Hemilabile Ligands as Mechanosensitive Electrode Contacts for Molecular Electronics

Abstract Single‐molecule junctions that are sensitive to compression or elongation are an emerging class of nanoelectromechanical systems (NEMS). Although the molecule–electrode interface can be engineered to impart such functionality, most studies to date rely on poorly defined interactions. We focused on this issue by synthesizing molecular wires designed to have chemically defined hemilabile contacts based on (methylthio)thiophene moieties. We measured their conductance as a function of junction size and observed conductance changes of up to two orders of magnitude as junctions were compressed and stretched. Localised interactions between weakly coordinating thienyl sulfurs and the electrodes are responsible for the observed effect and allow reversible monodentate⇄bidentate contact transitions as the junction is modulated in size. We observed an up to ≈100‐fold sensitivity boost of the (methylthio)thiophene‐terminated molecular wire compared with its non‐hemilabile (methylthio)benzene counterpart and demonstrate a previously unexplored application of hemilabile ligands to molecular electronics.

Water (20 mL) was added to the mixture, and the product was extracted with CH 2 Cl 2 (3 x 20 mL). Purification by column chromatography on silica (100 % hexanes) gave the title compound as a white solid (1.10 g, 95 %). 1

Preparation of 1b
Prepared by modifying a published procedure. 5 To a solution of 1a (0.5 g, 2.42 mmol) in anhydrous DMF (18 mL), Nbromosuccinimide (0.863 g, 4.85 mmol) was added portionwise in the dark, and the reaction mixture was stirred at room temperature for 1 hour under Ar atmosphere. After this time, the suspension was poured into 10 mL of saturated aqueous sodium thiosulfate, and the product extracted with hexane (3 x 20 mL). The combined organic phases were washed with brine, dried over MgSO 4 , filtered, and the solvent was removed in vacuo. Purification by column chromatography on silica (100 % hexanes) and subsequent recrystallisation from boiling hexane gave the title compound as a white solid (0.397 g, 45 %). 1

Preparation of 1
To a solution of 1b (247 mg, 0.68 mmol) in dry THF (18 mL) under Ar atmosphere that was cooled at -78 °C, nbutyllithium in hexane (caution! 1.61 M, 0.93 mL, 1.49 mmol) was added dropwise while stirring. The resulting yellow solution was stirred for 45 minutes before the dropwise addition of dimethyl disulfide (0.13 mL, 1.49 mmol).
The mixture was then allowed to reach room temperature and it was stirred overnight. The solvent was then removed in vacuo and the crude solid was taken up in CH 2 Cl 2 (25 mL) and washed with water (2x10 ml) and brine (15 mL).
The organic layer was then dried over MgSO 4 , filtered, and the solvent was evaporated in vacuo to give an off-white solid. Purification by column chromatography on silica (100 % hexanes) and subsequent recrystallization from boiling methanol afforded the title compound as white needle-like crystals (0.095 g, 47 %

Preparation of 2a
A solution of 2,2'-bithiophene (10.0 g, 60.2 mmol) in acetic acid (50 mL) and chloroform (100 mL) was cooled to 0 °C, and bromine (caution! 16 mL, 300 mmol) was added dropwise over 1 hour. The mixture was then allowed to return to room temperature and it was stirred for 2 hours, followed by warming to 60°C for 2 hours. The mixture was then poured onto 300 mL of ice-cooled methanol and filtered. The residue was recrystallized four times from boiling ethanol to afford the title compound as off-white solid (14.2 g, 49 %). 1

Preparation of 2b
To a solution of 2a (4.22 g, 8.8 mmol) in dry THF (115 mL) under Ar atmosphere that was cooled at -78 °C, nbutyllithium in hexane (caution! 1.58 M, 11mL, 17.5 mmol) was added dropwise while stirring. The resulting yellow solution was stirred for 45 minutes and dimethyl disulfide (1.7 mL, 18.4 mmol) was added dropwise. The mixture was allowed to reach room temperature and it was stirred overnight. The solvent was then removed in vacuo, the resulting oil was picked up in CH 2 Cl 2 (50 mL) and washed with water (2 x 25 ml) and brine (25 mL). The organic layer was then dried over MgSO 4 , filtered, and the solvent evaporated in vacuo to give an orange oil that was purified by recrystallisation from 1:1 CH 2 Cl 2 :hexanes at -61 °C to give the title compound as pale-yellow solid (2.84 g, 77 %). 1

Preparation of 2
To a solution of 2b (600 mg, 1.44 mmol) in dry THF (80 mL) under Ar atmosphere that was cooled at -78 °C, nbutyllithium in hexane (caution! 1.59 M, 1.8 mL, 2.9 mmol) was added dropwise while stirring. The resulting dark yellow solution was stirred for 30 minutes and a solution of carbamoyl chloride (0.14 mL, 1.44 mmol) in THF (1 mL) S6 was then added dropwise. The mixture was left stirring at -78 °C for 2.5 hours, and after this time it was warmed at -40 °C and a saturated aqueous solution of NH 4 Cl (10 mL) was added. The resulting suspension was allowed to return to room temperature whilst stirring overnight. The organic layer was then separated, and the aqueous layer was extracted once with hexanes (20 mL). The combined organic layers were dried over MgSO 4 , filtered, and the solvent was removed in vacuo to give a dark brown oil that was purified by column chromatography on silica (2 % EtOAc in hexanes) to give a dark purple solid. CHNS microanalysis was not satisfactory, so the solid was dissolved in CH 2

Preparation of 3
To a solution of 4,4'-diiodobiphenyl (1 g, 2.46 mmol) in THF (45 mL) that was cooled at -78 °C, n-butyllithium in hexane (caution! 1.30 M, 3.8 mL, 4.92 mmol) was added dropwise. The resulting milky white suspension was stirred for 30 minutes, and then dimethyl disulfide (0.486 g, 5.16 mmol) was added dropwise at -78 °C. The solution returned clear, and it was stirred at room temperature for 3 hours. After this time, water (40 mL) was added dropwise, and the layers were separated. The aqueous layer was extracted with chloroform (3 x 30 mL), the combined organics were washed with brine (3 x 15 mL), dried with MgSO 4 and filtered. After removal of the solvent in vacuo, the resulting solid was recrystallised from boiling ethanol to give the title compound as white reflective plates (0.147 g, 25 %).

Preparation of 4
A solution of 2,2':5',2"-terthiophene (0.5 g, 2.01 mmol) in dry THF (40 mL) was cooled to -78 ºC and n-butyllithium (caution! 1.55 M, 2.6 mL, 4.00 mmol) was added dropwise whilst stirring. The yellow suspension was stirred for 30 minutes at that temperature and dimethyl disulfide (0.36 mL, 4.05 mmol) was added dropwise. The mixture was S7 allowed to return to room temperature and was then stirred for 4 hours. Water (100 mL) and CH 2 Cl 2 (30 mL) were added, the phases were separated, and the aqueous phase was extracted with CH 2 Cl 2 (3 x 25 mL). The combined organic phases were then washed with brine, dried over MgSO 4 and concentrated under vacuum. The resulting yellow crude product was then purified by column chromatography (20 % CH 2 Cl 2 in hexanes) to afford a slightly impure In a typical experiment, a fixed bias is applied to the junction (200 mV in this study), and the Au tip is driven into the sample until the conductance of the junction is >5 G 0 , and then abruptly retracted of 1 nm. The desired modulation is then applied for 100 ms, and the junction is stretched again. Thousands of consecutive traces are collected, and these are then sliced between the stretches by analyzing the signal imposed to the piezo actuator and cutting where its second derivative is above 0.1.  Figure S5: Example of piezo signal of a modulation experiment (in nm) and its second derivative, which is used to slice a single trace into individual modulation "snippets". Second derivative threshold is shown as red dashed line.

S9
The transimpedance amplifier signal is converted to current by applying the conversion factor (10 6 or 10 5 V/A, depending on the bandwidth and sensitivity needed), and then conductance is determined by Ohm's law ( ⁄ ) and divided by the quantum of conductance G 0 (≈ 77.48 µS). The individual slices are then fed into a sorting algorithm, which takes the average of the conductance of the first and last modulation and checks that both are below 0.1 G 0 and above the noise level of the preamplifier (10 -5 G 0 ). This filters out the slices where the tip is in contact or shallow interaction with the substrate, those where no molecular bridge is present, and those where the molecular bridge did not survive the whole modulation process, leaving only the slices relative to molecular junctions.

STM-BJ data
As discussed in the introduction, a large spread of conductance values is generally found in oligothiophene-based molecular wires, and we focussed our investigation on the bridged bithiophenes 1 and 2 which were found to result in sharper conductance histogram peaks. 6 However, while measurements on compound 1 indeed produces sharper histograms than measurements on 5,5'-bis(methylthio)-2,2'-bithiophene, the span of conductance values is still significantly larger than in the simple biphenyl 3 ( Figure S8).

Additional Piezo-Modulation Data
In addition to the 3 Å modulation presented in the manuscript, we recorded data with a smaller (2 Å) and a larger (4 Å) amplitude. Results are presented here, as 2D density maps. In the main paper, we presented only a 2 ms "snippet" of the 10 kHz modulation trace. Further data is presented here.  Figure S18: Histogram (100 bins per decade) for the full 50 ms high-speed modulation trace shown in Figure S16. The peak heights for the compressed and relaxed junction are different, but the integrated area of the two gaussian fittings is equivalent (49.7 and 50.1). This data is in good agreement with the histograms presented for the full low-speed dataset in Figure 3 of the main text.
We also performed experiments on 4 at 2 Å square-wave modulation. The compound shows slightly reduced modulation amplitude, but still a clear HIGH-LOW conductance switching.    Figure S20. The grey region shows the Fermi energy where the amplitude of conductance is in agreement with measured values.
A single Breit-Wigner resonance, using the formula ( ) ( ) , was then fitted to the transmission curves presented here and in the manuscript, using the maximum of the LUMO resonance (the closer to E F ) as the value of ( ), therefore parametrising the value of . An example for compound 1 can be found in Figure S22. The results are summarised in Figure S23. As discussed in the manuscript, a clear pattern of increasing as the electrodes are compressed can be found in compounds 1, 2 and 4, but no pattern was found in 3.