Molecular Probing of the Microscopic Pressure at Contact Interfaces

Obtaining insights into friction at the nanoscopic level and being able to translate these into macroscopic friction behavior in real-world systems is of paramount importance in many contexts, ranging from transportation to high-precision technology and seismology. Since friction is controlled by the local pressure at the contact it is important to be able to detect both the real contact area and the nanoscopic local pressure distribution simultaneously. In this paper, we present a method that uses planarizable molecular probes in combination with fluorescence microscopy to achieve this goal. These probes, inherently twisted in their ground states, undergo planarization under the influence of pressure, leading to bathochromic and hyperchromic shifts of their UV–vis absorption band. This allows us to map the local pressure in mechanical contact from fluorescence by exciting the emission in the long-wavelength region of the absorption band. We demonstrate a linear relationship between fluorescence intensity and (simulated) pressure at the submicron scale. This relationship enables us to experimentally depict the pressure distribution in multiasperity contacts. The method presented here offers a new way of bridging friction studies of the nanoscale model systems and practical situations for which surface roughness plays a crucial role.


Materials and Methods
Chemicals and solvents for syntheses, purification, and analyses were purchased from Sigma-Aldrich unless otherwise specified.The acceptor building block 5-bromo-4-methylthiophene-2-carbaldehyde was acquired from Fluorochem Ireland, while 2,5-dibromo-3-methylthiophene was purchased from Abcr GmbH.With the exception of ethanol, solvents were dried using 4Å molecular sieves (Carl Roth GmbH) before reactions.NMR spectra were recorded using a Bruker AV II 400 spectrometer, and subsequently analyzed with MestReNova.The infrared spectrum of molecule 1 in Figure S19 was taken using a PerkinElmer Spectrum™ 3 spectrometer.

Synthesis of Molecule 1 and 3
The synthesis of the new probe 1 was achieved by the method modified from the previous studies developed by Suga et al. and Goto et al. 1,2 In the first step, dibromo-methylthiophene was reacted with lithium disopropylamide (LDA) to generate Li-Br exchange. 3The intermediate was directly transferred to a dry reaction flask with CuCl2 under argon at -78 ˚C, whereupon it formed compound b.The compound was reduced by Zn powder; in the final step, a nucleophilic substitution of amine to bromine was performed to afford compound d.Next, we coupled the thiophene with d with Stille coupling.In the final step, the molecule was modified with a Knoevenagel condensation to afford a dicyano group or a cyanoacetic acid group.Detailed synthetic steps are described below.

Synthesis of molecule 2
To elucidate the planarization effect, we synthesized a model system utilizing the planar molecule 2. The dihedral angle between the donor and acceptor moieties was 0.4˚ as determined by DFT calculation using the PBE0 hybrid functional and the CC-pVTZ basis set and the Polarizable Continuum Solvent model for toluene with the default settings in Gaussian16.Compound 2 is related to a series of compounds that have been employed in organic photovoltaics. 4,5Details regarding the synthetic steps of the molecule can be found in Scheme S1 and the subsequent descriptions.

Synthesis of 6-(5-(di-p-tolylamino)thieno[3,2-b]thiophen-2-yl)nicotinaldehyde
To a stirred solution of N, N-di-p-tolylthieno [3,2-b]thiophen-2-amine (1.34 g, 4.00 mmol)in THF (40 mL) was added dropwise n-BuLi (1.6 M in hexane, 2.8 mL, 4.40 mmol) at -78ºC under argon and the reaction was continued for an additional 1 hr.Tributyltin chloride (1.3 mL, 4.60 mmol) was added to the solution at -78ºC, and it was brought back to ambient temperature.The mixture was extracted after 16 hrs with ether and washed with brine.The organic solution was dried over MgSO4 and concentrated to give the reactant for the coupling reaction.The reactant was added to the reaction flash with 6-bromonicotinaldehyde (0.74 g, 4.00 mmol), and bis(triphenylphosphine) palladium(II) dichloride (0.28 g, 0.40 mmol).The reaction mixture was refluxed for 16 hrs in toluene (40 mL) under Ar atmosphere.After cooling to room temperature, the solvent was removed under reduced pressure, and the excess organotin reagent was removed by reprecipitation in pentane.The crude product was purified by column chromatography with DCM, and reprecipitation from DCM/MeOH to afford the title compound as a red solid (0.96 g, 60%).

Surface Immobilization
In Scheme S2, we show the surface functionalization procedure for 1 and 2 in a manner similar to our previous studies. 6,7Clean coverslips were silanized by immersing them in a mixture of 2 mL of APTES, 80 mL of 96% EtOH, and 2 mL of 99% acetic acid, undergoing a 30 mins reaction at room temperature.The cover slips were rinsed with absolute ethanol and annealed in an oven at 1 bar, 130 °C for 24 hrs.Silanized coverslips were placed in a reaction flask with molecule 1 or 2 (20 mg, 0.044 or 0.035 mmol), 1-[bis(dimethylamino)-methylene]-1H-1,2,3triazolo[4,5b]pyridinium 3-oxid hexafluorophosphate (HATU) (46 mg, 0.12 mmol), and N,N-diisopropylethylamine (DIPEA) (30 mg, 0.23 mmol).65 mL of anhydrous dimethylformamide (DMF) was added.The immobilization was continued for 16 hours under argon at room temperature.The coverslips were cleaned with absolute ethanol with sonification to remove excess reagents and residual DMF after the immobilization.Surface homogeneity of the coverslip anchored with molecule 1 was tested by using confocal microscope as demonstrated in Figure S1.

Scheme S2.
Procedures for immobilizing molecules 1 and 2 on glass.

Contact Images and Photophysical Properties
Fluorescence lifetimes and fluorescence images were measured using a MicroTime 200 confocal microscope (PicoQuant GmbH) with an Olympus IX-71 microscope body and a 50 × 0.45 N.A. objective (SLMplan, Olympus).Various excitation wavelengths were generated with an NKT Supercontinuum Laser (SuperK Extreme Supercontinuum, NKT Photonics).Depending on the wavelength, different excitation filters were utilized: a FF409em filter (Semrock) for wavelengths between 455 and 476 nm and a bandpass (482/18, Semrock) for 479-491 nm to block unwanted light.A dichroic mirror (Z488RDC, Chroma) was used to reflect the excitation light, and the emitted/reflected light from the sample was filtered by the same dichroic mirror, a 488 nm notch filter (NF01-488U-25, Semrock), and a long pass filter (550FGL, Thorlabs).The light was then detected using time-resolved single photon counting (TCSPC) with a PDM Series detector (PicoQuant GmbH).Decay time traces were processed and fitted with SymPhoTime64 using the Tailfit method.Steady-state emission spectra were measured using an EMCCD camera (PhotonMAX, Princeton Instruments/Acton) attached to a spectrometer (Spectra Pro-150, Acton Research Instruments).The collected emission was filtered (496LP, Chroma) to exclude the excitation light.The wavelength of the spectrograph was calibrated with 3 rd order polynomial fitting using the NKT Supercontinuum Laser at 500 nm, 530 nm, 560 nm, and 600 nm with a fixed power of 1.1 µW.
The contact experiment was realized by using a rheometer eccentrically mounted with a polymer bead (polystyrene (PS) or poly(methylmethacrylate)(PMMA)) on the scanning stage of the microscope.Reflective images were first captured on coverslips without probe molecules to identify the contact zone.Afterward, coverslips were switched to samples modified with either probe 1 or 2, minimizing photobleaching under continuous intensive laser exposure.The topographies of the polymer beads were recorded prior to the contact experiments using an optical profilometer (Keyence) with the pixel size set at 277 nm/pixel.Contact topographies are displayed in Figures S3 and S6.These topographies were used in the boundary element method (BEM) simulations for contact areas using an elastic-fully plastic model.The simulated images shown in Figure 5 were rescaled with a bilinear method to match the pixel size of the experimental images.Parameters were derived from earlier studies, with hardness measured by matching experimental and simulated contact areas as in Figure S5.Input parameters for simulations are in Table S2.
Emission spectra under contact are measured with the spectrometer on the confocal microscope as demonstrated in Figure S4.For excitation spectra and contrast analyses on the confocal microscope, a 50/50 beam splitter (Thorlabs) replaced the dichroic mirror.Fluorescence intensities at various excitation wavelengths were recorded using the image scanning mode with the scanning stage constantly moving in the contact zone or free surface to prevent possible photobleaching.The data was converted into time trace files using SymPhoTime64 with the intensity binned per second.Each wavelength had a fixed acquisition time of 10 seconds, i.e. 10 measurements for a single wavelength, for statistic validity.Throughout the measurements, the laser power was fixed at 1.1 µW to minimize the photobleaching, and the power was calibrated using a power meter (PM160, Thorlabs) prior to the experiment.All images and data were analyzed by using ImageJ and Matlab. 8 assess the probe's response under hydrostatic pressure, molecule 3 was dissolved in toluene and placed in a custom pressure cell, detailed in ref [6].In summary, the device features a stainless-steel container with four sapphire windows.The sample solution, in a quartz cell with a movable Teflon piston cap, was positioned in a chamber filled with heptane.Pressure, up to ~3 kbar, was applied via a high-pressure screw piston pump.The device was integrated into a fluorescence spectrometer (SPEX Fluorolog 3-22 fluorimeter, Horiba), and emission was detected in right-angle mode.Spectral properties or molecule 3 under hydrostatic pressure are shown in Figure S8.

Figure S4
Emission spectra from immobilized molecule 1 at the free surface and at the contact interface, captured using the spectrometer on the microscope.The interface is wetted with DMSO to match the refractive index.
Figure S5 shows the simulated contact area based on the topography from Figure S3, adjusting the material's hardness (plasticity) between 50 MPa and 1000 Mpa.Comparing the experimental and simulated contact areas, we determined the hardness to be 200 Mpa.Table S2 lists the parameters for the contact simulation, sourced from a previous publication. 7 In Figure S6, we display the topography of the polymer bead used for the contact experiment shown in Figure 5. Figure S7 presents the theoretical pressure image for a perfectly smooth bead with a radius of 1.45 mm.Using these pressure images, we plotted the distribution illustrated in Figure 5D.In Figure S8, we show the absorption and emission spectra of compounds 1 and 3 in THF.For hydrostatic pressure experiment, we present excitation and emission spectra of the model compound 3 in a toluene solution, serving as a stand-in for 1 which was too poorly soluble for this experiment.Molecule 3 displays a significant red shift in its spectra compared to molecule 1.This is attributed to the stronger electron-withdrawing effect of the two cyano groups vs. a cyano and a COOH or CONR group, which further depresses the LUMO level of the molecule.Despite this difference, all three compounds can be expected to respond similarly to hydrostatic pressure.Details of the experimental setup can be found in the Materials and Methods section.Spectra were normalized based on peak intensity at 320 nm.The emission maxima shift by 12 nm (250 cm -1 ) as pressure is increased from 0.1 Mpa to 270 Mpa.In the excitation spectra, we observed a red-shift of 14 nm (670 cm -1 ) of the peak as pressure increased.
Concurrently, the excitation intensity (or absorbance) also rises.However, this shift is not as pronounced as the effects observed at contact interfaces (Figure 2B): the intensity increased by 170% at 580 nm.In Figure S9 we show the relative single point energy (  ), the excitation energy (  ), and the oscillator strength (f) as a function of the dihedral angle ( −−− ) of the twisted molecules (cf.data in Table S1).
The pixel size was 229 nm/pixel.PMMA (1.45 mm diameter) and PS (1.55 mm diameter) polymer beads (Cospheric, US) were used without further modification.The contact image of probe 2 is presented in FigureS2, showing a negative contrast where brightness diminishes upon static contact.

Figure
Figure S5 (A) Contact simulations for different hardness values.(B) Comparison of the simulated and the experimental contact areas, indicating that 200 Mpa is the hardness value that gives the best agreement.This parameter was further used in the BEM simulation for the experiment to determine the local pressure as shown in Figure 5.

Figure S6 .
Figure S6.Topography of the polymer bead used for contact experiment in Figure 5.The scale bar represents 20 µm.

Figure S7 .
Figure S7.Theoretical pressure image of a perfectly smooth, i.e. single asperity, polymer bead with a radius of 1.45 mm.The scale bar represents 20 µm.

Figure S8
Figure S8 Absorption (A) and emission (B) spectra of 1 and 3 in THF and DMSO.Excitation (C) and emission (D) spectra of molecule 3 in toluene under hydrostatic pressure from 0.1 to 270 MPa.

Figure S9 .
Figure S9.Computed values of the relative single point energy   (PBE0/CC-pVTZ), excitation energy   , and oscillator strength f (CAM-B3LYP/CC-pVTZ) as a function of dihedral angle  −−− of the three model probes.

Table S2 .
Input parameters for BEM simulation.E is the Young's modulus, ν the Poisson ratio, and H the plastic hardness of the material.