[ Home | Projects | Publications | People | Recent Theses | Recent Presentations | Directions | Help ]
Global Modeling of Singlet-Delta Oxygen Production in Glow Discharges with Application to Oxygen-Iodine Lasers1
D. Shane Stafford2
Mark J. Kushner3
2 Department of Chemical and Biomolecular Engineering
3 Department of Electrical and Computer Engineering
University of Illinois at Urbana-Champaign
1406 W. Green St., Urbana, IL 61801
I. Conventional Oxygen-Iodine Lasers
Chemical oxygen-iodine laser (COIL) operation on the I(2 P1/2) » I(2P3/2) electronic transition in iodine has been pursued for more than three decades because of its high efficiency and potential multi-kilowatt cw power. The relatively short 1.315 micron wavelength is especially attractive because of its high optical coupling with most common metals and high transmission in optical fiber. These properties are desirable for such uses as anti-missile defense systems and remote dismantling of nuclear reactors. The traditional chemical oxygen-iodine laser (COIL) achieves I2 dissociation and ionization in a supersonic nozzle via I2 injection into a stream of excited singlet-delta O2. Some of the excited O2 dissociates the molecular iodine; the rest then excites the atomic iodine in a near resonant transfer to create population inversion and lasing on the I(2P1/2) » I(2P3/2) transition.
Conventional COILs generate the singlet-delta O2 metastable at yields up to 70% by reaction of Cl2 in basic H2O2. However, this method is less attractive for airborne missile defense systems because of the complexity, high weight, and operational hazards associated with the liquid chemical storage and pumping systems. For airborne systems, low system weight demands and reduced complexity have driven the development of all gas phase singlet-delta O2 generators. Research has shown that singlet-delta O2 yields up to 32% can be produced by rf discharges along with significant quantities of atomic oxygen, singlet-sigma O2, and O3. This type of electrical discharge COIL has been termed the ElectriCOIL (Fig. 1).
Despite the reports of yields in 20 - 30% range, the singlet-delta O2 has not been produced at high enough pressures and flow rates at sufficiently low temperatures to insure population inversion in iodine. Thus, no ElectriCOIL experiment has yet demonstrated laser gain, and more research is needed to understand the oxygen discharge electrodynamics. At UIUC we are simulating gas phase singlet-delta O2 generation via rf discharges in pure O2 and in mixtures with inert diluents such as He.
III. GlobalKin Model
To model this complex reactive system we are using GlobalKin, a software package developed at UIUC for zero-dimensional batch plasma reaction kinetics that has been modified to simulate temporally invariant one-dimensional plug flow. Using a specified species and reaction set, GlobalKin calculates reaction rates for various neutrals and ions and integrates them along the plug flow length. In addition, various electron impact reaction rates are calculated for specified power depositions in the discharge region, where GlobalKin first solves the spherically symmetric Boltzmann equations for the electron energy distributions (Fig. 2).
The GlobalKin results are validated with experiments on an ElectriCOIL run by CU Aerospace and the UIUC Chemical Laser Group. Various parameterizations are run based on discharge length, power deposition, system pressure, diluent ratios, etc. which are then used to direct further experimental work on the ElectriCOIL. Fig. 3 shows some recent results from the simulation of a 20 cm long discharge in pure O2.
|Figure 3. Species densities and gas temperature in a 20 cm discharge in pure O2 at 3 Torr. Inlet velocity is 1000 cm/s and power deposition is 0.5 W/cm3.|
Results of a large scale parameterization have verified that singlet-delta O2 yield scales primarily with specific energy deposition in the gas, where the specific energy is taken on an inlet O2 basis. Yield increases linearly at low specific energy deposition, rising to a maximum at 5-8 eV/O2 molecule before falling off again as dissociation into O atoms dominates. Fig. 4 shows a small range of the singlet-delta O2 and atomic O yield results from the parameterization. Actual specific energy depositions ranged to 250 eV/O2 molecule, continuing the trends shown in Fig. 4.
|Figure 4. Singlet-delta O2 and atomic O yield results from a 256 run parameterization on discharge velocity, pressure, power, and diluent (He) ratio.|
Although the singlet-delta O2 yield is clearly a strong function of specific energy deposition, there is still a large amount of variance between runs at identical specific energies. This remaining scatter can be attributed to diluent, pressure, and power effects. The combined effects of O2 partial pressure and mixture ratio on singlet-delta O2 yield are shown in Fig. 5. The animation shows how the yield increases as the partial pressure of O2 increases, but in all cases the peak yield occurs when large amounts of the He diluent are present. This can be attributed to the lower reduced electric field (E/N) caused by increasing pressure and by addition of diluent, allowing the discharge to operate at a more efficient E/N for excitation of singlet-delta O2.
|Figure 5. Animation showing yields as a function of O2 mole fraction (abscissa) and specific energy deposition (ordinate) as the O2 partial pressure is increased from 0.2 - 7.2 Torr.|
Reducing the specific power deposition of the discharge also increases singlet-delta O2 yield, as shown in Fig. 6. The yield shows a peak near 1 W/cm3, suggesting that very low powered discharges are not necessary to achieve the best yields, and indeed may not be capable of sustaining the glow. Power depositions of 1 W/cm3 (0.33 W/cm3/Torr O2 in Fig. 6) are still low compared to current ElectriCOIL technology, and would require moderately long discharges (~1 m) to achieve the 5 - 8 eV/molecule O2 necessary to obtain the best yield.
Although the GlobalKin model reproduces experimental results adequately, it is still inherently a zero-dimensional model. Currently, a full one-dimensional time variant model is in development. Future research will more accurately describe the effect of specific energy deposition as well as the secondary effects of diluents, pressure, and power deposition using the 1-D model.
1This work is supported by the National Science Foundation (CTS 99-74962, CTS03-15353) and AFOSR/AFRL.