Attempts to Synthesize a Thiirane, Selenirane, and Thiirene by Dealkylation of Chalcogeniranium and Thiirenium Salts.

Thiiranium salts [Ad2 SR]+ X- (5, 8, 9, 11, 12; X- =Tf2 N- (Tf=CF3 SO2 ), SbCl6 - ) and seleniranium salts [Ad2 SeR]+ X- (14, 16, 17, 23-25; X- =Tf2 N- , BF4 - , CHB11 Cl11 - , SbCl6 - ) are synthesized from strained alkene bis(adamantylidene) (1). The disulfides and the diselenides (Me3 SiCH2 CH2 E)2 (4, 13), (tBuMe2 SiCH2 CH2 E)2 (7, 22), and (NCCH2 CH2 E)2 (10, 15; E=S, Se) have been used. The thiirenium salts [tBu2 C2 SR]+ X- (34) and [Ad2 C2 SR]+ X- (35, 36) are prepared from the bis-tert-butylacetylene (2) and bis-adamantyl-acetylene (3) with R=Me3 SiCH2 CH2 and tBuMe2 SiCH2 CH2 . Attempts to cleave off the groups Me3 SiCH2 CH2 , tBuMe2 SiCH2 CH2 , and NCCH2 CH2 resulted in thiiranes 27, 30. No selenirane Ad2 Se (33) is formed from seleniranium salts, instead cleavage to the alkene (1) and diselenide (13, 15) occurs. The thiirenium salt [Ad2 C2 SCH2 CH2 SiMe3 ]+ Tf2 N- (35) does not yield the thiirene Ad2 C2 S (37), the three-membered ring is cleaved, forming the alkyne (3) and disulfide (4). All compounds are characterized by ESI mass spectra, NMR spectra, and by quantum chemical calculations. Crystal structures of the salts 8, 12, 25, 17, 26, 36 and the thiiranes 27, 30 are presented.

Introduction

Thiiranium and thiirenium salts are long known. In recent years, we have prepared selenirenium und tellurirenium salts[ , ] and also the saturated seleniranium and telluriranium salts, see also ref. [6]. These are key intermediates in electrophilic addition reactions of organochalcogen cation equivalents [RE+X−] to alkynes and alkenes.[ , , , , , , , ]

However, the knowledge about uncharged C2‐chalcogen three‐membered rings is limited almost to thiiranes. Seleniranes have been described as unstable intermediates and also in substance, but no structural proof for this class of compounds exists. The postulated instability, meaning the easy decomposition into alkene and elemental selenium, calls into question their existence, see here the discussion about seleniranes and telluriranes by Braunschweig. Unexpectedly, results by Sardar et al. are surprising, who claim to have made seleniranes at high temperatures, which have very unusual 77Se NMR shifts in the region of 600–700 ppm. So far, telluriranes have not been mentioned in the literature. The search for thiirenes has been going on for a long time. These would be of great theoretical interest as antiaromatic cyclic 4π‐electron compounds.[ , ] This would explain their extreme instability. The calculated nucleus‐independent chemical shift (NICS (1)) value of −2.2 ppm does not allow a clear statement on the question of antiaromaticity. With natural bond orbital (NBO), a hybridization of 3s1.8 3p4.16 is calculated for the S atom in thiirene (see Table S1 in the Supporting Information). Early synthetic attempts employed photolysis of 1,2,3‐thiadiazoles. At best, thiirenes have been characterized by the matrix isolation technique at very low temperatures in combination with IR spectroscopy, but structural proof is lacking. Thiirensulfoxides and thiirensulfones are known, however, but attempts of deoxygenation to thiirenes failed. Our long‐running work with such unstable three‐membered ring chalcogen compounds has motivated us to erase this white spot with some new attempts. Our plan for generating thiiranes, seleniranes, and thiirenes was first to generate the C2‐chalcogen three‐membered ring motive by electrophilic addition of an [RE+X−] equivalent to an alkene or alkyne under formation of chalcogeniranium and thiirenium salts.[ , , ] The positively charged S or Se atoms should bear a cleavable group, so that after cleavage the uncharged thiiranes, seleniranes, and thiirenes should be obtained. The cleavage of a Me3SiCH2CH2 group should be done by F−, whereas the NCCH2CH2 group can be cleaved of by bases such as OH−. Both are well known protecting groups for the SH and the SeH functions[ , ] (see Scheme 1). We wanted to test these methods by the dealkylation reactions of thiiranium salts under formation of stable thiiranes. Then, this procedure should be transferred to the preparation of seleniranes und thiirenes. The expected highly unstable seleniranes should be stabilized by the use of the extreme sterically demanding alkene bis(adamantylidene) Ad=Ad (1). The latter has enabled us already to achieve the isolation of the extremely unstable telluriranium salts.

Scheme 1

Synthetic concept for thiiranes, seleniranes, and thiirenes.

Thiirenes should be stabilized by the sterically demanding alkynes tBuC≡CtBu (2) and diadamantyl acetylene AdC≡CAd (3).[ , ] The interesting question is whether F− or OH− would indeed attack at the protecting group, or if the reaction would occur at the C or chalcogen ring atoms.

Results and Discussion

We used RE+ electrophiles for the preparation of the three‐membered ring salts that we developed previously: either by two‐electron oxidation of the disulfides and diselenides with XeF2 in combination with fluoride ion acceptors like Tf2NSiMe3, BF3 ⋅OEt2, or Me3Si+CHB11Cl11 −,[ , ] or by chlorination with SO2Cl2. The reagent [RE+SbCl6 −] is generated subsequently by reaction with SbCl5.[ , ]

Thiiranium salts

The disulfide (Me3SiCH2CH2S)2 (4) is oxidized with XeF2/Tf2NSiMe3 and reacted in the presence of the alkene 1, yielding the thiiranium salt [Ad2SCH2CH2SiMe3]+Tf2N− (5). The Me3Si group in 4 is not attacked by XeF2. By chlorination of 4 with SO2Cl2 and subsequent reaction with SbCl5 and 1, the chloroethyl thiiranium salt [Ad2SCH2CH2Cl]+SbCl6 − (6) is formed, which is identified by ESI‐MS. Here, the Me3Si group is cleaved off and substituted by Cl. When using the more stable tBuMe2Si group in the disulfide (tBuMe2SiCH2CH2S)2 (7), the method with SO2Cl2/SbCl5 and 1 produces the salt [Ad2SCH2CH2SiMe2 tBu]+SbCl6 − (8). With 7, XeF2/Tf2NSiMe3, and 1, the salt 9 is formed.

The cyanoethyl thiiranium salts 11 and 12 are reacted with (NCCH2CH2S)2 (10) and 1, either with XeF2/Tf2NSiMe3, or with SO2Cl2/SbCl5 (see Scheme 2).

Scheme 2

Synthesis of the thiiranium salts with the Me3SiCH2CH2, tBuMe2SiCH2CH2, and NCCH2CH2 groups.

In the ESI mass spectra, the new thiiranium salts show the expected molecular peaks for the cations and anions. The 13C NMR spectra show the characteristic signal of the three‐membered ring C atoms at 96–99 ppm, compared with 67–105 ppm in ref. [1e] and [25]. The less soluble SbCl6 − salts crystallize well, and the crystal structures of [Ad2SCH2CH2SiMe2 tBu]+SbCl6 − (8) and [Ad2SCH2CH2CN]+SbCl6 − (12) have been obtained (see Figures 1 and 2). These new thiiranium ions have C−C bond lengths of 150.0–150.6 pm within the three‐membered ring, C−S bond lengths of 189.4–191.8 pm, and C‐S‐C angles of 46.4°. The C−S bond to the substituent at the S atom are a little shorter (182.3–183.8 pm) than the C−S bonds within the ring. Known thiiranium salts have quite similar bond parameters.[ , , ]

Figure 1

Molecular structure of [Ad2SCH2CH2SiMe2 tBu]+SbCl6 − ⋅EtCN (8), with the probability ellipsoids drawn at 50 %. The anion and solvent have been omitted for clarity. Selected bond parameters [pm/°]: C1−C11 150.6(6), S−C1 190.1(5), S−C11 191.6(5), S−C21 183.8(5), C21−C22 153.2(7), Si−C22 189.8(5); C1‐S‐C11 46.47(19).

Figure 2

Molecular structure of [Ad2SCH2CH2CN]+SbCl6 − ⋅CH2Cl2 (12), with the probability ellipsoids drawn at 50 %. The anion and solvent have been omitted for clarity. Selected bond parameters [pm/°]: C1−C9 150.0(4), S−C1 191.8(3), S−C9 189.4(3), S−C23 182.3(3), C22−C23 153.1(5), C21−C22 146.5(1), N−C21 113.9(5); C1‐S‐C9 46.35(12).

Seleniranium salts

These methods of synthesis can only be applied in part to the Se compounds. The diselenide (Me3SiCH2CH2Se)2 (13) reacts with XeF2/Tf2NSiMe3 and 1, forming the seleniranium salt 14, which decomposes already at room temperature with formation of elemental selenium (see Scheme 3). However, 13C and 77Se NMR spectra can be obtained directly from the reacting solution at −40 °C, which are in accordance with the structure of 14. In particular, the 13C signal of the ring C atom at 110 ppm and the 77Se signal of the highly shielded Se atom at 35 ppm are typical values. With the diselenide (NCCH2CH2Se)2 (15) both synthetic routes are successful, namely with the oxidation by XeF2/BF3 ⋅OEt2, and the chlorination with SO2Cl2, reaction with SbCl5 and addition of 1. For this sake, the synthesis of the cyanoethyldiselenide (15) had to be improved, as according to ref. [24a] only mixtures of di‐ and triselenide are obtained. The initially prepared cyanoethyl selenocyanate (18) reacts with LiEt3BH, and oxidation with O2 selectively gives the diselenide 15 (see Scheme 4). The instability of the seleniranium salts with the Me3SiCH2CH2 group contrasts with the much higher stability of the non‐functionalized seleniranium salts. Therefore, we tried to prepare the salts carrying the tBuMe2SiCH2CH2 group, as we assumed that these would be more stable. The corresponding diselenide (tBuMe2SiCH2CH2Se)2 (22) is synthesized from the vinylsilane (19). Hydroboration with 9‐borabicyclononane (9‐BBN) and oxidation of the borane with H2O2 gives the silylethanol (20), see also ref. [29], followed by bromination with PPh3/CBr4 affording tBuMe2SiCH2CH2Br (21). This, however, reacts with Li2Se2 to diselenide, which is contaminated with considerable amounts of triselenide. The product mixture is reduced in liquid NH3/THF with Na and afterwards reoxidized with O2 to the diselenide 22 (see Scheme 4). The disulfide (tBuMe2SiCH2CH2S)2 (7) is prepared similarly with Li2S2.

Scheme 3

Synthesis of the seleniranium salts carrying the Me3SiCH2CH2 and NCCH2CH2 groups.

Scheme 4

Synthesis of the diselenides 15 and 22.

(tBuMe2SiCH2CH2Se)2 (22) reacts with 1 and XeF2/Tf2NSiMe3 or XeF2/Me3Si+CHB11Cl11 −, and here also with SO2Cl2/SbCl5 to the seleniranium salts 23–25 (see Scheme 5). The salts carrying the tBuMe2SiCH2CH2 group decompose slowly in solution at room temperature, for example, during NMR measurements. They can be isolated as solid compounds and are stable for prolonged time at −20 °C. The 13C NMR spectrum shows signals that are typical for the three‐membered ring at 110 ppm, and 77Se signals at high field 24–39 ppm. These values are very close to those of the [Ad2SeEt]+ ion. ESI mass spectra have the correct molecular peaks for cations and anions.

Scheme 5

Synthesis of the seleniranium salts carrying the tBuMe2SiCH2CH2 group.

Crystal structure determinations of the new seleniranium salts [Ad2SeCH2CH2SiMe2 tBu]+SbCl6 − (25) and [Ad2SeCH2CH2CN]+SbCl6 − (17) prove the structures (see Figures 3 and 4). By reacting of 22 with SO2Cl2/SbCl5 and 1, the side product [Ad2SeCH2CH2Cl]+SbCl6 − (26) is isolated and characterized by ESI‐MS and crystal structure determination (see Figure S1 in the Supporting Information). Here also, a partial cleavage takes place even of the tBuMe2Si group. The structural parameters within the three‐membered rings in these seleniranium salts (C−C bond lengths of 147.9–148.9 pm, C−Se bond lengths of 205.5–209.7 pm, C−Se bond lengths to the groups at the Se atoms of 196.5–200.2 pm, and C‐Se‐C bond angle of 41.8–42.4°) are almost identical to those of our [Ad2SeEt]+ salt, see also ref. [6a].

Figure 3

Molecular structure of [Ad2SeCH2CH2SiMe2 tBu]+SbCl6 − ⋅CH2Cl2 (25), with the probability ellipsoids drawn at 50 %. The anion and solvent have been omitted for clarity. Selected bond parameters [pm/°]: C1−C2 147.9(4), Se−C1 206.0(3), Se−C2 207.3(3), Se−C21 197.3(3), C21−C22 152.8(4), Si−C22 189.6(3); C1‐Se‐C2 41.94(11).

Figure 4

Molecular structure of [Ad2SeCH2CH2CN]+SbCl6 − ⋅CH2Cl2 (17), with the probability ellipsoids drawn at 50 %. The anion and solvent have been omitted for clarity. Selected bond parameters [pm/°]: C1−C19 148.9(9), Se−C1 209.7(5), Se−C19 207.7(5), Se−C21 200.2(6), C21−C22 149.1(9), C22−C23 146.7(10), N−C23 111.1(9); C1‐Se‐C19 41.8(2).

Dealkylation of the thiiranium salts—thiiranes

We began the attempts to cleave off the protecting groups with the thiiranium salts. Salt 5 reacts surprisingly selectively with Bu4NF in THF and under cleavage of the Me3SiCH2CH2 group at −40°. The thiirane Ad2S 27 was isolated. Then, we tried to combine the preparation of the thiiranium salt and its dealkylation in a one‐pot procedure. Alkene 1 reacts with disulfide 4 and XeF2/BF3 ⋅OEt2 and the formed thiiranium salt is reacted with the Bu4NF solution without isolating it. The expected thiirane 27 is obtained in quantitative yield. Its molecular structure has been determined by crystal structure determination (see Figure 5).

Figure 5

Molecular structure of thiirane 27, with the probability ellipsoids drawn at 50 %. Selected bond parameters [pm/°]: C1−C11 150.73(16), S−C1 184.92(12), S−C11 185.11(12); C1‐S‐C11 48.08(5).

Cis‐cyclooctene (28) as a simple sterically unstrained alkene is reacted in a one‐pot procedure with 4 and XeF2/BF3 ⋅OEt2, and subsequently with Bu4NF. The preparation of the thiirane 30 was also successful, as proven by the cis‐structure in the crystal. Thiirane 30 exists in the solid state in two different forms in a 1:1 ratio, and these are mirror images of each other. Figure 6 shows both molecules, the S atoms pointing to the viewer. In the crystalline state they are oriented in a different manner.

Figure 6

Molecular structure of thiirane 30, with the probability ellipsoids drawn at 50 %. Selected bond parameters [pm/°]: C11−C18 147.2(6), S1−C11 183.4(5), S1−C18 183.1(5); C11‐S1‐C18 47.38(18), C12‐C11‐C18‐C17 −1.902(720); C1−C8 148.9(6), S2−C1 184.0(5), S2−C8 184.2(5); C1‐S2‐C8 47.71(19), C2‐C1‐C8‐C7 0.039(743).

If trans‐cyclooctene 29 is subjected to the same protocol, the same cis‐thiirane 30 is formed as with the cis‐alkene; the products have identical 13C NMR spectra. Addition of the Me3SiCH2CH2S+ electrophile to the trans‐alkene results in a configuration inversion of the strained eight‐membered ring system. Cis–trans isomerization of thiiranium ions occur via ring‐opened carbocations.[ , ] There are incorrect data in ref. [30]: a postulated trans‐9‐thiabicyclo[6.1.0]nonane has exactly the same 13C NMR data as our cis‐compound 30. For other 13C NMR data of the trans‐isomer, see ref. [31].

The reaction of Bu4NF with the thiiranium salt 9 carrying the tBuMe2SiCH2CH2 group does not give the thiirane 27, but the cleavage of the tBuMe2SiCH2CH2S group is observed. The three‐membered ring is attacked instead: the 13C NMR spectrum shows only the alkene 1 and the disulfide (tBuMe2SiCH2CH2S)2 (7), and only traces of the thiirane 27. The tBuMe2SiCH2CH2 group is clearly too stable as a protecting group for our purposes (see Scheme 6).

Scheme 6

Cleavage of the protecting groups from the thiiranium salts.

The NCCH2CH2 group in the thiiranium salts 11 and 12 is also not a useful protecting group for the liberation of the thiirane 27. Instead, 11 and 12 react after 3 h at 0 °C with CsOH in H2O/THF or MeOH/THF to mixtures of 1 and the thiirane 27 in the ratios of 1:3 to 1:1, according to the 13C NMR spectra. Besides the dealkylation, a lot of cleavage takes place of the three‐membered ring.

The structures of the thiirane 27 and thiiranium salts 8 and 12 have C−C bonds of similar lengths (150.0–150.7 pm). In the thiirane, the C−S bonds are a little shorter (185 pm) than in the salts (189.4–191.8 pm). In 27, the C‐S‐C angle is consequently a little larger (48.1°) compared with 8 and 12 (46.4–46.5°).

Attempts to dealkylate seleniranium salts—selenirane?

For the sake of detection and preparation of a selenirane, reactions were undertaken to cleave off the Me3SiCH2CH2 group from seleniranium salts. Learning from the results with the thiiranium salts, we have used for these reactions solely compounds with this protecting group. The [Ad2SeCH2CH2SiMe3]+ salts turned out to be unstable, so these reactions were done in a one‐pot procedure, meaning the preparation of the seleniranium salts is followed by immediate reaction with Bu4NF. Because of the expected instability of the target compound, the preparation of the seleniranium salt and the dealkylation were done at −40 °C.

We calculated by DFT the 13C and 77Se NMR shifts of the selenirane ring, together with those of the thiiranium and seleniranium ions and the thiirane 27. Considering the small differences between the calculated and experimental data, the selenirane should have a 13C signal at about 85 ppm and a 77Se resonance at about 70 ppm (see Scheme 7).

Scheme 7

Calculated 13C and 77Se NMR chemical shifts [ppm] (δ=σ ref−σ comp; by using GIAO‐B3PW91/cc‐pVTZ//B3PW91/6‐311+G(d,p) theory), relative to TMS and Me2Se, including experimental data.

First, we prepared the seleniranium salt 14 by reacting 1, the diselenide 13, and XeF2/Tf2NSiMe3 at −40 °C. A solution of Bu4NF in THF is added at −78 °C, followed by stirring at −40 °C. The 13C NMR measurement of the reaction solution at −78 °C shows only the compounds 1 and 13, whereas the 77Se spectra show the presence of 13 and an organoselenium trifluoride, most likely Me3SiCH2CH2SeF3 by a new signal at 1187 ppm (for the NMR spectra of RSeF3, see refs. [32], [33]). There is no indication of the selenirane 33. The F− ion possibly attacks the selenium atom of the seleniranium ring, which is cleaved to the alkene 1 and the selenium monofluoride Me3SiCH2CH2SeF (31). The latter is unstable and disproportionates according to 3RSeF→R2Se2+RSeF3 in a known manner[ , , ] to the diselenide (Me3SiCH2CH2Se)2 (13) and the trifluoride Me3SiCH2CH2SeF3 (32; see Scheme 8).

Scheme 8

Attempts for preparation of a selenirane.

We also attempted a reaction of 1, 13, XeF2/BF3 ⋅OEt2, and Bu4NF, with the same results.

The reaction of the seleniranium salt 16 with CsOH in THF/H2O at −40 °C cleaves the three‐membered ring selectively to the olefin 1 and the diselenide 15, a selenirane is not observed (see Scheme 8).

Thiirenium salts

The thiirenium salt 34 is prepared by oxidation of the disulfide 4 with XeF2/BF3 ⋅OEt2 in the presence of di‐tert‐butylacetylene (2). The synthesis succeeds also for the thiirenium salts 35 and 36 by reaction of the diadamantylacetylene (3) with 4 and XeF2/Tf2NSiMe3 or the disulfide 7 and XeF2/Tf2NSiMe3 (see Scheme 9). These compounds crystallize badly and often are obtained as oils. However, single crystals have been obtained of the compound 36. Figure 7 shows the crystal structure.

Scheme 9

Synthesis of thiirenium salts.

Figure 7

Molecular structure of [Ad2C2SCH2CH2SiMe2 tBu]+Tf2N− (36), with the probability ellipsoids drawn at 30 %. The anion has been omitted for clarity. Selected bond parameters [pm/°]: C21−C22 128.9(11), S−C21 185.4(11), S−C22 181.0(9), S−C23 182.6(6), C23−C24 152.6(7), Si−C24 188.7(6); C21‐S‐C22 41.2(3).

The 13C NMR signals of the ring C atoms of these thiirenium ions are in the same region (113–115 ppm) as in the unsubstituted thiirenium salts.[ , , , ]

The structure of 36 has a C=C bond length of 128.9 pm within the three‐membered ring, the S−C bonds are 185.4 and 181.0 pm long, to the tBuMe2SiCH2CH2 group 182.6 pm. The C‐S‐C angle is 41.2°. This three‐membered ring is quite similar to those in the non‐substituted alkylthiirenium salts.[ , , ]

Attempts towards dealkylation of thiirenium salts—thiirene?

We tried to cleave off the Me3SiCH2CH2 group from the thiirenium salt 35 by reaction with Bu4F at −60 °C. 13C NMR shows that, analogously to the seleniranium salts, a selective attack by the F− ions occurs at the sulfur atom. Alkyne 3 is liberated, and there is no indication of the formation of thiirene 37. No cleavage of the Me3SiCH2CH2 group is observed (see Scheme 10).

Scheme 10

Attempts for syntheses a thiirene.

Summing up: The method of cleaving off the protecting groups works only with thiiranium salts, where the Me3SiCH2CH2 group leaves and a thiirane is liberated. In thiiranium salts with the tBuMe2SiCH2CH2 or NCCH2CH2 groups, in seleniranium salts with the Me3SiCH2CH2 and NCCH2CH2 groups, and in thiirenium salts with the Me3SiCH2CH2 group, the attack by F− or OH− takes place at the heteroatom of the three‐membered ring. In these cases, the RS+ or RSe+ part is cleaved off, and an alkene or alkyne is formed. In sterically unprotected thiiranium and seleniranium ions, the attack by nucleophiles happens mostly at the C atom of the three‐membered rings,[ , , , , , , , ] very similar to the thiirenium ions. This is also the case during the fluorothiolation and fluoroselenation of alkenes and the fluoroselenation of alkynes.[ , , ] But there are examples known where the RS+ group is transferred from thiirenium ions to transition metal complexes or C nucleophiles.[ , ] Finally, we want to point to the transition of RS+ and RSe+ groups in thiiranium, seleniranium, or thiirenium ions to alkenes and alkynes, where the C−C‐multiple bond preferably interacts with the heteroatom of the three‐membered ring.[ , , , ]

We tried to understand the different reactivity of the compounds carrying the Me3SiCH2CH2 groups by calculating the HOMOs and LUMOs of the thiiranium (A), seleniranium (B), and thiirenium ions (C), as shown in Figure 8. In all three ions, the HOMO is spread over the entire molecule including the Me3SiCH2CH2 group. The LUMO, however, is localized at the three‐membered ring, especially at the heteroatom. This is also the case in the unsubstituted ions [Ad2EEt]+ (D, E) and those with the cyanoethyl group [Ad2ECH2CH2CN]+ (F, G; E=S, Se, see Figure S2 in the Supporting Information). In an orbital controlled attack by the F− ions there should be an interaction with the LUMO of the three‐membered ring. Therefore, the cleavage of the RSe+ group from B or the cleavage of the RS+ group from C by the F− ion is more likely than the attack at the Me3Si group. The lower lying LUMO energy in the ions B and C relative to A should favor the attack at the three‐membered ring. As the NBO‐calculated positive charge on the S atom in thiiranium ion A is lower than at the heteroatoms in B and C, an attack by the F− ions at the Si atom in A in competition to the S atom is comprehensible.

Figure 8

HOMO and LUMO of the ions A–C, orbital energies [Hartrees], and NBO charges, calculated at the B3PW91/6‐311+G(d,p) level of theory.

Conclusion

Thiiranium salts [Ad2SCH2CH2SiMe3]+X− are dealkylated by the F− ion, giving the thiirane Ad2S. Thiiranium salts with the tBuMe2SiCH2CH2 or NCCH2CH2 group, seleniranium salts [Ad2SeCH2CH2SiMe3]+X−, and [Ad2SeCH2CH2CN]+X− as well as thiirenium salts [Ad2C2SCH2CH2SiMe3]+X− are attacked by F− or OH− at the chalcogen atom of the three‐membered ring, under cleavage of the RS+ or RSe+ groups and liberation of the alkene and alkynes. The synthesis of a selenirane and a thiirene remains an open challenge.

Experimental Section

General

Dried solvents and argon as protection gas were used. 11B, 13C, 19F, 29Si, and 77Se NMR spectra were recorded with a JEOL ECZ 400R or with a JEOL ECS 400 spectrometer (128.25, 100.51, 376.13, 79, and 76.24 MHz, respectively). Chemical shifts δ are reported in ppm relative to BF3 ⋅OEt2 (11B), Me4Si (13C, 29Si), CFCl3 (19F), and Me2Se (77Se).

For ESI‐TOF mass spectra, the samples were measured from CH3CN, CH3CN/CH2Cl2, CH3OH, or CH3CN/CH3OH solutions with an Agilent 6210 ESI‐TOF, Agilent Technologies, Santa Clara, CA, USA. The solvent flow rate was adjusted to 4 μL min−1, spray voltage set to 4 kV, drying gas flow rate was set to 15 psi (1 bar; ESI‐TOF=electrospray ionization–time of flight). Non‐ionic compounds were analyzed with a HR‐EI‐MS (Autospec Premier, Waters Co., Milford, MA, USA) using 80 eV electron energy.

Quantum chemical calculations

Calculations were performed with Gaussian 16 on a high‐performance computer system SOROBAN at Zedat, Freie Universität Berlin, https://www.zedat.fu‐berlin.de/HPC/Soroban.

The olefin Ad=Ad (1), trans‐cyclooctene (29), the alkynes tBuC≡CtBu (2) and AdC≡CAd (3), the Si compounds Tf2NSiMe3 and Me3Si+ CHB11Cl11 were prepared according to known procedures.

1: 13C NMR (CDCl3): δ=133.17 (C=C), 39.64 (8×CH2), 37.37 (2×CH2), 31.93 (4×CH), 28.60 (4×CH) ppm.

2: 13C NMR (CDCl3): δ=87.08 (C≡C), 31.58 (CMe 3), 27.14 (CMe3) ppm.

3: 13C NMR (CDCl3): δ=87.59 (C≡C), 43.55 (6×CH2), 36.47 (6×CH2), 29.19 (Cq), 28.19 (6×CH) ppm.

Syntheses of (Me3SiCH2CH2S)2 (4) and (Me3SiCH2CH2Se)2 (13)

Li2S2 and Li2Se2 are prepared by reacting Li (420 mg, 60 mmol) with S (1.92 g, 60 mmol) or Se (4.47 g, 60 mmol) in liq. NH3 (150 mL). After evaporation of the NH3, THF (50 mL) was added. Under stirring, Me3SiCH2CH2Br (10.87 g, 60 mmol) was added dropwise followed by stirring at room temperature for 12 h. Addition of water and extraction with CH2Cl2 was followed by vacuum distillation.

4: Yield: 6.2 g (78 %); b.p.: 86 °C/0.19 mbar; EI‐MS: [266, M +] (C10H26S2Si2); 13C NMR (CDCl3): δ=34.77 (CH2S), 17.23 (1 J Si,C=47.9 Hz, CH2Si), −1.60 (1 J Si,C=51.0 Hz, CH3Si) ppm.

13: Yield: 7.61 g (70 %); b.p.: 100 °C/0.08 mbar; EI‐MS: [362, M +] (C10H26Se2Si2); 77Se NMR (CDCl3): δ=355.8 ppm; 13C NMR (CDCl3): δ=24.92 (1 J Se,C=69.8 Hz, CH2Se), 19.55 (1 J Si,C=47.2 Hz, CH2Si), −1.63 (1 J Si,C=50.9 Hz, CH3Si) ppm.

Synthesis of (NCCH2CH2Se)2 (15)

NCCH2CH2SeCN (18) was prepared from NCCH2CH2Br (13.4 g, 0.1 mol) and KSeCN (14.4 g, 0.1 mol) in DMF (50 mL) according to ref. [27].

To 18 (9.54 g, 60 mmol) in THF (100 mL) was added dropwise at −78 °C LiEt3BH (65 mL, 65 mmol, 1 m in THF),[ , ] followed by 1 h stirring at this temperature. After warming to room temperature and further stirring for 1 h, O2 was bubbled through the reaction solution for 30 min. THF was removed under vacuum. The remainder was dissolved in H2O (300 mL) and extracted with CH2Cl2. The product was purified by column chromatography on silica gel with mixtures of CH2Cl2/hexane: 300 mL 30:70, 450 mL 40:60, 150 mL 50:50, 150 mL 60:40, 150 mL 70:30, 150 mL 80:20.

18: Yield: 13.1 g (82 %); b.p.: 133 °C/0.09 mbar; EI‐MS: [160, M +] (C4H4N2Se); 77Se NMR (CDCl3): δ=244.2 ppm; 13C NMR (CDCl3): δ=117.97 (CN), 101.46 (1 J Se,C=237.5 Hz, SeCN), 23.24 (1 J Se,C=59.4 Hz, CH2Se), 20.21 (CH 2CN) ppm.

15: Yield: 5.2 g (49 %); EI‐MS: [268, M +] (C6H8N2Se2); 77Se NMR (CDCl3): δ=326.5 ppm; 13C NMR (CDCl3): δ=119.14 (CN), 23.00 (1 J Se,C=81.0 Hz, CH2Se), 19.79 (CH 2CN) ppm.

Syntheses of (tBuMe2SiCH2CH2S)2 (7) and (tBuMe2SiCH2CH2Se)2 (22)

Under argon, vinylsilane 19 (14.2 g, 0.1 mol) was added dropwise to a stirred solution of 9‐BBN (230 mL, 0.114 mol, 0.5 m in THF), followed by heating at reflux for 2 h, see ref. [29]. After cooling to room temperature, H2O (100 mL) was carefully added, then NaOH (14 g in 120 mL H2O) and H2O2 (120 mL, 30 %) were added, followed by heating at reflux for 1.5 h. After extraction with Et2O, very pure silylethanol 20 was obtained.

Into a stirred solution of 20 (12 g, 75 mmol) and CBr4 (27.4 g, 82.5 mmol) in CH2Cl2 (80 mL), Ph3P (21.6 g, 82.5 mmol) were added within 5 min at 0 °C, followed by 2 h stirring at 0 °C. After evaporation of the THF, the product was dissolved in pentane, filtered through silica gel, and the bromide 21 was vacuum distilled.

Compounds 7 and 22 were synthesized by producing Li2S2 or Li2Se2 from Li (208 mg, 30 mmol) and S (962 mg, 30 mmol) or Se (2.37 g, 30 mmol) in liq. NH3 (150 mL). Li2S2 or Li2Se2 were reacted with 4 and 13 similar to the described method above with tBuMe2SiCH2CH2Br 21 (6.7 g, 30 mmol). After aqueous workup and extraction with Et2O, the disulfide 7 was purified by column chromatography on silica gel with hexane.

The raw product of the selenium compound was freed from triselenide by dissolving in liq. NH3 (100 mL) and dry THF (50 mL). Na (1.4 g, 60 mmol) was added and the blue solution was stirred for 1 h. After evaporation of the NH3 and addition of H2O (100 mL), O2 was bubbled through the solution for 1 h. Purification was done by extraction with Et2O and column chromatography in hexane.

20: Yield: 16 g (100 %); 13C NMR (CDCl3): δ=59.42 (CH2OH), 26.35 (CMe 3), 17.80 (1 J Si,C=46.0 Hz, CH2Si), 16.27 (CMe3), −6.09 (1 J Si,C=49.8 Hz, CH3Si) ppm; 29Si NMR (CDCl3): δ=6.31 ppm.

21: Yield: 11.6 g (69 %); b.p.: 49–51 °C/0.2 mbar; 13C NMR (CDCl3): δ=32.33 (CH2Br), 26.42 (CMe 3), 20.41 (1 J Si,C=43.0 Hz, CH2Si), 16.60 (CMe3), −6.26 (1 J Si,C=50.2 Hz, CH3Si) ppm; 29Si NMR (CDCl3): δ=8.34 ppm.

7: Yield: 4.54 g (86 %); EI‐MS: [350.1959, M +] (C16H38S2Si2); 13C NMR (CDCl3): δ=35.28 (CH2S), 26.61 (CMe 3), 16.69 (CMe3), 13.37 (1 J Si,C=46.4 Hz, CH2Si), −6.16 (1 J Si,C=50.0 Hz, CH3Si) ppm; 29Si NMR (CDCl3): δ=8.61 ppm.

22: Yield: 4.2 g (63 %); EI‐MS: [446, M +] (C16H38Se2Si2); 77Se NMR (CDCl3): δ=361.5 ppm; 13C NMR (CDCl3): δ=26.69 (CMe 3), 25.37 (1 J Se,C=69.9 Hz, CH2Se), 16.77 (CMe3), 15.74 (1 J Si,C=45.7 Hz, CH2Si), −6.16 (1 J Si,C=49.9 Hz, CH3Si) ppm; 29Si NMR: (CDCl3): δ=8.70 ppm.

Syntheses of the thiiranium und seleniranium salts

Method A: 1/R2S2 or R2Se2/XeF2/fluoride ion acceptor

Dry CH2Cl2 (15 mL) was condensed on Ad=Ad 1 (2 mmol, 537 mg) and the disulfide 4 (1 mmol, 267 mg), 7 (1 mmol, 351 mg), 10 (1 mmol, 172 mg), or diselenide 13 (1 mmol, 360 mg), 15 (1 mmol, 266 mg), 22 (1 mmol, 445 mg) at −196 °C. XeF2 (1 mmol, 169 mg), and Tf2NSiMe3 (2 mmol, 707 mg), BF3 ⋅OEt2 (2 mmol, 284 mg), or Me3Si+CHB11Cl11 (2 mmol, 1.34 g) were added at −78 °C, followed by stirring for 30 min at this temperature and 2 h at −40 °C. Half of the solvent was pumped off, and by slow addition of hexane (30 mL), the salt was crystallized. The product was filtered off, washed with hexane (3×20 mL), and dried in vacuum. Compound 14 was measured by NMR spectroscopy after addition of a little CD2Cl2.

Method B: R2S2 or R2Se2/SO2Cl2/SbCl5/1

Dry CH2Cl2 (15 mL) was condensed on disulfide 7 (1 mmol, 351 mg), 10 (1 mmol, 172 mg), or diselenide 15 (1 mmol, 266 mg), 22 (1 mmol, 445 mg) at −196 °C. SO2Cl2 (1.1 mmol, 135 mg) was added at −20 °C, followed by stirring for 30 min at this temperature. SbCl5 (2 mmol, 598 mg) was slowly added at −78 °C, followed by stirring for 10 min at this temperature. Then, Ad=Ad 1 (2 mmol, 537 mg) was added and stirred for 2 h at −40 °C. Half of the solvent was pumped off, and by slow addition of hexane (30 mL), crystallization initiated. The salt was filtrated off, washed with hexane (3×20 mL), and dried in vacuum.

Thiiranium salts 5, 9, 11, according to method A

[Ad2SCH2CH2SiMe3]+Tf2N− (5): Yield: 0.99 g (73 %); ESI‐MS (CH3CN/CH2Cl2): [401.2692]+ (C25H41SSi+), [279.9180]− (C2F6NO4S2 −); 13C NMR (CD2Cl2): δ=119.88 (q, 1 J F,C=321.9 Hz, CF3), 95.87 (ring C), 38.55 (2×CH2), 38.19 (2×CH2), 38.08 (2×CH2), 37.02 (2×CH2), 36.07 (2×CH2), 33.57 (2×CH), 30.31 (2×CH), 27.53 (CH2S), 26.50 (4×CH), 16.17 (1 J Si,C=42.5 Hz, CH2Si), −2.66 (1 J Si,C=52.0 Hz, CH3Si) ppm; 29Si NMR (CD2Cl2)): δ=3.52 ppm.

[Ad2SCH2CH2SiMe2 tBu]+Tf2N− (9): Yield: 1.16 g (80 %); ESI‐MS (CH3CN): [443.3198]+ (C28H47SSi+), [279.9248]− (C2F6NO4S2 −); 13C NMR (CD2Cl2): δ=119.83 (q, 1 J F,C=321.6 Hz, CF3), 96.15 (ring C), 38.57 (2×CH2), 38.21 (2×CH2), 38.09 (2×CH2), 37.03 (2×CH2), 36.05 (2×CH2), 33.56 (2×CH), 30.31 (2×CH), 27.91 (CH2S), 26.50 (2×CH), 26.48 (CMe 3), 25.91 (2×CH), 16.43 (CMe3), 12.37 (CH2Si), −7.15 (MeSi); 29Si NMR (CD2Cl2): δ=10.05 ppm.

[Ad2SCH2CH2CN]+Tf2N− (11): Yield: 580 mg (46 %); ESI‐MS (CH3CN): [354.2268]+ (C23H32NS+), [279.9181]− (C2F6NO4S2 −); 13C NMR (CD2Cl2): δ=119.71 (q, 1 J F,C=321.3 Hz, CF3), 115.99 (CN), 99.00 (ring C), 38.94 (2×CH2), 38.54 (2×CH2), 37.85 (2×CH2), 36.49 (2×CH2), 35.95 (2×CH2), 33.93 (2×CH), 30.19 (2×CH), 26.40 (2×CH), 26.30 (2×CH), 25.18 (CH2S), 16.23 (CH 2CN) ppm.

Thiiranium salts 8, 12, according to method B

[Ad2SCH2CH2SiMe2 tBu]+SbCl6 − (8): Yield: 1.46 g (94 %); ESI‐MS (CH3OH): [443.3177]+ (C28H47SSi+), [334.7208]− (SbCl6 −); 13C NMR (CD2Cl2): δ=96.75 (ring C), 38.91 (2×CH2), 38.42 (2×CH2), 38.38 (2×CH2), 37.27 (2×CH2), 36.15 (2×CH2), 33.78 (2×CH), 30.63 (2×CH), 28.12 (CH2S), 26.62 (2×CH), 26.59 (2×CH), 26.15 (CMe 3), 16.62 (CMe3), 12.70 (CH2Si), −6.63 (MeSi) ppm.

Single crystals of 8 are prepared by slow cooling of a solution in EtCN from room temperature to −80 °C.

[Ad2SCH2CH2CN]+SbCl6 − (12): Yield: 1.13 g (82 %); ESI‐MS (CH3CN): [354.2286]+ (C23H32NS+), [334.7125]− (SbCl6 −).

Single crystals of 12 were obtained by slow addition of pentane into a solution CH2Cl2 at −20 °C until cloudiness begins to set in, and slow cooling from room temperature to −80 °C.

Seleniranium salts 14, 16, 23, 24 according to method A

[Ad2SeCH2CH2SiMe3]+Tf2N− (14): Low‐temperature NMR measurement at −40 °C: 77Se NMR (CD2Cl2): δ=35.3 ppm; 13C NMR (CD2Cl2): δ=119.08 (q, 1 J F,C=321.0 Hz, CF3), 109.66 (ring C), 39.38 (2×CH2), 39.33 (2×CH2), 39.20 (2×CH2), 38.19 (2×CH2), 37.05 (2×CH2), 33.56 (2×CH), 31.14 (2×CH), 27.55 (CH2Se), 26.63 (4×CH), 15.47 (CH2Si), −2.72 (MeSi) ppm.

[Ad2SeCH2CH2CN]+BF4 − (16): Yield: 0.8 g (82 %); ESI‐MS (CH3CN/CH2Cl2): [402.1845]+ (C23H32NSe+), [87.0002]− (BF4 −); 77Se NMR (CD2Cl2): δ=24.1 ppm; 13C NMR (CD2Cl2): δ=117.22 (CN), 111.87 (ring C), 39.97 (2×CH2), 39.71 (2×CH2), 39.48 (2×CH2), 38.09 (2×CH2), 36.66 (2×CH2), 34.28 (2×CH), 31.29 (2×CH), 26.96 (2×CH), 26.87 (2×CH), 24.15 (CH2Se), 15.78 (CH 2CN) ppm; 19F NMR (CD2Cl2)): δ=−151.04 ppm.

[Ad2SeCH2CH2SiMe2 tBu]+Tf2N− (23): Yield: 0.78 g (51 %); ESI‐MS (CH3CN): [491.2611]+ (C28H47SeSi+), [279.9217]− (C2F6NO4S2 −); 77Se NMR (CD2Cl2): δ=44.8 ppm.

[Ad2SeCH2CH2SiMe2 tBu]+CHB11Cl11 − (24): Yield: 1.32 g (65 %); ESI‐MS (CH3CN): [491.2597]+ (C28H47SeSi+), [521.7698]− (CHB11Cl11 −); 77Se NMR (CD2Cl2): δ=39.1 ppm; 13C NMR (CD2Cl2): δ=111.12 (ring C), 46.73 (CHB11Cl11 −), 39.81 (2×CH2), 39.73 (2×CH2), 39.52 (2×CH2), 38.51 (2×CH2), 36.52 (2×CH2), 33.96 (2×CH), 31.55 (2×CH), 28.22 (CH2Se), 26.79 (4×CH), 26.11 (CMe 3), 16.64 (CMe3), 12.03 (CH2Si), −6.78 (MeSi) ppm; 11B NMR (CD2Cl2): δ=−2.52, −10.14, −13.18 ppm.

Seleniranium salts 17, 25, according to method B

[Ad2SeCH2CH2CN]+SbCl6 − (17): Yield: 0.67 g (66 %); ESI‐MS (CH3CN): [402.1730]+ (C23H32NSe+), [334.7094]− (SbCl6 −); 77Se NMR (CD2Cl2): δ=27.4 ppm; 13C NMR (CD2Cl2): δ=115.01 (CN), 100.00 (ring C), 40.28 (CH2), 39.99 (CH2), 39.87 (CH2), 38.58 (CH2), 36.53 (CH2), 34.58 (CH), 31.82 (CH), 26.90 (CH), 26.81 (CH), 23.82 (CH2Se), 16.86 (CH 2CN) ppm.

Single crystals were obtained by slow addition of Et2O into a solution CH2Cl2 at −20 °C until cloudiness begins to set in, and slow cooling from −20 °C to −80 °C.

[Ad2SeCH2CH2SiMe2 tBu]+SbCl6 − (25): Yield: 1.37 g (93 %); ESI‐MS (CH3CN): [491.2635]+ (C28H47SeSi+), [334.7133]− (SbCl6 −).

Single crystals were obtained by slow addition of Et2O into a solution CH2Cl2 at −20 °C until cloudiness begins to set in, and slow cooling from −20 °C to −80 °C.

Single crystals of the chloroethyl compound 26 were obtained in the same manner.

Attempts to dealkylate the thiiranium salts, the thiiranes 27 and 30

Cleavage of the Me3SiCH2CH2 group?

A solution of Bu4NF in THF (2 mL, 1 m, 2 mmol) was added to the thiiranium salt 5 (1 mmol, 682 mg), dissolved in dry THF (5 mL) at −196 °C, followed by 2 h of stirring at −78 °C. The solvent was removed completely and the remainder was purified by chromatography in a mixture of CH2Cl2 (10 mL) and hexane (20 mL) with silica gel.

27: Yield: 210 mg (70 %).

One‐pot method

The thiiranium salt [Ad2SCH2CH2SiMe3]+BF4 − was made by reacting 1 (2 mmol, 537 mg), (Me3SiCH2CH2S)2 (4) (1 mmol, 267 mg), XeF2 (1 mmol, 169 mg), and BF3 ⋅OEt2 (2 mmol, 284 mg) in CH2Cl2 (15 mL) according to method A. Into the solution, Bu4NF in THF (4 mL, 1 m, 4 mmol) was added dropwise at −78 °C, followed by stirring for 2 h at −40 °C. The solvents were pumped off and the remainder was purified by chromatography in a mixture of CH2Cl2 (10 mL) and hexane (20 mL) with silica gel.

27: Yield: 600 mg (100 %); EI‐MS: 300 (33, M +, C20H28S), 268 (100, M +−S); 13C NMR (CDCl3): δ=71.61 (ring C), 38.72 (4×CH2), 38.53 (4×CH2), 37.91 (2×CH2), 35.06 (4×CH), 27.87 (2×CH), 27.27 (2×CH) ppm.

The thiiranium salt [C8H14SCH2CH2SiMe3]+BF4 − was prepared from cis‐cyclooctene (28; 2 mmol, 220 mg), (Me3SiCH2CH2S)2 (4) (1 mmol, 267 mg), XeF2 (1 mmol, 169 mg), and BF3 ⋅OEt2 (2 mmol, 284 mg) in CH2Cl2 (15 mL) according to method A, and reacted and worked up in the same manner as described for 27 with Bu4NF (2 mmol, 2 mL, 1 m in THF).

Thiirane 30: Yield: 190 mg (67 %); EI‐MS: 142 (74, M +, C8H14S), 109 (64, M +−SH), 67 (100); 13C NMR (CD2Cl2): δ=41.09 (ring C), 29.67 (CH2), 29.45 (CH2), 26.43 (CH2) ppm.

Thiirane 30 was also obtained from trans‐cyclooctene (29) in the same manner. Yield: 30 250 mg (44 %); 13C NMR (CD2Cl2): δ=41.11 (ring C), 29.66 (CH2), 29.45 (CH2), 26.42 (CH2) ppm.

Cleavage of the tBuMe2SiCH2CH2 group?

Thiiranium salt 9 (1 mmol, 724 mg) was reacted in dry THF (5 mL) at −78 °C with a solution of Bu4NF in THF (2 mL, 1 m, 2 mmol), followed by stirring for 2 h at this temperature. The solvent was pumped off completely in vacuum. The remainder is dissolved in CH2Cl2 (10 mL) and hexane (20 mL) and purified by chromatography over silica gel.

The 13C NMR spectrum shows the presence of alkene 1 and disulfide 4 as the main products, and thiirane 27 in less than 10 % in the product mixture.

Cleavage of the NCCH2CH2 group?

Thiiranium salt 11 (1 mmol, 635 mg), dissolved in THF (5 mL), was reacted with CsOH (2.5 mmol, 375 mg) in H2O or MeOH (3 mL) for 30 min at −40 °C, and stirred for 30 min at this temperature and 3 h at 0 °C. After removal of the solvent, the mixture was purified by chromatography.

13C NMR spectra indicate a mixture of thiirane 27 and alkene 1 in the ratio 1:1 (reaction in MeOH) and 3:1 (reaction in H2O).

Single crystals of the thiiranes 27 and 30 were obtained by slow cooling of solutions in pentane from room temperature to −80 °C.

Attempts to dealkylate seleniranium salts

The seleniranium salt [Ad2SeCH2CH2SiMe3]+Tf2N− (14) was prepared by reacting (Me3SiCH2CH2Se)2 (13; 2 mmol, 537 mg), XeF2 (1 mmol, 169 mg), and Tf2NSiMe3 (2 mmol, 707 mg) in CH2Cl2 (15 mL) according to method A. A solution of Bu4NF in THF (4 mL, 1 m, 4 mmol) was added dropwise to the reaction mixture at −78 °C, followed by stirring for 30 min at −78 °C and 2 h at −40 °C. Half of the solvent was pumped off under vacuum, some CD2Cl2 was added, and the NMR spectrum of this solution was measured at −40 °C. [Ad2SeCH2CH2SiMe3]+BF4 −, made from 1, 13, XeF2, and BF3 ⋅OEt2, were reacted with Bu4NF in the same manner.

13C NMR (CD2Cl2, at −40 °C): δ=132.86, 39.26, 36.99, 31.59, 28.36 (Ad=Ad 1); 24.64, 18.95, −2.40 (diselenide 13); 119.59 (q, 1 J F,C=321.1 Hz, Tf2N−); 57.86, 23.40, 19.46, 13.40 (Bu4N+); 67.59, 25.44 (THF); 54.16 (CH2Cl2) ppm; 77Se NMR (CD2Cl2, at −40 °C): δ=339.9 (diselenide 13) and 1187.5 (Me3SiCH2CH2SeF3 32) ppm, signal intensity 13/32 2:1.

Syntheses of the thiirenium salts 34–36

To disulfide 7 (1 mmol, 351 mg) or 4 (1 mmol, 267 mg) and alkyne 2 (2 mmol, 277 mg) or 3 (2 mmol, 589 mg) dry CH2Cl2 (15 mL) were condensed at −196 °C. At −78 °C, XeF2 (1 mmol, 169 mg) and BF3 ⋅OEt2 (2 mmol, 284 mg) or Tf2NSiMe3 (2 mmol, 707 mg) were added, and stirred at this temperature for 30 min and for 2 h at −40 °C. Half of the solvent was pumped off and after slow addition of hexane (30 mL), the salt was crystallized. The product was filtered off, washed with hexane (3×20 mL), and dried in vacuum.

[tBu2C2SCH2CH2SiMe3]+BF4 − (34): Yield: 545 mg (76 %); ESI‐MS (CH3CN/CH3OH): [271.1930]+ (C15H31SSi+), [87.0097]− (BF4 −); 13C NMR (CD2Cl2): δ=114.84 (ring C), 42.20 (CH2S), 32.96 (CMe3), 27.69 (CMe 3), 13.79 (CH2Si), −2.48 (MeSi) ppm.

[Ad2C2SCH2CH2SiMe3]+Tf2N− (35): Yield 35: 934 mg (66 %); ESI‐MS (CH3CN): [427.2832]+ (C27H43SSi+), [279.9196]− (C2F6NO4S2 −).

[Ad2C2SCH2CH2SiMe2 tBu]+Tf2N− (36): Yield 36: 1.94 g (65 %); ESI‐MS (CH3CN): [469.3356]+ (C30H49SSi+), [279.9232]− (C2F6NO4S2 −); 13C NMR (CD2Cl2): δ=119.84 (q, 1 J F,C=321.5 Hz, CF3), 113.28 (ring C), 43,47 (SCH2), 40,04 (Ad, CH2), 35.42 (Ad, CH2), 34.56 (CMe3), 27.59 (CMe 3), 26.00 (Ad, CH), 16.48 (Ad, Cq), 10.43 (CH2Si), −6.96 (MeSi) ppm; 29Si NMR (CD2Cl2): δ=10.27 ppm.

Single crystals of 36 were obtained by careful addition of Et2O and pentane to a solution in CH2Cl2 until clouding, filtration, and slow cooling from room temperature to −80 °C (CH2Cl2/Et2O/pentane 1:1:1).

Attempts to dealkylate thiirenium salts

Thiirenium salt 35 (1 mmol, 708 mg) was dissolved in dry CH2Cl2 (5 mL) and at −78 °C a solution of Bu4NF in THF (2 mL, 1 m, 2 mmol) was added dropwise. After stirring for 2 h at this temperature and 3 h at −60 °C, half of the solvent was pumped off, and some CD2Cl2 was added.

13C NMR (CD2Cl2, at −60 °C): δ=86.66, 42.76, 35.78, 28.56, 27.64 (AdC≡CAd 3); 119.38 (q, 1 J F,C=321.1 Hz, Tf2N−); 57.20, 23.12, 19.19, 13.28 (Bu4N+); 67.28, 25.23 (THF); 54.17 (CH2Cl2); −2.67 (4) ppm.

Crystal structure determinations

Single crystals were grown by slow cooling to −80 °C in appropriate solvents and transferred onto the diffractometer under cooling and exclusion of moisture: Bruker Smart CCD 1000 TU diffractometer, MoKα irradiation, scan width 0.3 deg in ω, full sphere by 2400 frames, usually 20 s per frame. After multi‐scan absorption corrections (SADABS) by equalizing symmetry‐equivalent reflections. The structures were solved and refined with the SHELX programs. All atoms except hydrogen were refined anisotropically. Hydrogen atoms were refined isotropically in positions located by difference Fourier maps or placed in pre‐calculated positions, depending on the quality of the data contain the supplementary crystallographic data for these compounds. The data can be obtained free of charge from the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data request/cif. The experimental details of all determined structures of this paper are collected in Table S2 in the Supporting Information. The structure figures have been generated with the program DIAMOND.

Conflict of interest

The authors declare no conflict of interest.

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

Supplementary

Click here for additional data file.