An experimental and theoretical study of spin-spin coupling in chlorosilanes

Article abstract of DOI:10.1021/jp060704g

An experimental and theoretical study of the absolute value of the one-bond spin-spin coupling constant |¹J(Si,H)| in SiH _nCl_4-n (n = 0-4) dissolved in THF-d₈ is presented. We found |¹J(Si,H)| to increase with an increasing number of chlorine substituents, and the quantitative changes were found to differ from the values previously reported for the same compounds dissolved in cyclohexane-d₁₂. We also report on the variations in | ¹J(Si,H)| as a function of temperature, which we found to be linearly temperature dependent for the chlorine-substituted silanes and temperature independent for SiH₄. Furthermore, the temperature dependence of |¹J(Si,H)| varied between the different chlorosilanes. Solvent-solute interactions were studied by quantum chemical DFT calculations. The variations in chloro-silane bond lengths upon adduct formation and the different adduct interaction energies may explain the temperature dependences of the coupling constants.

Full text of DOI:10.1021/jp060704g

J. Phys. Chem. A 2006, 110, 9801-9804
9801
An Experimental and Theoretical Study of Spin-Spin Coupling in Chlorosilanes
Knut Thorshaug,* Ole Swang, Ivar M. Dahl, and Anja Olafsen
Department of Hydrocarbon Process Chemistry, SINTEF Materials and Chemistry,
P.O. Box 124, Blindern, N-0314 Oslo, Norway
ReceiVed: February 2, 2006; In Final Form: June 16, 2006
An experimental and theoretical study of the absolute value of the one-bond spin-spin coupling constant
1
1
|
J(Si,H)| in SiH
n
Cl4-n (n ) 0-4) dissolved in THF-d
8
is presented. We found | J(Si,H)| to increase with an
increasing number of chlorine substituents, and the quantitative changes were found to differ from the values
previously reported for the same compounds dissolved in cyclohexane-d12. We also report on the variations
1
in | J(Si,H)| as a function of temperature, which we found to be linearly temperature dependent for the chlorine-
1
substituted silanes and temperature independent for SiH
4
. Furthermore, the temperature dependence of | J(Si,H)|
varied between the different chlorosilanes. Solvent-solute interactions were studied by quantum chemical
DFT calculations. The variations in chloro-silane bond lengths upon adduct formation and the different
adduct interaction energies may explain the temperature dependences of the coupling constants.
Introduction
techniques. THF-d8 was dried according to standard procedures
prior to use. SiCl4 (Aldrich) and MgH2 (Alfa Aesar) were used
as received.
Sample Preparation. In a typical preparation, MgH2 (28 mg,
Nuclear magnetic resonance (NMR) spectroscopy may be
used to obtain molecular information by measuring the chemical
shifts and spin-spin coupling constants, and in the case of
1
.1 mmol) and THF-d8 (0.76 mL) were added to an NMR tube.
1
chloro-substituted silanes SiHnCl4-n (n ) 0-4), both the H and
The mixture was carefully shaken before SiCl4 (75 µL, 0.66
mmol) was added. Finally, the tube was sealed under Ar. The
reaction was allowed to take place for 3 days at room
temperature, forming a mixture of the various chlorosilanes from
which the reported data were obtained.
2
9
Si nuclei can easily be probed. To the best of our knowledge,
the number of systematic studies of all the said compounds n )
0
-4 is limited to the two papers by Ebsworth and Turner, who,
1
in the early 1960s, reported both the coupling constants and
the chemical shifts in the nonpolar solvent cyclohexane-d12.
2
1
Instrument. H NMR spectra were measured on a Varian
The recent developments within the field of theoretical chem-
istry have made it possible to calculate various parameters
2
9
Gemini 2000 using 32 scans and Si NMR on a Varian VXR-
3
00S using 5000 scans.
3
relevant to the NMR observables, but the calculations are
Variable Temperature. The instrument temperature calibra-
mostly concerned with molecules in the gas phase. Various
levels of quantum chemical calculations of NMR parameters
tion was performed by employing a methanol thermometer in
the range 194.3 K < T < room temperature in steps of 10 K.
Computational Details. DFT calculations were carried
4
for SiH4 in the gas phase have been presented in the literature,
but in the case of chloro-substituted derivatives, such calcula-
tions have not been reported.
5
6
out using the Spartan 5.1 (BP86 functional and DNP basis
7
8
9
set ) and Gaussian 03 (B3LYP functional and aug-cc-pvtz
Although experimental data on the SiHnCl4-n (n ) 0-4)
system have already been presented, we believe further studies
of the compounds are merited due to the advances in the NMR
instruments and computational possibilities made over the past
decades. Furthermore, the presence of a polar NMR solvent,
e.g. THF-d8, may perturb the silanes (n ) 0-4) to a significant
extent, resulting in changes in the one-bond spin-spin coupling
constants compared to the values obtained in less polar solvents.
Thus, by combining experiments and quantum chemical calcula-
tions, the aim of the present work is to present new and reliable
data on the absolute value of the one-bond spin-spin coupling
1
0
basis set ) program packages. For the B3LYP results, coun-
terpoise calculations for estimation of basis set superposition
errors (BSSE) were performed using the algorithm built into
Gaussian 03.
For each silane-THF adduct, a number of different conform-
ers were investigated to ensure that the global energy minimum
was found. From the calculated IR spectra, we could confirm
that each stationary state is a true minimum, as witnessed by
the absence of imaginary vibrational frequencies.
1
Results and Discussion
constants | J(Si,H)| for chlorosilanes in THF-d8 and to make
an attempt at rationalizing the observed experimental data.
Chlorine Substitution. The absolute value of each one-bond
1
spin-spin coupling constant | J(Si,H)| was measured as the
Experimental and Computational Methods
1
separation of the Si-satellites in the H NMR spectra or as the
separation between the signals in the various multiplets in the
General Considerations. All manipulations were performed
using standard techniques for handling of air-sensitive com-
pounds, either inside a glovebox or using standard Schlenk
29
Si NMR spectra. As seen from Table 1 and Figure 1, the devi-
1
29
ations between the H and Si NMR experimentally determined
1
|
J(Si,H)| values are in the range 0.1-0.8 Hz. Although we note
1
that H NMR systematically provides a slightly lower value,
the difference is insignificant.
*
Corresponding author. Telephone: (+47) 98243936. Fax: (+47)
2
2067350. E-mail: knut.thorshaug@sintef.no.
1
0.1021/jp060704g CCC: $33.50 © 2006 American Chemical Society
Published on Web 07/22/2006

9802 J. Phys. Chem. A, Vol. 110, No. 32, 2006
Thorshaug et al.
TABLE 1: Experimentally Determined H and 29Si NMR
1
Data for SiH
n
Cl4-n (n ) 0-4)a
1H NMR
29Si NMR
this workb
literaturec
this workd
1
1
1
δ
H
| J(Si,H)|
| J(Si,H)|
δ
Si
| J(Si,H)|
SiCl
SiHCl
4
-
-
-
-18.55
-9.52
-16.83
-50.87
-95.84
-
3
6.09
5.32
4.48
3.14
377.5
308.5
257.4
203.0
362.9
288.0
238.1
202.5
378.0
308.9
258.2
203.1
SiH
SiH
SiH
2
3
4
Cl
Cl
2
a
Chemical shifts δ are reported in ppm and the one-bond spin-
1
b
spin coupling constants | J(Si,H)| are reported in Hz. T ) 296 K, in
THF-d are relative to residual THF protons at
8
; the chemical shifts δ
H
Figure 1. Experimentally observed one-bond spin-spin coupling
c
d
1
1
.73 ppm. Data from Ebsworth and Turner (ref 1). T ) 296 K, in
THF-d
| J(Si,H)| in SiH
n
Cl4-n (n ) 0-4) dissolved in THF-d
8
(this work) and
8
; the chemical shifts δSi are relative to TMS at 0 ppm.
cyclohexane-d12 (ref 1).
TABLE 2: Calculated Bonding Energies (kcal/mol)
between Chlorosilanes and THF Computed at Different
Levels of Theory
E(B3LYP/
E(BP86/
E(B3LYP/
aug-cc-pvtz)
DNP)
aug-cc-pvtz)
+ BSSE
SiHCl
3
0.8
2.0
1.5
0.1
1.6
2.6
3.1
0.7
1.3
2.4
2.9
0.6
SiH
SiH
SiH
2
Cl
3
Cl
4
2
1
The observed H chemical shifts increase monotonically from
.14 to 6.09 ppm with increasing chlorine substitution, in line
3
2,11
29
1
with the trend reported by Ebsworth and Turner. For the Si
Figure 2. The correlation between the difference in the | J(Si,H)|
chemical shifts, also in line with previous reports,12 a maximum
8
values measured in THF-d (this work) and cyclohexane-d12 (ref 1)
and the theoretically calculated silane-THF bonding energies.
δSi value is observed for SiHCl3. Thus, the reliability of the
1
|
J(Si,H)| values to be discussed in further detail below is sup-
silanes compared to SiH4. In Figure 2 we have illustrated the
correlation between the difference in the experimentally deter-
ported by the trends in the observed chemical shifts.
From Figure 1, we see that successive substitution of H with
1
1
mined | J(Si,H)| values for the two solvents (THF-d and cyclo-
Cl results in an approximately linear response in | J(Si,H)|. The
8
1
1
hexane-d ) and the calculated silane-THF bonding energies.
observed increase in | J(Si,H)| with increasing Cl substitution
12
1
As seen, there is a reasonable qualitative agreement between
the data, although we note that for the higher level of theory,
there is a slight deviation from the expected trend for the highly
Cl-substituted silanes.
agrees well with previous findings. However, a careful inspec-
tion of Figure 1 shows that there are significant deviations
between the data reported herein and the data previously reported
1
in the literature.
1
When our | J(Si,H)| values at 296 K, obtained by probing
Due to the weak solvent-SiH4 interactions in THF-d8, a
comparison between the calculated gas phase and experimentally
determined coupling constants should be of value. However,
we note that, depending on the chosen method, SCF, CAS, DFT,
1
the H nucleus, are compared to the values reported for chloro-
silanes in cyclohexane-d12, it is seen that for SiH4 the values
differ by 0.5 Hz, which is within the range of our experimental
variations (vide supra). Thus, although the two chosen solvents
1
SOPPA, or SOPPA(CCSD), theoretically derived J(Si,H) values
have very different polarities, the effect exerted on the measured
in the range from -186.10 to -243.74 Hz have been reported
1
13,14
|
J(Si,H)| for SiH4 is insignificant.
for SiH4.
Among these data, we find the calculated value
1
4
For the various chloro-substituted molecules, on the other
of -202.541 Hz reported by Sauer et al. to show the best
1
1
hand, the data reported previously by Ebsworth and Turner
does not compare equally well to the data reported herein. In
numerical agreement with our measured | J(Si,H)| values for
SiH4. This is reasonable, as their calculations were performed
at a higher level of theory (CCSD) than the others. A certain
amount of fortuity may not be ruled out, however, as the
calculations model a gas-phase molecule rather than a molecule
in solution, and the SOPPA approximation may introduce errors
as well. For chlorine-substituted silanes, theoretically calculated
values are currently not available from the literature.
1
our experiments, the | J(Si,H)| values at 296 K were found to
be 14.6-20.9 Hz larger than the earlier values, as reported in
Table 1. It is reasonable to ascribe this discrepancy to the
stronger interactions between the polar and soft chlorosilanes
and the polar THF-d8 used in the present work compared to the
1
,2
nonpolar cyclohexane-d12 used by Ebsworth and Turner.
1
To put this hypothesis on firmer ground, we studied THF-
silane interactions by DFT calculations. We found that THF-
silane adduct formation may indeed occur by coordination of
the O atom to the H-rich side of the silane. Calculated bonding
energies between THF and the different silanes are given in
Table 2. The good qualitative agreement between the results
from different methods is gratifying.
Our measured | J(Si,H)| values in THF-d8 are systematically
larger than the results reported for cyclohexane-d12. Further DFT
calculations serve to elucidate the factors, giving raise to this
difference: Coordination of THF-d8 perturbs the geometry of
the silane, and although the changes in the Si-H bond lengths
(Figures 3-6) are rather subtle, the coupling constant is known
1
5
to depend strongly on atomic distances. The change in the
calculated average Si-H bond length upon THF coordination
is in the range 0.1-0.4%, whereas the concomitant change in
Regardless of the level of theory, the calculated interaction
between THF and the silanes is higher for the chloro-substituted

Spin-Spin Coupling in Chlorosilanes
J. Phys. Chem. A, Vol. 110, No. 32, 2006 9803
Figure 3. THF-SiH
4
complex. Si-H Bonding distances in picometers.
is 148.3 pm.
Si-H distance for unperturbed SiH
4
Figure 7. Experimentally observed one-bond spin-spin coupling
1
|
J(Si,H)| at various temperatures in silane and the various chlorosilanes
dissolved in THF-d .
8
1
TABLE 3: Estimated Slopes (a) and | J(Si,H)| at 0 K (J
0
)
Obtained by Fitting J ) aT + J to the Experimentally
Determined | J(Si,H)| Values at Various Temperatures
T
0
1
a
a
J
0
a
J
0
SiHCl
3
-0.131
-0.247
415.8
380.9
SiH
SiH
3
4
Cl
-0.140
0.006
299.5
201.3
SiH
2
Cl
2
a
1
The temperature dependence of | J(Si,H)| is shown in Figure 7.
is reported in Hz.
a is reported in Hz/K, whereas J
0
Figure 4. THF-SiH
3
Cl complex. Bonding distances in picometers.
Cl is 147.8 pm.
Si-H distance for unperturbed SiH
3
Figures 3-6 depict the silane-THF adducts, with the Si-H
bond lengths given in picometers. The calculations suggest
that the Si-H distance, except for one of the two pseudo-
symmetrically independent Si-H bonds in SiH4, is shortened
upon THF coordination to the silane. All of these changes agree
qualitatively with the experimental observations of a stronger
spin-spin Si-H coupling for chlorosilanes dissolved in THF-d8
compared to the values measured in cyclohexane-d12. Further-
more, we see that for SiH4, THF coordination causes one of
the Si-H bonds to shorten, whereas the other is elongated. Thus,
according to the calculations, on average, the influence of the
solvent (THF-d8 vs cyclohexane-d12) should be lower for SiH4
than for any of the chloro-substituted silanes. It is gratifying
that this is in accordance with the experimental observations.
Figure 5. THF-SiH
2
Cl
2
complex. Bonding distances in picometers.
Cl is 147.3 pm.
Si-H distance for unperturbed SiH
2
2
1
Temperature Dependence. We studied | J(Si,H)| in the
temperature range 213 K < T < 273 K, and the plots are shown
1
in Figure 7. In a separate experiment we measured | J(Si,H)|
1
at 296 K. As seen from Figure 7, the | J(Si,H)| measured at
2
96 K agrees very well with the estimated value obtained by
extrapolation of the temperature series and we obtained the same
1
numerical values for | J(Si,H)|, regardless of whether we probed
1
29
the H or the Si nuclei at 296 K (Table 1).
1
It is interesting to note that | J(Si,H)| is temperature inde-
3
Figure 6. THF-SiHCl complex. Bonding distances in picometers.
Si-H distance for unperturbed SiHCl is 146.7 pm.
3
pendent for SiH4, whereas for the chloro-substituted silanes,
significant temperature dependences are observed. Over the
1
14
1
|
J(Si,H)| is in the range 4-8%. For SiH4, Sauer et al. cal-
temperature range studied, each set of | J(Si,H)| data may be
culated the coupling constant’s dependence on geometry defor-
mations by computing the former for a large number of different
geometries. Further, they parametrized the results, allowing us
to estimate the effect of the perturbation. If we enter the silane
geometry from our geometry optimizations on the silane-THF
complex into the expression from Sauer et al., we arrive at a
coupling constant change of 10.8 Hz due to geometry changes.
This is on the same order of magnitude as the discrepancies
observed between the chlorinated silanes in the two different
solvents. The lack of discrepancy in the case of SiH4 may be
ascribed to the weak bonding and concomitant short lifetime
of the SiH4-THF adduct.
fitted by a linear equation given as JT ) aT + J0, where T is
1
the absolute temperature, JT and J0 are | J(Si,H)| at any T and
T ) 0 K, respectively, and a is an empirical constant. In Table
3 we list the results of the linear fits.
1
For all of the chloro-substituted compounds, | J(Si,H)| de-
creases with an increase in the temperature. The J0 values pre-
sented in Table 3 are obtained under the assumption that J(T)
remains linear down to 0 K, which should be kept in mind when
the numerical values are interpreted. However, given the large
deviations between the estimated J0 values, we find it reasonable
to postulate that J0 will increase with increasing chlorine sub-
stitution on the species also at 0 K.

9804 J. Phys. Chem. A, Vol. 110, No. 32, 2006
Thorshaug et al.
Acknowledgment. We thank “Norsk Hydros fond for
SINTEF” for financial support, Aud M. Bouzga for recording
the NMR spectra, and Trygve Helgaker for valuable discussions
on coupling constants.
References and Notes
(
1) Ebsworth, E. A. V.; Turner, J. J. J. Chem. Phys. 1962, 36, 2628-
634.
2) Ebsworth, E. A. V.; Turner, J. J. J. Phys. Chem. 1963, 67, 805-
2
8
3
1
(
07.
(
52.
3) Helgaker, T.; Jaszunski, M.; Ruud, K. Chem. ReV. 1999, 99, 293-
(4) (a) Van Wazer, J. R.; Ewig, C. S.; Ditchfield, R. J. Phys. Chem.
Figure 8. The correlation between the slope of the linear fit to the
989, 93, 2222-2230. (b) Jameson, C. J.; Jameson, A. K. Chem. Phys.
1
observed temperature dependence of | J(Si,H)|, i.e., a value from Table
Lett. 1988, 149, 300-305. (c) Sauer, S. P. A.; Raynes, W. T. J. Chem.
3
, and the calculated silane-THF bonding energy.
Phys. 2000, 113, 3121-3129.
(
5) Spartan 5.1; Wavefunction, Inc., 18401 Von Karman Ave., Suite
370, Irvine, CA 92612. http://www.wavefun.com.
6) (a) Becke, A. D.; Phys. ReV. A 1988, 38, 3098-3100. (b) Perdew,
J. P. Phys. ReV. B 1986, 33, 8822-8824.
7) The developers of the Spartan program do not, as far as we can
1
The temperature dependence of | J(Si,H)| is significantly
(
more pronounced for the C2V-symmetric SiH2Cl2 than for the
C3V-symmetric SiHCl3 and SiH3Cl, as seen from the slope of
the linear fits in Figure 7 and Table 3. Although SiH3Cl has
the highest electrical dipole moment µ of the three chloro-
substituted silanes, it does not show the most pronounced tem-
perature dependence. Thus, the slopes a reported in Table 3
and shown graphically in Figure 7 cannot be related to µ alone.
The calculated silane-THF bonding energies, given in Table
, are of a magnitude that makes the adduct lifetimes strongly
temperature dependent at and around room temperature (RT ≈
.3 kJ/mol). The tendency in bonding energies compare well
with the tendency in temperature dependence of the coupling
constants, as illustrated in Figure 8. While we do not wish to
draw any firm conclusions based on these results, they may
illustrate a possible cause of the measured temperature depend-
ences, as the time-averaged atomic distances in the adducts will
vary as a function of the adduct lifetimes.
(
see, reference the numerical basis sets in any of their documentation.
However, there is strong reason to believe that the basis sets closely resemble
those described by Delley (Delley, B. J. Chem. Phys. 1991, 94, 7245-
7
250 and references therein).
(8) Frisch, M.-J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K.
N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li,
X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.;
Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.;
Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.;
Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels,
A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson,
B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03,
Revision B.04; Gaussian, Inc., Pittsburgh, PA, 2003.
2
2
Conclusions
(9) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
Under the given experimental conditions, we found that the
observed absolute value of the one-bond spin-spin coupling
(10) Kendall, R. A.; Dunning, T. H., Jr.; Harrison, R. J. J. Chem. Phys.
1992, 96, 6796-6806.
1
(11) The chemical shifts δH in THF-d reported by us in Table 1 and
constant | J(Si,H)| increases with an increasing number of
8
the values reported in cyclohexane-d12 in ref 2 deviate by 0.02-0.12 ppm.
chlorine substituents in the chlorosilane SiHnCl4-n (n ) 0-4).
The quantitative changes observed in THF-d8 studied herein
were found to differ from the values previously reported for
Note: In ref 2, the chemical shifts are reported on the τ-scale.
(
12) Van Wazer, J. R.; Ewig, C. S.; Ditchfield, R. J. Phys. Chem. 1989,
93, 2222-2230.
the same compounds in cyclohexane-d12. The variations in
(13) (a) Lazzeretti, P.; Rossi, E.; Zanasi, R. J. Chem. Phys. 1984, 80,
315-318. (b) Nair, A. C.; Chandra, P. Theor. Chim. Acta 1994, 89, 261-
72. (c) Kirpekar, S.; Jensen, H. J. Aa.; Oddershede, J. Theor. Chim. Acta
1
|
J(Si,H)| with temperature were found to be linearly temper-
2
1
ature dependent for all the chlorine-substituted silanes, and the
997, 95, 35-47. (d) Malkina, O. L.; Salahub, D. R.; Malkin, V. G. J.
1
temperature dependence of | J(Si,H)| varies between the dif-
Chem. Phys. 1996, 105, 8793-8800. (e) Kirpekar, D.; Enevoldsen, T.;
Oddershede, J.; Raynes, W. T. Mol. Phys. 1997, 91, 897-907.
ferent chlorosilanes, being most pronounced for the C2V sym-
1
(14) Sauer, S. P. A.; Raynes, W. T.; Nicholls, R. A. J. Chem. Phys.
001, 115, 5994-6006.
metric SiH2Cl2. | J(Si,H)| in SiH4 was found to be temperature
2
independent. Quantum chemical DFT calculations indicate that
(15) Selected studies of the relation between bond length and spin-
solute/solvent adduct formation changes the Si-H distances to
spin coupling: (a) Raynes, W. T.; Geertsen, J.; Oddershede, J. Chem. Phys.
Lett. 1992, 197, 516-524. (b) Kirpekar, S.; Sauer, S. P. A. Theor. Chem.
Acc. 1999, 103, 146-153. (c) Pecul, M.; Jaszunski, M.; Sadlej, J. Chem.
Phys. Lett. 1999, 305, 139-146. (d) Ruden, T. A.; Lutnaes, O. B.; Helgaker,
T.; Ruud, K. J. Chem. Phys. 2003, 118, 9572-9581. (e) Helgaker, T.
Personal communication.
1
an extent that may explain the dependence of | J(Si,H)| upon
the choice of solvent. Further, the calculated adduct formation
energies correlate well with the temperature dependence of the
coupling constants.