2022 Volume 63 Issue 5 Pages 748-751
Silver nanoparticles (AgNPs) on the surface of cellulose nanofibers (CNF) were immobilized using a high-pressure wet-type jet mill to increase their stability while maintaining their catalytic activity. As a raw starting material, an aqueous silver nitrate solution was mixed with a CNF suspension. The mixture was then processed five times using the high-pressure wet-type jet mill at a discharge pressure of 100 MPa. AgNPs with an average particle size of 3.8 ± 0.8 nm were immobilized and well-dispersed on the surface of CNF, as observed by transmission electron microscopy. The catalytic activities of the AgNPs coated with PVP (Ag–PVP) and the prepared AgNPs-supported on CNF composite (Ag–CNF) were evaluated. As a result, the catalytic activity of Ag–CNF was discovered to be 2.32 times greater than that of Ag–PVP. The catalytic activity was measured in terms of the number of runs because Ag–CNF can be easily recovered by centrifugation or filtration. Even the catalytic activity of Ag–CNF after five runs was maintained at 62%, but the catalytic activity gradually decreased as the number of runs increased.
Fig. 2 Transmission electron microscopy images and particle size distribution histograms of (a) Ag–CNF and (b) Ag–PVP, respectively.
Metal nanoparticles (NPs) have been extensively researched in recent years on a wide range of properties,1–4) including catalytic activity.5) The vast majority of those NPs are synthesized from the “bottom-up” approach. For example, silver NPs (AgNPs) have been prepared via a few different routes, such as chemical and photochemical reduction,6–9) laser ablation,10) processes involving emulsion,11) and ultrasonic irradiation.12,13) Surfactants such as cetyltrimethylammonium bromide (CTAB) and polyvinylpyrrolidone (PVP) must be used in these methods to stabilize the NPs state, suppress grain growth, and prevent aggregation.14) In other words, the AgNPs lose their activity without those agents because of the smaller surface area due to the grain growth. Polysaccharides, such as carbohydrate (starch)15) and carboxymethyl cellulose (CMC),16–18) are used as both a stabilizing agent and reductant in the green synthesis of AgNPs. However, those agents would cover the particle surface for stabilization while decreasing the catalytic activity of NPs. The same covering effects apply to CTAB or PVP as an immobilizing matrix for AgNPs.
Cellulose nanofibers (CNF) are made by pulverizing pulp cellulose to nanometer size; they have excellent physical and chemical properties such as high elastic modulus, low linear thermal expansion, high chemical stability, nontoxicity, and thixotropic properties.19,20) It is emphasized that CNF is crystalline and maintains the shape of fibers. A technique known as high-pressure wet-type jet milling is one of the most appropriate techniques for the fabrication of CNF. The high-pressure wet-type jet mill (jet mill) can apply high shearing forces to the pulp to fabricate CNF. Fujii and Furutani focused on exposing the reducing functional groups present in the structure of CNF by applying high shearing forces with the jet mill. Fujii and Furutani have reported that AgNPs precipitate owing to the reduction of CNF when CNF was pulverized with the jet mill in silver ions.21,22) In addition, AgNPs were immobilized on the surface of the CNF. It is considered CNF to be an outstanding matrix for immobilizing AgNPs. In this case, it is suggesting that the AgNPs–CNF composite (Ag–CNF) may retain AgNPs’ high surface activity because CNF does not completely cover the AgNPs surface, unlike CTAB, PVP, or polysaccharides. This paper will report the evaluation of the catalytic activity of Ag–CNF, and compare the activity with that of PVP-coated AgNPs (Ag–PVP).
Furutani and Fujii have already described the details of the Ag–CNF preparation procedure.22) In brief, separate 0.8% suspensions of CNF (product code: BiNFi-s FMa-100, Sugino Machine Ltd., Toyama, Japan) and 74 mM AgNO3 solution (FUJIFILM Wako Pure Chemical Corp., Osaka, Japan) was prepared, and the two were mixed in a volume ratio of 1:1. The suspension mixture was subjected to the jet mill (100 MPa; HJP-25001, Sugino Machine Ltd., Toyama, Japan) five times to obtain Ag–CNF. The jet milling is advantageous because it atomizes the mixture while also reducing Ag+ to Ag0 to produce Ag–CNF in a single step.
Ag–PVP was purchased from EM Japan Co., Ltd. (Product code: NP-AG-6-5) as the reference.
2.2 CharacterizationThe morphology of the samples was observed by transmission electron microscopy (TEM, JEOL Ltd., Tokyo, Japan, JEM-2100plus; accelerating voltage: 200 kV). For the observation, a drop of diluted sample suspension was developed on a Cu mesh with a carbon film (Product code: UHR-C10, OkenShoji Co., Ltd., Tokyo, Japan) and air-dried. By measuring the size of 100 particles in the TEM images, the average particle size was determined. The silver content of the Ag–CNF was determined using inductively coupled plasma–atomic emission spectroscopy (ICP–AES, Thermo Fisher Scientific Inc. Japan, iCAP6500Duo): a suitable amount of Ag–CNF was dissolved in hot HNO3, and the filtrate was measured using the ICP. For the identification of the crystalline phases, the obtained samples were subjected to vacuum freeze-drying. The crystalline phases of the samples were identified using X-ray diffraction (XRD, RIGAKU Corp., Osaka, Japan, Smart Lab; Cu-Kα). It was operated at 45 kV and 200 mA, in the range of 2θ = 5°–60°.
2.3 Catalytic activitiesThe catalytic activity of Ag–CNF and Ag–PVP was evaluated on bleaching methylene blue trihydrate (MB, IUPAC; 3,7-bis(Dimethylamino)phenothiazin-5-ium chloride, FUJIFILM Wako Pure Chemical Corp., Osaka, Japan) to leucomethylene blue by NaBH4 was investigated. Aqueous solutions containing 0.05 mM MB and 100 mM NaBH4 were prepared. Three types of aqueous suspensions were prepared: CNF (1 mass%), Ag–CNF (1 mass%), and Ag–PVP (37.9 mg/L). In this case, the silver content in the last two suspensions was the same: 37.9 mg/L.
The MB solution (2.7 mL) and the suspensions (0.1 mL each) were placed in a vial bottle and mixed thoroughly. Later, the NaBH4 (0.2 mL) was added to the vial bottle, and the mixture was pipetted well. After 20 min, the suspension was centrifuged. The absorption spectrum of the supernatant was measured in the range of 400–800 nm using an ultraviolet–visible spectrophotometer (UV–Vis, Shimadzu Corp., Kyoto, Japan, UV-3100PC). Equation (1) was used to calculate the reaction ratio (%) using absorbance at 662 nm.
| \begin{equation} \text{Reaction ratio}\ (\%) = (\text{A}_{0} - \text{A}_{20})/\text{A}_{0} \times 100 \end{equation} | (1) |
Where A0 represents the initial absorbance and A20 represents the absorbance at 20 min. After each run, the catalyst was centrifuged out from the suspension, the optical absorption spectrum of the supernatant was measured using a UV–Vis scanning spectrophotometer. The recovered residue was washed and re-served: this photometric evaluation procedure was repeated five times.
Kinetics analysis was monitored the absorbance at 662 nm at 1-min intervals. The measured suspension was not centrifuged, but it was confirmed to not affect the absorbance at 662 nm.
Figure 1 illustrates the XRD pattern of an Ag–CNF sample that was jet milled five times. The cellulose I type crystallites (JCPDS 00-056-1718) were responsible for the peaks at 14.9°, 16.4°, 22.7°, and 34.5° (○).23) The others, at 38.1° and 44.2° (●), were attributable to metallic silver (JCPDS 01-004-0783). It was confirmed that the reduction of Ag+ to metallic Ag occurs by processing the mixture containing both CNF and Ag+.
X-ray diffraction pattern of the sample prepared using the jet mill.
Figure 2 shows the TEM images and the size distribution histograms of (a) Ag–CNF and (b) Ag–PVP, respectively. The size distribution histograms of Ag–CNF, and Ag–PVP particles were derived from the particle size measurement in the TEM images shown in Fig. 2. The dark nanometer-sized spherical dots embedded, immobilized, and uniformly dispersed within the CNF were about 10–20 nm in width and connected each other in the image of Ag–CNF. The dark dots in Fig. 2(a) were AgNPs, according to the XRD pattern. It should be noted that these AgNPs did not contain any protective agents such as PVP. Figure 2(b) showed that the Ag–PVP particles were also spherical and dispersed well without aggregation. The mean particle size for Ag–CNF and Ag–PVP was 3.8 ± 0.8 nm and 5.7 ± 2.0 nm, respectively. Because the mode (most frequent value) of the size for Ag–CNF and Ag–PVP is 3 and 5 nm, respectively, it appears that Ag–CNF was composed of slightly smaller particles with a sharper and narrower size distribution than Ag–PVP.
Transmission electron microscopy images and particle size distribution histograms of (a) Ag–CNF and (b) Ag–PVP, respectively.
The optical absorption spectra in Fig. 3 present the catalytic activity of both AgNPs on bleaching MB by NaBH4. Here the spectra for the MB solution with or without CNF were also illustrated. It was confirmed that the MB solution exhibits a strong and distinct peak at 662 nm. By comparing the absorbance of the main peak at 662 nm, the absorbance derived from the MB solution contained Ag–PVP slightly reduced, whereas the absorbance of the MB solution contained CNF suspension barely reduced or yielded nearly the same profile as the MB solution. Meanwhile, the absorbance spectrum of the MB solution contained Ag–CNF showed no peaks, indicating that MB was completely reduced. It was revealed that the Ag–CNF had high catalytic activity.
Ultraviolet–Visible adsorption spectra for the catalytic reduction of methylene blue after reduction for 20 min.
AgNPs are difficult to separate by membrane filter or centrifugation. However, because Ag–CNF is a material in which AgNPs are immobilized on CNF, it can be easily recovered by centrifugation. Using this recovery, it was determined whether repeated use of the recovered Ag–CNF would maintain the catalytic activity. Figure 4 shows the catalytic activity concerning the number of runs. The catalytic activity gradually decreased as the number of uses increased. The decrease could be attributed to the formation of silver oxide on the surface of the AgNPs, as well as a decrease in the amount of Ag–CNF recovered due to centrifugation. Despite this, even after five runs, it remained at 62%.
Relationship between reaction ratio and number of runs.
Because the MB solution exhibits a strong and distinct peak at 662 nm, the absorbance of this peak can be used to calculate the amount measure of MB reduced. Figure 5 is the Langmuir–Hinshelwood plot24) of ln(At/A0) vs. the reduction time t. According to the model, eq. (2) correlates At/A0 with Ct/C0 as
| \begin{equation} \mathrm{Ln} (\text{A}_{\text{t}}/\text{A}_{0}) = \mathrm{Ln} (\text{C}_{\text{t}}/\text{C}_{0}) = -\text{kt} \end{equation} | (2) |
Plots of Ln(At/A0) versus time for the reduction of MB using CNF, Ag–PVP, and Ag–CNF. ●: Blank (MB), □: MB + CNF, ○: MB + Ag–PVP, △: MB + Ag–CNF.
Here, k is the rate constant whereas C0 and Ct represent the MB concentration at time t = 0 (initial state) and at time t, respectively. The model assumes a pseudo-first-order reduction due to excess concentration of NaBH4 over MB. This assumption holds under the current experimental conditions. Figure 5 shows that all of the systems have good linear correlations. Blank and CNF represent the MB solution and suspension of CNF in the MB solution, respectively, whereas Ag–PVP and Ag–CNF represent MB solutions suspending Ag–PVP particles and Ag–CNF composite, respectively. The decrease in ln(A0/At) for Blank means NaBH4 in the system bleached MB, though a little, without the catalysts. The decrease of the absorbance derived from the MB solution contained CNF suspension, which is similar to Blank, indicating that CNF has no catalytic activity on MB bleaching. Figure 5 also shows that Ag–CNF has higher catalytic activity than Ag–PVP, as shown in Fig. 3. The rate constants for the systems derived directly from the slope of the lines are shown in Fig. 5, indicating that the rate constant for Ag–CNF was 2.32 times greater than that of Ag–PVP. Finally, the catalyst activity is proportional to the number of active sites, which is determined by the surface area.
Table 1 lists the total surface area (TSA), total volume (TV), and specific surface area (SSA) of AgNPs in Ag–CNF, and Ag–PVP. All silver particles were assumed to be spherical with a density of 10.49 g·cm−3 in this study. TSA and TV were calculated from the size of 100 particles derived from TEM images, respectively. The mass was calculated from TV and density and converted to the SSA.
Ag–CNF has 1.94 times greater SSA than Ag–PVP. Taking this result into consideration, the difference in catalytic activity (2.32 times) is thought to be due to the effect of PVP covering the surface of AgNPs. In addition, the entire surface of AgNPs within Ag–CNF is not exposed because AgNPs are immobilized on the CNF matrix. As a result, Ag true catalytic activity in Ag–CNF is even higher. It seems that high catalytic activity was demonstrated because AgNPs did without coating anything for the Ag–CNF. Furthermore, Ag–CNF, in which AgNPs are immobilized on CNF, has the advantage of being easy to recover, and the catalytic activity after five runs can be maintained at 62%.
Ag–CNF was prepared by processing a mixture containing an aqueous silver nitrate solution and a CNF suspension five times using the jet mill at a discharge pressure of 100 MPa. As the reference, Ag–PVP, which coated AgNPs with PVP, was purchased. As a result of comparing the catalytic activity of Ag–CNF and Ag–PVP, it was discovered that Ag–CNF has approximately 2.32 times the activity of Ag–PVP. The catalytic activity after five runs was maintained at 62% as a result of the repeated examination, but the catalytic activity gradually decreased.
The authors gratefully acknowledge Prof. A. Osaka of Okayama University for discussion.
