Oxygen discharges diluted with argon:
Dissociation processes
J. T. Gudmundssona,b,∗ and E. G. Thorsteinssona,b,
a
Science Institute, University of Iceland, Reykjavik, Iceland;
b
Department of Electrical Engineering, University of Iceland, Reykjavik, Iceland;
∗
[email protected]
Results and discussion
• Addition of Ar to the O2 discharge, while ashing photoresist, increases the plasma density and etch rate approximately by a factor
of 2 (Takechi and Lieberman, 2001).
• Kitajima et al. (2006) report on increased density of metastable O(1D)
atoms in highly argon diluted oxygen plasmas due to quenching of
metastable argon atoms.
• These claims are explored using a global (volume averaged) model.
• Electron impact dissociation of the O2 molecule accounts for roughly
25 – 30 % of the O(3P) creation
• The chamber is assumed to be made of stainless steel, cylindrical with
R = 10 cm and L = 10 cm.
• The applied power is 500 W, the gas flow rate 50 sccm.
• The metastable oxygen atom is created mainly through electron impact excitation of the oxygen atom, roughly 75 – 65 % contribution,
decreasing with increased argon content.
1
0.9
(a)
• We find the contribution of dissociation by quenching of the argon
metastable Arm by molecular oxygen to be negligible in the creation
of atomic oxygen, both O(3P) and O(1D).
0.8
0.7
Ri/ ΣRi
0.5
0.3
quenching by O2
electron impact dissociation
0.2
0.1
7
deexcitation
0.4
6
25 % Ar
wall quenching
5
dissociative attachment
20
40
50 % Ar
75 % Ar
dissociation by Arm
0
0
60
80
100
e
[Ar] / ([Ar] + [O2]) [%]
T
• We use a global (volume averaged) model to study the dissociation
processes and the presence of negative ions and metastable species in
a low pressure high density O2/Ar discharge.
• In addition to electrons the oxygen discharge consists of molecular oxy1∆ ),
gen in ground state O2(3Σ−
),
metastable
molecular
oxygen
O
(a
g
2
g
3Σ+, A3∆ , c1Σ−), atomic oxygen in ground state
)
and
O
(A
O2(b1Σ+
u
2
g
u
u
O(3P), metastable atomic oxygen O(1D), ozone O3, the positive ions
−, O− and O−.
O+ and O+
and
the
negative
ions
O
2
2
3
• We use a significantly revised reaction set for the oxygen discharge
(Gudmundsson, 2004)
• The argon discharge consists of argon atom in the ground state
Ar(3s23p6), metastable argon Arm (the metastable levels 1s5 and 1s3),
radiatively coupled levels Arr (the levels 1s4 and 1s2), Ar(4p) and positive argon ions Ar+.
• Electrons are assumed to have a Maxwellian energy distribution in the
range 1 – 7 eV.
0.6
4
3
2
1
0.9
(b)
1
0.8
0
0
10
0.7
0.6
Ri/ ΣRi
The global (volume averaged) model
• The destruction of O(3P) is mainly through wall recombination and
electron impact excitation.
[V]
Introduction
0.5
Figure 5: The electron temperature versus pressure. The applied
power is 500 W, the gas flow rate 50 sccm and the argon content 25,
50 and 75 %.
electron impact excitation
0.3
0.2
electron impact ionization
0.1
pumped out
0
0
0.9
2
10
wall recombination
0.4
1
1
10
p [mTorr]
20
40
60
80
100
[Ar] / ([Ar] + [O2]) [%]
0.8
100
Figure 3: The reaction rates for (a) the creation of the O(3P) atom
and (b) the loss of the O(3P) atom versus fractional argon flowrate
[Ar]/([Ar] + [O2]).
γ
0.6
0.5
0.4
0.3
0.2
0.1
0
1
0
50
100
p [mTorr]
(a)
150
0.9
Ri/ ΣRi
0.4
0.3
0.2
0.1
0
0
60
50
40
30
20
20
40
60
[Ar]/([Ar] + [O ]) [%]
80
100
dissociation by Arm
dissociative attachment
20
40
60
80
100
[Ar] / ([Ar] + [O ]) [%]
2
Conclusions
• The electron temperature increases with increased argon content due
to the higher ionization potential of argon compared to atomic and
molecular oxygen.
(b)
0.8
0.6
Ri/ ΣRi
Pabs = 1200 W
Figure 6: The fractional dissociation versus the fractional argon
flow rate.
electron impact dissociation of O2
• The fractional dissociation increases with increased argon content due
to increased electron impact dissociation with higher electron temperature
0.7
atomic oxygen density [m−3]
70
0.5
0.9
20
Pabs = 500 W
2
1
10
80
excitation
0.6
Comparison to experiments
Pabs = 300 W
0
0
0.7
• The recombination coefficient of oxygen atoms at the chamber walls is
given by a best fit through the measured data in the pressure range 2
– 150 mTorr. For the pressure below 2 mTorr we assume the recombination rate coefficient increases linearly with decreasing pressure from
0.5 at 2 mTorr to 1 at vacuum.
90
10
0.8
Figure 1: The recombination coefficient of oxygen atoms at the
chamber walls for stainless steel as a function of pressure. The measured data is taken from, o Singh et al. (2000), × Matsushita et
al. (1997), △ Mozetic̆ and Zalar (2000), Booth and Sadeghi (1991)
and ∗ Gomez et al. (2002).
[0] / ([O] + [O2])
0.7
deexcitation
3
0.5
quenching by O( P)
0.4
0.3
0.2
Acknowledgments
wall quenching
0.1
0
0
19
10
ionization
pumped out
20
40
60
80
100
[Ar] / ([Ar] + [O ]) [%]
2
18
10
0
20
40
60
p [mTorr]
80
100
Figure 2: The absolute density of atomic oxygen as a function of
pressure argon oxygen ratio [Ar]/[O2] = 0.63/0.37. The measured
data + is compared to global model calculation (solid line). The
applied power was 150 W and the gas flow rate 33.5 sccm. The
inductive discharge chamber was made of stainless steel 20 cm in
diameter and 10 cm long. The measured data is taken from Hsu
et al. (2006).
Figure 4: The reaction rates for (a) the creation of the O(1D) atom
and (b) the loss of the O(1D) atom versus fractional argon flowrate
[Ar]/([Ar] + [O2]).
• The majority of the ground state atomic oxygen O(3P) originates from
the metastable state O(1D)
• About 20 – 40 %, depending on the argon content, is created by electron impact deexcitation of the metastable atom O(1D).
• Quenching of O(1D) by the O2 molecule accounts for over 30 % of the
O(3P) creation for pure oxygen discharges and falls to become negligible with increased argon content.
This work was partially supported by the Icelandic Research Fund and
the University of Iceland Research Fund.
References
J. T. Gudmundsson, A critical review of the reaction set for a low
pressure oxygen processing discharge, Technical Report RH-17-2004,
Science Institute, University of Iceland (2004).
C.-C. Hsu, M. A. Nierode, J. W. Coburn, and D. B. Graves, Journal of
Physics D: Applied Physics 39, 3272 (2006).
T. Kitajima, T. Nakano, and T. Makabe, Applied Physics Letters 88,
091501 (2006).
K. Takechi and M. A. Lieberman, Journal of Applied Physics 90, 3205
(2001).