Jingyuan
Zhang
ab,
Xiaodong
Li
*ab,
Mu
Zhang
ab,
Qi
Zhu
ab and
Xudong
Sun
ab
aKey Laboratory for Anisotropy and Texture of Materials (Ministry of Education), Northeastern University, Shenyang, Liaoning 110819, China. E-mail: xdli@mail.neu.edu.cn
bResearch Center for Advanced Ceramic Materials, School of Materials Science and Engineering, Northeastern University, Shenyang, Liaoning 110819, China
First published on 11th November 2021
Herein, we present a clean, green and economical method to prepare flexible and ultrathin silver films for electromagnetic interference (EMI) shielding by thermal decomposition of metal–organic decomposition (MOD) inks. The non-particulate ink synthesized by complexing silver oxalate with 1,2-diaminopropane in ethanol solvent can be easily decomposed to pure silver at a quite low temperature in air. The efficient electrical transport network constructed by the interconnected nanoparticles of the silver film results in outstanding electrical properties and shielding performance, with an absorption-dominated mechanism. By curing the ink at 130 °C for 30 min, a more effective network structure for electron transport can be obtained. By the spin coating deposition method, the silver film with only a 200 nm thickness has an ultrahigh conductivity of 8.46 × 106 S m−1 and excellent shielding effectiveness of 51.1 dB at 10.3 GHz in the X band, which is the highest shielding effectiveness of the film with such thin thickness so far. The factors affecting the EMI shielding performance of silver films were elucidated, including the heat treatment process, ink concentration and spin coating times, and the relationship between the microstructure and the properties of the film was established. This work indicates that silver metal–organic decomposition inks will be a favorable choice for the industrial production of EMI shielding applications.
Recently, Jia et al. dispersed Ag flakes into water with waterborne polyurethane (WPU) and a fluorocarbon surfactant to obtain a conductive ink.18 The obtained Ag/WPU coating with a thickness of 10 μm had high conductivity (14300 S cm−1) and shielding effectiveness (74.5 dB), demonstrating that conductive inks have many advantages in electromagnetic shielding applications. Generally, conductive inks can be applied to additive manufacturing for various printing and coating technologies to achieve large-scale production.19–21 However, due to the poor stability of conductive particle inks, surfactants are needed to disperse the particles into the solvent, which requires a high thermal temperature to remove.22,23 Furthermore, for inkjet printing, particle inks can result in a clogged nozzle easily. Therefore, it is urgent to develop a stable and particle free conductive ink suitable for electromagnetic shielding materials and optimize its preparation process to understand the relationship between its microstructure and properties. The metal–organic decomposition (MOD) ink of silver carboxylates and amine in an ethanol solvent was stable in air for several months.24,25 Pure-phase silver films can be obtained via the thermal decomposition of this non-particulate ink at a quite low temperature in air, which is a clean and energy-saving method with high utilization of silver. The silver film thermally treated at 150 °C for 30 min has low resistivity (8.6 μΩ cm),24 demonstrating that it may be a potential candidate for the next generation of electromagnetic shielding materials.
In this work, we introduced a silver MOD ink to prepare flexible silver films of ultra-thin thickness for electromagnetic shielding applications. The silver MOD inks in this work were synthesized by complexing silver oxalate powder with 1,2-diaminopropane. The wet films deposited by the spin coating method were cured at different temperatures and times. The effects of the heat treatment process on the microstructure, conductivity and electromagnetic shielding properties of the films were analyzed. The electromagnetic shielding mechanism of the nanoparticle silver film was also explained. In addition, the thickness of the film was altered by changing the ink concentration and spin coating times, and their effects on the microstructure, electrical properties and electromagnetic shielding performance of the film were explored. Finally, the relationships between the microstructure, electrical properties and electromagnetic shielding properties of silver films were established.
The 3 cm × 3 cm PET films used as the substrates were cleaned with ethanol and deionized water, followed by O2 plasma treatment to achieve hydrophilic surfaces.26 The prepared MOD ink was spin-coated on the pretreated PET substrate at 3000 rpm for 20 s to obtain wet films. Then, the wet films were thermally treated on a hotplate at different temperatures and times. For films with multilayers, spin-coating and thermal treatment were repeated several times and the repeated times corresponded to the film layers.
The EMI shielding performance was analyzed using a vector network analyzer (VNA, Keysight N5222A, USA) in the frequency range of 8.2–12.4 GHz (X band). Before testing, the films were cut to suit the waveguide sample holder. After the measurement, the S parameters (S11, S12, S21, and S22) were collected to calculate the power coefficients (R, A, and T) and shielding effectiveness (SER, SEA, and SET) according to the following equations.27,28
R = |S11|2 = |S22|2 | (1) |
T = |S12|2 = |S21|2 | (2) |
A = 1 − R − T | (3) |
SER = −10log(1 − R) | (4) |
SEA = −10log(T/1 − R) | (5) |
SET = SER + SEA (when SER ≥ 10 dB) | (6) |
Fig. 1 (a) Image of silver oxalate-1,2-diaminopropane ink, (b) image of the silver film, and (c) the TG-DSC curve of the ink. |
The SEM images (Fig. 2a1–a6) show that the film is composed of interconnected nanoscale Ag particles with some pores among them. The ink all turned into pure silver in the whole temperature range investigated, as seen from the XRD patterns (Fig. 2b) of the six fabricated films. The intensity of the peaks gradually increased with the increase of the curing temperature, corresponding to the increased crystallite size and the growing amount of decomposed silver in the film. The nanoparticles of silver are less than 100 nm, hence, the grain size of Ag films can be calculated by the Scherrer formula according to the XRD data, as shown in Fig. 2c. The grain size gradually increases with the temperature from 16 nm at 90 °C to 25 nm at 150 °C, afterwards becoming stable at 25 nm. It can be concluded that the internal microstructure of the film is a nanonetwork structure composed of more than 20 nanometer sized silver particles with a large number of pores. Although the porosity of the film is very small, nanometer-sized pores are dispersed among the nanoparticles, resulting in a large number of interfaces among the particles in the film. To further explore the residual organic matter in the films, FT-IR analysis was performed on the above films, as shown in Fig. 2d. In the FTIR spectrum of the ink, the broad and strong band at 3700–3020 cm−1 is originated from the O–H bond of ethanol and the amine group of the complexing agent 1,2-diaminopropane.24 The absorption band at 2980–2860 cm−1 corresponds to the C–H stretching vibration of the organic solvent.24 The peaks between 1760 cm−1 and 1240 cm−1 are attributed to the stretching vibration of the amine group and carbonyl in oxalate.33,34 After heat treatment at 90 °C for 30 min, the broad FTIR spectrum of the product turned into a strong and sharp peak at 3306 cm−1, which is the vibration peak of the amine group,24 indicating that the ethanol solvent had volatilized and 1,2-diaminopropane remained in the decomposition product of the ink. A large number of sharp peaks appear at 1760–1240 cm−1, also indicating that 1,2-diaminopropane and oxalate remained and were wrapped on the surface of the silver nanoparticles. The small peak at 2210 cm−1, corresponding to carbon oxide,35 is likely due to oxalate partially decomposing into CO2 and adsorbing on the decomposition products. As the temperature was increased to 100 °C, the peak at 3306 cm−1 disappeared, indicating that 1,2-propanediamine had volatilized. In the local expansion of the FTIR spectrum (shown as Fig. S2, ESI†), two weak peaks are found at 1383 cm−1 and 2168 cm−1, corresponding to the symmetric stretching vibration of carboxylate36 and CO adsorbed on the product surface, respectively. With the increase in temperature, the intensity of the two remaining peaks decreased gradually. When the temperature was increased to 170 °C, no peak was found. Therefore, the FTIR results, being consistent with the TG-DSC analysis, suggest that the residual organic matter decomposes more thoroughly with the increase of curing temperature and there is almost no organic matter in the film at 170 °C.
The electrical conductivity of the films cured at different temperatures is shown in Fig. 3a. The conductivity increases with the increase of curing temperature, attributed to more contact of the particles and less porosity of the films, which is a result of the decomposition degree of the ink. The formed nanoparticles at 90 °C are wrapped with the undecomposed ink, which blocks the conductive path. On the other hand, the film cured at 170 °C completely decomposed with the largest grain size and the smallest porosity, forming the most effective electron transport network structure. Therefore, it has the largest conductivity of up to 8.76 × 106 S m−1, which is close to the conductivity of bulk silver (6.3 × 107 S m−1). The same analysis methods were applied to the films cured for different times ranging from 1 min to 60 min and similar results (Fig. S3a–d, ESI†) can be seen with extension of the curing time, the grain size of the silver films is enhanced, while the porosity is reduced, and thereby, the conductivity is improved.
Fig. 3b displays the EMI shielding effectiveness (SE) of the films cured at different temperatures measured in the frequency range of 8.2–12.4 GHz. As we can see, the electromagnetic shielding effectiveness of the film increases with the increase of curing temperature, which is similar to the changing trend of the conductivity. The film treated at 90 °C has an average effectiveness of only 0.45 dB mainly due to its poor conductivity, and the film treated at 100 °C shows a low conductivity of 2.02 × 105 S m−1, hence, an average SE of 18.4 dB was observed. The films treated above 110 °C exhibited an outstanding EMI shielding performance, attributed to their ultra-high conductivity. The shielding effectiveness is slightly improved between 130 °C and 170 °C, reaching 49.68 dB at 130 °C and 55.25 dB at 170 °C, respectively. Similarly, with the increase of curing time, the conductivity increases from 1.73 × 106 S m−1 to 4.81 × 106 S m−1, and the corresponding electromagnetic shielding effectiveness increases from 42.93 dB to 50.94 dB, as shown in Fig. S3(d) and (f) (ESI†). The improvement of SE tends to be trivial, when the curing time is longer than 30 min. In consideration of the energy-saving synthesis of silver films with high shielding performance, the appropriate curing process is set as 130 °C for 30 min. The low curing temperatures make it possible that our ink can be applied to various flexible substrates. Many printing and coating technologies can be adopted for this ink to deposit the silver film or patterns, which makes the fabrication process easier and more economical. In this experiment, we select the spin coating method to fabricate more uniform films. By this fabrication method, the silver film possessed a high shielding effectiveness of 49.68 dB with a thickness of 400 nm.
To clarify the outstanding EMI shielding performance of the ultra-thin silver films in this work, a theoretical analysis was carried out according to eqn (7) for electrically thin materials:37,38
(7) |
δ = (πfμσ)−1/2 |
It is well known that shielding materials weaken the penetration ability of electromagnetic waves by means of reflection and absorption. The interaction mechanism between the incident electromagnetic wave and the shielding material is illustrated in Fig. 3e. The power coefficients R, A and T (shown as Fig. 3e) represent the ratio of the power of reflected, absorbed and transmitted electromagnetic waves to the incident electromagnetic waves, respectively, calculated from the S parameters using eqn (1)–(3). Fig. 3c shows the power coefficients R, A and T of silver films cured at different temperatures. R values are always greater than the corresponding A values, indicating that the incident electromagnetic waves on the silver film are mainly reflected off. This is determined by the nature of the shielding material itself due to the big impedance mismatch between silver and air.39 As we can see from Fig. 3e, the absorption of electromagnetic waves occurs after the surface reflection. Due to the large R value of silver films, most of the electromagnetic waves are reflected off and only a small part is incident with the silver film. Therefore, the reflection and absorption capability of silver films to electromagnetic waves cannot be judged according to the power coefficients.40 The EMI shielding effectiveness is used to characterize the ability of shielding materials to attenuate electromagnetic waves. The total SE (SET) is the sum of reflection shielding effectiveness (SER), absorption shielding effectiveness (SEA) and multiple reflection shielding effectiveness (SEM). As SEA is higher than 10 dB, SEM can be ignored.41,42 SER, SEA and SET are calculated from the power coefficient according to eqn (4)–(6). The contribution of SER and SEA to the total SE is usually used to elucidate the shielding mechanism.40,43,44 As observed in Fig. 3d, the average SER and SEA both increase initially with the increase of temperature. Although the values of SER become stable afterwards, the values of SEA always increase with the SET, indicating that SER contributes less than SEA. Furthermore, the value of SEA is always higher than the SER at any temperature. For example, the SER and SEA of the silver film heat treated at 130 °C for 30 min are 38.3 dB and 11.25 dB, accounting for 77% and 23% of the total SE (49.68 dB), respectively. Accordingly, the shielding mechanism of the silver films is mainly dominated by absorption. The network structure providing a large number of interfaces among the nanoparticles may be the reason for the absorption-dominated shielding mechanism. The incident electromagnetic waves are reflected and scattered in the pores formed among the nanoparticles, resulting in an effective absorption loss, as schematically shown in Fig. 3e.
Fig. 4 Surface (the left column) and cross-sectional (the middle and right columns) SEM images of films with different concentrations. (a1–a3) 0.25 M, (b1–b3) 0.5 M, (c1–c3) 1 M, and (d1–d3) 2 M. |
Fig. 5a displays the relationships between the sheet resistance, conductivity and roughness of silver films prepared with inks of different concentrations. Except for the 0.25 M film, with the increase of the ink concentration, the changing trend of sheet resistance is consistent with that of film roughness, ranging from 0.59 Ω sq−1 to 0.76 Ω sq−1, and the conductivity decreases greatly from 8.46 × 106 S m−1 to 2.63 × 106 S m−1. The low solid content of the 0.25 M ink results in nonsufficient nanoparticles to establish an efficient electrical network. Therefore, the 0.25 M film exhibits an inferior electrical performance, with 2.58 Ω sq−1 sheet resistance and 1.94 × 106 S m−1 conductivity.
Fig. 5 (a) RMS roughness, sheet resistance and conductivity of films with different concentrations. (b) EMI SE in the X band. |
Fig. 5b shows the EMI shielding effectiveness (SE) of the films of different concentrations in the frequency range of 8.2–12.4 GHz. It can be seen that the electromagnetic shielding performances of silver films are mainly related to their electrical properties. In detail, the 0.25 M film with the worst electrical performance exhibits the lowest shielding performance. However, the 0.5 M film has the best conductivity and shielding effectiveness, with an average SE of up to 51.1 dB. Due to the decline of electrical properties affected by the large surface roughness, the shielding effectiveness of high concentration films (1 M, 2 M) decreases gradually. Accordingly, different ink concentrations cause the change of film surface roughness, and then affect their electrical properties and electromagnetic shielding performances.
Then, we explore the relationship between surface roughness, sheet resistance and resistivity. According to the SEM images, due to the uniform morphology of 0.25 M films, the roughness can be detected by atomic force microscopy, as shown in Fig. 7a and Fig. S4 (ESI†). However, considering the uneven surfaces of other concentration films and the big variety of the roughness in different micro regions, a laser confocal microscope was used to observe the large area region (Fig. S5, ESI†) and obtain the average surface roughness, as shown in Fig. 7b–d. The changing trends of the roughness of low concentration films (0.25 M and 0.5 M) and high concentration films (1 M and 2 M) with layers are different. The roughness of low concentration films increases with layers, while that of high concentration films decreases, in good agreement with the above SEM observations. Fig. 7a–d also present the relationships between the sheet resistance, conductivity and roughness of silver films of 0.25 M, 0.5 M, 1 M and 2 M with different layers. It can be seen that, except for 0.25 M films, the changing trend of the sheet resistance of films with different layers is consistent with that of the roughness. With the increase of layers, the sheet resistance of 0.5 M films increases slightly from 0.59 Ω sq−1 to 0.77 Ω sq−1, and the sheet resistance of high concentration films decreases greatly. For high concentration films, the large shift of sheet resistance with layers is due to the significant change of thickness. As other studies reported, the sheet resistance of the film decreased with the increase of thickness.48 The abnormal changing trend of the roughness and sheet resistance of 0.25 M films between the single layer and three layers is mainly because of the insufficient silver nanoparticles of the 0.25 M–1 L film to establish an effective electrical transport network. Therefore, for the 0.25 M–3 Ls film, enough silver particles make the sheet resistance decrease significantly and the conductivity increase obviously. While for the 0.25 M–6 Ls film, the increase of surface roughness leads to the increase of sheet resistance. Therefore, it can be considered that the film roughness and a significant change of film thickness have an effect on the sheet resistance. For all the fabricated films, unlike the sheet resistance, the conductivity of films has no obvious changing trend with the layers and has no direct change relationship with the roughness.
The EMI shielding effectiveness of four concentration films with different layers in the X-band is shown in Fig. 7e–h and i displays the relationship between the shielding effectiveness, sheet resistance and conductivity of the above films. It can be found that the change of electromagnetic shielding effectiveness of all the fabricated films is only related to the sheet resistance, and the changing trend is opposite to that of the sheet resistance. We calculated their skin depths and found that for all films, the ratios of thickness to skin depth are all less than 1.3, indicating that eqn (7) can be used to calculate their theoretical SE. The average SE and theoretical SE of all the films are presented in Fig. 7j. It can be seen that there is little difference between the theoretical calculation values and the measured values, which not only demonstrates the reliability of the measured values, but also confirms that the shielding effectiveness of the conductive thin layer is directly related to the sheet resistance according to eqn (7). Furthermore, it is found that the measured values of most films are slightly larger than the theoretical values, which may be due to the network structure of nanoparticles. The incident electromagnetic waves can be reflected multiple times in the porosity among the nanoparticles of the silver film, which increases the energy loss of electromagnetic waves. Moreover, the increase of multilayer film thickness and the introduction of pores between interlayers will also make the transmission path of incident electromagnetic waves become longer and the multiple reflected loss will be enhanced, which will further improve the electromagnetic shielding effectiveness.
From the above analysis, it can be concluded that the shielding effectiveness of silver films fabricated by this method is related to sheet resistance. Multilayers will affect the roughness and thickness of films, which will further affect the sheet resistance and the shielding effectiveness. For low concentration films (0.5 M), increasing the layers has an adverse effect on the shielding performance by increasing the film roughness. However, for high concentration films, multilayer films (1 M, 2 M) show better shielding performance, attributed to the lower roughness and larger thickness. Therefore, among all the fabricated films, the 2 M–6 Ls film exhibits the best shielding performance, with an average SE of 61.36 dB. Besides the high conductivity, the ultrathin thickness is another advantage of this method. Considering this, the values of SE divided by sample thickness, SE/t, are presented in Fig. 7k. The SE/t of single layer films with any concentration is larger than that of their corresponding multilayer films. Among them, the 0.5 M–1 L film has the largest SE/t, up to 256 dB μm−1. In comparison with other studies related to silver coatings or films (Table 1), this work has the best SE/t, highlighting the ultrathin characteristic and high shielding performance of this silver film prepared from the silver MOD ink. Furthermore, the silver films in this work have better flexible reliability. After multiple bending tests, the electromagnetic shielding effectiveness of the silver films is still above 50 dB (Fig. S6, ESI†), demonstrating that the silver films can be well applied in flexible wearable devices.
Sample | Fabrication method | Frequency/GHz | Conductivity/S m−1 | Thickness/μm | SE/dB | SE/t/dB μm−1 |
---|---|---|---|---|---|---|
AgNPs/PLA microfibers49 | Electroless deposition | 8.2–12.4 | 254 | 1500 | 50 | 0.033 |
AgNPs@CFs nonwoven fabrics50 | Electroless deposition | 8.2–12.4 | 3.3 × 104 | 500 | 111 | 0.222 |
PP/PDA/AgNPs/PDMS51 | Chemical reduction | 8.2–12.4 | 8.12 × 103 | 350 | 71.2 | 0.2 |
CNFs@PDA@AgNPs composite films52 | Electroless deposition | 8.2–12.4 | 1 × 106 | 167.2 | 93.8 | 0.56 |
AgNW/cellulose38 | Dip-coating | 1 | 6.75 × 103 | 160 | 48.6 | 0.3 |
PAN@SiO2–Ag-PFDT film53 | Electroless deposition | 8.2–12.4 | 1.78 × 104 | 50 | 81.01 | 1.62 |
Ag/WPU coating18 | Silver flakes ink | 8.2–12.4 | 1.43 × 106 | 10 | 73.6 | 7.36 |
Ag film54 | Atmospheric pressure plasma reduction | 0.1 | 4.41 × 106 | 0.93 | 60.49 | 65 |
Ag film55 | Spray coating | 1–1.5 | 1.2 × 106 | 0.2 | 45 | 225 |
0.5 M–1 L Ag film (this work) | Silver MOD ink | 8.2–12.4 | 8.46 × 106 | 0.2 | 51.1 | 256 |
Footnote |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1ma00918d |
This journal is © The Royal Society of Chemistry 2022 |