Fourier-transform infrared observation of SiCn chains. I. the v4(σ) mode of linear SiC9 in Ar at 10 K

Article abstract of DOI:10.1063/1.479062

The linear SiC₉ cluster has been detected for the first time in Fourier-transform infrared spectra observed when the products from the laser evaporation of Si/C rods were trapped in Ar at ～10 K. Comparison of ¹³C isotopic shift measu

Full text of DOI:10.1063/1.479062

Fourier-transform infrared observation of SiCn chains. I. The ν4(σ) mode of
linear SiC9 in Ar at 10 K
X. D. Ding, S. L. Wang, C. M. L. Rittby, and W. R. M. Graham
Citation: J. Chem. Phys. 110, 11214 (1999); doi: 10.1063/1.479062
View online: http://dx.doi.org/10.1063/1.479062
View Table of Contents: http://jcp.aip.org/resource/1/JCPSA6/v110/i23
Published by the AIP Publishing LLC.
Additional information on J. Chem. Phys.
Journal Homepage: http://jcp.aip.org/
Journal Information: http://jcp.aip.org/about/about_the_journal
Top downloads: http://jcp.aip.org/features/most_downloaded
Information for Authors: http://jcp.aip.org/authors
Downloaded 25 Aug 2013 to 141.161.91.14. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jcp.aip.org/about/rights_and_permissions

JOURNAL OF CHEMICAL PHYSICS
VOLUME 110, NUMBER 23
15 JUNE 1999
Fourier-transform infrared observation of SiCn chains. I. The ␯4„␴… mode
of linear SiC9 in Ar at 10 K
X. D. Ding, S. L. Wang, C. M. L. Rittby,a) and W. R. M. Grahama)
Department of Physics and Astronomy, Texas Christian University, Fort Worth, Texas 76129
͑Received 2 February 1999; accepted 19 March 1999͒
The linear SiC9 cluster has been detected for the first time in Fourier-transform infrared spectra
observed when the products from the laser evaporation of Si/C rods were trapped in Ar at ϳ10 K.
Comparison of 13C isotopic shift measurements with the results of density functional theory
calculations ͑DFT͒ confirms the identification of the ␯4(␴) mode of SiC9 at 1935.8 cmϪ1. © 1999
American Institute of Physics. ͓S0021-9606͑99͒00123-3͔
INTRODUCTION
are all monosilicon-carbon clusters, and since many astro-
physical molecules have a carbon chain backbone, it is rea-
sonable to expect that SiCn clusters based on long carbon
chains are potentially important in this context. Using a tech-
nique involving the Nd-YAG laser evaporation of silicon-
carbon rods has now enabled us to produce SiCn species
consisting of long carbon chains with a single terminal Si.
Silicon-carbon clusters have generated considerable ex-
perimental and theoretical interest because of their novel
structures and their many technological applications. They
have also attracted the attention of astrophysicists because
several small silicon-carbon molecules have been discovered
in astronomical environments, in particular, SiC,1 SiC2,2,3
23,24
When a Si atom is added to one or both ends of C4
4
and SiC4 have been observed by radioastronomy in the cir-
and C5,26 both experiment and theory have shown that linear
structure of the original carbon chain is retained. In this pa-
per, we report the detection for the first time of a new
monosilicon-carbon cluster, SiC9, which like C9 is found to
be linear. Based on a comparison of measured 13C isotopic
shifts with the results of DFT calculations, the ␯4(␴) mode
cumstellar shell of the carbon star IRCϩ10216. SiC was first
identified by Bernath et al.5 In the laboratory and subse-
quently detected1 in the envelope of IRCϩ10216. In con-
trast, both SiC2 and SiC4 were first discovered in space and
then detected in the laboratory. Extensive laboratory studies
of electronic and rotational spectra for both SiC1,5–9 and
is assigned to an absorption at 1935.8 cmϪ1
.
10–13
SiC2
have also been reported.
Beginning with the earliest work by Weltner and
McLeod on SiC2,14 matrix investigations have played a sig-
nificant role in establishing the vibrational spectra and struc-
tures of silicon-carbon clusters. Other matrix infrared work
to establish the vibrational fundamentals of SiC2 followed,15
including Fourier-transform infrared FTIR measurements in
this laboratory.16 The collaboration of FTIR measurements
and ab initio theory has resulted in the first observations of
vibrational fundamentals of the other triatomic Si-C cluster,
THEORETICAL CALCULATIONS
To aid in the identification of the vibrational spectrum of
linear SiC9 we have performed density functional calcula-
tions at the standard B3LYP/cc-pVDZ level using the
GAUSSIAN94/DFT program suite.27 In addition to the linear
isomer, one could possibly expect to be able to form a cyclic
isomer of SiC9 just as is the case for the isovalent cluster
C10. Our experimental reaction dynamics, however, is likely
to strongly favor linear chains and no further consideration
will be given to the cyclic isomer here.
Si2C;17 the tetra-atomics, rhombic Si2C2 and rhomboidal
18
Si3C;19 the penta-atomics, linear SiC3Si20 and planar pen-
tagonal Si3C2;21 and the hexa-atomic cluster, linear SiC4Si.22
Matrix measurements of vibrational spectra have also pro-
vided guides for subsequent gas phase measurements at high
resolution. For example, the first observation of the ␯1(␴)
3
The geometry of the ⌺ linear ground state of SiC9 was
first optimized at the B3LYP/cc-pVDZ level, resulting in the
geometry parameters and dipole moment given in Table I.
Because of the on going interest in the observation of the
rotational spectra of linear SiCn chains, an additional set of
calculations was performed to investigate the behavior of the
equilibrium rotational constant ͑see Table II͒ and the dipole
moment for increasing values of n. The equilibrium rota-
tional constant is a direct consequence of the optimized ge-
ometry and the atomic masses. In the limit of large n, the
SiCn rotational constants should behave similarly to those of
pure carbon chains, which can be shown to go as ϳnϪ3 for
large n. This trend can be seen in Fig. 1. In Fig. 2 we show
the trends for the calculated dipole moment. The chains with
even n are singlet states whereas the ones with odd n have
triplet ground states. The addition of carbon atoms gives rise
and ␯3(␴u) fundamentals of SiC4 and SiC3Si,20 respec-
23
tively, by matrix FTIR in this laboratory quickly led to high-
resolution diode laser measurements in the gas phase.24,25
Although we have successfully formed and characterized
a variety of Si-C clusters by means of Ar matrix studies, it
has proved difficult to produce clusters containing more than
four C atoms using conventional high temperature evapora-
tion from a Knudsen cell. Since the three Si-C clusters de-
tected so far in astronomical sources, SiC, SiC2, and SiC4,
a͒
Authors to whom correspondence should be addressed; electronic mail:
w.graham@tcu.edu and m.rittby@tcu.edu
0021-9606/99/110(23)/11214/7/$15.00
11214
© 1999 American Institute of Physics
Downloaded 25 Aug 2013 to 141.161.91.14. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jcp.aip.org/about/rights_and_permissions

J. Chem. Phys., Vol. 110, No. 23, 15 June 1999
Ding et al.
11215
TABLE I. B3LYP/cc-pVDZ optimized geometry parameters for SiC9. The
equilibrium rotational constant for this geometry is Beϭ0.225 GHz with a
dipole moment of ␮ϭ8.81 Debye.
Parameter value
Bond
͑Å͒
Si-C1
C1-C2
C2-C3
C3-C4
C4-C5
C5-C6
C6-C7
C7-C8
C8-C9
1.733
1.286
1.295
1.282
1.287
1.289
1.281
1.302
1.298
to additional nonpolar bonds in the chain. One would thus
expect a near linear dependence of the dipole moment with
increasing n. As can be seen in Fig. 2, this is indeed the case
with slightly larger values for chains with even values of n.
The harmonic frequencies and infrared intensities for
linear SiC9 calculated at the B3LYP/cc-pVDZ level are
given in Table III. Because of their low absolute intensity,
the bending modes are unlikely to be observed in our experi-
ment. We focus instead on the behavior of the stretching
modes. Table IV gives the isotopically shifted frequencies
resulting from single 13C, 29Si, or 30Si substitution. As
Tables III and IV indicate, the modes involving the silicon
atom, i.e., the ones with substantial silicon isotopic shifts,
have as expected low frequencies but are, in addition, low in
absolute intensity. These modes are therefore unlikely as
candidates for observation and the presence of silicon in any
observed band has to be established by other means. Of the
remaining vibrational modes our best candidates for obser-
vation are thus ␯1(␴), ␯3(␴), ␯4(␴), and ␯6(␴).
FIG. 1. Equilibrium rotational constants calculated at the B3LYP/cc-pVDZ
level for linear SiCn chains with 2рnр10. Filled circles for even n, filled
triangles for odd n.
a range of levels of theory. The theoretical isotopic shifts for
the most intense band, ␯6(␴u) ͑1998.0 cmϪ1 in an Ar ma-
trix͒, did, however, show great variation. Because of con-
taminating bands we were at the time unable to establish if
the experimental shifts agreed with a pattern calculated at
any of the different levels of theory. New experiments,
where the contaminating bands have been eliminated, show,
however, that isotopic shifts calculated at the B3LYP/cc-
From our previous studies of pure linear carbon chains
we know that caution is warranted when considering the high
frequency modes of such clusters. In the case of C9 we found
that the normal mode displacements for the highest fre-
quency modes were very sensitive to the level of theory at
which they were calculated.28 As a result, there is the possi-
bility that a given theoretical calculation can give erroneous
isotopic shift patterns as well as unreliable infrared intensi-
ties. In our study, the ␯7(␴u) ͑1601.0 cmϪ1 in Ar matrix͒
stretching mode of C9 was found to be ‘‘well-behaved’’ and
could be identified based on its isotopic shifts calculated for
TABLE II. B3LYP/cc-pVDZ equilibrium rotational constants for linear
SiCn chains with 2рnр10.
n
Be ͑GHz͒
2
3
4
5
6
7
8
9
10
6.184
2.696
1.506
0.906
0.601
0.414
0.302
0.225
0.173
FIG. 2. Dipole moments calculated at the B3LYP/cc-pVDZ level for linear
SiCn chains with 2рnр10. Filled circles for even n, filled triangles for
odd n.
Downloaded 25 Aug 2013 to 141.161.91.14. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jcp.aip.org/about/rights_and_permissions

11216
J. Chem. Phys., Vol. 110, No. 23, 15 June 1999
Ding et al.
TABLE III. B3LYP/cc-pVDZ harmonic frequencies and infrared intensities
for linear SiC9.
Harmonic frequency
Infrared intensity
FIG. 3. Normal mode displacements for the ␯4(␴) mode of linear SiC9.
Vibrational mode
͑cmϪ1
͒
͑km/mol͒
␯1(␴)
␯2(␴)
␯3(␴)
␯4(␴)
␯5(␴)
␯6(␴)
␯7(␴)
␯8(␴)
␯9(␴)
2177
2130
2005
1937
1691
1343
992
654
353
722
615
515
388
268
176
95
2305
4
525
1034
12
167
35
13
frequency of the candidate band discussed below. The
shifted peaks are labeled in a scheme similar to that pre-
sented in Table V. There is hardly any resemblance between
the two patterns, a behavior similar to that encountered with
C9. The positive identification in the latter case, however,
indicates that our present B3LYP/cc-pVDZ calculation could
be able to predict the isotopic shifts of SiC9 relatively well.
Care should, however, be taken when comparing any of the
isotopic data in Table IV with experimental results, in par-
ticular with regards to high frequency modes.
10
␯
␯
␯
␯
␯
␯
␯
␯
10(␲)
11(␲)
12(␲)
13(␲)
14(␲)
15(␲)
16(␲)
17(␲)
0.03
1.1
1.8
0.02
5.8
0.7
4.7
0.7
EXPERIMENT
36
Matrix samples were prepared by vaporizing a Si/C or
pure C rod using a focused Nd:YAG ͑GCR-11, Spectra-
Physics͒ pulsed laser operating at 532 nm. The laser ablation
procedure typically required about 15–30 min for silicon-
carbon cluster spectrum or 5–10 min for a pure carbon clus-
ter spectrum. The evaporated species were condensed in
solid Ar ͑Matheson, 99.9995% purity͒ on a gold-plated sur-
face cooled to ϳ10 K by a closed cycle refrigerator ͑APD
Cryogenics, Displex͒ in a vacuum of 10Ϫ7 Torr or better dur-
ing the experiment.
The Si/C rods were made by mixing silicon powder
͑Alfa, 99.999%͒ and carbon powder (12C: Alfa, 99.9995%
and 13C: Isotec, 99.3%͒ with any desired ratio and isotopic
enrichment and then pressing the mixtures into rods. A de-
gassing procedure was used to remove any trapped gas by
heating the rods to ϳ200 °C under vacuum for more than
24 h.
pVDZ level can be used to identify the ␯ mode of C9, albeit
6
with somewhat larger discrepancies between theory and ex-
periment than was found for the ␯7(␴u) mode. With regards
to the calculated infrared intensities, the ␯5(␴u) mode
͑2078.1 cmϪ1͒ was predicted at the DFT level to be 0.84
times that of the ␯6(␴u) mode ͑1998.0 cmϪ1͒,28 but was
actually measured to be only 0.02 times as intense. The ob-
served intensity was thus smaller than the calculated, by a
factor of ϳ40.
In light of the discussion above, care should be taken
when interpreting the isotopic shifts, as well as the infrared
intensities, from our B3LYP/cc-pVDZ calculation. The ex-
perimental discussion below will focus on the ␯4(␴) mode
of SiC9, see Fig. 3. To illustrate the sensitivity of this mode
to the level of calculation we performed a calculation of
isotopic shifts at the Hartree–Fock ͑HF͒ level ͑with the cc-
pVDZ basis set͒ and compared the results with our B3LYP/
cc-pVDZ results. Figure 4 shows a comparison of the simu-
lated spectra resulting from single 13C substitution. The main
absorption, labeled a, has been scaled to coincide with the
FTIR absorption spectra of the matrix samples were re-
corded in the region of 550–3900 cmϪ1 at a resolution of 0.2
cmϪ1 using a Bomem DA 3.16 Fourier-transform spectrom-
eter equipped with a liquid nitrogen cooled MCT ͑Hg-Cd-
Te͒ detector and KBr beamsplitter. Details of the optical sys-
tem have been reported previously.29 All frequencies
reported were measured to Ϯ0.1 cmϪ1
.
TABLE IV. B3LYP/cc-pVDZ singly substituted isotopic shift frequencies ͑cmϪ1͒ for the ␴ modes of linear
SiC9.
Isotopomer
Si-C-C-C-C-C-C-C-C-C
␯
␯
␯
␯
␯
␯
␯
␯
␯
9
1
2
3
4
5
6
7
8
28-12-12-12-12-12-12-12-12-12 2177.4 2129.8 2004.6 1936.9 1691.2 1342.5 991.9 654.0 352.8
28-13-12-12-12-12-12-12-12-12 2177.2 2125.5 2004.2 1925.4 1678.2 1330.4 984.4 653.1 352.0
28-12-13-12-12-12-12-12-12-12 2173.6 2119.3 1997.6 1912.9 1688.5 1341.3 985.6 651.3 352.5
28-12-12-13-12-12-12-12-12-12 2164.5 2129.3 1980.5 1936.9 1679.2 1337.6 991.0 719.6 352.8
28-12-12-12-13-12-12-12-12-12 2171.3 2116.5 1989.0 1928.4 1691.1 1331.3 990.7 718.4 352.8
28-12-12-12-12-13-12-12-12-12 2177.2 2101.4 2000.4 1935.7 1674.8 1341.7 985.7 717.2 352.5
28-12-12-12-12-12-13-12-12-12 2163.6 2119.0 2000.9 1920.3 1689.8 1336.7 985.2 716.5 352.0
28-12-12-12-12-12-12-13-12-12 2155.3 2127.4 1992.3 1936.7 1682.4 1332.3 990.7 715.8 351.5
28-12-12-12-12-12-12-12-13-12 2170.6 2120.7 1995.5 1918.2 1683.4 1342.2 990.9 719.9 350.9
28-12-12-12-12-12-12-12-12-13 2176.7 2127.7 2003.0 1930.8 1683.2 1334.4 984.8 721.9 350.6
29-12-12-12-12-12-12-12-12-12 2177.4 2129.7 2004.6 1936.8 1691.1 1342.1 990.8 651.1 350.0
30-12-12-12-12-12-12-12-12-12 2177.4 2129.7 2004.6 1936.8 1691.0 1341.8 989.9 648.4 347.2
Downloaded 25 Aug 2013 to 141.161.91.14. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jcp.aip.org/about/rights_and_permissions

J. Chem. Phys., Vol. 110, No. 23, 15 June 1999
Ding et al.
11217
mixture of 50% Si and 50% C powder. Si powder contains
the 28Si, 29Si, and 30Si isotopes in 92.2%, 4.7%, and 3.1%
natural abundance, respectively.
In the pure carbon cluster spectrum ͑bottom, Fig. 5͒ all
of the linear clusters ranging from C3 to C9, except C8, are
observed with particularly high yields of C3, C6, and C9. The
characteristic (e ) mode of cyclic (D3h)C6, which this lab
v4
Ј
reported earlier,30 also appears at 1694.9 cmϪ1
.
In the silicon-carbon cluster spectrum ͑top, Fig. 5͒ all of
the pure carbon clusters, appearing in the pure carbon spec-
trum, are again observed with very high yields of C3, C6, and
C9. However, in addition to the known Cn bands, there are
new absorptions which appear only when Si is present. After
careful examination, many of these new bands can be as-
signed to modes of silicon-carbon clusters, which we have
reported previously. The 1741.3 and 824.3 cmϪ1 bands are,
respectively, the ␯ and ␯ modes of linear SiC2,16 the
1
2
2080.1 cmϪ1 band is the ␯ mode of linear SiC4,23 and the
1
1955.2 and 898.9 cmϪ1 absorptions are the ␯ and ␯ modes,
3
4
respectively, of linear SiC3Si.20 Of the new bands, which
remain unassigned, the absorption at 1935.8 cmϪ1 is ex-
tremely intense. Because this band is not observed in the
pure carbon spectrum, it cannot belong to a carbon cluster.
There are thus only three possibilities left for its identity.
One is that it is a band of a contaminant; however, this can
easily be eliminated because, as indicated in the top spec-
trum of Fig. 5, other benchmark contaminant bands such as
SiH4, H2O, SiN2, SiCO, CO, and CO2 are very weak. An-
other possibility is that the carrier of the 1935.8 cmϪ1 ab-
sorption is a pure silicon cluster; however, this is unlikely
since the vibrational modes for the pure silicon clusters31,32
(Sin , nϭ2–8͒ have been predicted and, in many cases,
observed33 to lie below 700 cmϪ1. Thus the third possibility,
that it is a silicon-carbon cluster, is the most plausible expla-
nation for the carrier of the 1935.8 cmϪ1 absorption.
All of the silicon-carbon clusters identified in the top
spectrum in Fig. 5 are in the form of either SiCn ͑such as
SiC2 and SiC4͒ or SiCnSi ͑such as SiC3Si͒, i.e., a carbon
chain with a Si atom bonded to one or both of the terminal
carbons. None of the known silicon-carbon clusters with
more than two silicon atoms, such as Si3C19 or Si3C2,21 is
observed in our experiments. It is therefore reasonable to
FIG. 4. Comparison of simulated spectra resulting from single 13C substi-
tution calculated at the HF/cc-pVDZ and B3LYP/cc-pVDZ levels of theory.
The absorptions are labeled according to the scheme presented in Table V.
RESULTS AND DISCUSSION
Figure 5 shows typical FTIR spectra in the 700 to 2350
cmϪ1 region, which were obtained using the laser ablation
technique. The mid-IR ͑infrared͒ region shown is where
most of the stretching modes of silicon-carbon clusters are
expected to be located. Two spectra are plotted together for
easy comparison. The bottom spectrum is observed for a
sample rod made from pure C powder ͑containing 1% 13C in
natural abundance͒, and the top one, for a rod made from a
TABLE V. Comparison of observed vibrational frequencies ͑cmϪ1͒ of the ␯4(␴) mode for 13C substituted
isotopomers of SiC9 with the predictions of B3LYP/cc-pVDZ calculations.
Isotopomer
Observed
B3LYP/cc-pVDZ
Scaleda
Difference
Si-C-C-C-C-C-C-C-C-C
␯
␯
␯
⌬␯
28-12-12-12-12-12-12-12-12-12
28-13-12-12-12-12-12-12-12-12
28-12-13-12-12-12-12-12-12-12
28-12-12-13-12-12-12-12-12-12
28-12-12-12-13-12-12-12-12-12
28-12-12-12-12-13-12-12-12-12
28-12-12-12-12-12-13-12-12-12
28-12-12-12-12-12-12-13-12-12
28-12-12-12-12-12-12-12-13-12
28-12-12-12-12-12-12-12-12-13
͑A͒
͑B͒
͑C͒
͑D͒
͑E͒
͑F͒
͑G͒
͑H͒
͑I͒
1935.8
1925.3
1912.2
overlapped
1928.4
1934.3
1918.6
overlapped
1917.4
1936.9
1925.4
1912.9
1936.9
1928.4
1935.7
1920.3
1936.7
1918.2
1930.8
1935.8
1924.3
1911.8
1935.8
1927.3
1934.6
1919.2
1935.6
1917.1
1929.7
0.0
1.0
0.4
¯
1.1
Ϫ0.3
Ϫ0.6
¯
0.3
͑J͒
1931.5
1.8
aResults of the B3LYP/cc-pVDZ calculation scaled by a factor of 1935.8/1936.9.
Downloaded 25 Aug 2013 to 141.161.91.14. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jcp.aip.org/about/rights_and_permissions

11218
J. Chem. Phys., Vol. 110, No. 23, 15 June 1999
Ding et al.
FIG. 5. Typical FTIR spectra of pure carbon clusters ͑bottom͒ and silicon-carbon clusters ͑top͒ produced by the laser evaporation of C and Si/C rods,
respectively. The top spectrum is shifted 5 cmϪ1 to higher frequency relative to the lower spectrum to facilitate comparison.
expect that the silicon-carbon cluster with absorption at
1935.8 cmϪ1 is also of the form of SiCn or SiCnSi. Further-
more, the high yields of C3, C6, and C9 make the SiC3, SiC6,
SiC9, SiC6Si, and SiC9Si clusters ͑noticing that SiC3Si has
already been identified21͒ leading candidates.
The two FTIR spectra showing the region near the
1935.8 cmϪ1 band in Fig. 6 were recorded under identical
conditions using two different sample rods. For the bottom
spectrum, which is just a portion of the top spectrum in Fig.
5, a rod made from a mixture of 50% Si and 50% C powder
was used. The top spectrum was observed using a rod fabri-
cated from a 50% Si, 45% C, and 5% 13C mixture, i.e., with
10% 13C isotopic enrichment of the carbon powder. As a
result of this enrichment, new absorption bands due to
13C-substituted isotopomers appear in the upper spectrum.
In the vicinity of the 1935.8 cmϪ1 band, there is only one
intense absorption, the ␯4(␴u) mode of C6 at 1952.5 cmϪ1
whose 13C isotopic shifts may mix with those of the 1935.8
FIG. 6. Comparison of FTIR spectra produced by the
evaporation of a Si/C rod with 10% 13C enrichment
͑top͒ and with 1% natural abundance of 13C ͑bottom͒.
The letters identifying the isotopomer bands of SiC9
correspond to the labeling scheme shown in Table V.
Downloaded 25 Aug 2013 to 141.161.91.14. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jcp.aip.org/about/rights_and_permissions

J. Chem. Phys., Vol. 110, No. 23, 15 June 1999
Ding et al.
11219
TABLE VI. B3LYP/cc-pVDZ predicted shifts for modes of SiCn chains (5рnр9) at frequencies of ␻
у1800 cmϪ1
.
Frequency
͑cmϪ1
Isotopic shifts ͑scaleda͒
Species
SiC5
Mode
͒
IR intensity
͑cmϪ1
͒
2116.3
1966.4
2225.1
2140.8
1877.3
2134.3
2093.6
1949.1
2240.9
2191.6
2057.4
1885.6
2177.4
2129.8
2004.6
1936.9
1935.8
315
492
4694
108
1275
1595
12
140
958
4141
5854
43
1.3
1.1
0.1
4.5
2.6 13.0 18.1 32.3
7.5 14.5 27.9 29.8
8.9 22.0 28.2
9.0 12.0 21.1 28.1
5.1 10.3 11.9 13.7 19.5 20.3
␯1(␴)
␯2(␴)
␯1(␴)
␯2(␴)
␯3(␴)
␯1(␴)
␯2(␴)
␯3(␴)
␯1(␴)
␯2(␴)
␯3(␴)
␯4(␴)
␯1(␴)
␯2(␴)
␯3(␴)
␯4(␴)
SiC6
SiC7
SiC8
1.3
5.4
3.4
0.4
0.6
0.1
0.0
0.1
1.5
5.9
0.2
0.5
0.3
0.0
1.8
0.9
1.4
1.8
0.7
1.8
6.0
0.2
2.0
1.5
0.2
2.9
2.7
6.9
2.0
1.3
5.3
8.6
0.7
2.2
3.5
1.2
1.5
6.1
6.2 14.5 26.4
5.3 11.5 27.8 33.3
8.2 13.3 25.1 26.9
5.1
1.5
5.7
7.6
2.9 20.1 21.2 31.7
7.6 8.1 21.2 25.1
9.0 11.3 22.9
8.9 11.5 12.2 14.7 15.1
SiC9
2305
5
525
1034
3.4
3.9
4.1
6.1
4.3
5.4
8.3
6.7
6.1 11.5 12.3 19.6
9.5 9.9 12.1 25.8
8.8 11.9 15.0 23.2
8.5 11.5 16.6 18.7 24.0
7.4 10.5 17.2 18.4 23.6
Observed shifts for
aThe scale factor is the 1935.8 cmϪ1 frequency of the observed absorption divided by the predicted frequency
of the mode.
cmϪ1 band. Fortunately, all of these isotopic shifts for both
single and double 13C substitutions have previously been
identified by this lab.34 The 1929.1 cmϪ1 band corresponding
to a single 13C substitution, and the 1913.5, 1922.6, and
1923.1 cmϪ1 bands, to double 13C substitutions, can thus be
definitely excluded as shifts for the 1935.8 cmϪ1 band. The
1915.7, 1926.8, and 1930.0 cmϪ1 bands can also be elimi-
nated as potential 13C isotopic shifts because they appear in
the bottom spectrum recorded without 13C isotopic enrich-
ment. It is concluded, therefore, that the seven remaining
bands, 1934.3, 1931.5, 1928.4, 1925.3, 1918.6, 1917.4, and
1912.2 cmϪ1 and labeled F, J, E, B, G, I, and C, respectively
͑see Table V͒, all correspond to single 13C isotopic shifts of
the 1935.8 cmϪ1 band.
shifts in Table VI can mainly be attributed to the remaining
interactions between ␯4(␴) and the other modes SiC9 of
similar frequency.
As can be seen from the B3LYP/cc-pVDZ results in
Table III, ␯4(␴) is predicted to be one of the three most
intense modes, and its frequency of 1936.9 cmϪ1 is ex-
tremely close to the 1935.8 cmϪ1 frequency of the candidate
absorption. A detailed comparison between the observed iso-
topomer frequencies and the B3LYP/cc-pVDZ predictions in
Table V demonstrates the excellent agreement. It should be
noted that the 29Si and 30Si isotopic shifts for the ␯ mode
4
arising from their respective 4.7% and 3.1% natural abun-
dances are not observable because, as shown in Table IV,
their predicted shifts of 0.1 cmϪ1 would be overlapped by the
envelope of the main band. The good agreement between the
observed and predicted mode frequencies and the very good
agreement between the measured and calculated isotopic
shifts ͑as shown in Table V͒ strongly support the assignment
The existence of seven 13C shifts means that there are at
least 7 carbon atoms in the unidentified silicon-carbon clus-
ter if it is of the form SiCn , or at least 13 carbon atoms due
to the center of symmetry if it is of the form SiCnSi. Thus,
from the five candidate silicon-carbon clusters mentioned
earlier, SiC9 is left as the potential carrier of the 1935.8 cmϪ1
band. The frequencies of the ␴ modes and their isotopic
shifts have been calculated for all linear SiCn and SiCnSi
(nр9) clusters at the B3LYP/cc-pVDZ level. Comparison
of the 1935.8 cmϪ1 frequency and its isotopic shifts with the
predicted shifts for modes with ␻у1800 cmϪ1 ͑see Table
VI͒ confirms that only the ␯4(␴) mode of SiC9 fits the mea-
surements. It should be noted that, as is the case for pure
carbon clusters, the accuracy of calculated isotopic shifts
presented in Table VI depends on the accuracy of the full
force constant matrix. In particular, one would expect larger
uncertainties in these values when two or more modes have
similar frequencies. As noted above, we have found that cal-
culations at the B3LYP/cc-pVDZ level seem to give reliable
results for linear carbon chains. It is therefore expected that
the calculated force constant matrix for SiC9 is relatively
accurate. The discrepancies between observed and calculated
of the 1935.8 cmϪ1 absorption to the ␯ mode of SiC9.
4
By contrast, the calculated IR intensities of the vibra-
tional modes of SiC9 ͑see Table III͒ may appear at first
glance to weaken this assignment. The ␯ mode predicted at
1
2176.4 cmϪ1 is calculated to be the most intense mode, or
2.4 times the intensity ␯ mode we have identified in our
4
spectra as the 1935.8 cmϪ1 band, yet there is no evidence of
a suitable band in the 2000–2200 cmϪ1 region. As was indi-
cated earlier, however, the calculated infrared intensities in
the high frequency range can be expected to be unreliable
just as was the case for C9 where the intensity of the ␯5(␴u)
was overestimated by a factor of ϳ40. If the predicted inten-
sity of the ␯ mode of SiC9 was overestimated by a similar
1
factor, this mode would then be only 0.06 times that of the
␯ mode, making it too weak to be observable in the present
4
work. A similar uncertainty in the DFT intensity predictions
must apply to the ␯ mode which has not been observed in
3
the present work, although it is predicted at 2005.4 cmϪ1
Downloaded 25 Aug 2013 to 141.161.91.14. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jcp.aip.org/about/rights_and_permissions

11220
J. Chem. Phys., Vol. 110, No. 23, 15 June 1999
Ding et al.
with an intensity ϳ0.5 that of the observed ␯ mode at
13 M. Bogey, C. Demuynck, J. L. Destombes, and A. D. Walters, Astron.
Astrophys. 247, L13 ͑1991͒.
4
1935.8 cmϪ1
.
14 W. Weltner, Jr., and D. McLeod, Jr., Chem. Phys. 41, 235 ͑1964͒.
15 V. E. Bondybey, J. Phys. Chem. 86, 3396 ͑1982͒.
CONCLUSION
The linear SiC9 cluster produced by trapping the prod-
ucts of the laser evaporation of Si/C rods in an Ar matrix has
been detected for the first time. FTIR measurements of the
␯4(␴) fundamental at 1935.8 cmϪ1 and its 13C isotopic shifts
are in excellent agreement with the predictions of B3LYP/
cc-pVDZ DFT calculations. SiC9 is the first silicon-carbon
cluster with more than six atoms discovered so far.
16
´
J. D. Presilla-Marquez, W. R. M. Graham, and R. A. Shepherd, J. Chem.
Phys. 93, 5424 ͑1990͒.
17
18
19
20
21
22
´
J. D. Presilla-Marquez and W. R. M. Graham, J. Chem. Phys. 95, 5612
͑1991͒; C. M. L. Rittby, J. Chem. Phys. 95, 5609 ͑1990͒.
´
J. D. Presilla-Marquez, S. C. Gay, C. M. L. Rittby, and W. R. M. Graham,
J. Chem. Phys. 102, 6354 ͑1995͒.
C. M. L. Rittby, J. Chem. Phys. 96, 6768 ͑1992͒; J. D. Presilla-Marquez
and W. R. M. Graham, ibid. 96, 6509 ͑1992͒.
´
´
J. D. Presilla-Marquez and W. R. M. Graham, J. Chem. Phys. 100, 181
͑1994͒; C. M. L. Rittby, ibid. 100, 175 ͑1994͒.
´
J. D. Presilla-Marquez, C. M. L. Rittby, and W. R. M. Graham, J. Chem.
ACKNOWLEDGMENTS
Phys. 104, 2818 ͑1996͒.
´
J. D. Presilla-Marquez, C. M. L. Rittby, and W. R. M. Graham, J. Chem.
Grants from the Welch Foundation ͑Graham, No.
P-0786; Rittby, No. P-1259͒ and the TCU Research and Cre-
ative Activities Fund in support of this research, and from
the W. M. Keck Foundation for the Bomem spectrometer,
are gratefully acknowledged.
Phys. 106, 8367 ͑1997͒.
23 P. A. Withey and W. R. M. Graham, J. Chem. Phys. 96, 4068 ͑1992͒.
24 A. Van Orden, R. A. Provencal, T. F. Giesen, and R. J. Saykally, Chem.
Phys. Lett. 237, 77 ͑1995͒.
25 A. Van Orden, T. F. Giesen, R. A. Provencal, H. J. Hwang, and R. J.
Saykally, J. Chem. Phys. 101, 10237 ͑1994͒.
26 S. Hunsicker and R. O. Jones, J. Chem. Phys. 105, 5048 ͑1996͒.
27 M. J. Frisch, G. W. Trucks, H. B. Schlegel, P. M. W. Gill, B. G. Johnson,
M. A. Robb, J. R. Cheeseman, T. Keith, G. A. Petersson, J. A. Montgom-
ery, K. Raghavachari, M. A. Al-Laham, V. G. Zakrzewski, J. V. Ortiz, J.
B. Foresman, J. Cioslowski, B. B. Stefanov, A. Nanayakkara, M. Challa-
combe, C. Y. Peng, P. Y. Ayala, W. Chen, M. W. Wong, J. L. Andres, E.
S. Replogle, R. Gomperts, R. L. Martin, D. J. Fox, J. S. Binkley, D. J.
Defrees, J. Baker, J. P. Stewart, M. Head-Gordon, C. Gonzalez, and J. A.
Pople, GAUSSIAN 94, Revision E1 ͑Gaussian Inc., Pittsburgh, PA, 1995͒.
28 R. H. Kranze, P. A. Withey, C. M. L. Rittby, and W. R. M. Graham, J.
Chem. Phys. 103, 6841 ͑1995͒.
1 J. Cernicharo, C. A. Gottlieb, M. Guelin, P. Thaddeus, and J. M. Vrtilek,
Astrophys. J. 341, L25 ͑1989͒.
2 B. Kleman, Astrophys. J. 123, 162 ͑1956͒.
3 P. Thaddeus, S. E. Cummins, and R. A. Linke, Astrophys. J. 283, L45
͑1984͒.
4 M. Ohishi, N. Kaifu, K. Kawaguchi, A. Murakami, S. Saito, S. Yama-
moto, S. Ishikawa, Y. Fujita, Y. Shiratori, and W. M. Irvine, Astrophys. J.
345, L83 ͑1989͒.
5 P. F. Bernath, S. A. Rogers, L. C. O’Brien, C. R. Brazier, and A. D.
McLean, Phys. Rev. Lett. 60, 197 ͑1988͒.
6 C. R. Brazier, L. C. O’Brien, and P. F. Bernath, J. Chem. Phys. 91, 7384
͑1989͒.
29 L. N. Shen, T. J. Doyle, and W. R. M. Graham, J. Phys. Chem. 93, 1597
͑1990͒.
7 M. Ebben, M. Drabbels, and J. J. ter Meulen, Chem. Phys. Lett. 176, 404
͑1990͒.
30 S. L. Wang, C. M. L. Rittby, and W. R. M. Graham, J. Chem. Phys. 107,
6032 ͑1997͒.
8 T. J. Butenhoff and E. A. Rohlfing, J. Chem. Phys. 95, 3939 ͑1991͒.
9 R. Mollaaghababa, C. A. Gottlieb, J. M. Vrtilek, and P. Thaddeus, Astro-
phys. J. 352, L21 ͑1990͒.
31 K. Raghavachari and C. M. Rohlfing, J. Chem. Phys. 89, 2219 ͑1988͒.
32 R. Fournier, S. B. Sinnott, and A. E. DePristo, J. Chem. Phys. 97, 4149
͑1992͒.
10 D. L. Michalopoulos, M. E. Geusic, P. R. R. Langridge-Smith, and R. E.
Smalley, J. Chem. Phys. 80, 3556 ͑1984͒.
33 S. Li, R. J. Van Zee, W. Weltner, Jr., and K. Raghavachari, Chem. Phys.
Lett. 243, 275 ͑1995͒.
11 T. J. Butenhoff and E. A. Rohlfing, J. Chem. Phys. 95, 1 ͑1991͒.
12 C. A. Gottlieb, J. M. Vrtilek, and P. Thaddeus, Astrophys. J. 343, L29
͑1989͒.
34 R. H. Kranze and W. R. M. Graham, J. Chem. Phys. 98, 71 ͑1993͒.
Downloaded 25 Aug 2013 to 141.161.91.14. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jcp.aip.org/about/rights_and_permissions