Optical emission spectroscopy and actinometry in CCl4-Cl2 radiofrequency discharges

Radiofrequency discharges fed with CCl₄-Cl₂ mixtures have been studied in the pressure range 0.3 to 0.6 torr by means of emission spectroscopic actinometry with Ar, He, and N₂ as actinometers. Two different reactors, a parallel plate and a capacitively coupled tubular one, have been utilized for this study to obtain information for a large range of electron energy distributions. Analysis of the experimental results demonstrates the following: the utilization of actinometry and its range of validity, the importance of electron attachment to CCl_x species during the plasma decomposition process, and the effects of the presence of chlorine and “glowpolymer” in the discharge medium.     

Optical emission spectroscopy of electron-cyclotron-resonance-heated helium mirror plasmas

In this experiment emission spectroscopy in the 3000–5000 Å range has been utilized to determine the electron temperature (15–60 eV) and ion density (2–5 x 10¹¹ cm^–3) of helium plasmas produced by the Michigan mirror machine⁽¹⁾ (MIMI). The plasma is generated and heated by whistler-mode electron-cyclotron resonance (ECR) waves at 7.43 GHz with 400–900 W power in 80-ms-long pulses. Gas fueling is provided at the midplane region by a leak valve with a range in pressure of 3 x 10 to 2 x 10⁴ Torr. Emission line intensities are interpreted using a model of the important collisional and radiative processes occurring in the plasma. The model examines secondary processes such as radiation trapping, excitation transfer between levels of the carne principle quantum number, and excitation front metastable states for plasmas in the parameter range of MIMI (n _c = 1–6 x 10¹¹ cm^–3). Front the analysis of line intensity ratios for neutral helium, the electron temperature is measured and its dependence upon the gas pressure and microwave power is determined. These temperatures agree with those obtained by Langmuir probe measurements. Art analysis of the line intensity ratio between singly ionized helium and neutral helium yields a measurement of the ion density which is in good agreement with electron density measurements made by a microwave interferometer.     

Characterization of laser induced plasmas by optical emission spectroscopy: A review of experiments and methods

Advances in characterization of laser induced plasmas by optical emission spectroscopy are reviewed in this article. The review is focused on the progress achieved in the determination of the physical parameters characteristic of the plasma, such as electron density, temperature and densities of atoms and ions. The experimental issues important for characterization by optical emission spectroscopy, as well as the different measurement methods are discussed. The main assumptions of the methods, namely the optical thin emission of spectral lines and the existence of local thermodynamic equilibrium in the plasma are evaluated. For dense and inhomogeneous sources of radiation such as laser induced plasmas, the characterization methods are classified in terms of the optical depth and the spatial resolution of the emission used for the measurements. The review deals firstly with optically thin spatially integrated measurements. Next, local measurements and characterization in not optically thin conditions are discussed. Two tables are included that provide reference to the works reporting measurements of electron density and temperature of laser induced plasmas generated with diverse samples.     

Optical emission spectroscopy as a real time diagnostic tool for plasma-assisted deposition of TiN

Real-time optical emission spectroscopy (OES) was used to monitor the deposition of TiN both from mixtures of tetrakis(dimethylamino)titanium (TDMATi)-N₂ and TiCl₄-H₂-N₂ in an electron cyclotron resonance chemical vapor deposition system. The accurate control of the ratio of the emission intensities of ionized nitrogen at 391.4 nm and molecular nitrogen at 357.7 nm (N ₂ ⁺ /N₂) led to low temperature deposition of stoichiometric TiN (Ti/N ≈ 1) and very low resistivity in both cases. It was found that high ion density plasmas are crucial for a considerable reduction of the deposition temperature while maintaining good film quality. OES shows that the abundance of certain excited plasma species is not only dependent on the gas mixture and the deposition parameters, such as total pressure and microwave power, but also is strongly affected by the magnetic field configuration. The deposition rate and the film resistivity can be related to the emission intensity ratio, I(N ₂ ⁺ )/I(N₂). Finally, the two processes are compared in terms of the quality of as-deposited and heat-treated films. The comparison shows that the films obtained with TDMATi exhibit lower resistivity and are thermally more stable than with TiCl₄.     

Optical emission spectroscopy diagnostics of inductively-driven plasmas in argon gas at low pressures

An optical emission spectroscopy method for determination of electron temperature, electron density and gas temperature is developed and applied for diagnostics of inductively-driven argon discharges in a cylindrical geometry. The discharges are maintained at frequency 27 MHz, applied power varied in the limits P = (90 – 160) W and gas pressure in the range p = (1.1 – 117.3) Pa. The method combines measurements of emission spectral line intensities and profile broadenings with a collisional-radiative model of argon plasma at low pressure. The model is employed for investigation of the plasma kinetics governing the population densities of 3p⁵4s and 3p⁵4p argon configuration levels, treated separately. In the numerical calculations the electron density and electron temperature are varied whereas the values of the third plasma parameter — the gas temperature — are involved as obtained data from the experiments. Comparison of the experimental results of the line-intensity ratios with those calculated by the model yields the values of the electron density and temperature. The dependence of the electron temperature, electron density and gas temperature on the discharge conditions is obtained and discussed in the study.     

A model of the chemical processes occurring in CF4/O2 discharges used in plasma etching

A model has been developed in an attempt to explain the chemistry which occurs in plasmas produced in mixtures of CF₄ and O₂. Emphasis is placed on gas-phase free radical reactions, and the predictions of the model are compared with experimental results. Dissociation rates following electron impact are deduced mainly from experimental observations although relative dissociation rates have been calculated. An important assumption of the model is that CF₂ can be produced as a primary dissociation product following electron impact. Furthermore, this process is favored over that producing CF₃ by more than a factor of 2. Experimental evidence is presented to support this assumption. Although the model agrees well with experiment on the total amount of fluorine produced, some discrepancy exists between the predicted and measured values of [F₂]. It is suggested that the higher concentrations detected in the experiments resulted from recombination of F atoms in the sampling region. The agreement for concentrations of CO₂, CO, and COF₂ is generally better than a factor of 2 over a wide range of experimental conditions.     

Analysis of relaxing laser-induced plasmas by absorption spectroscopy: Toward a new quantitative diagnostic technique

Broad-band near UV absorption spectroscopy was used to analyze atmospheric laser-induced plasmas formed on metallic and refractory targets. When the common emission spectroscopy only provides the density of the radiating atomic excited states, the technique reported in this paper is able to achieve high spatial resolution in the measurement of absolute number densities in expanding laser-induced plasmas. The reliability and the versatility of this technique, which is based on the comparison between results of the numerical integration of the radiative transfer equation and experimental spectra, were tested on different targets. The evolutions in time and space of the absolute population of the plasma species originating from metallic alloys (Al–Mg and Cu–Ni) and refractory materials (C/SiC) were achieved over large time scales. Owing to its accuracy, this absorption technique (that we call “LIPAS” for Laser Induced Plasma Absorption Spectroscopy) should bring a new and enhanced support to the validation of collisional-radiative models attempting to provide reliable evolutions of laser-induced plasmas.     

EPR investigation of plasma-chemical resist etching in O2 and O2/CF4 discharges

During the etching of AZ 1350 photoresist in O₂ and O₂/CF₄ discharges, ground-state concentrations of atoms (O, F, and H), and small radicals (OH, HO₂, RO₂) were measured in the discharge afterglow by EPR spectroscopy. In the case of CF₄/O₂ discharges, the dependence of O and F atom concentrations on the etch time reflects both surfäce oxidation and fluorination reactions in accordance with existing etch models. In the case of high-rate resist etching in pure O₂ discharges, high concentrations of product radicals (H, OH and HO₂) were detected and compared with resist free O₂/H₂O discharges. Kinetic modeling of the afterglow reactions reveals that the mean lifetime and, accordingly, the diffusion length of the etchant species O(³P) is drastically reduced in rapid reactions with OH and HO₂. The results are used to simulate both etch homogeneity and the loading effect in a simple etch model.     

On mass- and emission spectroscopic CVD process monitoring of organometallics/O2 discharges

Mass spectroscopic analysis of neutrals and ions from a deposition plasma shows that the decomposition of the organometallic precursor compounds [La(thd)₃, Cu(acac)₂, and Al isopropoxide] in the plasma .starts with the abstraction of complete ligands. The mass spectra of plasma ions sensitively indicate the incomplete oxidation of the organic fragments with increasing organometallic partial pressure. The concentration of carbon-rich ions in the oxygen-based deposition plasma correlates with the carbon content of the deposited oxide films. Specific emissions of the precursor compounds (e.g., Cu atomic lines and LaO bands) can be used to monitor the precursor partial pressure; however, there is some interference with sputter emission from the deposited films. During La₂O₃ deposition, optical emission of oxidation products (e.g., OH, CO, CO₂ bands) was used to regulate the precursor partial pressure in the discharge with a closed-loop control circuit.On leave from Friedrich Schiller University, Institute of Physical Chemistry, Lessingstrasse 10, D-O-6900 Jena, Germany.On leave from California State University, Department of Chemistry, 2555 East San Ramon Avenue, Fresno, California 93740-0070.     

A model for the etching of Si in CF4 plasmas: Comparison with experimental measurements

A model has been developed to describe the chemistry which occurs in CF₄ plasmas and the etching of Si both in the plasma and downstream. One very important feature of this model is that for discharge residence times which vary by more than an order of magnitude, the amount of CF₄ consumed is low and relatively constant. This is because the gas-phase combination reactions between F and both CF₃ and CF₂ lead to the rapid reforming of CF₄. The model predicts that CF₂ is a major species in the gas phase and that the [F] detected as a sample point downstream is a very sensitive function of [CF₂]/[F] in the discharge. Even though the calculations show that [F] in the discharge varies only slightly over the wide range of experimental conditions considered, large variations in [F] at the sample point occur because the [CF₂]/[F] ratio in the discharge changes. The concentrations of C₂F₆ and SiF₄ are predicted to within a factor of 2 over a very wide range of experimental conditions. This confirms the importance of gas-phase free radical reactions in the etching of Si.     

Hydrocarbon plasma emission as a means to monitor plasma polymerization,revisited

The degree of aromaticity of toluene plasma polymer is shown to decrease with decreasing toluene flow rate into the plasma (constant rf power). Optical emission spectroscopy (OES) of the plasma is predictive; a plasma emitting a higher relative benzyl radical signal results in the deposition of a more aromatic plasma polymer. Such results have been earlier reported in the literature. However, in this work the geometry of OES observation assured that only plasma emissions from the vicinity of the deposition site were observed. Under these conditions it is found that plasma pressure must be kept constant for the predictive nature of OES results to hold.     

On-line monitoring of a frontal chromatographic separation using inductively coupled plasma atomic emission spectroscopy

A test array is described employing a destructive analytical technique for the long-term monitoring of an industrial-scale separation process. As an example, we chose frontal chromatography as the separation and ICP/AES as the analytical method. The feed solution of the process was conveyed by a process pump via the separation unit to a sample station, where a small portion was diverted and transported by a roller pump into the spectrometer. We equipped our array with different loops for operating the process, calibrating the instrument and verifying the calibration. We obtained identical results for the different loops by absorbing the pulsation of the process pump and arranging for identical suction lines of the spectrometer pump. Based on the results, we redesigned the sample station for a technical application using only commercially available parts.     

Investigation of radical reactions in C2Cl4 and C2Cl4/O2 discharges by EPR,mass, and emission spectroscopy

The reaction of tetrachloroethylene, C₂Cl₄, with O(³P) atoms as well as the plasma decomposition of C₂Cl₄ and C₂Cl₄/O₂ mixtures have been investigated by combined application of electron paramagnetic resonance (EPR) and emission and mass spectroscopies. C₂Cl₄ plasma decomposition is shown to proceed primarily to the formation of CCl₃ radicals and chlorine-deficient products, which are ultimately involved in the formation of carbonaceous layers. A simple reaction model accounts for all the detected stable and radical species, encountered during the plasma decomposition. The model also enables order-of-magnitude estimates of decomposition rate constants to be made. The suppression of the formation of both carbonaceous layers and products C_mCl_n (m3) in C₂Cl₄/O₂ discharges is explained using results of an investigation of elementary reactions in the system C₂Cl₄/O(³P)/O₂. The stable products of C₂Cl₄/O₂ discharges, i.e., COCl₂, CCl₄, and C₂Cl₆, respectively, are shown to originate from recombination of the peroxy radicals CCl₃OO and C₂Cl₅OO.     

Characterization of laser-induced plasmas by emission spectroscopy with curve-of-growth measurements. Part I: Temporal evolution of plasma parameters and self-absorption

Laser-induced plasmas have been characterized by emission spectroscopy, including the measurement of curves of growth. The plasmas have been generated in air at atmospheric pressure using an infrared Nd:YAG laser from a set of Fe–Ni alloys with varying Fe concentrations. The procedure used provides, in addition to the apparent temperature T and electron density N_e, a parameter N′l (the atom number density for 100% concentration times the length of the plasma along the line-of-sight), relevant to obtain the self-absorption and the intensity of the emission lines. The temporal evolution of the plasma parameters has been deduced from the measurement and fitting of the curves of growth. A fast temporal decrease of N′l is obtained for ions, whereas a gradual increase takes place for neutral atoms. The temporal evolution of the line intensity in the optically thin limit and the self-absorption of neutral atom and ion lines have been obtained experimentally and calculated from the evolution of the plasma parameters. The usefulness of the curve-of-growth method in measurements with time integration, in spite of the fast variation of the plasma parameters, has been demonstrated.     

A simple line shape technique for electron number density diagnostics of helium and helium-seeded plasmas

The results of an experimental study of the He I 447.1 nm line and its forbidden component at high electron number density are presented and compared with profiles calculated using computer simulation method. Michelson interferometer at 632.8 nm was used to measure plasma electron number density in the range (1–7) × 10²³ m^− 3 while electron temperatures for the same experimental conditions in the range of 25 000 K to 35 000 K were determined using several spectroscopic techniques. The agreement of experimental overall line shape with computer simulation results is within 10% of what is well within theoretical and experimental uncertainty. This favorable comparison enabled the development of a simple approximate formula for the evaluation of electron number density from the measurement of wavelength separation between peaks of allowed and forbidden lines. This technique of plasma diagnostics is not sensitive to the presence of self-absorption of strong He I allowed line. The derivation of approximate formula with estimated accuracy of 15% was followed by detailed comparison with other experimental and theoretical data.     

Characterization of laser-induced plasmas by emission spectroscopy with curve-of-growth measurements. Part II: Effect of the focusing distance and the pulse energy

Laser-induced plasmas generated with different focusing distances and pulse energies have been characterized by a method based in emission spectroscopy that includes the measurement and calculation of curves of growth. An infrared Nd:YAG laser is used to generated the plasmas from Fe–Ni samples placed in air at atmospheric pressure. The characterization method provides a reduced set of plasma parameters (N_e, T, N′l, αA) that describe the line emission in optically thin and optically thick conditions. For a pulse energy of 100 mJ, the plasma parameters for varying focusing distances are obtained. The apparent (population averaged) temperatures for neutral atoms and ions are shown to be different in the plasmas generated with all the focusing distances. For each pulse energy (in the range 20–100 mJ), the plasmas generated with the optimum focusing distance, which corresponds to a constant value of irradiance, have been investigated. In these conditions, simple laws have been obtained for the variation of the plasma parameters with the pulse energy E: the electron density N_e and the apparent temperature T are independent of E while linear relations with E are obtained for the parameters N′l, αA. These simple laws lead to a quadratic dependence on E of the line intensities in the optically thin limit and to a variation of the intersection concentration C_int that characterizes self-absorption as E^− 1.     

The Optical Spectrum of Au2+

The electronic structure of the Au₂⁺ cation is essential for understanding its catalytic activity. We present the optical spectrum of mass-selected Au₂⁺ measured via photodissociation spectroscopy. Two vibrationally resolved band systems are observed in the 290–450 nm range (at ca. 440 and ca. 325 nm), which both exhibit rather irregular structure indicative of strong vibronic and spin-orbit coupling. The experimental spectra are compared to high-level quantum-chemical calculations at the CASSCF-MRCI level including spin-orbit coupling. The results demonstrate that the understanding of the electronic structure of this simple, seemingly H₂⁺-like diatomic molecular ion strictly requires multireference and relativistic treatment including spin-orbit effects. The calculations reveal that multiple electronic states contribute to each respective band system. It is shown that popular DFT methods completely fail to describe the complex vibronic pattern of this fundamental diatomic cation.     

Optical emission diagnostics in cascade arc plasma polymerization and surface modification processes

Optical Emission Spectroscopy (OES) was used to identify reactive species and their excitation states in low-temperature cascade arc plasmas of N₂, CF₄, C₂F₄, CH₄, and CH₃OH. In a cascade arc plasma, the plasma gas (argon or helium) was excited in the cascade arc generator and injected into a reactor in vacuum. A reactive gas was injected into the cascade arc torch (CAT) that was expanding in the reactor. What kind of species of a reactive gas, for example, nitrogen, are created in the reactor is dependent on the electronic energy levels of the plasma gas in the cascade arc plasma jet. OES revealed that no ion of nitrogen was found when argon was used as the plasma gas of which metastable species had energy less than the ionization energy of nitrogen. When helium was used, ions of nitrogen were found. While OES is a powerful tool to identify the products of the cascade arc generation (activation process), it is less useful to identify the reactive species that are responsible for surface modification of polymers and also for plasma polymerization. The plasma surface modification and plasma polymerization are deactivation processes that cannot be identified by photoemission, which is also a deactivation process. © 1998 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 36: 1583–1592, 1998