Generation of atomic iodine via fluorine for chemical oxygen-iodine laser

Article abstract of DOI:10.1016/j.chemphys.2007.02.023

A method of the chemical generation of atomic iodine for a chemical oxygen-iodine laser (COIL) using atomic fluorine as a reaction intermediate was studied experimentally. This method is based on the reaction between F₂ and NO providing F atoms, and the reaction of F with HI resulting in iodine atoms generation. Atomic iodine was produced with efficiency exceeding 40% relative to initial F₂ flow rate. This efficiency was nearly independent on pressure and total gas flow rate. The F atoms were stable in the reactor up to 2 ms. An optimum ratio of the reactants flow rates was F₂:NO:HI = 1:1:1. A rate constant of the reaction of F₂ with HI was determined. The numerical modelling showed that remaining HI and IF were probably consumed in their mutual reaction. The reaction system was found suitable for employing in a generator of atomic iodine with its subsequent injection into a supersonic nozzle of a COIL.

Full text of DOI:10.1016/j.chemphys.2007.02.023

Chemical Physics 334 (2007) 167–174
www.elsevier.com/locate/chemphys
Generation of atomic iodine via fluorine for chemical
oxygen–iodine laser
a,
a
a
a
*
ˇ
ˇ
´
´
´
´
Vıt Jirasek , Otomar Spalek , Miroslav Censky , Irena Pickova ,
a
b
´
Jarmila Kodymova , Ivo Jakubec
a
Institute of Physics, Academy of Sciences, Na Slovance 2, 182 21 Prague, Czech Republic
b
ˇ
Institute of Inorganic Chemistry, Academy of Sciences, 250 68 Rezˇ, Czech Republic
Received 12 December 2006; accepted 28 February 2007
Available online 12 March 2007
Abstract
A method of the chemical generation of atomic iodine for a chemical oxygen–iodine laser (COIL) using atomic fluorine as a reaction
intermediate was studied experimentally. This method is based on the reaction between F2 and NO providing F atoms, and the reaction
of F with HI resulting in iodine atoms generation. Atomic iodine was produced with efficiency exceeding 40% relative to initial F2 flow
rate. This efficiency was nearly independent on pressure and total gas flow rate. The F atoms were stable in the reactor up to 2 ms. An
optimum ratio of the reactants flow rates was F2:NO:HI = 1:1:1. A rate constant of the reaction of F2 with HI was determined. The
numerical modelling showed that remaining HI and IF were probably consumed in their mutual reaction. The reaction system was found
suitable for employing in a generator of atomic iodine with its subsequent injection into a supersonic nozzle of a COIL.
ꢀ 2007 Elsevier B.V. All rights reserved.
Keywords: Atomic iodine; Atomic fluorine; Chemical oxygen–iodine laser
1. Introduction
In conventional COIL, atomic iodine is produced by disso-
ciation of molecular iodine, and energy of 4–6 molecules of
singlet oxygen O2(1Dg) is consumed in this process [3]
P
A process of chemical generation of atomic iodine was
studied with the aim to use it for the chemical oxygen–iodine
laser (COIL) operation. This laser is the shortest wavelength
high-energy chemical laser operating on the electronic
transition of the 2P term of iodine atom at 1.315 lm [1,2]
1
3
I2 þ nO2ð DgÞ ! 2I þ nO2ð Þ; n ¼ 4–6:
ð3Þ
g
The energy of O2(1Dg) saved by avoiding this process might
be utilized for the laser pumping and thus a laser power in-
crease (by 25%) [4]. Another drawback of using molecular
iodine as a precursor of iodine atoms in the COIL opera-
tion is the fast quenching reaction [5]
2
2
Ið P1=2Þ ! Ið P3=2Þ þ hm:
ð1Þ
The population inversion on this transition is maintained
by a near resonant collisional energy transfer from the
metastable excited oxygen, O2(1Dg), to iodine atom in the
2
2
Ið P1=2Þ þ I2 ! Ið P3=2Þ þ I2;
ground state
k4 ¼ 3:5 ꢀ 10ꢁ11 cm3 molecꢁ1 sꢁ1
:
ð4Þ
P
2
1
2
3
Ið P3=2Þ þ O2ð DgÞ ! Ið P1=2Þ þ O2ð Þ;
Gaseous molecular iodine prior to its dissociation to atoms
is produced by evaporation of solid or liquid iodine. Con-
trolling of this process is difficult due to a time-varying
interfacial area and low heat conductivity of solid or liquid
iodine. Besides, the overall iodine management must be
heated to 80–100 ꢁC to prevent solid iodine precipitation.
g
k2 ¼ 7:6 ꢀ 10ꢁ11 cm3 molecꢁ1 sꢁ1
:
ð2Þ
*
Corresponding author. Tel.: +420 266 052 737; fax: +420 286 890 527.
´
E-mail address: jirasek@fzu.cz (V. Jirasek).
0301-0104/$ - see front matter ꢀ 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.chemphys.2007.02.023

168
V. Jira´sek et al. / Chemical Physics 334 (2007) 167–174
Two methods of atomic iodine generation that avoid
Atomic iodine may be also partially lost in ternary reac-
tions with other molecules [22–24,14]
using of molecular iodine were studied. The first one uti-
lizes a decomposition of some iodine donors in electrical
discharge [6–8]. A significant drawback of this method is
a strict requirement for a minimum discharge volume due
to a very fast homogenous and heterogenous recombina-
tion of iodine atoms. The second, chemical method is based
on releasing iodine atoms from hydrogen iodide in its fast
reaction with F or Cl atoms produced in gaseous chemical
reaction [9–11]. In our laboratory, the chemical generation
(using Cl atoms) was for the first time successfully applied
for the supersonic COIL operation [12]. A laser power of
430 W with chemical efficiency of 13% was achieved.
An experimental generation of atomic iodine via atomic
fluorine produced by reaction F2 + H2 was published in
[11]. Another source of F atoms for atomic iodine produc-
tion was studied in our laboratory [13]. The present paper
is a continuation of this work.
IþIþN2 ! I2 þN2
0:07
k14 ¼ 6:15ꢀ10ꢁ34ðT=298Þ expð7433=RTÞ cm6 molecꢁ2 sꢁ1 ð14Þ
IþIþI2 ! I2 þI2
k15 ¼ 2:68ꢀ10ꢁ31ðT=298Þꢁ3:06 expð6568=RTÞ cm6 molecꢁ2 sꢁ1 ð15Þ
IþIþHI ! I2 þHI
k16 ¼ 8:13ꢀ10ꢁ32ðT=298Þꢁ3:5 cm6 molecꢁ2 sꢁ1
IþNOþN2 ! INOþN2
ð16Þ
k17 ¼ 1:8ꢀ10ꢁ32ðT=298Þꢁ1 cm6 molecꢁ2 sꢁ1
;
ð17Þ
and in the reaction with formed INO [25]
I þ INO ! I2 þ NO k18 ¼ 1:66 ꢀ 10ꢁ10 cm3 molecꢁ1 sꢁ1
:
ð18Þ
Iodine atoms can also react with molecular fluorine, and
recombine on the walls [26,27]
2. Description of the reaction system
I þ F2 ! IF þ F k19 ¼ 1:9 ꢀ 10ꢁ14 cm3 molecꢁ1 sꢁ1 ð19Þ
I þ I þ wall ! I2 þ wall c20 ¼ 1:
ð20Þ
The atomic iodine generation proceeds in a two-step
reaction process. In the first step, atomic fluorine is formed
in the reaction of nitrogen oxide with molecular fluorine
[14]
3. Summary of the previous investigations on this reaction
system
NO þ F2 ! NOF þ F
k5 ¼ 7 ꢀ 10ꢁ13 expðꢁ1150=TÞ cm3 molecꢁ1 sꢁ1
:
ð5Þ
The results of measurements presented in [13] were
obtained with the flow rates of reactive gases in the range
of 50–500 lmol sꢁ1, and at total pressures of 1.5–9 kPa.
All three reactants, F2, NO, and HI were diluted with N2
in the ratio 1:9. A maximum yield of iodine atoms related
to the initial F2 flow rate, I/F2,0 = 0.27, was attained at
approximately equimolar ratio of flow rates of reaction
gases. The yield decreased with pressure and was only
slightly lower with an excess of NO over F2. On the other
hand, an excess of F2 over NO significantly reduced a life-
time of I atoms due to the reaction (19). This effect was
compensated by the generation of further iodine atoms
from HI and F atoms produced in (19), when also HI
was in an excess over NO. With increasing flow rates of
reactants, a final temperature of the reaction mixture
increased from 320 K to 700 K.
In the present work, higher flow rates of reactants (up to
1.5 mmol sꢁ1) and higher pressures (up to 20 kPa) were
employed, in order to adequately simulate the conditions
needed for the COIL operation. The generation of atomic
iodine outside the laser cavity with a subsequent injection
of I atoms into the COIL cavity downstream the supersonic
nozzle throat is planned. In comparison with previous exper-
iments, a flow rate of diluting gas (nitrogen) was reduced in
order to limit the loss of F and I atoms in ternary reactions.
In the second step, the formed fluorine atoms react with
hydrogen iodide to atomic iodine [15]
2
F þ HI ! HF þ Ið P3=2
Þ
k6 ¼ 1:5 ꢀ 10ꢁ10 expðꢁ5057:8=RTÞ cm3 molecꢁ1
s
ꢁ1: ð6Þ
As the reaction product HF is a strong quencher of I(2P1/2
)
[16], it is necessary to replace HI by DI in the laser system
[17]:
2
2
HF þ Ið P1=2Þ ! HF þ Ið P3=2
Þ
k7 ¼ 1:3 ꢀ 10ꢁ11 cm3 molecꢁ1 sꢁ1
ð7Þ
ð8Þ
2
2
DF þ Ið P1=2Þ ! DF þ Ið P3=2
Þ
k8 ¼ 1:05 ꢀ 10ꢁ14 cm3 molecꢁ1 sꢁ1
:
Formation of F atoms is accompanied by several loss-pro-
cesses [18–21]:
F þ NO þ NO ! NOF þ NO
k9 ¼ 1:8 ꢀ 10ꢁ31ðT=298Þꢁ1 cm6 molecꢁ2 sꢁ1
F þ NO þ N2 ! NOF þ N2
ð9Þ
k10 ¼ 1:8 ꢀ 10ꢁ31ðT=298Þꢁ1 cm6 molecꢁ2 sꢁ1
F þ F þ N2 ! F2 þ N2
ð10Þ
k11 ¼ 6 ꢀ 10ꢁ34 cm6 molecꢁ2 sꢁ1
ð11Þ
ð12Þ
ð13Þ
4. Experimental
F þ F þ wall ! F2 þ wall c12 ¼ 0:05
Two different reactors were used. The cross-section of
the first reactor is shown in Fig. 1. This experimental set-
F þ I2 ! IF þ I k13 ¼ 4:3 ꢀ 10ꢁ10 cm3 molecꢁ1 sꢁ1
:

V. Jira´sek et al. / Chemical Physics 334 (2007) 167–174
169
through the wedge-shaped windows perpendicularly to
the gas flow. The windows were coated with a thin layer
of Teflon in order to protect glass against HF. The probe
beam emitter/detector unit was mounted on an assembly
of the motorized linear positioning equipment controlled
by PC. In this configuration, the probe beam could move
along the cell on the path within a distance from 12 mm
to 60 mm from the axis of the injectors 4 (dotted area in
Fig. 1). The probe beam moved with a velocity of 4 mm sꢁ1
in the direction of the gas stream. The gas temperature was
measured with a jacketed Ni–CrNi thermocouple 6 (1 mm
o.d.) inserted perpendicularly to the gas flow 5 mm down-
stream from the end of the windows. The gas exiting the
reactor was exhausted with a rotary pump (capacity
25 m3 hꢁ1) through a liquid nitrogen trap, where most of
the NOF, IF, HI, NO, and I2 was condensed out of the
gas flow. The rotary pump operated with oil chemically
inert to fluorine, type NC 1/14 (Leybold Ltd., Germany),
consisting of perfluorinated polyether. The residual HF in
the exhausting gas was adsorbed in a column filled with
aluminosilicates.
Fig. 1. Scheme of atomic iodine reactor with movable NO injector. 1 –
inlet NO/N2, 2 – inlet F2/N2, 3 – region of F atom generation, 4 – HI/N2
injectors, 5 – optical cavity for atomic I detection, 6 – thermocouple, 7 –
input flange, 8 – delta-shaped channel.
up enabled to vary a time delay between NO and HI injec-
tions into F2 flow thanks to the axial shifting of NO injec-
tor. The tube injector 1 made of stainless steel with the
outer diameter 8 mm was used for introducing NO/N2 mix-
ture (10% NO) through 24 holes (1.1 mm i.d.) drilled near
the injector end in three rows along the tube perimeter.
This injector was inserted into a delta-shaped entrance
space of the reactor body having the rectangular cross-sec-
tion. The F2/N2 mixture (10% F2) was introduced through
the ring-shaped insert 2, and the HI/N2 mixture (9.6% HI)
through two injectors 4 into the rectangular reactor cavity
(having a cross-section of 8 · 25 mm). The second part of
this reactor equipped with two side windows served for
the optical detection of atomic iodine.
The molar flow rate of atomic iodine nI was calculated
from the equation
n_I ¼ cIV =NA;
ð21Þ
The cross-section of the second reactor is shown in
Fig. 2. The device is composed of a coaxial reactor 3 for
F atoms generation (F-reactor) inserted into a space of
delta-shaped entrance of the rectangular reactor where
atomic iodine is produced (I-reactor). The reactor 3 con-
sists of the inner tube for NO inlet 1 (outer diameter
4 mm), and the outer tube for F2 inlet. In this arrangement,
non-diluted NO was injected into the F2/N2 flow through 6
holes (0.3 mm i.d.) drilled in one row along the tube perim-
eter. Fluorine atoms were generated in the space between
these holes and exit from the F-reactor. The gas with F
atoms exited the reactor 3 through 24 holes (1.1 mm i.d.)
drilled along the perimeter of the outer tube in three rows
distant 1.5 mm between each other.
A concentration of iodine atoms was measured by
means of a special method called Iodine Scan Diagnostics
(ISD) [28]. This diagnostics based on the near IR spectros-
copy employs a narrow-band tunable diode probe laser for
monitoring gain or absorption on the I*–I transition at
1.315 lm. Gas temperature (in following text denoted as
TISD) can be calculated from the Gaussian full width after
deconvolution of the Voigt profile of measured absorption
– frequency curves. In our device, the probe beam passed
where a local concentration (number density) of atomic io-
dine, cI, was evaluated from ISD data for the central line of
the gas channel, NA is the Avogadro constant and V is a
volumetric flow rate of gas mixture through the detection
cell. This was evaluated from the modified equation of state
V ¼ n_totRT =p;
ð22Þ
where n_tot is the total molar flow rate representing the sum
of molar flow rates of all gas components, T is the gas tem-
perature, and p is the pressure in the detection cell.
5. Experimental results
5.1. Measurements in the reactor with movable NO injector,
and 10% mixtures of reacting gases with nitrogen
In this experimental configuration (Fig. 1), atomic
iodine production was studied for different F2, NO and
HI flow rates, and different time interval between NO
and HI injection into F2 flow. Fig. 3 shows an example
of I concentration recorded along the optical cell for differ-
ent HI flow rates. These data were measured at a time
interval of 0.64 ms between NO and HI injections. Under
these conditions, a maximum concentration of atomic
iodine was measured at a distance of 15–30 mm from the
HI injection, corresponding to a time interval of 0.5–
1 ms. The ascending parts of the curves obviously reflect
increasing conversion of reacting gases into I atoms. The
descending parts are due to the loss reactions. A higher
iodine concentration at a higher HI flow rate follows from
the acceleration of reaction (6). A decrease in this con-
centration at increased HI flow rate is explained by an
Fig. 2. Scheme of atomic iodine reactor with embedded reactor for
generation of F atoms. 1 – inlet NO, 2 – inlet F2 + N2 mixture, 3 –
generator of F atoms, 4 – HI injectors, 5 – optical cavity for atomic I
detection, 6 – thermocouple, 7 – input flange, 8 – delta-shaped channel.

170
V. Jira´sek et al. / Chemical Physics 334 (2007) 167–174
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
5.2. Measurements with embedded F reactor, 10% mixtures
of F2 and HI, and non-diluted NO
1
2
3
4
This device (Fig. 2) was designed to simulate a model sit-
uation in COIL where F atoms are generated in a separate
reactor and then injected together with DI into the primary
flow of singlet oxygen. The non-diluted NO was used to
reduce the unfavorable effect of N2 diluent on the rate of
ternary loss reactions. Besides, the temperature is higher
at lower dilution resulting in the acceleration of the binary
production reactions and deceleration of the ternary loss
reactions. The pressure pF in the F-reactor was calculated
using a formula for the molar flow rate through the sonic
orifice [29]
0
10
20
30
40
50
60
70
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
jþ1
ꢀ
ꢁ
distance from HI injector, mm
jꢁ1
jRT F
W
2
n_ ¼ cdpFA
;
ð23Þ
Fig. 3. Concentration profile of atomic iodine along gas flow at different
HI flow rate, 484 lmol sꢁ1 F2 and 484 lmol sꢁ1 NO; 1 – 119 lmol sꢁ1 HI,
2 – 206 lmol sꢁ1 HI, 3 – 308 lmol sꢁ1 HI, 4 – 460 lmol sꢁ1 HI, pcell = 3.7–
5.2 kPa, sNO–HI = 0.64 ms, sHI–ISD = 0.45–2 ms.
j þ 1
where TF, W, and j are the total temperature, the molecu-
lar weight and the adiabatic constant of the F2 + NO reac-
tion mixture, cd is the discharge coefficient (equal to 0.6 for
holes with sharp edges), and A is the total area of the injec-
tor holes. The temperature TF in the F reactor was esti-
mated using the enthalpy-balance from the temperature
at the position of the HI injection evaluated by the extrap-
olation of the TISD course along the optical cell.
An example of the I concentration profile along the opti-
cal cell for different HI flow rates is shown in Fig. 5. The
time interval between the NO admixing and the gas exit
from the F-reactor (sNO–F) was 14 ls, the interval between
the F-reactor exit and HI admixing, sF@HI, was 0.8 ms.
Although the maximum iodine concentration measured in
this experimental configuration was only slightly higher
than in the previous one (Fig. 3) for similar F2 and NO
flow rates, its decay with a distance from the HI injector
was slower.
6
5
4
3
2
1
2
3
1
0
0
10
20
30
40
50
60
70
distance from HI injector, mm
4.0
3.5
3.0
2.5
2.0
Fig. 4. Concentration profile of atomic iodine along gas flow at different
NO–HI time interval, 608 lmol sꢁ1 F2, 642 lmol sꢁ1 NO and
502 lmol sꢁ1 HI, pcell = 7.6–9 kPa, sNO–HI = 0.64 ms (curve 1), 0.8 ms
(curve 2), and 1.015 ms (curve 3), sHI–ISD = 0.45–2 ms.
increasing effect of the ternary loss reactions (10), (14), and
(17), as the N2 concentration in the reactor increased
together with the HI flow rate. A prolonging the interval
between NO and HI injection up to ꢂ1 ms had no signifi-
cant effect on atomic iodine production (see Fig. 4). In this
experimental arrangement and at pressure above 5 kPa, the
yield of atomic iodine was rather low. It ranged from 3% to
13% relative to the initial F2 flow rate, and from 8% to 35%
relative to the HI flow rate. The yield was a bit lower at
higher flow rates of reaction gases due to ternary reactions
including N2. The temperature measured by the thermo-
couple was 550–650 K. The temperature evaluated from
the ISD measurements was rather scattered in the range
400–900 K.
1.5
1.0
1
2
3
4
5
0.5
0.0
0
10
20
30
40
50
60
70
distance from HI injector, mm
Fig. 5. Concentration profile of atomic iodine along gas flow at different
HI flow rate, 459 lmol sꢁ1 F2 and 433 lmol sꢁ1 NO; 1 – 172 lmol sꢁ1 HI,
2 – 188 lmol sꢁ1 HI, 3 – 292 lmol sꢁ1 HI, 4 – 294 lmol sꢁ1 HI, 5 –
333 lmol sꢁ1 HI, pF = 6.6–8.1 kPa, pcell = 3.3–3.9 kPa, sNO–F = 14 ls,
s
NO–HI = 0.8 ms, sHI–ISD = 0.45–2 ms.

V. Jira´sek et al. / Chemical Physics 334 (2007) 167–174
171
8
7
6
5
4
3
2
1
0
14
F2
0.40 0.59 0.48
0.81 1.1 0.8
1.52 1.69 1.53
NO
HI
1
2
3
12
10
8
3
6
2
1
4
1
2
3
2
0
0
100
200
300
400
500
600
700
800
900 1000 1100
0
10
20
30
40
50
60
70
nHI mol s-1
distance from HI injector, mm
Fig. 7. Effect of F2, NO and HI flow rate (mmol/s) on iodine concen-
Fig. 6. Concentration of atomic iodine in dependence on HI flow rate,
tration. pF = 4.4–17.3 kPa, pcell = 1–3 kPa,
sNO–F = 0.01 ms, sF–HI =
794 lmol sꢁ1 F2 and 718 lmol sꢁ1 NO, pF = 14 kPa, pcell = 7–8 kPa,
0.29 ms.
sNO–F = 14 ls (1), 90 ls (2), 240 ls (3), sF–HI = 1.1 ms, sHI–ISD = 1 ms.
did not result in a substantial decrease in the atomic iodine
yield. The number density of I atoms reached a high value
of 1 · 1016 cmꢁ3 at a pressure of 3 kPa. This value
increased up to 4.2 · 1016 cmꢁ3 at a pressure of 12.5 kPa,
which was attained by choking the vacuum pipe at the
gas exit. The atomic iodine generation was not affected
by increasing the time interval sF–HI up to 1.14 ms
(Fig. 8). It confirmed the results of modelling that the life-
time of F atoms in given reaction mixture is much longer
than the lifetime of Cl atoms in the system of atomic iodine
generation via Cl atoms [9,10]. A production of F atoms is
affected however by a substantial extension of the reaction
time in the F reactor – see Table 1. This effect is caused by
higher pressure in this reactor (8–21 kPa), and thus an
acceleration of the ternary loss-processes.
A dependence of the I concentration on the HI flow rate
is shown in Fig. 6. These data were measured at different
time intervals sNO–F. The ISD probe was at a fixed position
corresponding to a time interval sHI–ISD of 1 ms. It can be
seen that the atomic iodine production was not changed
when the time interval sNO–F was increased from 14 ls to
240 ls. This fact demonstrates rather long lifetime of F
atoms. A decrease in the I concentration with increasing
HI flow rate beyond a certain value demonstrates the effect
of ternary loss reactions when diluted HI is used.
In this experimental arrangement, the yield of atomic
iodine was 8–15% relative to the F2 flow rate, and 11–
37% relative to the HI flow rate. These values are slightly
higher in comparison with measurements performed with
the diluted NO, which is ascribed to a reduced effect of ter-
nary reactions rather than to an effect of increased temper-
ature (by 50 K only).
7
6
5
4
3
5.3. Measurements with non-diluted NO and HI
The results of measurements presented above clearly
indicated that a further increase in the atomic iodine yield
required a reduction in the diluent (N2) content in the reac-
tion mixture. Therefore, non-diluted NO and HI were used
together with 20% F2/N2 mixture in next sets of experi-
ments. In the first series, the NO was introduced through
the inner injector (1 in Fig. 2), the F2/N2 mixture through
the inlet 2, and the HI gas through two injectors 4. Fig. 7
shows the effect of different flow rates of reactants on the
measured iodine concentration. The results in Fig. 7 dem-
onstrate that the iodine concentration increased substan-
tially with increasing the flow rates of all reactants. A
twofold increase in flow rates resulted in approximately
1.7 times higher iodine concentration. It means that, in
contrary to the experiments with 10% reactants HI and
NO, a higher gas pressure due to higher reactant flow rates
2
1
2
3
1
0
0
10
20
30
40
50
60
70
distance from HI injector, mm
Fig. 8. Effect of time interval between gas exit from F reactor and HI
injection (sF–HI)
1–0 ms, 2–0.29 ms, 3–1.14 ms; 0.81 mmol sꢁ1 F2,
1.04 mmol sꢁ1 NO, 0.8 mmol sꢁ1 HI, pF = 8.5 kPa, pcell = 1.8 kPa,
NO–F = 0.16 ms.
:
s

172
V. Jira´sek et al. / Chemical Physics 334 (2007) 167–174
Table 1
Dependence of atomic iodine yield on time interval sNO–F between NO admixing and exit from F-reactor
pcell
(kPa)
pF
(kPa)
sNO–F
(ms)
sF–HI
(ms)
TISD
(K)
TF
(K)
nF2
(mmol sꢁ1
nNO
(mmol sꢁ1
nHI
(mmol sꢁ1
[I] (·1015
)
I/F2,0
(%)
)
)
)
(cmꢁ3
)
1.75
1.76
4.00
3.85
9.31
8.57
20.73
18.76
0.3
6.4
0.3
1.09
1.09
1.09
1.09
1050
900
1140
970
1157
996
1324
1130
0.82
0.81
1.53
1.51
1.04
1.04
1.63
1.63
0.75
0.84
2.07
2.25
6
6
16
14.7
35.7
30.9
46.8
38.7
19.5
The experiments with more concentrated reactants
showed that a lifetime of F atoms is high even at a rather
high pressure (20 kPa in the F-reactor). The yield of iodine
atoms, 26–47% (related to F2), was much higher than with
diluted reactants. The gas temperature evaluated from ISD
measurement was from 900 K to 1500 K.
In order to evaluate the influence of reactions (24a),
(24b) on the composition of the reaction mixture, a simpli-
fied, constant-pressure, quasi-1-D kinetic modelling was
applied. The kinetic equations were integrated together
with the enthalpy-balance equation. The finite mixing of
reactants was modelled by means of the second-order
‘‘reactions’’
A theoretical work [30] predicted the possibility of the
reaction
F2;unmixed þ NOunmixed ! F2;mixed þ NOmixed
F2 þ HI ! products:
ð24Þ
k23 ¼ 1 ꢀ 10ꢁ14 cm3 molecꢁ1 sꢁ1
ð26Þ
ð27Þ
Because the rate constant of the reaction (24) was not
found in literature, we tried to estimate its value experi-
mentally. In this experiment, the HI gas was introduced
through the inner injector (labeled 1 in Fig. 2), the F2/N2
mixture through the inlet 2, and the NO through the injec-
tors 4. It was found that the concentration of atomic iodine
detected in the optical cavity decreased 7.4 times when the
premixing time of F2 + HI in the F-reactor was prolonged
from 0.01 ms to 0.88 ms. Assuming that this drop is caused
by the reaction (24) only, the second-order rate constant
k21 = 3.7 · 10ꢁ14 cm3 molecꢁ1 sꢁ1 was estimated from
these data. Owing to a rather high value of this constant,
the reaction (24) has to be considered in theoretical analysis
of this reaction system.
HIunmixed þ HIunmixed ! HImixed þ HImixed
k24 ¼ 1 ꢀ 10ꢁ14 cm3 molecꢁ1 sꢁ1
The rate constants k23 and k24 were adjusted to attain a
fast, but still rate-determining mixing, as it followed from
our previous modelling experience.
The modelling was performed for following input data:
the initial temperature 300 K, the total pressure 15 kPa and
time of 0.05 ms between NO and HI admixing into F2 flow.
Three different values of ratio of the gases flow rates were
considered, F2:NO:HI:N2 = 1:1:x:4, where x = 0.5, 0.75
and 1, respectively. The model included the reactions (5),
(6), (9)–(19), and (24). Four different sets of reaction rate
constants k21a, k21b, given in Table 2, were examined.
Relative concentrations resulting from these calcula-
tions are shown in Figs. 9–12. The IF concentration
decreased dramatically from the case 1 to the case 4 for
the ratio F2:HI = 1:1 (see curves 1A–4A in Fig. 9a). When
the HI flow rate was decreased, the difference between cases
1–4 significantly decreased and was nearly zero at
F2:HI = 1:0.5 (see Fig. 9b). This behaviour was explained
as follows. At certain conditions, a very fast reaction (13)
(k13 = 4.3 · 10ꢁ10) serves as a ‘‘backward’’ reaction to
(24b); the IF is formed predominantly in (19) and its rate
of formation is not affected by the rate of (24b) up to
k21 = 1 · 10ꢁ12. Such conditions occur when sufficient con-
centration of F atoms for the reaction (13) is at disposal.
By increasing the HI flow rate, F atoms are more fastly
consumed in (6), reaction (13) does not apply and the reac-
6. Numerical modelling and discussion
It is supposed [30] that the reaction (24) has two-steps
F2 þ HI ! HF þ IF
IF þ HI ! HF þ I2;
ð24aÞ
ð24bÞ
where k21b ꢃ k21a. From this assumption and our estima-
tion of k21a followed that the reaction (24b) is sufficiently
fast to remove both HI (unreacted with F in reaction (6))
and IF (produced in reactions 13, 19). It is desired to keep
the IF concentration low during the COIL operation be-
cause of the fast quenching of I* atoms [31]
2
2
IFþIð P1=2Þ ! IFþIð P3=2Þ k22 ¼ 1:3ꢀ10ꢁ11 cm3 molecꢁ3 sꢁ1
:
ð25Þ
Table 2
Values of parametrized rate constants used in numerical modelling
On the other hand, the reaction (24b) may be beneficial in a
COIL because HI quenches O2(1Dg), probably via the HO2
radical, as we experimentally proved recently [32]. It also
follows from our estimation of k21 that the premixing of
F2 and HI at higher pressures prior to the NO injection
should be avoided.
Case
k21a (cm3 molecꢁ1 sꢁ1
)
k21b (cm3 molecꢁ1 sꢁ1
)
1
2
3
4
0
0
1 · 10ꢁ15
1 · 10ꢁ14
1 · 10ꢁ14
1 · 10ꢁ14
1 · 10ꢁ13
1 · 10ꢁ12

V. Jira´sek et al. / Chemical Physics 334 (2007) 167–174
173
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
0.30
0.25
0.20
0.15
0.10
0.05
0.00
1A
2A
3A
4A
1A
3A
1B
3B
1C
3C
0.0
0.5
1.0
1.5
2.0
0.0
0.5
1.0
1.5
2.0
Time (ms)
Time (ms)
Fig. 9a. Calculated relative concentration of IF in dependence of time
after HI admixing. 1, 2, 3, 4 – cases corresponding to Table 2. A –
F2:HI = 1:1.
Fig. 10. Calculated relative concentration of I. Key as in Fig. 9.
is, however, given by a relative ratio of the rates of the pro-
duction reaction (6) vs. loss reactions for F atoms (9)–(11).
At lower HI flow rate, the rate of (6) is lower and the loss of
F atoms from (9)–(11) is higher. This fact is confirmed by a
higher production of NOF at F2:HI = 1:0.5 (Fig. 11).
The difference between curves 1A and 3A in Fig. 12
demonstrates the effect of reactions (24a) and (24b) and
the ratio F2:HI on HI concentration:HI unreacted with F
atoms is consumed in reactions with F2 and IF. At a low
excess of HI over F atoms, the effect of reactions (24a),
(24b) is however negligible (compare the curves 1C and
3C).
The modelling revealed that in the case of fast reactions
(24a), (24b), the ratio of flow rates F2:HI = 1:1 is advanta-
geous both for atomic iodine production and a removal of
IF and excessive HI.
The experimental and modelling results of our work may
be compared with results published by Helms et al. [33]. In
their work, F atoms produced in the reaction (5) were mixed
with I2 to produce vibrationally excited IF. A low depen-
dence of atomic fluorine production on the pressure and
Fig. 9b. Calculated relative concentration of IF in dependence on time
after the HI admixing; 1, 2, 3, 4: the cases corresponding to Table 2; B –
F2:HI = 1:0.75; C – F2:HI = 1:0.5.
1.0
0.8
0.6
0.4
tion (24b) significantly reduces the IF concentration. At
given modelling conditions (a time interval between HI
admixing to F2 + NO), this change in behaviour occurs
at F2:HI = 1:0.5.
The calculations further showed that the rate constants
k21a and k21b have only a small effect on the production of
atomic iodine – see Fig. 10.
It was found experimentally that decreasing the HI:F2
or HI:NO flow rate ratio below 0.75 results in a decrease
in the iodine concentration, in spite of the fact that [I]/
[HI]0 < 0.4 (see Fig. 6). A sudden change in the first deriv-
ative of [I]/[F2]0 may be observed in Fig. 10 at a time
around 0.25–0.35 ms. Before this time, a steep rise on the
curve may be attributed to a very fast reaction (6) involving
F atoms produced before HI is completely entrained to the
flow. A following slow increase in the I atoms production is
driven by a slower reaction (5). The maximum of I atoms
1A
3A
1C
3C
0.2
0.0
0.0
0.5
1.0
1.5
2.0
Time (ms)
Fig. 11. Calculated relative concentration of NOF. Key as in Fig. 9.

174
V. Jira´sek et al. / Chemical Physics 334 (2007) 167–174
4027), and the Czech Science Foundation (Grant. No. 202/
05/0359).
References
[1] W.E. McDermott, N.R. Pchelkin, D.J. Benard, R.R. Bousek, Appl.
Phys. Lett. 32 (1978) 469.
[2] R.G. Derwent, B.A. Trush, Chem. Phys. Lett. 9 (1971) 591.
[3] J.F. Hon, D.N. Plummer, P.G. Crowell, J. Erkilla, G.D. Hager, C.A.
Helms, AIAA Paper 94-2422, 1994.
[4] M. Endo, D. Sugimoto, H. Okamoto, K. Nanri, T. Uchiyama, S.
Takeda, T. Fujioka, Jpn. J. Appl. Phys. 39 (2000) 468.
[5] J.I. Cline, S.R. Leone, J. Phys. Chem. 95 (1991) 2917.
[6] H. Okamoto, T. Hirata, K. Shinoda, D. Sugimoto, M. Endo, T.
Fujioka, AIAA Paper 2000-2492.
[7] B. Quillen, W. Schall, in: CD Proceedings of COIL R&D Workshop,
Stuttgart, Germany, 2003, DLR Stuttgart, Germany, 2003, p. 1.
[8] P.A. Mikheyev, A.A. Shepelenko, A.V. Voronov, N.V. Kuprayev,
Russ. J. Quantum Electron. 32 (2002) 1.
Fig. 12. Calculated relative concentration of HI. Key as in Fig. 9.
ˇ
ˇ
´
´
´
[9] V. Jirasek, O. Spalek, J. Kodymova, M. Censky, Chem. Phys. 269
flow rates at low dilution by N2 measured by these authors
is in accordance with our results. They obtained the F yield
of 20–30% relative to the initial F2 flow rate. We have found
that this efficiency can exceed 40% under conditions where
the third gas consuming the F atoms (i.e. HI in our case)
is admixed shortly after F2 + NO mixing place.
(2001) 167.
ˇ
ˇ
´
´
´
[10] O. Spalek, V. Jirasek, M. Censky, J. Kodymova, I. Jakubec, G.D.
Hager, Chem. Phys. 282 (2002) 147.
[11] K.B. Hewett, G.D. Hager, P.G. Crowell, Chem. Phys. 308 (2005)
159.
ˇ
ˇ
´
´
[12] O. Spalek, M. Censky´, V. Jirasek, J. Kodymova, I. Jakubec, G.D.
Hager, IEEE J. Quantum Electron. 40 (2004) 564.
ˇ
ˇ
´
´
[13] M. Censky´, O. Spalek, V. Jirasek, J. Kodymova, I. Jakubec, Collect.
7. Conclusions
Czech. Chem. Commun. 71 (5) (2006) 739.
[14] D.L. Baulch, J. Duxbury, S.J. Grant, D.C. Montague, J. Phys. Chem.
Ref. Data 10 (Supplement) (1981) 1.
[15] N. Johnatan, C.M. Melliar-Smith, S. Okuda, D.H. Slater, D. Timlin,
Mol. Phys. 22 (1971) 561.
[16] R.D. Coombe, A.T. Pritt Jr., J. Chem. Phys. 66 (1977) 5214.
[17] A.T. Pritt Jr., D. Patel, J. Chem. Phys. 81 (3) (1984) 1337.
[18] W.B. DeMore, S.P. Sander, D.M. Golden, R.F. Hampson, M.J.
Kurylo, C.J. Howard, A.R. Ravishankara, C.E. Kolb, M.J.
MolinaChemical Kinetics and Photochemical Data for Use in
Stratospheric Modeling. Evaluation Number 12, vol. 97-4, JPL
Publication, 1997, p. 1.
[19] C.J. Ultee, Chem. Phys. Lett. 46 (1977) 366.
[20] G.P. Kota, J.W. Coburn, D.B. Graves, J. Vac. Sci. Technol. A 17
(1999) 282.
[21] E.H. Appelman, M.A.A. Clyne, J. Chem. Soc., Faraday Trans. 1
(1975) 2072.
[22] J.K.K. Ip, G.J. Burns, Chem. Phys. 56 (1972) 3155.
[23] J.A. Blake, G.J. Burns, J. Chem. Phys. 54 (1971) 1480.
[24] H.W. Chang, G. Burns, J. Chem. Phys. 64 (1976) 349.
[25] H. Van Den Bergh, J.J. Troe, Chem. Phys. 64 (1976) 736.
[26] H.V. Lilenfeld, G.R. Bradburn, J. Phys. Chem. 91 (1987) 1881.
[27] G.P. Perram, G.D. Hager, Air Force Weapons Laboratory Technical
Report, AFWL-TR-88-50, October, 1988.
[28] S.J. Davis, M.G. Allen, W.J. Kessler, K.R. McManus, M.F. Miller,
P.A. Mulhall, SPIE Paper 2702-17, 1996.
[29] Handbook of Fluid Dynamics, CRC Press, New York, 1998.
[30] X.-Y. Han, Z.-Y. Wang, L.-S. Wang, C.-S. Yang, B. Yu, Acta Phys.–
Chim. Sinica 20 (11) (2004) 1364.
A method of the chemical generation of atomic iodine
for a COIL using atomic fluorine as an intermediate species
was studied. The results showed that by using a low-diluted
F2 and non-diluted NO and HI, atomic iodine can be pro-
duced with a rather high efficiency, i.e. with the yield
exceeding 40% relative to the initial F2 flow rate. This effi-
ciency is only weakly dependent on the pressure and resi-
dence time of F atoms in the reactor up to 2 ms. In the
studied system the lifetime of F atoms is significantly
longer than the lifetime of chlorine atoms during I atoms
generation via Cl. Only extension of the (F2 + NO) reac-
tion time (>5 ms) at a high pressure (>8 kPa) lowered the
production of F atoms. The optimum ratio of reactant flow
rates is F2:NO:HI = 1:1:1. A rough estimation of the rate
constant of the reaction of F2 with HI was determined from
the experimental data with
a
value of 3.7 · 10ꢁ14
cm3 molecꢁ1 sꢁ1. The numerical modelling showed that
an excessive HI and the reaction product IF are probably
consumed in their mutual reaction. A number density of
atomic iodine generated in this reaction system can be very
high (up to ꢂ4 · 1016 cmꢁ3 at 20 kPa), which is sufficient
for the COIL operation provided that a subsequent injec-
tion of iodine atoms from the side reactor into the super-
sonic nozzle of this laser will follow.
[31] M.A. Chowdury, A.T. Pritt, D. Patel, D.J.J. Benard, Chem. Phys. 84
(1984) 6687.
ˇ
ˇ
´
´
´
´
[32] V. Jirasek, O. Spalek, M. Censky, J. Kodymova, I. Pickova, I.
Jakubec, in: Proceedings of the XVI International Symposium on Gas
Flow and Chemical Lasers and High Power Lasers Conference, 2006,
vol. 6236, Gmunden, Austria, in press.
Acknowledgments
This work was supported by the USAF European Office
for Research and Development (Grant No. FA 8655-05-M-
[33] C.A. Helms, L. Hanko, K. Healy, G. Hager, G.P. Perram, J. Appl.
Phys. 66 (12) (1989) 6093.