Measuring the rate constant of the reaction between chlorine atoms and CHF2Br by Cl atom resonance fluorescence

Article abstract of DOI:10.1134/S0023158416030101

The rate constant of the reaction between Cl atoms and CHF₂Br has been measured by chlorine atom resonance fluorescence in a flow reactor at temperatures of 295–368 K and a pressure of ~1.5 Torr. Lining the inner surface of the reactor with F-32L fluoroplastic makes the rate of the heterogeneous loss of chlorine atoms very low (k_het ≤ 5 s^–1). The rate constant of the reaction is given by the formula k = (4.23 ± 0.13) × 10^–12e^{(–15.56 ± 1.58)/RT} cm³ molecule^–1 s^–1 (with the activation energy in kJ/mol units). The possible role of this reaction in the extinguishing of fires producing high concentrations of chlorine atoms is discussed.

Full text of DOI:10.1134/S0023158416030101

ISSN 0023ꢀ1584, Kinetics and Catalysis, 2016, Vol. 57, No. 3, pp. 308–312. © Pleiades Publishing, Ltd., 2016.
Original Russian Text © I.K. Larin, A.I. Spasskii, E.M. Trofimova, N.G. Proncheva, 2016, published in Kinetika i Kataliz, 2016, Vol. 57, No. 3, pp. 308–312.
Measuring the Rate Constant of the Reaction between Chlorine
Atoms and CHF2Br by Cl Atom Resonance Fluorescence
I. K. Larina, A. I. Spasskiia, E. M. Trofimovaa, *, and N. G. Pronchevab
aInstitute of Energy Problems of Chemical Physics, Russian Academy of Sciences,
Moscow , 117829 Russia
bKeldysh Institute of Applied Mathematics, Russian Academy of Sciences, Moscow, 125047 Russia
*eꢀmail: eltrofimova@yandex.ru
Received August 19, 2015
Abstract—The rate constant of the reaction between Cl atoms and CHF2Br has been measured by chlorine
atom resonance fluorescence in a flow reactor at temperatures of 295–368 K and a pressure of ~1.5 Torr. Linꢀ
ing the inner surface of the reactor with Fꢀ32L fluoroplastic makes the rate of the heterogeneous loss of chloꢀ
rine atoms very low (khet
≤
5
s–1). The rate constant of the reaction is given by the formula
k
= (4.23 0.13)
×
10–12e(–15.56
RT cm3 molecule–1 s–1 (with the activation energy in kJ/mol units). The possible role of this
1.58)/
reaction in the extinguishing of fires producing high concentrations of chlorine atoms is discussed.
Keywords: chlorine atom, refrigerants, resonance fluorescence, reaction rate constant, fire extinguishing
DOI: 10.1134/S0023158416030101
Saturated hydrocarbons that contain bromine atoms was prohibited as well. Although the ODP of CHF2Br
(refrigerants) are usable in fire extinguishing [1]. The is as low as 0.73 [8], the contribution from this comꢀ
hypothetical extinguishing mechanism [2] is based on pound to ozone layer depletion cannot be neglected
the inhibition of chain combustion reactions by a refrigꢀ either, because the chain length in ozone destruction
erant. The inhibitor reacts with H atoms, which are cycles involving bromine up to an altitude of 40 km far
chainꢀbranching agents in combustion:
exceeds the chain length in chlorine cycles. In the
lower stratosphere, the chain lengths differ by a factor
of several tens or hundreds [9]. Nevertheless, CHF2Br
is still employed in fire extinguishing in exceptional
cases (e.g., coal mines and naval vessels).
H• + RBr
→
HBr + R•,
H2O + Br•,
(I)
(II)
HBr + OH•
→
HBr + O•
→
OH• + Br•.
(III)
The compound that was earlier used particularly
widely in fire extinguishing is CF3Br, which is the most
effective refrigerant from the standpoint of both comꢀ
bustion inhibition and flame speed reduction, espeꢀ
cially in fire emergencies in naval vessels and warꢀ
planes. However, the manufacturing of this refrigerant
was banned by the Montreal Protocol in 1987 [3] for
the reason that it has a very high ozone depletion
potential (ODP). Its ODP is 16 [4], which means that
this compound can reach the stratosphere and destroy
ozone molecules through chain reactions.
In view of this fact, researchers began seeking alterꢀ
native reagents that would not be so harmful to the
ozone layer but would be sufficiently effective in fire
extinguishing. CHF2Br turned out to be the best alterꢀ
native [5]. Its advantage over CF3Br is that its molecule
contains a hydrogen atom, which allows it to react with
the OH• radical [6, 7] to yield the highly active radical
Note the following additional advantage of
CHF2Br over CF3Br. The radical R• that results from
reaction (I) can subsequently react either with molecꢀ
ular bromine resulting from the recombination of broꢀ
mine atoms or with molecular oxygen:
R• + Br2
→
RBr + Br•,
(IV)
i
R• + O2 + M
→
+ M.
(V)
RO2
Reaction (IV) regenerates molecules of the fireꢀextinꢀ
guishing agent, and reaction (V) leads to the decay of
R• radicals. Investigating the photolysis of CHF2Br
and CF3Br, we determined the ratio of the rate conꢀ
stants of reactions (IV) and (V) [11]. For CF3Br
kIV
fire extinguishing, CHF2Br can be used at lower conꢀ
centrations than CF3Br
It is of interest to study the reaction between
/kV = 0.47; for CHF2Br kIV/kV = 4.8. Therefore, in
.
CF2Br•, which can subsequently react with an oxygen CHF2Br and chlorine atoms. The appearance of chloꢀ
molecule to turn into a lowꢀactivity radical. This rine atoms in the atmosphere is mainly due to their
diminishes the probability of this refrigerant reaching formation from sea salt aerosols in coastal and oceanic
the upper atmosphere. However, later this compound regions of our planet as a result of the interaction of
308

MEASURING THE RATE CONSTANT OF THE REACTION
309
sodium chloride with ozone and other oxidizers [12], at 118.9 nm. Emitted photons were detected in the
but there are anthropogenic sources of chlorine atoms 117–133.8 nm range with a homemade photoionizaꢀ
as well. Chlorine atoms are emitted into the tropoꢀ tion counter. The counter was filled with a mixture of
sphere by municipal solid wasteꢀfired combined heat argon and nitric oxide (230 and 10 Torr, respectively).
and power plants [13] and by the combustion of orgaꢀ The longꢀwavelength detection limit of the counter
nochlorine materials used in industrial construction. (133.8 nm) was determined by the NO ionization
Another possible source of chlorine atoms is ignition potential (–8.7 eV [18]). The lamp and counter were
at municipal and industrial waste dumping sites [14]. fabricated using MgF2 glass, so the shortꢀwavelength
The concentration of Cl atoms in the atmosphere can detection limit was 117 nm. This allowed us to use a
reach ~105 molecule/cm3, which is only a few times chlorine line at 118.9 nm, taking measurements in a
lower than the concentration of OH• radicals [15]. At spectral range in which there would be almost no
the same time, the rate constants of the reactions of absorption by dioxygen even if the latter were present
CHF2Br with chlorine atoms, which play a significant in the reactor. The signal from the counter was passed
role in atmospheric chemistry, can be one order of through a frequency meter operating in the pulse
magnitude large than the rate constants of the reacꢀ counting mode and was then input into a computer for
tions between this refrigerant and OH• [16].
accumulation and subsequent processing.
A signalꢀtoꢀnoise ratio of 2 was attained at a chloꢀ
rine atom concentration of 1.2
Here, we report measurement of the rate constant
of the reaction between the chlorine atom and CHF2Br
for the reason that both reactants can be present in the
air during fire extinguishing at power plants and other
industrial enterprises. As was mentioned above, chloꢀ
rine atoms result from the combustion of polymers,
which are widely used in the construction of these
objects. Furthermore, the reaction between Cl and
CHF2Br can occur in fire extinguishing in coal mines,
since the chlorine content of coal can be up to 0.3–
0.7% and, in the pyrolysis of the latter, chlorine can
pass practically entirely into the gas phase.
×
1010 atm/cm3. Absoꢀ
lute Cl sensitivity calibration was carried out by titraꢀ
tion with C2H6, which was described in detail in our
earlier publication [19]. Measurements demonstrated
that the chlorine atom concentration is proportional
to the flow rate of molecular chlorine being passed
through the reactor. The coefficient of proportionality
depended on the discharge conditions, so the system
needed to be calibrated anew in each experiment.
Reactor Design and Introducing the Reactants
The reactor that we used was a quartz cylinder with
an inner diameter of 1.7 cm (Fig. 1). The inner surface
of the reactor was lined with Fꢀ32l fluoroplastic to
diminish the rate of the heterogeneous loss of atoms
and radicals. The diluent gas (helium) and CHF2Br
were introduced into the reactor through side holes.
The mass flow rates of the reactants and diluent gas
were determined by measuring the pressure drop in a
calibrated volume in a certain time. The pressure variꢀ
ation was monitored with a reference manometer.
Highꢀpurity helium was used in all experiments. The
molecular chlorine supply line consisted of two glass
tubes and Teflon shutoff valves. No vacuum grease was
used here.
EXPERIMENTAL
Measurement Methods and Obtaining the Reactants
Chlorine atoms were detected by the atomic resoꢀ
nance fluorescence (RF) method in the photon countꢀ
ing mode.
The experimental setup consisted of a source of
chlorine atoms, a chlorine atom detection system, a
reactor, and a reactant feeding system. Chlorine atoms
were generated by passing heliumꢀdiluted molecular
chlorine through a Breid resonator. The source of
microwave energy was a Luchꢀ3M microwave therapy
device (Russia), whose output power was 2.5 W.
The chlorine atom detection zone was consisted of
tubes soldered crosswise into the reactor. The tubes
ended with quartz sockets, in which the brass cones of
a resonance lamp, photon counter with collimators,
and Wood’s horn were secured using a vacuum grease.
Molecular chlorine was synthesized by oxidizing
HCl with potassium permanganate and was purified by
lowꢀtemperature distillation. It was stored in glass botꢀ
tles and was introduced through a capillary into the He
stream being passed through the resonance lamp and
the source of chlorine atoms. Ethane to be used in the
calibration of the absolute chlorine atom sensitivity of
the system was stored in a glass bottle and was directly
introduced into the reactor through a side inlet.
The source of resonance radiation was a flowꢀ
through lamp operating on a Cl2 + He mixture at conꢀ
centrations of
1
×
1013 and
1
×
1017 molecule/cm3.
Discharges in the lamp were initiated with a Breid resoꢀ
nator energized from the aboveꢀmentioned Luchꢀ3M
generator. The body of the resonance lamp was made
from a quartz tube with a diameter of 6 mm and a wall
thickness of 1 mm. The pressure in the lamp was varied
RESULTS AND DISCUSSION
Chlorine atoms mixed with helium were introꢀ
between 0.5 and 1 Torr and was recorded with a Sapfir duced into the reactor through a movable tube, with
24 D manometer (Russia) with an accuracy of the airtightness of the joint ensured by using a Teflon
0.05 Torr. The lamp emitted a chlorine resonance line gasket. The RF signal of chlorine atoms along the
KINETICS AND CATALYSIS Vol. 57
No. 3 2016

310
LARIN et al.
the assumption that chlorine atoms can react both with
CHF2Br and with the reactor wall, the total Cl atom
Microwave resonator
Cl2, He
consumption rate is
d[Cl]
dt
(1)
= −kVI[CHF2Br][Cl] − khet[Cl],
O2
where kVI is the rate constant of the bimolecular reacꢀ
tion between Cl and CHF2Br (cm3 molecule–1 s–1) and
khet is the rate constant of the heterogeneous loss of
chlorine on the reactor wall (s–1).
By dividing all terms of Eq. (1) by [Cl] and inteꢀ
grating the result with respect to time, we obtain
C2H6
CHF2Br
ln[Cl] = –
The following designations are used here:
tant contact time, which is equal to z is the disꢀ
v ;
0
k
VI[CHF2Br]
τ
–
khet
τ
+ const.
(2)
τ
is the reacꢀ
z
/
tance between the end of the tube through which
atomic chlorine is introduced and the detection zone;
is the linear velocity of the reactants, determined
v
0
To manometer
from the Qtotal
is the total flow rate of the diluent gases and reactants
in the reactor; is the pressure in the reactor (Torr);
is the crossꢀsectional area of the reactor (cm2).
Note that the flow velocity in the reactor did not
=
p S relationship; Qtotal (cm3 Torr s–1)
v
0
p
S
Detection zone
exceed 700 cm/s and the Reynolds number was
below 50, so the flow in the reactor was certainly lamꢀ
Fig. 1. Schematic of the reactor.
inar.
If no CHF2Br is fed into the reactor, Eq. (1) will
appear as
ln(J0/J)
0.10
0.08
0.06
0.04
0.02
d[Cl]0
(3)
= −khet[Cl]0.
dt
After division of Eq. (3) by [Cl]0 and integration,
we obtain
ln[Cl]0 = –khet
where [Cl]0 is the concentration of chlorine atoms at a
zero CHF2Br concentration.
τ
+ const,
(4)
After simple rearrangements, expression (2) will
then appear as
[Cl]0
(5)
ln
= kVI[CHF2Br]τ.
[Cl]
Since the RF signal from chlorine atoms,
J
, is proꢀ
~ [Cl], Eq. (5)
0
0.005
0.010
0.015
0.020
0.025
, s
portional to their concentration, i.e.,
turns into
J
τ
Fig. 2. ln(
J
/
J
)
as a function of
τ
at a temperature of 295 K,
3.7
J0
0
(6)
ln = kVI[CHF2Br]τ.
a gas pressure of 1.5 Torr in the reactor, [Cl]
≈
×
J
12
3
14
10 molecule/cm , and [CHF Br] = 5.9
×
10 moleꢀ
2
J0
3
Figure 2 plots
as a function of τ for a temperꢀ
ln
cule/cm .
J
ature of 295 K. The pressure in the reactor was
1.5 Torr, the concentration of chlorine atoms was
reactor axis (J0) was measured at different distances
from the detection zone. Next, CHF2Br diluted with
helium or oxygen from a calibrated vessel was introꢀ
duced into the reactor, and it reacted there with chloꢀ
rine atoms:
Cl• + CHF2Br
The consumption of chlorine atoms in reaction (VI)
was monitored by measuring their RF signal ( ). Under stant of reaction (VI) in the literature. This values was
~3.7
×
1012 molecule/cm3, and the CHF2Br concenꢀ
tration was 5.9
×
1014 molecule/cm3.
The rate constant of reaction (VI) derived from the
slope of the straight line plotted in Fig. 2 is
→
HCl + CF2Br•.
(VI)
k
VI = (7.1 0.5)
We found only one numerical value of the rate conꢀ
×
10–15 cm3 molecule–1 s–1
.
J
KINETICS AND CATALYSIS Vol. 57
No. 3 2016

MEASURING THE RATE CONSTANT OF THE REACTION
311
It was of interest to compare our data with the rate
constants of the reaction between CHF2Br and the
OH· radical:
ln(
3.4
k
1015) [cm3 molecule–1 s–1]
×
3.2
3.0
2.8
2.6
2.4
2.2
2.0
1.8
CHF2Br + OH•
→
CF2Br• + H2O.
(VII)
The following expression was reported [7] for the rate
constant of reaction (VII) experimentally measured in
the temperature range from 298 to 460 K:
k
VII = (9.3 1.03)
(7)
×
10–13 –11.06/RT cm3 molecule–1 s–1.
e
CONCLUSIONS
2.6
2.8
3.0
3.2
3.4
1000/
The rate constants of reaction (VI) measured in this
study and the rate constants of reaction (VII) calcuꢀ
lated via Eq. (7) are listed in the table. Clearly, kVII at
any temperature is only slightly larger than the rate
constant of reaction (VI). Therefore, the rate of the
reaction between CHF2Br and the chlorine atom will
exceed the rate of the reaction between CHF2Br and
OH· only when the concentration of chlorine atoms is
higher than the hydroxyl concentration. This situation
is possible in the Earth’s coastal and oceanic regions,
where chlorine atoms result from the interaction of
sodium chloride with ozone and other oxidizers, as
was mentioned above. Note, however, that process
(VII) is preferable in this case, since it yields a neutral
product (H2O), while reaction (VI) yields hydrogen
chloride, a corrosive substance that can cause acid
rains. Reaction (VI) can be advantageous in fire extinꢀ
guishing when high chlorine atom concentrations are
generated by the burning material, because in this case
replacement of toxic atomic chlorine by hydrogen
chloride, which can then be removed with water, is a
way of reducing the concentration of poisonous subꢀ
stances in the fire zone.
T
, K–1
Fig. 3. Rate constant of the reaction between the chlorine
atom and CHF Br as a function of temperature in the
295–358 K range.
2
calculated for a temperature of 296 K by correlating
experimental data with a complex scheme of the reacꢀ
tion [20]:
k
VI = (5.8 0.98) .
10–15 cm3 molecule–1 s–1
×
This value is in good agreement with the value
obtained by us and with theoretical calculations by
Zhang et al. [21], who arrived at the following relaꢀ
tionship:
k
VI = 4.8 e
10–12 –16.48/RT cm3 molecule–1 s–1
×
(Hereafter, the activation energy is in kJ/mol units.)
At 296 K, the value of this constant is
k
VI = 6.2 .
10–15 cm3 molecule–1 s–1
×
We measured the rate constant of reaction (VI) at
313, 333, and 358 K as well (Fig. 3). In this temperaꢀ
ture range, kVI is given by the expression
REFERENCES
k
VI = (4.23 0.13)
1.58)/
1. Tapscott, R.E., Sheinson, R.S., Babushok, V.,
Nyden, M.R., and Gann, R.G., NIST Technical
Note 1443, Gaithersburg, Md.: National Inst. of Stanꢀ
dards abd Technology, 2001.
×
10–12 (–15.56
e
RT cm3 molecule–1 s–1
,
which is in good agreement with theoretical data [21].
2. Noto, T., Babushok, V., Hamins, A., and Tsang, W.,
Rate constants of the reactions of CHF2Br with the chlorine
atom and OH• radical
Combust. Flame, 1998, vol. 112, p. 147.
3. Montreal Protocol on Substances That Deplete the Ozone
Layer. United Nations Environment Programme
(UNEP): 2006 Assessment Report of the Halons Techniꢀ
cal Options Committee, Nairobi, 2006.
kVI
×
1015,
kVII ×
1015,
T
, K
cm3 molecule–1 s–1
(this work)
cm3 molecule–1 s–1
(data from [7])
4. Platt, U. and Honniger, G., Chemosphere, 2003, vol. 52,
p. 325.
292
313
333
358
7.1
11.3
15.8
22.7
9.7
13.7
17.6
23.2
5. Richard, G. and Gann, R.G., National Institute of
Standards and Technology. EDP Sciences, 1998, p. 1.
6. Hsu, K.J. and DeMore, W.B., J. Phys. Chem., 1995,
vol. 99, p. 1235.
7. Orkin, V.L. and Khamaganov, V.G., J. Atmos. Chem.
1993, vol. 16, p. 169.
,
KINETICS AND CATALYSIS Vol. 57
No. 3 2016

312
LARIN et al.
8. Scientific Assessment of Ozone Depletion 2006: Global 14. Buekens, A. and Huang, H., J. Hazard. Mater., 1998,
Ozone Research and Monitoring Project, Report no. 50
,
vol. 62, p. 1.
Geneva, Switzerland, p. 572.
15. Vogt, R., Crutzen, P.J., and Sandler, R., Nature, 1996,
vol. 383, p. 323.
16. Donaghy, T., Shanahan, I., Hande, M., and Fitszꢀ
9. Larin, I.K., Khimicheskaya fizika ozonovogo sloya
(Chemical Physics of the Ozone Layer), Moscow:
GEOS, 2013.
patrick, S., Int. J. Chem. Kinet., 1993, vol. 25, p. 273.
17. Boichenko, M.V., Slashchinin, G.A., and Shakhvorosꢀ
10. Zholudev, E.R., Cand. Sci. (Eng.) Dissertation
,
tova, T.S., Aktual’nye Vopr. Sudeb. Med. Ekspert. Prakt.
,
Kemerovo: Research Inst. of Mine Rescue Work, 2006.
2011, no. 17, p. 1.
11. Larin, I.K., Belyakova, T.I., Messineva, N.A.,
18. Kikoin, I.K., Tablitsy fizicheskikh velichin (Tables of
Physical Data), Moscow: Atomizdat, 1976.
19. Larin, I.K., Spasskii, A.I., Trofimova, E.M., and Turꢀ
Spasskii, A.I., and Trofimova, E.M., Kinet. Catal.
2014, vol. 55, p. 549.
,
kin, L.E., Kinet. Catal., 2003, vol. 44, p. 202.
20. Bilde, M., Sehested, J., Mogelber, T.E., Wallington, T.J.,
and Nielse, O.J., J. Phys. Chem., 1996, vol. 100, p. 7050.
21. Zhang, H., Li, Z.S., Liu, J.Y., Sheng, L., and Sun, C.C.,
12. McFiggans, G., Cox, R.A., Mossinger, J.C., Allan, B.J.,
and Plane, J.M.C., J. Geophys. Res., 2002, vol. 107, p. 15.
13. Van den Berg, M., Birnbaum, L., Bosveld, A.T.C., Brunꢀ
strom, B., Cook, P., Feeley, M., Giesy, J.P., Hanberg, A.,
Hasegawa, R., Kennedy, S.W., Kubiak, T., Larsen, J.C.,
Van Leeuwen, F.X.R., Liem, A.K.D., Nolt, C., et al.,
Environ. Health Perspect., 1998, vol. 106, p. 775.
J. Mol. Struct.: THEOCHEM, 2004, vol. 674, p. 23.
Translated by D. Zvukov
KINETICS AND CATALYSIS Vol. 57
No. 3 2016