Preparation of silver nanoparticles by chemical

Colloids and Surfaces A: Physicochem. Eng. Aspects 256 (2005) 111–115

Preparation of silver nanoparticles by chemical reduction method Hongshui Wanga , Xueliang Qiaoa,∗ , Jianguo Chena , Shiyuan Dingb a

State Key Laboratory of Plastic Forming Simulation and Die and Mould Technology, Huazhong University of Science and Technology, Wuhan, Hubei 430074, PR China b Hubei Xinyin Noble Metal Co. Ltd., Shiyan, Hubei, PR China Received 16 March 2004; accepted 29 December 2004 Available online 30 January 2005

Abstract In the solution containing polyvinyl pyrrolidone (PVP), silver nitrate was reduced by the glucose, and silver particles were generated. The possible reaction process is discussed in this paper. Sodium hydroxide was used to enhance the reaction velocity. When the mole ratio of NaOH to AgNO3 was ranged from 1.4 to 1.6, the colloid kept stable and no Ag+ was traced. The particles and colloids were also analyzed by the X-ray diffraction (XRD), transmission electron microscopy (TEM) and UV–visible (UV–vis) spectrophotometer. The TEM photo indicated that with the increase in PVP, the particles dispersed better; and if the weight ratio of PVP to AgNO3 is no less than 1.5, the particles dispersed individually in a colloid form. The agglomeration of particles also was influenced by the mixing speed of the reactants. The XRD spectrums showed that the particles were silver simple substance if the reductant was sufficient and the mixing speed of the reactants was slow enough. © 2005 Elsevier B.V. All rights reserved. Keywords: Silver; Nanoparticles; Agglomeration

1. Introduction The area of nanoparticles research has witnessed tremendous growth due to the unusual chemical and physical properties demonstrated by this intermediate state of matter [1]. Due to their small size, these crystallites exhibit novel material properties that largely differ from the bulk properties [2]. Many reports on quantum size effect on photochemistry [3–5], nonlinear optical properties of semiconductor [6,7] or the emergence of metallic properties with the size of the particles [8–10] have appeared during the past. Nanoparticles of noble metals are of great interest today because of their possible applications in microelectronics [11–14]. Silver particles play an important role in the electronic industry. In recent years, with the higher integrated density of the electronic components (small size and precision of the electronic components), there are growing demands for a decrease in the thickness of conductive

∗

Corresponding author. Tel.: +86 27 87541540; fax: +86 27 87541540. E-mail address: [email protected] (X. Qiao).

0927-7757/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfa.2004.12.058

films and a further narrowing of the width of printed circuits and the space between these circuits. It is thus required that the powders (to form the conductive films and printing the circuits on the basement) composing the paste should have as small in diameter as possible, and the synthesis of these particles is an important task. The chemical way is often employed to synthesize silver colloidal metal particles [15–17].

2. Experimental procedures All chemicals used in the experiment were analytic reagent (AR). The silver nitrate was provided by Hubei Xinying Noble Metal Co. Ltd., glucose and polyvinyl pyrrolidone (PVP) were obtained from China Medicine (Group) Shanghai Chemical Reagent Corp. The sodium hydroxide was purchased from Tianjin Chemical Reagent Corp. Silver nanoparticles were prepared by reducing the silver nitrate in PVP aqueous solution. Glucose was used as reducer and sodium hydroxide to accelerate the reaction.

112

H. Wang et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 256 (2005) 111–115

Silver nitrate solution (A) was prepared by adding 3.4 g of AgNO3 into 20 ml distilled water. The PVP solution (B) was prepared by dissolving PVP, glucose and sodium hydroxide in 60 ml distilled water together. Solution B was heated to 60 ◦ C and stirred hard, and solution A was added into B drop by drop. After all the silver nitrate solution was added, the mixed solution was stirred for 10 min more. The particles were separated by centrifugation, and the solid products were washed with distilled water several times until no NO3 − could be traced. Characterizations of the particles were achieved by different techniques. X-ray diffraction (XRD) data were taken with Cu K␣ radi˚ on the powder diffractometer operated ation (λ = 1.5418 A), in the θ/2θ mode primarily in the 35–85◦ (2θ) range and stepscan of 2θ = 0.5◦ . Samples were prepared as uniform thin films supported on the slides. The transmission electron microscopy (TEM) was performed with a Jeol electron microscope (model JEM 100CX II). Samples were prepared by dispersing a drop of the colloid on a copper grid, which was covered by carbon film, and the solvent was evaporated. Extinction spectra were measured on a UV-2010 UV–vis spectrophotometer. All spectra were obtained from the particles immersed in water.

3. Results and discussion 3.1. Possible reaction process As shown in Fig. 1, the solution without PVP can be kept for 50 h at 60 ◦ C, and the UV–vis spectra were superposed completely with the silver nitrate solution, but 10 min after the addition of PVP, an obvious change can be seen on the curve, the peak at 300 nm was decreased dramatically, and a new peak appeared at about 420 nm. These changes can be explained as follows: the silver ions were reduced and the silver particles were generated. The possible reaction between glucose and silver ion in PVP solution can be written as follows: Ag+ + PVP → Ag(PVP)+

(1)

CH2 OH (CHOH)4 CHO + 2[Ag(PVP)]+ + 2OH− → CH2 OH (CHOH)4 COOH + 2Ag(PVP) ↓ + H2 O (2) 2Ag+ + 2OH− → Ag2 O + H2 O

(3)

Ag2 O + CH2 OH (CHOH)4 CHO + 2PVP → CH 2 OH (CHOH)4 COOH + 2Ag(PVP) ↓

(4)

Fig. 1. The absorption spectra of the silver ions solution. (A) The UV–vis spectrum of silver nitrate with a concentration of 1 mol/l. (B) The UV–vis spectrum of silver nitrate with glucose solution was kept for 2, 10 and 50 h at 60 ◦ C. (C) The UV–vis spectrum after 10 min of the PVP was added in B.

In the solution containing PVP, silver ions were reduced in two possible paths. The first is Eqs. (1) and (2). Ag+ was compounded with PVP firstly and complex ions were generated. The hydroxyl ion may undergo a nucleophilic addition reaction to glucose producing gluconate ions [15], and then it reduces silver ion to silver atom. The other two equations were the second path of the reaction. Ag+ reacted with hydroxyl ion, and then the product, Ag2 O, was reduced by glucose and silver particles were generated [18]. In the process of reaction, the disperser formed a protection layer on the surface of Ag2 O or Ag particles. The pH of the solution also plays an important role during the reaction process. Under lower pH, the reaction proceeded in the first pattern, and with the increasing of the pH, the second one became the dominant pattern gradually. The relative ratio of glucose to silver was actually kept at 2 throughout this study. This ratio should be sufficient to reduce all silver ions in the solution.


113

Table 1 Effects of sodium hydroxide Sample number

NaOH:AgNO3 (mole ratio)

Ag+ was traced

Deposition

1 2 3 4 5

1 1.2 1.4 1.5 1.6

Yes Yes Yes No No

No No No No Yes

3.2. Effects of sodium hydroxide According Eq. (2), the addition of alkali is favored for higher reducing ability; however, it had an adverse effect on particle agglomeration. Silver colloids were destabilized by sodium hydroxide and deposited. Samples with different sodium hydroxide were prepared. Each sample was divided into two parts. The first part was separated by centrifugation, and the Cl− was introduced to check whether Ag+ exsited in the supernate. The other part was standing to observe whether there was deposition precipitated 24 h later. As shown in Table 1, no Ag+ could be monitored until the mole ratio (nNaOH : nAgNO3 ) is over 1.4, and when it came to 1.6, particles were deposited out after 24 h. When nNaOH : nAgNO3 was ranged from 1.4 to 1.6, Ag+ had not reacted completely, but it could not be monitored by Cl− , and they were absorbed on the surface of particles, and formed electric double layer; it protected the colloid from agglomeration. With increased nNaOH : nAgNO3 , OH− diffused into the layer, and the adsorption layer was thinned. When nNaOH : nAgNO3 came to 1.6, the layer was destroyed by OH− , particles with a PVP layer have more chance to collide with each other, and the particles would agglomerate by the intertwist of PVP or the conjunction of silver atoms.

Fig. 2. The possible process of the agglomarate.

3.3. The inﬂuence of dispersant The use of dispersant has two purposes, one is to generate complex compound with the precursor, and control the process of the reaction as discussed in part 3.1, the other is to protect particles from growth and agglomeration (Fig. 2). Fig. 3 is the TEM photograph with different ratio of disperser to AgNO3 , with the increase in the disperser, the particle size has no obvious changes (the diameters of particles ranges from 20 to 80 nm), but the dispersibility becomes far better. In the reactions, the pH value of solution B was very high, reaction will follow the second pattern, when the AgNO3 solution was added. The particle size was controlled by the size of Ag2 O (20–70 nm), because the generation of Ag2 O is more quickly than the formation of protecting layer. When the amount of the disperser is not enough, it cannot form a complete protection layer, and the particles will agglomerate easily. With more disperser added, it can form a more perfect layer quickly, and the layer protects the particles from agglomeration and growth. As shown in Fig. 3, if the

Fig. 3. The TEM photograph with different rate of disperser to AgNO3 .

114


Fig. 5. The XRD spectrum of silver particles.

Fig. 6. The XRD spectrum of the samples containing Ag2 O.

Fig. 4. The influence of the mixing speed of reactants.

weight ratio (wPVP : wAgNO3 ) is about 1.5, the particles can disperse individually. 3.4. The inﬂuence of the mixing speed of reactants When the reactants were mixed, the generated particles needed time to disperse in the solution, and disperser also needed time to form a protection layer. If the mixing speed is too high, new particles will generate by side of the former ones and form bigger particles together. As shown in Fig. 4, the agglomerations in Fig. 4c are the biggest, and the particles in Fig. 4b are dispersed individually. When the speed is about 1 drop per second, the particles can disperse sufficiently before the new ones are generated, but if the speed was too slow, the reaction time become longer, and the particles will impact with each other and agglomerate because of the Brownian movement [19]. 3.5. XRD study All samples were analyzed by XRD. The spectrum of the sample used for Fig. 2 was shown in Fig. 5. There are five peaks corresponding to the interplanar distance (d): 2.3571, 2.0420, 1.4438, 1.2316 and 1.1795; they are all agreeing with

the spectrum of the silver. This indicated that the particles were silver simple substance. The curves of other samples were also agreed with Fig. 5. If the solutions were mixed in a high velocity, the XRD spectrums would be like Fig. 6. There are strong peaks at the point where d is 2.7233 and 1.6665; they are matching with the diffraction curves of the Ag2 O. The reason that the oxide was remained can be explained as follows: as the reducing process was slower than the generation of the Ag2 O, the Ag2 O particles are generated too fast to be reduced entirely before agglomerating, and then the agglomerations were protected by PVP immediately. So Ag2 O formed in the middle of the agglomeration with part of Ag and part of PVP on the surface, which made the reducing reaction not keep on. As a result, the Ag2 O was remained in the agglomerations.

4. Conclusions Well-dispersed silver particles with 20–80 nm size and spherical shape were prepared by reducing silver nitrate with glucose in the presence of protective agent PVP. The addition of the sodium hydroxide enhanced the reaction velocity. When the mole ratio of NaOH:AgNO3 is in range from 1.4 to 1.6, the colloid keeps stable and no Ag+ could be traced. The PVP protected the silver particles from growth and agglomeration, when the weight ratio of PVP:AgNO3 is no less than 1.5, the particles could be dispersed individually. When the mixing speed of the reactant was about 1 drop per second, the colloid got excellent dispersing ability. If the mixing speed of the reactants was too high, the Ag2 O remained in the particles.


Acknowledgement Thanks are due to Shuizhou Cai for the test of UV–vis. The authors are also grateful to Xiaoxia Yang for the help in the experiment.

References [1] A. Henglein, Top. Curr. Chem. 143 (1988) 113. [2] G.A. Ozin, Adv. Mater. 4 (1992) 612. [3] M.G. Bawendi, M.L. Steigerwald, L.E. Brus, Annu. Rev. Phys. Chem. 41 (1990) 477. [4] K. Kayanuma, Phys. Rev. B 38 (1988) 9797. [5] A. Eychmuller, A. Hasselbarth, L. Katsikas, et al., Phys. Chem. 95 (1991) 79. [6] Y. Wang, Acc. Chem. Res. 24 (1991) 133. [7] Y. Yuan, J.H. Fendler, I. Cabusso, Chem. Mater. 4 (1992) 312.

115

[8] G. Levi, J. Partigny, J.P. Marsault, et al., J. Phys. Chem. 96 (1992) 926. [9] M. Kerker, D.S. Wang, H. Chew, Appl. Opt. 19 (1980) 4159. [10] P. Matejka, B. Vlchova, J. Vohlidal, et al., J. Phys. Chem. 96 (1992) 1360. [11] R.P. Andres, J.D. Bielefeld, J.I. Henderson, et al., Science 273 (1996) 1690. [12] G. Schimd, Chem. Rev. 92 (1992) 1709. [13] W.A. Deheer, Rev. Nod. Phys. 65 (1993) 611. [14] K. Fukumi, A. Chayahara, K. Kadono, et al., J. Appl. Phys. 75 (1994) 3075. [15] K. Chou, C.Y. Ren, Mater. Chem. Phys. 64 (2000) 241. [16] P. Christophe, L. Patricia, P.J. Marie-Paule, Phys. Chem. 97 (1993) 12974. [17] S.A. Vorobyova, A.I. Lesnikovich, N.S. Sobal, Coll. Surf. A 152 (1999) 375. [18] Z. Zhang, B. Zhao, L. Hu, et al., J. East China Univ. Sci. Technol. 21 (4) (1995) 423. [19] B. Zhao, H. Ma, Z. Zhang, et al., J. East China Univ. Sci. Technol. 22 (2) (1996) 221.