β-trimethylsilyl-2-propyl cation has been formed by the gas phase protonation of allyl-trimethylsilane and characterized by infrared multiple photon dissociation spectroscopy. The experimental Cβ-Cα+ stretching feature at 1586 cm−1, remarkably blue-shifted with respect to a C−C single bond stretching mode, is indicative of high double bond character, a signature of β-stabilizing effect due to hyperconjugation of the trimethylsilyl group in the β-position with respect to the positively charged carbon. Density functional theory calculations at the B3LYP/6-311++G(2df,2p) level yield the optimized geometries and IR spectra for candidate isomeric cations and for neutral and charged reference species.

One remarkable aspect of organosilicon chemistry is the exceptional stabilizing effect exerted on carbocations by a silyl group in the β position to the carbon holding the formal positive charge.1–3 The β-silyl effect on carbocations has been extensively studied, both experimentally and computationally.4–10 Carbocations bearing a β-silyl group are significantly stabilized relative to unsilylated analogues. The stabilization is markedly orientation-dependent, due to the hyperconjugative interaction between the Si−Cβ σ-bond and the formally empty 2p(Cα+) orbital as represented in structure A in Figure 1. Computational studies have shown that the hyperconjugative interaction leads to a lengthening of the Si−Cβ bond while the Cβ−Cα+ bond becomes shorter.6 Another interesting feature is the Si−Cβ−Cα+ bond angle which reflects the degree of hyperconjugation and eventually the bridging structure character (B in Figure 1).

FIG. 1.

“Open” (a) and “bridged” (b) form of a β-silyl-substituted carbocation.

FIG. 1.

“Open” (a) and “bridged” (b) form of a β-silyl-substituted carbocation.

Close modal

Several nuclear magnetic resonance spectroscopic investigations of β-silyl-substituted carbocations, bearing various groups in the α-position, have reported characteristic features assigned to the operation of β-silyl-hyperconjugation.11–17 However, due to the great reactivity of β-silyl-substituted carbocations in solution, all the species so far examined in the condensed phase are structurally highly complex. Indeed a spectroscopic characterization of an exemplary, simple β-silyl-substituted alkyl cation, where the hyperconjugation effect of the trimethylsilyl group should be evident, is not straightforward either in solution or in the solid-phase. Interestingly, mass spectrometry allows the generation of naked, prototypical β-silylated carbocations prone to be investigated by spectroscopic techniques. In the last decade, a powerful structurally diagnostic assay based on tandem mass spectrometry coupled with a radiation source allowing infrared multiple photon dissociation (IRMPD) spectroscopy has been developed and successfully applied to solve a variety of structural problems.18–24 In the past few years, IRMPD spectroscopy has allowed to elucidate the structure of numerous ions from very simple, fundamental species25–28 to (bio)molecular29,30 and cluster ions.31,32 This powerful methodology is based on the use of IR light generated by a free electron laser (FEL) at CLIO or FELIX research facilities, possessing high fluence and wide tunability. Also methods based on the tagging approach for the IR photodissociation of mass selected clusters have been recently applied to solve structural problems regarding various SixHy+ ions.33,34

Here we report the IRMPD spectrum of a simple β-trimethylsilyl-substituted carbocation, namely, the one that is obtained by protonation of allyl-trimethylsilane, (CH3)3Si–CH2–CH=CH2. The protonation is effected in the cell of an Fourier transform ion cyclotron resonance (FT-ICR) mass spectrometer by the t-butyl cation, ((CH3)3C+) a mild protonating agent (Eq. (1)),

\begin{eqnarray}&&({\rm CH}_3)_3 {\rm C}^ + + ({\rm CH}_3)_3 {\rm Si} {-} {\rm CH}_2 {-} {\rm CH} {=} {\rm CH}_2\nonumber\\&&\quad \to ({\rm CH}_3)_3 {\rm Si} {-} {\rm CH}_2 {-} {\rm CH}^ + {\rm CH}_3 + {\rm C}_4 {\rm H}_8.\quad\end{eqnarray}
( CH 3)3C++( CH 3)3 Si CH 2 CH = CH 2( CH 3)3 Si CH 2 CH + CH 3+C4H8.
(1)

In order to gain information about the structure of the ion of interest, the IRMPD spectrum of the so-formed [C6H15Si]+ ion is compared with the calculated IR spectra of potential candidate geometries and some other ions or neutral species taken as reference. The optimized structures for possible isomers corresponding to general formula [C6H15Si]+ have been obtained from B3LYP/6-311++G(2df,2p) calculations and are presented in Figure 2, while Cartesian coordinates are reported in the supplementary material (Table IS).38 

FIG. 2.

Optimized geometries and relative energies (kJ mol−1, in parentheses) for representative [C6H15Si]+ structures calculated at the B3LYP/6-311++G(2df,2p) level.

FIG. 2.

Optimized geometries and relative energies (kJ mol−1, in parentheses) for representative [C6H15Si]+ structures calculated at the B3LYP/6-311++G(2df,2p) level.

Close modal

Protonation of allyl-trimethylsilane is expected to yield a β-trimethylsilyl-2-propyl cation. Two geometries can be considered for the β-trimethylsilyl-2-propyl cation, an “open” form depicted as A in Figure 1 (R=CH3, R=H) where the cationic charge on carbon is stabilized by hyperconjugative interaction between the vacant p orbital and the Si−Cβ σ-bond and a “bridged” form (B) where the stabilization mode involving silicon leads to a cyclic structure. Without any constraint, however, all attempts to optimize the bridged form of [C6H15Si]+ evolve towards structure 1, which may be seen as intermediate between A and B. The same final geometry is observed when the optimization was performed at the MP2/6-311++G(2df,2p) level. A comparison of significant geometrical parameters obtained from the two computational levels is provided in Table IIS of the supplementary material.38 Three other potential isomers of [C6H15Si]+ are also taken in consideration. The first one named 2 in Figure 2 is the dimethyl-2-methylpropyl-silyl cation that may be formed from 1 by a rearrangement process involving the shift of a methyl group from silicon to the positively charged Cα. Similar isomerization processes have been highlighted to occur within [C5H11Si]+ ions formed either by the reaction of (CH3)3Si+ ions with acetylene or by direct protonation of trimethylsilylacetylene, (CH3)3Si–C≡CH, when the reactions are run at low pressure in the cell of an FT-ICR mass spectrometer.35,36 Upon inspection of the [C5H11Si]+ potential energy surface, silacations, having the positive charge on silicon, are found to be more stable with respect to isomers holding the positive charge on a carbon atom. In contrast, among [C6H15Si]+ ions, 1 is largely more stable than silacations such as 2 or 3, by 31 kJ mol−1 or 44 kJ mol−1, respectively. Isomer 3 can be formed from 1 by a proton shift from Cβ to Cα, giving a high energy isomer 4, followed by a 1,2-shift of a methyl group from silicon to Cβ. In another isomer, named 5 in Figure 2, a cyclopropane unit can be seen as edge-coordinated to (CH3)3Si+. This ion can be formally obtained by protonation of allyl-trimethylsilane on the methine unit of the allyl group, followed by ring closure to the three-membered ring. This latter isomer is 77 kJ mol−1 higher in energy relative to 1.

The only fragment ion originating from the IRMPD process of [C6H15Si]+ is an ion at m/z 73, namely the trimethylsilyl cation ((CH3)3Si+) (Eq. (2)):

\begin{equation}[{\rm C}_5 {\rm H}_{11} {\rm Si}]^ + \mathop {\longrightarrow} \limits^{{\rm nh}v} ({\rm CH}_3)_3 {\rm Si}^ + + {\rm C}_3 {\rm H}_6.\end{equation}
[C5H11 Si ]+ nh v( CH 3)3 Si ++C3H6.
(2)

This observed fragmentation product suggests an unrearranged structure for the sampled ion, prone to cleave the trimethylsilyl group as stable fragment. Different fragmentation paths are expected from isomers such as 2 or 3, characterized by a rearranged bonding framework.35,36 The IRMPD spectrum is obtained by plotting the photofragmentation yield R (R = −log[Iparent/(Iparent + Ifragment)], where Iparent and Ifragment are the integrated intensities of the mass peaks of the precursor and of the fragment ion, respectively) as a function of the frequency of the IR radiation.37 The IRMPD spectrum of [C6H15Si]+ in the 690–1790 cm−1 range is plotted in Figure 3(a). The calculated IR spectrum of the stable isomer 1 is presented in Figure 3(b), while the upper panels show the calculated IR spectra of neutral molecules related to the sampled species, as detailed in the following discussion. A comparison of the experimental IRMPD spectrum of protonated allyl-trimethylsilane with the IR spectra of the candidate isomers displayed in Figure 2 is provided in Figure 1S of the supplementary material.38 The IRMPD spectrum of [C6H15Si]+ comprises two sections: the first one (690–1450 cm−1) was recorded using a FEL electron energy of 40 MeV, with an average power of 550 to 500 mW. The main feature is a broad absorption band at 840 cm−1, not well resolved from a second feature at 883 cm−1 even when the spectrum was recorded using an attenuator decreasing the laser power by a factor of three (green trace in Figure 3(a)). Two distinct features are present at 997 and 1283 cm−1. The second region (1140–1790 cm−1) was recorded using a FEL electron energy of 45 MeV, with an average laser power varying from 1000 to 840 mW upon increasing the wavenumber (red trace in Figure 3(a)). This region shows a wide absorption from 1380 to 1450 and a distinct band at 1586 cm−1, besides confirming the feature at 1283 cm−1. The reported IRMPD spectrum of [C6H15Si]+ is in very good agreement with the calculated IR spectrum of isomer 1. Relying on the typically good accord between the IRMPD spectrum and the linear IR spectrum of the sampled species as obtained by hybrid density functional theory calculations, the vibrational modes of [C6H15Si]+ can be assigned. Table IIIS of the supplementary material38 summarizes the experimental IRMPD features, listed together with the IR resonances calculated for the most stable isomer 1. A frequency scaling factor of 0.99 is uniformly adopted.

FIG. 3.

(a) Experimental IRMPD spectrum of [C6H15Si]+ obtained with full and attenuated (by a factor of 3, green) laser power, together with computed IR spectra of (b) β-trimethylsilyl-2-propyl cation (1), (c) allyl-trimethylsilane (6), (d) 1-propyl-trimethylsilane (7), (e) Z-1-propenyl-trimethylsilane (8), (f) E-1-propenyl-trimethylsilane (9), (g) 5,5-dimethyl-pent-2-yl cation (10); all calculated at the B3LYP/6-311++G(2df,2p) level in the spectral range of 690–1790 cm−1.

FIG. 3.

(a) Experimental IRMPD spectrum of [C6H15Si]+ obtained with full and attenuated (by a factor of 3, green) laser power, together with computed IR spectra of (b) β-trimethylsilyl-2-propyl cation (1), (c) allyl-trimethylsilane (6), (d) 1-propyl-trimethylsilane (7), (e) Z-1-propenyl-trimethylsilane (8), (f) E-1-propenyl-trimethylsilane (9), (g) 5,5-dimethyl-pent-2-yl cation (10); all calculated at the B3LYP/6-311++G(2df,2p) level in the spectral range of 690–1790 cm−1.

Close modal

IRMPD spectroscopy has thus allowed us to establish that the gas phase protonation of allyl-trimethylsilane leads to β-trimethylsilyl-2-propyl cation (isomer 1), whose structure appears to be influenced by the operation of the β-silyl effect. In this respect, it is interesting to search for spectroscopic and structural features that may characterize the β-stabilizing effect due to hyperconjugation of the trimethylsilyl group. To this purpose, the experimental IRMPD spectrum of protonated allyl-trimethylsilane is compared with the calculated linear IR spectra of neutral and ionic species, chosen to provide useful references and reported in Figure 4.

FIG. 4.

Structural parameters of optimized structures for selected neutral and ionic species calculated at the B3LYP/6-311++G(2df,2p) level. Distances are given in Å.

FIG. 4.

Structural parameters of optimized structures for selected neutral and ionic species calculated at the B3LYP/6-311++G(2df,2p) level. Distances are given in Å.

Close modal

Theoretical studies have indicated that, in order to stabilize the cationic center, the (CH3)3Si−Cβ sigma bond must be about parallel to the vacant p orbital on Cα+. In this way, π-bond character is built into the Cβ−Cα bond at the expense of weakening the adjoining (CH3)3Si−Cβ sigma bond. The effect on the geometry is a lengthening of the (CH3)3Si−Cβ bond and a shortening of Cβ−Cα bond while the (CH3)3Si group is moving towards Cα+. These expected changes in geometric parameters are confirmed by calculations. Selected bond distances are reported in Figure 4 where the β-trimethylsilyl-2-propyl cation 1 is compared with neutral allyl-trimethylsilane (6) and neutral 1-propyl-trimethylsilane (7). Indeed, the optimized geometries of 6 and 7 show a relatively similar Cβ−Cα bond length, of 1.494 Å and 1.540 Å, respectively, while for β-trimethylsilyl-2-propyl cation 1 this bond length becomes 1.365 Å, a value close to the Cβ=Cα double bond length of 1.337 and 1.335 Å calculated for neutral Z-1-propenyl-trimethylsilane and E-1-propenyl-trimethylsilane reported as 8 and 9, respectively, in Figure 4. The partial double bond character in 1 is experimentally confirmed by the IR absorption observed at 1586 cm−1, close to the range of a typical stretching frequency for a carbon-carbon double bond. The calculated IR spectrum presents a Cβ−Cα+ stretching vibration at 1597 cm−1 for isomer 1, which may be compared with the corresponding modes in neutral examples such as Z-1-propenyl-trimethylsilane (8) and E-1-propenyl-trimethylsilane (9). Within these molecules Cβ=Cα stretching modes are calculated to be active at 1650 and 1658 cm−1, respectively, as reported in Table IVS of the supplementary material.38 In contrast, a single bond character is observed for the Cβ−Cα bond of 1-propyl-trimethylsilane (7), displaying a stretching frequency of 1062 cm−1. That the double bond character imparted to the Cβ−Cα+ bond of 1 should be ascribed to the presence of the β-trimethylsilyl group is further confirmed by noting that in the all-carbon analogue of cation 1, namely in 5,5-dimethyl-pent-2-yl cation (10), the Cβ−Cα+ bond is 1.401 Å long with a stretching frequency of 1549 cm−1. In cation 1 the corresponding values are calculated equal to 1.365 Å and 1597 cm−1. The two ions 1 and 10 differ by the presence of a t-butyl group in 10 replacing the trimethylsilyl group present in 1. In comparing the two species, a markedly superior hyperconjugative ability by the β-trimethylsilyl group appears clearly, as further testified by the Si−Cβ−Cα angle of 97.82° while little perturbation of the tetrahedral angle is observed in 10 ((CH3)3C−Cβ−Cα angle equal to 106.5°). Finally, the most pronounced band in the IRMPD spectrum includes a component on the low energy side at 840 cm−1 that, upon viewing the animation of the calculated normal modes, can be assigned to a combination of CH bending and Si−Cβ stretching vibrations. It is significant that this band is somewhat red-shifted relative to the corresponding one in the IR spectra of allyl-trimethylsilane (6) (but also of 7, 8, and 9), in agreement with a weakening of the Si−Cβ bond due to the operation of β-silyl hyperconjugation. The same evidence becomes even clearer in the region of the (CH3)3Si−Cβ stretch mode (calculated at 320 cm−1 for 1 and 570 cm−1 for 6) that is unfortunately beyond experimental access. In summary, IRMPD spectroscopy provides unambiguous indication of the important role played by a silyl substituent on a carbon atom adjacent to a positively charged carbenium center, known as β-silyl effect.

Dedicated to Professor Helmut Schwarz in celebration of his 70th birthday. Financial support has been provided by Università di Roma “La Sapienza” and by the European Union (Project No. IC010-03). The authors wish to thank J. M. Ortega and all the CLIO team for their support during the experiments.

1.
E. W.
Colvin
,
Chem. Soc. Rev.
7
,
15
(
1978
).
2.
J. B.
Lambert
and
Y.
Zhao
,
J. Am. Chem. Soc.
118
,
7867
(
1996
).
3.
J. B.
Lambert
,
Y.
Zhao
,
R. W.
Emblidge
,
L. A.
Salvador
,
X.
Liu
,
J.-H.
So
, and
E. C.
Chelius
,
Acc. Chem. Res.
32
,
183
(
1999
).
4.
S. G.
Wierschke
,
J.
Chandrasekhar
, and
W. L.
Jorgensen
,
J. Am. Chem. Soc.
107
,
1496
(
1985
).
5.
X.
Li
and
J. A.
Stone
,
J. Am. Chem. Soc.
111
,
5586
(
1989
).
6.
H.-U.
Siehl
and
T.
Müller
, in
The Chemistry of Organic Silicon Compounds
, edited by
Z.
Rappoport
and
Y.
Apeloig
(
John Wiley and Sons, Ltd.
,
1998
), p.
595
.
7.
F.
Cacace
,
M. E.
Crestoni
, and
S.
Fornarini
,
J. Am. Chem. Soc.
114
,
6776
(
1992
).
8.
B.
Chiavarino
,
M. E.
Crestoni
, and
S.
Fornarini
,
J. Am. Chem. Soc.
120
,
1523
(
1998
).
9.
S.
Fornarini
, in
The Chemistry of Organic Silicon Compounds
, edited by
Z.
Rappoport
and
Y.
Apeloig
(
John Wiley and Sons, Ltd.
,
2001
), p.
1027
.
10.
N.
Goldberg
and
H.
Schwarz
, in
The Chemistry of Organic Silicon Compounds
, edited by
Z.
Rappoport
and
Y.
Apeloig
(
John Wiley and Sons, Ltd.
,
1998
), p.
1105
.
11.
A.
Klaer
and
T.
Muller
,
J. Phys. Org. Chem.
23
,
1043
(
2010
).
12.
H.
Mayr
and
R.
Pock
,
Tetrahedron
42
,
4211
(
1986
).
13.
T.
Muller
,
R.
Meyer
,
D.
Lennartz
, and
H.-U.
Siehl
,
Angew. Chem. Int. Ed.
39
,
3074
(
2000
).
14.
H.-U.
Siehl
, in
Recent Developments in Carbocation and Onium Ion Chemistry
, edited by
K. K.
Laali
(
American Chemical Society
,
2007
), p.
1
.
15.
H. U.
Siehl
,
F. P.
Kaufmann
,
Y.
Apeloig
,
V.
Braude
,
D.
Danovich
,
A.
Berndt
, and
N.
Stamatis
,
Angew. Chem. Int. Ed.
30
,
1479
(
1991
).
16.
H. U.
Siehl
,
F. P.
Kaufmann
, and
K.
Hori
,
J. Am. Chem. Soc.
114
,
9343
(
1992
).
17.
R.
Meyer
,
K.
Werner
, and
T.
Muller
,
Chem.-Eur. J.
8
,
1163
(
2002
).
18.
J. R.
Eyler
,
Mass Spectrom. Rev.
28
,
448
(
2009
).
19.
N. C.
Polfer
,
Chem. Soc. Rev.
40
,
2211
(
2011
).
20.
N. C.
Polfer
and
J.
Oomens
,
Mass Spectrom. Rev.
28
,
468
(
2009
).
21.
J.
Roithova
,
Chem. Soc. Rev.
41
,
547
(
2012
).
22.
M. A.
Duncan
,
Int. J. Mass Spectrom.
272
,
99
(
2008
).
23.
T. D.
Fridgen
,
Mass Spectrom. Rev.
28
,
586
(
2009
).
24.
L.
MacAleese
and
P.
Maitre
,
Mass Spectrom. Rev.
26
,
583
(
2007
).
25.
B.
Chiavarino
,
M. E.
Crestoni
,
O.
Dopfer
,
P.
Maitre
, and
S.
Fornarini
,
Angew. Chem. Int. Ed.
51
,
4947
(
2012
).
26.
B.
Chiavarino
,
M. E.
Crestoni
,
S.
Fornarini
,
J.
Lemaire
,
P.
Maitre
, and
L.
MacAleese
,
J. Am. Chem. Soc.
128
,
12553
(
2006
).
27.
B.
Chiavarino
,
M. E.
Crestoni
,
S.
Fornarini
,
J.
Lemaire
,
L.
MacAleese
, and
P.
Maitre
,
ChemPhysChem
6
,
437
(
2005
).
28.
B.
Chiavarino
,
M. E.
Crestoni
,
S.
Fornarini
,
J.
Lemaire
,
L.
MacAleese
, and
P.
Maitre
,
ChemPhysChem
5
,
1679
(
2004
).
29.
F.
Lanucara
,
B.
Chiavarino
,
M. E.
Crestoni
,
D.
Scuderi
,
R. K.
Sinha
,
P.
Maitre
, and
S.
Fornarini
,
Inorg. Chem.
50
,
4445
(
2011
).
30.
U. J.
Lorenz
,
J.
Lemaire
,
P.
Maitre
,
M. E.
Crestoni
,
S.
Fomarini
, and
O.
Dopfer
,
Int. J. Mass Spectrom.
267
,
43
(
2007
).
31.
B.
Chiavarino
,
M. E.
Crestoni
,
S.
Fornarini
,
S.
Taioli
,
I.
Mancini
, and
P.
Tosi
,
J. Chem. Phys.
137
,
024307
(
2012
).
32.
B.
Chiavarino
,
M. E.
Crestoni
,
S.
Fornarini
,
D.
Scuderi
, and
J.-Y.
Salpin
,
J. Am. Chem. Soc.
135
,
1445
(
2013
).
33.
M.
Savoca
,
M. A. R.
George
,
J.
Langer
, and
O.
Dopfer
,
Phys. Chem. Chem. Phys.
15
,
2774
(
2013
).
34.
M.
Savoca
,
J.
Langer
, and
O.
Dopfer
,
Angew. Chem. Int. Ed.
52
,
1568
(
2013
).
35.
B.
Chiavarino
,
M. E.
Crestoni
, and
S.
Fornarini
,
Chem. Commun.
2002
,
1418
.
36.
H.-U.
Siehl
,
S.
Brixner
,
C.
Coletti
,
N.
Re
,
B.
Chiavarino
,
M. E.
Crestoni
,
A.
De Petris
, and
S.
Fornarini
,
Int. J. Mass Spectrom.
334
,
58
(
2013
).
37.
J.
Lemaire
,
P.
Boissel
,
M.
Heninger
,
G.
Mauclaire
,
G.
Bellec
,
H.
Mestdagh
,
A.
Simon
,
S.
Le Caer
,
J. M.
Ortega
,
F.
Glotin
, and
P.
Maitre
,
Phys. Rev. Lett.
89
,
273002
(
2002
).
38.
See supplementary material at http://dx.doi.org/10.1063/1.4818729 for experimental methods, Cartesian coordinates (Table IS), geometrical parameters of isomer 1 calculated at B3LYP/6-311++G(2df,2p) and at MP2/6-311++G(2df,2p) level of theory (Table IIS), experimental and calculated vibrational modes of isomer 1 (Table IIIS); geometrical parameters (Table IVS); zero point energies (Table VS), and experimental IRMPD spectrum and computed IR spectra (isomer 1-5) (Figure 1S).

Supplementary Material