Method of multi-layer near-field electrohydraulic printing and sintering of nano-silver ink prepared by liquid phase reduction

The liquid phase reduction method has become one of the hot spots in the preparation of nano-silver ink due to its advantages of simple equipment, convenient operation, low production cost, and easy industrialization.^1,2 The nanoparticle silver ink prepared by this method has been widely used in organic diodes, field effect transistors, radio frequency identification (RFID) tags, low-cost sensors, and other flexible electronic products.^3–6 The liquid reduction method needs to control the reaction temperature and precipitate silver nanoparticles out of the dispersant by dropping the reducing agent.⁷ Because the reaction temperature and the addition of the reducing agent are difficult to be controlled accurately, the particle size distribution is not uniform and easy to agglomerate.⁸ In addition, this method also has the disadvantages of the low collection rate of silver powder and high impurity content, which will cause uneven printing pattern and poor electrical conductivity after sintering, thus restricting its application and popularization in large-scale industrial production.⁹ 

Although increasing the content of silver in the conductive ink can improve the conductivity of the conductive pattern after sintering, the nanoparticles dispersed in the ink are easy to aggregate and precipitate because of the van der Waals forces and gravity, which can cause choking of the printer nozzles or non-uniform printed patterns.¹⁰ An additive dispersant (usually a polymer) is always employed to prevent or slow down the agglomeration.¹¹ However, the dispersant can cause an initially high resistance of the printed patterns even after the sintering treatment.¹² Therefore, it is difficult to prepare high-concentration nanoparticle ink by the liquid phase reduction method. In order to improve the conductivity of the printed pattern, there are two main approaches at present.^13–15 One is to add inhibitors in the reduction reaction, configure the mixed ink of multi-scaled AgNPs and silver nanorods (AgNRs), and improve the overall conductivity of the micro-conductive pattern by building a bridge between the separated nanoparticles through nanorods. Tang et al.¹⁶ prepared a low concentration multi-size silver particle (AgNPs) and silver nanorods (AgNRs) mixed ink and obtained a conductive pattern with an electrical resistivity of 14.7 Ω cm after sintering. Ye et al.¹⁷ prepared a low-concentration Ag nanowire ink and applied it in the manufacture of transparent electrodes. The other is to print multi-layer silver nano-ink. By increasing the thickness of the conductive pattern, the uneven phenomenon of the printing pattern caused by the low ink concentration can be avoided so as to improve its electrical conductivity. Cui et al.¹⁸ developed a gold nano-ink protected by two layers of overlapping polymers [polyvinyl pyrrolidone (PVP) and acrylic resin (AR)]. After 50 layers of continuous printing and sintering for 3 h at 500 °C, the conductive pattern obtained was close to the conductivity of bulk-like gold metals. Chen et al.¹⁹ used the near-field electrohydrodynamic printing method to print multiple layers of PEDOT patterns on photographic paper and investigated the relationship between the number of printed layers and the electrical conductivity. Qin et al.²⁰ studied the relationship between the number of printing layers and the electrical conductivity of Cu.

Although multi-particle ink multi-layer printing can effectively improve the electrical conductivity of conductive patterns, it still faces some problems. First of all, due to the low solid content of ink, with the increase in the thickness of the print pattern, the viscosity of the ink is not enough to support the stability of the shape of the print pattern, which will make the ink permeate around after printing, affecting the resolution of the print pattern.^21,22 In addition, due to the low content of ink particles and high printing thickness, the density of the sintered printing pattern is poor, which affects the adhesion of the conductive pattern to the substrate.

To solve this problem, the simultaneous printing and drying (SPAD) method was adopted in this study. With the polyethylene terephthalate (PET) film as the substrate, mixed ink of multi-scaled AgNPs and silver nanorods (AgNRs) prepared by the liquid reduction method was printed. The ink was printed with high precision by the near-field electrohydraulic printing method, and a heating device was installed on the collecting plate; the micropattern printing and solvent evaporation are carried out simultaneously. By controlling the drying time, the bottom pattern is solidified, and then the next layer of printing is carried out. The SPAD method makes the multi-layer printing pattern more stable and improves the resolution of the multi-layer printing pattern. Finally, the printed multi-layer micropattern was sintered by the hot/pressure sintering method to obtain the micropattern with good compactness and good conductivity. The results show that the resistivity of the micro-conductive pattern is only 1.89*10⁻⁸ Ω m under the optimal sintering parameters, which is close to the resistivity of bulk silver.

Silver nitrate (AgNO₃, produced by Sinopharm Chemical Reagent Co., Ltd., GR) was used as the source of silver. Polyvinyl pyrrolidone (PVP K-30, produced by Sinopharm Chemical Reagent Co., Ltd., GR), ethylene glycol (EG, produced by Jiangsu Qiangsheng Chemical Co., Ltd., AR), anhydrous ethanol (C₂H₆O, produced by Chinasun Specialty Products Co., Ltd., China), and glycerol (C₃H₈O₃, produced by Sigma-Aldrich China, Shanghai) were used as the solvent and the dispersant. The PET film (produced by Henan Ya’an Insulation Material Factory Co. Ltd., China) was used as the print base.

Conductive ink prepared in the laboratory was used in this experiment. The ink used silver nitrate as the silver source, ethylene glycol as the reducing agent, and PVP as dispersant. Silver nitrate and PVP were added into ethylene glycol according to the weight ratio of 1:2, after magnetic stirring for 15 min, and then the silver particles were reduced by ultrasonic heating for 15 min. The purification of the silver nanoparticles was conducted by centrifugation. They are separated from the PVP and ethylene glycol by centrifugation at 9000 rpm for 15 min. The obtained precipitates were then dissolved in ethanol and were centrifuged three times in succession. The resulting sample is dried at room temperature to form a silver nanoparticle powder. Silver powder, ethanol, ethylene glycol, and glycerol were mixed in a weight ratio of 5:10:9:1. Conductive ink of multi-scaled AgNPs and silver nanorods with a weight ratio below 25 wt. % was prepared. Its viscosity is 2 cps.

The near-field electrohydraulic printer as shown in Fig. 1(a) is used to print the configured ink. The high-voltage-power supplier is connected between the nozzle and the collecting plate, and the electrostatic field is formed between the nozzle and the substrate. The height of the nozzle and the flow of the syringe pump can be adjusted by the control system. The printing nozzle was a stainless-steel nozzle with an inner diameter of 0.33 mm and an outer diameter of 0.51 mm. When printing, the input voltage was set to 2000 V, the syringe pump flow was set to 0.2 ml/h, and the distance between the ink nozzle and substrate was set to 0.5 mm. The state of the ink droplet at the nozzle was observed with a CCD high-resolution industrial camera. After the droplet reached the ideal Taylor cone, the motion platform began to move at a speed of 40 mm/s. The temperature of the collecting plate was set to 70 °C and the length of the print line was set to 30 mm. After each printing was completed, the ink was heated and dried on the substrate for a period of time, and then the nozzle was returned to the starting point to print the next layer. The maximum number of printing layers for conductive patterns is set to four. The near-field electrohydraulic printer working principle is shown in Fig. 1(b). After printing the multi-layer micro-conductive pattern, the sample was put into a vacuum drying oven and dried for 20 min at 70 °C. The surface morphology of the dried pattern was observed by an optical microscope (DMI3000m/DFC450). The thickness of the dried pattern was measured by a profilometer (DEKTAKXT).

A 30-12H hot/pressure sintering machine as shown in Fig. 2(a) was used to sinter the printed samples in which the upper heating plate is fixed and the height of the lower heating plate can be adjusted by a jack. The temperature controller is, respectively, connected with the upper and lower heating plates. After the temperature reached the preset value, the sample was wrapped with a yellow light film and put into the heating plate. The height of the lower heating plate is raised by a jack and allowed to come into contact with the upper heating plate. The pressure value is observed and adjusted between the two heating plates through the pressure gauge, and the 30-12H hot pressing sintering machine working principle is shown in Fig. 2(b).

After sintering, a multimeter (VC9801A+) was used to test the resistance value of the sintered sample.

The low solid content would make nano-silver ink poor viscosity. In multi-layer printing, the flow of the lower layer ink can make the ink on top unstable, causing it to seep into the surrounding area. Therefore, it is necessary to heat and dry each layer of patterns after printing. In order to determine the most appropriate drying time, we set the temperature of the worktable at 70 °C and dried 0, 5, 10, 20, 30, and 40 s after the printing of each layer of pattern. After printing the pattern, the linewidth is obtained by calculating the average value of five randomly selected points. Measurement results are shown in Fig. 3; it is found that the SPAD method can effectively reduce the width of the printed pattern. With the increase in drying time, the width of the print pattern gradually decreases and tends to be stable. For the two-layer printing pattern, it takes only 5 s to produce a 97 µm print linewidth, which is close to a one-layer printing pattern (95 µm). The linewidth of the three-layer printing pattern tends to change gently when the drying time is more than 10 s and reaches 98 µm after 20 s. The linewidth of the four-layer printing pattern needs to be dried for 30 s to reach 101 µm. By comparison, it can be found that the higher the number of printed layers of the sample, the longer the drying time required. The increased thickness of the pattern would slow down the transfer of heat and increase the drying time for the newly printed pattern. The drying time for three-layer and two-layer printing patterns is much shorter than for the four-layer printing pattern.

In order to observe the effect of drying time on the surface morphology of the printing pattern, an optical microscope was used to observe the three-layer printing pattern in different drying times, and the results are shown in Fig. 4. By comparing Fig. 4(a) with Fig. 4(b), it can be found that heating and drying the substrate can effectively reduce the linewidth of the printed pattern. By comparing Fig. 4(c) with Fig. 4(d), it can be found that extending the drying time of the three-layer printing line to 20 s can make the edge of the printed pattern more even, which is similar to the edge of the one-layer printing line. This is because the drying time is prolonged, and the conductive pattern can continue to absorb heat to improve the surface temperature after curing so that the newly printed ink can dry faster, thus improving the resolution of the printed pattern. As for the high-layer print pattern, a better printing effect can be obtained by appropriately extending the drying time. In the subsequent experiments, we set the drying time of the two-layer, three-layer, and four-layer print lines as 5, 20, and 40 s.

Pressing the sample during sintering can not only strengthen the adhesion of the sintered pattern but also make the conductive pattern denser and improve its electrical conductivity. In order to explore the effect of pressure on the sintering effect, we printed several layers of the conductive pattern with a different number of layers and dried them in a vacuum drying oven at 70 °C for 20 min to make the printed patterns solidify thoroughly. Then, the samples were placed in the hot/pressure sintering machine at 110 °C to sinter for 20 min at the pressure of 6000, 9000, 12 000, 15 000, and 18 000 psi, respectively. The resistance values measured after sintering are shown in Fig. 5. The resistance of the one-layer printing pattern reached the lowest point of 3.9 Ω when the pressure is 6000 psi. As the pressure increases, the resistance of the printing pattern continues to rise. When the pressure is 18 000 psi, the one-layer printing pattern appears to be non-conductive. The resistance value of the two-layer printing pattern decreases first and then rises with the increase in the pressure. The minimum value is 2.4 Ω when the pressure is 9000 psi. The curve of the three-layer printing pattern’s resistance change is similar to that of the two-layer printing pattern, and the resistance value reaches the minimum value of 2.4 Ω at 9000 psi. The four-layer printing pattern is non-conductive at 6000 psi, and the minimum value is 1.6 Ω when the pressure is 12 000 psi. The optimal sintering pressure of the printing pattern with different layers is different, and the optimal sintering pressure increases with the increase in thickness. If the sintering pressure is too low, there will still be a certain interspace among the particles in the printed pattern, resulting in the pattern not being dense enough, which affects its electrical conductivity. If the pressure is too high, the printing pattern will be pressed into the interior of the substrate due to roughness on the surface of the pressure plate, resulting in the surface morphology becoming uneven and the effective conductive volume becoming small, which affects its electrical conductivity. Through the above experiments, the optimal sintering pressure of the one-layer, two-layer, three-layer, and four-layer printing patterns was determined to be 6000, 9000, 9000, and 12 000 psi, respectively.

In order to explore the optimal sintering temperature, we printed several conductive patterns with different layers. The sintering pressure of the one-, two-, three-, and four-layer printing patterns is set at 6000, 9000, 9000, and 12 000 psi, respectively. Sintering is done for 15 min at the temperatures of 70, 90, 110, and 130 °C, respectively, and the results are shown in Fig. 6.

With the increase in temperature, the resistance value of all kinds of printed patterns gradually decreases and tends to be stable. As the sintering temperature increases, the resistance values of all samples decrease. The trend flattens out when the temperature is above 90 °C. However, at the sintering temperature of 130 °C, the PET substrate will appear yellow as shown in Fig. 7(a). This is because the actual temperature of the sample under high pressure is greater than the set temperature of the heating plate. If the sintering temperature exceeds the limit temperature of PET material, it is easy to make the substrate deformation, affecting its mechanical properties. In order to ensure the reliability of the printed silver conductive pattern, we set the optimal sintering temperature at 90 °C.

In order to determine the best sintering time, we set the temperature to 90 °C according to the previous experiment and the sintering pressure of the one-, two-, three-, and four-layer printing patterns is set at 6000, 9000, 9000, and 12 000 psi, respectively. The samples are placed into the hot/pressure sintering machine and taken out every 3 min to measure their resistance values. The resistance values measured after sintering for 3, 6, 9, 12, and 15 min are shown in Fig. 8.

With the increase in sintering time, the resistance of the printing pattern gradually decreases and tends to be stable. When the sintering time is more than 9 min, the resistance value of all kinds of printing patterns does not change significantly. Considering the actual production efficiency, we set the optimal sintering time as 9 min.

Through the previous experiments, the optimal sintering parameters of the one-, two-, three-, and four-layer printing patterns were determined. Since the number of printing layers and the thickness of sintered conductive pattern are different, a profilometer is used to measure the thickness of sintered samples and calculate their resistivity in order to study its electrical conductivity better. Under the optimal sintering pressure, the resistivity of one-, two-, three-, and four-layer conductive patterns is 9.12 * 10⁻⁸ Ω m, 4.26 * 10⁻⁸ Ω m, 1.89 * 10⁻⁸ Ω m, and 2.51 * 10⁻⁸ Ω m, respectively. Scanning electron microscopy was used to observe the surface morphology of the sintered pattern, and the results are shown in Fig. 9.

By comparing Fig. 9(a) with Fig. 9(b), it can be found that the fewer printed layers result in fewer conductive path of the connection after sintering, which make the electrical conductivity poor. With the increase in the printing layers, the pattern becomes denser. When the number of printing layers reaches three layers as shown in Fig. 9(c), the surface porosity of the conductive pattern after sintering is only 2%. However, if the number of printing layers increases to four layers as shown in Fig. 9(d), the density will not be further improved, so the resistivity of the four-layer printed sample is similar to that of the three-layer printed pattern. Therefore, for the printed pattern with more than four layers, the electrical conductivity can only be improved by increasing the thickness of the printed pattern. However, the drying time required by four-layer printing is much higher than that of three-layer printing. Considering the actual manufacturing efficiency, three-layer printing is determined to be the most reasonable.

To demonstrate the superiority of this method, the resistivities of patterns fabricated by different printing and sintering methods, using silver nano-inks with different mass ratios, are presented in Table I.

As can be seen from Table I, patterns printed with low-concentration silver particle ink (mass fraction is below 25 wt. %) have sintered resistivity between 10.3 and 30 * 10⁻⁸ Ω m.^23–26 However, the resistivity of sintered silver ink with a mass fraction of 35 wt. % is only between 2.7 and 6.5 * 10⁻⁸ Ω m.^27,28 This means that for single-layer printing with low-concentration inks, it is difficult to obtain a highly conductive pattern only by changing printing and sintering method. The SPAD method is used for multi-layer printing without influencing the resolution of the conductive pattern, solidified by hot/pressure sintering. In addition, the silver particles in multi-layer printed conductive pattern are in close contact under high pressure, which increases the contact area, promoting the compactness among the internal particle to reduce the resistivity, which is 1.89 * 10⁻⁸ Ω m in this study, close to the resistivity of bulk silver.

In order to further study the relationship between the number of printing layers and the reliability of the conductive pattern, we sintered the printing patterns with a different number of layers according to the optimal sintering parameters determined previously.

Figure 10(a) shows the peel-off tests’ results of each layer printing pattern under the optimal sintering parameters. R₀ is the resistance of the printed silver pattern before the test, and R is the resistance of the printed silver pattern after the test. The results show that with the increase in stripping times, the resistance of the printing pattern increases and tends to be stable. After 100 times of stripping tests, the relative resistance (R/R₀) of the one-, two-, three-, and four-layer printing patterns increased to 1.65, 1.45, 1.43, and 1.32 times, respectively. After the completion of the test, the silver wire still adhered to the PET substrate and does not peel off, indicating that the silver film has strong adhesion to the substrate. This is because hot/pressure sintering can make the printing pattern denser, increase the contact area between the printing film and the substrate, and play a positive role in improving the bonding strength. It is worth noting that the higher the number of printed patterns, the more pressure they can be able to withstand during sintering, which will lead to stronger adhesion of layers to the substrate.

Figure 10(b) shows the bending test results of each layer of the printing pattern under optimal sintering parameters. The bending strain was set at 0.6% and the sample was subjected to 600 bending tests, with the resistance measured once for every 50 bending tests. For each layer of printing pattern, the resistance increases and tends to be stable with the increase in bending times. When bending times reach 600 times, the relative resistance (R/R₀) reaches 3.35, 2.36, 1.97, and 1.51, respectively. This indicates that increasing the number of printing layers can not only reduce the resistance value but also improve the mechanical flexibility of the conductive pattern effectively. As can be seen from Fig. 11, after 600 bending tests on the three-layer printing pattern, LED bulb could still be lit. This shows that its reliability is good and it can be used in the manufacture of flexible sensor parts.

Through experiments, we found that the SPAD method can effectively improve the surface morphology of the multi-layer printing pattern. The temperature of the worktable is set to 70 °C, and after each layer is printed, it is dried for an appropriate time so that the low-concentration ink with a lower viscosity can complete multi-layer high-thickness printing without causing edge diffusion. The drying time is related to the number of printing layers. The higher the number of printing layers, the longer the drying time is required. Hot/pressure sintering can effectively improve the densification of multi-layer printing silver wire. At the sintering temperature of 120 °C, sintering pressure of 9000 psi, and sintering time of 10 min, the resistivity of the sintered three-layer printing wire can reach the lowest value, which is close to the resistivity of bulk silver. Continuing to increase the number of printing layers can reduce the resistance value of the printed pattern and improve its reliability but will greatly prolong the drying time. In addition, the conductive pattern prepared by this method can still maintain good electrical conductivity after peel-off tests and bending tests. It shows that this method can be applied to the manufacture of flexible electronic equipment.

F.L.H. contributed to conceptualization; T.C.Z. and Z.H.Y. derived the methodology; T.C.Z. and Z.H.Y. performed the validation; T.C.Z., Z.H.Y., C.Y., and C.L.T. performed formal analysis; F.L.H. found resources; T.C.Z. and Z.H.Y. performed data curation; T.C.Z. and Z.H.Y. helped in writing and preparing the original draft; F.L.H. and Z.H.Y. contributed to writing—review and editing; F.L.H. supervised this study; F.L.H. helped in project administration; and F.L.H. helped in funding acquisition. All authors have read and agreed to the published version of the manuscript.

Additional financial support was provided by the Top-level Talent Project of Zhejiang Province. Support of equipment and technology provided by College of Information Science and Engineering at Jiaxing University is gratefully acknowledged.

This research was funded by the Basic Public Welfare Research Program of Zhejiang Province (Grant No. LGG20E050021) and the Science and Technology Bureau of Jiaxing City (Grant No. 2018AY11009).

The authors declare no conflicts of interest.

The data that support the findings of this study are available from the corresponding author upon reasonable request.