Influence of chemical composition on radiations of microwave discharge lamp

22nd International Symposium on Plasma Chemistry
July 5-10, 2015; Antwerp, Belgium
Influence of chemical composition on radiations of microwave discharge lamp
M. Hamady1 and G. Zissis2
1
2
Department of Physics, Faculty of Sciences, Lebanese University, El-Hadath, Beirut, Lebanon
Université de Toulouse; LAPLACE (Laboratoire Plasma et Conversion d’Energie); Toulouse, France
Abstract: This paper study the influence of chemical composition on the photometric
properties of a conventional metal halide (MH) plasma lamp containing Hg doped with TlI
sustained by a microwave (mw) electromagnetic field. The plasma is assumed to be fully
mixed, implying a constant mercury/thallium ratio throughout the discharge tube. Plasma
composition, radial gas temperature, luminous efficacy and CCT have been computed.
Keywords: plasma lamps, ray tracing, plasma chemical composition
2. Chemical composition calculation
The radiation transport in HID lamp clearly depends on
the chemical composition of the discharge. For pure Hg
discharges, the plasma composition is relatively simple to
be calculated compared to other Hg doped discharges.
The discharge studied here contains mercury doped with
thallium iodide, and a low pressure (a few Torr) of argon
as a buffer gas. The chemical composition of the plasma
depends on the quantities of each material present in the
lamp, as well as the temperature profile.
In the
temperature range between 1000 K to 6000 K, the
chemical species present in the plasma are the monatomic
species (Hg, Tl and I), the molecular species (HgI, HgI 2 ,
TlI and I 2 ) and the charged species (Hg+, Tl+, I− and e).
To determine the profiles of densities and partial
pressures of the eleven chemical species present in the
discharge, it is necessary to have eleven independent
P-I-2-25
equations. These relationships are derived from the laws
governing the state of the plasma in local thermodynamic
equilibrium.
Thus, chemical equilibrium and ionic
equations lead to seven equations. In addition four
equations
expressing
the
charge
conservation,
conservation of total pressure, and also the two atomic
ratios: the mercury to thallium ratio (R 1 ) and the thallium
to iodine ratio (R 2 ).
The atomic ratio R 1 = mercury/thallium is expressed by:
R1 =
Hg PHg + PHgI + PHgI 2 + PHg + =constant
=
PTl + PTlI + PTl +
Tl
(1)
The atomic ratio R 2 = thallium/iodine is expressed by:
R2 =
PTl + PTlI + PTl +
Tl
=1
=
I
PHgI + 2 PHgI 2 + PI + 2 PI 2 + PTlI + PI −
(2)
The chemical species present in the plasma are shown in
Fig. 1 for R 1 = 148.
5ensity of species ( m-3 )
1. General
The development of new, efficient electrodeless HID
light sources, based on the “resonant cavity” [1] concept
(so-called plasma lamps) has aroused considerable
interest in recent years. When electrical power in the
lamp is sufficiently high, these lamps can achieve higher
luminous efficacies than conventional HID lamps [1],
with photometric properties comparable to or exceeding
those of the current generation of LED lamps. Further,
plasma lamps are considerably smaller than either the
equivalent LED arrays or conventional HID lamps, and
therefore provide excellent optical control. Theoretical
analysis of plasma lamps has to date been limited to pure
Hg discharges [1, 2]. This paper provides an initial
approach to understanding the performance of the
photometric properties of MH plasma lamps once the
chemical composition of the plasma changes. Gas
temperature profile of a microwave sustained HgTlI
discharges has been calculated using a simple numerical
model [1]. A ray tracing model [3] has been used later to
compute the radiation flux emitted by all visible spectral
lines. The influence of the variation of the chemical
composition on the photometric properties of a
conventional metal halide is therefore discussed.
1025
Hg
1024
TlI
1023
I
1022
Tl
Hg+
1021
Tl+ HgI
1020
1019
e-
I-
HgI2
I2
1018
1017
0.0
0.2
0.4
0.6
0.8
Reduced Radius
1.0
Fig. 1. Plasma composition.
1
3. Temperature profile calculation
In the calculations described here, a simple “skin depth”
numerical model [1] was used to compute the radial
temperature profile in a microwave generated discharge at
frequency 2.45 GHz. The lamp’s parameters (radius
R = 3.9 mm, inter-electrode distance L = 7.2 cm) and the
net emission coefficient are obtained from Bouaoun [4].
The discharge was assumed to be in LTE in the T 010 mode
and the gas temperature was computed from the
simplified Elenbaas- Heller equation
∇ ⋅ κ g ∇Tg + σ e E z2 − U rad = 0
(3)
where T g is the gas temperature, Κ g is the coefficient of
thermal conductivity of the gas, σ e is the electrical
conductivity, E z (r) is the axial electric field at radius r,
computed from the “skin depth” equation [1] and U rad
represents the net energy transported by radiation from
each point in the discharge. In the calculations, Κ g and σ e
were calculated self consistently and values of U rad (T g )
were taken Bouaoun [4]. The total electrical power in the
discharge is
Table 1. Influence of changing the chemical composition
of plasma.
Discharge
W elec (W)
R1 = 100
R1 = 150
R1 = 200
214
214
214
R
0
where n e is the electron density and W rad and W therm
represent the electrical power dissipated as radiation and
heat respectively.
x 10
5613
5580
5376
22
R1=200
R1=150
R1=100
5
Density of Tl (m-3)
(4)
CCT (K)
5. Discussion
The influence of changing the chemical composition of
the plasma was presented. The density of Hg seems to be
dominant in the discharge as shown in Fig. 1 and stay
relatively unchanged for different atomic ratio R1.
However, the density of thallium increases once the
atomic ratio R1 decrease (Fig. 3) and play an important
role in improving the photometric properties of the lamp
as shown in Table 1. This is a direct consequence of the
difference between the ionization energies where the
energy of ionization Tl is relatively low compared to that
of Hg (E Tl = 6.11 eV and E Hg = 10.43eV).
6
Welec = 2πRL ∫ neσ e E 2z (r )rdr = Wrad + Wtherm
η (lm/W)
101
93
86
4
3
2
1
4. Results
The obtained temperature profile is consistent with
operation of plasma lamps at low power, where the short
skin depth prevents the microwave power from
penetrating into the discharge [1] as shown in Fig. 2.
Temperature (K)
6000
5000
4000
3000
2000
1000
0
0.2
0.6
0.4
Reduced Radius
0.8
1
Fig. 2. Temperature profile of the discharge.
The calculated gas temperature profile was then used as
input into a 3D radiation model [3] to compute the
radiation flux from the visible spectra of Hg and Tl. The
obtained results of luminous efficacy and CCT for
different atomic ratios R1 are summarised in Table 1.
2
0
0
0.2
0.6
0.4
Reduced Radius
0.8
1
Fig. 3. Density of Tl for different atomic ratio.
The results described in this paper provide an initial
approach to understanding the performance of MH
plasma lamps containing contains mercury doped with
thallium iodide. The net emission coefficient for this
discharge as a function of temperature was included in the
solution to the Elenbaas-Heller equation, and this was
coupled with the “skin effect” model for the electric field,
developed by Waymouth [1], to compute the radial gas
temperature profile in 2.45 GHz microwave generated
discharges containing Hg and TlI. A radiation transport
model was then used to calculate the visible radiation flux
from these discharges. These results are, however, based
on a relatively simple numerical model, and the
development of a more sophisticated model, together with
experimental measurements should provide more
quantitative information on the performance of mw
discharges. Planned improvements to the model include
inclusion of molecular radiation in the HgTlI model and
the inclusion of a solution to Maxwell’s equations to
replace the current “skin effect” model for the electric
field.
P-I-2-25
6. References
[1] J.F. Waymouth.
Microwave Discharges:
Fundamentals and Applications. (C.M. Ferreira and
M. Moisan; Eds.) (New York: Plenum Press) 427
(1993)
[2] S. Offermanns. J. Appl. Phys., 67, 115 (1990)
[3] M. Hamady, G.G. Lister, M. Aubès and G. Zissis.
J. Phys. D: Appl. Phys., 44, 105201 (2011)
[4] M. Bouaoun, H. Elloumi, L. Troudi, A. Chammam,
K. Charrada and M. Stambouli. J. Phys. D: Appl.
Phys., 43, 185205 (2010)
P-I-2-25
3