Metal–nonmetal transition in the copper–carbon nanocomposite films

doi:10.1016/j.physb.2010.06.035

Physica B: Condensed Matter

Volume 405, Issue 18, 15 September 2010, Pages 3949-3951

https://doi.org/10.1016/j.physb.2010.06.035 Get rights and content

Abstract

We prepared Cu nanoparticles in a-C:H thin films by co-deposition of RF-sputtering and RF-PECVD methods at room temperature. By increasing Cu content in these films a nonmetal–metal (N–M) transition is observed. This transition is explainable by the power law of percolation theory. The critical metal content is obtained 56% and the critical exponent is obtained 1.6, which is larger than the exponent for 2 dimension systems and smaller than the one for 3 dimension systems. The electrical conductivity of dielectric samples was explained by tunneling. Activation tunneling energy that was obtained from temperature dependence of electrical resistivity correlates with near infrared absorption peak of samples and both of them depend on Cu content of thin films. In the early stage of N–M transition, by increasing metal content, a peculiar effect of metallic to nonmetallic state occurs.

Introduction

Transport processes in disordered materials constitute an important class of problems, in view of their relevance to understanding and modeling of a wide variety of phenomena in natural and industrial processes [1].Composites of dielectric and metal exhibit nonmetal–metal (N–M) transition that arise from variation of chemical composition, temperature, stress or magnetic field [2], [3]. The problem of N–M transition above a critical diameter of nanoparticle has been studied too [4]. Two fundamentally different classes of theories have been used to explain the conductivity transition in these systems [5]. In the first class, theories such as percolation have been applied to mixtures which are inhomogeneous on the macroscopic scale. In the second class, quantum theories of localization by various models such as Anderson localization, variable rang hopping and the scaling theory of localization of non-interacting electrons has been developed [5]. Tunneling and percolation in metallic-insulator composite materials films with thickness 2–5 μm was investigated by Toker et.al. They argued that all-connected tunneling network can be reduced to a well-defined percolation network [6]. Dielectric behavior of a metal–polymer composite with low percolation threshold was studied for disks with a thickness about 1.5 mm [7]. Dramatic change in the physical properties of composites occurs when filler particles form a percolating network through the composite, particularly when the difference between the properties of the constitutive phase is large. By use of electric conductivity and dielectric properties, recent studies on the physical properties of composites near percolation are reviewed [8].

In this work, we prepared Cu nanoparticle in carbon thin films with various Cu content by co-deposition of RF-sputtering and RF-PECVD. The difference between the properties of the constitutive phase in our composite is large. Our specific deposition conditions including room temperature and non-wet chemical deposition are prerequisites for applications in optical and electronic devices. Deposition process was reported in details previously [9]. Early study on N–M transition was reported previously too [10]. Here, we explain variation of electrical resistivity versus Cu content for thin films with thickness about 100 nm. By variation of Cu content, N–M transition is observed and is explainable by power law percolation. We argue on resulted percolation threshold and critical exponent. A near infrared absorption for dielectric samples is observed whose energy is comparable with electron activated tunneling energy obtained from temperature dependence of electrical resistivity. The activation energy and near infrared absorption peak are depended on Cu concentration. These composites can be interesting materials for IR detection due to their intense absorption in near infrared region.

Section snippets

Experimental details

Cu nanoparticles in a-C:H thin films were prepared by using acetylene gas and Cu target in a capacitance coupled RF-PECVD system with 13.56 MHz power supply. The reactor consists of two electrodes with different size. The smaller one was Cu plate and was used as powered electrode. The other electrode was grounded via the body of the stainless steel chamber. Deposition was performed at room temperature on the glass and silicon substrates over grounded electrode. The growth was done in constant RF

Results and discussion

The room temperature electrical resistivity versus Cu content of the films is shown in Fig. 1. Three conduction regions are distinguished: dielectric, metallic and in between a transition region. In the dielectric region the Cu content is less than 45%, the electrical resistivity is large and the temperature coefficient of electrical resistivity is negative. In the metallic region, the Cu content is more than 55%, the electrical resistivity is small and the temperature coefficient of electrical

Conclusion

We observed an N–M transition for Cu nanoparticles in carbon films by increasing of Cu concentration. This N–M transition was explained by percolation model. The critical exponent obtained for our data indicates that we have semi-two dimensional random network conduction. The electrical conductivity of the dielectric samples is explainable by tunneling models. Activation tunneling energy that was obtained from temperature dependence of electrical resistivity is comparable with energy of near

Acknowledgments

The authors would like to thank Mr. A. Baghizadeh, Mrs. Vaseghinia for RBS and AFM measurements.

References (15)

A. Mobius et al.
Solid State Mater. Sci.
(1999)
T. Ghodselahi, M.A. Vesaghi, A. Shafiekhani, M. Ahmadi, Proceedings of SPIE, 2008, 6988:...
M. Sahimi
Heterogeneous Materials I, Linear Transport and Optical Properties
(2003)
R. Gomez-Escoto et al.
Rev. Mex. Fis. S
(2007)
K. Chatterjee et al.
J. Appl. Phys.
(2004)
K. Epstein et al.
Phys. Rev. B
(1983)
D. Toker et al.
Phys. Rev. B
(2003)
C.W. Nan et al.
Annu. Rev. Mater. Res.
(2010)

There are more references available in the full text version of this article.

Cited by (24)

Nonmetal—Metal transition in carbon films embedded by Ni nanoparticles: The temperature coefficient of resistivity (TCR), Raman spectra and surface morphology
2018, Optik
Citation Excerpt :
Fig. 6(b) shows the surface roughness of films with Ni surface area 1.78, 3.21, 3.92 and 4.64% are estimated as 3.83, 3.30, 0.54 and 3.79 nm, respectively. The surface roughnesses of films from 1.78 to 3.92% are increased however, at range from 3.92 to 4.64% these characteristics are decreased [23–25]. Using Tencor Alpha −step 500 profiler the thicknesses of films with Ni surface areas 1.78, 3.21, 3.92and 4.64% were measured as about 700, 350, 200 and 450 nm, respectively.
In this paper, carbon films embedded by Ni nanoparticles were deposited on glass substrates. Films using four mosaic targets made of carbon and nickel with Ni surface areas 1.78, 3.21, 3.92 and 4.64% are made. With increasing Ni content in these films a nonmetal–metal transition especially above room temperature region is observed. This transition is explainable by electrical measurements in temperature range 15–500 K. With increasing Ni surface areas from 1.87 to 4.64%, Ni content of films increases from59.68 to 92.89%. The electrical conductivity of dielectric films was explained by tunneling effect. Activation tunneling energy that was obtained from temperature dependence of electrical resistivity correlates both with G and D peaks of Raman spectra and depend on Ni content of films. With increasing Ni surface areas from1.78 to 3.92% due to increasing graphen like carbon, a nonmetal–metal and from 3.92 to 4.64% due to increasing amorphous carbon interconnections of Ni nanoparticles, a peculiar effect of metallic to nonmetallic state occurs. The RMS roughness of films in 3.92% has minimum value about 0.54 nm.
Growth control of carbon nanotubes using nanocomposite nickel/carbon thin films
2017, Thin Solid Films
Citation Excerpt :
Such processes allow adjusting the size, the distribution and the shape of the metal nanoparticles by simply tuning the deposition parameters without the need of applying any intentional external heating to the substrate during deposition. Since the nature of the metal inserted in the amorphous carbon matrix can be simply controlled by using different targets, many systems have been considered, including nc-Ti/C [27–31,39,40], nc-Co/C [41–44], nc-Pt/C [45–47], nc-Ni/C [25,33,35–37,47–62] and nc-Cu/C [23,38,47,63–77]. Among the studied systems, nc-Ni/C thin films appeared to be very promising candidates for the growth of CNTs [78–80].
Recent papers have demonstrated that the growth of carbon nanotubes (CNTs) by plasma enhanced chemical vapor deposition (PECVD) was possible using nanocomposite nickel/carbon (nc-Ni/C) thin films as catalysts. In this study, the growth of CNTs by PECVD in H₂/C₂H₄ atmosphere was achieved using nc-Ni/C thin films deposited by a hybrid plasma process combining the sputtering of a nickel target and the deposition of hydrocarbon by PECVD using Ar/CH₄ atmosphere. In order to identify the most favorable conditions to obtain dense CNTs arrays using nc-Ni/C thin films, the Ni content in the catalyst as well as the growth conditions of the CNTs were varied. Films containing 40, 55 and 65 at.% of Ni were selected for this study. The growth temperature of the CNTs was varied between 500 and 700 °C whereas the electrical power applied to the PECVD source was tuned from 30 to 50 W. Scanning electron microscopy and Raman spectroscopy were employed to probe the morphology and the structure of the CNT's. Depending on the chemical composition of the nc-Ni/C thin films, different trends were observed. No CNTs were obtained neither for the highest nickel content (i.e. %Ni = 65 at.%) nor for the lowest growth temperature (i.e. 500 °C). On the other hand, for temperatures exceeding 500 °C, while a high power on the PECVD source (i.e. 50 W) was found to be necessary to obtain CNTs in the case of films with a moderate Ni content (i.e. %Ni = 55 at.%), a lower power (i.e. 30 W) was sufficient for the film with the lowest Ni content (i.e. %Ni = 40 at.%). This difference in behavior was attributed to the differences in microstructure of nc-Ni/C thin films which is directly related to their chemical composition.
Electrical behavior of nickel/carbon nanocomposite thin films
2017, Carbon
The relationship between the microstructure and the electrical behavior of nickel/carbon nanocomposite (nc-Ni/C) thin films is reported. A particular attention was accorded to the role of the chemical composition and the nature of the amorphous carbon matrix on the electrical behavior of the material. The nc-Ni/C thin films were synthesized using two different cold plasma processes both allowing to finely control the chemical composition, structure and morphology of the films. The first process combines magnetron sputtering of a nickel target and the deposition of hydrogenated carbon by plasma enhanced chemical vapor deposition using methane as a precursor. The second process consists in the co-sputtering of a nickel and a graphite target in pure argon plasma. For the two deposition processes, a similar increase in the electrical conductivity with the Ni content was observed and attributed to the percolation of the Ni nanograins through the carbon matrix. The percolation threshold was evaluated around 41 at.% of Ni for the two deposition processes. For lower Ni concentrations, the microstructural study indicated that the electrical conductivity is governed by the nature of the amorphous carbon phase which is found to be dependent on the deposition process as revealed by μ-Raman spectroscopy.
Fabrication LSPR sensor chip of Ag NPs and their biosensor application based on interparticle coupling
2015, Applied Surface Science
Citation Excerpt :
Metal nanoparticles are attached very well to the glass substrate in this method. The interparticle distance can be controlled by acetylene gas pressure in this method similar to our previous report about Cu NPs deposition [22,23]. The prepared LSPR sensor chip of Cu@CuO NPs by this method had been used for detection of CO molecule, previously [24].
We introduce a simple method to synthesize localized surface plasmon resonance (LSPR) sensor chip of Ag NPs on the hydrogenated amorphous carbon by co-deposition of RF-Sputtering and RF-PECVD. The X-ray photoelectron spectroscopy revealed the content of Ag and C atoms. X-ray diffraction profile and atomic force microscopy indicate that the Ag NPs have fcc crystal structure and spherical shape and by increasing deposition time, particle sizes do not vary and only Ag NPs aggregation occurs, resulting in LSPR wavelength shift. Firstly, by increasing Ag NPs content, in-plan interparticles coupling is dominant and causes redshift in LSPR. At the early stage of agglomeration, out-plane coupling occurs and in-plane coupling is reduced, resulting a blueshift in the LSPR. By further increasing of Ag NPs content, agglomeration is completed on the substrate and in-plan coupling rises, resulting significant redshift in the LSPR. Results were used to implement biosensor application of chips. Detection of DNA primer at fM concentration was achieved based on breaking interparticles coupling of Ag NPs. A significant wavelength shift sensitivity of 30 nm and a short response time of 30 min were obtained, where both of these are prerequisite for biosensor applications.
Fabrication Localized Surface Plasmon Resonance sensor chip of gold nanoparticles and detection lipase-osmolytes interaction
2014, Applied Surface Science
Citation Excerpt :
In particular, acetylene gas plays an important role in the formation of Au NPs in plasma and assists to immobilize them on the glass substrate through creation of a-C:H intermediate material. The inter-particle distance can be controlled by acetylene gas pressure in this method in which we have reported for Cu NPs deposition previously [14,15]. The LSPR analysis is applied to detect analyte concentration, protein–protein interaction and DNA hybridization [1].
Co-deposition of RF-sputtering and RF-PECVD from acetylene gas and Au target were used to prepare sensor chip of gold nanoparticles (Au NPs). Deposition conditions were optimized to reach a Localized Surface Plasmon Resonance (LSPR) sensor chip of Au NPs with particle size less than 10 nm. The RF power was set at 180 W and the initial gas pressure was set at 0.035 mbar. Transmission Electron Microscopy (TEM) images and Atomic Force Microscopy (AFM) data were used to investigate particles size and surface morphology of LSPR sensor chip. The Au and C content of the LSPR sensor chip of Au NPs was obtained from X-ray photoelectron spectroscopy (XPS). The hydrogenated amorphous carbon (a-C:H) thin film was used as intermediate material to immobilize Au NPs on the SiO₂ substrate. The interaction between two types of osmolytes, i.e. sorbitol and trehalose, with Pseudomonas cepacia lipase (PCL) were detected by the prepared LSPR biosensor chip. The detection mechanism is based on LSPR spectroscopy in which the wavelength of absorption peak is sensitive to the refractive index of the environment of the Au NPs. This mechanism eliminates the use of a probe or immobilization of PCL on the Au NPs of LSPR sensor chip. The interaction between PCL and osmolytes can change refractive index of the mixture or solution. We found that unlike to trehalose, sorbitol interacts with the PCL. This interaction increases refractive index of the PCL and sorbitol mixture. Refractive index of PCL in the presence of different concentration of sorbitol was obtained by Mie theory modeling of LSPR peaks. This modeling stated that the present LSPR sensor chip has sensitivity as high as wavelength shift of 175 nm per refractive index. Moreover, the detection of such weakly interaction between bio-molecules cannot be achieved by other analysis.
Morphology, optical and electrical properties of Cu-Ni nanoparticles in a-C:H prepared by co-deposition of RF-sputtering and RF-PECVD
2011, Applied Surface Science
Citation Excerpt :
It is reduced to 3 KΩ and 500 Ω for the samples with 10 and 15 nm Ni over layer respectively. For the sample with resistance of 500 Ω, electrical conduction arises from percolation of NPs [11]. Regarding above results, one may correlate LSPR absorption peak to electrical resistance of samples.
We report optical and electrical properties of Cu–Ni nanoparticles in hydrogenated amorphous carbon (Cu–Ni NPs @ a-C:H) with different surface morphology. Ni NPs with layer thicknesses of 5, 10 and 15 nm over Cu NPs @ a-C:H were prepared by co-deposition of RF-sputtering and RF-Plasma Enhanced Chemical Vapor Deposition (RF-PECVD) from acetylene gas and Cu and Ni targets. A nonmetal–metal transition was observed as the thickness of Ni over layer increases. The surface morphology of the sample was described by a two dimensional (2D) Gaussian self-affine fractal, except the sample with 10 nm thickness of Ni over layer, which is in the nonmetal–metal transition region. X-ray diffraction profile indicates that Cu NPs and Ni NPs with fcc crystalline structure are formed in these films. Localized Surface Plasmon Resonance (LSPR) peak of Cu NPs is observed around 600 nm in visible spectra, which is widen and shifted to lower wavelengths as the thickness of Ni over layer increases. The variation of LSPR peak width correlates with conductivity variation of these bilayers. We assign both effects to surface electron delocalization of Cu NPs.

View all citing articles on Scopus

View full text

Metal–nonmetal transition in the copper–carbon nanocomposite films

Abstract

Introduction

Section snippets

Experimental details

Results and discussion

Conclusion

Acknowledgments

Solid State Mater. Sci.

Heterogeneous Materials I, Linear Transport and Optical Properties

Rev. Mex. Fis. S

J. Appl. Phys.

Phys. Rev. B

Phys. Rev. B

Annu. Rev. Mater. Res.