Supporting Information. Fabrication of Strain Gauges via Contact Printing: A Simple Route to Healthcare Sensors Based on Cross-Linked Gold

Transcription

1 Supporting Information Fabrication of Strain Gauges via Contact Printing: A Simple Route to Healthcare Sensors Based on Cross-Linked Gold Nanoparticles Bendix Ketelsen #,&, Mazlum Yesilmen #,&, Hendrik Schlicke #,, Heshmat Noei, % Chun-Hao Su, Ying-Chih Liao, and Tobias Vossmeyer*,# # Institute of Physical Chemistry, University of Hamburg, Grindelallee 117, Hamburg, Germany % DESY NanoLab, Deutsches Elektronen-Synchrotron DESY, Hamburg, Germany Department of Chemical Engineering, National Taiwan University, Taipei 10617, Taiwan & these authors contributed equally present address: Fraunhofer Center for Applied Nanotechnology CAN, Grindeallee 117, Hamburg, Germany * tobias.vossmeyer@chemie.uni-hamburg.de S-1

2 1. Four Point Flexural Bending Test Setup The setup consisted of a Nema 23 stepper motor which was connected to a threaded rod with a pitch of 1 mm. The rod was mounted on a stamp with two adjustable pins which could be pressed towards two other pins attached to a ground plate. The stepper motor was triggered by a TB6560 stepper motor drivers board which was connected to a micro controller (Elegoo, MEGA 2560 R3). A voltage of 24 V was applied to the stepper motor. A Python script executed on a PC was used to set the test parameters. The resistance measurements were done using two Keithley 2601A Sourcemeters. For performing the bending tests the GNP films were glued onto FR4 circuit board bars using Loctite 401. Before usage, each circuit board was freed from photo-resist with acetone and the underlying copper layer was removed in an etchant bath (77 vol% demineralized water, 20 vol% hydrochloric acid (37 %), 3 vol% hydrogen peroxide (30 %)). The circuit board was cut into 16 cm long and 1 to 3 cm wide bars. The central part of the circuit board bar was then roughened with abrasive paper (grain size: 380) and cleaned with acetone. The sensor probe was attached to the very center of the bar and small tabs at both ends were kept to contact the sensor via crocodile clips. The electrode gap was placed perpendicular to the long axis of the circuit board bar. A commercial strain gauge (SGT-1/350-TY11, Omega, gauge factor: 2.14) was used as reference sensor. The spacing of the top and lower pins of the bending machine was set to 4 cm and 8 cm, respectively. During a typical measurement the top pins were lowered by one millimeter, resulting in a total strain of the sensor of ~0.12 %.To ensure a smooth movement of the pins the step size of the stepper motor was set to 1/16 steps. Photographs of the setup are shown in Figure S1. S-2

3 Figure S1. Photographs of the custom-built flexural bending setup. a) Keithley 2601A source meters for resistance measurements, b) stepper motor drivers board, c) stepper motor mounted on top of the testing case, d) power source for stepper motor, e) computer for data acquisition, f) open test chamber with inserted sample and commercial sensor, both glued onto a circuit board bar, g) circuit board bar with glued GNP/PI sample. S-3

4 2. X-Ray Photoelectron Spectroscopy To confirm the exchange of the GNPs original dodecylamine ligands by 9DT cross-linker molecules during film preparation, X-ray photoelectron spectroscopy (XPS) measurements were performed on a GNP 4nm film deposited onto a silicon wafer following the LbL spin-coating procedure. Figure S2a, b and c show N 1s, Br 3d and Cl 2p regions, respectively. No signals were observed in the nitrogen N 1s core level region ( ev), confirming that the original dodecylamine ligands of the GNPs were efficiently exchanged by the 9DT cross-linker. Further, no signals were observed in the energy range around 70 ev for bromine Br 3d and around 198 ev for chlorine Cl 2p. These findings indicate that also reaction precursors (tetrachloroauric acid, tetraoctylammonium bromide) were efficiently removed during GNP synthesis, subsequent work-up, and GNP film preparation. Figure S2. XPS spectra of a) the N 1s, b) the Br 3d, and c) the Cl 2p core levels. No signals were observed for the respective elements. d) Deconvoluted XPS spectrum of sulfur S 2p. The solid red and blue lines correspond to fits of doublets (S 2p 3/2 and S 2p 1/2 ) assigned to free thiols and gold-bound sulfur. The darker gray solid line represents the envelope function. S-4

5 Figure S2d depicts the deconvoluted XPS spectrum of sulfur S 2p. In agreement with Han et al.,[1] who used the LbL spin-coating procedure in a recent study, both, gold bound (S 2p 3/2 / S 2p 1/2 doublet at ~162 ev, S-Au) as well as hydrogen-bound (S 2p 3/2 / S 2p 1/2 doublet at ~164 ev, thiol S-H) sulfur were observed. We assign the dominating fraction of free thiols in our samples to the presence of a significant amount of only partly bonded and free cross-linker molecules trapped within the final composite film. S-5

6 3. SEM Images of GNP 4nm and GNP 7nm Films Before and After Transfer Figures S3 and S4 show SEM micrographs of GNP films prepared from both particle samples, i.e., GNP 4nm and GNP 7nm. The films, which were deposited onto glass slides and then transferred onto PI foil, were imaged at 5.000x (Figure S3) and x (Figure S4) magnification. Figure S3. Low magnification images of GNP 4nm (left) and GNP 7nm (right) films before and after transfer from glass slides onto PI foil. The contamination on the GNP 4nm film observed after transfer onto PI foil may consist of PMMA residues, which were not removed completely by washing the films with acetone. S-6

7 Figure S4. High magnification images of GNP 4nm and GNP 7nm films on glass and PI. Obvious changes of the film s morphology upon transfer from glass substrates onto PI foil are not observable. S-7

8 4. Sensor Stability Figure S5 shows the data of strain-response measurements of the PI foil supported sensors. The sensors were stored under ambient conditions for ~3 weeks and measurements were performed on the dates indicated in the figure. The GNP 4nm as well as the GNP 7nm films did not show significant changes of the gauge factor (slope of curves). Hence, aging related to storing under ambient conditions for a period of ~3 weeks has no significant influence on the sensors sensitivity. Only a faint decrease in sensitivity is indicated in case of the GNP 4nm film. The gauge factors were determined as the slopes of the strain-response curves with average values of (16.1 ± 0.6) and (24.3 ± 0.3) for the GNP 4nm and GNP 7nm film, respectively. Figure S5. Resistive strain-response curves of a) the GNP 4nm and b) the GNP 7nm sensor supported on PI foil. The unprotected GNP films were stored for a period of ~3 weeks under ambient air and measurements were done on the dates indicated. S-8

9 5. Resistive Responses of PDMS Supported GNP 7nm Strain Gauges Figure S6 shows the resistive responses of a PDMS supported GNP 7nm film to tensile strain. A fairly linear resistive response is observed. Three consecutive strain-relaxation curves were recorded. For these measurements the sensor was fixed onto the PCB stripe using adhesive tape. The measurements reveal a gauge factor of ~25, nearly the same as in case of the PI-supported GNP 7nm sensor. Due to inhomogeneous deformations of the rather thick (~0.5 mm) PDMS substrate, as well as delamination problems of the tape used to fix the sensor during the bending test, the data points show broader scattering than those measured for the PI supported sensors. Figure S6. Resistive strain-response curves of a PDMS supported GNP 7nm film. [1] C. Han, S. J. Percival, B. Zhang, Langmuir 2016, 32, S-9