SEQUENTIAL PECVD/PVD TECHNIQUE USED FOR THE SYNTHESIS OF COPPER-POLYPYRROLE NANOCOMPOSITES WITH SURFACE PLASMON RESONANCE EFFECT

Romanian Reports in Physics, Vol. 64, Supplement, P. 1345–1353, 2012

Dedicated to Professor Ioan-Iovitz Popescu’s 80th Anniversary

SEQUENTIAL PECVD/PVD TECHNIQUE USED FOR THE SYNTHESIS OF COPPER-POLYPYRROLE NANOCOMPOSITES WITH SURFACE PLASMON RESONANCE EFFECT V. SATULU, C. STANCU, V. ION, M. FILIPESCU, B. MITU*, G. DINESCU National Institute for Laser, Plasma and Radiation Physics, PO-Box MG-36, 077125, Măgurele, Romania *[email protected] Received November 14, 2012

Abstract. The synthesis of copper-polypyrrole composites was conducted by combining, in a sequential substrate exposure, the plasma polymerization of pyrrole vapors and magnetron sputtering of Cu target. The optical properties indicate strong absorption band around 620 nm, associated with the surface plasmon resonance due to Cu inclusion in the polypyrrole matrix. Key words: plasma polymerization, magnetron sputtering, polypyrrole-like material, Cu nanoparticles, surface plasmon resonance.

1. INTRODUCTION The metal-polymer composites are among the most studied materials nowadays due to their possible application in light emitting diodes (LED), thin film transistors (TFT), gas sensors [1], solar cells, catalytic systems [2] and biomedical surfaces [3]. Copper-polymer composites are especially investigated thanks to their optical properties resulted from the generation of surface plasmon resonance [4] and catalytic activity, as well as their possible use for sensors applications [5]. In order to accomplish the conditions for use in microelectronic fields and sensor applications, the polymeric matrix used for the copper nanoparticles inclusions should be conductive; for this purpose, polypyrrole is one of the potential candidates [6]. Although most of the methods for the synthesis of the metal-

1346

V. Satulu, C. Stancu, V. Ion, M. Filipescu, B. Mitu, G. Dinescu

2

polymer composites implies the use of wet chemistry [7, 8], utilization of nonequilibrium plasma discharges became an attractive alternative due to its environmentally-friendly character, possibility of tailoring the ratio between the polymer and metal concentration, and the versatility regarding the implementation into the existing technological lines [9]. The present contribution focuses on the synthesis of copper-polypyrrole-like composite materials by a combined PECVD/PVD method. The substrate is alternatively exposed to an Ar plasma in which pyrrole vapors are introduced under controlled flow rates, and to a magnetron sputtering plasma of copper target. The plasma generated by each source was investigated by Optical Emission Spectroscopy (OES) and mass spectrometry (MS) in order to reveal the species involved in composite synthesis. The obtained materials have been investigated as regarding the optical, morphological and compositional properties by means of Spectroscopic Ellipsometry (SE), Atomic Force Microscopy (AFM), and Fourier Transformed Infrared spectroscopy (FTIR) techniques. 2. EXPERIMENTAL SET-UP The deposition of copper-polypyrrole composites was carried out in a vacuum chamber system, initially pumped down to a base pressure of 2×10-2 mbar, provided with two plasma sources and the plasma diagnostic tools, as shown in Fig. 1. In the upper part is mounted a classical PECVD plasma source, consisting in a RF powered electrode designed as a shower for facilitating the homogeneous introduction of the monomer. The pyrrole (C4H5N) precursor is introduced in vapor phase at controlled flow of 2.8 sccm through a control, evaporator and mixing system (CEM, Bronkhorst) by a small argon carrier flow (20 sccm). A capacitively coupled RF plasma (13.56 MHz, 20 W power) is running between the active electrode and the grounded substrate holder, placed at 6 cm from the RF one. The established pressure in the plasma polymerization step is 0.13 mbar. On a lateral side of the chamber was mounted, at 8 cm from the central part of the substrate holder, the magnetron sputtering gun, provisioned with a 2” diameter copper target. Argon flow (10 sccm) was injected in front of the target in order to efficiently sputter Cu atoms upon discharge generation in RF field at 100 W. The working pressure during the sputtering step of the deposition was established at 7.3×10-2 mbar. A pre-sputtering time (of 10 min) was set prior every step of metallic deposition in order to insure the target cleaning. The grounded substrate holder, which accommodates 1×1 cm2 substrates of Si (100) double side polished, was inclined at 45 degrees in respect to both PECVD and magnetron

3

Sequential PECVD/PVD technique

1347

sputtering sources. The deposition process of copper–polypyrrole (Cu-PPY) composites consists in the sequential exposure of the substrate to the PECVD plasma working in argon/pyrrole admixture for 1 min and respectively to the PVD plasma with copper target for the next minute. The first and the last step of each deposition was the polymeric one, for an effective covering of the copper particles. For comparison purposes, continuous layers of polypyrrole-like and Cu thin films were deposited, using exposure time of the substrates in the range from 5–15 min in each case.

Fig. 1 – Experimental set-up used for sequential PVD/PECVD deposition of Cu-polypyrrole composite.

The plasma diagnostics was ensured by OES investigations in the region 200–1000 nm using a Bruker spectrograph (grating 1200 mm-1) equipped with an Andor CCD camera, and respectively by a mass spectrometer (EQP 1000, Hiden Analytical) mounted inside the reactor, laterally and equidistant in respect to both plasma sources. The base pressure inside the mass spectrometer was 2.5×10-7 mbar, while the pressure during measurements was reaching 5.5×10-7 mbar. The surface topography was investigated by Atomic Force Microscopy (AFM), using a Park Systems apparatus, model: XE-100, working in non-contact mode. The chemical structure of the metal-polymer composites was determined by Fourier Transform Infrared spectroscopy (FTIR) measurements performed with a JASCO 6300 spectrometer in the range 400–4000 cm-1, with a resolution of 4 cm-1. A Woollam Variable Angle Spectroscopic Ellipsometer (WVASE 32, J.A. Woollam Company, Inc) equipped with a HS-190 monocromator with xenon lamp was used to evaluate the thickness and to determine the optical constants of the obtained materials. The ellipsometric angles (ψ and ∆) have been experimentally measured at an angle of 700 of the light beam in the spectral range 300–1000 nm with a step of 2 nm.

V. Satulu, C. Stancu, V. Ion, M. Filipescu, B. Mitu, G. Dinescu

1348

4

3. RESULTS AND DISCUSSION 3.1. PLASMA INVESTIGATION Typical emission spectra for each of the plasmas involved in the synthesis of the composite Cu-PPY material are presented in Fig. 2. The plasma generated by the magnetron sputtering plasma source with Cu target is dominated by the emission of Cu I lines, the most intense ones being visible at 323.57 nm and 327.39 nm, as well as Ar I (e.g. at 696.54 nm) and Ar II lines; however, small bands associated to the presence of impurities in the reaction chamber are also visible, through the N2 emission bands of the Second Positive System (SPS, C3Π-B3Π) [10]. The most important feature of the PECVD discharge generated in Ar/C4H5N is the emission of CN Violet System B2Σ+ - X2Σ+ with the band head (0,0) at 388.3 nm, suggesting the decomposition of the monomer by plasma. At the same time, no emission coming from the CH and C2 related bands were detected, pointing out that still the plasma polymerization process takes place under soft power conditions [11]. 320

340

360

380

660

680

700

PECVD plasm a Ar/C 4H 5N

Cu lines

PVD plasm a Ar/Cu Ar lines

Intensity (a.u.)

N 2 SPS system 3

3

C Π -B Π

320

340

360

CN Violet System 2 + 2 + B Σ -X Σ

380

660

680

700

W avelength (nm )

Fig. 2 – Typical OES spectra of the PECVD discharge working in Ar/C4H5N mixture (black line) and respectively of PVD magnetron argon plasma with copper target (red line).

The mass spectra of neutral species present in the reactor during magnetron sputtering of Cu (not shown) are dominated by the Ar peak (40 amu) and the peaks at 63 and 65 amu, corresponding to the Cu isotopes. At the same time, small traces of water, nitrogen and oxygen coming from the residual atmosphere were also noticed. The ionic mass spectra evidence the presence of Cu ions and the remaining of just a few Ar ions, since they are consumed during the target sputtering process [12]. The mass spectra of the neutrals present in the argon/pyrrole mixture plasma reveals the incomplete dissociation of the monomer in the plasma as the signal

5

Sequential PECVD/PVD technique

1349

associated to (C4H5N) at 67 amu is still present, and the generation of smaller radicals, like C2H3N, C2HN and C3H3 at 41 amu and respectively at 39 amu [13]. The signal at 40 amu and 28 amu incorporate both the contribution of the pyrrole formed species (C2H2N, C2H4,) and those of the working gas (Ar) and impurities (N2), respectively. 3.2. MATERIAL CHARACTERIZATION The AFM investigations of a plasma polypyrrole layer and respectively for a Cu layer deposited using a deposition time of 1 min, equal to that used during a synthesis cycle of Cu-PPY composites are presented in Fig. 3 (a and b). The polymeric layer evidences a very smooth deposit, with a few features present on the surface and a roughness RMS of only 0.3 nm over a scanned area of 3×3 µm2.

a)

b)

Fig. 3 – AFM images of: a) plasma polypyrrole film obtained during the PECVD step; b) metallic covering obtained during PVD step.

The metallic deposition obtained also during one cycle shows the formation of small particles, below 20 nm diameter and 1 nm height and only a few larger particles sparsely distributed on the surface. The roughness RMS in this case is around 0.6 nm. Instead, when several cycles of sequential polymeric-metallic deposition is performed, the surface topography is revealing the formation of agglomeration of small particles with spherical shape, as shown in the inset of Fig. 4. We assimilate such topography to the formation of core-shell type structures consisting of nanometer sized Cu nanoparticles coated with a thin polypyrrole layer. The processed AFM image obtained over 3×3 µm2 scanned area allowed the determination of typical dimension of the structures in the range of 50–150 nm diameter, and respectively an average of 14–15 particles/µm2. The roughness RMS of the composite material was increased up to 8.9 nm.

1350

V. Satulu, C. Stancu, V. Ion, M. Filipescu, B. Mitu, G. Dinescu

6

Fig. 4 – Histogram of the particles diameters obtained during the synthesis of Cu-PPY composite material; in the inset: typical AFM image of Cu-PPY composites containing spherical nanoparticles.

In Figure 5 are presented the FTIR spectra of a plasma polypyrrole thin film and that of a Cu-PPY composite material, in respect to the spectrum of the initial liquid pyrrole monomer. Several FTIR bands evidence that the initial structure of the monomer is conserved upon plasma polymerization of pyrrole, e.g. the band at 728 cm–1 associated to the out of plane deformation vibration in pentacyclic ring, the sequence of 3 bands positioned at 1430, 1449, 1529 cm-1 related to the C=C in plane vibrations and two small peaks at 3105, 3134 cm-1 related to the =C-H stretching bond in aromatic compounds [14]. Nevertheless, the presence of intense bands associated to the aliphatic stretching bonds of CHx (x = 2,3) in the region 2800–3000 cm-1 corroborated to the formation of – C ≡ N and –N ≡ C triple bonds observed in the range 2150–2250 cm–1 indicate the ring breaking and the crosslinking character of the deposit [15]. One should notice here that the copper – polypyrrole composite present larger peaks in the region 1600–1800 cm–1 and 3100–3600 cm–1 associated to the presence of hydroxyl groups which may induce the oxidation of copper. The spectro-ellipsometry data were fitted by using an optical model needed for the generation of theoretical curves. It comprises 4 layers of different materials: the first one is the Si substrate (bulk), the second is the native SiO2 layer with a thickness of approximately 3 nm under normal substrate cleaning procedure, the investigated layer and a fourth uppermost layer which counts for the surface roughness. This superficial layer was approximated, in a Bruggeman approach, by a combination of 50% material with optical characteristics identical to those of the main deposit and 50% air. The optical constants of the first two layers were taken from the literature [16]. The polypyrrole-like layer could be successfully

7

Sequential PECVD/PVD technique

1351

approximated to a Cauchy-dispersive material with an Urbach absorption [17]. The experimental and fitted curves of the ellipsometric angles, as well as the obtained refractive index n and absorption coefficient k of the obtained polymeric film are presented in Fig. 6, evidencing very low values of the absorption coefficient (below 5×10-3) at wavelengths above 400 nm. A deposition rate of around 15 nm/min was obtained under these deposition conditions. C=C in plane NH, NH2 ring vibrations

=C-H stretching

CHx stretching

bending vibrations

Cu-PPPy

Absorbance (a.u.)

III

-C N, -N=C stretching

NH, NH2 stretching

Plasma polypyrrole

C4H5N monomer

800

1200

1600

2000

2800

3200

3600

4000

-1

Wavenumber (cm )

Fig. 5 – FTIR spectra of pyrrole initial monomer (black curve); polypyrrole-like thin film (red curve) and respectively composite Cu-polypyrrole material. 200

60

125

50

100

40 75

30 20

50

10

25 1000

400

600

800

Wavelength (nm)

a)

1.75 0.14 0.12

1.70

∆°

Ψ°

80

0.10 0.08

1.65

0.06

n

1.60

0.04 0.02

k 1.55

400

600

800

Absorption coefficient

70

Ψ model Ψ experimental 175 ∆ model ∆ experimental 150

Refractive index

90

0.00 1000

Wavelength (nm)

b)

Fig. 6 – a) Experimental (points) and modeled (lines) ellipsometric angles (ψ and ∆); b) refractive index and absorption coefficient of polypyrrole-like thin film.

For the case of composite Cu-polypyrrole material, the optical model comprises as main layer a mixture of polypyrrole having the optical constants equal to those previously determined and copper. In a combined EMA-Cauchy Urbach

V. Satulu, C. Stancu, V. Ion, M. Filipescu, B. Mitu, G. Dinescu

1352

8

model, the experimental curves could be fitted within reasonable errors (MSE = 42.11) according to the curves presented in Fig. 7a. We should point out that the total thickness of ~110 nm for the 6 sequences of 1 min pyrrole plasma and 5 sequences of 1 min Cu plasma is in accordance to the findings for the individual polypyrrole and Cu layers, respectively. The copper percentage in the material was determined as 2.36%. 26

60

22

40

16 14

30

12

Ψ model 20 Ψ experimental ∆ model 10 ∆ experimental

10 8 400

600

800

1000

∆°

Ψ°

18

Refractive index

50

20

1.92

0.54

1.90

0.52

1.88

0.50

n

0.48

1.86

0.46 1.84

Absorption coefficient

70

24

k 0.44

1.82

400

Wavelength (nm)

600

800

1000

Wavelength (nm)

a)

b)

Fig. 7 – a) Experimental (points) and modeled (lines) ellipsometric angles (ψ and ∆); b) refractive index and absorption coefficient of Cu-polypyrrole composite materials.

The resulted optical constants, presented in Fig. 7b, evidence a strong absorption band in the wavelength region around 620 nm, typical for the generation of Cu surface plasmon resonance [18]. At the same time, the broader absorption in the region 360–550 nm can be attributed to the polaron/bipolaron transition [7], specific to the conductive polymer layers, probably enhanced by the presence of conductive copper particles in the composite structure. Nevertheless, further investigation of the composite optical properties (e.g. by means of UV-Vis absorption) will be performed in order to find a better correlation between the experimental and fitted ellipsometric angles and to establish more accurately the absorption bands. 4. CONCLUSIONS The synthesis of copper-polypyrrole nanocomposites was conducted by combining two deposition methods, namely PECVD of pyrrole vapors and PVD (magnetron sputtering) of Cu target in a sequential exposure of the substrate to the plasma sources. The plasma diagnostics during plasma polymerization step evidenced the optical emission of CN radicals and the formation of various pyrrole fragments, while, during the magnetron sputtering of copper, atomic and ionic Cu was observed by means of OES and mass spectrometry, respectively.

9

Sequential PECVD/PVD technique

1353

The AFM investigations showed the agglomeration of small particles having spherical shape and diameter in the range of 50–150 nm. FTIR spectra indicated the partial conservation of the pyrrole structure, as well as the appearance of new bonds related to the fragmentation and recombination of the monomer during plasma polymerization process. The spectroscopic-ellipsometry data evidenced a strong absorption peak around 620 nm, associated with the generation of surface plasmon resonance due to the Cu inclusion of nanometric size in the polypyrrole matrix. Acknowledgements. This work was supported by the Romanian Ministry of Education, Research, Youth and Sport under the Human Resources Programme, contract TE_229/2010 Metalpolymer nanocomposites obtained by combined plasma techniques.

REFERENCES 1. H. Goktas, F.G. Ince, A. Iscan, I. Yildiz , M. Kurt, I. Kaya, Synthetic Metals, 159, 2001–2008 (2009). 2. E. Dilonardo, A. Milella, F. Palumbo, G. Capitani, R. d’Agostino, F. Fracassi, Plasma Proc. Polym., 7, 51–58 (2010). 3. L. Yue, M. Zhou, Q. Chen, J. Weng, Y. Zhang, Vacuum, 83, 1200–1203 (2009). 4. D. Salz, B. Mahltig, A. Baalmann, M. Wark, N. Jaeger, Phys. Chem. Chem. Phys., 2, 3105–3110 (2000). 5. C. Walter, V. Bruser, A. Quade, K.-D. Weltmann, Plasma Proc. Polym., 6, 803–812 (2009). 6. T. V. Vernitskaya, O. N. Efimov, Russ. Chem. Rev., 66, 443 (1997). 7. K. Mallick, M. J. Witcomb, M. S. Scurrell, Europ. Polym. J., 42, 670–675 (2006). 8. V. Chaudhary, A. K. Thakur, A. K. Bhowmick, J Mater Sci, 46, 6096–6105 (2011). 9. H. Biederman, O. Kylian, M. Drabik, A. Choukourov, O. Polonskyi, P. Solar, Surf. Coat. Technol., 211, 127–137 (2012). 10. R.W.B. Pearse, A.G. Gaydon, The identification of molecular spectra, Chapman and Hall, London, 1976, 4th ed. 11. M.C. Kim, S.H. Cho, J.G. Han, B.Y. Hong, Y.J. Kim, S.H. Yang, J.-H. Boo, Surf. Coat. Technol., 169–170, 595–599 (2003). 12. B. Mitu, V. Satulu, G. Dinescu, Proc. 20th Int. Symp. Plasma Chem., Philadelphia, USA, July 24–29; http://www.ispc-conference.org/ispcproc/ispc20/237.pdf (2011). 13. *** NIST Chemistry WebBook Database. 14. G. Socrates, Infrared Characteristic Group Frequencies, J.Wileys&Sons Ltd, 2nd ed., 1994. 15. J. Zhang, M.Z. Wu, T.S. Pu, Z.Y. Zhang, R.P. Jin, Z.S. Tong, D.Z. Zhu, D.X. Cao, F.Y. Zhu, J.Q. Cao, Thin Solid Films, 307, 14–20 (1997). 16. C.M. Herzinger, B. Johs, W.A. McGahan, J.A. Woollam,W. Paulson, J. Appl. Phys., 83, 3323–3336 (1998). 17. H. Fujiwara, Spectroscopic Ellipsometry Principles and Applications, Published by Maruzen Co. Ltd, Tokyo, Japan, ISBN-13:978-0-470-01608-4. 18. D. B. Pedersen, S. Wang, J. Phys. Chem. C, 111, 17493–17499 (2007).