Upload
gaetan
View
215
Download
0
Embed Size (px)
Citation preview
Characterization of microwave plasma for polymer surface modification
using FTIR emission spectroscopy
Maryam Mavadat1,2,a, Stéphane Turgeon2,b, André Ricard3,c and Gaétan Laroche1,2,d
1Laboratoire d’ingénierie de surface, Centre de Recherche sur les Matériaux Avancés,
Département de génie des mines, de la métallurgie et des matériaux, 1065 avenue de la médecine, Université Laval, Québec, G1V 0A6, Canada
2Centre de recherche du CHUQ, Hôpital Saint-François d’Assise, 10 rue de l’Espinay, Québec,
G1L 3L5, Canada
3 LAPLACE, Université Paul Sabatier, 118 route de Narbonne, 31062 Toulouse, France
Keywords: Plasma emission spectroscopy, Polymer surface treatment, Infrared spectroscopy, Plasma temperature
Abstract. Infrared (IR) emission spectroscopy measurements were performed in N2 microwave
discharges at pressures ranging from 0.5 to 3 Torr and powers of 200 and 300 W. Although
emission spectroscopy in the infrared region has rarely been investigated, this technique has
nevertheless provided numerous key data. For instance, numerical generation of spectra to match
experimental FTIR emission data allowed estimating the plasma temperature.
1. Introduction
Non-equilibrium plasmas have a great impact on polymeric material science, in that they allow
modifying the outermost surface layer, in ‘cold’ processes, conferring materials a tailored surface
composition and properties such as adhesion, wettability, and biocompatibility. Different plasma
external parameters (pressure, input power and gas flow) affect the characteristics of the plasma and
consequently the surface chemistry of plasma modified polymers. In this context, the knowledge of
fundamental processes occurring in the plasma is a prerequisite for further process control. The
plasma temperature, which most of the time is estimated by rotational temperature (Tr), is one of the
most important plasma parameters. As a matter of fact the chemistry of the discharge can be
influenced by the gas temperature, since it governs the reaction rate of active species generation
through dissociation, excitation, and ionization processes, and consequently, accurate knowledge of
Tr is required to understand and control plasma processes.
In the present work in situ, non-intrusive diagnostic (optical emission spectroscopy) in the
infrared spectral region were performed in a N2 microwave plasma discharge using FTIR emission
spectrometer. The data recorded from emission spectroscopy were used to determine the
temperature of the N2 microwave discharge as a function of gas pressure and microwave power by
comparing the experimental and numerically generated spectra. The results of plasma temperature
as a function of pressure are presented.
Advanced Materials Research Vol. 409 (2012) pp 797-801Online available since 2011/Nov/29 at www.scientific.net© (2012) Trans Tech Publications, Switzerlanddoi:10.4028/www.scientific.net/AMR.409.797
All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of TTP,www.ttp.net. (ID: 132.174.255.116, University of Pittsburgh, Pittsburgh, USA-02/12/14,04:13:46)
2. Experimental set-up
Fig. 1 presents the experimental set-up.
Figure 1. Experimental set-up for MW discharge and emission spectroscopy.
The low pressure plasma reactor illustrated in Fig. 1 is described in detail elsewhere [1]. Briefly, the
experiments were performed in a microwave reactor purchased from Plasmionique Inc. (Varennes,
QC, Canada). The N2 gas (99.998% purity) used for the plasma characterization was introduced in
the discharge tube by means of a mass flow controller. The plasma column was created in N2 gas at
pressures ranging from 0.5 to 3 Torr. The power delivered to the launcher was 200 and 300 W.
As aforementioned, the plasma characterisation was performed using IR emission spectroscopy.
The near infrared spectra were recorded with a FTLA2000 FTIR (770-3300 nm) spectrometer
purchased from ABB-Bomem (Québec, QC, Canada) at a resolution of 2 cm-1
. For infrared
collection, the IR light exited through a ZnSe feedthrough window, was expanded through a
concave ZnSe lens then was redirected by a concave gold mirror into the FTIR’s entrance port. All
IR optics was purchased from International Scientific Products Corp. (Irvington, NY, USA). The
great advantage of near infrared emission spectroscopy is that it makes it possible to record features
of the N2 first positive (1st positive) system that are not observable through classical UV-VIS
emission spectroscopy.
The detector used during each experiment was a thermoelectrically cooled Indium Arsenide
(InAs) semiconductor diode which is sensitive in the range of 3000-14000 cm-1
(710-3300 nm). One
hundred interferograms were routinely co-added and Fourier-transformed, thereby enabling us to
record spectra with an acceptable signal-to-noise ratio and reasonable acquisition time.
3. Results and discussion
Infrared emission spectrum of the N2 microwave discharge at 3 Torr is shown in Fig. 2. In the
present study, the observed infrared spectra consisted mostly of N2 1st positive transitions. In
addition, spectra recorded at low gas pressures P <1 Torr allowed visualizing various NI atomic
transitions. In the illustrated spectrum, the infrared features are related to different vibrational
transitions originating from the 1st positive system of nitrogen.
798 THERMEC 2011 Supplement
Figure 2. Nitrogen plasma emission spectrum in the 630-1830 nm range (300W, 3 Torr, 80 sccm).
Rotational temperature. In principle, the rotational temperature of N2 can be determined using
various bands from either the first negative [2] or second positive system [3] through a Boltzmann
plot, provided that the spectrometer resolution is high enough to separate the rotational structure of
the bands or by fitting numerical models to the band envelope when medium to low resolution
spectrometers are used [4]. However, despite the characteristics of the 1st positive system that make
it a more reliable tool to estimate gas temperature (low excitation threshold, long radiative lifetime,
higher predissociation level, and high emission intensity), its complex structure with 27 rotational
branches has hampered its widespread use [4-6].
Inspired by Biloiu et al. [7] the plasma gas temperature was evaluated by calculating and
generating theoretical spectra closely matching the bands originating from 0-0 transition in the N2
1st positive system which is observed in the range of 1025-1055 nm. To that end, a synthetic
spectrum which was automated in MATLABTM
[7] has been used. However, such a procedure
requires the knowledge of the FTIR spectrometer apparatus function which, in turn, has to be
convolved with the calculated spectrum. To that end, the shape parameters of the 826 nm ArI line
taken from the spectrum of an Ar-Hg source were used, considering that the instrumental
broadening was modeled by a pseudo-Voight function:
( ) ( ) ( )( )
202 22
0
4ln 2 4ln 2 2, exp 1 .
4
wf p w p p
w w wλ λ
ππ λ λ
= − − + − + −
where p and 1−p are the relative magnitudes of the Gaussian and Lorentzian functions contributions,
respectively, w is the full width at half maximum of the line (FWHM), and λ0 is the central
wavelength. This allowed determining that p value was 0.5 while w was 0.2 nm for the
experimental setup used in the present study.
It is worth mentioning that the value of p has an error of ± 0.1. However, this has minimal effect on
the temperature calculation as can be seen from the error bars on the curves presented in Fig. 4.
Advanced Materials Research Vol. 409 799
Fig. 3 shows an example of a measured ro-vibrational N2 spectrum along with the corresponding
spectrum calculated using the aforementioned protocol. As can be seen, an excellent match between
both calculated and experimental spectra was observed, with a confidence level better than 95%.
Figure 3. Experimental spectrum (solid line) and corresponding numerically generated (dotted line) spectrum of the 0-0
band for the best fit gas temperature. In this example, the evaluated rotational temperature (Tr) was 1009 K
This method was thereafter used to characterize the pressure-dependence of the rotational
temperature in nitrogen plasmas generated at powers of 200 W and 300 W, respectively. As seen in
Fig. 4, Tr increased with pressure and power due to the more frequent collisions occurring between
the plasma species which lead to a decrease in electron temperature with concomitant increase of
gas temperature. These results are in good agreement with previous results on pure N2 discharges [1,
4].
Figure 4. Rotational temperature as a function of pressure for plasmas generated at 200 W and 300W.
The present results clearly highlighted that the numerical simulation of the spectrum of the 0-0
transition of the nitrogen first positive system allows calculating rotational temperatures with better
sensitivity as compared to using the P1/P2 ratio from the 2-0 transition [1, 4]. On one hand, the 0-0
transition gives rise to spectra with higher emission intensity as compared to the 2-0 transition. On
the other hand, the P1/P2 ratio from 2-0 transition revealed to be not sensitive enough to discriminate
subtle temperature changes.
800 THERMEC 2011 Supplement
Conclusion
Optical emission spectroscopy in the near IR spectral region was used for the diagnostic of low
pressure nitrogen microwave discharges currently used for modifying the surface of biomedical
polymers. The experimental infrared spectra consisted mostly of N2 1st positive system transitions.
Rotational temperatures were accurately measured by fitting the N2 1st positive system spectra
recorded from plasmas generated at 0.5 to 3 Torr of N2 with powers of 200 W and 300 W. Under
these conditions, the N2 rotational temperatures ranged from 646 to 1012 K.
References
[1] Mavadat M, Ricard A, Sarra-Bournet C and Laroche G 2011 Journal of Physics D: Applied
Physics Accepted
[2] Williamson J M and DeJoseph C A 2003 Journal of Applied Physics 93 1893-8
[3] Tonnis E J and Graves D B 2002 Journal of Vacuum Science & Technology A: Vacuum,
Surfaces, and Films 20 1787-95
[4] Britun N, Gaillard M, Ricard A, Kim Y M, Kim K S and Han J G 2007 Journal of Physics D:
Applied Physics 40 1022-9
[5] Biloiu C, Sun X, Harvey Z and Scime E 2006 Determination of rotational and vibrational
temperatures of a nitrogen helicon plasma. Review of Scientific Instruments) pp 10F117-4
[6] Ricard A, Nouvellon C, Konstantinidis S, Dauchot J P, Wautelet M and Hecq M 2002 Journal
of Vacuum Science & Technology A: Vacuum, Surfaces, and Films 20 1488-91
[7] Biloiu C, Sun X, Harvey Z and Scime E 2007 Journal of Applied Physics 101 073303-11
Advanced Materials Research Vol. 409 801
THERMEC 2011 Supplement 10.4028/www.scientific.net/AMR.409 Characterization of Microwave Plasma for Polymer Surface Modification Using FTIR Emission
Spectroscopy 10.4028/www.scientific.net/AMR.409.797