Molecular Enhanced Separation Concept (ESC): Removing the Functional Subunit from the Electrode by Molecular Design

: A new concept to improve the reliability of functional single molecule junctions is presented using the E -field triggered switching of Fe II bis-terpyridine complexes in a mechanically controlled break junction experiment as model system. The complexes comprise a push-pull ligand sensing the applied E -field and the resulting distortion of the Fe II ligand field is expected to trigger a spin-crossover event reflected in a sudden jump of the transport current. By molecular engineer-ing, the active centre of the complex is separated from the gold Hz, 3F, CF 3 ), –72.42 (d, 1 J P,F = 707.9 Hz, PF 6 ). HRMS (ESI-ToF): calcd. for [C 77 H 70 F 3 FeN 7 O 2 S 2 ] 2+ 650.7163, found 650.1768.


Introduction
Since the visionary claim of Gordon Moore [1] predicting the ongoing miniaturization of electronic circuits to improve their performance, alternative concepts to complement metal-oxide semiconductor (CMOS) technology moved into the focus of interest. Among others, single molecule devices are particularly appealing from a scientific perspective. Molecules are the smallest objects providing the structural diversity required enabling the programming of a particular electronic function into their structure. Furthermore, synthetic chemistry developed in the past centuries provides the skills required for the controlled assembly of tailor-made molecular structures. electrodes in order to eliminate undesired side-effects. Two aspects are considered to isolate the central metal ion, namely the spacing by introducing additional alkynes, and the steric shielding achieved by bulky isopropyl groups. With this small series of model complexes, a pronounced correlation is observed between the occurrence of bistable junctions and the extent of separation of the central metal ion, affirming the hypothesized Enhanced Separation Concept (ESC).
The concept of single molecule electronics began with the hypothesized single molecule rectifier discussed theoretically by Aviram and Ratner. [2] Since then, not only the validity of their hypothesis has been demonstrated, but numerous electronic functions like transistors, [3][4][5] diodes, [6,7] wires, [8,9] switches [10][11][12] (for an overview see ref. [13] ) and memory devices [14][15][16] (for an overview see ref. [17] ) have been realized on a single-molecule level. These experimental achievements only became possible due to the parallel engagements and developments in collaborating disciplines like synthetic chemistry, experimental physics, and theory-based modelling. While single-molecule junctions based on scanning microscopy technologies often profit from the analytical power of the integration set-up providing structural insight into the spatial arrangement of the molecular structure inside the junction, this is not the case for junctions based on mechanically controlled break junctions (MCBJ) or electrode pairs prepared by electromigration. In those set-ups the transport features of the junction is in many cases the only analytical information available. The limitation in analytical diversity has been compensated by automation enabling to assemble similar junctions repeatedly collecting enough data allowing for statistical extraction of characteristic features.
Of particular interest are molecular switches as potential memory devices. [18] These junctions display at least two different conductance states, which can be switched by an external trigger. Numerous systems have been reported being addressed by various stimuli like e.g. light, [19,20] mechanical manipulation, [11,[21][22][23][24] electrochemical gating, [25,26] the applied voltage, [27] or the electrical current. [28,29] We became recently interested in E-field triggered molecular junctions and developed a variety of model compounds sensing this parameter.
Examples range from turnstile-type macrocycles [30] over tripodal rigid 9,9′-bisspirofluorene platforms exposing a dipole moment with a terminal coordination site [31,32] to spin crossover complexes with geometrical arrangements defined by tailor-made ligands based on heterocycles. [33] The integration of these Fe II terpyridine (tpy) derivatives exposing various functional groups in molecular junctions turned out to be challenging. Attempts based on scanning tunneling microscopy (STM) experiments suffered from the limited stability of the dicationic complex, [34] and MCBJ studies displayed a rich variety in the characteristics of the junctions pointing at numerous different arrangements of the structure between both electrodes. [33] While the extensive collection of transport data combined with the systematic variation of the molecular structure enabled to spot the correlation between junctions displaying the bistability expected for an E-field triggered spin cross-over and the strength of the complexes' dipole moments, the poor control over the molecules arrangement inside the junction remained unpleasant.
Here, we report our attempts to increase the control over the complexes arrangement in the single-molecule junction by improving the design of the immobilizing and electrode bridging tpy-ligand.

Molecular Design
In Figure 1 the already reported heteroleptic Fe II tpy complexes 1-3 are displayed together with the E-field triggered switching concept. The tpy-ligand that all three compounds 1-3 have in common exposes two acetyl protected terminal thiol anchor groups. The task of this ligand is to immobilize the complex inside the junction by bridging both electrodes. The terminal thiol anchor groups are mounted in para-position with respect to the 4-phenylpyridine subunit of the tpy-ligand to promote current transport through the central metal ion. The second tpy-ligands of the heteroleptic complexes 1-3 are exposing an electron withdrawing group (EWG) on one side and an electron donating group (EDG) on the other side creating a dipole moment, which can align in an E-field.
The Fe II tpy complexes are expected to have octahedral coordination spheres and their strong crystal fields result in a considerable splitting of the energies of the 3 orbitals with t 2g symmetry (d xy , d xz , and d yz ) compared to the 2 orbitals with e g symmetry (d z 2 and d x 2 -y 2). The 6 coordinating electrons are thus following the Pauli-principle and are occupying pairwise the 3 orbitals with lower energy (d xy , d xz , and d yz ) resulting in the low-spin electron configuration (bottom of Figure 1b). By applying a voltage across the junction an E-field builds up distorting the octahedral coordination sphere. Distortion from the octahedral arrangement reduces the strength of the energy splitting between the orbitals with t 2g and e g symmetries. The more comparable in energy they become the more Hund's-rule dictates that each of the 5 orbitals must first be filled by a single electron each with parallel spin resulting in the high spin electron configuration (top of Figure 1b). [35] The spin cross-over process should be reflected in the junctions transport current as the Fe center is integrated in the current path and its highspin configuration is reported to be of increased electronic transparency. [36,37] The investigation of these junctions however turned out to be challenging and an increased occurrence of the expected bistability with the increased dipole moment in the series 3 to 1 was consistent with the hypothesized spin cross-over switching mechanism. [33] The still mediocre yield of functional junctions however remained incomprehensible. The working hypothesis was that the Fe-tpy complexes adopt a large variety of conformations inside the junction. In particular, the polar EWG and EDG groups exposing lone-pair comprising heteroatoms were suspected to compete with the thiol anchor groups and thereby increasing the diversity of molecular junctions.
To investigate the validity of this hypothesis, the active subunit exposing functional groups (EWG and EDG) competing with the intended anchor groups should be moved further away from the electrode surfaces. As sketched in Figure 2a, two molecular design parameters addressing this task have been identified: 1) the distance of the "active subunit" and the electrode surface and 2) the bulkiness of substituents sterically separating the "active subunit" from the electrode (bulky shield in Figure 2a). For simplicity, we named the approach "Enhanced Separation Concept" (ESC). To verify the ESC, the heteroleptic Fe II model complexes 4 and 5 were considered as optimized derivatives of the parent compound 2 (Figure 2b). While 4 enhances the distance between electrode and Fe-center by introducing an additional alkyne group, derivative 5 further enhances the steric shielding by additional isopropyl groups in meta-positions with respect to the thiol anchor groups.

Synthesis
The series of tpy-ligands of the 1 st generation have been assembled by engaging the dichloro-tpy precursor 9 into a Suzuki cross-coupling reaction. [33,38] To introduce an additional alkyne unit, a cross-coupling protocol for the also Pd-catalyzed Sonogashira reaction was optimized. The assembly of both target complexes is displayed in Scheme 1.
For the assembly of 4, the terpyridine compound 11 was required. Starting with commercial 4-bromobenzenethiol 6, the thiophenol was first protected by treating 6 in an excess of trimethyl(vinyl)silane with di-tert-butyl peroxide (DTBP) as radical initiator. With this variation of a reported procedure, [39] the TMS-ethyl protected thiophenol 7 was obtained in 98 % yield as colorless oil. In a Sonogashira coupling reaction, the alkyne synthon was introduced by substituting the bromine of 7 with a trimethylsilyl (TMS)-masked acetylene. The crude reaction product was liberated from its alkyne protection group by treatment with K 2 CO 3 in a mixture of methanol (MeOH) and tetrahydrofuran (THF). This adapted procedure [40] provided the alkyne 8 in 55 % yield after column chromatography (CC). 4,4′′-Dichloro-2,2′:6′,2′′-terpyridine 9 was synthesized according to our already reported procedure, [38] and its chlorine atoms were substituted by the alkyne 8 in a variation of the Sonogashira cross-coupling reaction which does not require the addition of copper salts. The latter being an interesting aspect reducing the frequently observed competing copper catalyzed oxidative acetylene coupling. The catalytic system was assembled in-situ by adaptation of a reported protocol [41] using Pd(MeCN) 2 Cl 2 as palladium source and 2-dicyclohexylphosphino-2′,4′,6′-triisopropylbiphenyl (X-Phos) as ligand in propionitrile with Cs 2 CO 3 as base. The reaction was refluxed and monitored by thin layer chromatography (TLC). Work up after 3.5 hours provided the tpy-ligand 10 in very good 93 % yield after CC. Transprotection with tetrabutylammonium fluoride (TBAF) and acetyl chloride (AcCl) in THF provided the desired tpy-ligand 11 in very good 95 % isolated yield. The push-pull tpy-ligand 12 was assembled following our already reported procedure. [33] With both ligands 11 and 12 in hands, the heteroleptic complex 4 was synthesized using a simple but effective statistical strategy. Thus, both ligands were suspended together with FeCl 2 in a mixture of dichloromethane (DCM), MeOH, and water (H 2 O). The complexes were precipitated by addition of an excess of ammoni-umhexafluorophosphate (NH 4 PF 6 ) and the desired heteroleptic complex was separated from the homoleptic ones by preparative reversed-phase HPLC giving the desired Fe II tpy 2 derivative 4 as dark purple solid in 19 % isolated yield.
The assembly of the second complex 5 was expected to be more challenging due to the bulky isopropyl groups, which might also interfere with the intended coupling steps. However, a very comparable strategy was applied. The required alkyne 17 was assembled in 4 steps from the commercially available 4bromo-2,6-diisopropylaniline 13. The order of functional group transformations on this building block turned out to be crucial as initial attempts first introducing the masked thiol substituent reduces the reactivity of the halide in para-position for the subsequent coupling reaction to a level making the further development of the building block impossible. Thus, the alkynesynthon was introduced first. The amine was transformed into an iodine in a Sandmeyer reaction following a classical protocol: Treatment of 13 with sodium nitrite in a mixture of sulfuric acid, H 2 O, and acetonitrile (MeCN) at 0°C provided the diazonium salt, which was subsequently converted into the iodine 14 by quenching with potassium iodide. Work-up followed by CC provided 14 as light brown solid in 89 % isolated yield. To introduce the alkyne-synthon, we hoped that the increased reactivity of the iodine substituent in Pd-catalyzed coupling conditions would outperform the competing sterically less buried bromine in 14. To further increase the potential steric issues, we were forced to use the triisopropylsilyl (TIPS) masked acetylene as the more compact TMS-masking group used above is not surviving the rather harsh reaction conditions required for the introduction of the tert-butyl masked thiol substituent. To our delight we found optimized Sonogashira conditions mainly addressing the iodine even though elevated temperatures were required and the substitution was slow. Using Pd(dppf )Cl 2 and CuI as catalytic system in an argon saturated mixture of diisopropylamine (DIPA) and toluene at 80°C the arylalkyne 15 was slowly formed from the iodoaryl 14 and the TIPS-acetylene. The course of the reaction was monitored by gas-chromatography (GC) and the disappearance of TIPS-acetylene was compensated by the addition of additional equivalents. After 2 weeks the reaction was worked up in spite of remaining iodoaryl 14 and the desired arylalkyne 15 was isolated in 54 % yield after CC. Also the substitution of the bromine in 15 with tert-butylmercaptane was enabled by Pd-catalyzation. The aryl bromide 15 and tert-butylmercaptane were dissolved in 1-methyl-2-pyrrolidone with Pd(OAc) 2 as Pd-source and 1,4-bis(diphenylphosphino)butane as ligand of the catalyst and K 2 CO 3 as base. After 24 h at 100°C the building block 16 was isolated in poor 39 % isolated yield after CC. The alkyne 17 was liberated quantitatively by treatment with TBAF in DCM at room temperature. For the twofold cross-coupling between the alkyne 17 and the dichloro-tpy 9 the same reaction conditions as for the sterically less demanding alkyne 8 in the assembly of 10 have been applied. Thus 9 and 2.5 equivalents of 17 were exposed to the catalyst assembled from Pd(MeCN) 2 Cl 2 and X-Phos in situ and the base Cs 2 CO 3 in refluxing propionitrile for 16 h. In spite of the elongated reaction time compared with the assembly of 10, the tpy-ligand exposing tert-butyl masked thiols 18 was iso-lated in only 65 % yield by CC reflecting the sterically more demanding situation in this case. The transprotection from the tert-butyl groups in 18 to the acetyl groups in 19 was achieved almost quantitatively by treatment with bismuth(III) trifluoromethanesulfonate and AcCl in MeCN/toluene at room temperature. The Fe II tpy 2 target compound 5 was obtained by a similar protocol as applied above for the assembly of 4. Equimolar amounts of both ligands 12 and 19 were stirred for 1 h together with FeCl 2 in a DCM/MeOH mixture. The crude was transferred to an aqueous solution and the Fe II tpy 2 complexes were precipitated by NH 4 PF 6 . The heteroleptic target complex 5 was isolated by preparative reversed-phase HPLC as dark purple solid in 45 % yield.
The identity of the Fe II tpy 2 target complexes was fully corroborated by their analytical data. For their transport analyses in a mechanically controlled break junction (MCBJ) fresh solutions were prepared as the dissolved heteroleptic complexes 4 and 5 are in an equilibrium slowly forming both homoleptic complexes. Stability experiments were performed with complex 1 and a half-life time (t 1/2 = 24 h) exceeding substantially the duration of the deposition procedure (5 minutes) was determined. [33] While comparable stability behaviors were observed for the heteroleptic complexes 2 and 4, the stability investigations of 5 revealed its considerably increased stability probably arising from the steric shielding of the metal ion. According to preliminary kinetic studies (see supporting information), the complex 5 decomposes in acetonitrile at 25°C with a half-life time of about t 1/2 = 316 h. Thereby, the heteroleptic complex not only forms an equilibrium with its homoleptic relatives, but slowly decomposes irreversibly into another yet unknown compound.
The target structures and their precursors and intermediates on the synthetic path have been characterized by 1 H-and 13 C-NMR spectroscopy and mass spectrometry.

Transport Experiments
Similarly to our previous study on complex 2, [33] single-molecule-transport measurements of compound 4 and 5 were performed in a MCBJ in vacuo at liquid helium temperature in the case of complex 4 and at liquid nitrogen temperature for complex 5. A schematic of the experiment is shown in Figure 3. The sample consists of a lithographically patterned gold wire suspended onto a flexible substrate coated with a polyimide film, which can be bent to stretch the gold wire. When the wire ruptures, two atomically sharp electrodes are formed and the distance between them can be tuned with sub-nanometer precision.
In analogy to our already reported transport studies with complex 2, [33] the MCBJ sample was decorated with the complex under investigation prior to the breaking and measuring sequence. A freshly prepared solution of either 4 or 5 (0.5 mM in acetonitrile) was deposited on the MCBJ sample and the solvent was evaporated. The measurement chamber was evacuated to 10 -7 mbar and cooled down. At low temperature, the gold wire was repeatedly broken and fused in the presence of molecules hundreds of times and current-voltage characteristics were recorded during every breaking cycle. The breaking traces were first divided into two main categories, namely the ones forming a molecular junction and the empty ones. While the first category displays current plateaus indicative of molecules bridging the gap between the electrodes, the empty junctions show an exponential decrease of the current with the displacement. The left pie chart of Figure 5 shows the fraction (blue segment) forming a molecular junction for each complex 2, 4 and 5.
At low temperature the molecular junctions are stable enough to record the IV characteristic. Based on the analysis of the IV behaviors the molecular junctions were further divided into the ones displaying bistability features (orange segment in the right pie charts of Figure 5) and the ones without. A spin crossover event in the central metal complex is expected to result in an immediate variation of the transport current, which is displayed by bistability features in the recorded IV. In  Eur. J. Org. Chem. 2019, 5334-5343 www.eurjoc.org The variety of the IV traces from different switching molecular junctions most likely reflects the diversity of both the spatial arrangements of the complex inside the junction and the atomistic realization of the electrodes' surfaces/shapes. 542, 1315 and 653 breaking cycles were measured in the case of complex 2, 4 and 5, respectively. Figure 5 shows the proportion of breaking events presenting a molecular signature. Compared to the 7 % already recorded for 2, [33] the ratio of molecular junctions increased to 37 % and 21 % with the new complexes 4 and 5. While it is tempting to correlate the increased likeliness of molecular junction formation with the expanded structure of their terpyridine ligand comprising acetylene spacers, the data set with only two additional complexes is too small for reliable conclusions. In an earlier study we recorded for five different complexes of the dimension of 2, molecular junction formation probabilities between 6 % and 25 % [33] documenting the large and to some extent random dispersion of the ratios of molecular junctions. However, in spite of the small experimental data set, the trend of larger ratios of molecular junctions with the two complexes with an immobilizing tpy-ligand enlarged by ethynyl subunits (4 and 5) supports the molecular design hypothesis that more exposed anchor groups facilitate the immobilization of the compound between both electrodes.
Even more impressive is the substantial increase in molecular junctions displaying bistability within the series 2, 4, and 5 (orange segment in Figure 6). The ratio of junctions with bistable IVs increases from 43 % for 2, over 56 % for 4, to climax in 98 % for 5. Both molecular design optimizations (the increasing of the spacing between electrode and central complex and the enlarged shielding of the complex by bulky substituents) are improving the formation of junctions comprising a functional complex. The substantial improvement by the introduction of isopropyl groups in 5 probably is a combination of several aspects. Their bulkiness results in a better defined immobilization of the complexes via the S-anchor groups and disfavors contacts between the phenyl-π-system of the immobilizing tpyligand and the electrode surface. Furthermore, their bulkiness leads to a better organization of the entire complex by filling up the empty space in-between both ligands. Thus, the more densely packed complex 5 offers less interaction surface for πgold interactions, [42] which are likely to be mainly responsible for molecular junctions without switching events. The fixation of the push-pull tpy-ligand by such interactions prevents any Figure 6. Statistics of the switching bias voltage for junctions with compound 5′.
Eur. J. Org. Chem. 2019, 5334-5343 www.eurjoc.org alteration of the coordination sphere and thus also the intended switching. Further details concerning the nature of the molecular switching was collected for the junctions with 5 by analyzing the statistics of the bias voltage at which the switching between two conductance states occurs. As displayed in Figure 6, the switching occurs more often at higher bias voltage values with a maximum around -0.9 V in the negative region. The symmetrical distribution of the events mirrored at about zero bias was expected for switching events depending exclusively on the magnitude of the electric field and not on its polarity.
The bias voltage required to initiate the switching is of hundreds of meV, which is much larger than the thermal energy at 77 K (≈ 6 meV) or even at room temperature (25 meV). Therefore, we expect that thermal effects do not play an important role in triggering the bistability of the molecule. On the other hand, junction stability is usually affected by the higher temperature, yet the higher stability of the junctions comprising 5 allowed their investigation at ≈ 77 K.

Conclusions
A new molecular design approach named "Enhanced Separation Concept" (ESC) to improve the reliability of functional molecules in single molecule junctions is presented. This revolves around the principle of reducing the interaction between the electrode and active subunit. This is achieved by increasing the separation between active subunit and the electrode surface (4, 5) and the addition of bulky substituents to shield the active subunit (5). The ESC was benchmarked on the two model complexes 4 and 5, which both comprise the same push-pull Fe II tpy ligand that can be distorted under electric field by triggering a spin-crossover transition. The occurrence of bistable single molecule junctions is analyzed by integrating the complexes in MCBJ experiments at low temperature. A substantial increase of bistable junctions is observed by separating the functional Fe II (tpy) 2 center in the series 2, 4, and 5 with 43, 56 and 98 %, respec-tively. The findings not only corroborate the hypothesized "Enhanced Separation Concept", but also demonstrate the optimization power of molecular design.
Motivated by these results we are currently developing the complexes further with foci on both, their chemical stability and their spatial arrangement inside the junction.

Experimental Section
General Procedures: All commercial available chemicals were used without further purification. Dry solvents were used as crown cap and purchased from Acros Organics and Sigma-Aldrich. NMR solvents were obtained from CIL Cambridge Isotope Laboratories, Inc. (Andover, MA, USA) or Sigma-Aldrich. All NMR experiments were performed on Bruker Avance III or III HD, two or four-channel NMR spectrometer operating at 400.13, 500.13 or 600.27 MHz proton frequency. The instruments were equipped with direct observe BBFO, indirect BBI or cryogenic four-channel QCI (H/C/N/F) 5 mm probes all with self-shielded z-gradient. The experiments were performed at 298K or 295K. All chemical shifts (δ) are reported in parts per million (ppm) relative to the used solvent and coupling constants, (J) are given in Hertz (Hz). The multiplicities are written as: s = singlet, d = doublet, t = triplet, hept = heptet, dd = doublet of doublet, m = multiplet. Gas Chromatography (GC-MS) was performed on a Shimadzu GC-MS-QP2010 SE gas chromatograph system, with a ZB-5HT inferno column (30 m × 0.25 mm × 0.25 mm), at 1 mL/min He-flow rate (split = 20:1) with a Shimadzu mass detector (EI 70 eV). Flash column chromatography (FCC) was performed with SiliaFlash® P60 from SILICYCLE with a particle size of 40-63 μm (230-400 mesh) and for TLC Silica gel 60 F 254 glass plates with a thickness of 0.25 mm from Merck were used. The detection was observed with a UV-lamp at 254 or 366 nm. Gel Permeation Chromatography (GPC) was performed on a Shimadzu Prominence System with PSS SDV preparative columns from PSS (2 columns in series: 600 mm × 20.0 mm, 5 μm particles, linear porosity "S", operating ranges: 100-100.000 g mol -1 ) using chloroform as solvent. For HPLC a Shimadzu LC-20AD and a LC-20AT HPLC was used equipped with a diode array UV/Vis detector (SPD-M10A VP from Shimadzu, λ = 200-600 nm) and a column oven Shimadzu CTO-20AC for analytical measurements. The used column was a Reprosil 100 C18, 5 μm, 250 × 16 mm; Dr. Maisch GmbH. For preparative HPLC a Shimadzu LC-20Ap and a LC-20Ap was used equipped with a diode array UV/Vis detector (SPD-20A from Shimadzu, λ = 200-600 nm). The used column was a Reprosil 100 C18, 10 μm,250 × 30 mm; Dr. Maisch GmbH. High-resolution mass spectra (HRMS) were measured as HR-ESI-ToF-MS with a Maxis 4G instrument from Bruker or as HR-EI-MS spectrometry with a DFS double-focusing (BE geometry) magnetic sector mass spectrometer (ThermoFisher Scientific, Bremen, Germany). Mass spectra were measured with electron ionization (EI) at 70 eV, solid probe inlet, a source temperature of 200°C, an acceleration voltage of 5 kV, and a resolution of 10'000. The instrument was scanned between e.g. m/z 300 und 350 at scan rate of 100-200 s/decade in the electric scan mode. Perfluorokerosene (PFK, Fluorochem, Derbyshire, UK) served for calibration.

{2-[(4-Ethynylphenyl)thio]ethyl}trimethylsilane (8):
Compound 8 was prepared in two steps by adapting a previously reported literature procedure. [40] 2-[(4-Bromophenyl)thio]ethyltrimethylsilane (7, 810 mg, 2.80 mmol, 1.00 equiv.), TMS-acetylene (58.9 mg, 310 μmol, 0.11 equiv.), Pd(PPh 3 ) 2 Cl 2 (199 mg, 280 μmol, 10 mol-%) and CuI (810 mg, 2.80 mmol, 1.00 equiv.) were charged into a Schlenk tube, dissolved in diethylamine (50 mL) and degassed for 30 min before the reaction vessel was sealed and heated to 60°C for 18 h. Subsequently the reaction was cooled down to room temperature before being quenched by the addition of water and dilution with DCM. Saturated NH 4 Cl solution (30 mL) and brine were added prior to repeated extraction of the mixture with DCM (4 × 30 mL). The combined organic layers were dried with MgSO 4 , filtered and concentrated under reduced pressure to yield the crude product which was now purified by FCC using a gradient of solvents reaching from pure n-hexane to a mixture of n-hexane/EtOAc (17:3). The column yielded 821.4 mg of a mixture of starting material and TMS protected 8. In the second step 767 mg (at least 2.5 mmol) of the described mixture were charged into a round-bottomed flask and K 2 CO 3 (873 mg, 6.25 mmol) was added. The reagents were dissolved in a 1:1 mixture of MeOH and THF (15 mL each) and the mixture was stirred for 1 h at room temperature, until the deprotection was finished (monitored by TLC). After DCM and brine were added to the reaction mixture the same was extracted with DCM (4 × 30 mL). The combined organic layers were dried over MgSO 4 and concentrated to yield the crude product which was now purified by FCC using a gradient of solvents reaching from pure n-hexane to a mixture of n-hexane/DCM (9:1). Thus the desired product {2-[(4ethynylphenyl)thio]ethyl}trimethylsilane (8) was obtained quantitatively (338 mg, 1.44 mmol) from this reaction step. Apart from this the original starting material was reisolated (283 mg, 0.98 mmol). Therefore, the desired target compound 8 was isolated in an overall yield of 55 % (362 mg, 1.54 mmol), whereas the starting material was recovered in a yield of 37 % (303 mg, 1.05 mmol). The spectroscopic data for this compound was identical to those reported in literature. [40]