Directional Building Blocks Determine Linear and Cyclic Silicon

Subscriber access provided by UNIV OF DURHAM

Directional Building Blocks Determine Linear and Cyclic Silicon Architectures Eric A. Marro, Eric M. Press, Maxime A. Siegler, and Rebekka S. Klausen J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b02541 • Publication Date (Web): 11 Apr 2018 Downloaded from http://pubs.acs.org on April 11, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

Directional Building Blocks Determine Linear and Cyclic Silicon Architectures Eric A. Marro, Eric M. Press, Maxime A. Siegler, Rebekka S. Klausen* Department of Chemistry, Johns Hopkins University, 3400 N. Charles St, Baltimore, MD 21218 ABSTRACT: Silicon nanomaterials combine earth abundance and biodegradability with exceptional electronic properties. Strategic synthesis promises access to novel architectures with well-defined surface structure, size, and shape. Herein, we describe a fivestep synthesis of functional macrocyclic polysilanes. Comparison of the materials isolated from isomeric building blocks provides evidence that building block directionality controls the shape of the resulting nanomaterial. Infrared (IR) and 1H and 29Si NMR spectroscopies, coupled to computational data, provide evidence of a well defined Si−H and Si−Me terminated structure. The intrinsic porosity and the polarization arising from the hydridic character of the Si−H bond suggest applications in lithium-ion batteries, which are supported by quantum chemical calculations.

Introduction. We describe the strategic synthesis of macrocyclic poly(cyclosilane)s, materials bridging the chemical space between molecules, polymers, and nanoparticles. Nanoscale silicon possesses unusual properties with great promise in energy, optoelectronic, and biomedical applications.1 Silicon’s high theoretical charge capacity has motivated extensive studies in lithium-ion batteries.2–4 Quantum confinement introduces a tunable band gap5,6 and porous silicon (p-Si) has been intensely investigated for optoelectronic applications since Canham’s discovery of light emission.7,8 Silicon nanomaterials appeal in biological contexts due to their biodegradability and low toxicity.9–12 Free-standing silicon nanoparticles and nanocrystals exhibit similar promise and their properties are tunable by particle size and surface modification.13–19 While bottom-up synthetic approaches have led to great advances,20,21 control of shape, size, and interface and surface structure is a topic of significant current interest.15,22–25 The identification of new synthetic strategies and building blocks promises access to novel well-defined silicon nanostructures. Strategic synthesis is an established tool for the construction of complex organic nanomaterials26,27 that has more recently found application in the preparation of inorganic nanostructures.28,29 Notable examples include superatoms,30–32 hierarchical nanoparticle assemblies,33 and magic number precursors to indium phosphide quantum dots.34,35 However, few examples of the multistep strategic synthesis of complex silicon-based nanomaterials are known. Marschner described the synthesis of molecular sila-adamantane, the polycyclic core of crystalline silicon, via a Lewis acid-mediated skeletal rearrangement.36 We recently demonstrated the synthesis of a linear poly(cyclosilane), a novel conjugated polymer in which the repeat unit is a six-membered ring of silicon atoms.37

Figure 1. Comparing directionality in carbon-based π-conjugated and silicon-based σ-conjugated building blocks. Optimized structures of a linear trimer of 1,4Si6 and a cyclic hexamer of 1,3Si6. DFT B3LYP/6-31G(d). Blue = silicon; pink = hydrogen. Methyl groups omitted for clarity.

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 27

Scheme 1. Building block synthesis. a) Synthesis of 1,4Si6 via formation of potassium disilanide 2 and salt metathesis with Cl(SiMe2)2Cl. b) Significant rearrangement is observed in the coupling of potassium disilanide 5 and dichlorodimethylsilane. c) Magnesium disilanide 7 provides higher isolated yields of 6. Cyclosilane 6 is straightforwardly elaborated to the desired monomer 1,3Si6. 18-cr-6 = eighteen-crown-6; TfOH = trifluoromethanesulfonic acid; LAH = lithium aluminum hydride.

The preparation of complex silicon nanostructures is challenging in part due to the limited structural complexity of commercially available precursors. This is in contrast to organic chemistry where ortho-, meta-, and para-functionalized isomers of a given aromatic building block are readily available. The different directionalities of these organic building blocks give rise to unique and predictable nanomaterial shapes (Figure 1). A para-functionalized aromatic ring templates the growth of linear materials like the poly(phenylene ethynylene) (PPE) derivatives used in organic light-emitting diodes.38,39 meta-Functionalized arenes template macrocycles, like the anion-binding cyanostar.40,41 Building block directionality is an essential design concept in coordination polymers, supramoelcular assemblies, and framework materials.42–47 We explore the ability of chair-like cyclohexasilane rings to act as directional precursors to distinct nanomaterial structures. In contrast to cyclohexane rings, cyclohexasilanes are optically active in the UV region.48,49 Alkylated cyclohexasilanes are also bench stable, while hexasilabenzene (Si6H6) remains elusive.50 The attractive potential properties of a macrocyclic polysilane are an incentive for exploring the synthesis and oligomerization of a “meta”-like cyclohexasilane. Curvature is an important structural frontier in organic electronic materials and conjugated macrocycles exhibit exceptional optical and physical properties.51–59 Macrocycles have wideranging applications in supramolecular chemistry and represent enduring challenges for chemical synthesis.60 As described above, porosity in silicon, especially with nanoscale dimensions, introduces unique functionality to a powerful and ubiquitous semiconductor.8,9,61 Here, we show that dehydrocoupling polymerization of a novel 1,3-functionalized oligosilane building block provides an ensemble of macrocyclic polysilanes. The macrocyclic poly(cyclosilane)s are characterized by a suite of vibrational and NMR spectroscopies, complemented by density functional theory (DFT) calculations. The polarization of the macrocycles suggests opportunities in cation binding and quantum

chemical calculations support the ability of highly directional Si−H bonds to coordinate Li+. Results and Discussion. 1. Synthesis of 1,3Si6. We envision employing cyclohexasilane 1,3Si6 as a precursor to macrocyclic poly(cyclosilane)s via dehydrogenative polymerization (Figure 1). In earlier work, we reported the dehydrocoupling polymerization of structural isomer 1,4Si6 to linear poly(cyclosilane)s.37 The synthesis of site-selectively functionalized cyclohexasilanes is nontrivial. Reductive coupling of dichlorodimethylsilane yields dodecamethylcyclohexasilane (Si6Me12).62 Chlorination of Si6Me12, while yielding a mixture of structural isomers and diastereomers, is a popular route to substituted cyclohexasilanes via substitution and chromatographic separation.63,64 Reductive heterocouplings are unselective and yield complex mixtures.65 Salt metathesis of dianions is more promising for the synthesis of isomerically pure functionalized cyclosilanes. Marschner reported that crown ether-activation of potassium tert-butoxide (KOt-Bu/18-cr-6) is an excellent method for the preparation of terminal disilanides.66–68 We identified α,ωpotassium diphenylsilanide 2 as a useful intermediate in the synthesis of 1,4Si6 (Scheme 1a).37,69 Coupling to dichlorotetramethyldisilane (Cl(SiMe2)2Cl) provides 3. Protonolysis of the aromatic rings with trifluoromethanesulfonic acid (TfOH) yields an electrophilic silyl tetratriflate, which is reduced with lithium aluminum hydride (LAH) to yield crystalline 1,4Si6. Based on this scalable five-step synthesis, we identified the coupling of potassium diphenylsilanide 5 and dichlorodimethylsilane (Me2SiCl2) as a key step in the synthesis of 1,3functionalized cyclohexasilane 6. However, significant skeletal rearrangement was observed in the salt metathesis reaction, providing in 49% yield an inseparable mixture of the structural isomers 3 and 6 and unidentified byproducts (Scheme 1b). The ratio of 6 to 3 is 6.8:1. A solution to this synthetic challenge arose from crystallographic analyses of potassium diphenylsi-


Page 3 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


lanides 2 and 5, which provided insight into the nature of silanide reactivity.69 These are very strong nucleophiles due to weak ion-pairing between silicon and potassium in both solution and the solid state: instead of a contact ion pair, cation-π interactions are observed between the phenyl rings of the silanide and the potassium cation.69 The weak ion pairing is attributed to silicon’s “soft anion” character, which contributes to weak charge-controlled interactions. Based on this insight, we investigated countercations that form more covalent interactions with silicon. We found that magnesium diphenylsilanide 7, arising from the addition of magnesium bromide to 5,70 couples cleanly with Cl2SiMe2 to yield 6 in 84-93% yield (Scheme 1c). The crystal structure of 6 obtained from x-ray data analysis is shown in Figure 2a. A chair conformation is observed in the solid state. Subsequent dearylation with TfOH71,72 and LAH reduction yielded the desired building block in 70% yield over two steps. 1,3Si6 is a clear, colorless liquid at room temperature (Figure 2b).

a

b

Figure 2. a) X-ray crystal structure of cyclosilane 6. Displacement ellipsoids shown at 50% probability. Hydrogens omitted for clarity. Blue = silicon, black = carbon. b) Image of clear, colorless liquid 1,3Si6 monomer.

2. Polymerization Studies. Primary and secondary monosilanes (e.g. phenylsilane and diphenylsilane) are polymerized by transition metal-catalyzed dehydrocoupling, yielding polysilanes with the general formula [SiR2]n (R = H, alkyl, or aryl) and hydrogen gas.73,74 Group 4 metallocenes are among the best-studied dehydrocoupling polymerization catalysts.75–79 These complexes are known to be sensitive to the degree of monomer substitution, with primary silanes (RSiH3) producing the highest molecular weight polymers.78–80 Secondary silanes (R2SiH2) typically yield a mixture of dimers and trimers.78 This is due to the step-growth polymerization mechanism, in which dimerization of the secondary silane provides a molecule with unreactive tertiary silane end groups. Bifunctional monomers like 1,3Si6 and 1,4Si6 retain reactive secondary end groups after dimerization. This combination of monomer properties and the steric sensitivity of the promoter predicts that the greater reactivity of the chain ends of the growing poly(1,3Si6) compared to internal tertiary sites facilitates propagation and suppresses chain branching. Dehydrocoupling polymerization of 1,3Si6 was initiated by a series of early transition metal metallocenes (Table 1).73 The highest molecular weight polymer was obtained with zirconocene dichloride in combination with n-butyllithium (Cp2ZrCl2/n-BuLi, entry 1), a catalytic system first reported by Corey.78–80 Variation of the group 4 metal center did not lead to higher degrees of polymerization (entries 2-3), as observed in 1,4Si6 polymerization.81 A several day induction period was observed with dimethylzirconocene (Cp2ZrMe2),75 however,

performing the reaction neat resulted in a reasonable degree of polymerization (entry 4). High polymer molecular weights are only observed at high reaction extents in step growth polymerization and several factors including catalyst deactivation and competitive depolymerization under forcing conditions contribute to modest degrees of polymerization in silane dehydrocoupling.73,81 Table 1. Metallocene-initiated dehydrocoupling polymerization of 1,3Si6. Catalyst

Mn (kDa)

Mw (kDa)

Đ

Cp2ZrCl2/n-BuLia

3.26

4.78

1.47

10

a

1.04

1.35

1.30

3

Cp2HfCl2/n-BuLia

1.32

1.46

1.11

4

2.69

3.43

1.28

8

Cp2TiCl2/n-BuLi Cp2ZrMe2

b

a

Typical polymerization conditions: [metallocene]:[monomer]0 = 1:18, toluene, rt, 24 h. b Reaction conditions: [metallocene]:[monomer]0 = 1:18, neat, 24 h. c Determined by GPC at 254 nm relative to a polystyrene standard. d The ratio of poly(1,3Si6) Mn and 1,3Si6 Mn, as determined by GPC.

A unimodal distribution of molecular weights is shown by gel permeation chromatography (GPC) (Figure 3). The number average molecular weight (Mn) of poly(1,3Si6) corresponds to an average degree of polymerization of 10, comparable to the average degree of polymerization of 8 observed with poly(1,4Si6) under similar polymerization conditions.37,81 The similar molecular weight distributions rule out the possibility that spectroscopic differences in poly(1,3Si6) and poly(1,4Si6) arise from differences in chain length.

Figure 3. Gel permeation chromatography (GPC) analysis of poly(1,3Si6). Determined relative to a polystyrene standard at 254 nm (THF, [poly(1,3Si6)] = 1 mg mL-1, 40 °C, 0.35 mL min-1, 10 µL injection). The blue arrow indicates a low molecular weight shoulder attributed to linear oligomers.

Poly(1,3Si6) is stored in a nitrogen atmosphere glove box to minimize autoxidation of the reactive Si−H bonds.82 Under these conditions, poly(1,3Si6) is stable and no decomposition is observed over several weeks. Poly(1,3Si6) is highly soluble in dichloromethane, chloroform, tetrahydrofuran, and toluene, as well as in simple hydrocarbons like pentane, whereas



poly(1,4Si6) is sparingly soluble in chlorinated and ethereal solvents. The directionality of the 1,3Si6 monomer templates two possible nanomaterial architectures: poly(1,3Si6) is predicted to have either a linear or a macrocyclic structure (Scheme 2). Cyclic oligomers (e.g. the cyclohexasilane) arising from intramolecular chain end combination are typically observed in polymerization of monosilanes, under both dehydrogenative and reductive coupling conditions.73,83,84 Extensive investigation supports the assignment of poly(1,3Si6) to an ensemble of macrocycles with between 5 to 15 repeat units, as well as linear oligomers that are too short to cyclize (a shoulder in the size exclusion chromatogram assigned to short oligomers is labeled with a blue arrow in Figure 3). These conclusions are made on the basis of analysis of molecular weight distribution, comparison of experimental and calculated spectroscopic data, vibrational and absorbance spectroscopy, and polymer end group characterization. Scheme 2. Cyclic and linear polymers can arise from polymerization of 1,3Si6.

Page 4 of 27

n-(1,3Si6)6

n-(1,3Si6)6 Methyl groups omitted Figure 4. DFT optimized structures of n-(1,3Si6)6, shown with and without methyl groups. Blue = silicon, black = carbon, pink = hydrogen. DFT B3LYP/6–31G(d).

Table 2. Selected structural parameters for c-(1,3Si6)6 and n-(1,3Si6)6. Linear Minimum

Maximum

Mean

Si-Si Bond Length (Å)

2.37

2.39

2.38

Torsion Angle (deg)

93.6

96.1

94.7

Ring Angle (deg)

107

119

112

Butterfly Minimum

3. Density Functional Theory (DFT) Calculations: Structure. DFT calculations provide insight into the structure and expected characteristics of linear and cyclic forms of poly(1,3Si6). The cyclic hexamer c-(1,3Si6)6 and linear hexamer n-(1,3Si6)6 were chosen for these studies as the best compromise between computational tractability and experimental relevance. Geometry optimization was performed at the B3LYP/6-31g(d) level of theory. The linear structure adopts a helical conformation due to an approximately 95° torsion angle between rings (Figure 4 and Table 2). Depending on the starting structure, two different conformations of c(1,3Si6)6 are identified as energetic minima, a nonsymmetric conformer assigned to the C1 point group and a butterfly-like high-symmetry conformer assigned to the D3d point group (Figure 5). The nonsymmetric conformer is higher in energy by 7.65 kcal mol-1 relative to the butterfly conformer. Both optimized c-(1,3Si6)6 conformers are approximately 2 nm across, with a central pore between 5 and 7 Å wide. Both structures exhibit twisting between adjacent cyclohexasilane rings. In the butterfly conformer, rings alternate up and down and torsion angles are relatively small (between 1.4 and 14°). The same up-down alternation exists in the C1 conformer, but torsion angles vary more significantly, between a minimum of 7.8° and a maximum of 125° (Table 2). This results in modest curvature. The exo and endo faces are labeled in Figure 5c-d.

Maximum

Mean


2.37

2.40

2.38

Torsion Angle (deg)

1.35

14.2

6.65

Ring Angle (deg)

106

115

110

Nonsymmetric Minimum

Maximum

Mean


2.37

2.41

2.38

Torsion Angle (deg)

7.84

125

58.9

Ring Angle (deg)

106

117

111

Every cyclohexasilane ring in the linear and cyclic structures adopts a chair conformation. Evidence of significant ring strain is not apparent (Table 2). Within each ring, bond angles are close to the expected value for a tetrahedral center (109.5°). The average calculated Si-Si bond length (2.38 Å) is typical of oligosilanes and corresponds well with the Si-Si bond lengths observed in the crystal structure of 1,4Si6 (2.34± 0.001 Å).37 In both conformers of c-(1,3Si6)6, all hydrides are in the axial position, while the larger neighboring 1,3Si6 rings are in the equatorial positions. This is most easily seen in a simplified representation of the c-(1,3Si6)6 skeleton in which methyl groups are omitted for clarity (Figure 5b and 5d). The central pore is lined with Si−H bonds. Pairs of diaxial Si−H’s alternate orientation between the two faces of the macrocycle. In the C1 conformer, the curvature results in the endo face having the directional Si−H pairs oriented towards the interior of the pore, while the exo face is somewhat less organized.


Page 5 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5. DFT optimized structures of c-(1,3Si6)6. a) Conformation with D3d symmetry. b) Conformation with C1 symmetry. Methyl groups omitted for clarity where indicated. Blue = silicon, black = carbon, pink = hydrogen. DFT B3LYP/6–31G(d).



Page 6 of 27

Figure 6. Structural characterization of poly(cyclosilane)s by NMR spectroscopy (benzene-d6). Resonances are labeled as secondary (SiH2) or tertiary (SiH) silanes. The assigned nuclide is italicized. a) Cropped 1H NMR spectra of (top to bottom) poly(1,4Si6), 1,4Si6, poly(1,3Si6), and 1,3Si6. Only the SiH region is shown. b) 1H-29Si HSQC spectrum of poly(1,3Si6). The cross-peak is labeled with its coordinates. c) Cropped 29Si {1H} DEPT NMR spectra of (top to bottom) poly(1,4Si6), 1,4Si6, poly(1,3Si6), and 1,3Si6. Only the SiH region is shown. d) 29Si INEPT+ NMR of (top to bottom) poly(1,4Si6), 1,4Si6, poly(1,3Si6), and 1,3Si6. Only the SiH region is shown. HSQC = heteronuclear single quantum coherence; DEPT = distortionless enhanancement by polarization transfer; INEPT+ = insensitive nuclei enhanced by polarization transfer.

4. NMR Spectroscopy. The calculations described above point to two features that distinguish the macrocyclic from the linear structure: high symmetry and the absence of SiH2 end groups. 1H and 29Si NMR spectroscopies provide compelling evidence that poly(1,3Si6) is both highly symmetric and lacking resonances consistent with SiH2 functional groups. Comparison of experimental NMR spectra of the “para” and “meta”-like structural isomers 1,4Si6 and 1,3Si6 supports the ability of these building blocks to template unique product architectures. The 1H NMR spectra of monomers and polymers are shown in Figure 6a. A sharp singlet at δ 3.28 is assigned to the SiH2 groups of 1,3Si6. Rapid ring inversion on the NMR time scale causes the inequivalent axial and equatorial protons to coalesce.88 The SiH2 groups of 1,4Si6 have a similar chemical shift (δ 3.27) as observed in 1,3Si6. Upon polymerization, however, poly(1,3Si6) and poly(1,4Si6) exhibit very different 1 H NMR spectra. Linear poly(1,4Si6) has several broad resonances of roughly equal intensity in the typical Si–H region (3.5 to 3.0 ppm). These resonances are assigned to internal tertiary silanes (SiH) and secondary (SiH2) end group silanes

arising from a linear polymer structure on the basis of 1H-29Si HSQC NMR spectroscopy.37 In poly(1,3Si6), the SiH region of the 1H NMR spectrum is dominated by one broad singlet at δ 3.25. The 1H NMR spectrum of poly(1,3Si6) is more consistent with the high symmetry macrocyclic structure than a linear polymer structure. Our initial assignment of the poly(1,3Si6) δ 3.26 resonance to a silane proton is supported by 1H-29Si HSQC NMR spectroscopy, which shows a cross-peak between this resonance and a family of 29Si NMR resonances centered around an intense peak at δ –112 (Figure 6b). Further support for the assignment of the correlated resonances to tertiary silanes arises from 29Si NMR spectroscopy.89,90 The degree of H-substitution strongly influences the chemical shift of the 29Si nuclide. 29Si {1H} DEPT NMR spectra for 1,3Si6, poly(1,3Si6), 1,4Si6, and poly(1,4Si6) are shown in Figure 6c. Poly(1,3Si6)’s δ –112 resonances are observed at lower frequency than the δ –98.7 resonance assigned to the secondary silane (SiH2) group of the monomer. A less intense resonance (labeled with an asterisk) is assigned to the secondary silane end groups of short linear oligomers that have not macrocyclized.


Page 7 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Coupling constants are consistent with the proposed assignments. In the 29Si INEPT+ NMR spectrum, the δ –112 resonances appear as a family of overlapping doublets. The coupling constant (J = 160.9 Hz) is consistent with typical one-bond coupling constants between proton and silicon (1JSi-H = 420 – 75 Hz). The resonance assigned to the end group of uncyclized short oligomers (labeled with an asterisk, Figure 6d) has the expected triplet splitting pattern (J = 169.9 Hz). 5. Vibrational Spectroscopy. Experimental attenuated total reflectance-Fourier transform infrared (ATR-FTIR) spectroscopy and calculated vibrational spectra for the linear and cyclic hexamers provide additional evidence for a highsymmetry structure featuring Si3SiH functional groups. Figure 7 compares the experimental ATR-FTIR spectra of monomer and polymer. Both materials show resonances consistent with a Si−H stretching frequency and with a Si−Me stretching frequency. While the Si−Me stretching frequency changes minimally on polymerization, the Si−H stretching frequency shifts to lower frequency by appx. 30 cm-1 (Table 3).

ment with the simulated spectrum of the cyclic hexamer than the linear hexamer (Figure S1). The Si–H region of the n(1,3Si6)6 simulated spectrum is complex, as multiple factors (axial or equatorial, degree of substitution, and symmetry) influence the frequency. Higher frequency resonances correspond to vibrations of the secondary silane end groups, such as the symmetric stretch of the equatorial Si–H end group at 2075 cm-1, while internal Si–H stretching frequencies are found at lower frequency (Figure 8a). The higher symmetry of the cyclic conformers leads to simulated spectra that are simpler than observed for the linear hexamer. Both conformers have similar simulated IR spectra, with features at appx. 2050 and 2060 cm-1 (Figures S1 and 8b). We note that comparison of the calculated spectra of 1,3Si6 (Figure S2) and c-(1,3Si6)6 reproduces the experimentally observed 30 cm-1 shift to lower frequency of the Si–H stretching frequency upon macrocyclization. Good agreement of the experimental and calculated IR spectra of TMS3SiH (Figure S3, ν(Si–H) = 2051 cm-1 (calc.), 2049 cm-1 (expt.)) supports the validity of our calculations. Calculated IR spectra using the M06-2X functional (includes dispersion forces)87 were less successful in reproducing the experimental spectrum of TMS3SiH (Figure S4). After bench top handling, both TMS3SiH and poly(1,3Si6) show resonances consistent with Si-O-Si features, suggesting that the weak Si3SiH bond accelerates autoxidation.82

Figure 7. Cropped ATR-FTIR spectra of 1,3Si6 (dashed line) and poly(1,3Si6) (solid line). Only the a) Si–H and b) Si–Me stretching frequencies are shown. Data obtained from neat or pressed samples at room temperature.

Table 3. Selected functional group stretching frequencies of 1,3Si6, poly(1,3Si6) and TMS3SiH. 1,3Si6 (cm-1)

Poly(1,3Si6) (cm-1)

TMS3SiH (cm-1)

ν(Si−H)

2077

2046

2049

ν(Si−Me)

1243

1243

1243

The shift to lower frequency is consistent with a change from a secondary to tertiary silyl-substituted silane. Tris(trimethylsilyl)silane (TMS3SiH) is a representative silylsubstituted tertiary silane. The Si−H bond in TMS3SiH is significantly weaker than in triethylsilane (79.0 versus 90.1 kcal mol-1) due to the weakening effect of silyl-substitution compared to alkyl-substitution.85,86 Weaker bonds resonate at lower frequencies and the Si−H stretching frequency in TMS3SiH (2049 cm-1) is observed at a lower frequency than in 1,3Si6 (2077 cm-1) or triethylsilane (2107 cm-1). The observed close agreement of the poly(1,3Si6) and TMS3SiH Si–H stretching frequencies (2046 versus 2049 cm-1, Table 3) support assignment to a tertiary silyl-substituted silane. Selected Si–H stretching frequencies from simulated IR spectra of the linear and cyclic hexamers are shown in Figure 8. The experimental observation of a single broad Si–H stretching frequency centered at 2049 cm-1 is in better agree-

Figure 8. Optimized structures of a) n-(1,3Si6)6 and b) C1 c(1,3Si6)6 labeled with calculated stretching frequencies of selected Si–H bonds. Methyl groups omitted for clarity. DFT B3LYP/6– 31G(d).

5. Absorbance Spectroscopy and Time-Dependent DFT Calculations. The optical properties of extended silanes are very different from the structurally analogous long-chain alkanes.48 Oligo- and polysilanes strongly absorb ultraviolet light, while alkanes have higher energy excitations (

Directional Building Blocks Determine Linear and Cyclic Silicon

Recommend Documents