - see front matter q 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-2860(00)00759-6

58
B.J. van der Veken et al. / Journal of Molecular Structure 560 (2001) 57±86
Fig. 1. The cis and trans conformers of methylaminothiophosphoryl
di¯uoride.
conformer for these simple organophosphorus mole-
cules. In order to provide additional information on
the conformational stability of organophosphorus mole-
cules a study of methylaminothiophosphoryl di¯uoride,
CH3N(H)P(ˆS)F2 was initiated. There has been a
previously reported [17] infrared study and, in this
study, the mid-infrared spectra of the vapor phase
(4000±600 cm21) and the solid (4000±350 cm21)
were recorded. The authors concluded that the cis
form (Fig. 1) is the more stable rotamer in the gas
whereas they could not make a conformational assign-
ment for the molecule in the solid state. In order to
expand the vibrational and structural data available on
this molecule and to obtain more conformational infor-
mation we have recorded the infrared and Raman
spectra of methyl aminothiophosphoryl di¯uoride and
its three isotopomer derivatives, CH3N(D)P(ˆS)F2,
CD3N(H)P(ˆS)F2 and CD3N(D)P(ˆS)F2, and have
compared the data among the three physical states.
Additionally, ab initio calculations using both the 6-
31G(d) and 6-3111G(d,p) basis sets have been carried
out at the restricted Hartree±Fock (RHF) and/or with
full electron correlation by the perturbation method to
second order (MP2) [18] to obtain the conformational
stability, structural parameters, and fundamental
frequencies for comparison to the experimental values.
The results of this spectroscopic and theoretical study
are reported herein.
Fig. 2. Mid infrared spectra of gaseous (A); amorphous solid (B);
and crystalline solid (C) CH3N(H)P(S)F2.
phase reaction of methylamine with chlorothiophos-
phoryldi¯uoride (2:1 molar ratio). These compounds
were mixed and after a reaction time of one to two
hours, the mixture was transferred to a low tempera-
ture, low pressure fractional distillation column for
puri®cation. The isotopic species were prepared in
the same manner by using CH3ND2, CD3NH2 or
CD3ND2 as the starting materials. Methylamine-d3
was obtained by the reaction of CD3NH2HCl (Janssen
Chimica) with NaOH in H2O solution. Methylamine-
d2 and methylamine-d5 were prepared by several
successive exchanges of methylamine and methyl-
amine-d3, respectively, with concentrated NaOD in
D2O solution. The different isotopic methylamines
were puri®ed by vacuum distillation and checked for
isotopic purity by their vapor phase infrared spectra.
Chlorothiophosphoryldi¯uoride was prepared by
partial ¯uorination of thiophosphoryl trichloride
with freshly sublimed antimony tri¯uoride (Janssen
Chimica), followed by vacuum fractionation.
2. Experimental
Methylaminothiophosphoryl di¯uoride, CH3N(H)P
(yS)F2, was prepared according to the procedure of.
Cavell et al. [17]. This method is based on the vapor

B.J. van der Veken et al. / Journal of Molecular Structure 560 (2001) 57±86
59
Fig. 4. Mid infrared spectra of gaseous (A); amorphous solid (B);
and crystalline solid (C) CH3N(D)P(S)F2.
Fig. 3. Mid infrared spectra of gaseous (A); amorphous solid (B);
and crystalline solid (C) CD3N(H)P(S)F2.
the vapor onto a cooled Si-window, using the Bruker
model IFS 113v spectrometer equipped with Mylar
beamsplitters, a Globar source, a DTGS detector and
a liquid helium cooled Si bolometer. A total of 100
interferograms were co-added and Happ±Genzel
apodized so as to obtain a resolution of 0.5 cm21. The
far-infrared spectra of the vapors were obtained on this
same instrument with the sample contained in a 30 cm
cell equipped with polyethylene windows. The spectral
data were treated the same as for the solid.
The Raman spectra (3600±10 cm21) (Figs. 10±13)
were recorded on a SPEX model 1403 Raman spectro-
meter equipped with a Spectra-Physics model 2020
argon ion laser operating on the 514.5 nm line. The
laser power at the samples was varied from 0.5 to
1.5 W depending on the physical state under investi-
gation. The spectra of the liquids were obtained from
the samples sealed in a glass capillary, and variable
temperature experiments in this phase were carried
out in a Miller±Harney cell [19]. The solid samples
The mid-infrared spectra (Figs. 2±5) of the
vapors were obtained on a Bruker model IFS 113v
Fourier transform spectrometer equipped with a Ge/
KBr beamsplitter, a Globar source and a broad band
MCT detector. The spectra were recorded at a
different vapor pressure up to 10 mbar in a 30 cm
cell ®tted with KBr windows. The interferograms
were recorded 250 times with a resolution of
0.5 cm21 and transformed with a Happ±Genzel
apodization function. The corresponding spectra of
the solids were recorded on the same instrument by
condensing the samples on a CsI plate held at
boiling liquid nitrogen temperature. The amorphous
samples obtained in this way were annealed until no
further changes in the spectra were observed.
The far-infrared spectra (Figs. 6±9) of the solids
were obtained by condensing the compound from

60
B.J. van der Veken et al. / Journal of Molecular Structure 560 (2001) 57±86
Fig. 6. Far infrared spectra of gaseous (A); amorphous solid (B); and
crystalline solid (C) CH3N(H)P(S)F2.
Fig. 5. Mid infrared spectra of gaseous (A); amorphous solid (B);
and crystalline solid (C) CD3N(D)P(S)F2.
were obtained by condensation from the liquids (in a
sealed glass capillary) and annealing, using a
Miller±Harney cell, or by condensing the samples
onto a blackened brass block maintained at about
77 K by boiling liquid nitrogen. Depolarization
measurements were made with the standard SPEX
accessories. The spectra were measured with a resolu-
tion of 2.0 cm21. Because of the relatively low vapor
pressure of the compounds, no vapor phase Raman
spectra could be recorded. The observed infrared
and Raman bands are listed in Tables 1±4.
The near-infrared spectra of the vapor phase
were also recorded with the Bruker IFS 113v
spectrometer, equipped with a Si/CaF2 beamsplitter,
a Globar source and a broad band MCT detector.
For the gas phase the sample was examined by using
the maximum vapor pressure of approximately
10 mbar in a 10 m cell ®tted with quartz windows.
Interferograms were recorded 1000 times with a
resolution of 0.5 cm21 and transformed with a
Happ±Genzel apodization function.
3. Ab initio calculations
Because there are no previously reported experi-
mental data with regard to the structure of methyla-
minothiophosphoryl di¯uoride, the initial molecular
structure was determined by ab initio calculations.
The ab initio calculations were made with the
program Gaussian-98 [20] at the restricted
Hartree±Fock (RHF) level and with full electron
correlation (second order Moller±Plesset perturbation
theory) using Gaussian type basis functions with the
6-31G(d) and/or 6-3111G(d,p) basis sets. Density
function theory (DFT) calculations were also made
with Gaussian-98 [20], which were restricted to the
hybrid B3LYP method with the 6-31G(d) basis set.
The energy minima with respect to the nuclear coor-
dinates were obtained by the simultaneous relaxation
of all the geometric parameters using the gradient
method of Pulay [21]. All the calculations converged
on the near-trans and near-cis conformers, instead of
the perfect trans and cis conformer (Fig. 1). The

B.J. van der Veken et al. / Journal of Molecular Structure 560 (2001) 57±86
61
Fig. 7. Mid infrared spectra of gaseous (A); amorphous solid (B);
and crystalline solid (C) CD3N(H)P(S)F2.
Fig. 8. Mid infrared spectra of gaseous (A); amorphous solid (B);
and crystalline solid (C) CH3N(D)P(S)F2.
structural parameters as determined with these calcu-
lations are given in Table 5 for both conformations. It
should be noted that for the near-trans conformer, the
dihedral angle SPNC ranged from 172.88 from the
MP2/6-31G(d) calculation to 174.88 from the RHF/
6-31G(d) calculation. Such a small dihedral angle
away from the planar structure makes this conformer
effectively trans in the spectroscopic study. In order to
assist our experimental study, calculations were
carried with the SPNC dihedral angle ®xed at 1808.
The force ®elds which resulted were utilized for the
normal coordinate analysis stated below.
In order to obtain a more complete description of
the molecular motions involved in the normal
modes, we have also carried out a normal coordi-
nate analysis. This analysis was performed utilizing
ab initio calculations and the classic Wilson GF
matrix method [22] with the computer programs
written by Schachtschneider [23]. The force ®elds
in Cartesian coordinates were calculated by the
Gaussian-98 program [20] with the MP2/6-
31G(d) basis set. Internal coordinates (Fig. 14)
were used to calculate the B matrix and using this
B matrix, the force ®eld in Cartesian coordinates
was then converted to the force ®eld in internal
coordinates [24]. The resulting force ®eld can be
obtained from the authors. Initially, all scaling
factors were kept ®xed at a value of 1.0 to repro-
duce the pure ab initio calculated vibrational
frequencies. Subsequently, scaling factors of 0.88
for carbon±hydrogen stretches, 0.90 for carbon±
hydrogen bends, heavy atom stretches and methyl
torsions and 1.0 for heavy atom bends and asym-
metric torsion, and the geometric average of scaling
factors for interaction force constants were used to
obtain the ®xed scaled force ®eld and resultant
wavenumbers. A set of symmetry coordinates
were used (Table 6) to determine the corresponding
potential energy distributions (PEDs). A comparison
between the observed and calculated infrared inten-
sities, Raman activities, depolarization ratios and
PEDs are given in Tables 7±10.

62
B.J. van der Veken et al. / Journal of Molecular Structure 560 (2001) 57±86
Fig. 9. Mid infrared spectra of gaseous (A); amorphous solid (B);
and crystalline solid (C) CD3N(D)P(S)F2.
Fig. 10. Raman spectra of liquid (A); amorphous solid (B); and
crystalline solid (C) CH3N(H)P(S)F2.
4. Conformational analysis and stability
pattern is observed for the ®rst overtone of n(NH) in
the d3 isotopomer, and in that of n(ND) in d1
molecule. Therefore, we conclude that the doublet
structure is not due to Fermi resonance, but is due to
the presence of a second conformer. In contrast to the
patterns observed for d0, d1 and d3, the ®rst overtone of
n(ND) in d4 shows a triplet structure. Hence, in one or
both conformers this vibration must be in Fermi reso-
nance with another overtone. No attempts were
made to give a detailed assignment of these three
transitions.
In a series of papers [6±8,10±12] we have shown
that infrared band contour simulation can be used to
identify the conformers present in the gaseous state.
The a priori knowledge of the orientation of a 2m=2Q
for the simulation of gas phase contours restricts the
fundamentals that can be used to those that are suf®-
ciently characteristic, whereas also the contours
calculated for the proposed conformers must differ
suf®ciently from each other to allow for conclusive
conformational results to be drawn.
As suggested by the ab initio calculations, we can
expect the existence of two conformers, the near-cis
and trans rotamers, in the ¯uid phases. The vapor
phase infrared spectra (Figs. 2±5) of the four isotopic
derivatives indeed show the appearance of a weaker
second band in the PS stretching region, which can
only be assigned to a second conformer. This is
clearly substantiated by the low temperature study
where the PS stretch of the second conformer disap-
pears upon crystallization, for each of the four
isotopic species investigated.
As will be shown below, n(NH) and n(ND) are
observed as a doublet in the ¯uid phases of each of
the isotopic derivatives. If this doublet structure were
due to Fermi resonance, it is unlikely that the ®rst
overtone of n(NH) and n(ND) would give rise to a
similar doublet structure. In Fig. 15, the fundamental
and the ®rst overtone of n(NH) in the infrared spec-
trum of the d0 vapor are shown. It is clear that the ®rst
overtone also occurs as a doublet. A very similar
For the present compound, n(NH) and n(ND) are

B.J. van der Veken et al. / Journal of Molecular Structure 560 (2001) 57±86
63
Fig. 12. Raman spectra of liquid (A); amorphous solid (B); and
crystalline solid (C) CH3N(D)P(S)F2.
Fig. 11. Raman spectra of liquid (A); amorphous solid (B); and
crystalline solid (C) CD3N(H)P(S)F2.
highly characteristic vibrations. However, the
contours observed experimentally for these vibrations
are strongly thermally broadened, so that their P, Q
and R structure has almost completely disappeared.
Therefore, these contours cannot be used to abstract
conformational information. In addition to not
showing conformational splitting, n(CH3) is also
found to be in strong Fermi resonance. Both
phenomena are indications against the use of the
ns(CH3) contour in simulations. Of the other funda-
mentals, only n(PyS) shows a well de®ned conformer
doublet (vide infra). Hence this vibration, which has
been successfully used in simulations before [11], was
selected.
The pure type A, B and C infrared contours of the
trans conformers of CH3N(H)P(yS)F2 and its d1, d3
and d4 isotopomers were calculated using the rota-
tional constants obtained from the microwave data
[25]. Those for the near-cis conformer were calculated
using the rotational constants obtained by assuming
for the near-cis conformer the bond lengths and
valency angles from Table 5. All pure type contours
were calculated to a maximum J level of 150, using a
slit of 0.5 cm21. With the assumption that the direc-
tion of 2m=2Q for the n(PyS) fundamental is parallel
to the PyS bond, the directional cosines for 2m=2Q in
the principal axis system were calculated. These are
given in Table 11. The directional cosines were used
to obtain the hybrid contours for the corresponding
fundamentals as described earlier [26]. Additional
convolutions, with Gaussian or Lorentzian functions,
of the theoretical hybrids were necessary to account
for thermal effects. The reproduction of the full width
at half height of the Q branch was used as a criterion
for the slit width used in the additional convolutions.
As can be seen from Table 11, the n(PyS) contour for
the trans conformer is almost pure type A, whereas
the hybrid for the cis conformer is expected to show
an appreciable amount of B type character.
In Fig. 16 the experimental contour of the n(PyS)
doublet (B) is compared to the contours simulated
with the high frequency component assigned to either

64
B.J. van der Veken et al. / Journal of Molecular Structure 560 (2001) 57±86
metry is not observed in the d0 compound, it is
unlikely to be due to anharmonicity of the n(PyS).
From the infrared spectra of the d1 molecule it can be
seen, however, that this compound contains an appre-
ciable amount of the d0 compound as isotopic
impurity. The latter shows its n(PyS) vibrations
only slightly shifted from those in the d1 compound.
By admixing an amount of contour C of Fig. 16, at the
correct frequencies, to account for the isotopic
impurity, the observed asymmetry is reproduced, as
is shown by contour C of Fig. 17. Also, in this way the
low frequency component is correctly reproduced,
con®rming the origin of the phenomena.
In order to determine the enthalpy difference
between the trans and near-cis conformers, a
variable temperature Raman study of the liquid
phase was undertaken utilizing an experimental
technique similar to that described by Miller and
Harney [19]. As can be seen from the Raman
spectra in Fig. 18, the lines assigned to the PS
stretch of the trans and near-cis conformers show
signi®cant intensity changes upon cooling. Since
these lines are very strong and well separated
from each other and from other lines, they were
selected for the temperature study.
Fig. 13. Raman spectra of liquid (A); amorphous solid (B); and
crystalline solid (C) CD3N(D)P(S)F2.
For each of the isotopes at a series of temperatures
between 77 and 21388C, this doublet was recorded
(Table 12). A least squares curve ®tting of these
spectra, using Gauss/Lorentz sum bands to ®t the
observed curves, allowed the relative intensities of
the two bands to be expressed as band areas. From
these data, DH was found to be 354 ^ 13 cm21
(4.24 ^ 0.16 kJ/mol) in d0, 404 ^ 18 cm21 (4.83^
0.22 kJ/mol) in d3, 369 ^ 15 cm21 (4.42 ^ 0.18 kJ/
mol) in d1 and 368 ^ 19 cm21 (4.40 ^ 0.23 kJ/mol)
in the d4 derivative. The average value of
368 ^ 15 cm21 (4.41 ^ 0.18 kJ/mol) clearly indicates
that the trans conformer is the more stable form in the
liquid state.
Assuming the absorption coef®cient for n(PyS)
to have comparable values in both conformers,
and making the usual assumption of a negligible
entropy difference between the trans and near-cis
conformer, the vapor phase infrared relative
intensities of the n(PyS) doublet suggest that in
the vapor phase the trans conformer is even more
stable relative to the cis than in the liquid phase.
The Raman spectrum of the crystalline solid is also
the near-cis conformer (A) or the trans conformer (C)
and vice versa for the low frequency component. It is
clear that from the simulation of the low frequency
component no conclusion can be drawn. However, the
simulation of the high frequency component with the
near-cis hybrid leads to a split Q branch, clearly
absent in the experimental contour. The agreement
of the experimental high frequency contour with the
one simulated using the trans conformer con®rms that
the intense high frequency n(PyS) component must
be due to the trans conformer. A similar result is
obtained from the simulations of the n(PyS) doublet
in the d3 molecule.
In Fig. 17 the experimental contour of the n(PyS)
doublet in the d1 derivative (B) is compared to the
contours simulated with the high frequency compo-
nent assigned to the near-cis (A) or the trans (D)
rotamer. It is clear again that simulation A is to be
rejected. Simulation D, however, does not reproduce
the asymmetry in the P and R branch of the experi-
mental high frequency component. As the same asym-

Table 1
Observed infrared and Raman frequencies ( cm21) and vibrational assignments for CH3N(H)P(S)F2 (abbreviations used: v, very; s, strong; m, moderate; w, weak; br, broad; sh,
shoulder; p, polarized; R, Q and P refer to band contours in the gas phase infrared spectrum)
Infrared
Gas
Raman
Assignments
a
Rel. int. Amorphous Rel. int. Crystalline Rel. int. Liquid Rel. int. Amorphous Rel. int. Crystalline Rel. int. ni
3397
3391
3325
w
w
m
crystal effect
3390
3328
vw
w
crystal effect
3480
3446
3023
m
3328
s
3352
p,br vw 3332
vw
n1
NH stretch
vw
vw
n 0
n2
NH stretch (near-cis)
CH3 antisymmetric stretch
1
3013
2999
2956
2936
vw
vw
vw
vw
3012
3000
2960
2939
vw
vw
w
2989
2957
2934
2900
vw
vw
w
2993
2956
sh
2992
2956
2936
w
w
w
n16 CH3 antisymmetric stretch
n3
2963
2925
m
p m
CH3 symmetric stretch
vw
vw
vw
2909
2844
sh
2852 R
2847 Q
2842 P
w
2840
vw
2843
vw
p w
2841
w
2844
w
Fermi resonance
1478
1454
1433
m
1477
1452
1438
1394
1389
1376
1126
w
1470
1460
1436
sh
1478
1454
1436
vw
vw
vw
1475
1457
1439
vw
vw
vw
n4 CH3 antisymmetric deformation
n17 CH3 antisymmetric deformation
w
vw
vw
vw
m
vw
vw
p vw
n5
n6
n8
CH3 symmetric deformation
crystal effect
1391
1160
1113
s
1397
1138
1110
911
s
1394
1124
1107
913
vw
1399
1129
1102
914
vw
vw
vw
vw
vw
1373
vw
NH deformation
crystal effect
vw
s
sh
vs
sh
vs
vs
vs
p w
p vw
p w
p w
1130
1124
1102
sh
CN stretch
vw
vw
n18 CH3 out-of-plane rock
n7
1105
934
915
878
863
s
CH3 rock
sh
m
sh
vs
923
877
867
vs
vs
sh
927
881
867
vw
vw
vw
n9
PF2 symmetric stretch
n19 PF2 antisymmetric stretch
n10 NP stretch
865
866
867
704 R
699 Q
695 P
m
686
s
684
680
677
s
684
p vs
679
vs
687
675
m
n11 PS stretch
m
m
vs
657 R
651 Q
645 P
w
638
500
458
w
640
457
p s
p s
638
456
vw
m
n 0
PS stretch (near-cis)
11
sh br
vs
500
458
sh bv
vs
n21 NH out-of-plane bend
n12 PF2 wag
465 R
461 Q
vs
455
m

Table 1 (continued)
Infrared
Raman
Rel. int. Amorphous Rel. int. Crystalline Rel. int. Liquid Rel. int. Amorphous Rel. int. Crystalline Rel. int. ni
Assignments
a
Gas
456 P
440
sh
vs
vs
440
416
390
vw
vs
s
n 0
CNP bend (near-cis)
14
418 Q
414 Q
400 R
392Q
417
vs
415
389
p m
vw
416
391
m
417
376
m
n20 PF2 rock
vw
n 0
PF2 symmetric deformation
(near-cis)
13
385 P
378
350
vs
m
vw
n13 PF2 symmetric deformation
n21 NH out-of-plane bend
348
300
sh
s
345
vw
336 R
325 Q
305 R
299 Q
w
w
n 0
PF2 rock (near-cis)
20
302
298
257
253
s
299
254
p m
w
301
255
182
m
302
297
257
252
w
w
w
w
n14 CNP bend
sh
m
m
crystal effect
242 br vw
233 br vw
253
240
vw
sh
w
n22 PF2 twist
n 0
n 0
PF2 twist (near-cis)
NPS bend (near-cis)
22
15
194
182
vw
vw
180
95
vw
vw
180
135
112
vw
w
vw
180
92
vw
vw
vw
n15 NPS bend
n23 CH3 torsion
n24 asymmetric torsion
m
95
66
52
w
w
m
81
49
34
s
s
lattice modes
ni0 refer to the near-cis conformer.
a

Table 2
Observed infrared and Raman frequencies ( cm21) and assignments for CD3N(H)P(S)F2 (abbreviations used: v, very; S, strong; m, moderate; w, weak; br, broad; sh, shoulder; p,
polarized; R, Q and P refer to band contours in the gas phase infrared spectrum)
Infrared
Gas
Raman
Assignments
a
Rel. int. Amorphous Rel. int. Crystalline Rel. int. Liquid Rel. int. Amorphous Rel. int. Crystalline Rel. int. ni
3393
3318
m
3391
3330
vw
vw
crystal effect
3482
3447
m
3328
s
vs
3357
p brw
3333
2968
br w
vw
n1
NH stretch
vw
n1
NH stretch (near-cis)
3013
2991
2972
vw
vw
vw
2968
vw
vw
2965
2841
p vw
p vw
2925
2852 R
2847 Q vw
2842 P
2845
vw
2265
w
2260
2243
sh
2261
2253
2243
2226
2206
2133
2114
2087
1388
1380
1364
vw
vw
vw
vw
sh
2255
2249
sh
w
2261
2253
2243
w
n2
CD3 antisymmetric stretch
CD3 antisymmetric stretch
vw
vw
2241
vw
vw
2241
br w
n16
2206
2133
2112
2087
vw
w
2140
2093
1382
w
2133
2113
2085
vw
ww
w
vw
vw
w
2133
2087
1374
p w
p m
vw
2132
2113
2085
w
Fermi resonance
vw
m
vw
w
m
vs
n3
n6
CD3 symmetric stretch
crystal effect
m
1388
m
m
1377
vw
1377
1362
vw
vw
NH deformation
crystal effect
m
1138 R
1133 Q
1127 P
1082
s
s
1124
1098
s
1125
1120
1102
s
1119 p, vw
1121
1101
vw
vw
1118
1100
1061
vw
vw
vw
n5
CD3 symmetric deformation
sh
m
m
1093
1063
vw
vw
n4n8
CD3 antisymmetric
deformation, CN stretch
CD3 antisymmetric deformation
1066
1054
vw
vw
1065
vw
1064
1056
vw
sh
n17
n 017,n 0 CD3 symmetric deformation
(near-cis)
5
CD3 antisymmetric deformation
(near-cis)
973
965
m
m
973
s
966
m
967
p w
966
vw
969
905
893
874
859
w
n7
CD3 rock
vw
vw
vw
vw
913
876
vs
vs
887
857
vs
vs
891
874
w
905
877
sh
914
881
864
sh
n9
PF2 symmetric stretch
PF2 antisymmetric stretch
CD3 out-of-plane rock
vs
br vw
vw
sh
n19
n18

Table 2 (continued)
Infrared
Raman
Rel. int. Amorphous Rel. int. Crystalline Rel. int. Liquid Rel. int. Amorphous Rel. int. Crystalline Rel. int. ni
Assignments
NP stretch
a
Gas
810 R
805 Q vs
800 P
802
685
s
808
s
795
p vw
796
vw
803
vw
n10
791
687
681
670
m
789
686
681
669
vw
vw
vw
s
crystal effect
PS stretch (Isotopic impurity)
crystal effect
ww
vw
vw
s
crystal effect
680 R
677 Q
m
667
632
s
662
s
667
634
p vs
p s
664
632
vs
662
vs
n11
n11
PS stretch
646
454 R
vw
vw
vw
PS Stretch (near-cis)
480
446
sh br
vs
480
446
sh br
vs
n21
n12
ND out-of-plane bend
PF2 wag
449 Q
444 P
417 Q
s
445
p m
443
m
445
416
m
m
414 Q vs
392 R
415
383
vs
m
416
vs
414
389
p m
vw
415
383
rn
n20
n 0
PF2 rock
386 Q vs
vw
PF2 symmetric deformation
(near-cis)
13
380 P
375
347
vs
w
373
291
vw
m
n13
n21
PF2 symmetric deformation
NH out-of-plane bend
336 Q vs
328 P
343
289
250
s
333
288
253
vw
p m
w
294 R
288 Q vs
281 P
vs
w
291
s
289
253
m
w
n14
CNP bend
254
251
w
w
254
250
w
crystal effect
PF2 twist
250
233
sh br
sh
n22
n 0
m
PF2 twist (near-cis)
NPS bend (near-cis)
NPS bend
22
181
171
vw
vw
n 0
15
172
94
vw
vw
170
122
106
69
ww
m
s
171
vw
169
vw
n15
n23
n24
CD3 torsion
90
77
48
34
vw
vw
m
asymmetric torsion
Lattice modes
w
51
m
m
ni0 refer to the near-cis conformer.
a

Table 3
Observed infrared and Raman frequencies ( cm21) and assignments for CH3N(D)P(S)F2 (abbreviations used: v, very; s, strong; m, moderate; w, weak; br, broad; sh, shoulder; p,
polarized; R, Q and P refer to band contours in the gas phase infrared spectrum)
Infrared
Gas
Raman
Assignments
a
Rel. int. Amorphous Rel. int. Crystalline Rel. int. Liquid Rel. int. Amorphous Rel. int. Crystalline Rel. int. ni
3395
3325
w
m
3395
3325
vw
vw
crystal effect (d0)
3481
3446
3024
w
3325
3010
w
3350
p
3325
vw
NH stretch (d0)
vw
vw
NH stretch (d0, near-cis)
CH3 antisymmetric stretch
n16 CH3 antisymmetric stretch
sh
3014
2999
2990
2962
2956
2935
2906
vw
vw
vw
vw
sh
3015
3000
2992
2963
w
n2
2990
2958
sh
2892
2957
vw
w
vw
vw
w
2990
2957
vw
vw
2965
2921
m
w
p w
n3
CH3 symmetric stretch
2927
2902
vw
vw
vw
vw
2929
2905
vw
vw
2931
2906
vw
vw
2909
2842
vw
2853 R
2848 Q
2842 P
w
2839
vw
2846
vw
p w
2841
w
2847
2562
w
Fermi resonance
2843
2564
vw
vw
vw
crystal effect
Fermi resonance
crystal effect
2534
2468
vw
m
2537
2480
p w
p w
2539
2470
vw
w
2511
2468
w
m
2510
2467
vw
vw
2588
2562
2510
w
n1
n 0
ND stretch
sh
vw
ND stretch (near-cis)
1
2377
1474
1454
1435
vw
w
1474
1453
1439
1395
1387
vw
vw
vw
w
1473
1457
1440
vw
vw
vw
1474
1459
1435
vw
vw
vw
n4 CH3 antisymmetric deformation
n17 CH3 antisymmetric deformation
w
vw
1460
1437
vw
p vw
n5
CH3 symmetric deformation
crystal effect (d0)
w
crystal effect (d0)
1392
1265
rn
s
1398
1276
w
m
1382
1280
1264
w
m
w
NH deformation (d0)
CN stretch
1265
1130
1113
932
p vw
p vw
sh
1265
1124
vw
vw
1277
1263
1130
vw
vw
vw
n8
n8
n7
crystal effect
CN stretch (d0)
1140
1116
sh
s
1108
932
vs
s
1117
1105
933
m
vs
m
1117
1100
939
vw
vw
vw
CH3 rock
1101
931
vw
vw
n18 CH3 out-of-plane rock
n9
942 Q vs
932 Q
vw
PF2 symmetric stretch
914
m
917
898
m
m
925
899
vw
vw
PF2 symmetric stretch (d0)

Table 3 (continued)
Infrared
Raman
Rel. int. Amorphous Rel. int. Crystalline Rel. int. Liquid Rel. int. Amorphous Rel. int. Crystalline Rel. int. ni
Assignments
a
Gas
878
893
864
s
888
868
m
s
890
870
vw
894
868
vw
sh
890
878
861
vw
vw
vw
vs
s
sh
ww
n6
ND deformation
850 Q
844 Q
853
843
687
683
vs
vs
m
s
854
845
vw
vw
n19 PF2 antisymmetric stretch
n10 NP stretch
842
682
vs
s
847
680
vw
847
675
vw
vs
683
673
rn
vs
crystal effect
698 R
692 Q
687 P
651 Q
s
678
m
p vs
n11 PS stretch
643 Q vw
465 Q
632
457
417
vw
vs
635
456
415
p s
632
454
416
w
m
rn
n 0
PS stretch (near-cis)
11
462 Q vs
417 Q
455
417
s
p rn
p rn
452
417
m
m
n12 PF2 wag
412 Q vs
396 R
vs
vs
n13 PF2 symmetric deformation
393 Q vs
389 Q
n 0
PF2 rock (near-cis)
20
390
s
390
375
332
303
s
s
s
s
387
300
254
vw
m
391
300
256
182
vw
m
389
374
323
301
vw
vw
vw
w
n21 ND out-of-plane bend
n20 PF2 rock
340
300
280
vw
m
340
299
280
w
s
n21 ND out-of-plane bend
n14 CNP bend
n 0
sh
vw
ND out-of-plane bend (near-cis)
21
250
230
w
250
m
w
w
251
w
n22 PF2 twist
223
w
sh
n 0
n 0
PF2 twist (near-cis)
NPS bend (near-cis)
22
15
193
181
vw
vw
180
95
vw
vw
182
135
112
95
vw
sh
m
vw
178
130
95
vw
sh
n15 NPS bend
n23 CH3 torsion
n24 Asymmetric torsion
vw
vw
w
67
82
51
35
vw
vw
vw
52
w
Lattice modes
ni0 refer to the near-cis conformer.
a

Table 4
Observed infrared and Raman frequencies ( cm21) and assignments for CD3N(D)P(S)F2 (abbreviations used: v, very; S, strong; m, moderate; w, weak; br, broad; sh, shoulder; p,
polarized; R, Q and P refer to band contours in the gas phase infrared spectrum)
Infrared
Gas
Raman
Assignments
a
Rel. int. Amorphous Rel. int. Crystalline Rel. int. Liquid Rel. int. Amorphous Rel. int. Crystalline Rel. int. ni
3395
3329
m
s
3395
3325
vw
vw
crystal effect (d3)
3482
3446
w
3330
m
3332
vw
NH stretch (d3)
vw
NH stretch (d3, near-cis)
3016
2995
2975
vw
vw
vw
2972
vw
2975
vw
2971
p vw
2983
2958
2841
sh
vw
vw
2538
2500
m
s
2537
2496
vw
vw
crystal effect
2587
2564
rn
sh
2510
rn
2520
2429
w
2512
2432
w
n1
ND stretch
ND stretch (near-cis)
Fermi resonance
n 0
1
2454
2425
2259
2253
2243
2222
sh
2454
2422
2260
2254
2245
2224
vw
vw
m
2433
2255
m
m
vw
vw
2265
sh
sh
vw
vw
vw
vw
n2
CD3 antisymmetric stretch
w
2243
2237
w
2243
2223
w
2249
2226
2196
2135
vw
2248
2223
(2190)
2134
2115
w
vw
vw
n16 CD3 antisymmetric stretch
Fermi resonance
sh
vw
p vw
vw
vw
2142
w
2135
2113
vw
rn
2135
2115
vw
vw
p w
w
2136
2116
w
Fermi resonance
Fermi resonance
vw
vw
2100 R
2094 Q rn
2089 P
2085
2087
1389
1372
1240
m
w
2088
p rn
2086
m
2087
rn
n3
CD3 symmetric stretch
crystal effect (d3)
NH deformation (d3)
CN stretch
1386 R
1382 Q
1376 P
1230
m
1392
1240
w
m
vs
vs
vs
1237
1107
vw
1237
1107
wvw
vw
n8
1235
1123
sh
m
1125
1108
m
m
1138 R
1133 Q
1126 P
w
1115
s
p vw
1107
vw
n5
n4
CD3 symmetric deformation
1100
1062
1055
969
sh
vw
vw
s
1100
1065
1052
971
sh
1081
975
w
s
vw
vw
vs
1065
1061
970
sh
vw
1063
1056
968
vw
vw
w
1064
1056
970
vw
vw
vw
CD3 antisymmetric deformation
n17 CD3 antisymmetric deformation
n7 CD3 rock
p vw

Table 4 (continued)
Infrared
Raman
Rel. int. Amorphous Rel. int. Crystalline Rel. int. Liquid Rel. int. Amorphous Rel. int. Crystalline Rel. int. ni
Assignments
a
Gas
940
vs
927
895
vs
m
931
924
894
873
856
851
845
s
930
p vw
928
vw
936
921
895
882
860
849
849
vw
vw
vw
vw
vw
vw
vw
n9
PF2 symmetric stretch
s
s
875
vs
s
vs
vs
874
859
vw
vw
880
859
vw
vw
n19 PF2 antisymmetric stretch
n18 CD3 rock
852 Q
848 P
853
800
vs
s
n6
ND deformation
s
805 R
801 Q vs
801
786
682
678
669
666
s
792
p vw
793
680
vw
vw
801
784
vw
vw
n10 NP stretch
crystal effect
s
vw
vw
m
vs
crystal effect (d3)
682
665
659
vw
m
PS stretch (d3)
crystal effect
crystal effect
676 R
671 Q
666 P
643 R
m
664
s
660
vs
664
p vs;
661
vs
vs
n11 PS stretch
637 Q Vw
453 R
627
445
vw
vs
630
444
p s
628
443
vw
m
n 0
PS stretch (near-cis)
11
449 Q Vs
443 P
444
417
s
p m
444
415
387
m
n12 PF2 wag
415 Q
410 Q
392 R
386 Q
S
S
417
vs
vs
414
381
p m
ww
415
390
m
m
n13 PF2 symmetric deformation
n 0
PF2 rock (near-cis)
20
388
380
m
m
388
370
365
m
m
sh
vw
vw
n21 ND out-of-plane bend
n20 PF2 rock
382 Q
332
286 Q
Vw
M
330
290
br m
s
331
290
s
s
n21 ND out-of-plane bend
n14 CNP bend
287
252
w
287
253
170
m
289
247
m
w
282 Q
214 br
M
W
n 0
ND out-of-plane bend (near-cis)
21
247
170
w
w
247
m
vw
w
n22 PF2 twist
n 0
n 0
PF2 twist (near-cis)
CNP bend (near-cis)
22
14
182
170
vw
vw
170
105
vw
m
vw
168
117
vw
sh
n15 NPS bend
n23 CD3 torsion

Table 4 (continued)
Infrared
Raman
Rel. int. Amorphous Rel. int. Crystalline Rel. int. Liquid Rel. int. Amorphous Rel. int. Crystalline Rel. int. ni
Assignments
a
Gas
90
w
93
68
50
vw
vw
m
95
83
50
33
vw
vw
vw
vw
n24 asymmetric torsion
lattice modes
ni0 refer to the near-cis conformer.
a

Table 5
Ê
Ab initio structure (bond lengths in A, bond angles in degrees, rotational constants in MHz and dipole moments in debyes) obtained for the near-trans and near-cis conformers of
methylaminothiophosphoryl di¯uoride
Parameters
RHF/6-31G(d)
MP2/6-31G(d)
B3LYP/6-31G(d)
MP2/6-3111G(d,p)
Near-trans
Near-trans
Near-cis
1.460
Near-trans
Near-cis
1.463
Near-trans
Near-cis
1.464
Near-cis
1.464
R(C±N)
R(P±N)
R(PyS)
r(P±F1)
r(P±F2)
r(N±H1)
r(C±H2)
r(C±H3)
r(C±H4)
/CNP
/NPS
1.460
1.621
1.900
1.556
1.553
0.997
1.081
1.083
1.081
125.0
117.4
115.7
116.5
115.5
108.8
111.8
109.6
174.8
125.0
2123.3
12.3
1.464
1.638
1.891
1.593
1.587
1.013
1.090
1.093
1.090
123.2
117.4
116.4
117.4
114.7
108.7
112.4
108.8
172.8
125.4
2122.4
18.3
1.467
1.647
1.912
1.594
1.588
1.012
1.093
1.096
1.092
124.1
117.0
116.2
117.3
114.2
108.8
112.6
109.1
173.0
125.5
2122.2
18.1
1.464
1.635
1.887
1.593
1.586
1.010
1.090
1.093
1.090
124.0
118.3
115.8
117.2
114.5
108.6
112.4
109.2
173.9
125.8
2122.4
20.0
1.624
1.899
1.558
1.552
1.000
1.079
1.084
1.081
1.640
1.891
1.594
1.586
1.016
1.089
1.094
1.091
1.646
1.912
1.594
1.588
1.015
1.091
1.096
1.094
1.637
1.887
1.594
1.584
1.012
1.090
1.094
1.090
126.4
124.0
125.6
123.6
119.0
114.3
117.0
115.5
109.4
111.8
108.6
27.0
118.3
114.9
118.1
115.5
108.5
112.4
108.7
31.6
118.5
115.0
117.6
115.8
108.8
112.1
109.2
24.5
118.7
114.4
117.8
115.6
108.9
112.5
108.4
33.1
/F1PS
/F2PS
/H1NP
/H2CN
/H3CN
/H4CN
t(SPNC)
t(F1PSN)
t(F2PSN)
t(H1NPS)
t(H2CNP)
t(H3CNP)
t(H4CNP)
umau
126.7
2122.0
188.8
231.6
89.4
127.3
2120.9
187.0
234.7
86.1
126.9
2121.5
186.4
225.8
94.7
127.0
2121.2
185.8
240.6
80.5
154.9
275.8
37.0
159.5
281.0
42.0
159.7
281.0
42.2
159.2
280.6
41.7
2150.0
2152.6
2144.0
2158.4
4.724
1.656
0.253
5.012
3830
4.584
4.308
1.892
0.431
4.725
3743
4.399
4.057
1.718
0.390
4.423
3691
4.057
3.430
2.179
0.432
4.514
3772
4.386
umbu
umcu
umtu
A
1.611
0.024
4.858
1.436
0.046
4.628
1.462
0.008
4.312
0.770
0.027
4.453
2656
2585
2558
2600
B
1757
2258
1642
1742
2254
1637
1722
2228
1609
1735
2251
1643
C
2(E 1 1031)
DE ( cm21
1605
1598
1574
1597
0.852492
0.850438
451
1.755879
1.754419
320
3.578564
3.577259
286
2.290805
2.289042
387
)

B.J. van der Veken et al. / Journal of Molecular Structure 560 (2001) 57±86
75
Fig. 14. Internal coordinates of methylaminothiophosphoryl di¯uoride.
shown in Fig. 18, and, in spite of the crystal ®eld
5. Vibrational assignment
splitting, it is clear that it is the high frequency
n(PyS) which remains in this phase. This shows
that in the crystalline solid, the compound has the
trans conformation.
The trans conformer of methylaminothio-
phosphoryl di¯uoride has a plane of symmetry that
coincides with the plane formed by the a and b
Table 6
Symmetry coordinates for CH3N(H)PSF2
Species
A0
Descriptiona
Symmetry coordinates
NH stretch
S1 ˆ P
CH3 antisymmetric stretch
CH3 symmetric stretch
CH3 antisymmetric deformation
CH3 symmetric deformation
NH deformation
CH3 rock
S2 ˆ 2r1 2 r2 2 r3
S3 ˆ r1 1 r2 1 r3
S4 ˆ 2b1 2 b2 2 b3
S5 ˆ b1 1 b2 1 b3 2 a1 2 a2 2 a3
S6 ˆ h 2 d
S7 ˆ 2a1 2 a2 2 a3
S8 ˆ Q
CN stretch
PF2 symmetric stretch
NP stretch
PS stretch
S9 ˆ X1 1 X2
S10 ˆ T
S11 ˆ R
PF2 wag
S12 ˆ m1 1 m2 2 f1 2 f2
S13 ˆ 4u 2 m1 2 m2 2 f1 2 f2
S14 ˆ e
PF2 symmetric deformation
CNP bend
NPS bend
S15 ˆ r
A00
CH3 antisymmetric stretch
CH3 antisymmetric deformation
CH3 rock
S16 ˆ r2 2 r3
S17 ˆ b2 2 b3
S18 ˆ a2 2 a3
S19 ˆ X1 2 X2
S20 ˆ m1 1 f1 2 m2 2 f2
S21 ˆ m1 2 m2 2 f1 1 f2
S22 ˆ t1
PF2 antisymmetric stretch
PF2 rock
PF2 twist
CH3 torsion
asymmetric torsion
S23 ˆ t2
a
Not normalized.

Table 7
Observed and calculated frequencies ( cm21) and potential energy distributions for methylaminothiophosphoryl di¯uoride
Vib. no. Descriptiona
trans
Near-cis
Ab initioa
Fixed
IR
Raman Calc. Obs.c
act. dep.
PED
Ab
Fixed
IR
Raman Calc. Obs.c
PED
scaledb int.
initioa scaledb int.
act.
dep.
A0 n1
NH stretch
3682
3493
3034
84.7 36.0 0.24
7.6 71.6 0.69
3480 100S1
3023 99S2
3609
3242
3424
3041
60.3 118.2
4.2 29.4
0.27
0.73
3446 100S1
92S2
n2
n3
n4
n5
CH3 antisymmetric 3235
stretch
CH3 symmetric
stretch
3122
2929
1504
1439
34.5 107.6 0.00
2963 99S3
3116
1580
1516
2923
1499
1438
37.7 109.0
8.3 20.9
0.06 2909* 93S3
CH3 antisymmetric 1585
deformation
18.2 14.7 0.75 (1477) 89S4
0.67
0.74
73S4,10S7
CH3 symmetric
deformation
NH deformation
CH3 rock
1516
1.0
8.7 0.60 (1438) 100S5
1.7
6.9
96S5
n6
n7
n8
1453
1201
1161
1379
1143
1101
141.7
145.3
63.7
0.7 0.51
4.1 0.34
0.9 0.73
1391 77S6
1472
1164
1178
1397
1106
1119
69.3
60.2
94.8
2.0
3.0
2.7
0.70
0.75
0.64
69S6,10S7
1113 39S8,48S7
1160 25S7,30S8,
22S10,21S6
37S7,26S6,14S8,14S10
47S8,29S7,11S18
CN stretch
n9
PF2 symmetric
stretch
952
901
702
469
399
299
174
907
857
673
460
398
297
281.3
138.8
0.7 0.23
2.8 0.18
923 66S9,16S11
952
906
660
381
367
448
907
863
627
379
364
438
323.0
163.3
2.8
1.6
0.28
0.73
0.06
0.73
0.53
0.65
44S9,26S11
n10
n11
n12
n13
n14
NP stretch
867 47S10,22S8,
20S11
42S10,28S9,13S8
PS stretch
PF2 wag
40.7 14.4 0.06
699 36S11,31S9,
12S12
11.2 20.7
651 59S11,19S9,14S10
392 58S12,25S13
31S13,25S20,22S12
15.4
15.8
13.0
0.4
9.5 0.27
4.6 0.53
4.5 0.69
1.5 0.74
461 45S12,25S11
,
,
34.4
17.1
66.8
2.3
3.4
2.4
3.6
2.2
19S14
PF2 symmetric
deformation
CNP bend
(378) 80S13,12S15
299 24S14,32S12
26S15,11S13
440 16S14,36S24,26S20
n15
NPS bend
173
(180) 48S15,40S14
199
197
0.74
0.67
194* 46S15,32S14
89S16
A00 n16
CH3 antisymmetric 3209
stretch
3011
11.9 51.8 0.75 (2999) 100S16
8.3 20.3 0.75 (1452) 93S17
3206
3007
15.7 72.8
10.5 16.7
n17
CH3 antisymmetric 1547
deformation
1468
1560
1480
0.70
78S17,11S4
n18
n19
CH3 rock
1192
906
1131
861
0.0
3.5 0.75 (1124) 92S18
1189
905
1129
861
12.2
4.1
1.0
0.64
0.71
81S18
87S19
PF2 antisymmetric
stretch
159.2
0.8 0.75
877 95S19
181.9
n20
n21
PF2 rock
409
377
408
358
1.6
2.4 0.75
0.83 0.75
414 70S20
326
514
324
486
2.7
2.2
0.69
325 39S20,20S13,10S15
53S21,14S14,14S15
NH out-of-plane
bend
77.8
(350) 35S21,36S20,16S13
81.3
1.69 0.74
n22
n23
n24
PF2 twist
280
121
70
278
116
70
17.9
25.5
3.0
3.6 0.75
0.8 0.75
0.1 0.75
(253) 63S22,10S24
(135) 78S23,18S24
(112) 68S24,18S22
233
110
73
232
105
73
9.6
1.5
0.8
3.5
0.5
0.3
0.75
0.60
0.75
233 82S21
87S22
CH3 torsion
asymmetric torsion
100S23
a
For convenience, the vibrations are ordered according to the trans conformer.
b
c
Scaled ab initio calculations with factors of 0.88 for C±H stretches, 0.9 for C±H bends, CH3 torsion and other stretches, 1.0 for heavy atom bends and asymmetric torsion.
Wavenumbers in parenthesis are taken from the infrared or Raman spectrum of the solid, except the one indicated with an asterisk from the Raman spectrum of the liquid.

Table 8
Observed and calculated frequencies ( cm21) and potential energy distributions for methylaminothiophosphoryl di¯uoride-d3
Vib. no. Descriptiona
trans
Near-cis
Ab
Fixed IR
Raman Calc. obs.c
PED
Ab
Fixed IR
Raman Calc. obs.c PED
act. dep.
initioa scaledb int.
act.
dep.
initioa scaledb int.
A0 n1
NH stretch
3682 3493
85.2 36.0
4.7 35.7
0.24
0.71
3482 100S1
2265 99S2
3609 3424
2406 2257
60.4 119.0 0.27 3447 100S1
n2
n3
n4
n5
CD3 antisymmetric 2401 2253
stretch
2.9 16.7 0.69
91S2
CD3 symmetric
stretch
2240 2101
26.1 52.8
0.00
0.75
0.20
2093 99S3
2237 2098
1142 1083
1121 1064
1450 1376
25.7 53.3 0.04
98S3
CD3 antisymmetric 1141 1083
deformation
19.0
95.4
5.4
0.8
1082 89S4,11S5
1133 40S5,45S8
44.9
32.5
4.3 0.73
9.7 0.64
55S4,14S17,10S5
16S5,42S4,28S17
CD3 symmetric
deformation
NH deformation
CD3 rock
1197 1136
n6
n7
n8
n9
1440 1367
171.2
93.9
0.8
3.9
3.6
2.7
0.26
0.38
0.72
0.01
1382 83S6
86.1
178.4
44.0
4.3 0.69
2.0 0.60
0.7 0.66
1.5 0.30
81S6
1005
961
973 38S7,13S14,11S10
1082 16S8,35S5,28S10,12S4
913 55S9,25S11
1004
961
29S7,11S5,11S14
44S8,45S5
61S9,10S11
CN stretch
PF2 symmetric
stretch
1131 1073
83.3
1199 1138
938
891
230.5
935
888
166.2
n10
n11
n12
n13
NP stretch
PS stretch
834
682
455
398
794
651
447
397
105.6
1.1
0.73
0.07
0.31
0.52
805 22S10,26S7,16S8,15S11
677 33S11,25S9,13S7,12S10
449 52S12,22S11,15S14
(375) 81S13,12S15
830
650
376
364
790
618
374
361
136.2
4.2 0.75
20S10,39S7,14S11
646 53S11,18S10,18S9
386 69S12,14S13
22.2 14.6
5.0 20.3 0.06
PF2 wag
18.2
15.2
8.7
4.6
36.2
3.5 0.74
2.2 0.53
PF2 symmetric
deformation
CNP bend
12.6
31S13,34S20,12S12
n14
n15
288
162
286
161
10.6
0.6
4.6
1.3
0.67
0.74
288 31S15,26S12,22S14,11S13
(169) 45S14,44S15
436
182
428
180
48.1
3.0
4.1 0.65
15S14,25S24,24S20,14S13,12S15
NPS bend
1.9 0.73 181* 42S15,42S14
A00 n16
CD3 antisymmetric 2383 2236
stretch
8.2 26.9
0.75
2241 99S16
2379 2232
1124 1066
10.8 35.9 0.73
90S16
n17
CD3 antisymmetric 1116 1058
deformation
2.9
8.7
0.75 (1065) 98S17
38.3
4.1 0.74
56S17,14S10,13S5,10S8
n18
n19
CD3 rock
922
905
875
861
2.3
3.5
0.7
0.75
0.75
(859) 93S18
876 92S19
915
905
868
861
14.7
3.6 0.59
0.9 0.65
90S18
88S19
PF2 antisymmetric
stretch
157.5
174.0
n20
n21
PF2 rock
403
374
402
355
1.4
2.3
0.75
414 71S20,12S21
320
502
318
475
2.8
2.2 0.69
1.69 0.74
30S20,25S13,14S15,11S14
62S21,11S15
ND out-of-plane
bend
77.8
0.83 0.75
(347) 34S21,42S20,12S13
81.3
n22
n23
n24
PF2 twist
278
98
275
93
17.1
15.1
4.5
3.7
0.6
0.0
0.75
0.75
0.75
(251) 66S22,21S20
232
81
231
77
10.0
1.0
3.4 0.75
0.3 0.63
0.2 0.75
233 82S22
CD3 torsion
asymmetric torsion
(122) 81S23,15S24
99S23
63
62
(106) 61S24,15S21,13S23
67
67
0.4
100S24
a
For convenience, the vibrations are ordered according to the trans conformer.
Scaled ab initio calculations with factors of 0.88 for C±H stretches, 0.9 for C±H bends, CH3 torsion and other stretches, 1.0 for heavy atom bends and asymmetric torsion.
Wavenumbers in parenthesis are taken from the infrared or Raman spectrum of the solid, except the one indicated with an asterisk from the Raman spectrum of the liquid.
b
c

Table 9
Observed and calculated frequencies ( cm21) and potential energy distributions for methylaminothiophosphoryl di¯uoride-d1
Vib. no. Descriptiona
trans
Near-cis
Ab
Fixed IR
Raman Calc. obs.c
act. dep.
PED
Ab
Fixed IR
Raman Calc. Obs.c PED
act. dep.
initioa scaledb int.
initioa scaledb int.
A0 n1
ND stretch
2698 2560
3235 3034
3122 2929
54.7 16.5 0.26
6.7 72.2 0.68
34.4 107.1 0.00
2588 99S1
3024 99S2
2965 99S3
2641 2506
3242 3041
3116 2923
1577 1496
1516 1438
1333 1266
1171 1114
47.3 59.2 0.27 2562 99S1
n2
CH3 antisymmetric stretch
CH3 symmetric stretch
3.8 28.9 0.72
37.8 108.2 0.06
7.5 18.8 0.67
92S2
n3
93S3
n4
CH3 antisymmetric deformation 1584 1502
13.2 14.4 0.75 (1474) 91S4
75S4,14S17
n5
CH3 symmetric deformation
ND deformation
CH3 rock
1516 1438
932 886
1.1
116.5
129.6
228.2
159.4
136.6
8.7 0.61 (1439) 100S5
3.0
102.5
86.4
6.4 0.75
2.3 0.73
3.5 0.70
4.4 0.49
0.8 0.27
0.8 0.48
95S5
n6
1.1 0.66
4.3 0.35
1.3 0.54
1.2 0.08
1.6 0.21
878 26S6,28S922S8
1116 53S7,34S8
1265 23S8,25S10,24S6,23S7 916
23S6,29S7,21S8,19S10
45S7,39S8
n7
1195 1137
1303 1237
n8
CN stretch
870
936
848
619
376
346
471
197
119.9
268.3
141.3
19S8,28S6,16S19,11S7,11S10
32S9,25S6,19S11
31S10,41S9,18S6
n9
PF2 symmetric stretch
NP stretch
975
877
698
466
398
298
172
927
834
668
457
396
296
171
942 38S9,28S6
983
891
651
378
350
480
199
n10
n11
n12
n13
n14
n15
844 43S10,22S11,21S6
692 35S11,31S9,11S12
462 46S12,24S11,19S14
412 81S13,12S15
PS stretch
35.8 14.9 0.06
7.1 20.9 0.07
643 58S11,17S10,17S9
PF2 wag
15.8
15.6
12.7
0.4
9.1 0.28
4.4 0.52
4.5 0.69
1.5 0.75
29.2
3.2 0.69
2.7 0.51
2.4 0.75
45S12,32S13
PF2 symmetric deformation
CNP bend
21.4
11.5
2.8
40S13,30S12,14S24,11S14
31S14,27S15,14S13
300 32S14,25S15,25S14
(178) 49S15,40S14
NPS bend
2.1 0.74 193* 44S15,32S14
A00 n16
n17
n18
CH3 antisymmetric stretch
3209 3011
11.8 51.8 0.75 (2999) 100S16
8.4 20.3 0.75 (1453) 93S17
3206 3007
1560 1480
1188 1128
15.7 72.6 0.67
11.1 16.8 0.70
89S16
77S17,12S4
CH3 antisymmetric deformation 1547 1468
CH3 rock
1188 1127
0.1
155.4
9.8
3.3 0.75 (1100) 93S18
3.6
177.3
48.6
2.2
4.2 0.61
1.3 0.56
3.1 0.71
1.52 0.73
2.8 0.74
0.5 0.60
0.3 0.75
90S18
n19
n20
PF2 antisymmetric stretch
PF2 rock
906
392
321
268
118
67
861
392
304
267
112
66
0.8 0.75
2.2 0.75
0.83 0.75
3.4 0.75
0.7 0.75
0.1 0.75
850 95S19
904
417
305
221
109
71
860
409
290
218
104
71
75S19
(375) 80S20
393 52S20,31S24,11S21
280 23S21,38S20,18S22
223 59S22,20S24,14S15
86S23
n21
n22
ND out-of-plane bend
PF2 twist
77.8
6.8
(332) 56S21,21S12
(250) 75S22,14S20
(135) 70S23,24S24
(112) 66S24,15S22
14.1
2.8
n23
n24
CH3 torsion
19.7
2.7
asymmetric torsion
0.6
100S24
a
For convenience, the vibrations are ordered according to the trans conformer.
b
c
Scaled ab initio calculations with factors of 0.88 for C±H stretches, 0.9 for C±H bends, CH3 torsion and other stretches, 1.0 for heavy atom bends and asymmetric torsion.
Wavenumbers in parenthesis are taken from the infrared or Raman spectrum of the solid, except the one indicated with an asterisk from the Raman spectrum of the liquid.

Table 10
Observed and calculated frequencies ( cm21) and potential energy distributions for methylaminothiophosphoryl di¯uoride-d4
Vib. no. Descriptiona
trans
Near-cis
Ab
Fixed IR
Raman Calc. Obs.c PED
dep.
Ab
Fixed IR
Raman Calc. Obs.c PED
dep.
initioa scaledb int. act.
initioa scaledb int. act.
A0 n1
ND stretch
CD3 antisymmetric 2401 2253
stretch
CD3 symmetric
stretch
CD3 antisymmetric 1141 1082
deformation
CD3 symmetric
deformation
ND deformation
CD3 rock
CN stretch
PF2 symmetric
stretch
2698 2560
54.2 16.7 0.26 2587 99S1
4.6 35.8 0.71 2265 99S2
2641 2506
2406 2257
47.1 58.6 0.29 2564 99S1
n2
n3
n4
n5
2.8 16.8 0.69
25.7 53.3 0.04
2.5 6.1 0.69
7.4 2.1 0.68
91S2
2240 2101
26.1 52.7 0.00 2094 99S3
8.6 5.6 0.75 1081 92S4
6.4 2.0 0.60 1133 76S5
2237 2098
1136 1078
1172 1113
98S3
69S4,28S17
77S5
1170 1110
n6
n7
n8
n9
885 840
1021 972
89.8 1.5 0.13
33.4 6.7 0.36
1280 1215 333.8 0.6 0.74 1230 40S8,30S10,24S6
852 43S6,19S10,14S11,13S9
975 36S7,27S8,13S5,12S6
890 846 110.6 1.1 0.29
1014 965 17.4 4.7 0.75
1290 1225 175.9 0.7 0.68
979 934 323.9 1.0 0.11
27S6,44S9,20S10
31S7,30S8
40S8,26S10,25S6
33S9,22S11,15S6
969 924 222.5 0.7 0.12
940 51S9,15S11,14S6
n10
n11
n12
n13
NP stretch
PS stretch
PF2 wag
PF2 symmetric
deformation
CNP bend
832 792
678 648
452 444
397 395
95.8 1.0 0.70
20.5 15.0 0.07
18.6 0.56 0.32
15.1 4.5 0.51
801 18S10,27S7,18S8,13S11
671 32S11,24S9,13S10,12S7
449 53S12,22S11,15S14
410 82S13,11S15
815 774
644 612
373 370
345 341
90.7 4.6 0.73 12S10,40S7,14S6,10S8
3.4 20.5 0.06 637 52S11,20S10,17S9
29.4 3.3 0.71
19.3 2.4 0.51
53S12,21S13
42S13,22S12,17S24,12S14
n14
n15
287 284
162 160
10.4 4.5 0.67
0.6 1.3 0.74 (168) 44S15,45S14
286 31S15,25S12,23S14,10S13
181 180
462 453
3.8 1.8 0.73 182* 42S14,39S15
14.8 2.7 0.73
NPS bend
28S15,22S13,21S14
A00 n16
CD3 antisymmetric 2383 2236
stretch
CD3 antisymmetric 1116 1058
deformation
8.1 26.9 0.75 2243 99S16
2379 2232
1122 1064
915 868
10.7 35.9 0.73
90S16
n17
2.7 8.6 0.75 (1052) 98S17
0.1 3.3 0.75 (860) 95S18
3.9 8.0 0.68
12.4 3.6 0.59
70S17,26S4
n18
n19
CD3 rock
916 870
90S18
90S19
PF2 antisymmetric 905 861 157.0 0.7 0.75
stretch
875 93S19
905 861 163.5 0.9 0.74
n20
n21
PF2 rock
388 388
316 299
8.9 2.2 0.75 (370) 81S20,12S23
77.8 0.83 0.75 (331) 69S21,14S12
412 405
302 286
44.1 3.4 0.70 386 53S20
2.2 1.52 0.73 282 21S21,33S20,21S22
ND out-of-plane
bend
PF2 twist
CD3 torsion 96
asymmetric torsion 60
n22
n23
n24
267 266
6.7 3.5 0.75 (247) 77S22,13S20
12.3 0.5 0.75 (117) 73S23,18S24
3.8 0.0 0.75 (105) 59S24,14S23,12S22
219 216
14.8 2.7 0.74 214 58S22,21S24,16S15
1.0 0.3 0.63
0.3 0.2 0.74
92
60
81
65
77
65
98S23
100S24
a
For convenience, the vibrations are ordered according to the trans conformer.
b
c
Scaled ab initio calculations with factors of 0.88 for C±H stretches, 0.9 for C±H bends, CH3 torsion and other stretches, 1.0 for heavy atom bends and asymmetric torsion.
Wavenumbers in parenthesis are taken from the infrared or Raman spectrum of the solid, except the one indicated with an asterisk from the Raman spectrum of the liquid.

80
B.J. van der Veken et al. / Journal of Molecular Structure 560 (2001) 57±86
Fig. 15. Fundamental and ®rst overtone bands of n(N±H) in the mid
and near-infrared vapor phase spectra of CH3N(H)P(S)F2.
principal axes of the molecule and belongs to the Cs
point group. Therefore, the vibrations belonging to the
A0 block should exhibit A, B or A/B hybrid type
infrared band envelopes and polarized Raman lines,
whereas the A00 modes should have C-type band
contours and depolarized Raman lines. For the near-
cis conformer which has C1 symmetry, such distinc-
tion between the vibration modes can not be made.
Our assignments are also based on isotopic effects.
group frequencies and the assignments are supported
by the normal coordinate analyses.
Fig. 16. Comparison of experimental with simulated pro®les for the
yS stretch: (A) simulation using the near-cis and trans conformer
pro®les for the high and low frequency band, respectively; (B)
experimental pro®le; (C) simulation using the trans and near-cis
conformer pro®les for the high and low frequency band for
CH3N(H)P(S)F2, respectively.
Figs. 10±13. Observed frequencies are listed in
Tables 1±4. In view of the results discussed in the
previous paragraph, the fundamentals tabulated and/
or discussed below, except where otherwise stated,
belong to the trans conformer. With the near-cis
conformer present in the ¯uid phases, for the latter
it is implied that, with a few exceptions where they
are observed separately, the near-cis fundamentals
appear accidentally degenerate with the corre-
sponding trans fundamentals. The intense vapor
phase infrared band at 3480 cm21 for d0, and
3482 cm21 for the d3 derivative, are assigned to the
NH stretch of the trans conformer, whereas the
weaker band at 3446 and 3447 cm21, respectively,
are due to the same fundamental in the near-cis
conformer. The ND fundamentals of the trans rotamer
are observed at 2588 cm21 in the d1 and 2587 cm21 in
the d4 derivative. In both cases, the vapor phase ND
stretch shows a low frequency shoulder at 2562 cm21
Mid-infrared spectra of CH3N(H)P(yS)F2 and its
deuterated species are shown in Figs. 2±5, the far-
infrared spectra in Figs. 6±9 and Raman spectra in
Table 11
Directional cosines for the 2m~ =2Q of the PS stretch in the principal
axis system of CH3N(H)P(S)F2, CH3N(D)P(S)F2, CD3N(H)P(S)F2
and CD3N(D)P(S)F2
trans
cis
ucos au
ucos bu
ucos gu ucos au
ucos bu
ucos gu
d0
d1
d3
d4
0.954
0.951
0.944
0.941
0.300
0.309
0.330
0.338
0.000
0.000
0.000
0.000
0.277
0.418
0.152
0.216
0.961
0.908
0.988
0.976
0.000
0.000
0.000
0.000

B.J. van der Veken et al. / Journal of Molecular Structure 560 (2001) 57±86
81
Fig. 18. Raman spectra of the PS doublet (near-cis and trans
conformer) in the liquid phase at 748C (A); 198C (B); 2358C (C);
2728C (D); 21058C (E); and the PS stretch in the crystalline phase
(F) of CH3N(H)P(S)F2.
solid, both in the infrared and Raman spectra, a
doublet nNX (X ˆ H,D) is observed. This is caused
by factor group splitting as a consequence of the
presence of two or more molecules in the unit cell
of the crystals.
Fig. 17. Comparison of experimental with simulated pro®les for the
PyS stretch of CH3N(D)P(S)F2: (A) simulation using the near-cis
and trans conformer pro®les for the high and the low frequency
band, respectively; (B) experimental pro®le; (C) simulation using
the trans conformer pro®les of both the d0 and d1 derivatives for the
high frequency band and the simulation using the near-cis
conformer pro®les of both the d0 and d1 derivatives for the low
frequency band; (D) simulation using the trans and near-cis
conformer pro®les for the high and low frequency band, respec-
tively.
In the CH3 stretching region of the d0 and d3
isotopic species, more than three lines are observed
in the Raman spectra of the liquids. Two of these
bands at 2844 and 2956 cm21 are intense and strongly
polarized and are assigned as a Fermi-doublet of the
symmetric CH3 stretch with the ®rst overtone of the
symmetric CH3 deformation. The shoulder at
2909 cm21 is assigned to the CH3 stretch of the
near-cis conformer. The two near degenerate antisym-
metric CH3 stretching motions are assigned to the
strongly overlapping, depolarized doublet at
2993 cm21. Both the infrared and Raman spectra of
the crystalline solid show a very complex pattern. The
occurrence of crystal site effects in the spectrum of the
solid for the NH stretches suggests the possible exis-
tence of similar effects for the CH3 stretches, which
contributes to the complexity of the observed pattern.
For the d3 and d4 species, the CD3 symmetric stretch is
assigned to the strongly polarized Raman line
observed at 2087 cm21 in d3 and at 2088 cm21 in d4
isotopomer in the spectra of the liquids. The other
polarized bands, at 2135 cm21 for d3, and at 2133
and 2226 cm21 for d4, are caused by Fermi resonances
in d1 and 2564 cm21 in d4 isotopopomers, respec-
tively, caused by the near-cis conformer. In the liquid
phase, as shown by the Raman spectra, the N±H
stretching mode is observed at substantially lower
wavenumber. These shifts can be explained by the
existence of hydrogen bonds and the broad band at
3352 cm21 in d0, and 3357 cm21 in d3, show a high-
frequency shoulder due to the presence of the second
conformer. Taking into account the expected H/D
shift, the doublet structure of n(ND) in the Raman
spectra of the liquid, at 2534 and 2468 cm21 in d1
and at 2510 and 2433 cm21 in d4, cannot be due to
the presence of two conformers, but must be due to
Fermi resonance between the trans ND stretch and an
overtone and/or combination band. In the crystalline

82
B.J. van der Veken et al. / Journal of Molecular Structure 560 (2001) 57±86
Table 12
Temperature intensity ratios of the PS stretch for the conformational stability of liquid methylaminothiophosporyl di¯uoride and its isotopic
derivatives
T (8C)
104´(1/T), in K21
K ˆ Itrans=Icis
ln K
CH3N(H)P( ˆ S)F2
74
43
28.80
31.65
34.20
38.00
42.00
46.10
49.75
52.00
55.25
59.50
64.10
68.50
73.53
1.714
2.010
2.369
3.024
3.697
4.441
5.440
6.253
6.997
8.617
10.629
13.314
18.137
0.539
0.698
0.862
1.107
1.308
1.491
1.694
1.833
1.946
2.154
2.364
2.589
2.898
19
210
235
256
272
284
292
2105
2117
2127
2138
CD3N(H)P( ˆ S)F2
70
43
29.16
31.65
34.42
38.17
42.02
45.87
49.75
52.91
55.25
59.17
1.698
1.983
2.382
3.072
3.858
4.775
6.159
7.180
8.053
9.247
0.529
0.685
0.868
1.122
1.350
1.563
1.818
1.971
2.086
2.224
17
211
235
255
272
284
292
2104
CH3N(D)P( ˆ S)F2
77
72
28.57
28.98
31.25
33.67
38.46
42.55
46.30
50.00
52.63
55.56
59.52
63.29
1.756
1.879
1.989
2.560
3.356
4.063
5.166
6.046
6.900
8.355
8.979
10.965
0.563
0.631
0.688
0.940
1.211
1.402
1.642
1.799
1.932
2.123
2.195
2.395
47
24
213
238
257
273
283
293
2105
2115
CD3N(D)P( ˆ S)F2
69
46
29.24
31.35
33.44
38.31
42.02
46.08
49.73
52.63
55.56
58.48
61.73
1.885
2.135
2.437
3.351
4.164
5.099
6.016
6.689
8.001
8.601
11.002
0.634
0.758
0.891
1.209
1.42G
1.629
1.795
1.900
2.080
2.152
2.405
26
212
235
256
272
283
293
2102
2111

B.J. van der Veken et al. / Journal of Molecular Structure 560 (2001) 57±86
83
of the ns(CD3) with the ®rst overtones and/or combi-
nation bands of CD3 deformations. The antisymmetric
A0 and A00 CD3 stretches of the d3 and d4 derivatives
are found accidentally degenerate at 2241 and
2249 cm21, respectively.
The CH3 symmetric deformation, ds(CH3), and the
CH3 antisymmetric deformations, das(CH3) and
d 0as(CH3) are assigned to the Raman bands observed
at 1436, 1470 and 1460 cm21, respectively, for the
liquid. The analogous fundamentals in the d3 deriva-
tive are assigned to the bands at 1114, 1093 and
1063 cm21. Using the H/D isotope effects, the NX
(X ˆ H,D) in-plane bend can be assigned to the
bands at 1391 cm21 (d0), 1382 cm21 (d3), 878 cm21
(d1) and 852 cm21 (d4).
Using these arguments, in the Raman spectra of the
liquid, the g(NH) in the d0 isotope can be assigned at
345 cm21 whereas the tCH3 in the same isotopomer is
assigned at 135 cm21. As can be expected from the
hydrogen bonding in the liquid phase, g(NH) in the
vapor phase occurs some 20 cm21 below the value in
the liquid phase. In the spectrum of the solid there
appear to be two bands which can be attributed to
this mode where it is possible that the higher
frequency band is associated with a stronger hydrogen
bond. These bonds are found at 390 and 332 cm21. in
the spectrum of the d1 solid and are only slightly
shifted (388 and 331 cm21) in the spectrum of the
crystalline d4 molecule.
The CH3 rockings and the CN stretch are expected
to occur in the same spectral region [13]. Detailed
comparison of the spectra of the different isotopic
species allows the two r(CH3) modes to be assigned
in the Raman spectra accidentally degenerate at
1107 cm21 in d0 and at 1113 cm21 in d1, whereas
they split into two bands at 967 and 859 cm21 in the
species and 970 and 859 cm21 in d4. The C±N stretch
is observed at 1124 cm21 (d0), 1265 cm21 (d1),
1093 cm21 (d3) and 1230 cm21 (d4). It may be
observed that from d1 to d2 and from d3 to d4 species
this mode is shifted to higher wavenumbers, opposite
to what is expected on the basis of the mass effect.
This shift must be due to different couplings between
the r(CH3) and n(CN) modes in the different isotopic
derivatives.
The PF2 stretches, nas(PF2) and ns(PF2) are found to
be quite insensitive to deuteration and are assigned to
the two most intense infrared bands. The NP stretch
can be expected to be sensitive to the deuteration of
the N±H bond. Therefore, we assign n(NP) to the
band at 847 cm21 in d1, and at 792 cm21 in the d4
compound, whereas the analogous fundamental in
the d0 and d3 isotopomers occurs almost accidentally
degenerate with the antisymmetric PF2 stretch at 866
and 795 cm21, respectively.
6. Discussion
It should be noted that the description of several of
the vibrations in the `®nger print' spectral region is
rather arbitrary because of extensive mixing as indi-
cated by the PEDs. Also the mixing changes signi®-
cantly with the isotopic substitution. For example, for
the normal species, CH3N(H)P(S)F2, the NH bend
(A0) is a relatively pure mode (77%) at 1391 cm21
with 21% contribution to the 1113 cm21 band.
However with deuteration (N-d) it is spread over
four bands at 1265 (24%), 942 (28%), 878 (26%)
(ND bend) and 844 (21%) cm21 where the relatively
strong infrared band at 1265 cm21 assigned as the CN
stretch (23%) but it is made up of 23% CH3 rock, and
25% NP stretch in addition to the 24% contribution
from the ND Bend. The CH3 rock assigned at
1113 cm21 for the ªlightº species with 48% contribu-
tion from this motion and 39% from the CN stretch is
only slightly changed to 53% CH3 rock and 34% CN
stretch with its assignment at 1116 cm21 for the N-d
molecule. For the ªlightº compound the PF2
symmetric stretch at 923 cm21 is 66% this motion
and 16% PS stretch whereas the corresponding band
at 942 cm21 for the N-d molecule there is now 38% of
the PF2 symmetric stretch and 28% N±D bend. There-
fore, the molecular motions for n6 through n10 are
signi®cantly different for the N-d molecule compared
to those for the normal species except for the CH3
rock.
As mentioned earlier, the extremely strong and
polarized Raman doublet at 684 and 640 cm21 are
characterized as the PS stretches of the trans and
near-cis conformers, respectively. The CH3/CD3
substitution will strongly in¯uence the CH3 torsional
frequency, whereas the N±H out-of-plane deforma-
tion will be shifted by deuteration of the N±H bond.
The enthalpy difference obtained from the Raman
spectrum of liquid may differ signi®cantly from that in

84
B.J. van der Veken et al. / Journal of Molecular Structure 560 (2001) 57±86
Fig. 19. Calculated infrared spectra of methylaminothiophosphoryl
di¯uoride: (A) mixture of the trans and near-cis conformers; (B)
pure near-cis; and (C) pure trans.
Fig. 20. Comparison of experimental and calculated Raman spectra
of methylaminothiophosphoryl di¯uoride: (A) observed Raman
spectrum of the liquid; (B) simulated Raman spectrum of mixture
of the trans and near-cis conformers; (C) simulated Raman spec-
trum of the pure near-cis conformer; and (D) simulated Raman
spectrum of the pure trans conformer.
the vapor state. However, since both stable confor-
mers have similar dipole moments one does not
expect that there will be a change in which conformer
is the stable one in the gas, i.e. the trans form will be
the more stable rotamer in the gaseous phase. This
expected result is also supported by the ab initio
calculations where the trans conformer is predicted
to be more stable than the near-cis form by the
range of 451 cm21 (RHF/6-31G(d)) to 286 cm21
(B3LYP/6-31G(d)). Since these values are relatively
large, it is expected that the ab initio calculations predict
the correct conformer to be the more stable form.
The uncertainties listed for the enthalpy values are
the statistical uncertainties and do not take into
account any underlying combination bands or
problems with association, etc. In order to minimize
such problems it is better to obtain the enthalpy differ-
ence from several conformer pairs rather than from
just one pair. Nevertheless, one does not expect the
error to be more than 15±20% from the Raman data
which were taken over a relatively large temperature
range so it is clear that the trans form is the more
stable conformer.
The simulated infrared (Fig. 19) and Raman (Fig.
20) spectra were predicted using ®xed scaled wave-
numbers and infrared intensities determined from the
MP2/6-31G(d) calculations. Infrared intensities were
calculated based on the dipole moment derivatives
with respect to the Cartesian coordinates. The deriva-
tives were taken from the ab initio calculations trans-
formed to normal coordinates by:
!
ꢀ
ꢁ
X
2mu
2mu
ˆ
Lij
2Qi
2Xj
j
Where Qi is the ith normal coordinate, Xj is the jth

B.J. van der Veken et al. / Journal of Molecular Structure 560 (2001) 57±86
85
Cartesian displacement coordinate, and Lij is the trans-
formation matrix between the Cartesian displacement
coordinates and normal coordinates. The infrared
intensities were then calculated and the predicted
infrared spectra of the near-cis and trans conformers
of normal compound are shown in Fig.19B and C,
respectively. The mixture of the two conformers is
shown in Fig. 19A. These spectra can be compared
to the experimental spectra and they provide support
for the assignment of the observed bands to the indi-
cated fundamentals, and more importantly, for the
identi®cation of the conformer bands. For example,
the conformer pair for the PS stretching mode is easily
identi®ed from the predicted spectra. Although the PC
stretch is predicted too weak and the CH3 stretch is
predicted too strong, the overall simulated infrared
spectrum closely resembles the observed spectrum,
which provides excellent evidence for the quality of
the ab initio calculations.
cis conformers of the normal compound are shown in
Fig. 20D and C, respectively. In Fig. 20B the
predicted Raman spectra of the mixtures of the two
conformers are shown using the experimentally deter-
mined DH value of 368 cm21 with the trans
conformer the more stable rotamer. These spectra
should be compared with the experimental Raman
spectrum of the liquid which is shown in Fig. 20A.
They closely match in both band positions and inten-
sities. Spectroscopic evidence for the conformational
stability can be found by this comparison and the
conclusion can be drawn that the trans conformer is
the thermodynamically preferred conformer.
Using scaling factor of 0.88 for carbon±hydrogen
stretches, 0.90 for all the other vibrations except 1.0
for the heavy atom bends and asymmetric torsion, the
wavenumbers for the fundamentals are predicted from
the MP2/6-31G(d) calculations to better than 10 cm21
comparing with the experimental values. These
results show that these two scaling factors are suf®-
cient to predict the wavenumbers for the observed
fundamentals for these types of molecules. The ab
initio predictions for the various isotopomers were
valuable in making vibrational assignments for the
various deuterated species since it was not possible
to obtain the N-d species with high isotopic purity.
Additional support for the proposed vibrational
assignment was obtained from the Teller±Redlich
product rule.
The structural parameters calculated for the two
different conformers of CH3NHP(S)F2 are very
similar with major differences being the NPS and
CNP angles (Table 5). For the near-cis conformer
both the CNP and NPS angle are predicted to open
by 1±28 which is consistent with the steric interaction
of the methyl group with the sulfur atom. However,
these differences are not predicted by the MP2/6-
3111G(d,p) calculation with the larger basis set.
Since the structural parameters are predicted to differ
by such a small amount it is not surprising that the
force constants for the two conformers also have very
small differences which leads to most of the funda-
mentals for the two conformers to have similar values.
The evaluation of the Raman activity by using the
analytical gradient methods has been developed
[27,28]. The activity Sj can be expressed as:
Sj ˆ gjꢀ45a2j 1 7bj2†;
where gj is the degeneracy of the vibrational mode j,
aj is the derivative of the isotropic polarizability and
bj is that of the anisotropic polarizability. The Raman
scattering cross sections, 2sj/2V, which are propor-
tional to the Raman intensities, can be calculated from
the scattering activities and the predicted wavenum-
bers for each normal mode using the relationship
[29,30]:
0
1
!
!
4
24p4
h
B
B
C
C
2sj
2V
ꢀn0 2 nj†
1 2 exp
ꢂ
ꢃ
ˆ
B
@
C
A
Sj;
8p2cnj
2hcnj
kT
45
where n0 is the exciting wavenumber, vj is the vibra-
tional wavenumber of the jth normal mode, and Sj is
the corresponding Raman scattering activity. To obtain
the polarized Raman scatteringcross sections, the polar-
izabilities are incorporated into Sj by Sj‰ꢀ1 2 rj†=ꢀl 1
rj†Š where rj is the depolarization ratio of the jth normal
mode. The Raman scattering cross sections and
calculated ®xed wavenumbers are used together with
Lorentzian function to obtain the calculated spectra.
The predicted Raman spectra of the trans and near-
Acknowledgements
The NFWO (Belgium) is thanked for ®nancing

86
B.J. van der Veken et al. / Journal of Molecular Structure 560 (2001) 57±86
[17] R.G. Cavell, T.L. Charleton, W. Sim, J. Am. Chem. Soc. 93
(1971) 1130.
spectroscopic equipment and for travel grants. Prof.
Dr. H. Desseyn is thanked for his interest in this work
and H. Mulders is thanked for technical assistance.
Travel between our two laboratories was provided
by a NATO grant. We also acknowledge P. Schilder
who recorded some of the earliest vibrational data.
[18] C. Moller, M.S. Plesset, Phys. Rev. 46 (1934) 618.
[19] F.A. Miller, B.M. Harney, Appl. Spectrosc. 24 (1970) 291.
[20] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A.
Robb, J.R. Cheeseman, V.G. Zakrzewski, J.A. Montgomery
Jr., R.E. Stratmann, J.C. Burant, S. Dapprich, J.M. Millam,
A.D. Daniels, K.N. Kudin, M.C. Strain, O. Farkas, J. Tomasi,
V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C.
Adamo, S. Clifford, J. Ochterski, G.A. Petersson, P.Y. Ayala,
Q. Cui, K. Morokuma, D.K. Malick, A.D. Rabuck, K. Ragha-
vachari, J.B. Foresman, J. Cioslowski, J.V. Ortiz, B.B.
Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.
Gomperts, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham,
C.Y. Peng, A. Nanayakkara, M.W. Wong, J.L. Andres, C.
Gonzalez, M. Challacombe, P.M.W. Gill, B. Johnson, W.
Chen, M.W. Wong, J.L. Andres, M. Head-Gordon, E.S.
Replogle, J.A. Pople, Gaussian 98 (Revision A.6), Gaussian
Inc., Pittsburgh, PA, 1998.
References
[1] J.R. Durig, A.W. Cox Jr., J. Chem. Phys. 63 (1975) 2303.
[2] J.R. Durig, A.W. Cox Jr., J. Chem. Phys. 64 (1976) 1930.
[3] J.R. Durig, A.W. Cox Jr., J. Chem. Phys. 80 (1976) 2493.
[4] J.R. Durig, Y.S. Li, J. Mol. Spectrosc. 70 (1978) 27.
[5] J.R. Durig, B.J. Streusand, Appl. Spectrosc. 34 (1980) 65.
[6] R.L. Odeurs, B.J. van der Veken, M.A. Herman, J.R. Durig, J.
Mol. Struct. 117 (1984) 235.
[7] G.H. Pieters, B.J. van der Veken, A.J. Barnes, T.S. Little, J.R.
Durig, J. Mol. Struct. 125 (1984) 243.
[21] P. Pulay, Mol. Phys. 17 (1969) 197.
[22] E.B. Wilson, J.C. Decius, P.C. Cross, Molecular Vibrations,
McGraw Hill, New York, 1955.
[8] B.J. van der Veken, R.L. Odeurs, M.A. Herman, J.R. Durig,
Spectrochim. Acta 40 (1984) 563.
[23] J.H. Schachtschneider, Vibrational Analysis of Polyatomic
Molecules, Parts V and VI, Technical Report Nos. 231 and
57, Shell Development Co., Houston, TX, 1964 and 1965.
[24] G.A. Guirgis, X. Zhu, Z. Yu, J.R. Durig, J. Phys. Chem. 104
(2000) 4383.
[9] P. Groner, J.S. Church, Y.S. Li, J.R. Durig, J. Chem. Phys. 82
(1985) 3894.
[10] B.J. van der Veken, P. Coppens, R.D. Johnson, J.R. Durig, J.
Chem. Phys. 83 (1985) 1517.
[11] B.J. van der Veken, P. Coppens, R.D. Johnson, J.R. Durig, J.
Phys. Chem. 90 (1986) 4537.
[25] B.J. van der Veken, R.S. Sanders, P. Schilder, W. Zhao, J.R.
Durig, submitted for publication.
[12] B.J. van der Veken, R.S. Sanders, J.R. Durig, J. Mol. Struct.
216 (1990) 113.
[26] R.L. Odeurs, B.J. van der Veken, M.A. Herman, J. Mol.
Struct. 98 (1983) 221.
[13] J.R. Durig, R.J. Harlan, P. Groner, J. Phys. Chem. 93 (1989)
3041.
[27] M.J. Frisch, Y. Yamaguchi, J.F. Gaw, H.F. Schaefer III, J.S.
Binkley, J. Chem. Phys. 84 (1986) 531.
[14] B.J. van der Veken, P. Coppens, R.S. Sanders, F.F. Daeyaert,
J.R. Durig, J. Mol. Struct. 272 (1992) 305.
[28] R.D. Amos, Chem. Phys. Lett. 124 (1986) 376.
[29] G.W. Chantry, in: A. Anderson (Ed.), The Raman Effect, vol.
1, Marcel Dekker, New York, 1971 (chap. 2).
[15] B.J. van der Veken, R.S. Sanders, R.J. Harlan, J.R. Durig,
Spectrochim. Acta 49A (1993) 1833.
[16] S. Wolfe, Acc. Chem. Res. 5 (1972) 102.
[30] P.L. Polavarapu, J. Phys. Chem. 94 (1990) 8106.