Plasma Science and Technology, Vol.16, No.1, Jan. 2014
Preparation of Copper Nanoparticles Using Dielectric Barrier Discharge at Atmospheric Pressure and its Mechanism∗ DI Lanbo (底兰波), ZHANG Xiuling (张秀玲), XU Zhijian (徐志坚) College of Physical Science and Technology, Dalian University, Dalian 116622, China
Abstract
Dielectric barrier discharge (DBD) cold plasma at atmospheric pressure was used for preparation of copper nanoparticles by reduction of copper oxide (CuO). Power X-ray diffraction (XRD) was used to characterize the structure of the copper oxide samples treated by DBD plasma. Influences of H2 content and the treating time on the reduction of copper oxide by DBD plasma were investigated. The results show that the reduction ratio of copper oxide was increased initially and then decreased with increasing H2 content, and the highest reduction ratio was achieved at 20% H2 content. Moreover, the copper oxide samples were gradually reduced by DBD plasma into copper nanoparticles with the increase in treating time. However, the average reduction rate was decreased as a result of the diffusion of the active hydrogen species. Optical emission spectra (OES) were observed during the reduction of the copper oxide samples by DBD plasma, and the reduction mechanism was explored accordingly. Instead of high-energy electrons, atomic hydrogen (H) radicals, and the heating effect, excited-state hydrogen molecules are suspected to be one kind of important reducing agents. Atmospheric-pressure DBD cold plasma is proved to be an efficient method for preparing copper nanoparticles.
Keywords: copper, atmospheric-pressure cold plasma, dielectric barrier discharge (DBD), optical emission spectra (OES), excited-state hydrogen molecules
PACS: 52.70.-m, 52.77.-j DOI: 10.1088/1009-0630/16/1/09 (Some figures may appear in colour only in the online journal)
1
Introduction
ported noble-metal ions using Ar and O2 glow discharge plasmas at low pressure [16] . Jang et al. prepared TiO2 supported Pd and Pd-Ag catalysts by air and H2 radio frequency (RF) cold plasmas at low pressure [17] . However, most of these experiments operated at low pressure with vacuum systems adopted. Dielectric barrier discharge (DBD), which was widely used to generate atmospheric-pressure cold plasma, has been recently adopted for preparing copper nanoparticles [18−23] . It was found that DBD plasma reduction was a fast and facile method for reducing copper ions, in which atomic hydrogen (H) radicals were often thought to be the reducing agents. For example, Sawada et al. found that copper oxide could be reduced at 100 ◦ C by DBD plasma at near atmospheric pressure [19] , and they thought that atomic hydrogen (H) radicals were the reducing agent with the help of vacuum ultraviolet (VUV) line absorption. In our previous work, excited-state hydrogen molecules were suspected to be another efficient kind of species for reducing metal ions. However, to the best of our knowledge, there has been no report on reducing of copper ions by using excited-state hydrogen molecules. In this study, DBD cold plasma was used to reduce copper oxide to prepare copper nanoparticles at atmospheric pressure, and XRD was used to characterize the samples. In order to illustrate the reduction
In recent years, great efforts have been devoted to the synthesis of metal nanoparticles because of their special properties and various applications. Among these metals, inexpensive copper nanoparticles have drawn much research attention due to its potential applications in conductive paste, catalysis, solar cells, and magnetic storage media, and thus have become industrially important materials [1] . Different methods have been developed to synthesize copper nanoparticles, such as chemical reduction, microemulation method, electrolytic synthesis, sol-gel method, vacuum vapor deposition, and so on [2−5] . However, these methods may cause large copper nanoparticles or be not environmentally friendly. Thus, it would be useful if an environmentally friendly and facile reduction method can be available. Cold plasma treatment is a green technique and has recently attracted significant attentions for efficient reduction of metal ions [6−17] . It was found that the metal nanoparticles prepared by cold plasma had smaller particle size, high dispersion, thereby exhibiting high activity and selectivity. Moreover, high-energy electrons were often thought to be the reducing agents in these studies. For example, Liu et al. synthesized supported noble-metal catalysts by reducing sup-
∗ supported by National Natural Science Foundation of China (No. 21173028), the Science and Technology Research Project of Liaoning Provincial Education Department of China (No. L2013464), and the Scientific Research Foundation for the Doctor of Liaoning Province of China (No. 20131004)
41
Plasma Science and Technology, Vol.16, No.1, Jan. 2014
3
mechanism, OES spectra were also observed during the reduction of copper oxide.
3.1
2
Experiment
before ICuO(111) after ICuO(111)
× 100%,
Characterization of the samples
It is well known that the addition of Ar gas into H2 gas may produce more active hydrogen species because of the collision of hydrogen molecules with the energetic metastable Ar atoms (e.g., Ar 3 P0 metastable at 11.55 eV), so the influence of H2 content on the reduction of copper oxide (copper oxide: 0.2 mg · cm−2 ) by DBD plasma was investigated. The XRD spectra and the corresponding reduction ratio of the copper oxide samples treated by DBD plasma at different H2 contents and 4 min treatment were shown in Fig. 1(a) and (b). Fig. 1(a) indicated that the copper oxide samples could not be reduced when pure Ar gas was used as the working gas. However, when the mixture of Ar gas and H2 gas was used as the working gas, the XRD spectra of the copper oxide samples treated by DBD plasma exhibited characteristic diffraction peaks at around 2θ = 43.3◦ and 50.4◦ , which can be indexed as (111) and (200) reflections of the metallic copper (JCPDS No. 04-0836). From Fig. 1(b), it could be seen that the reduction ratio of copper oxide was first increased and then decreased with the increase in H2 content, and the highest reduction ratio was achieved at 20% H2 content. This may be attributed to the fact that an appropriate addition of Ar gas into H2 gas may produce more active hydrogen species, and therefore facilitating the reduction of copper oxide.
Copper oxide (CuO, Shenyang Xinxing Reagent Factory, 99%) was used as received without any further purification. The size of CuO calculated by Scherrer’s formula from the half-maximum peak-width values of the (111) reflections of XRD patterns was ca. 28.1 nm. A certain amount of copper oxide was dispersed in deionized water with intense stirring, grown on microscope slides (26 mm× 35 mm×1.0 mm) by a simple coating method, and then dried at 373 K for 2 h for further investigation. The schematic diagram of DBD plasma device for preparing copper nanoparticles at atmospheric pressure had been described previously [24,25] . A reaction cell made of quartz was placed between the high-voltage and the ground electrodes, both of which were made of stainless steel plates (Φ50 mm). The microscope slides supported copper oxide was placed in the reaction cell, and the discharge gap was 4 mm. All the individual gases used in this work were of high-purity grade (>99.99%). The gas flow rates were adjusted and controlled by a mass flow controller system (SevenStar Co., China). A mixture of Ar and H2 (or N2 ) with a total flow rate of 100 mL·min−1 was used as working gas. The plasma treatment was conducted by applying a 36 kV sine-wave high voltage at a frequency of 14.1 kHz without extra heating. XRD characterization of the copper oxide samples before and after DBD plasma treatment was carried out on a rotating anode X-ray diffractometer (DX-2700, Dandong Haoyuan, China) with graphitemonochromatized Cu Kα1 radiation (λ=1.54178 ˚ A). The crystallite size of the copper nanoparticles synthesized by DBD plasma was calculated by Scherrer’s formula from the half-maximum peak-width values of the (111) reflections of XRD patterns taken for each sample. Copper oxide conversion into copper (η) was defined as follows: η=
Results and discussion
(1)
before after where ICuO(111) and ICuO(111) were the intensity of (111) reflection of copper oxide before and after DBD plasma treatment. Optical emission spectra (OES) during the reduction processes were observed by an AvaSpec-ULS2048 multichannel fiber optic spectrometer (Avantas, Netherlands) in the spectral range of 200-1100 nm. The resolution of AvaSpec-ULS2048 multichannel fiber optic spectrometer was 0.16 nm. The rotational temperature (Trot ) was determined by fitting the spectrum with SPECAIR in the range of 366-382 nm, which corresponds to the ∆ν = −2 rotational band sequence of N2 (C-B). This temperature had been shown to be equal to the rotational temperature of ground-state N2 , which is the thermally equilibrated (translational and rotational) gas temperature (Tg ) [26] .
Fig.1 (a) XRD spectra and (b) the corresponding reduction ratio of the copper oxide samples treated by DBD plasma at different H2 contents. (Copper oxide: 0.2 mg · cm−2 , treatment time: 4 min)
In order to investigate the rate of reduction of copper oxide, XRD spectra of the copper oxide samples treated by DBD plasma at different time instants were observed and the corresponding copper oxide conversions were calculated, as shown in Fig. 2(a) and (b). Obviously, the copper oxide samples were gradually reduced by the DBD plasma into copper nanoparticles with the increase in treating time; however, the average reduction rate was decreased. The following reduction 42
DI Lanbo et al.: Preparation of Copper Nanoparticles Using DBD at Atmospheric Pressure and its Mechanism scheme may exist: the active species transported onto a solid surface initially reduced the copper oxide at the surface. As the reduction proceeded, the interface gradually shifts from the surface into the inner region. The active species diffused into the reduced copper layer, and the consumption of the active species resulted in the decrease in reduction rate. Moreover, the crystallite size of the copper nanoparticles calculated by Scherrer’s formula from the half-maximum peak-width values of the (111) reflections of XRD patterns taken for the sample synthesized by DBD plasma was ca. 35 nm, which was much smaller than that prepared by atmospheric pressure plasma jet (50-100 nm) [27] .
were detected in this work. For example, characteristic peaks corresponding to Hα transition (656.3 nm), which was relative easier to appear, could not be detected (inset in Fig. 4(a)). Thereby, atomic hydrogen radicals were not the reducing agents in this work. It was found that copper oxide could be reduced at the temperature above 500 K in H2 gas [28] . For this reason, the gas temperature (Tg ) after 15 min stabilization time was estimated by fitting the spectrum with SPECAIR in the range of 366-382 nm, as shown in Fig. 4(b). The gas temperature was ca. 480 K, which was below 500 K. In addition, the experiment was performed with an interval of 10 min between two operations. Each treatment just took 4 min. In our previous work [29] , it was found that the gas temperature estimated in Ar+N2 was about the same as that in pure Ar gas. In this work, CuO could not be reduced when pure Ar was used as the working gas. In other words, the copper oxide was not reduced by the heating effect. Instead of high-energy electrons, atomic hydrogen (H) radicals and the heating effect, excited-state hydrogen molecules are now suspected to be one kind of important reducing agents.
Fig.2 (a) XRD spectra of the copper oxide samples treated by DBD plasma at different time instants, and (b) the corresponding copper oxide conversion. (Copper oxide: 0.2 mg·cm−2 , treatment time: 16 min)
3.2
Reduction mechanism
In order to investigate the mechanism of DBD plasma in the reduction of copper oxide at atmospheric pressure, experiments were conducted by using pure Ar, 80%Ar+20%N2 , and 80%Ar+20%H2 as the working gases, respectively, and the corresponding XRD spectra were shown in Fig. 3. Copper oxide was completely reduced into copper nanoparticles by using the mixture of 80%+20%H2 as the working gas; however, it could not be reduced by using pure Ar gas or the mixture of 80%Ar+20%N2 as the working gas. High-energy electrons were thought to be a kind of important active species for reducing metal ions. However, in this study, copper oxide could not be reduced by using pure Ar gas or the mixture of 80%Ar+20%N2 as the working gas. This may be due to the fact that the energy of electrons in this work was not high enough to reduce the copper oxide. In order to further investigate the mechanism of DBD plasma for reduction of copper oxide at atmospheric pressure, OES spectra were observed during the reduction processes by using pure Ar, 80%Ar+20%N2 , and 80%Ar+20%H2 as the working gases, respectively, as shown in Fig. 4(a). Atomic hydrogen (H) radicals were often thought to be the other important kind of active species for reducing metal ions. Interestingly, no peaks corresponding to atomic hydrogen (H) radicals
Fig.3 XRD spectra of the copper oxide samples treated by DBD plasma using (a) pure Ar, (b) 80%Ar+20%N2 , and (c) 80%Ar+20%H2 as the working gases, respectively. (Copper oxide: 0.2 mg·cm−2 , treatment time: 16 min)
Fig.4 OES spectra during the reduction processes by DBD plasma using (a) pure Ar, (b) 80%Ar + 20%N2 , and (c) 80%Ar + 20%H2 as the working gases, respectively
43
Plasma Science and Technology, Vol.16, No.1, Jan. 2014 The possible reactions are illustrated as follows: e∗ + H2 → H∗2 + e ∗
∗
∗
H∗2
(2)
e + Ar → Ar + e Ar + H2 → H∗2
+ Ar
+ CuO → Cu + H2 O
∗
2 3
(3) (4)
4
(5)
5
H∗2
where e is the energetic electron, is the excitedstate hydrogen molecule, Ar∗ is the metastable Ar atom. All in all, the schematic diagram of the reduction mechanism by DBD cold plasma at atmospheric pressure is shown in Fig. 5. Atmospheric-pressure DBD cold plasma has been proved to be an environmentally friendly and efficient method for preparing copper nanoparticles. Different from previous works, excitedstate hydrogen molecules are thought to be another kind of efficient active species for reducing copper oxide. Further work will be done to verify the existence of the excited-state hydrogen molecules, and the relationship between them and the reduction rate is to be disclosed.
6 7 8 9 10 11 12 13 14 15 16 17
Fig.5 The schematic diagram of the mechanism for copper oxide reduction by DBD cold plasma
18 19
4
Conclusions
20
Copper nanoparticles were successfully prepared by reduction of copper oxide (CuO) in atmosphericpressure DBD cold plasma. Influences of H2 content and the treating time on the reduction of copper oxide by DBD plasma were investigated. Analyses of the XRD spectra show that the highest reduction ratio was achieved at 20% H2 content. Moreover, the copper oxide samples were gradually reduced by DBD plasma into copper nanoparticles with the increase in treating time; however, the average reduction rate was decreased as a result of the diffusion of the active hydrogen species. The reduction mechanism of DBD plasma was also explored. Instead of high-energy electrons, atomic hydrogen (H) radicals and the heating effect, excited-state hydrogen molecules are suspected to be one kind of important reducing agents. Atmosphericpressure DBD cold plasma has been proved to be an efficient method for preparing copper nanoparticles.
21 22 23 24 25 26 27 28 29
(Manuscript received 30 August 2013) (Manuscript accepted 29 October 2013) E-mail address of DI Lanbo:
[email protected]
References 1
Wei W, Lu Y, Chen W, et al. 2011, Journal of the American Chemical Society, 133: 2060 Dang T M D, Le T T T, Fribourg-Blanc E, et al. 2011, Advances in Natural Sciences: Nanoscience and Nanotechnology, 2: 015009 Vazquez-Vazquez C, Banobre-Lopez M, Mitra A, et al. 2009, Langmuir, 25: 8208 Theivasanthi T and Alagar M. 2011, International Journal of the Physical Sciences, 6: 3662 He J, Ichinose I, Kunitake T, et al. 2002, Langmuir, 18: 10005 Klein A N, Cardoso R P, Pavanati H C, 2013, Plasma Science and Technology, 15: 70 Liang X, Liu C J, Kuai P. 2008, Green Chemistry, 10: 1318 Wang Z J, Xie Y B, Liu C J. 2008, The Journal of Physical Chemistry C, 112: 19818 Wang H, Liu C. 2011, Applied Catalysis B: Environmental, 106: 672 Liu X, Mou C Y, Lee S, et al. 2012, Journal of Catalysis, 285: 152 Chen Y T, Wang H P, Liu C J, et al. 2012, Journal of Catalysis, 289: 105 Zhou C M, Wang X, Jia X L, et al. 2012, Electrochemistry Communications, 18: 33 Buitrago-Sierra R, Garc´ıa-Fern´ andez M J, Pastor-Blas M M, et al. 2013, Green Chemistry, 15: 1981 Zhang X L, Di L B, Zhou Q. 2013, Journal of Energy Chemistry, 22: 446 Zou J J, Zhang Y P, and Liu C J. 2006, Langmuir, 22: 11388 Shi C K, Hoisington R, and Jang W L B. 2007, Industrial and Engineering Chemistry Research, 46: 4390 Sawada Y, Tamaru H, Kogoma M, et al. 1996, Journal of Physics D: Applied Physics, 29: 2539 Sawada Y, Taguchi N and Tachibana K. 1999, Japanese Journal of Applied Physics, 38: 2539 Inui H, Takeda K, Kondo H, et al. 2010 Applied Physics Express, 3: 126101 Tajima S, Tsuchiya S, Matsumori M, et al. 2011, Thin Solid Films, 519: 6773 Yamamoto Y, Akiyama H, Ooka K, et al. 2012, Current Applied Physics, 12: S63 Zhao P, Zheng W, Meng Y D. 2013, Journal of Applied Physics, 113: 123301 Di L B, Xu Z J, Zhang X L. 2013, Catalysis Today, 211: 143 Di L B, Xu Z J, Wang K, Zhang X L. 2013, Catalysis Today, 211: 109 Donnelly V M, Malyshev M V. 2000, Applied Physics Letters, 77: 2467 Nakahiro H, Zhao P, Ogino A, et al. 2012 Applied Physics Express, 5: 056201 Kim J Y, Rodriguez J A, Hanson J C, et al. 2003, Journal of the American Chemical Society, 125: 10684 Di L B, Li X S, Zhao T L, et al. 2013, Plasma Science and Technology, 15: 64
Galletti A M R, Antonetti C, Marracci M, et al. 2013, Applied Surface Science, 280: 610
44
