Coupled Molecular Switching Processes in Ordered Mono- and Multilayers of Stimulus-Responsive Rotaxanes on Gold Surfaces

Interfaces provide the structural basis for function as, for example, encountered in nature in the membrane-embedded photosystem or in technology in solar cells. Synthetic functional multilayers of molecules cooperating in a coupled manner can be fabricated on surfaces through layer-by-layer self-assembly. Ordered arrays of stimulus-responsive rotaxanes undergoing well-controlled axle shuttling are excellent candidates for coupled mechanical motion. Such stimulus-responsive surfaces may help integrate synthetic molecular machines in larger systems exhibiting even macroscopic effects or generating mechanical work from chemical energy through cooperative action. The present work demonstrates the successful deposition of ordered mono- and multilayers of chemically switchable rotaxanes on gold surfaces. Rotaxane mono- and multilayers are shown to reversibly switch in a coupled manner between two ordered states as revealed by linear dichroism effects in angle-resolved NEXAFS spectra. Such a concerted switching process is observed only when the surfaces are well packed, while less densely packed surfaces lacking lateral order do not exhibit such effects.


General Methods
Synthetic reactions were conducted under a dry argon atmosphere. Dry solvents were purchased from ACROS Organics and used as received.
Diethylether (Et 2 O), hexane and ethylacetate were purchased from VWR and destilled prior to use by common laboratory methods. Ethanol (EtOH), dichloromethane (DCM), dimethylformamide (DMF), and acetonitrile (ACN) used for surface experiments were purchased from Carl Roth or VWR in HPLC grade and used as received. Silica gel 60M (0.04-0.063 mm, Macherey-Nagel) was used for column chromatography.
Dialysis was performed using cellulose ester tubes (MWCO 1000 and 2000 Da; Spectrumlabs) from Roth. All gold substrates were purchased from Georg Albert PVD and stored under argon prior to use. Gold substrates used for XPS and NEXAFS were prepared onto polished singlecrystal Si(100) wafers which have been coated with a 9 nm titanium adhesion layer and 30 nm gold (rms-roughness < 0,5nm). Semitransparent Au substrates (20 nm) used for transmission-UV/Vis-spectroscopy were prepared on borosilicate glass with a 1 nm titanium adhesion layer (. All surface experiments were performed in gamma-sterilized tubes (Orange Scientific).
Multilayers were prepared on pyridine-terminated self-assembled monolayers (PST) as template layer. As metal sources, tetrakis(acetonitrile)palladium(II) tetrafluoroborate and iron(II) tetrafluoroborate hexahydrate were used, respectively. For iron(II) deposition, the samples S1 were immersed in a 1 mM solution of the metal salt in ethanol for 30 min at r.t. For palladium(II) deposition, the samples were immersed in a 1 mM acetonitrile solution of the metal salt for 10 min at r.t. Deposition of the rotaxane took place by immersing the samples in a 1 mM solution of Rot6 in dichloromethane for 24 h at r.t. For multilayer construction, both steps were alternatingly repeated until the desired layer number was reached. Chloride addition was obtained by immersing the samples in a 1 mM solution of tetrabutyl ammonium chloride in dichloromethane for 2 h at r.t. Removal of the chloride ion was performed by immersing the sample in a 1 mM solution of sodium tetrakis [3,5-bis(trifluoromethyl)phenyl]borate in dichloromethane for 2 h at r.t. After each deposition step, the samples were immersed in the corresponding solvent (ethanol, DMF or dichloromethane) for another 10 min, dried vigorously in a stream of argon and stored under argon before characterization. For deposition of the terpyridine-terminated monolayer consisting of one part 12-(2,2′:6′,2″-terpyridin-4′-yl)dodecane-1-thiol (TDT) and three parts decanethiol (DT) as well as the azide-terminated SAM 1,2-bis(11-azidoundecyl)disulfide (AUD) the gold surfaces were immersed for 24 h in a 1mM ethanol solution of these molecules after cleaning with HCl. The deposition of Rot3 was performed in DCM (1 mM) with 1 mol-% of the Cu(I)-catalyst Cat.
Microcontact printing (µCP) was performed using stamps patterned with dots (diameter = 10 µm; spacing = 5 µm) which were produced from polydimethylsiloxane and the Sylgard 184 curing agent (Dow Corning) by casting them against a silicon master and curing for 16 h at 60 °C.
The master was produced photolithographically under clean-room conditions at MESA+. Previous to inking and µCP, the corresponding stamps and surfaces have been rinsed vigorously using MiliQ-water and EtOH for 30 s each. Inking of the PDMS stamps was carried out using a 1 mM solution of octadecanethiol ODT in ethanol for 15 min followed by drying in an argon stream. Transfer of the pattern to the surfaces was done manually by contacting the stamp and the gold substrate for 10 min. Afterwards, the substrates were rinsed with MiliQwater and EtOH for 30 s each. Backfilling of the spacings between the dots was carried out by immersing the patterned substrates into 1 mM solutions of PST in DMF. 1 Exact masses were measured on either an Agilent 6210 ESI-ToF mass spectrometer or an ESI-FTICR Ionspec QFT-7, Varian Inc. instrument.

Instrumentation and Data
Transmission UV/Vis spectra were recorded on a Varian Cary 50 UV/Vis spectrophotometer. A spectrum of the underlying SAM was used as background and subtracted from all multilayer spectra.
XPS measurements were carried out with an AXIS Ultra DLD electron spectrometer manufactured by Kratos Analytical, UK. XP spectra were recorded using monochromated Al K α excitation at a pass energy of 40 eV for all detail spectra and 80 eV for the survey spectrum. The source-to-analyser angle was 60°. Emission angles of 0° and 60° with respect to the surface normal were used. The binding energy scales of XP spectra were corrected for charging using an electron binding energy of 83.96 eV for the Au 4f 7/2 level of the gold substrate. 2 XP spectra were analyzed with Unifit 2013 fitting software (Unifit Scientific Software GmbH, Leipzig, Germany) and all peak fits were performed with a Lorentzian-Gaussian sum function peak-shape model. The FWHM values in the N 1s and C 1s spectra were constrained to be equal for each component per spectrum. Peak fits and integrated peak areas were obtained after subtraction of Shirley backgrounds (Au 4f 7/2 , C 1s, F 1s Fe 2p 3/2 and Ni 2p 3/2 ). In case of N 1s, this approach was not applicable because that low intensity peak is superimposed by the intense loss S2 structure of the Au 4d photoemission doublet. The application of a Shirley background requires a higher count rate at the upper binding energy limit of the energy window used for background determination, but in the given case (due to superposition of Au 4d) we have a lower count rate at the upper binding energy limit. As a workaround, a linear background was used in this case. The uncertainty of measurement stays at an acceptable level. AFM analysis was performed using a Multimode Nanoscope V (Bruker, Nanoscope 8.10) in tapping mode under ambient conditions. The data obtained from the patterned substrates was analysed using the open-source Gwyddion software. 3 ToF-SIMS imaging was performed on a reflectron-type ToF-SIMS IV instrument (ION-TOF, Münster, Germany) equipped with a 25 keV bismuth liquid metal ion gun (LMIG) as the primary ion source mounted at 45° with respect to the sample surface. The LMIG was operated at 0.5 μA emission current using the novel "collimated burst alignment" (CBA) mode optimized to achieve high lateral (< 200 nm) as well as a high mass resolution (R ~ 5000). 4  NEXAFS 5 spectra were measured at the HESGM CRG dipole magnet beam line at the synchrotron radiation source BESSY II (Berlin, Germany). The spectra were acquired in the partial electron yield (PEY) mode using a channel plate detector with a retarding voltage of -150 V and incident angles of linearly polarized synchrotron light of 30° (electric field vector upright to surface plane) and 90° (electric field vector parallel to the surface plane). The resolution E/ΔE of the monochromator at the carbonyl π* resonance of CO (hν = 287.4 eV) was in the order of 2500. Raw spectra were divided by ring current and monochromator transmission function. The latter was obtained with a freshly sputtered Au sample. Energy alignment of the energy scales was achieved by using an I 0 feature referenced to a C 1s  π* resonance measured with a fresh surface of highly ordered pyrolytic graphite (HOPG; Advanced Ceramic, Cleveland, USA) at 285.4 eV. 6

Preparation and Characterization of New Compounds
The synthesis of diterpyridine rotaxane Rot6 was realized by a modified literature known synthesis of Rot5 followed by a Suzuki crosscoupling. The precursors were synthesized according to literature-known procedures. 7  (dp-axle), 4H), 2.26 (s, Cy, 8H), 2.03 (s, CH 3 , 24 H), 1.58 (bs, Cy, 12H) ppm. 13    used to determine the exchange rate between the two triazol-stations upon switching after chloride addition, it has been shown that the shuttling process as well as the rotation of the axle are both fast processes at least on the NMR timescale, as we could not reach the coalescence temperature of both processes at 223 K (which is the lowest temperature, we can get with our instrument) the rates could not be determined. S15

Additional Surface-Characterization Data
Fig. S16: SR-XP survey spectrum of AUD-Rot3 (top) and AUD (bottom) (excitation energy: 700 eV). One can easily see that the C 1s as well as the O 1s signal increase after click-reaction. The C 1s increase is clearly due to the deposition of the rotaxane. In contrast, the increase of the O 1s signal is rather surprising. One reasonable explanation is the following: Not all azides are used to click Rot3 to the AUD surface as the size of the rotaxane is certainly significantly larger than that of an azidoalkylthiol. Consequently, some azides should be left. According to the XPS data, this is however not the case and we conclude that remaining azides decomposed by nitrogen loss. According to earlier data, 11 this decomposition process leads to the formation of imines by 1,2-H shifts within the nitrene intermediate. In the presence of traces of water, imines are prone to hydrolysis yielding aldehydes thus rationalizing the increased signal for oxygen in the XPS spectra.