A Universal Nano‐capillary Based Method of Catalyst Immobilization for Liquid‐Cell Transmission Electron Microscopy

Abstract A universal nano‐capillary based method for sample deposition on the silicon nitride membrane of liquid‐cell transmission electron microscopy (LCTEM) chips is demonstrated. It is applicable to all substances which can be dispersed in a solvent and are suitable for drop casting, including catalysts, biological samples, and polymers. Most importantly, this method overcomes limitations concerning sample immobilization due to the fragility of the ultra‐thin silicon nitride membrane required for electron transmission. Thus, a straightforward way is presented to widen the research area of LCTEM to encompass any sample which can be externally deposited beforehand. Using this method, NixB nanoparticles are deposited on the μm‐scale working electrode of the LCTEM chip and in situ observation of single catalyst particles during ethanol oxidation is for the first time successfully monitored by means of TEM movies.

Nanocapillary and SECCM setup. Nanocapillaries were prepared with a laser-based micropipette puller (Sutter P-2000) using quartz capillaries of 1.2 and 0.9 mm outer and inner diameter, respectively. The pulled capillaries were filled with the Catalyst nanoparticle containing ink using a MicroFil needle (MF34G-5, World Precision Instruments). A Pt wire was inserted into the capillary and the ink-filled capillary was mounted on a specifically designed capillary holder. The working electrode on the LCTEM chip was connected via its Pt contact pad using a Cu tape and the chip was positioned on a x,y,z-piezocube (P-611.3S nano cube, PI) controlled by an analogue amplifier (E-664, PI). The piezo cube is fixed on positioning system consisting of three stepper motors (Owis) controlled with a LStep PCIe (Lang) controller. The stepper motors are used to preposition the capillary at around 10 µm above the working electrode of the chip controlled optically with the aid of a video microscope camera (DMK 21AU04, The Imaging Source). Automatic approach is started with an approach speed of 100 nm/s, while applying a voltage difference of 100 mV between the Pt wire inside the capillary and the working electrode of the chip and monitoring the current with a current amplifier (ELC-03XS, NPI). Initially only noise is measured as the electric circuit is not closed, however, immediately upon contact of the drop hanging from the opening of the capillary with the surface, a distinct increase in the measured signal is observed, which can be used to stop the approach in an automatic feedback loop. The stop criterion was usually set to the detection of a current higher than 20 pA. The capillary is then retracted manually and moved to another position above the chip for the next approach in order to increase the catalyst loading. The set-up is built on a vibration damping table (RS 2000, Newport) with S-2000 stabilizers (Newport) in a Faraday cage with thermal isolation panels (Vaku-Isotherm). The piezo cube, stepper motors and current amplifier are controlled via a FPGA card (PCIe-7852R, National Instruments) with a LabVIEW software adapted on the basis of the program obtained from the Unwin group, University of Warwick (WEC-SPM).

Determination of the reference voltage by open circuit potential measurements.
The open circuit potential of the reference electrode on the chip was measured vs a conventional Ag/AgCl/3 M KCl in a solution containing 0.15 M NaOH and 2 M ethanol using a Jaissle potentiostat/galvanostat (1002 PC.T). A value of -185 mV was obtained.

Open-cell electrochemical measurements.
After successful deposition of catalyst nanoparticles, the Pt contact pads of the chip for the working and counter electrodes were connected with flexible pins to the current amplifier. A 0.5 µL of a solution containing 0.15 M NaOH and 2 M EtOH was dropped on the chip covering all electrodes but not the connection pads. CVs were recorded from -0.2 to 1 V vs Pt with a scan rate of 50 mV/s.

This supporting information contains 3 movies recorded during the in situ observation with LCTEM.
S1: Movie of In situ TEM with normal speed. The single NiB particle was not affected by the electron beam illumination at the beginning. At around 19 th second, the voltage of chronoamperometry was applied to the working electrode and a significant morphological change could be observed.
S2. The observation of a group of particles shows the successful deposition on the glassy carbon electrode with the flowing electrolyte.