3.1 SEM analysis
The microscopic appearance of TOMCC/AgNPs was observed as shown in Fig. 2. The process of preparing TOMCC/AgNPs as shown in the figure preserves the shape of TOMCC while successfully incorporating silver nanoparticles into TOMCC. At the same time, it can be seen in Fig. 2 that these silver nanoparticles aggregate with each other to form a cluster structure. [46, 47] It is worth noting that the corresponding silver nanoparticles and their cluster structures can also be observed in the transverse section of TOMCC, as shown in Fig. 2d, which further indicates that silver nanoparticles are mixed in TOMCC. Further magnifying the surface morphology as Fig. 2e, it is not difficult to find that most of the silver nanoparticles exist in the spherical state, and some of the silver nanoparticles are also accompanied by angular polygon state. Furthermore, Fig. 2f shows the size distribution of silver nanoparticles in TOMCC. The particle size distribution of silver nanoparticles is mainly concentrated in the range of 20 to 90 nm, and the most widespread is between 50 and 70 nm, with the average size of silver nanoparticles being about 65 nm. The color and particle size of silver nanoparticles were similar to those reported by Y Yan et al. [48], which successfully prepared silver nanoparticles and incorporated them into TOMCC.
In fact, electroless plating is a general process, especially in the bonding process between metals and between metals and nonmetals. However, unlike metal, non-metal has no catalytic activity, so it is necessary to generate catalytic active parts on non-metal materials in advance to ensure the normal process of electroless plating reaction. In previous studies, people often use gold, platinum and other precious metals with catalytic activity as catalysts to complete their electroless plating reaction. Such as Jun MATSUI, etc. they employ two kinds of functional polymer nanosheets: one works as an adhesion layer and the other works as a template layer to adsorb gold nanoparticles, which works as a catalyst for the electroless plating. Jun MATSUI et al. used this method to successfully deposit copper on polyimide films. [49] Following the continuous research, people focus on autocatalysis. For instance, Jian dong Hu, etc. use APTMS as an adhesive reagent was used to attach the gold nanoparticles (AuNPs) onto the glass substrate. These AuNPs could be regarded as the preferential nucleation or catalytic sites for gold electroless reduction, which accelerated the reduction of Au3+ on the glass surface and effectively prevented the formation of gold metal in the bulk solution. Finally, a continuous gold layer was successfully prepared on the glass surface. [50] Although the method has been improved, the mechanical connection between the substrate and the catalytically active particles is still essential. In other words, mechanical force is still the main combination between the substrate and the coating, and substances such as functional polymer nanosheets and APTMS are still used as connection bridges.
In our study, we spare the intermediates that provide mechanical adhesion and instead by electrostatic adsorption and chemical bonding to achieve fixation the catalytic active site. TEMPO can selectively oxidize the C-6 hydroxyl groups on the surface of microcrystalline cellulose into carboxyl groups, [51, 52] which in turn ionize into COO− in aqueous solution. At the same time, the electrostatic adsorption occurs between COO− and Ag+, forming−COO-Ag+, which makes the silver ions adsorbed in TOMCC, and generates catalytically active silver nanoparticles through in situ reduction. Meanwhile, the O-Ag structure is formed to bind the silver nanoparticles with TOMCC through chemical bonds. Finally, more silver nanoparticles were generated around the catalytic active site by electroless plating and mixed in TOMCC. Compared with previous studies, it is simpler and cheaper, such electrostatic adsorption and chemical bonding make silver nanoparticles more dispersed and firm, reduce agglomeration of silver nanoparticles, and increase the stability of TOMCC/AgNPs. [53]
3.2 EDS、UV-vis、XRD analysis
The EDS patterns of the TOMCC/AgNPs were shown in Fig. 3. The peak of silver element appeared on the elemental composition diagram. The strong peak caused by silver between 2 keV and 3.5 keV can be seen from the diagram. After analysis of element content, The atom and mass fractions of silver on the surface of the TOMCC reached 85.92% and 97.89%, respectively.
To further prove that silver is composite in the form of nanoparticles in TOMCC, UV-visible diffuse reflection of TOMCC/AgNPs was performed. As shown in Fig. 4, the UV-vis diffuse reflection spectrum of TOMCC/AgNPs had a strong absorption peak about the wavelength of 426nm, which was attributed to the light absorption peak generated by the plasmon resonance effect of silver nanoparticles interspersed in TOMCC. [54 ~ 56] This proves that silver is present in TOMCC as nanoparticles.
The crystal structure of silver nanoparticles was analyzed by XRD shown as Fig. 5. The four distinct peaks of the silver nanoparticles were comprised in X-ray diffraction pattern, which corresponding to the (111), (200), (221), and (311) planes, at 2θ = 38.32°, 44.48°, 64.67° and 77.58°. The face centered cubic structure of silver was certified through these peaks. At the same time, there were no characteristic peaks of silver oxide and silver nitrate in the figure, which means that there were no impurities in the reaction except elemental silver. [57, 58]
3.3 XPS analysis
XPS was employed to further analyze the composition and surface states of the TOMCC immobilized with AgNPs. The binding energies of Ag 3d at about 371 eV, Ag 3p at 573 eV and 604 eV, and Ag 3s at 719 eV are distinct in Fig. 6 and demonstrate the presence of the silver element. [59] The Ag 3d core-level spectrum shown in Fig. 7 can be deduced from the two peak components at BEs of 368.43 eV and 374.43 eV for Ag 3d5/2 and Ag 3d3/2, respectively. Both peaks are attributed to the Ag0 species, thus confirming the presence of Ag, in good accordance with XRD results.
3.4 Raman analysis
The Fig. 8 illustrates the Raman spectra of TOMCC and TOMCC/AgNPs. The absorption band at approximately 229 cm− 1 in the TOMCC/AgNPs spectrum was attributed to the Ag-O stretching vibration [60]. This result further suggested that the interaction between Ag and the carboxyl groups of TOMCC molecules was linked by a chemical bond. The most significant feature in the Raman spectrum is the emergence of a very strong band at 1374 cm− 1 assigned to the CO2 symmetric stretch. The enhancement in the intensity of the CO2 stretching vibration suggests that the carboxyl groups are directly involved in the interaction with the silver surface. The remarkable red shift of 22 cm− 1 for the CO2 symmetric stretch (1374 cm− 1) observed in the Raman spectrum from that observed in normal Raman spectrum of the molecule in solution (1396 cm− 1) compares well with that observed for the carboxyl band of most carboxylic acids [61 ~ 63]. This red shift in the carboxyl symmetric stretching vibration also indicates direct attachment of the carboxylate group to the silver surface. In addition, the strong and sharp peaks at 1540 cm− 1 and 1597 cm− 1 suggested the direct binding of the carboxyl groups with the Ag surface [60]. The Raman spectra indicated that the carboxyl groups of TOMCC molecules covalently and directly interacted with AgNPs by monodentate interaction in aqueous solution.
3.4 BET analysis
The BET analysis was carried out for TOMCC/AgNPs in order to evaluate the surface area and porosity shown as Fig. 9 and Fig. 10. The IV nitrogen sorption isotherms for the TOMCC/AgNPs suggest the existence of different pore sizes from mesoporous to macro pores. [64] As well, the hysteresis between adsorption and desorption branches of the TOMCC/AgNPs also suggest the existence of different types of pores with no pore-blocking effect during desorption. [65] The pore size of TOMCC/AgNPs ranged from 10nm to 80nm, the average pore size was 32.59nm, and the BET specific surface area was 1.89m2g− 1.
Actual, from the perspective of the structure of cellulose, the surface of cellulose in its natural state has pore structure. In previous studies, it was reported that the specific surface area of cellulose could reach tens of m2g− 1 or even hundreds of m2g− 1 after some changes. In our study, these natural pores on the surface of TOMCC can become the providers of catalytic active sites through the adsorption of silver ions, and even can adsorb silver ions in the plating solution to generate silver nanoparticles. The specific surface area of the obtained TOMCC/AgNPs was very low, indicating that the silver nanoparticles filled most of the pores of TOMCC. However, a still has some pores in the measurement, indicating that the silver nanoparticles have not filled all the pores of TOMCC. Adsorption is only an essential step in the preparation process of silver nanoparticles, but more importantly, the silver ions absorbed can be dispersed in the cellulose through the electrostatic attraction of COO− and Ag+, so that the generated silver nanoparticles cannot be too much agglomerate. As shown in Fig. 2, the prepared silver nanoparticles have stable size and good dispersion.
3.5 Thermal performance analysis
The influence of temperature on the thermal behavior of TOMCC/AgNPs was studied by TG and DTG analysis. As shown in Fig. 11, the thermal behavior of TOMCC/AgNPs and TOMCC was different. When the temperature was increased from room temperature to about 249℃, TOMCC (b) exhibited a slight loss in weight was attributed to the loss of water mlecules. Whereas for TOMCC/AgNPs it happens at room temperature up to 300℃. When the temperature rises from 249℃ to 352℃, TOMCC experiences a rapid loss of weight, and TOMCC/AgNPS does the same between 300℃ and 370℃.As the temperature increased, the rate of weight loss to slow down. As a result, the TOMCC and TOMCC/AgNPs were reduced by 91.5% and 45% in mass, respectively. The latter exhibited a higher thermal performance than former, which was related to the silver particles.
In fact, the thermal behavior of cellulose is mainly divided into two processes. The first process is the loss of water molecules, and the second process is the degradation and decomposition of cellulose. The temperature of TOMCC/AgNPs is higher than that of TOMCC in both the first and second processes, which is inseparable from the binding of silver nanoparticles, especially in the first process. These increments might be caused by the increase of lateral forces in the bulk state due to restricted steric effect of hydroxy groups by the branching of silver nanoparticle with the chain of TOMCC molecules. This is also reflected in the Raman spectrum, where the peak of TOMCC at 2929cm− 1 is stronger than that of TOMCC/AgNPs.[66] The degradation and decomposition of cellulose involves the breaking of chemical bonds, which means that more energy is needed to break the Ag-O bond, making the thermal behavior of TOMCC/AgNPs more stable. [60]
3.6 Electromagnetic parameters
Electromagnetic parameters are generally expressed by complex relative permittivity
(εr = ε′-jε′′) and complex relative permeability (µr = µ′-jµ′′). The complex relative permittivity and complex relative permeability of TOMCC/AgNPs between 2 and 18 GHz are shown in Fig. 12 and Fig. 13. Considering that TOMCC/AgNPs shows inconspicuous magnetic property in GHz, frequency dependent complex permittivity was recorded and analyzed. It can be seen from Fig. 12 the frequency dependence of complex permittivity of TOMCC/AgNPs. The real part of complex permittivity shows obvious dispersion characteristic that ε′ value decreases with the increase of frequency from 22 to 6, due to the lags of induced charges following the external reversing electromagnetic filed. The imaginary part ε״ is between 1 and 6, and there are two peaks in the whole frequency range, which is related to the relaxation of TOMCC/AgNPs.
Generally speaking, the real part of complex permittivity represents the storage of electric field energy and the imaginary part relates to the attenuation ability of electric field energy. Tangent value of dielectric loss tanδ = ε″/ε′ corresponds to the attenuation factor of incident electromagnetic wave energy, as shown in Fig. 14. Compared with some related materials reported, tanδ of TOMCC/AgNPs is obviously higher than them, which showing relatively good microwave absorption performance. [67 ~ 71]
In order to better explain the absorption mechanism of the TOMCC/AgNPs, the dielectric properties were further analyzed. Figure 15 shows the relationship between the real part ε′ and imaginary part ε″ for TOMCC/AgNPs composites. The plot of ε″ versus ε′ of TOMCC/AgNPs is complex with about five semicircles, indicating that TOMCC/AgNPs exhibit strongest Debye relaxation and could convert much more electromagnetic wave energy to thermal energy under external electromagnetic field. These five semicircles may arise as follows: under the alternating electromagnetic field, the lags of induced charges from the silver-silver, silver-TOMCC interfaces result in relaxation as Fig. 16.[72 ~ 76] In addition, the relaxation phenomenon caused by the charge density difference between silver and oxygen (carboxyl group), as well as the functional group molecules such as hydroxyl and ether on TOMCC will also convert electromagnetic energy into thermal energy under the application of electric field.[77, 78]
In general, the combination of silver nanoparticles and TOMCC endows TOMCC/AgNPs with dielectric loss ability, which plays a role in attenuating the absorption of electromagnetic energy.
3.7 Microwave absorption
In order to further quantitatively investigate the microwave absorbing properties of TOMCC/AgNPs, the reflection loss are calculated. With above measured complex permittivity at a given frequency, reflection loss can be obtained from the following equations 3 and 4. [79] Where Zin was input impedance of the absorbing material, Z0 was the characteristic impedance of free space, ƒ was the frequency of electromagnetic waves, d was the thickness of the absorbing material, c was the speed of light, εr and µr were complex permittivity and complex magnetic permittivity of the composites. It was generally believed that RL= -10 dB means 90% of electromagnetic waves were lost and RL= -20 dB means 99% of electromagnetic waves were attenuated.
$${Z}_{in}={Z}_{0}{\left({\mu }_{r}/{\epsilon }_{r}\right)}^{1/2}tanh\left[j\left(2\pi fd/c\right){\left({\mu }_{r}{\epsilon }_{r}\right)}^{1/2}\right] 3$$
$$RL=20log\left|\frac{{Z}_{in}-{Z}_{0}}{{Z}_{in}+{Z}_{0}}\right| 4$$
Figure 17 showed the reflection loss of the TOMCC/AgNPs between 2 and 18 GHz. What’s more, the frequency of peak absorption value shifts to low frequency with the increased thickness, which can be explained by quarter-wavelength matching model. [76, 80] The optimal reflection loss value is -51 dB at 4.7 GHz with a corresponding effective absorption bandwidth is 4 GHz, and the thickness of the layer is 2mm. In addition, the reflection loss value of -46 dB at 11.2 GHz with a corresponding effective absorption bandwidth of 3 GHz, and the reflection loss value of -41 dB at 15.6 GHz with a corresponding effective absorption bandwidth of 4 GHz, which the thickness of layer former is 3mm, and latter 2mm.