Reactivity of Negative Ions with Trifluoromethyl Halides

Article abstract of DOI:10.1021/jp960355y

The kinetics of the reactions of selected anions (A(-)) with the trifluoromethyl halides CF3X (X = F, Cl, Br, I) in the gas phase were measured at 300 K.Reaction rate constants and product branching fractions were determined using a selected ion flow tube (SIFT) instrument.The chosen anions were C5F5N(-), o-, m-, and p-CF3C6H4CN(-), C6F5Br(-), C6F5Cl(-), C6F5CF3(-), C6F5COCH3(-), Fe(-), FeCO(-), SF6(-), SO(-), SO2(-), NO(-), NO2(-), and NO3(-).The reactivity of these systems varies from unreactive to collisional, and a variety of reaction types was found.The results of our present and previous measurements on A(-) + CF3X reactions show that, in cases where nondissociative electron transfer (NDET) is energetically allowed, the total reactivity tends to be high, approaching collisional.This suggests that reactivity in these cases is initiated and controlled by electron transfer from the anion.In addition, association reactions tend to be preempted when NDET is energetically allowed.A few exceptions to these tendencies were found and are discussed.

Full text of DOI:10.1021/jp960355y

J. Phys. Chem. 1996, 100, 10641-10645
10641
Reactivity of Negative Ions with Trifluoromethyl Halides
Robert A. Morris,* A. A. Viggiano, Thomas M. Miller, John V. Seeley, Susan T. Arnold, and
John F. Paulson
Phillips Laboratory, Geophysics Directorate (GPID), 29 Randolph Road,
Hanscom AFB, Massachusetts 01731-3010
Jane M. Van Doren
Department of Chemistry, College of the Holy Cross, Worcester, Massachusetts 01610-2395
ReceiVed: February 5, 1996; In Final Form: April 16, 1996X
The kinetics of reactions of selected anions (A-) with the trifluoromethyl halides CF3X (X ) F, Cl, Br, I) in
the gas phase were measured at 300 K. Reaction rate constants and product branching fractions were
determined using a selected ion flow tube (SIFT) instrument. The chosen anions were C5F5N-, o-, m-, and
p-CF3C6H4CN-, C6F5Br-, C6F5Cl-, C6F5CF3 , C6F5COCH3 , Fe-, FeCO-, SF6 , SO-, SO2 , NO-, NO2 ,
-
-
-
-
-
-
and NO3 . The reactivity of these systems varies from unreactive to collisional, and a variety of reaction
types was found. The results of our present and previous measurements on A- + CF3X reactions show that,
in cases where nondissociative electron transfer (NDET) is energetically allowed, the total reactivity tends to
be high, approaching collisional. This suggests that reactivity in these cases is initiated and controlled by
electron transfer from the anion. In addition, association reactions tend to be preempted when NDET is
energetically allowed. A few exceptions to these tendencies were found and are discussed.
Introduction
Experimental Section
The measurements were made under thermal conditions at
300 K using the selected ion flow tube (SIFT) instrument at
Phillips Laboratory. This apparatus has been described previ-
ously,9 and the technique and generic instrumentation have been
extensively reviewed.10 Therefore, we report here only those
details which are relevant to the present study.
We have previously reported kinetics measurements for a
number of negative ions reacting with the trifluoromethyl halides
CF3X (X ) F, Cl, Br, I).1,2 In those studies, trends in reactivity
began to emerge, but the number and variety of systems studied
were quite limited. It was found that the reactivity of O2- with
fully halogenated methanes, including CF3X, correlates with
both the electron affinities of the perhalomethanes and their rates
of thermal electron attachment.1 This was interpreted as an
indication that in these cases reactivity is initiated and governed
by electron transfer from the anion to the perhalomethane, i.e.,
that initial electron transfer is the rate-limiting step.
Substitution reactions at carbon occurring in solution via
initial electron transfer were first characterized in the 1960s by
Kornblum et al.3 and by Russell and Danen.4 Bunnett5 named
this mechanism “SRN1” and studied its applicability to aromatic
substitution. Initial electron transfer in the gas phase was
proposed by McDonald et al.6 to explain the reactivity of
(CH3O)2PO- with CF3X and by Jones et al.7 to explain the
reactivity of 17-electron transition-metal complex anions with
a variety of haloalkanes. Knighton and Grimsrud8 also at-
tributed anion reactivity trends in haloalkanes to an initial
electron transfer mechanism.
In order to investigate the extent of applicability of the initial
electron transfer mechanism to other systems, we have studied
the reactivity of the anions C5F5N-, o-, m-, and p-CF3C6H4-
CN-, C6F5Br-, C6F5Cl-, C6F5CF3-, C6F5COCH3-, Fe-, FeCO-,
SF6-, SO-, SO2-, NO-, NO2-, and NO3- with the trifluorom-
ethyl halides CF3X (X ) F, Cl, Br, I). Rate constants and
product branching fractions were measured at 300 K using a
selected ion flow tube (SIFT) instrument. The results are
compared with previously published measurements of other
A-/CF3X systems performed on the same SIFT apparatus,1,2
and some trends are apparent in the combined data.
The ions were formed in a remote electron impact ion source
from precursor chemicals as follows. NO- was formed from
N2O, Fe- and FeCO- were formed from Fe(CO)5 diluted in
-
CO, SF6 was formed from an SF6 impurity present in NF3,
SO- and SO2 were formed from SO2, NO2 and NO3 were
formed from NO2 which contained an impurity of HNO3, and
F- was formed from CF4. All of the organic anions were
formed in the ion source from the corresponding parent neutral
compounds, e.g., C5F5N- from C5F5N.
-
-
-
The ion of interest was mass selected in a quadrupole mass
filter and injected through a Venturi aspirator inlet into a
stainless steel flow tube 1 m in length. The ions were conveyed
along the flow tube by a fast flow (approximately 104 cm s-1
)
of He buffer gas maintained at approximately 0.4 Torr, except
in the case of the NO- reactions which were studied in an Ar
carrier gas at about 0.2 Torr; the Ar carrier was used in this
case because the NO- anion is destroyed in the flow tube by
thermal electron detachment in collisions with He.11 Neutral
CF3X reactant gas was added to the flow tube, and the reactant
and product ions were sampled through a 0.2 mm orifice in a
flat plate mounted on a blunt sampling cone. The ion current
reaching the flat plate was recorded in order to check for possible
electron detachment reactions.12 After being sampled, the ions
were mass analyzed in a second quadrupole mass spectrometer
and detected by a channel particle multiplier.
Rate constants were determined by the standard technique
of recording the pseudo-first-order attenuation of the reactant
ion count rate as a function of reactant neutral flow rate. The
reaction time was measured directly by time-of-flight tech-
niques.9 Buffer and reactant gas flow rates were maintained
* To whom correspondence should be addressed.
X Abstract published in AdVance ACS Abstracts, June 1, 1996.
S0022-3654(96)00355-3 CCC: $12.00 © 1996 American Chemical Society

10642 J. Phys. Chem., Vol. 100, No. 25, 1996
Morris et al.
and measured with flow controllers, and the flow tube pressure
was measured with a capacitance manometer. Product branch-
ing fractions were determined by recording and plotting the
fractional product ion abundances as a function of reactant
neutral flow rate. The effects of secondary reactions were
accounted for by extrapolating the branching fractions to zero
reactant neutral flow rate. The product branching measurements
were performed under conditions of low resolution in order to
minimize the effects of mass discrimination. The reported
branching percentages have been rounded off to the nearest five
percentage points, except in those cases where a very minor
product percentage was determined, e.g., 2%.
indication that the reaction will be slow, since highly exothermic
pathways may exist which do not involve electron transfer per
se, e.g., nucleophilic attack at carbon by O-. Jones et al.7 have
pointed out that, even in some cases where NDET is endother-
mic, electron transfer may occur within the collision complex,
since the attractive ion-dipole and ion-induced dipole interac-
tions supply additional energy with which to overcome the
endothermicity of electron transfer. In that model, a subsequent
step, halide transfer from CF3X-, would then take place within
the collision complex, leading to the observed exothermic
products.
The slow reaction of SF6- with CF3I does not contradict the
-
proposal of an initial electron transfer mechanism because SF6
Results
is generally unreactive or inefficient with respect to electron
transfer processes;22 initial electron transfer from SF6- to CF3I
would be expected to be inhibited, which would lead to the
observed low overall reactivity. Grimsrud et al.23 have sug-
The experimental results are presented in Table 1. The
reactions are ordered by increasing detachment energy of the
anion. The NO- reactions were carried out in Ar buffer gas11
and all others in He. Neutral products of reaction were not
detected but were inferred from mass conservation and ther-
mochemical considerations.13
The reactivity of the systems studied varies from fast, i.e.,
collisional,14,15 to immeasurably slow. A variety of reaction
types is observed, including nondissociative electron transfer
(NDET), halogen atom transfer, halide ion production, CF3
radical transfer, and association. CF4 does not react with any
of the studied anions with the exception of a slow electron
detachment “reaction” with the weakly-bound anion NO-.
-
gested that the low reactivity of SF6 toward electron transfer
is caused by the significant geometry change required in going
-
from SF6 anion to SF6 neutral.
-
In addition to the SF6 + CF3I case, anomalous behavior is
also found for the reaction of m-CF3C6H4CN- with CF3Br,
which is slow despite an apparent 0.24 eV exothermicity for
NDET. Possible explanations for this behavior include incorrect
thermochemistry as well as the possibility that the geometry of
this anion is significantly different from that of the parent
neutral, an argument similar to that offered above for SF6-. An
alternative explanation is that the delocalized nature of the
negative charge in the reactant anion leads to the observed
inefficiency of reaction. It should be noted that no NDET
product CF3X- was observed from any of the CF3C6H4CN-
reactions, while all of the other organic anions studied did form
some CF3I- in reaction with CF3I.
Discussion
A. Reactivity. A few general observations about reactivity
are offered in this section. In Table 2 we compare not only the
results of the present measurements but also the previously
reported data from our laboratory on reactivity of anions with
CF3X.1,2 The previous work involves the reactions of F-, O-,
and O2- with the complete CF3X series, as well as reactions of
X-, XO-, and XF- with the corresponding CF3X compound,
e.g., IF- + CF3I. The reactant anions and neutrals in Table 2
are ordered by increasing detachment energy and electron
affinity, respectively.16 Table 2 gives the measured overall
reaction rate constants; cases where the nondissociative electron
transfer product CF3X- was observed are marked with an
asterisk.
We have previously shown that the rates of the reactions of
-
O2 with fully halogenated methanes correlate with both the
electron affinities and the electron attachment rates of the
halomethanes,1 supporting the above mechanism. Earlier,
McDonald et al.6 suggested an initial electron transfer mecha-
nism to explain the reactivity of (CH3O)2PO- with CF3X. Jones
et al.7 showed a correlation between haloalkane electron
attachment rates and the rates for reactions of 17-electron
transition-metal anions and concluded that the reactivity of these
systems is governed by initial electron transfer. Knighton and
Grimsrud8 also concluded that initial electron transfer controls
the reactivity of a series of reactions of substituted benzene
anions and haloalkanes. Pan and Ridge24,25 invoked this
mechanism to explain the correlation between anion reactivity
and the electron affinities of a series of organic electrophiles.
Thus, a wealth of gas phase anion reactivity data support this
mechanism for a variety of neutrals including CF3X, although
the mechanism does not hold for all molecules.22
i. Bimolecular ReactiVity. In general, these reactions are fast
(i.e., at or near the collisional limit14,15) when NDET is
energetically allowed (exothermic or approximately thermo-
neutral21), regardless of the observed product channel(s). Actual
NDET reaction products, i.e., CF3X-, are seen in a number of
reactions with CF3I, which has the highest electron affinity (EA)
in the CF3X series. Formation of CF3Br- was found in only
one case, O2 + CF3Br,1 and no formation of CF3Cl- was
-
observed in any of the reactions. (CF3Cl- has been observed
in the gas phase from electron attachment to (CF3Cl)n clus-
ters18,19 but has not been found as a product from ion-molecule
reactions.)
ii. ReactiVity toward Association. Another general observa-
tion is that when association occurs, it tends to occur when
NDET is endothermic; however, there are many cases where
NDET is endothermic yet no association is found. NDET thus
appears to preempt association as do other exothermic bimo-
lecular channels when they are operating, in most cases. The
reactions of the CF3C6H4CN- ions and of SF6- are again notable
exceptions. For the reaction of SF6- with CF3I, slow association
is the sole reaction channel in spite of NDET being exothermic
and other bimolecular channels being available. This is
consistent with the initial electron transfer mechanism as
follows: since initial electron transfer is inhibited, association
is not preempted by electron transfer. The CF3C6H4CN- ions
associate with CF3Br and CF3I despite the apparent availability
The observation that reaction tends to be efficient in cases
where NDET is energetically allowed suggests that the mech-
anism for these cases involves initial electron transfer from the
anion to CF3X either at long range or within the collision
complex. The formation of halide ion products or other products
which are fragments of CF3X- may then arise from subsequent
dissociation of the CF3X- species either before or after the
complex dissociates. We note that there are also cases where
overall reaction is fast yet NDET is quite endothermic. Thus,
while energetically-allowed NDET generally implies a fast
overall reaction, endothermic NDET should not be seen as an

Reactivity of Negative Ions with Trifluoromethyl Halides
J. Phys. Chem., Vol. 100, No. 25, 1996 10643

10644 J. Phys. Chem., Vol. 100, No. 25, 1996
Morris et al.
TABLE 2: Rate Constants (10-9 cm3 s-1) for Overall
Reaction of Selected Negative Ions with CF3X
C6F5CF3-, and SO2-, by NDET for C5F5N-, and by association
for o-, m-, and p-CF3C6H4CN-. In the case of FeCO-,
decreasing halogen transfer with heavier CF3X is accompanied
by increases in several pathways. The decrease in the contribu-
tion by halogen atom transfer with heavier CF3X could be the
result of the increasing strength of the C-X- bond in the CF3X-
anion. If reaction is initiated by electron transfer from A-,
forming CF3X- which then transfers X- to A (the net result of
which is halogen atom transfer), then the process involves
breaking increasingly stronger C-X- bonds for heavier CF3X.
ii. Halide Ion Formation. The present results show forma-
tion of halide ion X- from CF3X in many cases, but production
of F- is not found in any of these cases. The halide formation
channel, when operating, generally increases with heavier CF3X.
In fact, some I- is formed from CF3I in every case except for
SF6-, NO2-, and NO3-, which react only by association, and
m-CF3C6H4CN- which reacts by association and I atom transfer.
The o- and p-CF3C6H4CN- isomers both react with CF3I to form
I- as well as the association adduct and the I atom transfer
product CF3C6H4CNI-.
CF3Cl
CF3Br
(0.91)
CF3I
(1.57)
CF4 (<0)b (>∼0)b,c
NO- (0.026)
0.0035
nr
0.33
0.63
<0.004
0.0010
nr
0.013
0.0020
nr
1.0
1.1
1.9
1.8
Fe- (0.151)
O2- (0.451)d
nr
nr
nr
nr
nr
nr
1.0*
0.021
0.12
0.34
0.22
nr
1.5*
0.81
0.71*
0.89
0.87
0.80*
0.81*
0.005
0.62*
1.25*
1.4*
0.56*
1.4
m-C8F3H4N- (0.67)
C5F5N- (0.68)
o-C8F3H4N- (0.70)
p-C8F3H4N- (0.76)
C6F5Cl- (0.82)
C6F5COCH3- (0.88)
SF6- (1.05)
nr
nr
nr
nr
nr
nr
nr
nr
nr
nr
nr
nr
nr
nr
0.21
0.86
nr
0.0034
nr
0.0075
0.46
0.066
nr
0.94
1.4
0.0005
C6F5CF3- (1.06)
SO2- (1.107)
SO- (1.125)
C6F5Br- (1.15)
FeCO- (1.157)
O- (1.461)d
1.9
0.063
NO2- (2.273)
ClO- (2.275)d
BrO- (2.353)d
IO- (2.378)d
nr
nr
0.0099
In the reactions of Fe- with CF3X, production of X- increases
0.13
ClF- (2.4)d,e
nr
from 15% Cl- to 50% Br- to 100% I-, with the reaction rate
BrF- (∼2.5)d,f
<0.003
increasing with heavier CF3X. The rate constant for Fe-
+
IF- (∼2.5)d,f
0.94
<0.005
I- (3.059)d
CF3I exceeds the collisional rate constant by 47%, which is
greater than the combined uncertainties in the measured and
calculated14,15 rate constants. Because of this apparent discrep-
ancy, we repeated the measurements and checked the calibration
of the flow meters and manometer. We are confident in the
measurement to within our usual absolute uncertainty of (25%;
therefore, we propose that the reaction is in fact faster than the
theoretical ion-dipole capture rate. The explanation likely
involves initial electron transfer at relatively long distancess
distances outside of the capture radius of the ion-molecule
orbiting complex. A theoretical examination of the curve
crossings in this system as well as the ionic radius of Fe- could
shed light on this possibility. It should be pointed out that
electron transfer at distances greater than the capture radius have
been reported previously.27,28
Br- (3.364)d
<0.003
0.084
F- (3.401)d
nr
nr
<0.003
<0.003
<0.0005 <0.0005
0.72
Cl- (3.613)d
NO3- (3.937)
0.027
a The notation nr denotes no evidence of reaction. An asterisk
indicates observation of the nondissociative electron transfer product
CF3X-. Values in parentheses are electron affinities of neutrals and
electron binding energies of anions (both in electronvolts; ref 16 unless
otherwise noted). b Reference 17. c References 18 and 19. d Rate con-
stants from ref 1. e Reference 20. f Estimated.
of exothermic NDET. Here the clustering might be explained
by electron transfer to CF3X followed by association. The
CF3C6H4CN- ions, however, also associate with CF3Cl for
which NDET is clearly endothermic. We note that no NDET
products, i.e., CF3X-, were observed in any of the reactions of
the CF3C6H4CN- ions, while all of the other organic anions
studied formed some CF3I- from reaction with CF3I.
Association is not observed for any of the CF4 “reactions”,
and the association rates for the other CF3X generally increase
with heavier CF3X, as expected from the increasing dipole
moment and polarizability of the heavier CF3X, both of which
lead to larger cluster bond strengths.
iii. CF3 Radical Transfer. Transfer of a CF3 radical from
CF3X to the reactant anion occurs as a minor channel in the
reactions of o- and p-CF3C6H4CN- with CF3Cl, where the
overall reactivity is low, but not at all with the more reactive
CF3Br and CF3I. The isomer m-CF3C6H4CN- does not undergo
CF3 transfer with any CF3X. Transfer of CF3 also was observed
in the cases of FeCO- with CF3Cl and CF3Br but not with CF3I.
-
-
iV. Association Reactions. The ions NO2 and NO3 react
with the CF3X series only by association and only for the heavier
B. Chemical Reaction Channels. i. Halogen Atom Trans-
fer. A major and sometimes dominant pathway in most of the
reactions of both Fe- and FeCO- is halogen atom transfer, a
channel found previously in work involving other transition
metal complex anions.7,26 The relative contribution from the
halogen transfer channel diminishes with heavier CF3X. This
same trend is found in the reactions of o-, m-, and p-CF3C6H4-
CN-, for which halogen atom transfer dominates for CF3Cl and
CF3Br but is a minor pathway for CF3I, where the main channel
is association. Halogen atom transfer is also observed in the
-
CF3X: measurable rates are found just for NO2 + CF3I and
NO3- + CF3I. At 0.43 Torr, the second-order association rate
constant for NO2- + CF3I is more than twice the value for NO3
-
+ CF3I. The NO3-‚CF3I adduct has additional internal degrees
of freedom and on that basis alone would be expected to display
a faster rate of formation, due to the resulting longer complex
lifetime. Therefore, the explanation for the difference in rates
must lie in the relative cluster bond strengths for NO2-‚CF3I
versus NO3-‚CF3I.29,30 These bond strengths are not available
-
-
reactions of C6F5COCH3-, C6F5CF3-, and SO2 with CF3Br
in the literature, but NO2 is known to bond more strongly to
- 31
polar molecules than does NO3
.
and CF3I, with a larger contribution from CF3Br than from CF3I
in each case. The anion C5F5N- is unreactive with CF3Cl but
undergoes halogen atom transfer with CF3Br; this pathway is
not seen for CF3I. Thus, all of the results show the same trend
of decreasing relative contribution from halogen atom transfer
with heavier CF3X, while the overall reactivity is increasing.
Association in competition with bimolecular channels was
observed for the anions o-, m-, and p-CF3C6H4CN- with CF3-
Cl, CF3Br, and CF3I, and in the reaction of SO- with CF3Br. A
pressure dependence study of these reactions might prove
interesting. In our previous work on F- + CF3Br and CF3I,2
we found that increasing pressure leads to larger rate constants
for association but not at the expense of the rate constants for
The decrease in percent halogen transfer with heavier CF3X is
offset by halide formation in the cases of Fe-, C6F5COCH3
,
-

Reactivity of Negative Ions with Trifluoromethyl Halides
J. Phys. Chem., Vol. 100, No. 25, 1996 10645
the other reaction channel, in this case halide formation, and
we therefore postulated that the two pathways arise from two
different intermediates.
financial support from the Air Force Office of Scientific
Research is gratefully acknowledged.
-
References and Notes
The slow association reaction of SF6 with CF3I was
discussed in section A.ii.
(1) Morris, R. A. J. Chem. Phys. 1992, 97, 2372.
(2) Morris, R. A.; Viggiano, A. A. J. Phys. Chem. 1994, 98, 3740.
(3) Kornblum, N.; Michel, R. E.; Kerber, R. C. J. Am. Chem. Soc.
1966, 88, 5662.
(4) Russell, G. A.; Danen, W. C. J. Am. Chem. Soc. 1966, 88, 5663.
(5) Bunnett, J. F. Acc. Chem. Res. 1978, 11, 413.
(6) McDonald, R. N.; Chowdhury, A. K.; Gung, W. Y.; DeWitt, K.
D. AdV. Chem. Ser. 1987, 215, 51.
(7) Jones, M. T.; McDonald, R. N.; Schell, P. L.; Ali, M. H. J. Am.
Chem. Soc. 1989, 111, 5983.
V. Collisional Electron Detachment. Collisional electron
detachment from NO- has been the subject of several
studies,11,32-35 and the efficiency of detachment depends on the
particle colliding with the NO-. The NO- detachment energy
is only 0.026 eV,36 essentially equal to kT at room temperature.
Electron detachment from NO- is seen in the present work in
collisions with CF4, with a modest rate constant value of 3.5 ×
10-12 cm3 s-1. No chemical reactions, i.e., bonds broken or
formed, are found for anions in collision with CF4. Since
exothermic reactive channels are available for the other CF3X
species, NO- detachment is preempted by fast reaction. We
cannot, however, rule out a very minor (∼5%) contribution by
NO- detachment in the reactive cases.
(8) Knighton, W. B.; Grimsrud, E. P. J. Am. Chem. Soc. 1992, 114,
2336.
(9) Viggiano, A. A.; Morris, R. A.; Dale, F.; Paulson, J. F.; Giles, K.;
Smith, D.; Su, T. J. Chem. Phys. 1990, 93, 1149.
(10) Smith, D.; Adams, N. G. AdV. At. Mol. Phys. 1988, 24, 1.
(11) McFarland, M.; Dunkin, D. B.; Fehsenfeld, F. C.; Schmeltekopf,
A. L.; Ferguson, E. E. J. Chem. Phys. 1972, 56, 2358.
(12) Viggiano, A. A.; Morris, R. A.; Paulson, J. F.; Ferguson, E. E. J.
Phys. Chem. 1990, 94, 7111.
(13) Lias, S. G.; Bartmess, J. E.; Liebman, J. F.; Holmes, J. L.; Levin,
R. D.; Mallard, W. G. J. Phys. Chem. Ref. Data Suppl. 1 1988, 17, 1.
(14) Su, T.; Chesnavich, W. J. J. Chem. Phys. 1982, 76, 5183.
(15) Su, T. J. Chem. Phys. 1988, 89, 5355.
(16) Miller, T. M. In Handbook of Chemistry and Physics; Lide, D. R.,
Ed.; CRC Press: Boca Raton, FL, 1995; pp 10-180.
(17) Preston, H. J. T.; Kaufman, J. J. Chem. Phys. Lett. 1977, 50, 157.
(18) Kuhn, A.; Illenberger, E. J. Phys. Chem. 1989, 93, 7060.
(19) Kuhn, A.; Illenberger, E. J. Chem. Phys. 1990, 93, 357.
(20) Nguyen, M. T.; Ha, T.-K. Chem. Phys. Lett. 1987, 136, 413.
(21) Bohme, D. K.; Mackay, G. I.; Schiff, H. I. J. Chem. Phys. 1980,
73, 4976.
(22) Ikezoe, Y.; Matsuoka, S.; Takebe, M.; Viggiano, A. A. Gas Phase
Ion-Molecule Reaction Rate Constants Through 1986; Maruzen Company,
Ltd.: Tokyo, 1987; see also references therein.
(23) Grimsrud, E. P.; Chowdhury, S.; Kebarle, P. J. Chem. Phys. 1985,
83, 1059.
(24) Pan, Y. H.; Ridge, D. P. J. Am. Chem. Soc. 1989, 111, 1150.
(25) Pan, Y. H.; Ridge, D. P. J. Am. Chem. Soc. 1992, 114, 2773.
(26) VanOrden, S. L.; Pope, R. M.; Buckner, S. W. Organometallics
1991, 10, 1089.
Vi. Other Reaction Types. The occurrence of nondissociative
electron transfer (NDET) was discussed in section A.i.
Formation of the dihalide anion ClF- was observed in the
reaction of NO- with CF3Cl but not with the other CF3X and
not from any other anions in the present study. In previous
work on CF3X reactions, we found production of XF- from
O- reactant anion for X ) Cl, Br, and I.1
CF3 product formation occurs in the reactions of Fe- with
-
CF3Cl and CF3Br, with decreasing percentages, respectively.
CF3 is formed in the case of FeCO- + CF3Cl, CF3Br, and
-
CF3I.
In the FeCO- reactions, formation of FeX- is observed along
with three other channels.
Transfer of F+ is a minor pathway in the reaction of SO-
with CF3I.
Conclusions
(27) Bowers, M. T.; Laudenslager, J. B. J. Chem. Phys. 1972, 56, 4711.
(28) Su, T.; Bowers, M. T. J. Chem. Phys. 1973, 58, 3027.
(29) Bates, D. R. J. Phys. B: At. Mol. Phys. 1979, 12, 4135.
(30) Herbst, E. J. Chem. Phys. 1980, 72, 5284.
(31) Keesee, R. G.; Castleman Jr., A. W. J. Phys. Chem. Ref. Data 1986,
15, 1011.
(32) Maricq, M. M.; Tanquay, N. A.; O’Brien, J. C.; Rodday, S. M.;
Rinden, E. J. Chem. Phys. 1989, 90, 3136.
(33) Rinden, E.; Maricq, M. M.; Grabowski, J. J. J. Am. Chem. Soc.
1989, 111, 1203.
(34) Morris, R. A.; Viggiano, A. A.; Paulson, J. F. J. Chem. Phys. 1990,
92, 2342.
(35) Viggiano, A. A.; Morris, R. A.; Paulson, J. F. J. Phys. Chem. 1990,
94, 3286.
A wide range of reactivity is found for the systems studied.
The reactions tend to be fast when nondissociative electron
transfer is energetically allowed, suggesting that reactivity in
these cases is determined by an initial electron transfer mech-
anism. Association is usually most prevalent when nondisso-
ciative electron transfer is endothermic. While the reactions
of SF6 and of the o-, m-, and p-CF3C6H4CN- ions do not
-
-
follow these tendencies, the generally low reactivity of SF6
toward electron transfer is consistent with them, and the behavior
of the CF3C6H4CN- species may be due to similar effects.
(36) Travers, M. J.; Cowles, D. C.; Ellison, G. B. Chem. Phys. Lett.
1989, 164, 449.
Acknowledgment. We thank Amy Miller, Paul Kebarle, and
two reviewers for helpful comments. Technical assistance from
John Williamson and Paul Mundis is appreciated. Partial
JP960355Y