Editorial Board

Oleic acid (OA, 90%), oleylamine (OAm, 70%), tetrabutylammonium bromide (TBAB, ≥99%), toluene (99.8%), ethanol (≥99.5%), octane (99+%), isopropanol (99.5%), and (3-mercaptopropyl) trimethoxysilane (MPTS, 95%) were obtained from Sigma-Aldrich, Co. Silver nitrate (AgNO₃, 99.9+%) was purchased from Alfa Aeser, Co. Cyclohexylbenzene (98%) was obtained from Acros Organics, Co. All reagents were used without further purification.

Silver NCs were synthesized following the methods of previous studies reported in the literature with minor modifications. AgNO₃ (0.68 g), OAm (2.0 mL), and OA (18.0 mL) were combined in a 100 mL three-neck flask. The mixture was degassed at 70 ^oC for 1 h. After degassing, the three-neck flask was filled with nitrogen gas and heated to 180 ^oC at a rate of 1 ^oC/min. After the temperature reached 180 ^oC, the reaction was stopped and the mixture solution was cooled in air to room temperature. After synthesis, the mixture solution was washed using ethanol and toluene. In this work, the washed silver NCs were redispersed in cyclohexylbenzene (CHB) and octane solvents with a concentration of 100 mg mL⁻¹.

An inkjet printer (FUJIFILM, Dimatix Materials PrinterDMP-2831) was used to pattern Ag NCs. X-ray diffraction spectroscopy (D/MAX-2500 V, Rigaku) was used to confirm the ligand exchange process. Ultraviolet–visible (UV–vis) spectroscopy (Cary 5000, Agilent Technologies) and Fourier transform infrared (FTIR) spectroscopy (LabRam ARAMIS IR2) were used to characterize optical and chemical properties. A probe station (MST-4000A, MSTECH) was used to measure electrical conductivity.

In this study, silver NC thin film patterns were deposited on a substrate using an inkjet printing process and the properties of the thin films were adjusted through the ligand exchange process (Fig. 1(a)). The most important aspect of fabricating thin films via inkjet printing is the adjustment of the viscosity and surface tension of the ink, as well as the adhesion with the substrate. If the viscosity and surface tension of the ink are inappropriate, inkjet printing will not be possible, necessitating a more complex patterning process. Furthermore, if the adhesion between the thin film and ink is not sufficient, problems such as lift-off can occur during the ligand exchange process following the patterning process.

Fig. 1. 
(a) Scheme for silver NC synthesis, inkjet printing, and ligand exchange process, (b) Re-We plot for confirming the Z value, (c), (d) Scanning electron microscopy (SEM) images of CHB 95 v/v% silver NC patterns with MPTS treatment and without MPTS treatment, respectively.

Silver NCs with a size of 4 nm were synthesized using OA and OAm as surface ligands n a wet chemical method. Therefore, the surfaces of the NCs were surrounded by long organic ligands. This makes the NCs suitable for inkjet printing based on their high stability. However, dispersion in a commonly used octane solvent is inappropriate for inkjet printing based on the low viscosity of such solvents. To achieve proper control of the viscosity and surface tension of the ink, we added CHB to the silver NC ink, which has a relatively high viscosity. In this study, an Re-We plot was used to identify solvents with appropriate levels of viscosity and surface tension with Z values between 4 and 14. As a result, we determined that inkjet printing is possible under conditions with CHB > 95 v/v% (Fig. 1(b)). We adopted a 95 v/v% solvent rather than a CHB 100 v/v% solvent because incorporating an octane solvent can reduce the coffee ring effect. Using inkjet printing with the conditions defined above, silver NC thin film patterns with a 300 μm line width were successfully deposited. Fig. 1(c) presents an SEM image of a deposited film. Next, we introduced the self-assembled monolayer (SAM) formation method to improve adhesion between the substrates and thin films. The SAM formation steps are as follows. 1) Conduct UV-ozone treatment above the glass substrate to form -OH on the surface. 2) Immerse the surface of the glass in an MPTS solution for more than 12 h to exchange -OH with -SH, which improves the adhesion between the silver NCs and substrate. A non-MPTS-treated substrate cannot form normal patterns, as shown in (Fig. 1(d)). Based on the two strategies defined above, we developed a direct patterning technique.

In this study, we developed a temperature and strain sensor by applying ligand treatment to silver NC thin film patterns formed through inkjet printing to adjust their electrical, electrothermal, and electromechanical characteristics. The temperature and strain sensors that we developed are resistive-type sensors that measure temperature and strain by analyzing changes in conductivity (changes in resistance) caused by changes in temperature and strain. The silver nanoparticles used in the fabrication of temperature and strain sensors are surrounded by long organic ligands, such as OA and OAm, and these long organic ligands do not exhibit electrical properties based on the large distances between particles. In this study, the TBAB ligand [15–16] was used to adjust conduction properties and silver NCs were used as the active layers of the sensor. Accordingly, after synthesizing 4 nm silver NCs, chemical, optical, and structural analysis was performed to confirm ligand exchange (Fig. 2).

Fig. 2. 
As-synthesized TBAB-ligand-treated silver NC thin films: (a) FTIR data, (b) UV–vis data, and (c) X-ray diffraction (XRD) data.

First, FTIR spectroscopy of the as-synthesized and TBAB-treated silver NC thin films on a glass substrate was conducted (Fig. 2(a)). In the case of the as-synthesized silver NCs, CH bonds were present in OA and OAm, resulting in a peak at approximately 2900 cm⁻¹. In contrast, the TBAB-treated silver NCs exhibited no peak around 2900 cm⁻¹. This indicates that ligand exchange occurred.

Similarly, the UV–vis spectroscopy results contained a peak at 432 nm for the as-synthesized silver NCs. By applying the Scherrer equation to these results, we determined that 4 nm silver NCs were successfully synthesized. The UV–vis results of the TBAB-ligand-treated silver NC thin films exhibited a broad localized surface plasmon resistance peak at approximately 432 nm, which is a result of sintering during the ligand exchange process. This is because during the process of TBAB treatment, the long organic ligands are transformed into short TBAB ligands and physical contact between the silver NCs occurs, resulting in sintering. The differences between the as-synthesized and TBAB-treated data confirm that ligand exchange occurred.

Finally, XRD spectroscopy revealed a diffraction peak at 2θ = 39.1° corresponding to the {111} plane for the as-synthesized silver NCs. The TBAB-treated silver NCs exhibited strong diffraction peaks at 2θ = 31.3°, 44.6°, and 38.2°, corresponding to the {200} and {220} planes of AgBr and {111} plane of Ag, respectively. This implies that ligand exchange and sintering were successfully occurred. After sintering, the average size of 32.84 nm for the TBAB-treated Ag NCs was calculated by using Scherrer's equation. Therefore, our chemical, optical, and structural analyses confirmed that 4 nm silver NCs were synthesized and that the TBAB ligand exchange process successfully occurred.

The electrical and electrothermal properties of the TBAB-ligand-treated silver NC thin film patterned temperature sensors fabricated through inkjet printing were investigated. The performance of the fabricated temperature sensors can be determined based on the temperature-resistance coefficient (TCR), which can be expressed as

where ΔR, R₀, and ΔT are the resistance change, initial resistance, and temperature change, respectively.

First, the electrical properties of the TBAB-treated silver NC thin film sensors were measured at 303 K and 323 K from 0 V to 10 V to confirm the resistance change (Fig. 3(a)). The results confirmed that the resistance was greater at 323 K than at 303 K and the resistance varied with temperature. Next, we measured the resistance changes in the temperature sensors at intervals of 5 K over the range from 303 K to 323 K to determine the relative resistance change (Fig. 3(b)). The relative resistance changes at 308, 313, 318, and 323 K were 0.01792, 0.03777, 0.05601, and 0.07648, respectively, confirming that the change in resistance was linear with the change in temperature. Finally, cycle tests were conducted at 303 K and 323 K to confirm the stability of the temperature sensors (Fig. 3(c)). The change in relative resistance based on variations in temperature was measured, and stable temperature-resistance properties were confirmed based on a constant resistance at each temperature.

Fig. 3. 
Temperature sensors with TBAB-treated silver NC thin films. (a) Electrical properties at 303 K and 323 K, (b) Resistance-temperature curve measured at 5 K intervals between 303 K and 323 K under 10 V, (c) Cycle testing data from 303 K to 323 K under 10 V.

Through the evaluation of electrothermal properties, the performance of the resistive-type temperature sensor fabricated using inkjet printing was confirmed. The TCR of the temperature sensor was 3.82 × 10⁻³ K⁻¹, and we confirmed that the resistance-type temperature sensor was successfully fabricated through the inkjet printing process.

To confirm the performance of the fabricated temperature sensors, measurements of the temperature change of a fan heater were conducted. The temperature sensor was placed in front of a fan heater equipped with heating/cooling functionality, and the change in a relative resistance was measured when activating the heating/cooling mode (Fig. 4(a)). The temperature change of the fan heater was measured using an IR detector, which measured the temperatures of the cooling and heating modes to be 20.9 ^oC and 130.6 ^oC, respectively. The temperature calculated based on the previously measured TCR (3.82 × 10⁻³ K⁻¹) and resistance changes was 134.2 ^oC, confirming that the temperature can be successfully measured by the proposed sensor without significant error. Therefore, we confirmed that the development of low-cost simple temperature sensors is feasible based on the inkjet printing process.

Fig. 4. 
Analysis using a fan heater. (a) Demo of the temperature sensor at 293 K and 403 K, (b) Comparison of measurement performances of the fabricated temperature sensors and an IR detector.

The electrical and electromechanical properties of TBAB-treated silver NC thin film strain sensors fabricated through inkjet printing were investigated. The performance of a strain sensor can be determined based on its gauge factor (GF), which is expressed as

where ΔR, R₀, and ε are the resistance change, initial resistance, and strain, respectively.

An I-V curve was measured under the conditions of ε = 0% and 0.2% to confirm the resistance change of the strain sensors (Fig. 5(a)). The results indicate that the resistance was higher at 0.2% strain compared to that at 0% strain and that the resistance changed with the change in strain. Next, a bending test was conducted under strains of 0% and 0.2% to investigate the stability of the sensors (Fig. 5(b)). The results indicate that the resistance changed as a result of changes in strain, confirming that the sensors achieved stable electromechanical properties by showing a certain level of resistance based on the strain conditions.

Fig. 5. 
TBAB-treated silver NC thin films on a 250 μm polyethylene terephthalate substrate strain sensor. (a) Electrical properties at 0% and 0.2% strain from 0 V to 10, (b) Bending test data at 0% to 0.2% strain at 10 V.

These measurements of the electromechanical characteristics show that the strain sensors fabricated by inkjet printing operate normally at low strain such as at 0.2%. The GF was calculated to be 30.6, which is sufficient for use as a strain sensor. Therefore, we confirm that the fabrication of temperature and strain-detectable multi-sensors was successful through the inkjet printing process.