- Email: [email protected]
Study on synthesis of ultrafine Cu–Ag core–shell powders with high electrical conductivity
Applied Surface Science 263 (2012) 38–44
Contents lists available at SciVerse ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Study on synthesis of ultrafine Cu–Ag core–shell powders with high electrical conductivity Yu-hsien Peng a,b,c , Chih-hao Yang b , Kuan-ting Chen e,∗ , Srinivasa R. Popuri d , Ching-Hwa Lee a , Bo-Shin Tang b a
Department of Environmental Engineering, Dayeh University, 168 University Rd., Dacun, Changhua 515, Taiwan, ROC Department of Research & Development, Oriental Happy Enterprise Co., No. 27, Xin’ai Rd., South Dist., Tainan 702, Taiwan, ROC c Center for General Education, Kun Shan University, No. 949, Dawan Rd., Yongkang Dist., Tainan 710, Taiwan, ROC d Department of Biological & Chemical Sciences, The University of the West Indies, Cave Hill Campus 11000, Barbados e Department of Resources Engineering, National Cheng Kung University, No.1, Da-Hsueh Road, Tainan 701, Taiwan, ROC b
a r t i c l e
i n f o
Article history: Received 21 February 2012 Received in revised form 8 August 2012 Accepted 18 August 2012 Available online 25 August 2012 Keywords: Composite materials Fourier transform infrared spectroscopy X-ray diffraction topography Electrical conductivity
a b s t r a c t Cu–Ag composite powders with high electrical conductivity were synthesized by electroless plating of silver sulfate, copper powders with eco-friendly sodium citrate as reducing agent, dispersant and chelating agent in an aqueous system. The influences of sodium citrate/Ag ratio on Ag coatings of Cu powders were investigated. Ag was formed a dense coating on the surface of Cu powders at a molar ratio of sodium citrate/Ag = 0.07/1. SEM showed an uniformity of Ag coatings on Cu powders. SEM-EDX also revealed that Cu cores were covered by Ag shells on the whole. The surface composition analysis by XPS indicated that without Cu or Ag atoms in the surface were oxidized. The resistivity measurements of Cu–Ag paste shows that they have closer resistivity as the pure silver paste’s after 250 ◦ C for 30 min heat-treatment (2.55 × 10−4 cm) and 350 ◦ C for 30 min heat-treatment (1.425 × 10−4 cm). © 2012 Elsevier B.V. All rights reserved.
1. Introduction Copper and silver are widely used in the electronic industry [1–3], sensors [4], catalysis [5,6], and optical or biological devices due to their high electrical conductivity. According to the rare content of silver in the earth’s crust and easy oxidization of copper strongly constrain their applications. Thus Cu–Ag core–shell powders instead of single copper or silver powders is currently considered to be a desirable option in electronic industry to solve the above problems [7]. Generally, powders with core–shell structure can be achieved by electroplating [8], electroless plating (also called spontaneous displacement reaction) [9,10], and vacuum process (evaporation, sputtering [11], etc.). It is well known that silver can be deposited on many substrates by electroplating [12]. However, the efficiency of electroplating and vacuum coating for powders is very low. Therefore, electroplating and vacuum processes are not feasible for commercial purpose depending on the efficiency for particles
Abbreviations: SC, sodium citrate. ∗ Corresponding author at: 168 University Rd., Dacun, Changhua 515, Taiwan, ROC. Tel.: +886 4 851 1338/886 6 261 8328; fax: +886 6 2616003. E-mail addresses: [email protected], [email protected] (K.-t. Chen). 0169-4332/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.apsusc.2012.08.066
which is extremely low. The spontaneous displacement reaction of silver coating that can be achieved on copper powder and has a higher deposition rate than electroplating, although there are still difficulties in reducing process steps and form a well silver coating on the surface of copper powder. According to the literature, the rate of replacement reaction is quite important and has great impact on the final product. Mancier et al. [13] prepared the Cu–Ag core–shell nanopowders by ultrasound-assisted electrochemistry followed by a displacement reaction. Ethylenediaminetetraacetic acid (EDTA) was used as the chelating agent to control the rate of Ag–Cu displacement. Xu et al. [14] reported a synthesis of silver coating process on fine copper powder via Cu displace with silver nitrate/ammonium hydroxide solution. The quantity of ammonia controlled the rate of copper dissolution. In the aforementioned references, many of these synthetic methods are unable for mass-production scale due to their low efficiency or the inadequancy of precious equipment. Furthermore, in most of the processes, sodium borohydride (NaBH4 ), hydrazine hydrate (N2 H4 ·H2 O) were used as reducing agent, hexadecyltrimethyl ammonium bromide (CTAB), polyvinylpyrrolidone (PVP) and EDTA, etc. were used as dispersant and chelating agent, respectively, which are hard to be removed from the final products or poisonous. Recently, sodium citrate (Na3 C6 H5 O7 ) was found to be more widely used for replacing the above chemicals as multifunctional agent [15–18], such as a reducing agent,
Y.-h. Peng et al. / Applied Surface Science 263 (2012) 38–44
39
Table 1 Proportions and compositions of composite powders. Sample
[SC]/[Cu]
Cu powders (g)
SC for dispersant (g)
[Cu]/[Ag+ ] molar ratio
0.015 M Ag2 SO4 (ml)
[SC]/[Ag+ ] molar ratio
SC for chelating (g)
1 2 3 4 5
0.0078:1 0.0078:1 0.0078:1 0.0078:1 0.0078:1
2.53 2.53 2.53 2.53 2.53
0.09 0.09 0.09 0.09 0.09
5.3:1 5.3:1 5.3:1 5.3:1 5.3:1
250 250 250 250 250
0:1 0.07:1 0.27:1 1.09:1 4.35:1
0 0.15 0.6 2.4 9.6
dispersant, chelating agent, etc., and it also suppress precipitation of copper hydroxides in Cu–Ag spontaneous displacement reaction [14,15,18–21]. In this article, we report a simple and novel method to fabricate Cu–Ag core–shell powders using copper powders, silver sulfate, SC (sodium citrate) as starting materials at room temperature. As compared with the results reported previously, this method is a relatively green synthetic method because the starting materials come from SC which serve as mutifuntional agents such as reducing agent, dispersant and chelating agent, and are nontoxic and easily available. Furthermore, the Cu–Ag powders prepared by this method are hardly oxidized, which have greet potential for application in many fields. 2. Experimental 2.1. Materials Copper powders with primary size less than 2 m (median size measured by Laser Diffraction Particle Size Analyzer is 7.36 m) was supported from Oriental Happy Enterprise Co., Ltd., Taiwan, silver sulphate (Ag2 SO4 ) was from Solar Technology Inc., Taiwan, sodium citrate (Na3 C6 H5 O7 ) was from Union Chemical Works Ltd., Taiwan, flake and spherical Ag powders with median size 6.1 m, 5.2 m, respectively were from Metalor Technologies, Taiwan. All chemicals were used as received without further purification. 2.2. Preparation of Cu–Ag powders The chemical reactions of copper powders, sodium citrate (SC) and Ag2 SO4 systems should occurred as following reactions (1)–(4) [12,22,23]. C6 H5 O7 3− + 2Ag+ → C5 H4 O5 2− + H+ + CO2 + 2Ag
(1)
(Ag)2 + + citrate− ↔ [(Ag)2 + · · ·(citrate)− ] +
Cu + 2Ag ↔ 2Ag + Cu
2+
3Cu2+ + 2C6 H5 O7 3− ↔ Cu3 (C6 H5 O7 )2
(2) (3) (4)
During the preparation of Cu–Ag powders, two solutions denoted as A and B were first prepared. For preparing solution A, 2.53 g Cu powders, 0.09 g SC and 100 ml deionized water were mixed at room temperature by a magnetic stirrer; for preparing solution B, 1.17 g Ag2 SO4 ,various amounts of SC and 250 ml deionized water were mixed to form [(Ag)2 + · · ·(citrate)− ]. Detailed experiment design was shown in Table 1. First, solution B was quickly dropped into solution A with a magnetic stirrer at 1000 rpm. After the color of reacting solution from red to purple and then dark green, the reaction was ended by stopping the motion of the magnetic stirrer which was immediately followed by the filtration and clean processes. The obtained composite powders after single step filtration were washed at least 3 times by deionized water, and then dried. 2.3. Preparation of metal paste and sample for SEM cross-sectional image To ascertain the oxidation resistance of synthetic powders, the resistivity () of synthetic powders were investigated, the prepared powders was mixed with an appropriate amount of ethyl acetate–terpineol solution by a three-roll mill. The paste was grounded to the required consistency, and screened on to the glass substrates for forming metal films, and then dried at various conditions in air ambient. To determine the structure of synthetic powders, the sample of synthetic powders for SEM cross-sectional image were fabricated. The synthetic powders was firstly added into epoxy resin. After solidification, the resin which contain synthetic powders were polished by Grinder and Polisher (PM2-200SA).
Fig. 1. (a) SEM images of Cu–Ag powders with [SC]/[Ag] = 0/1, (b) 0.07/1, (c) 0.27/1, (d) 1.09/1, (e) 4.35/1.
40
Y.-h. Peng et al. / Applied Surface Science 263 (2012) 38–44
2.4. Characterization The morphologies of the synthetic powders were observed by SEM (SEM, JEOL JSM-7001F, Japan). The phase composition of the prepared powders was determined by X-ray diffraction (XRD, DX2500, Mainland China) using Cu K␣ radiation ( = 0.1542 nm). FTIR spectra were collected using an JASCO FTIR-410 infrared spectrometer. The X-ray photoelectron spectroscopy (XPS) measurements were obtained using a PHI 5000 VersaProbe spectrometer equipped
with an Al K␣ X-ray source for chemical states and elemental composition. The energy dispersed spectrometry (EDS) of SEM (samples of [SC]/[Cu] = 0.0078 and [SC]/[Ag] = 0.07, [SC]/[Cu] = 0.0078 and [SC]/[Ag] = 0.27) and inductively coupled plasma-optical emission spectrometer (ICP-OES) were also used for the element identification of the products. The oxidation resistance of synthetic powders was tested by measuring sheet resistance of metal films which changed with the various conditions of thermal treatment in air ambient. The sheet resistances of metal films were measured by
Fig. 2. SEM-EDS image of Cu–Ag powders on (a) spherical particles, (b) hexagonal particles.
Y.-h. Peng et al. / Applied Surface Science 263 (2012) 38–44
41
Fig. 3. (a) SEM cross-sectional image of Cu–Ag powders with [SC]/[Ag] = 0.07/1, (b) EDS analysis of Cu–Ag powders with [SC]/[Ag] = 0.07/1, (c) SEM cross-section image of Cu–Ag powders with [SC]/[Ag] = 0.27/1, (d) EDS analysis of Cu–Ag powders with [SC]/[Ag] = 0.27/1.
a programmable milli-ohm meter (ABM-3245) on samples which have 10 mm × 10 mm metal film patterns.
powders were more completely covered by Ag at the molar ratios of [SC]/[Ag] = 0.07/1 and 0.27/1. 3.2. Characterization of Cu–Ag composite powders
3. Results and discussion 3.1. Effects of [SC]/[Ag] ratio on Cu–Ag composite powders Fig. 1 shows the SEM images of Cu–Ag composite powders with different [SC]/[Ag] molar ratios. It can be seen in Fig. 1a that the smaller Ag particles with dendritic morphology were already separated from Cu particles if SC was not present. In Fig. 1b–e, two kinds of particles are shown in the images when SC was added: hexagonal particles and spherical ones. It can be seen that the diameters of synthetic powders are in the 100–1000 nm range with [SC]/[Ag] = 0.07/1 and 0.27/1. The powders become non-uniform with increasing the [SC]/[Ag] molar ratio. In Fig. 1d and e, the amount of hexagonal particles increased and the number of spherical particles decreased. Further analysis by SEM-EDS as shown in Fig. 2. Fig. 2a shows the EDS analysis that the spherical particles contain 55.21 wt% of Cu, 40.15 wt% of Ag, and Fig. 2b shows that the hexagonal particles contain 91.84 wt% of Cu, 3.97 wt% of Ag. Both of two kinds of particles were contain silver, copper, small amount of oxygen and carbon, which were possibly introduced from three sources: (1) the adsorption of C6 H5 O7 − on synthetic powders surface, (2) oxidation of synthetic powders, or (3) background value from carbon conductive tape. The reason will be discussed later. Thus, synthetic
To ascertain the Cu–Ag core–shell structure of synthetic powders, the SEM cross-sectional image and EDS analysis were presented in Fig. 3. In Fig. 3a–d which have strong intensity of silver on the surface of synthetic powder, and strong intensity of copper on the core of synthetic powder. It can be determined that the synthetic powders with the molar ratios of [SC]/[Ag] = 0.07, [SC]/[Ag] = 0.27 have Cu–Ag core–shell structure. Furthermore, in Fig. 3a–d, the intensities of carbon and oxygen also detected in internal area of Cu–Ag powders and it can be determined as the background value from epoxy resin. XRD analysis was performed to determine the composition and crystal structure of the synthetic powders. Fig. 4 shows the XRD spectra of pure Cu, pure Ag, and synthetic powders. In Fig. 4, both Cu and Ag phases were clearly identified: Cu (1 1 1) orientation at 43.316◦ , Cu (2 0 0) at 50.448◦ , Cu (2 2 0) at 74.124◦ and Ag (1 1 1) orientation at 38.114◦ , Ag (2 2 0) at 64.441◦ , Ag (3 1 1) at 77.395◦ (Cu: JCPDS 04-0836, Ag: JCPDS 65-2871). Fig. 4c–g shows that the presence of the Cu2 O phase was not observed in the XRD patterns that can be determined our synthetic powders was not oxidative. Further, It can be seen in Fig. 4c that strong Ag (1 1 1) and Cu (1 1 1) peaks were identified when the molar ratio of [SC]/[Ag] = 0. In Fig. 4d–g, the peaks of Ag (1 1 1) and Cu (1 1 1) both decreases within the molar ratio of [SC]/[Ag] = 0.07/1 and 4.35/1.
42
Y.-h. Peng et al. / Applied Surface Science 263 (2012) 38–44
Fig. 4. XRD patterns of powders with different [SC]/[Ag] ratios.
Furthermore, the lowest Cu (1 1 1) peak was detected when the molar ratio of [SC]/[Ag] = 0.07/1, and then the Cu (1 1 1) peak increased when increasing the molar ratio of [SC]/[Ag]. This phenomenon can be explained by adding SC which helps to form a Ag coating on the Cu powders surface. Thus, when the Ag coating is dense and sufficient, the intensity of Cu (1 1 1) peak can be decreased. Consequently, the Ag coating can be more completely formed on Cu powders with the molar ratio of [SC]/[Ag] = 0.07/1. To ascertain the surface adsorption of SC, the Fourier transform infrared (FTIR) spectra were recorded on Cu–Ag particles as depicted in Fig. 5. For comparison, the spectrum recorded for neat SC is also superimposed in the same figure. In Fig. 5a, symmetric (s ) and asymmetric (as ) vibrations respectively emerged at the ranges of 1440–1350 cm−1 and 1650–1550 cm−1 which correspond to a characteristic stretching vibration of COO− groups. The bands between 2400–2375 cm−1 and 3550–3300 cm−1 , respectively correspond to a characteristic stretching vibration of C O groups (from ambient air) and OH− groups (H2 O). Fig. 5b shows the presence of bands at 1410 and 1575 cm−1 which indicated C6 H5 O7 − have existed on the particles surface. XPS was also performed to study the surface information of Cu–Ag powders. Fig. 6a shows the intensity of Ag peak is higher than Cu, which clearly shows that Ag is on the surface of the synthetic powders. This result is in agreement with SEM and XRD
Fig. 6. (a) XPS spectra of Cu–Ag powders, (b) XPS spectra of Ag, and (c) XPS spectra of Cu.
Fig. 5. FTIR spectra of (a) pure SC and (b) Cu–Ag powders with the molar ratio of [SC]/[Ag] = 0.07/1.
findings. Fig. 6b shows that Ag 3d5/2 (368.276 eV) and Ag 3d3/2 (374.276 eV) peaks are associated with zero-valent Ag at 368 and 374 eV, respectively. The good symmetry of Ag peaks proves that the valence of Ag does not change. In Fig. 6c, the major contributions of the Cu 2p3/2 (932.676 eV) and Cu 2p1/2 (952.876 eV) peaks are associated with zero-valent Cu. From the aforementioned experimental results which indicate the synthetic powders with Cu–Ag core–shell structure and without any impurity phases were observed.
Y.-h. Peng et al. / Applied Surface Science 263 (2012) 38–44
Fig. 7. Resistivity variations of printed patterns as function of heat-treatment temperature. The pattern with a size of 10 mm × 10 mm were screen printed with 50 m thickness and head for 30 min in air ambient.
3.3. Electrical properties of Cu–Ag powders A series of electric performance tests were carried out on Cu–Ag powders with different [SC]/[Ag] ratios. Heat-treatment in air ambient of printed patterns are necessary, because it will remove the solvent and cause sintering between particles. Adjusting of the heat-treatment temperature of the printed patterns were exhibited in Fig. 7. As the result shown that the resistivity decreases within the increasing temperature between 50 ◦ C and 250 ◦ C, which were possibly introduced from three sources: (1) solvent is persistently removed from the metal film, (2) when the temperature is higher than 175 ◦ C that the adsorped C6 H5 O7 − on synthetic powders surface is decomposed and it can help to reduce the resistivity, (3) the conduction path between particles has established by inter-particle neck formation of small Ag particles. Fig. 8b shows the particle size of Ag particles which on the surface of Cu–Ag powders are between 20 and 100 nm. In Fig. 8c
43
Fig. 9. XRD patterns of Cu–Ag powders after 250 ◦ C for 30 min heat-treatment.
which can be clearly seen the inter-particle neck formation of small Ag particles after 250 ◦ C for 30 min heat-treatment. It is considered that the melting point of the bulk silver is 961.78 ◦ C and then the result shows a sintering effect of nano-sized silver particles. Besides, the resistivity of Cu–Ag powders with the molar ratios of [SC]/[Ag] = 0.07/1 (2.75 × 10−4 cm), [SC]/[Ag] = 0.27/1 (7.5 × 10−4 cm), [SC]/[Ag] = 1.09/1 (6.25 × 10−4 cm), −4 [SC]/[Ag] = 4.35/1 (4.5 × 10 cm) are almost the same as spherical (2.55 × 10−4 cm) and flake silver powders (2.5 × 10−4 cm). Otherwise, the resistivity of Cu–Ag powders with the molar ratio of [SC]/[Ag] = 0/1 (1.25 × 10−3 cm) are still ten times higher than silver powders. Furthermore, after heat-treatment temperature at 350 ◦ C for 30 min, the resistivity of Cu–Ag powders with the molar ratio of [SC]/[Ag] = 0.27/1 is increased. Therefore, there may be the copper of Cu–Ag powders with the molar ratios of [SC]/[Ag] = 0/1, 0.27/1, 1.09/1 and 4.35/1 are oxidized because of insufficient silver coating. Fig. 9a, c–e shows the presence of the Cu2 O and CuO phase was observed in the XRD patterns, indicating that Cu was partially
Fig. 8. (a) SEM images of Cu–Ag powders with [SC]/[Ag] = 0.07/1, (b) enlarged Cu–Ag powders in selected area of (a), (c) inter-particle neck formation of Ag after 250 ◦ C heat-treatment.
44
Y.-h. Peng et al. / Applied Surface Science 263 (2012) 38–44
oxidized at the molar ratios of [SC]/[Ag] = 0/1, 0.27/1, 1.09/1 and 4.35/1. Thus, after 250 or 350 ◦ C for 30 min heat-treatment, Cu–Ag powders with the molar ratio of [SC]/[Ag] = 0.07/1 has almost same resistivity (1.425 × 10−4 cm) as spherical (1.43 × 10−4 cm) and flake silver powders (1.425 × 10−4 cm), and only Cu and Ag phase were observed. 4. Conclusions Ag-coated Cu powders were successfully fabricated by a simple and green electroless plating technique. The Ag coating on Cu surface was improved by adding SC. SEM cross-sectional image and EDS analysis which indentify the Cu–Ag composite powders do have clear core–shell structure. In the synthesized process, hindrance of SC played an important role in avoiding aggregation of Cu and synthetic Cu–Ag powders through COOH bonding with powders. As the molar ratio of [SC]/[Ag] = 0.07/1, a dense Ag coating was obtained on the Cu surface. After 350 ◦ C for 30 min heat-treatment, the synthetic powders has almost same resistivity (1.425 × 10−4 cm) as spherical (1.43 × 10−4 cm) and flake silver powders (1.425 × 10−4 cm), and no Cu2 O and CuO phase were observed. Acknowledgements The financial support for this research was provided by the Industrial Development Bureau, Ministry of Economic Affairs of the Republic of China (Taiwan) under Grant no. M10000121222 in the Conventional Industry Technology Development (CITD) program.
References [1] B.K. Park, D. Kim, S. Jeong, J. Moon, J.S. Kim, Thin Solid Films 515 (2007) 7706. [2] Y. Lee, J. Choi, K.J. Lee, N.E. Stott, D. Kim, Nanotechnology 19 (2008) 415604. [3] N.A. Luechinger, E.K. Athanassiou, W.J. Stark, Nanotechnology 19 (2008) 445201. [4] E.K. Athanassiou, R.N. Grass, W.J. Stark, Nanotechnology 17 (2006) 1668. [5] B.C. Ranu, A. Saha, R. Jana, Advanced Synthesis and Catalysis 349 (2007) 2690. [6] A. Sarkar, T. Mukherjee, S. Kapoor, Journal of Physical Chemistry C 112 (2008) 3334. [7] H.T. Hai, J.G. Ahn, D.J. Kim, et al., Surface and Coatings Technology 201 (2006) 3788. [8] E. Takeshima, K. Takatsu, Y. Kojima, T. Fujji, US Patent No. 4,954,235 (1990). [9] N. Feldstein, US Patent No. 4,305,997 (1981). [10] J. Kondo, K. Murakawa, K. Nomoto, F. Ishikawa, N. Ishida, J. Ishikawa, US Patent No. 4,956,014 (1990). [11] B. Wang, Z. Ji, F.T. Zimone, G.M. Janowski, J.M. Rigsbee, Surface and Coatings Technology 91 (1997) 64. [12] A. Vaskelis, E. Matulionis, E. Norkus, A. Jagminiene, R. Juskenas, Surface and Coatings Technology 82 (1996) 165. [13] V. Mancier, C. Rousse-Bertrand, J. Dille, J. Michel, Ultrasonics Sonochemistry 17 (2010) 690. [14] X. Xu, X. Luo, H. Zhuang, W. Li, B. Zhang, Materials Letters 57 (2003) 3987. [15] Z. Yang, H. Qian, H. Chen, J.N. Anker, Journal of Colloid and Interface Science 352 (2010) 285. [16] M.A. Smiddy, J.-E.G.H. Martin, A.L. Kelly, C.G. Kruif, T. Huppertz, Journal of Dairy Science 89 (2006) 1906. [17] M. Teramoto, Y. Sakaida, S.S. Fu, N. Ohnishi, H. Matsuyama, T. Maki, T. Fukui, K. Arai, Separation and Purification Technology 21 (2000) 137. [18] X. Ma, H. Li, G. Zhua, L. Kanga, Z.H. Liua, Colloids and Surfaces A: Physicochemical and Engineering Aspects 371 (2010) 71. [19] P.S. Gils, D. Ray, P.K. Sahoo, International Journal of Biological Macromolecules 46 (2010) 237. [20] S.C. Boca, C. Farcau, S. Astilean, Nuclear Instruments and Methods in Physics Research Section B 267 (2009) 406. [21] M. Kumar, G.B. Reddy, Physica E 42 (2010) 1940. ´ Bioinorganic Chemistry and Applications (2008) 436458. [22] S. Djokic, [23] Z.S. Pallai, P.V. Kamat, Journal of Physical Chemistry B 108 (2004) 945.
Contents lists available at SciVerse ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Study on synthesis of ultrafine Cu–Ag core–shell powders with high electrical conductivity Yu-hsien Peng a,b,c , Chih-hao Yang b , Kuan-ting Chen e,∗ , Srinivasa R. Popuri d , Ching-Hwa Lee a , Bo-Shin Tang b a
Department of Environmental Engineering, Dayeh University, 168 University Rd., Dacun, Changhua 515, Taiwan, ROC Department of Research & Development, Oriental Happy Enterprise Co., No. 27, Xin’ai Rd., South Dist., Tainan 702, Taiwan, ROC c Center for General Education, Kun Shan University, No. 949, Dawan Rd., Yongkang Dist., Tainan 710, Taiwan, ROC d Department of Biological & Chemical Sciences, The University of the West Indies, Cave Hill Campus 11000, Barbados e Department of Resources Engineering, National Cheng Kung University, No.1, Da-Hsueh Road, Tainan 701, Taiwan, ROC b
a r t i c l e
i n f o
Article history: Received 21 February 2012 Received in revised form 8 August 2012 Accepted 18 August 2012 Available online 25 August 2012 Keywords: Composite materials Fourier transform infrared spectroscopy X-ray diffraction topography Electrical conductivity
a b s t r a c t Cu–Ag composite powders with high electrical conductivity were synthesized by electroless plating of silver sulfate, copper powders with eco-friendly sodium citrate as reducing agent, dispersant and chelating agent in an aqueous system. The influences of sodium citrate/Ag ratio on Ag coatings of Cu powders were investigated. Ag was formed a dense coating on the surface of Cu powders at a molar ratio of sodium citrate/Ag = 0.07/1. SEM showed an uniformity of Ag coatings on Cu powders. SEM-EDX also revealed that Cu cores were covered by Ag shells on the whole. The surface composition analysis by XPS indicated that without Cu or Ag atoms in the surface were oxidized. The resistivity measurements of Cu–Ag paste shows that they have closer resistivity as the pure silver paste’s after 250 ◦ C for 30 min heat-treatment (2.55 × 10−4 cm) and 350 ◦ C for 30 min heat-treatment (1.425 × 10−4 cm). © 2012 Elsevier B.V. All rights reserved.
1. Introduction Copper and silver are widely used in the electronic industry [1–3], sensors [4], catalysis [5,6], and optical or biological devices due to their high electrical conductivity. According to the rare content of silver in the earth’s crust and easy oxidization of copper strongly constrain their applications. Thus Cu–Ag core–shell powders instead of single copper or silver powders is currently considered to be a desirable option in electronic industry to solve the above problems [7]. Generally, powders with core–shell structure can be achieved by electroplating [8], electroless plating (also called spontaneous displacement reaction) [9,10], and vacuum process (evaporation, sputtering [11], etc.). It is well known that silver can be deposited on many substrates by electroplating [12]. However, the efficiency of electroplating and vacuum coating for powders is very low. Therefore, electroplating and vacuum processes are not feasible for commercial purpose depending on the efficiency for particles
Abbreviations: SC, sodium citrate. ∗ Corresponding author at: 168 University Rd., Dacun, Changhua 515, Taiwan, ROC. Tel.: +886 4 851 1338/886 6 261 8328; fax: +886 6 2616003. E-mail addresses: [email protected], [email protected] (K.-t. Chen). 0169-4332/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.apsusc.2012.08.066
which is extremely low. The spontaneous displacement reaction of silver coating that can be achieved on copper powder and has a higher deposition rate than electroplating, although there are still difficulties in reducing process steps and form a well silver coating on the surface of copper powder. According to the literature, the rate of replacement reaction is quite important and has great impact on the final product. Mancier et al. [13] prepared the Cu–Ag core–shell nanopowders by ultrasound-assisted electrochemistry followed by a displacement reaction. Ethylenediaminetetraacetic acid (EDTA) was used as the chelating agent to control the rate of Ag–Cu displacement. Xu et al. [14] reported a synthesis of silver coating process on fine copper powder via Cu displace with silver nitrate/ammonium hydroxide solution. The quantity of ammonia controlled the rate of copper dissolution. In the aforementioned references, many of these synthetic methods are unable for mass-production scale due to their low efficiency or the inadequancy of precious equipment. Furthermore, in most of the processes, sodium borohydride (NaBH4 ), hydrazine hydrate (N2 H4 ·H2 O) were used as reducing agent, hexadecyltrimethyl ammonium bromide (CTAB), polyvinylpyrrolidone (PVP) and EDTA, etc. were used as dispersant and chelating agent, respectively, which are hard to be removed from the final products or poisonous. Recently, sodium citrate (Na3 C6 H5 O7 ) was found to be more widely used for replacing the above chemicals as multifunctional agent [15–18], such as a reducing agent,
Y.-h. Peng et al. / Applied Surface Science 263 (2012) 38–44
39
Table 1 Proportions and compositions of composite powders. Sample
[SC]/[Cu]
Cu powders (g)
SC for dispersant (g)
[Cu]/[Ag+ ] molar ratio
0.015 M Ag2 SO4 (ml)
[SC]/[Ag+ ] molar ratio
SC for chelating (g)
1 2 3 4 5
0.0078:1 0.0078:1 0.0078:1 0.0078:1 0.0078:1
2.53 2.53 2.53 2.53 2.53
0.09 0.09 0.09 0.09 0.09
5.3:1 5.3:1 5.3:1 5.3:1 5.3:1
250 250 250 250 250
0:1 0.07:1 0.27:1 1.09:1 4.35:1
0 0.15 0.6 2.4 9.6
dispersant, chelating agent, etc., and it also suppress precipitation of copper hydroxides in Cu–Ag spontaneous displacement reaction [14,15,18–21]. In this article, we report a simple and novel method to fabricate Cu–Ag core–shell powders using copper powders, silver sulfate, SC (sodium citrate) as starting materials at room temperature. As compared with the results reported previously, this method is a relatively green synthetic method because the starting materials come from SC which serve as mutifuntional agents such as reducing agent, dispersant and chelating agent, and are nontoxic and easily available. Furthermore, the Cu–Ag powders prepared by this method are hardly oxidized, which have greet potential for application in many fields. 2. Experimental 2.1. Materials Copper powders with primary size less than 2 m (median size measured by Laser Diffraction Particle Size Analyzer is 7.36 m) was supported from Oriental Happy Enterprise Co., Ltd., Taiwan, silver sulphate (Ag2 SO4 ) was from Solar Technology Inc., Taiwan, sodium citrate (Na3 C6 H5 O7 ) was from Union Chemical Works Ltd., Taiwan, flake and spherical Ag powders with median size 6.1 m, 5.2 m, respectively were from Metalor Technologies, Taiwan. All chemicals were used as received without further purification. 2.2. Preparation of Cu–Ag powders The chemical reactions of copper powders, sodium citrate (SC) and Ag2 SO4 systems should occurred as following reactions (1)–(4) [12,22,23]. C6 H5 O7 3− + 2Ag+ → C5 H4 O5 2− + H+ + CO2 + 2Ag
(1)
(Ag)2 + + citrate− ↔ [(Ag)2 + · · ·(citrate)− ] +
Cu + 2Ag ↔ 2Ag + Cu
2+
3Cu2+ + 2C6 H5 O7 3− ↔ Cu3 (C6 H5 O7 )2
(2) (3) (4)
During the preparation of Cu–Ag powders, two solutions denoted as A and B were first prepared. For preparing solution A, 2.53 g Cu powders, 0.09 g SC and 100 ml deionized water were mixed at room temperature by a magnetic stirrer; for preparing solution B, 1.17 g Ag2 SO4 ,various amounts of SC and 250 ml deionized water were mixed to form [(Ag)2 + · · ·(citrate)− ]. Detailed experiment design was shown in Table 1. First, solution B was quickly dropped into solution A with a magnetic stirrer at 1000 rpm. After the color of reacting solution from red to purple and then dark green, the reaction was ended by stopping the motion of the magnetic stirrer which was immediately followed by the filtration and clean processes. The obtained composite powders after single step filtration were washed at least 3 times by deionized water, and then dried. 2.3. Preparation of metal paste and sample for SEM cross-sectional image To ascertain the oxidation resistance of synthetic powders, the resistivity () of synthetic powders were investigated, the prepared powders was mixed with an appropriate amount of ethyl acetate–terpineol solution by a three-roll mill. The paste was grounded to the required consistency, and screened on to the glass substrates for forming metal films, and then dried at various conditions in air ambient. To determine the structure of synthetic powders, the sample of synthetic powders for SEM cross-sectional image were fabricated. The synthetic powders was firstly added into epoxy resin. After solidification, the resin which contain synthetic powders were polished by Grinder and Polisher (PM2-200SA).
Fig. 1. (a) SEM images of Cu–Ag powders with [SC]/[Ag] = 0/1, (b) 0.07/1, (c) 0.27/1, (d) 1.09/1, (e) 4.35/1.
40
Y.-h. Peng et al. / Applied Surface Science 263 (2012) 38–44
2.4. Characterization The morphologies of the synthetic powders were observed by SEM (SEM, JEOL JSM-7001F, Japan). The phase composition of the prepared powders was determined by X-ray diffraction (XRD, DX2500, Mainland China) using Cu K␣ radiation ( = 0.1542 nm). FTIR spectra were collected using an JASCO FTIR-410 infrared spectrometer. The X-ray photoelectron spectroscopy (XPS) measurements were obtained using a PHI 5000 VersaProbe spectrometer equipped
with an Al K␣ X-ray source for chemical states and elemental composition. The energy dispersed spectrometry (EDS) of SEM (samples of [SC]/[Cu] = 0.0078 and [SC]/[Ag] = 0.07, [SC]/[Cu] = 0.0078 and [SC]/[Ag] = 0.27) and inductively coupled plasma-optical emission spectrometer (ICP-OES) were also used for the element identification of the products. The oxidation resistance of synthetic powders was tested by measuring sheet resistance of metal films which changed with the various conditions of thermal treatment in air ambient. The sheet resistances of metal films were measured by
Fig. 2. SEM-EDS image of Cu–Ag powders on (a) spherical particles, (b) hexagonal particles.
Y.-h. Peng et al. / Applied Surface Science 263 (2012) 38–44
41
Fig. 3. (a) SEM cross-sectional image of Cu–Ag powders with [SC]/[Ag] = 0.07/1, (b) EDS analysis of Cu–Ag powders with [SC]/[Ag] = 0.07/1, (c) SEM cross-section image of Cu–Ag powders with [SC]/[Ag] = 0.27/1, (d) EDS analysis of Cu–Ag powders with [SC]/[Ag] = 0.27/1.
a programmable milli-ohm meter (ABM-3245) on samples which have 10 mm × 10 mm metal film patterns.
powders were more completely covered by Ag at the molar ratios of [SC]/[Ag] = 0.07/1 and 0.27/1. 3.2. Characterization of Cu–Ag composite powders
3. Results and discussion 3.1. Effects of [SC]/[Ag] ratio on Cu–Ag composite powders Fig. 1 shows the SEM images of Cu–Ag composite powders with different [SC]/[Ag] molar ratios. It can be seen in Fig. 1a that the smaller Ag particles with dendritic morphology were already separated from Cu particles if SC was not present. In Fig. 1b–e, two kinds of particles are shown in the images when SC was added: hexagonal particles and spherical ones. It can be seen that the diameters of synthetic powders are in the 100–1000 nm range with [SC]/[Ag] = 0.07/1 and 0.27/1. The powders become non-uniform with increasing the [SC]/[Ag] molar ratio. In Fig. 1d and e, the amount of hexagonal particles increased and the number of spherical particles decreased. Further analysis by SEM-EDS as shown in Fig. 2. Fig. 2a shows the EDS analysis that the spherical particles contain 55.21 wt% of Cu, 40.15 wt% of Ag, and Fig. 2b shows that the hexagonal particles contain 91.84 wt% of Cu, 3.97 wt% of Ag. Both of two kinds of particles were contain silver, copper, small amount of oxygen and carbon, which were possibly introduced from three sources: (1) the adsorption of C6 H5 O7 − on synthetic powders surface, (2) oxidation of synthetic powders, or (3) background value from carbon conductive tape. The reason will be discussed later. Thus, synthetic
To ascertain the Cu–Ag core–shell structure of synthetic powders, the SEM cross-sectional image and EDS analysis were presented in Fig. 3. In Fig. 3a–d which have strong intensity of silver on the surface of synthetic powder, and strong intensity of copper on the core of synthetic powder. It can be determined that the synthetic powders with the molar ratios of [SC]/[Ag] = 0.07, [SC]/[Ag] = 0.27 have Cu–Ag core–shell structure. Furthermore, in Fig. 3a–d, the intensities of carbon and oxygen also detected in internal area of Cu–Ag powders and it can be determined as the background value from epoxy resin. XRD analysis was performed to determine the composition and crystal structure of the synthetic powders. Fig. 4 shows the XRD spectra of pure Cu, pure Ag, and synthetic powders. In Fig. 4, both Cu and Ag phases were clearly identified: Cu (1 1 1) orientation at 43.316◦ , Cu (2 0 0) at 50.448◦ , Cu (2 2 0) at 74.124◦ and Ag (1 1 1) orientation at 38.114◦ , Ag (2 2 0) at 64.441◦ , Ag (3 1 1) at 77.395◦ (Cu: JCPDS 04-0836, Ag: JCPDS 65-2871). Fig. 4c–g shows that the presence of the Cu2 O phase was not observed in the XRD patterns that can be determined our synthetic powders was not oxidative. Further, It can be seen in Fig. 4c that strong Ag (1 1 1) and Cu (1 1 1) peaks were identified when the molar ratio of [SC]/[Ag] = 0. In Fig. 4d–g, the peaks of Ag (1 1 1) and Cu (1 1 1) both decreases within the molar ratio of [SC]/[Ag] = 0.07/1 and 4.35/1.
42
Y.-h. Peng et al. / Applied Surface Science 263 (2012) 38–44
Fig. 4. XRD patterns of powders with different [SC]/[Ag] ratios.
Furthermore, the lowest Cu (1 1 1) peak was detected when the molar ratio of [SC]/[Ag] = 0.07/1, and then the Cu (1 1 1) peak increased when increasing the molar ratio of [SC]/[Ag]. This phenomenon can be explained by adding SC which helps to form a Ag coating on the Cu powders surface. Thus, when the Ag coating is dense and sufficient, the intensity of Cu (1 1 1) peak can be decreased. Consequently, the Ag coating can be more completely formed on Cu powders with the molar ratio of [SC]/[Ag] = 0.07/1. To ascertain the surface adsorption of SC, the Fourier transform infrared (FTIR) spectra were recorded on Cu–Ag particles as depicted in Fig. 5. For comparison, the spectrum recorded for neat SC is also superimposed in the same figure. In Fig. 5a, symmetric (s ) and asymmetric (as ) vibrations respectively emerged at the ranges of 1440–1350 cm−1 and 1650–1550 cm−1 which correspond to a characteristic stretching vibration of COO− groups. The bands between 2400–2375 cm−1 and 3550–3300 cm−1 , respectively correspond to a characteristic stretching vibration of C O groups (from ambient air) and OH− groups (H2 O). Fig. 5b shows the presence of bands at 1410 and 1575 cm−1 which indicated C6 H5 O7 − have existed on the particles surface. XPS was also performed to study the surface information of Cu–Ag powders. Fig. 6a shows the intensity of Ag peak is higher than Cu, which clearly shows that Ag is on the surface of the synthetic powders. This result is in agreement with SEM and XRD
Fig. 6. (a) XPS spectra of Cu–Ag powders, (b) XPS spectra of Ag, and (c) XPS spectra of Cu.
Fig. 5. FTIR spectra of (a) pure SC and (b) Cu–Ag powders with the molar ratio of [SC]/[Ag] = 0.07/1.
findings. Fig. 6b shows that Ag 3d5/2 (368.276 eV) and Ag 3d3/2 (374.276 eV) peaks are associated with zero-valent Ag at 368 and 374 eV, respectively. The good symmetry of Ag peaks proves that the valence of Ag does not change. In Fig. 6c, the major contributions of the Cu 2p3/2 (932.676 eV) and Cu 2p1/2 (952.876 eV) peaks are associated with zero-valent Cu. From the aforementioned experimental results which indicate the synthetic powders with Cu–Ag core–shell structure and without any impurity phases were observed.
Y.-h. Peng et al. / Applied Surface Science 263 (2012) 38–44
Fig. 7. Resistivity variations of printed patterns as function of heat-treatment temperature. The pattern with a size of 10 mm × 10 mm were screen printed with 50 m thickness and head for 30 min in air ambient.
3.3. Electrical properties of Cu–Ag powders A series of electric performance tests were carried out on Cu–Ag powders with different [SC]/[Ag] ratios. Heat-treatment in air ambient of printed patterns are necessary, because it will remove the solvent and cause sintering between particles. Adjusting of the heat-treatment temperature of the printed patterns were exhibited in Fig. 7. As the result shown that the resistivity decreases within the increasing temperature between 50 ◦ C and 250 ◦ C, which were possibly introduced from three sources: (1) solvent is persistently removed from the metal film, (2) when the temperature is higher than 175 ◦ C that the adsorped C6 H5 O7 − on synthetic powders surface is decomposed and it can help to reduce the resistivity, (3) the conduction path between particles has established by inter-particle neck formation of small Ag particles. Fig. 8b shows the particle size of Ag particles which on the surface of Cu–Ag powders are between 20 and 100 nm. In Fig. 8c
43
Fig. 9. XRD patterns of Cu–Ag powders after 250 ◦ C for 30 min heat-treatment.
which can be clearly seen the inter-particle neck formation of small Ag particles after 250 ◦ C for 30 min heat-treatment. It is considered that the melting point of the bulk silver is 961.78 ◦ C and then the result shows a sintering effect of nano-sized silver particles. Besides, the resistivity of Cu–Ag powders with the molar ratios of [SC]/[Ag] = 0.07/1 (2.75 × 10−4 cm), [SC]/[Ag] = 0.27/1 (7.5 × 10−4 cm), [SC]/[Ag] = 1.09/1 (6.25 × 10−4 cm), −4 [SC]/[Ag] = 4.35/1 (4.5 × 10 cm) are almost the same as spherical (2.55 × 10−4 cm) and flake silver powders (2.5 × 10−4 cm). Otherwise, the resistivity of Cu–Ag powders with the molar ratio of [SC]/[Ag] = 0/1 (1.25 × 10−3 cm) are still ten times higher than silver powders. Furthermore, after heat-treatment temperature at 350 ◦ C for 30 min, the resistivity of Cu–Ag powders with the molar ratio of [SC]/[Ag] = 0.27/1 is increased. Therefore, there may be the copper of Cu–Ag powders with the molar ratios of [SC]/[Ag] = 0/1, 0.27/1, 1.09/1 and 4.35/1 are oxidized because of insufficient silver coating. Fig. 9a, c–e shows the presence of the Cu2 O and CuO phase was observed in the XRD patterns, indicating that Cu was partially
Fig. 8. (a) SEM images of Cu–Ag powders with [SC]/[Ag] = 0.07/1, (b) enlarged Cu–Ag powders in selected area of (a), (c) inter-particle neck formation of Ag after 250 ◦ C heat-treatment.
44
Y.-h. Peng et al. / Applied Surface Science 263 (2012) 38–44
oxidized at the molar ratios of [SC]/[Ag] = 0/1, 0.27/1, 1.09/1 and 4.35/1. Thus, after 250 or 350 ◦ C for 30 min heat-treatment, Cu–Ag powders with the molar ratio of [SC]/[Ag] = 0.07/1 has almost same resistivity (1.425 × 10−4 cm) as spherical (1.43 × 10−4 cm) and flake silver powders (1.425 × 10−4 cm), and only Cu and Ag phase were observed. 4. Conclusions Ag-coated Cu powders were successfully fabricated by a simple and green electroless plating technique. The Ag coating on Cu surface was improved by adding SC. SEM cross-sectional image and EDS analysis which indentify the Cu–Ag composite powders do have clear core–shell structure. In the synthesized process, hindrance of SC played an important role in avoiding aggregation of Cu and synthetic Cu–Ag powders through COOH bonding with powders. As the molar ratio of [SC]/[Ag] = 0.07/1, a dense Ag coating was obtained on the Cu surface. After 350 ◦ C for 30 min heat-treatment, the synthetic powders has almost same resistivity (1.425 × 10−4 cm) as spherical (1.43 × 10−4 cm) and flake silver powders (1.425 × 10−4 cm), and no Cu2 O and CuO phase were observed. Acknowledgements The financial support for this research was provided by the Industrial Development Bureau, Ministry of Economic Affairs of the Republic of China (Taiwan) under Grant no. M10000121222 in the Conventional Industry Technology Development (CITD) program.
References [1] B.K. Park, D. Kim, S. Jeong, J. Moon, J.S. Kim, Thin Solid Films 515 (2007) 7706. [2] Y. Lee, J. Choi, K.J. Lee, N.E. Stott, D. Kim, Nanotechnology 19 (2008) 415604. [3] N.A. Luechinger, E.K. Athanassiou, W.J. Stark, Nanotechnology 19 (2008) 445201. [4] E.K. Athanassiou, R.N. Grass, W.J. Stark, Nanotechnology 17 (2006) 1668. [5] B.C. Ranu, A. Saha, R. Jana, Advanced Synthesis and Catalysis 349 (2007) 2690. [6] A. Sarkar, T. Mukherjee, S. Kapoor, Journal of Physical Chemistry C 112 (2008) 3334. [7] H.T. Hai, J.G. Ahn, D.J. Kim, et al., Surface and Coatings Technology 201 (2006) 3788. [8] E. Takeshima, K. Takatsu, Y. Kojima, T. Fujji, US Patent No. 4,954,235 (1990). [9] N. Feldstein, US Patent No. 4,305,997 (1981). [10] J. Kondo, K. Murakawa, K. Nomoto, F. Ishikawa, N. Ishida, J. Ishikawa, US Patent No. 4,956,014 (1990). [11] B. Wang, Z. Ji, F.T. Zimone, G.M. Janowski, J.M. Rigsbee, Surface and Coatings Technology 91 (1997) 64. [12] A. Vaskelis, E. Matulionis, E. Norkus, A. Jagminiene, R. Juskenas, Surface and Coatings Technology 82 (1996) 165. [13] V. Mancier, C. Rousse-Bertrand, J. Dille, J. Michel, Ultrasonics Sonochemistry 17 (2010) 690. [14] X. Xu, X. Luo, H. Zhuang, W. Li, B. Zhang, Materials Letters 57 (2003) 3987. [15] Z. Yang, H. Qian, H. Chen, J.N. Anker, Journal of Colloid and Interface Science 352 (2010) 285. [16] M.A. Smiddy, J.-E.G.H. Martin, A.L. Kelly, C.G. Kruif, T. Huppertz, Journal of Dairy Science 89 (2006) 1906. [17] M. Teramoto, Y. Sakaida, S.S. Fu, N. Ohnishi, H. Matsuyama, T. Maki, T. Fukui, K. Arai, Separation and Purification Technology 21 (2000) 137. [18] X. Ma, H. Li, G. Zhua, L. Kanga, Z.H. Liua, Colloids and Surfaces A: Physicochemical and Engineering Aspects 371 (2010) 71. [19] P.S. Gils, D. Ray, P.K. Sahoo, International Journal of Biological Macromolecules 46 (2010) 237. [20] S.C. Boca, C. Farcau, S. Astilean, Nuclear Instruments and Methods in Physics Research Section B 267 (2009) 406. [21] M. Kumar, G.B. Reddy, Physica E 42 (2010) 1940. ´ Bioinorganic Chemistry and Applications (2008) 436458. [22] S. Djokic, [23] Z.S. Pallai, P.V. Kamat, Journal of Physical Chemistry B 108 (2004) 945.