Infrared and Raman spectroscopy of 9,9′-spirobifluorene, bis(2,2′-biphenylene)silane, and bis(2,2′-biphenylene)germane. Vibrational assignment by depolarization measurement and HF and density functional theory studies

Article abstract of DOI:10.1021/jp973159f

The infrared and Raman spectra of 9,9′-spirobifluorene (SBFL), bis(2,2′-biphenylene)silane (BBPS), and bis-(2,2′-biphenylene)germane (BBPG) are measured, and the vibrational frequencies and modes for these molecules are assigned by ab initio Hartree-Fock and Becke 3-Lee-Yang-Parr (B3LYP) density functional theory (DFT) calculations using the 4-31G and 3-21G basis sets for SBFL and for BBPS and BBPG, respectively. Assignment of some of the vibrational modes of SBFL is also confirmed by solution Raman spectroscopy with the depolarization method. Comparison of the calculated and experimental vibrational spectra reveals that the DFT calculations are quite accurate in predicting the vibrational frequencies, intensities, and modes. It is found that the central Si and Ge atoms dominantly enhance some of the IR intensities of the vibrations involving only the M-C₄ (M = Si, Ge) antisymmetric stretching motion. The enhancement of some of the infrared intensities is accounted for by the increase of the electric dipole moment involving the M-_C4 vibrations due to the presence of the small electronegative Si and Ge atoms in the central position.

Full text of DOI:10.1021/jp973159f

J. Phys. Chem. A 1998, 102, 1139-1145
1139
Infrared and Raman Spectroscopy of 9,9′-Spirobifluorene, Bis(2,2′-biphenylene)silane, and
Bis(2,2′-biphenylene)germane. Vibrational Assignment by Depolarization Measurement and
HF and Density Functional Theory Studies
,†
‡
§
|
†
Bong Hyun Boo,* Jaiwook Park, Hwan Gon Yeo, Sang Yeon Lee, Chan Jo Park, and
Jin Ho Kim†
Department of Chemistry, Chungnam National UniVersity, Taejon 305-764, Korea, Center for
Molecular Science, 373-1 Kusung-dong, Yusung-gu, Taejon 305-701, Korea, Department of Chemistry,
Pohang UniVersity of Science and Technology, Pohang 790-784, Korea, Agency for Defence DeVelopment,
Taejon 305-600, Korea, and Department of Industrial Chemistry, Kyungpook National UniVersity,
Taegu 702-701, Korea
ReceiVed: September 29, 1997
The infrared and Raman spectra of 9,9′-spirobifluorene (SBFL), bis(2,2′-biphenylene)silane (BBPS), and bis-
(
2,2′-biphenylene)germane (BBPG) are measured, and the vibrational frequencies and modes for these molecules
are assigned by ab initio Hartree-Fock and Becke 3-Lee-Yang-Parr (B3LYP) density functional theory
DFT) calculations using the 4-31G and 3-21G basis sets for SBFL and for BBPS and BBPG, respectively.
(
Assignment of some of the vibrational modes of SBFL is also confirmed by solution Raman spectroscopy
with the depolarization method. Comparison of the calculated and experimental vibrational spectra reveals
that the DFT calculations are quite accurate in predicting the vibrational frequencies, intensities, and modes.
It is found that the central Si and Ge atoms dominantly enhance some of the IR intensities of the vibrations
involving only the M-C (M ) Si, Ge) antisymmetric stretching motion. The enhancement of some of the
4
4
infrared intensities is accounted for by the increase of the electric dipole moment involving the M-C vibrations
due to the presence of the small electronegative Si and Ge atoms in the central position.
accuracy than do the HF and MP2 calculations.1
3-23
Even when
Introduction
a uniform scaling factor is applied to the computed vibrational
frequencies, DFT methods give better agreement with the
experimental frequencies than do the HF and MP2 meth-
Spiro compounds consist of two mutually perpendicular
π-systems connected via a common insulating tetracoordinate
atom. The structural characteristics of spiro compounds have
1
3,15-17,23
been theoretically1-7 and experimentally8-12 studied. Previous
ods.
Some examples of the calculation on the vibra-
15,16
tional frequencies of benzene are represented by Handy et al.
theoretical calculation indicates that the overlap integral of two
pπ orbitals in the different plane of spiro[4.4]nonatetraene is
about 20% of the adjacent p orbitals in planar π electron
1
8
and Wheeless et al., in which MP2 calculations obtained by
using 6-311G(d,p) basis set still underestimate the vibrational
frequencies corresponding to the ring torsion and overestimate
the vibrational frequency corresponding to a b2u mode, which
tends to dissociate into three acetylene molecules. On the other
hand, these frequencies are accurately predicted by the DFT
calculation using the Becke-Lee-Yang-Parr (BLYP) func-
1
systems.
The electronic-vibrational spectra of 9,9′-spirobifluorene
SBFL) are the subject of ongoing spectroscopic investigations.
(
Recently, a van der Waals complex of SBFL with Ar, N2 and
H2O was studied with laser-induced fluorescence spectroscopy
in the free jet expansion to observe one origin transition of the
1
5,16,18
tional.
-1
bare molecule at 33 039 cm and variation of the spectral shifts
with cluster size.10 The electronic and vibrational bands in the
S1 excited state were analyzed.1
Rauhut and Pulay employed BLYP and B3LYP functional
methods using the 6-31G* basis set to predict fundamentals of
20 small molecules such as benzene, ether, and methanol, of
which vibrational frequencies are exactly assigned. Meanwhile,
they derived uniform scaling factors of 0.995 and 0.963 for
BLYP and B3LYP, respectively, having root-mean-square (rms)
1,12
Published results show that the vibrational frequencies
0
appended to the 0 band origin of SBFL consist of fundamen-
0
10-12
tals, overtones, and combination bands.
Thus correct
-
1
19
assignment of the ground-state fundamentals is necessary to
elucidate the electronic-vibrational spectra.
Computational methods based on density functional theory
deviations of 26.2 and 18.5 cm , respectively. When these
workers applied these scaling factors to another 11 molecules
such as aniline, ethanol, and oxetane, the rms deviations turned
-
1
(
DFT) are becoming widely used. These methods predict
out to be 26.9 and 19.7 cm for the BLYP and B3LYP
methods, respectively, not very different from those of the
molecules mentioned above.
relatively accurate molecular structures and vibrations with
moderate computational effort. Published results indicate that
DFT methods reproduce experimental fundamentals with higher
We have measured the infrared and Raman spectra of 9,9′-
spirobifluorene (SBFL) and its Si and Ge analogues of bis(2,2′-
biphenylene)silane (BBPS) and bis(2,2′-biphenylene)germane
(BBPG) and performed the normal vibrational mode analyses
at the HF and B3LYP levels of theory using the 4-31G basis
†
Chungnam National University and Center for Molecular Science.
Pohang University of Science and Technology.
Agency for Defense Development.
Kyungpook National University.
‡
§
|
S1089-5639(97)03159-9 CCC: $15.00 © 1998 American Chemical Society
Published on Web 01/22/1998

1140 J. Phys. Chem. A, Vol. 102, No. 7, 1998
Boo et al.
set for SBFL and a relatively smaller basis set for BBPS and
BBPG such as 3-21G to examine whether the DFT method could
reproduce correctly or not the vibrational frequencies and
intensities of the Si- and Ge-containing molecules and to
elucidate how the central Si and Ge atoms affect the peak
positions and intensities in the infrared and Raman spectra.
We have assigned the fundamentals of the molecules by
comparing the experimental peak positions and peak intensities
with the theoretical ones. It is shown in this study that the
B3LYP calculation, even when the relatively small basis set
such as 3-21G is used, could reliably reproduce the experimental
vibrational frequencies. It is also found that the central Si and
Ge atoms dominantly enhance some of the IR intensities of the
vibrations involving only the M-C4 antisymmetric stretching
motion.
Figure 1. Comparison of the experimental (upper) and calculated
(lower) infrared spectra of 9,9′-spirobifluorene.
The scattered light was collected through a collecting lens and
focused on the slit of a Jobin-Yvon U-1000 double monochro-
mator with a focal length of 1 m. The monochromatized photon
was detected by a Hamamatsu R943-02 photomultiplier tube
which can be cooled to -20 °C. The crystal was mounted on
the specimen holder of a microscope with magnification set to
Experimental and Theoretical Methods
500. The magnification of 500 was chosen out of three available
Synthesis of 9,9′-spirobifluorene. The compound was
magnifications, 100, 500, and 1000, because it gave us the best
views of the crystal to irradiate the laser to the fine surfaces of
the crystal. Raman spectra of SBFL dissolved in carbon
tetrachloride were obtained by replacing the microscope with a
synthesized by the Clarkson and Gomberg method.2 The mp
4
24
of SBFL was 199-201 °C (not corrected) (198-199 °C ). The
1
purity was checked by TLC and H NMR and mass spectros-
1
copy. H NMR (80 MHz) (CDCl3): δ 7.84 (d, 4H, J ) 7.4
solution sample chamber. Saturated CCl solution was used to
4
Hz), 7.32 (dd, 4H, J ) 7.3 Hz, 7.4 Hz), 7.09 (dd, 4H, J ) 7.4
Hz, 7.3 Hz), 6.71 (d, 4H, J ) 7.4 Hz). MS (70 eV) m/z (relative
obtain Raman bands as strong as possible. The depolarization
ratio in the solution was obtained by dividing the count rate of
the scattered light of which the electric vector is perpendicular
to the electric vector of the incident light by the parallel count
+
+
intensity): (M + 1) 317 (47), M 316 (100), 315 (80), 289
(11), 157 (20).
-
1
Synthesis of 2,2-Dilithiobiphenyl (DLBP). The compound
rate. The spectral resolution was 0.15 cm at the slit width of
10 µm. The IR spectra of the KBr pellet were measured by a
Digilab-FTS-80 FTIR spectrometer. The spectral resolution was
2
5
was synthesized by the Gilman and Gorsich method. To a
solution of biphenyl (2.52 g, 16.3 mmol) and tetramethyleth-
ylenediamine (TMEDA) (5.80 mL, 38.4 mmol) in hexane (100
mL) was added n-buthyllithium (n-BuLi) (1.6 M in hexane, 24.6
mL, 39 mmol) dropwise at room temperature. The reaction
mixture was stirred at 60 °C for 2 h to obtain yellowish red
microcrystals of the complex of DLBP and TMEDA. The
crystals were washed with hexane and were dried in vacuo to
obtain 2.3 g in 50% yield.
-
1
2 cm .
The molecular geometries were optimized at the HF and
B3LYP levels of theory with basis sets of 4-31G for SBFL and
27
3-21G for BBPS and BBPG by using the Gaussian 94 program
at SERI in Korea. Vibrational frequencies were computed with
the HF/4-31G (SBFL), B3LYP/4-31G (SBFL), HF/3-21G
(BBPS and BBPG), and B3LYP/3-21G (BBPS and BBPG), and
Synthesis of Bis(2,2′-biphenylene)silane. This compound
they were uniformly scaled by 0.893 (HF/4-31G for SBFL),
2
6
was synthesized by the Gilman and Gorsich method. To a
solution of tetrachlorosilane (SiCl4) (0.43 mL, 3.8 mmol) in ether
0
0
.958 (B3LYP/4-31G for SBFL), 0.904 (HF/3-21G for BBPS),
.973 (B3LYP/3-21G for BBPS), 0.900 (HF/3-21G for BBPG),
(10 mL) at -78 °C was added , a solution of the complex of
and 0.968 (B3LYP/3-21G for BBPG). The frequencies and
normal modes were determined by diagonalizing the mass-
weighted force constant matrix.
DLBP and TMEDA (2.12 g, 7.51 mmol) in ether (20 mL)
dropwise through a canula. The reaction mixture was heated
to reflux for 2 h. Subsequently, 10 mL of benzene was added,
and the mixture was refluxed for 2 h. After quenching with
water the organic layer was separated, dried over anhydrous
MgSO4, and filtered. The solid obtained by removal of solvent
from the filtrate was recrystallized from hexane to give white
crystals of BBPS of 1.25 g in a quantitative yield.
Results and Discussion
We have performed a variety of normal vibrational mode
analyses of SBFL, BBPS, and BBPG. In Figures 1-6, we
present a comparison of the experimental and DFT theoretical
infrared and Raman spectra in the range except 1650 < ν <
Synthesis of Bis(2,2′-biphenylene)germane (BBPG). This
26
-1
compound was synthesized by the Gilman and Gorsich method.
2900 cm , in which other peaks except the overtone bands
Tetrachlorogermane (GeCl4) (0.23 mL, 2.0 mmol) and the
complex of DLBP and TMEDA (2.12 g, 7.51 mmol) were
allowed to react as described in the preceding experiment for
the preparation of BBPS to give 0.20 g of BBPG in 27% yield.
A Spectra-Physics 2060-7S Beam Lok Ar ion laser at the
wavelength of 514.5 nm was used for the sample excitation.
are not observed. It is noticed that the simulated spectra have
-1
a Lorentzian distribution, having a fwhm of 4 cm . As shown
in Figures 1 and 2, the vibrational frequencies and intensities
of the experimental infrared and Raman spectra of SBFL are
quite nicely reproduced by the DFT calculations showing almost
a one-to-one correspondence. However, the DFT method

Infrared and Raman Spectra of SBFL, BBPS, and BBPG
J. Phys. Chem. A, Vol. 102, No. 7, 1998 1141
Figure 2. Comparison of the experimental (upper) and calculated
Figure 4. Comparison of the experimental (upper) and calculated
(lower) Raman spectra of bis(2,2′-biphenylene)silane.
(lower) Raman spectra of 9,9′-spirobifluorene.
Figure 3. Comparison of the experimental (upper) and calculated
lower) infrared spectra of bis(2,2′-biphenylene)silane.
Figure 5. Comparison of the experimental (upper) and calculated
(
(lower) infrared spectra of bis(2,2′-biphenylene)germane.
slightly overestimates the vibrational frequencies above 2950
-
1
cm , corresponding to the C-H stretching vibration, and
greatly overestimates the infrared intensity of the C-H stretch-
ing vibration. It is noticed in Figures 2, 4, and 6 that the
theoretical Raman intensities are extracted from the ab initio
HF calculation since the DFT calculation using the Gaussian
94 program does not produce Raman intensity. The HF Raman
intensities are not consistent with the experimental ones in the
whole range. However, the experimental Raman intensity ratios
-
1
below 1650 cm are quite correctly reproduced by the HF
calculation.
For BBPS and BBPG, some discrepancy in the peak position
and intensity is observed between the experimental and theoreti-
cal spectra, as indicated in Figures 3-6. Especially, in the
-
1
region above 2950 cm , discrepancy between theory and
experiment is appreciably observed in the peak positions and
intensities.
Figure 6. Comparison of the experimental (upper) and calculated
lower) Raman spectra of bis(2,2′-biphenylene)germane.
(
As shown in Figures 7 and 8, some frequencies and intensities
of SBFL are quite close to those of BBPS and BBPG. However,
significant differences are observed in the IR intensities of the
influenced in the M-C4 (M ) Si and Ge) antisymmetric
stretching vibrations, in which the electric dipole moments of
the molecules are increased. This is further discussed below.
Under D2d symmetry, quantum mechanical calculation indi-
cates that the molecules include 88 fundamentals having various
symmetries of 20a1 + 9a2 + 10b1 + 20b2 + 29e. Identification
of some of the fundamentals is straightforward due to the well-
separated spectral positions and to the excellent agreement in
the IR and Raman intensity ratios between theory and experi-
-
1
fundamentals at 1131, 1567, and 1590 cm in BBPS and in
-
1
those at 1120, 1570, and 1585 cm in BBPG. All the
frequencies are related to the in-plane CH and CCC bending
and M-C4 (M ) C, Si, Ge) stretching vibrational modes. The
DFT theory shows that the sizable differences in the IR
intensities are caused by the differences in the electronegativities
of C (2.55), Si (1.90), and Ge (2.01). The effect is greatly

1142 J. Phys. Chem. A, Vol. 102, No. 7, 1998
Boo et al.
Figure 9. Raman spectra of the saturated solution of 9,9′-spirobifluo-
rene in CCl . “x” refers to the solvent peaks. The spectra denoted by
4
the solid and dotted lines were obtained with the electric vector of the
scattered light parallel and perpendicular to the electric vector of the
incident light, respectively.
Figure 7. Comparison of the experimental infrared spectra of 9,9′-
spirobifluorene, bis(2,2′-biphenylene)silane, and bis(2,2′-biphenylene)-
germane.
reproduce the experimental fundamentals with the rms deviations
-1
of 22.63 and 8.30 cm , respectively. Since these scaling factors
are quite close to the previously used scaling factors by
21
Langhoff, we confirmed that our assignment of the vibrational
frequencies and their modes are correctly made and that the
DFT method can correctly estimate the fundamentals of the spiro
compounds. It is verified again that the DFT method reproduces
fundamentals much better than does the HF method. Our
frequencies scaled with the reported factor by Langhoff21 are
in excellent agreement with experiment as shown in Table 1.
a1 Modes. Under D2d symmetry, the SBFL molecule involves
2
0 a1 modes. Among the fundamentals, 15 fundamentals that
we identify as a1 fundamentals are 336, 424, 704, 760, 1020,
1
1
1
088, 1152, 1176, 1224, 1297, 1352, 1444, 1480, 1580, and
-
1
608 cm . Among these, the fundamentals at 1020, 1224,
Figure 8. Comparison of the experimental Raman spectra of 9,9′-
spirobifluorene, bis(2,2′-biphenylene)silane, and bis(2,2′-biphenylene)-
germane.
-
1
297, and 1480 cm are more or less intense in the Raman
spectrum (Figure 9), and thus the modes could be double-
checked to be a1 modes by the depolarization measurement in
the solution Raman spectra using a saturated solution in CCl4.
The depolarization ratios measured experimentally are indicated
along with the theoretical one in Table 1. It is noticed that the
signal-to-noise ratio is quite low since the solubility of SBFL
in CCl4 is very low; therefore only four frequencies with the a1
mode are safely assigned by the depolarization method.
It is shown in the B3LYP calculation that most of the a1
fundamentals involve totally symmetric C-H and CCC in-plane
bending and stretching modes.
ment. However, straightforward identification of fundamentals
of BBPS and BBPG is not relatively feasible due to the
discrepancy between theory and experiment in the peak position
and intensity.
Almost all the modes are delocalized over the whole molecule
and thus cannot be assigned to several local bonds. This is a
characteristic feature of cyclic compounds, particularly aromatic
compounds.28 Therefore we provide in Tables 1 and 2 only
the approximate mode descriptions.
9
,9′-Spirobifluorene. The scaled fundamentals for SBFL are
a2 Modes. Identification of the a2 fundamentals is impossible
here since their modes are infrared and Raman inactive. The
DFT calculation indicates that most of the fundamentals
correspond to ring torsion and out-of-plane C-H and C-C-C
bending modes.
listed in Table 1 together with the experimental fundamentals.
The uniform scaling factors of 0.893 and 0.958 are used for
HF/4-31G and B3LYP/4-31G, respectively, the factor of 0.893
2
2
has been used in our previous calculations on fluorene,
pyrrole,2 and carbazole, and the factor of 0.958 has been used
by Langhoff and has been verified as one of the good scaling
factors.21 The theoretical peak positions shown in Figures 1
and 2 are scaled by 0.958.
3
23
b1 Modes. Among the 10 fundamentals attributable to b1,
only six fundamentals are identified as b1 mode. These are 48,
-1
156, 320, 796, 916, and 1004 cm . Most of the fundamentals
involve C-H out-of-plane bending and ring torsion modes.
b2 Modes. Among the 20 b2 modes, only 12 modes are
identified. These are 412, 657, 948, 1021, 1104, 1208, 1295,
The rms deviations of the scaled frequencies from experiment
-
1
-1
turned out to be 22.80 cm for HF/4-31G and 8.82 cm for
B3LYP/4-31G. It is noticed that the comparison between theory
and experiment is made for 50 frequencies listed in Table 1,
and the fundamentals are assigned on the basis of this DFT
calculation. It is shown in this study that for the 50 experimental
frequency data, the scaling factors of 0.891 for the HF
calculation and 0.961 for the DFT calculation slightly better
-1
1351, 1446, 1480, 1581, and 1628 cm . Most of the vibrations
include C-H in-plane bending and stretching and ring deforma-
tion modes.
e Modes. Among the 29 fundamentals with the e mode, 17
fundamentals are identified as the e mode. These are 419, 482,
514, 618, 636, 727, 750, 908, 946, 1004, 1031, 1164, 1176,

Infrared and Raman Spectra of SBFL, BBPS, and BBPG
TABLE 1: Comparison of the Experimental and Calculated Vibrational Frequencies of 9,9-Spirobifluorenea
HF/4-31G B3LYP/4-31G exptl
J. Phys. Chem. A, Vol. 102, No. 7, 1998 1143
approximated
type of mode
sym
a
no.b
freq
I
IR
I
Ram
DPc
freq
I
IR
freq
I
IR
I
Ram
DPc
1
7
150
330
419
705
765
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0.4
0.7
5
0.1
0.7
1
2
0.01
0.1
0.2
1
0.6
3
1
10
0.06
6
2
6
0.06
15
0.6
5
3
0.35
0.01
0.23
0.08
0.48
0.01
0.01
0.06
0.21
0.01
0.04
0.02
0.47
0.02
0.28
0.35
0.73
0.75
0.50
0.01
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
150
331
420
705
766
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
ring defm
ip CH and CCC bend
ring defm
14
16
32
40
60
68
70
72
76
80
85
87
90
97
336
424
704
16
5
18
6
89
5
37
6
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ring str, ip CH bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
CH str
CH str
CH str
totally sym CH str
ring tor
ring tor
ring tor
ring tor
oop CH and CCC bend
oop CH and CCC bend
oop CH bend
oop CH bend
oop CH and CCC bend
ring tor
ring tor
ring tor
oop CH and CCC bend
ring tor
oop CH bend
oop CH and CCC bend
oop CH bend
oop CH and CCC bend
oop CH and CCC bend
ring defm
760
1013
1083
1144
1154
1197
1288
1308
1452
1482
1587
1620
2985
2995
3005
3017
134
284
453
580
774
822
929
1005
1051
49
158
329
516
587
796
836
941
1006
1051
217
405
666
747
952
1014
1099
1151
1197
1020
1083
1148
1178
1221
1298
1354
1444
1471
1573
1596
3048
3058
3068
3080
134
275
440
565
740
785
876
942
983
47
156
319
501
574
762
800
886
943
984
216
407
665
748
948
1020
1088
1152
1176
1224
1297
1352
1444
1480
1580
1608
∼0
25
∼0
∼0
18
8
54
12
∼0
1
1
1
1
1
01
05
08
13
17
6
35
100
2
27
11
100
0
0
0
0
0
0
0
0
0
2
0.4
1
0.8
0
0.5
0.1
0.6
0.4
0.3
0.05
0.7
0
0.1
4
3
2
0.1
2
a
2
1
1
2
3
4
4
4
5
2
9
5
5
1
5
9
4
3
8
3
2
6
9
2
6
0
5
9
5
1
6
3
1
7
1
5
b
1
48
156
320
22
8
10
1
2
2
3
4
4
5
5
0
0
0
0
0
0
0
0.4
1
796
916
6
5
1004
6
5
b
2
1
3
3
5
6
6
7
7
412
657
ring defm
5
7
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH bend
0.05
0.4
2
3
0.2
2
948
1021
1104
5
5
7
18
?
13
1020
1105
1174
1206
3
1208
19
C(sp )-C str, ip CH and
CCC bend
7
8
8
8
9
9
9
04
09
12
16
1
7
1
4
6
3
6
8
1227
1282
1310
1451
1487
1585
1616
2985
2995
3005
3016
47
93
244
436
492
514
633
650
757
1
5
0.2
35
0.1
1
1
0.06
0.07
100
13
0.01
1
2
5
0.2
0.6
2
8
9
0.6
0.1
9
0.2
3
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
1222
1299
1352
1441
1474
1572
1591
3048
3058
3068
3080
46
92
240
424
481
513
629
644
735
6
0.2
2
29
0.01
1
1
0.5
1
100
9
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
CH str
1295
1351
1446
1480
1581
1628
2
2
40
?
2
4
28
0.6
10
0.4
27
18
∼1
1
1
1
1
CH str
CH str
CH str
e
2
0.01
0.6
2
ring tor and defm
ring tor and defm
ring tor, ip (CH+CCC) bend
ring tor
ring tor and defm
ring tor, ip (CH+CCC) bend
ip CH and CCC bend
ring tor, ip (CH+CCC) bend
oop CH and CCC bend
oop CH bend
4
0.003
0.2
0.01
0.04
0.5
0
0.5
0.002
0.1
0.07
10
17
20
23
27
29
33
37
43
3
419
482
514
618
636
727
750
21
3
3
0.4
0.5
2
7
7
29
56
100
5
18
57
0.01
780
898
99
0.01
752
864
oop CH bend

1144 J. Phys. Chem. A, Vol. 102, No. 7, 1998
Boo et al.
TABLE 1 (Continued)
HF/4-31G
B3LYP/4-31G
exptl
approximated
type of mode
sym
no.b
freq
936
I
IR
I
Ram
DPc
freq
904
I
IR
freq
908
I
IR
I
Ram
DPc
4
7
1
0.6
0.75
1
5
ip (CH+CCC) bend, oop
(
CH+CCC) bend
5
5
5
6
6
6
7
7
8
8
9
9
9
1
6
8
2
5
9
3
8
2
8
1
4
9
1001
1009
1037
1051
1101
1138
1164
1225
1304
1456
1483
1590
1614
2983
2993
3004
3016
0.2
2
0.3
0.04
0.04
0.02
0.07
6
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
947
985
0.2
0.02
1
946
5
oop CH and CCC bend
oop CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
ip CH and CCC bend
CH str
2
1005
1035
1103
1155
1179
1288
1305
1449
1473
1570
1594
3047
3056
3067
3079
1004
1031
7
7
0.01
0.1
6
2
0.00
0.06
2
6
0.3
19
5
0.02
1
3
1164
1176
1282
14
5
0.03
3
0.001
1
9
2
0.3
0.08
0.9
0.06
2
23
11
0.1
0.9
3
1446
1474
?
10
1604
5
1
1
1
1
02
06
10
14
9
10
0
50
2
11
0.7
8
0.2
51
CH str
CH str
CH str
a
Vibrational frequencies in cm-1. b Mode numbers are extracted from the output of the DFT calculation. c Depolarization ratio. d Approximate
mode descriptions are extracted from output of the DFT calculation. ip refers to in-plane; oop out-of-plane; tor to torsion, and the modes assigned
are sorted in the order of their contributions to the vibrational motions.
TABLE 2: Comparison of the Infrared and Raman Vibrational Frequencies and Intensities of 9,9′-Spirobifluorene (SBFL),
Bis(2,2′-biphenylene)silane (BBPS), and Bis(2,2′-biphenylene)germane (BBPG)a
freq (int %)
approximateb
sym
e
SBFL
BBPS
BBPG
type of mode
Infrared
749 (100)
1120 (16)
1303 (3)
1430 (52)
1461 (14)
1570 (10)
1585 (18)
750 (100)
1104 (7)
1282 (9)
1446 (?)
1446 (40)
1581 (4)
1604 (5)
751 (100)
1131 (35)
1309 (1)
1428 (59)
1459 (22)
1567 (13)
1590 (37)
oop(CH + CCC) bend, M-C
4
almost not in vibration
str
b
e
e
b
b
e
2
ip(CH + CCC) bend, M-C
ip(CH + CCC) bend, M-C
ip(CH + CCC) bend, M-C
ip(CH + CCC) bend, M-C
ip(CH + CCC) bend, M-C
ip(CH + CCC) bend, M-C
4
4
4
4
4
4
rocking
rocking
scissoring
str and scissoring
str and rocking
2
2
Raman
675 (16)
a
a
b
1
704 (18)
1020 (89)
1104 (7)
1152 (37)
1164 (14)
1224 (25)
1480 (54)
1580 (12)
1608 (100)
675 (18)
1039 (72)
1139 (10)
1123 (23)
1159 (16)
1231 (27)
1463 (58)
1563 (14)
1587 (100)
ip(CH + CCC) bend, M-C
ip(CH + CCC) bend, M-C
ip(CH + CCC) bend, M-C
ip(CH + CCC) bend, M-C
ip(CH + CCC) bend, M-C
ip(CH + CCC) bend, M-C
ip(CH + CCC) bend, M-C
ip(CH + CCC) bend, M-C
ip(CH + CCC) bend, M-C
4
4
4
4
4
4
4
4
4
scissoring
str
str
str
str
1
1035 (46)
1131 (11)
1119 (19)
1163 (19)
1231 (23)
1463 (53)
1563 (29)
1587 (100)
2
a
e
a
a
a
a
1
1
1
1
1
str
scissoring
scissoring
scissoring
a
Vibrational frequencies in cm-1. b Approximate mode descriptions are extracted from output of the DFT calculation. ip refers to in-plane; oop
to out-of-plane.
-
1
1
282, 1446, 1474, and 1604 cm . Most of the fundamentals
shown in Figure 8, the Raman intensities are almost invariant
with respect to the polarity change of the Si-C4 bonds due to
the presence of the low-electronegative central atom.
a2 Modes. Identification of the a2 vibrational frequencies is
also impossible here since their modes are infrared and Raman
inactive.
involve ring deformation and in-plane C-H and CCC bending
and C-H stretching modes.
Bis(2,2′-biphenylene)silane. We could assign only 25
fundamentals on the basis of the HF and DFT calculations. We
have optimized the scaling factors by minimizing the rms
deviation to yield 0.904 and 0.973 for HF/3-21G and B3LYP/
b1 Modes. Among the 10 fundamentals attributable to b1, only
-
1
3
-21G, respectively. When we use these scaling factors, the
one frequency 279 cm is identified as a b1 mode.
rms deviations of the scaled computed frequencies from
b2 Modes. Among the 20 b2 modes, only five modes are
-
1
-1
experiment are found to be 20.82 and 8.52 cm for HF/3-21G
and B3LYP/3-21G, respectively. It is noticed that since the
scaling factors are optimized using the fundamentals below 1650
identified. These are 530, 770, 1139, 1459, and 1567 cm .
-
1
The infrared intensities corresponding to 1131 and 1567 cm
are greatly enhanced in comparison with those of SBFL. The
DFT calculation indicates that the modes involve the Si-C4
antisymmetric stretching frequencies, of which vibrations change
the electric dipole moment.
e Modes. Among the 29 e frequencies, only 11 frequencies
are identified as the e mode. These are 407, 430, 479, 515,
-
1
cm , the rms values are relatively low. Figures 3 and 4 are
the results scaled by 0.973.
a1 Modes. Among the 20 fundamentals, eight fundamentals
that we identify as a1 fundamentals by Raman spectroscopy are
-
1
3
91, 675, 1039, 1123, 1287, 1463, 1563, and 1587 cm . As

Infrared and Raman Spectra of SBFL, BBPS, and BBPG
J. Phys. Chem. A, Vol. 102, No. 7, 1998 1145
References and Notes
-
1
6
19, 751, 761, 1159, 1258, 1428, and 1590 cm . As shown
-
1
in Figure 7 and Table 2, the IR intensity of 751 cm is invariant
with the change of the central atom. This is because the mode
involves only the out-of-plane C-H vibration; that is the Si-
C4 subunit is not in vibration. The IR intensity corresponding
(
(
1) Simmons, H. E.; Fukunaga, T. J. Am. Chem. Soc. 1967, 89, 5208.
2) Boschi, R.; Dreiding, A. S.; Heilbronner, E. J. Am. Chem. Soc.
1970, 92, 123.
(3) Hoffmann, R.; Imamura, A.; Zeiss, G. D. J. Am. Chem. Soc. 1967,
89, 5215.
-
1
to 1590 cm is, however, enhanced since the Si-C4 vibration
involves the stretching and rocking modes.
(
(
4) Goldstein, M. J.; Hoffmann, R. J. Am. Chem. Soc. 1971, 93, 6193.
5) Bischof, P.; Gleiter, R.; Haider, R. J. Am. Chem. Soc. 1978, 100,
Bis(2,2′-biphenylene)germane. We could safely assign only
1036.
2
0
6 fundamentals. The uniform scaling factors of 0.900 and
.968 for HF/3-21G and B3LYP/3-21G, respectively, are the
(6) Kao, J.; Radom, L. J. Am. Chem. Soc. 1978, 100, 760.
(7) Gordon, M. D.; Fukunaga, T.; Simmons, H. E. J. Am. Chem. Soc.
best values to minimize the rms deviations of the experimental
fundamentals from theory.
1976, 98, 8401.
(
8) Semmelhack, M. F.; Foos, J. S.; Katz, S. J. Am. Chem. Soc. 1973,
9
1
1
5, 7325.
The rms deviations of computed frequencies from experiment
(9) Schweig, A.; Weidner, U.; Hellwinkel, D.; Krapp, W. Angew. Chem.
-
1
turned out to be 22.90 and 15.43 cm for HF/3-21G and
B3LYP/3-21G, respectively.
973, 85, 360.
(10) Dantzig, N. A. v.; Piotrowiak, P.; Levy, D. H. Chem. Phys. Lett.
994, 223, 127.
a1 Modes. Among the 20 fundamentals, nine fundamentals
that we identify as a1 fundamentals are 275, 387, 675, 1035,
(
11) Dantzig, N. A. v.; Levy, D. H.; Vigo, C.; Piotrowiak, P. J. Chem.
Phys. 1995, 103, 4894.
12) Boo, B. H.; Choi, Y. S.; Kim, T.-S.; Kang, S. K.; Kang Y. H.;
-
1
1
119, 1231, 1287, 1563, and 1587 cm . The intensities of
(
the Raman bands at 675, 1035, 1119, 1231, 1463, 1563, and
Lee, S. Y. J. Mol. Struct. 1996, 377, 129.
(13) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J.
Phys. Chem. 1994, 98, 11623.
-1
1
587 cm listed in Table 2 are almost invariant with the change
of the central atom, indicating that the dipole changes do not
affect the Raman intensity.
a2 Modes. Identification of the a2 vibrational frequencies is
also impossible here since their modes are infrared and Raman
inactive.
(
14) Devlin, F. J.; Finley, J. W.; Stephens, P. J.; Frisch, M. J. J. Phys.
Chem. 1995, 99, 16883.
15) Handy, N. C.; Maslen, P. E.; Amos, R. D.; Andrews, J. S.; Murray,
(
C. W.; Laming, G. J. Chem. Phys. Lett. 1992, 197, 506.
(16) Handy, N. C.; Murray, C. W.; Amos, R. D. J. Phys. Chem. 1993,
7, 4392.
9
b1 Modes. Among the 10 fundamentals attributable to b1,
-
1
(17) El-Azhary, A. A.; Suter, H. U. J. Phys. Chem. 1995, 99, 12751.
only one frequency, 275 cm , is identified as a b1 mode.
b2 Modes. Among the 20 b2 modes, only seven modes can
be identified. These are 681, 758, 1051, 1120, 1307, 1461, and
(
18) Wheeless, C. J. M.; Zhou, X.; Liu, R. J. Phys. Chem. 1995, 99,
1
2488.
(19) Rauhut, G.; Pulay, P. J. Phys. Chem. 1995, 99, 3093.
-
1
1
1
570 cm . The IR intensities of fundamentals at 1120 and
(20) Johnson, B. G.; Gill, P. M. W.; Pople, J. A. J. Chem. Phys. 1993,
98, 5612.
-
1
570 cm are enhanced since the Ge-C4 vibrations include a
(
(
(
(
(
(
21) Langhoff, S. R. J. Phys. Chem. 1996, 100, 2819.
Ge-C antisymmetric stretching mode.
22) Lee, S. Y.; Boo, B. H. J. Phys. Chem. 1996, 100, 8782.
23) Lee, S. Y.; Boo, B. H. J. Phys. Chem. 1996, 100, 15073.
24) Clarkson, R. G.; Gomberg, M. J. Am. Chem. Soc. 1930, 52, 2881.
25) Gilman, H.; Gorsich, R. D. J. Am. Chem. Soc. 1955, 77, 6380.
26) Gilman, H.; Gorsich, R. D. J. Am. Chem. Soc. 1958, 80, 1883.
e Modes. Among the 29 e fundamentals, nine fundamentals
are identified as the e mode. These are 447, 488, 618, 668,
-
1
7
49, 1163, 1303, 1430, and 1585 cm . As shown in Figure 5
-
1
and Table 2, the IR intensity of 749 cm is invariant with the
change of the central atom. However, the IR intensity of
fundamental at 1585 cm is increased due to the presence of
the low electronegative Ge atom.
(27) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.; Petersson, G.
A.; Montgomery, J. A.; Raghavachari, K.; Al-Laham, M. A.; Zakrzewski,
V. G.; Ortiz, J. V.; Foresman, J. B.; Cioslowski, J.; Stefanov, B. B.;
Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala, P. Y.; Chen, W.;
Wong, M. W.; Andres, J. L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. P.; Head-
Gordon, M.; Gonzalez, C.; Pople, J. A. Gaussian 94, Revision C.2; Gaussian,
Inc.: Pittsburgh, PA, 1995.
-
1
Acknowledgment. This work is supported by the Ministry
of Education, Korea, through the Basic Science Research
Institute Program, 1996-1997, Project No. BSRI-96-3432.
B.H.B. is grateful to Center for Molecular Science (CMS) for
partial financial support.
(28) Szczepanski, J.; Vala, M.; Talbi, D.; Parisel, O.; Ellinger, Y. J.
Chem. Phys. 1993, 98, 4494.