Kinetics and mechanism of the thermal gas-phase reaction between NO 2 and perfluoropropene

Article abstract of DOI:10.1524/zpch.218.5.575.30504

The reaction of NO₂ with perfluoropropene (C₃F ₆) has been studied at 413.1, 421.0 and 432.8 K, using a conventional static system. The initial pressure of NO₂ was varied between 43.6 and 204.0 Torr and that of C₃F₆ between 10.2 and 108.5 Torr. Several experiments were made adding CF₄, varying its pressure from 338.8 to 433.6 Torr. Four products were observed: NO and perfluoropropene oxide (PFPO), formed in equivalent amounts, CF₃CF(NO ₂)CF₂NO₂ and CF₃C(O)CF ₂NO₂. The relation R = [PFPO]/([CF₃CF(NO ₂)CF₂NO₂] + [CF₃C(O)CF ₂NO₂]) increases with temperature and decreases as the concentration of NO₂ and the total pressure M increase. The yields of PFPO, based on the amount of C₃F₆ consumed, varied between 63 and 89% and those of CF₃CF(NO₂)CF ₂NO₂ between 0.33 and 0.08%. Increasing the temperature, the yields of CF₃C(O)CF₂NO₂ decreased from 0.04 to 0.01%. The reaction is homogenous and the consumption rate of perfluoropropene, -d[C₃F₆]/dt = k₁ [C ₃F₆][NO₂], is independent of the total pressure. The following mechanism is postulated to explain the experimental results: 1) C₃F₆ + NO₂ → CF ₃C^?FCF₂NO₂, 2) CF ₃C^?FCF₂NO₂ → PFPO + NO, 3) CF₃C^?FCF₂NO₂ +NO₂ + M → CF₃CF(NO₂)CF₂NO₂ + M, 4) CF₃C^?FCF₂NO₂ + NO₂ → CF₃C(O)CF₂NO₂ + FNO. k₁ = (4.57 ± 1.4) × 10⁶ exp(-(15.44 ± 1.2) kcal mol^-1/RT) dm³ mol^-1 s^-1. The value of k₂, the rate constant for the unimolecular dissociation of the radical CF₃C^?FCF₂NO₂, was found to be of order of 10¹⁴ s^-1.

Full text of DOI:10.1524/zpch.218.5.575.30504

Z. Phys. Chem. 218 (2004) 575–597
by Oldenbourg Wissenschaftsverlag, München
Kinetics and Mechanism of the Thermal
Gas-Phase Reaction between NO2 and
Perfluoropropene
By Rosana M. Romano1 and Joanna Czarnowski2 ,
∗
1
CEQUINOR (CONICET, UNLP) Departamento de Qu´ımica, Facultad de Ciencias
Exactas, Universidad de La Plata, 47 y 115, 1900 La Plata, Argentina
Instituto de Investigaciones Fisicoqu´ımicas Teo´ricas y Aplicadas, Sucursal 4, Casilla
de Correo 16, 1900 La Plata, Argentina
2
(Received December 23, 2003; accepted in revised form February 19, 2004)
Gas Phase / Thermal Reaction / CF3CFCF2 / NO2 / Perfluoropropene
Oxide / CF3CF(NO2)CF2NO2 / CF3C(O)CF2NO2 / IR
The reaction of NO2 with perfluoropropene (C3F6) has been studied at 413.1,
421.0 and 432.8 K, using
a conventional static system. The initial pressure of
NO2 was varied between 43.6 and 204.0 Torr and that of C3F6 between 10.2 and
108.5 Torr. Several experiments were made adding CF4, varying its pressure from
338.8 to 433.6 Torr. Four products were observed: NO and perfluoropropene oxide
(PFPO), formed in equivalent amounts, CF3CF(NO2)CF2NO2 and CF3C(O)CF2NO2.
The relation R = [PFPO]/([CF3CF(NO2)CF2NO2]+[CF3C(O)CF2NO2]) increases with
temperature and decreases as the concentration of NO2 and the total pressure M
increase. The yields of PFPO, based on the amount of C3F6 consumed, var-
ied between 63 and 89% and those of CF3CF(NO2)CF2NO2 between 0.33 and
0.08%. Increasing the temperature, the yields of CF3C(O)CF2NO2 decreased from
0.04 to 0.01%. The reaction is homogenous and the consumption rate of perflu-
oropropene, −d[C3F6]/dt = k1[C3F6][NO2], is independent of the total pressure. The
following mechanism is postulated to explain the experimental results: 1) C3F6 +
NO2 → CF3C•FCF2NO2, 2) CF3C•FCF2NO2 → PFPO+NO, 3) CF3C•FCF2NO2 +NO2 +
M → CF3CF(NO2)CF2NO2 +M, 4) CF3C•FCF2NO2 +NO2 → CF3C(O)CF2NO2 +FNO.
k1 = (4.57 1.4)×106 exp(−(15.44 1.2)kcalmol−1/RT ) dm3 mol−1 s−1. The value of k2,
the rate constant for the unimolecular dissociation of the radical CF3C•FCF2NO2, was
found to be of order of 1014 s−1
.
*
Corresponding author. E-mail: bdq782@infovia.com.ar
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

576
R. M. Romano and J. Czarnowski
1. Introduction
The reactions of NO2 with halogenated olefins are important from synthetic
point of view. NO2 can act as nitrating and oxidating agent. The products of
these reactions are nitro haloacetyl halides and NOX, where X = Cl or F, vici-
nal dinitro compounds and nitro ketones [1–3].
The dinitro compounds O2NCF2CF2NO2 [4–6], O2NCF2CFClNO2 [5, 6],
and O2NCF2CCl2NO2 [5] have been prepared by the nitration of CF2CF2,
CF2CFCl and CF2CCl2, respectively. The formation of CF3CF(NO2)CF2NO2
and CF3C(O)CF2NO2, as products of the reaction of NO2 with perfluoro-
propene, has been reported [7, 8]. In the study of the gas-phase reactions of NO
with CF2CFCl and C3F6, in addition to the nitroso compounds, the formation of
O2NCF2CFClNO2 and O2NCF2C(O)F and that of CF3CF(NO2)CF2NO2 and
CF3C(O)CF2NO2, were observed, respectively [9]. It was attributed to the NO2
present as impurity in NO. In the reaction between NO2 and CClHCCl2 only
the oxidation product HC(O)C(O)Cl was found [10].
Taking into account the products formed and the chemical behaviour of
the olefins studied, the reactions of NO2 with alkenes are readily understood
in terms of the free radical mechanism. The elucidation of the mechanisms
of these reactions is relevant to the knowledge of the nitration and oxidation
processes by NO2 in organic and inorganic chemistry.
The detailed kinetic and mechanistic studies were reported for the gas-
phase reactions of NO2 with CF2CF2 [11], CF2CFCl [12], CF2CCl2 [13] and
CClHCCl2 [14]. The products of nitration of CF2CF2 were O2NCF2C(O)F and
FNO, those of CF2CFCl were O2NCF2CFClNO2, O2NCF2C(O)F and ClNO
and those of CF2CCl2 were O2NCF2CCl2NO2, O2NCF2C(O)Cl and ClNO.
In the reaction of NO2 with trichloroethene, NO2 oxidizes CClHCCl2, giving
HC(O)C(O)Cl and ClNO as the only products.
Several reaction channels can be suggested for the reactions of NO2 with
olefins. As substantial uncertainties still remain in the nitration mechanisms, it
is of importance to study the proposed mechanisms from the theoretical point
of view to characterize the probable intermediates, that form in the nitration as
well as in the oxidation of alkenes by NO2 giving different end products and
to determine the relative contributions of the proposed reaction channels to the
global reaction.
In this work the kinetic and mechanistic investigation of the reaction be-
tween NO2 and perfluoropropene C3F6 was undertaken to elucidate the elemen-
tary steps involved and to characterize the products obtained. The experimental
results were compared with those obtained theoretically.
2. Experimental
All reactants were purchased commercially. The C3F6 (PCR, 97–98%) was
purified by repeated low-pressure trap-to-trap distillations on a vacuum line,
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

NO2 and Perfluoropropene Reaction: Kinetics and Mechanism . . .
577
the fraction distilling between 163 and 183 K being retained each time. NO
was eliminated from NO2 (Matheson 99.5%) by a series of freeze-pump-thaw
cycles in presence of O2 until disappearance of the blue colour of N2O3. Fi-
nally, the degassed NO2 was purified by fractional condensation using the
fraction that distilled between 213 and 243 K. CF4, used as inert gas, was bub-
bled through 98% analytical-grade H2SO4 and passed slowly through a Pyrex
coil at 153 K.
The reaction proceeded with pressure decrease. The experiments were
performed in a grease-free static system allowing pressure measurements at
constant volume and temperature. Two reaction vessels were used: a spherical
quartz bulb of 270 cm3 (S/V = 0.75 cm−1) and another one of similar dimen-
sion filled with small pieces of quartz tubing (S/V = 4.7 cm−1). The pressure
was measured with a quartz spiral gauge and the temperature maintained within
0.1 K using a Lauda thermostat.
The reaction was followed measuring the pressure decrease as a function of
time. 22 experiments were made at 413.1, 421.0 and 432.8 K. The initial pres-
sure of C3F6 was varied between 10.2 and 108.5 Torr and that of NO2 between
43.6 and 204.0 Torr. Several experiments were made in presence of CF4, vary-
ing its pressure from 338.8 to 433.6 Torr. Two experiments were made in the
reactor with S/V = 4.7 cm−1.
Infrared spectra were recorded on Perkin–Elmer 325 and Perkin–Elmer
1600 Series FTIR spectrometers, using 10-cm cells provided with NaCl and
KBr windows, respectively. A Bruker IFS66 instrument was used to record
the gas FTIR spectra of the isolated products between 4000 and 400 cm−1 at
room temperature, using 10-cm cells provided with KBr windows. The prod-
ucts were isolated by fractional condensation after the reaction was complete.
All quantum chemical calculations were performed with the GAUSSIAN
98 program package [15] under Linda parallel execution environment using
two coupled PC’s. Geometry optimisations were performed using standard
gradient techniques with simultaneous relaxation of all the geometric pa-
rameters. The vibrational properties computed correspond to minima in the
potential energy surfaces, with no imaginary frequencies. The wavenumbers
of the vibrational fundamentals calculated with the HF approximation were
scaled by a factor of 0.9 to account for the inherent overestimation of such
values.
3. Results
Four compounds were observed as reaction products: NO, perfluoropropene
oxide (PFPO), CF3CF(NO2)CF2NO2 and CF3C(O)CF2NO2. The perfluo-
ropropene oxide was identified comparing its infrared spectrum with the
wavenumbers reported by Mahler et al. [16] and also with that previously ob-
tained in our laboratory [17] for the pure perfluoropropene oxide, which was
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

578
R. M. Romano and J. Czarnowski
Fig. 1. FTIR spectrum of PFPO at room temperature, pressure 3 Torr, optical path 10 cm,
resolution 1 cm−1
.
prepared by oxidation of C3F6 with H2O2 in a basic medium at 313–323 K [18,
19], eliminating the non-consumed perfluoropropene by bromination and sep-
arating PFPO as volatile by fractional condensation at 173 K.
As far as we know, no detailed study of the IR spectrum of PFPO was re-
ported in the literature. Fig. 1 shows the FTIR gas phase spectrum of PFPO
with the wavenumbers and intensities listed in Table 1.
The theoretical study of this molecule was performed using different the-
oretical approximations in order to compare with the experimental results.
First, the geometry optimization of the molecule, with the simultaneous re-
laxation of all the geometrical parameters was undertaken. The optimized
form is shown in Fig. 2, while the geometrical parameters are presented in
Table 2. The IR spectrum was simulated to characterize the calculated struc-
ture as a minimum, and also to compare the theoretical wavenumbers and
IR intensities with the exprimental ones. As revealed by Table 1, there is
a very good agreement between the experiment and theory. A tentative as-
signment is also included in Table 1. The vibrational normal modes of this
molecule, particularly those involving the different C–F groups, are the result
of the coupling between different vibrations, being difficult to give a sim-
ple and reliable description. The most intense vibrational mode in the IR
spectrum (ν = 1276 cm−1), that we assigned to the νas CF2, could be bet-
ter described as a mixture of movements of the CF2 and CF3 groups. In
the same way, the deformation modes appearing in the table assigned to
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

NO2 and Perfluoropropene Reaction: Kinetics and Mechanism . . .
579
Table 1. Experimental FTIR vibrational data in cm−1 for PFPO and theoretical results ob-
tained with different approximations.
a
*
*
I rel.
HF/6-31+G
B3LYP/6-31+G
FTIR
ν(cm−1
Assignment
ν(cm−1
)
I rel.
ν(cm−1
)
)
I rel.
1604
1551
1365
1276
1234
1214
1163
1127
1021
3
29
15
100
32
36
35
27
70
1586
1384
1307
1260
1243
1194
1170
1040
832
763
719
608
572
565
528
495
424
341
304
264
243
155
131
48
41
15
100
42
40
57
47
71
1
2
12
3
< 1
1
< 1
2
1552
1346
1260
1209
1190
1153
1132
1012
802
762
711
604
569
555
524
495
419
344
303
261
246
156
131
47
39
14
100
40
36
42
32
73
< 1
1
11
2
< 1
1
< 1
1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
ν C1–C2
ν C2–C3
νas CF2
ν CF3
ν CF3
ν CF
νs COC
νas COC
δ CF2
δ CF3
δ CF2
δ CF
δ CF3
δoop COC
δ CF
δ CF
773
720
691
668
612
564
509
< 1
11
< 1
< 1
1
1
1
1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
a
scaled by a factor of 0.9.
δ CF vibrations are combined movements of the different C–F groups of the
molecule.
CF3CF(NO2)CF2NO2 and CF3C(O)CF2NO2 were characterized by com-
paring their infrared spectra with the calculated ones for these molecules,
*
using the B3LYP/6-31+G approximation. These compounds are mentioned
in the literature as products of the reaction of C3F6 with NO [9] and of
the nitration of C3F6 with NO2 [7, 8], but the information about these sub-
stances is scarce and their infrared spectra have not been reported. Fig. 3
shows the FTIR spectra of CF3CF(NO2)CF2NO2 and CF3C(O)CF2NO2,
whose corresponding vibrational wavenumbers are listed in Table 3. The the-
oretical study of these two molecules was initiated with a potential energy
surface scan, in order to find the minima over these surfaces. The struc-
*
tures shown in Fig. 4, as calculated by the B3LYP/6-31+G , correspond
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

580
R. M. Romano and J. Czarnowski
*
Fig. 2. B3LYP/6-31+G molecular model of PFPO.
Table 2. Selected geometrical parameters of PFPO calculated with different theoretical ap-
proximations.
*
*
Parameter
HF/6-31+G
B3LYP/6-31+G
C1–C2
C1–O
C2–O
C2–C3
C1–F1
C1–F2
C2–F3
C3–F4
C3–F5
C3–F6
C1C2O
C1OC2
OC1C2
OC1F2
C2C1F1
F1C1F2
C1C2C3
C1C2F3
OC2C3
F3C2C3
F4C3C2
F5C3C2
F6C3C2
F4C3F5
F4C3F6
F5C3F6
1.4403
1.3513
1.3892
1.5189
1.3051
1.3053
1.3201
1.3099
1.3133
1.3153
57.0
1.4612
1.3850
1.4263
1.5321
1.3350
1.3354
1.3484
1.3396
1.3439
1.3456
57.3
63.4
59.6
62.6
60.1
116.6
124.2
109.9
126.6
119.2
117.0
110.4
111.2
109.7
109.5
108.8
108.9
108.7
116.7
124.4
109.5
126.7
119.3
117.0
110.2
111.2
110.0
109.6
108.7
108.8
108.6
to the most stable forms of CF3CF(NO2)CF2NO2 and CF3C(O)CF2NO2,
respectively. These structures are characterized as true minima for which
no imaginary frequencies occur, the calculated IR spectra being presented
in Table 3.
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

NO2 and Perfluoropropene Reaction: Kinetics and Mechanism . . .
581
Fig. 3. FTIR spectra of CF3CF(NO2)CF2NO2, pressure 3 Torr, (top) and CF3C(O)CF2NO2,
pressure 3 Torr, (bottom) at room temperature, optical path 10 cm, resolution 1 cm−1
.
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

582
R. M. Romano and J. Czarnowski
Table 3. Experimental FTIR vibrational data in cm−1 for CF3CF(NO2)CF2NO2 and
*
CF3C(O)CF2NO2 and theoretical results obtained with the B3LYP/6-31+G approxima-
tion.
CF3CF(NO2)CF2NO2
CF3C(O)CF2NO2
*
*
FTIR
B3LYP/6-31+G
FTIR
ν(cm−1
B3LYP/6-31+G
)
ν(cm−1
)
I rel.
ν(cm−1
)
I rel.
)
I rel. ν(cm−1
I rel. Assignment
1806
14
ꢀ
ꢀ
1808
1789
1872
1694
23
ν (C=O)
νas (NO2)
ꢀ
ꢀ
1618
1615
1708
79
1622
1612
100
100
1701
1405
100
9
1552
1397
1401
1313
18
36
νs (NO2)
1347
1325
νas (CCC)
27
71
1271
1384
1268
88
47
1329
1265
1235
80
80
νas (CF2)
νs (CCC)


1273
1253
1243
1241
1234
1214
1201



1252
40













1257
1221
1207
55
36
30
1178
1167
1159
1146
1128
1122
1116
1107
1098
1094
1194
1172
1057
11
83
21
νas (CF3)
νas (CF3)
ν (CN)




1198
71
1187
1092
967
36
5
2
910
817
778
19
20
1
1054
1030
17
2
760
751
ꢁ
942
25
911
829
11
17
724
720
717
735
685
7
ꢁ
822
817
815
700
692
668
17
14
649
617
607
604
774
< 1
805
5
619
3
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

NO2 and Perfluoropropene Reaction: Kinetics and Mechanism . . .
Table 3. continued.
583
CF3CF(NO2)CF2NO2
CF3C(O)CF2NO2
*
*
FTIR
B3LYP/6-31+G
FTIR
ν(cm−1
B3LYP/6-31+G
)
ν(cm−1
)
I rel.
ν(cm−1
)
I rel.
)
I rel. ν(cm−1
I rel. Assignment
734
700
668
10
2
< 1
745
716
657
604
595
538
525
478
377
369
355
341
304
280
271
264
198
182
159
124
96
3
5
8
2
10
1
1
3
< 1
< 1
< 1
< 1
< 1
< 1
1
569
511
484
434
370
343
297
269
255
207
156
141
58
5
2
1
3
< 1
< 1
1
2
1
2
< 1
1
< 1
< 1
< 1
53
11
1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
62
46
40
The very strong characteristic bands of FNO at 1844 and 766 cm−1 [20] and
those of CF3C(O)F at 1899 and 1099 cm−1 [21] were not observed. Neverthe-
less, the strong band appearing in the FTIR spectrum of CF3C(O)CF2NO2 at
1029 cm−1, and assigned to SiF4, gives account of FNO through the following
reaction [11]:
SiO2 +4 FNO → SiF4 +2 NO+2 NO2 .
For analyzing the reaction mixture of 15 experiments carried out to the com-
plete consumption of the reactant in defect, the reaction vessel was rapidly
cooled to liquid air temperature and the mixture separated by fractional con-
densation. The first fraction Fr1, volatile at 153 K, was perfluoropropene
oxide and C3F6 if present, the second fraction Fr2 was CF3C(O)CF2NO2
volatile at 173 K. The third fraction Fr3, distilling at 193 K, consisted of
CF3CF(NO2)CF2NO2 and NO, remaining NO2 as residue.
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

584
R. M. Romano and J. Czarnowski
*
Fig. 4. B3LYP/6-31+G molecular model of CF3CF(NO2)CF2NO2 and CF3C(O)CF2NO2.
To determine the concentration of perfluoropropene oxide, PFPO, in pres-
ence of C3F6, an infrared calibration curve was constructed, using pure C3F6,
that allowed the conversion of the absorption intensity at 1766 cm−1 to the pres-
sure of C3F6. The amounts of PFPO in the Fr1 were obtained by difference.
The pure nitric oxide NO is volatile at 153 K (B.P. 111.3 K), but in the
respective condensates at 153 and 173 K it forms N2O3 in presence of NO2.
CF3CF(NO2)CF2NO2 and NO distilled together at 193 K. To obtain pure
CF3CF(NO2)CF2NO2, the fractions Fr3 of three preliminary experiments
were collected together and treated with O2 to convert NO to NO2 and
CF3CF(NO2)CF2NO2 was separated by fractional condensation as volatile at
193 K. An infrared calibration curve was constructed to convert the absorption
intensity at 1618 cm−1 to the pressure of CF3CF(NO2)CF2NO2.
As spectroscopic studies needed major amounts of PFPO, CF3C(O)CF2NO2
and CF3CF(NO2)CF2NO2, each fraction Fr1, Fr2 and Fr3 obtained from the ex-
periments carried out to the complete consumption of C3F6, was condensed in
a separate trap, respectively. Oxygen was aggregated to the condensate of the
fractions Fr3 to oxidize the present NO to NO2, thus allowing the separation of
CF3CF(NO2)CF2NO2 as volatile by distillation at 193 K.
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

NO2 and Perfluoropropene Reaction: Kinetics and Mechanism . . .
585
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

586
R. M. Romano and J. Czarnowski
The analytical data of 15 experiments are summarized in the Table 4, where
i and f signify initial and final, respectively. From these data the following can
be deduced:
[C3F6]consumed = [PFPO]+[CF3C(O)CF2NO2]+[CF3CF(NO2)CF2NO2]
[PFPO] ≈ [NO] .
The relation:
R = [PFPO]/ {[CF3C(O)CF2NO2]+[CF3CF(NO2)CF2NO2]}
increases with temperature and decreases as the concentration of NO2
and the total pressure M increase. The yields of PFPO, based on the
amount of C3F6 consumed, varied between 63 and 89% and those of
CF3CF(NO2)CF2NO2 between 0.33 and 0.08%. Increasing the temperature,
the yields of CF3C(O)CF2NO2 decreased from 0.04 to 0.01%.
The reaction is homogeneous as can be seen from the experiments carried
out in the reaction vessel with a larger surface-to volume ratio. At the con-
ditions of our work, the equilibrium constants were so high that N2O4 was
completely dissociated into NO2 [22].
The consumption rate of C3F6 and that of NO2 were found to be of first
order with respect to the concentration of each reactant, but the consumption
of NO2 was larger by a factor n, which depended on the relation R between the
products formed. It can be represented by:
−d[C3F6]/dt = −1/n (d[NO2]/dt) = k[C3F6][NO2] .
(1)
The rate constants k at different temperatures are summarized in the Table 5.
ꢂ
ꢂ
In this table
t is the reaction time,
pf is the final pressure decrease of the reaction and V =
p is the pressure decrease in the
ꢂ
ꢂ
time
t,
[C3F6]c/ t, where the amounts of C3F6 consumed, [C3F6]c, were com-
ꢂ
puted point by point by pR1 being R1 = ([C3F6]i − [C3F6]f)/
pf. The
amounts of NO2 consumed were calculated by pR2 where R2 = ([NO2]i −
ꢂ
[NO2]f)/
pf. The indices i, f and c denote initial, final and consumed. The
following expression was used to calculate the rate constants:
ꢃ
ꢅ
ꢆ
ꢄ
ꢄ
k t =
pn −
pn−1 R1
ꢇ
ꢃ
ꢅ
ꢈ
ꢄ
ꢄ
[CF3OF]i −0.5
pn +
pn +
pn−1 R1
ꢇ
ꢃ
ꢅ
ꢈ
ꢄ
ꢄ
× [NO2]i −0.5
pn−1 R2
,
where n = 1, 2, 3 . . . is the number of the successive pressure measurement
ꢂ
ꢂ
ꢂ
ꢂ
of each experimental run, t =
tn −
t
n−1, (
pn −
pn−1)R1
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

NO2 and Perfluoropropene Reaction: Kinetics and Mechanism . . .
587
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

588
R. M. Romano and J. Czarnowski
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

NO2 and Perfluoropropene Reaction: Kinetics and Mechanism . . .
589
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

590
R. M. Romano and J. Czarnowski
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

NO2 and Perfluoropropene Reaction: Kinetics and Mechanism . . .
591
Fig. 5. Arrhenius plot of k1.
Fig. 6. Second order plot of the reaction of C3F6 with NO2 at 421.0 K, first order with re-
spect to each reactant concentration. ( pR1)/[C3F6][NO2] = k t, where p is the pres-
ꢂ
ꢂ
sure decrease at the time interval t and R1 = ([C3F6]i −[C3F6]f)/
pf, being
pf
the final pressure of the reaction.
ꢂ
is the [CF3OF]c at the time interval t, and {[CF3OF]i − 0.5(
pn +
ꢂ
ꢂ
ꢂ
p
n−1)R1} and {[NO2]i −0.5(
pn +
p
n−1)R2} are the mean pres-
sures [CF3OF]m and [NO2]m at the time interval t, respectively. For n = 1,
ꢂ
ꢂ
tn−1 and
pn−1 are equal to 0.
The following mean values of the rate constants were obtained: k(413.1)
=
0.025, k(421) = 0.049 and k(432.8) = 0.0593 dm3 mol−1 s−1. The Arrhenius plot of
k is presented in Fig. 5. The temperature dependence can be expressed by:
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

592
R. M. Romano and J. Czarnowski
k = (4.57 1.4)×106 exp (−(15.44 1.2) kcal mol−1 /RT)dm3 mol−1 s−1.
(2)
In the Fig. 6 the plot of ( pR1)/[C3F6][NO2] vs. reaction time t at 421.0 K is
presented. In this plot, p is the pressure decrease in the interval t and R1 =
ꢂ
ꢂ
([C3F6]i −[C3F6]f)/
pf, where
pf is the final pressure decrease of the
reaction. The obtained straight line indicates that the reaction is of second order
and, therefore, of first-order with respect to the concentration of each reactant.
4. Discussion
The following mechanism was proposed to explain the experimental results:
As NO2 and CF3CF=CF2 are stable compounds, the reaction must proceed
via the rate-limiting addition of NO2 to the double bond of C3F6, analogously
to other reactions of NO2 with alkenes [11–14, 23–25]. These reactions are of
first order with respect to both the alkene and NO2 concentration. The first-
order dependence on the concentration of each reactant indicates the formation
of a common intermediate, in this case radical CF3C•FCF2NO2, which can de-
compose giving perfluoropropene oxide, or undergo association reaction with
the second NO2 forming dinitro vicinal product, or react directly with NO2
to give CF3C(O)CF2NO2 and FNO. The initial attack by NO2 at the terminal
position occurs by C−N attachment [26, 27].
The lack of formation of CF3C(O)F indicates that, within the tempera-
ture range of our work, perfluoropropene oxide does not decompose to give
CF3C(O)F and :CF2. The pyrolysis of PFPO was studied between 463 and
503 K and its major products were CF3C(O)F, CF2=CF2 formed by associa-
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

NO2 and Perfluoropropene Reaction: Kinetics and Mechanism . . .
593
tion of two :CF2, and perfluorocyclopropane arising by addition of a :CF2 to the
tetrafluoroethene [28].
The formation of nitrosyl fluoride FNO was postulated, but it was not ob-
served in the infrared spectra. It decomposes very rapidly on the SiO2 surface
giving SiF4, NO and NO2 [11].
Under conditions of our work, the reactions of NO and FNO with perflu-
oropropene were neglected. It was reported that nitric oxide reacts with C3F6,
either under pressure in an autoclave or in the reactor at atmospheric pressure,
when kept in absence of light for 1 month [9]. FNO reacts with C3F6 on heat-
ing at 423 K in presence of calcium sulfate on activated charcoal as a catalyst
and also when heated at 393 K for 13 h, under an autogenous pressure of about
200 atm [29].
The experimental results of our work were rationalized taking into ac-
count the data reported for the reaction CF• +NO2 studied by several research
groups [30–37], where the predominant 3process, independent of the total
pressure, is the formation of COF2 and FNO. Bevilacqua et al. [32] and
Vakhtin [34] reported that CF3NO2 was also formed in significant amounts
in the reaction of CF3 with NO2. The formation of COF2 and FNO through
a four-center transition state was proposed by Sugarava et al. [30], and Oum
et al. [35]:
F2C
O
F
NO
For the formation of O2NCF2C(O)F and ClNO in the reaction O2NCF2CFCl• +
NO2 [12] the following transition state was postulated:
F
O2NCF2C
O
Cl
NO
Theoretical calculations performed in our work for the reaction (2) lead
us to propose the following intermediate for the decomposition of radical
CF3C•FCF2NO2 into perfluoropropene oxide and NO:
NO
O
CF2
CF3C
F
The following intermediate, also based on theoretical calculations, was postu-
lated for the reaction (4) to explain the formation of CF3C(O)CF2NO2:
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

594
R. M. Romano and J. Czarnowski
ONO
F
CF3C
CF2NO2
Applying the steady-state approximation method to the mechanism, the follow-
ing equations are obtained:
−d[C3F6]/dt = k1[C3F6][NO2]
(3)
and
ꢉ
ꢊ
−d[NO2]/dt = k1[C3F6][NO2] 1+{1+k2/[[NO2] (k3[M]+k4)]}−1
(4)
where
[PFPO]
k2/[[NO2] (k3[M]+k4)] =
.
[CF3CF(NO2)CF2NO2]+[CF3C(O)CF2NO2]
(5)
Comparing Eqs. (3) and (4) with the Eq. (1), it can be seen that:
n = 1+{1+k2/[[NO2] (k3[M]+k4)]}−1
(6)
and
k1 = k =
(4.57 1.4)×106 exp (− (15.44 1.2) kcal mol−1/RT)dm3 mol−1 s−1
.
(7)
Very low preexponential factors A: 1.3×104, 1.7×106 and 3.16×106
dm3 mol−1 s−1, were found for the homogeneous additions of NO2 to CF2CF2
[11], CF2CFCl [12] and CF2CCl2 [13], respectively. The corresponding activa-
tion energies were 7.5, 10.8 and 10.5 kcal mol−1. The decrease of the preexpo-
nential factor in a bimolecular reaction is expected when the complexity of the
reacting species increases [38].
From the Eq. (5), the rate constant k2 can be estimated. The effective pres-
sure M was obtained using the expression:
M = [C3F6]+[NO2]+[P]+γ[CF4]
(8)
where P is the product sum and γ the relative collisional efficiency factor
for CF4, The reported value for this factor was about 0.34 [39]. The rela-
tive collisional efficiency of NO2 and those of the products were assumed
to be similar to that of alkene taken as unity. The rate constant for associa-
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

NO2 and Perfluoropropene Reaction: Kinetics and Mechanism . . .
595
Fig. 7. Calculated relative energies for the mechanisms proposed by the Eqs. (1) and (2).
tion reaction k3 was assumed to be equal to the collisional frequency factor
(1010 dm3 mol−1 s−1) and k4 was neglected taking k3[M] > k4. Thus the esti-
mated value of k2, the rate constant for the unimolecular dissociation of the
radical CF3C•FCF2NO2, was found to be of the order of 1014 s−1.
The proposed mechanisms was studied from a theoretical point of view,
*
using the B3LYP/6-31+G theoretical approximation. For the formation of
PFPO both proposed intermediates were found as stable structures for which
no imaginary frequencies occur. The radical species, CF3C•FCF2NO2, was pre-
dicted to be 6.22 kcal mol−1 higer in energy than the reaction mixture, C3F6 +
NO2, while the cycled structure was predicted to be 15.33 kcal mol−1 lower
in energy than the radical. The final products, PFPO and NO, were found
to be even 13.82 kcal mol−1 lower in energy than this intermediate structure.
A schematic representation of this mechanism is shown in Fig. 7.
The addition of a second NO2 molecule to the CF3C•FCF2NO2 radical can
occur either through a C−N or C−O attachment. In the first case the reac-
tion leads to the dinitro compound, CF3CF(NO2)CF2NO2, predicted to be at
30.87 kcal mol−1 lower energy than the reactancts according to the B3LYP/6-
*
*
31 G approximation. The attachment through the oxygen atom of the NO2
originates the formation of the CF3CF(ONO)CF2NO2, 11.27 kcal mol−1 more
stable than the reactancts, that decomposes to give CF3C(O)CF2NO2 and FNO,
21.46 kcal mol−1 even more stable. These two mechanisms are schematically
represented in Fig. 8.
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

596
R. M. Romano and J. Czarnowski
Fig. 8. Calculated relative energies for the mechanisms proposed by the Eqs. (3) and (4).
Acknowledgement
The authors thank Mr. Z. Czarnowski for helpful comments. This work was fi-
nancially supported by the Consejo Nacional de Investigaciones Cient´ıficas y
Te´cnicas. JC thanks the ANCyT (PICT 6786). RMR also thanks the Fundacio´n
Antorchas.
References
1. B. L. Dyatkin, E. P. Mochalina, and I. L. Knunyants, Russ. Chem. Rev. 35 (1966)
417.
2. D. A. Barr and R. N. Haszeldine, J. Chem. Soc. (1960) 1151.
3. A. V. Fokin and A. T. Uzun, Zh. Obshchei Khimii 36 (1966) 117.
4. D. D. Coffman, M. S. Raasch, G. W. Rigby, P. L. Barrick, and W. E. Hanford, J. Org.
Chem. 14 (1949) 747.
5. R. N. Haszeldine, J. Chem. Soc. (1953) 2075.
6. E. Bissel, J. Org. Chem. 26 (1961) 5100.
7. I. L. Knunyants, A. V. Fokin, and V. A. Komarov, Zh. Vses. Khim. Obshchestva im.
D. I. Mendeleeva 7 (1962) 709.
8. I. L. Knunyants, V. A. Komarov, and A. V. Fokin, Izv. Akad. Nauk SSSR, Ser. Khim.
(1966) 466.
9. J. D. Park, A. P. Stefani, G. H. Crowford, and J. R. Lacher, J. Org. Chem. 26 (1961)
3316.
10. J. Czarnowski, J. Chem. Soc. Perkin Trans. 2 (1991) 1459.
11. C. W. Spicer and J. Heicklen, Int. J. Chem. Kinet. 4 (1972) 575.
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM

NO2 and Perfluoropropene Reaction: Kinetics and Mechanism . . .
597
12. R. M. Romano, C. O. Della Vedova, and J. Czarnowski, Z. Phys. Chem. 216 (2002)
1203.
13. J. Czarnowski and H. J. Schumacher, Int. J. Chem. Kinet. 18 (1986) 907.
14. J. Czarnowski, Int. J. Chem. Kinet. 24 (1992) 679.
15. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheese-
man, V. G. Zakrzewski, J. A. Montgomery, Jr., R. E. Stratmann, J. C. Burant, S. Dap-
prich, J. M. Millam, D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi,
V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clif-
ford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, D. K. Mal-
ick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz,
A. G. Baboul, B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Gom-
perts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, Y. Peng, A. Nanayakkara,
C. Gonzalez, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong,
J. L. Andres, M. Head-Gordon, E. S. Replogle, and J. A. Pople, GAUSSIAN 98,
revision A.7, Gaussian Inc., Pittsburgh, PA, 1998.
16. W. Mahler and P. R. Resnick, J. Fluor. Chem. 3 (1973) 451.
17. M. dos Santos Afonso, Doctoral Thesis, 1982.
18. E. I. Du Pont de Nemours and Co, British Patent 904877, 1960.
19. D. Sianesi, A. Pasetti, and F. Tarli, J. Org. Chem. 31 (1966) 2312.
20. P. J. H. Woltz, E. A. Jones, and A. H. Nielsen, J. Chem. Phys. 20 (1952) 378.
21. J. Pacansky, R. J. Waltman, and Y. J. Ellinger, J. Phys. Chem. 98 (1994) 4787.
22. H. Blend, J. Chem. Phys. 53 (1970) 4497.
23. R. Atkinson, S. M. Aschmann, A. Winer, and J. N. Pitts Jr., Int. J. Chem. Kinet. 16
(1984) 697.
24. H. Niki, P. D. Marker, C. M. Savage, L. P. Breitenbach, and M D. Hurley, Int. J.
Chem. Kinet. 18 (1986) 1235.
25. J. L. Powell, J. H. Ridd, and J. P. B. Sandall, J. Chem. Soc., Chem. Commun. (1990)
402.
26. H. Shechter and D. E. Ley, Chemistry and Industry (1955) 535.
27. J. C. D. Brand and I. D. R. Stevens, J. Chem. Soc. (1958) 629.
28. P. J. Krusic, D. C. Roe, and B. E. Smart, Isr. J. Chem. 39 (1999) 117.
29. S. Andreades, J. Org. Chem. 27 (1962) 4163.
30. K. Sugawara, T. Nakanag, H. Takeo, and C.Matsumura, J. Phys. Chem. 93 (1989)
1894.
31. J. S. Francisco and Z. Li, J. Phys. Chem. 126 (1989) 528.
32. T. J. Bevilacqua, D. R. Hanson, and C. J. Howard, J. Phys. Chem. 97 (1993) 3750.
33. J. Sehested, O. J. Nielsen, C. A. Rinaldi, S. I. Lane, and J. C. Ferrero, Int. J. Chem.
Kinet. 28 (1996) 579.
34. A. B. Vakhtin, Int. J. Chem. Kinet. 29 (1997) 203.
35. K. W. Oum and G. Hancock, J. Phys. Chem. 101 (1997) 2634.
36. P. Pagsberg, J. T. Jodkowski, E. Ratajczak, and A. Sillesen, Chem. Phys. Lett. 286
(1998) 138.
37. C. Breheny, G. Hancock, and C. Morell, Phys. Chem. Chem. Phys. 2 (2000) 5105.
38. S. W. Benson, The Foundations of Chemical Kinetics, McGraw-Hill Book Co. Inc.
(1960).
39. S. C. Chan, B. S. Rabinovitch, J. T. Bryant, L. D. Spicer, T. Fuyimoto, Y. N. Lin, and
S. P. Pavlou, J. Phys. Chem. 74 (1970) 3160.
Brought to you by | Cornell University Library
Authenticated
Download Date | 10/19/16 7:53 AM