Rearrangement of 3-membered 1,1,2-trifluorobromonium and iodonium ions and comparison of trifluorochloronium to fluorocarbenium ions

Article abstract of DOI:10.1021/jo800472e

(Chemical Equation Presented) Reactions of chlorine (Cl₂) with 4-halo-1,1,2-trifluorobut-1-enes (1, 2, or 3) give open-ion intermediates A and E that are in equilibrium. The open-chloronium ions (E) rearrange to a five-membered-ring halonium ion during ionic chlorination of 3 when the number-4 halo-substituent is iodine. Three-membered-ring bromonium and iodonium ions from alkenes 1, 2, or 3 are rather symmetrical and similar in structure. Quantum chemical calculations show that five-membered-ring halonium ion intermediates are 11 to 27 kcal/mol more stable than the three-membered-ring halonium ions or the open-ions A and E. The five-membered-ring intermediates lead to rearranged products. Rearranged products increase as the number-4 halogen (Z) becomes more nucleophilic (Z: Cl < Br < I). Open chloronium ions from ionic chlorination of terminal fluorovinyl alkenes are compared to the open ions generated by protons to similar alkenes.

Full text of DOI:10.1021/jo800472e

Rearrangement of 3-Membered 1,1,2-Trifluorobromonium and
Iodonium Ions and Comparison of Trifluorochloronium to
Fluorocarbenium Ions
Dale F. Shellhamer,† Kevyn J. Davenport,† Heidi K. Forberg,† Matthew P. Herrick,†
Rachel N. Jones,† Sean J. Rodriguez,† Sunamita Sanabria,† Nicole N. Trager,†
Ryan J. Weiss,† Victor L. Heasley,† and Jerry A. Boatz*,‡
Department of Chemistry, Point Loma Nazarene UniVersity, San Diego, California 92106-2899, and Air
Force Research Laboratory, Edwards Air Force Base, California 93524-7680
dshellha@pointloma.edu
ReceiVed February 28, 2008
Reactions of chlorine (Cl2) with 4-halo-1,1,2-trifluorobut-1-enes (1, 2, or 3) give open-ion intermediates
A and E that are in equilibrium. The open-chloronium ions (E) rearrange to a five-membered-ring halonium
ion during ionic chlorination of 3 when the number-4 halo-substituent is iodine. Three-membered-ring
bromonium and iodonium ions from alkenes 1, 2, or 3 are rather symmetrical and similar in structure.
Quantum chemical calculations show that five-membered-ring halonium ion intermediates are 11 to 27
kcal/mol more stable than the three-membered-ring halonium ions or the open-ions A and E. The five-
membered-ring intermediates lead to rearranged products. Rearranged products increase as the
number-4 halogen (Z) becomes more nucleophilic (Z: Cl < Br < I). Open chloronium ions from
ionic chlorination of terminal fluorovinyl alkenes are compared to the open ions generated by protons
to similar alkenes.
Introduction
In an earlier study we reported on the structure and symmetry
of halonium ion intermediates from fluorosubstituted terminal
alkenes.1 Structures for these intermediates were assigned based
on the distribution of Markovnikov (M) to anti-Markovnikov
(aM) products when the halonium ions were opened by the
solvent methanol. These assignments were refined by quantum
chemical calculations to include structures expected in the gas
phase and from the solvent effects in methanol.2 Halonium ion
structures were found to be open-ion (A or E), unsymmetrical
(B or D), or symmetrical C depending on the halogen electro-
phile and on the position and number of vinylfluorines bonded
to the terminal alkene.1,2 Structures for the three-membered
halonium ions were assigned from bond angles obtained by
quantum chemical calculations.2 The bond angle X-C1-C2 (R)
gave the best description for halonium ion structures A and B,
while the X-C2-C1 bond angle (ꢀ) gave accurate descriptions
for D and E.2 Bond angles around 55-60° for R or ꢀ represent
structure C.2 Data from halogenation reactions of 4-bromo-1,1,2-
trifluorobut-1-ene (2) were explained by neighboring group
participation from the number-4 bromine with the three-
membered halonium ion intermediate to give a five-membered
ring trifluorotetramethylene bromoniun ion 5 (Scheme 1, Z )
Br).1 The steric effect and repulsive forces from the lone-pair
electrons on the fluorine atoms of 5 shield the carbon nucleus,
rendering it resistant to nucleophilic displacement.3 Thus, when
the five-membered ring intermediate is formed, attack by the
† Point Loma Nazarene University.
‡ Edwards Air Force Base.
(1) Shellhamer, D. F.; Allen, J. L.; Allen, R. D.; Gleason, D. C.; O’Neil
Schlosser, C.; Powers, B. J.; Probst, J. W.; Rhodes, M. C.; Ryan, A. J.;
Titterington, P. K.; Vaughan, G. G.; Heasley, V. L. J. Org. Chem. 2003, 68,
3932.
(2) Shellhamer, D. F.; Gleason, D. C.; Rodriguez, S. L.; Heasley, V. L.;
Boatz, J. A.; Lehman, J. J. Tetrahedron 2007, 62, 11608.
4532 J. Org. Chem. 2008, 73, 4532–4538
10.1021/jo800472e CCC: $40.75 2008 American Chemical Society
Published on Web 05/28/2008

Rearrangement of Trifluorobromonium and Iodonium Ions
TABLE 1. Halonium Ion Bond Angles (r, ꢀ) and Charge
SCHEME 1
Densities (C1, C2)a
bond angle
charges
structureb
based on bond
angle R
C2-C1-X X-C2-C1
Q
Q
run
Z
X
Y
(R)
(ꢀ)
(C2)
(C1)
1a Cl Cl Cl
1b Cl Cl Cl
93.8
50.8
68.3
70.2
94.0
68.2
70.0
94.6
50.9
68.1
69.9
47.4
89.6
71.2
73.4
47.3
71.3
73.5
46.9
89.4
71.4
73.7
+0.18 -0.07
-0.11 +0.19
-0.03 +0.05
-0.08 +0.02
+0.18 -0.07
-0.03 +0.05
-0.08 +0.02
+0.18 -0.07
-0.11 +0.19
-0.03 +0.05
-0.08 +0.02
A
E
2
3
Cl Br Br
Cl Br
C
C
A
C
C
A
E
C
C
D
D
I
4
5
Br Cl Cl
Br Br Cl
D
D
6
Br
I
I
I
Cl
7a
7b
8
Cl Cl
Cl Cl
Br Br
I
D
D
9
I
I
Br
a Geometries were calculated at the MP2/6-311++G(P,d) level
theory. Lodwin atomic charges were obtained by using a Mulliken
population analysis based upon symmetrically orthogalized orbitals.
b Structures A through E are shown in the Introduction. The symbol
nucleophile will be at the hydrocarbon rather than the fluoro-
carbon methylene of 5. These rearranged products (Scheme 1)
provide experimental evidence for formation of the tetrameth-
ylene halonium ion intermediates. Neighboring group participa-
tion is well-known, and anchimeric acceleration is greatest for
formation of five-membered rings.4
In this paper we investigate the influence that a 4-halo-
substituent has on formation of three-membered halonium ions
and their rearrangement to five-membered-ring intermediates
when chloronium, bromonium, and iodonium ions from alkenes
1, 2, and 3 are formed in aprotic solvent (Scheme 1). We also
compare the open-ion chloronium ions with the open-carboca-
tions from addition of a proton to terminal fluorosubstituted
alkenes. Ohta has shown by 13C NMR isotope shift studies that
hydrocarbon chloronium and bromonium ions can open to
ꢀ-halocarbenium ions in super acid.5
C
D means that the halonium ion is rather symmetrical like C with
some asymmetry approaching structure D.
reported in the gas phase,6 in superacid media,7 and for the
addition of trifluoroacetic acid to 5-chloro-1-hexene.8 The five-
membered trifluorochloronium ion 5 (Scheme 1, X ) Y ) Z
) Cl) is 19.7 kcal/mol more stable than the open-chloronium
ion A (Table 2, run 1). Calculations also show that the open-
chloronium ion E is only 1.4 kcal/mol less stable than A (Table
2, run 1). If the five-membered-ring 5 (Scheme 1, X ) Y ) Z
) Cl) is formed, then it is from intermediate E in equilibrium
with A. Our data show that a rearranged product through step
3 in Scheme 1 is plausible, but we suspect that formation of 5
(Scheme 1, X ) Y ) Z ) Cl) does not occur. Only 2%
rearranged product was found for chlorination of alkene 2 where
the number-4 halogen is bromine (Table 3, run 5). We would
expect very little rearrangement of the open-chloronium ion E
when the number-4 halogen is the less nucleophilic chlorine.
Bond angle data in Table 1 show that the three-membered-
ring bromonium and iodonium ions from alkene 1 resemble
structure C with some small asymmetry like D written as C D2
(Table 1, runs 2 and 3). Rearranged products from five-
membered-ring halonium ions 5 were not observed for reactions
of bromine (Br2), bromine monochloride (BrCl), or iodine
monobromide (IBr) with alkene 1 (Table 3, runs 2, 3, and 4).
These data show that the number-4 chlorine atom in alkene 1
does not function as a neighboring group participant with either
the three-membered bromoniun or iodonium ions even though
the five-membered-ring intermediates are more stable by 16.2
and 11.6 kcal/mol, respectively (Table 2, runs 2 and 3).
(2) Halonium Ions from 2. Electrophilic addition of Cl2 to
alkene 2 gives 2% of rearranged product (Table 3, run 5).
Intermediate 5 (Scheme 1, X ) Y ) Cl; Z ) Br) is 22.1 kcal/
mol more stable than the open-ion A (Table 2, run 4). Our
calculations did not find a local minimum E for an open-
Results and Discussion
A. Comparing the Effect of the Number-4 Substituent
(Z) on Intermediates from Alkenes 1, 2, and 3. (1)
Halonium Ions from 1. Quantum chemical calculations indicate
an open-ion structure A for the intermediate from reaction of
chlorine with alkene 1 (Table 1, run 1). Reaction of chlorine
(Cl2) with 1 cannot provide experimental evidence for a five-
membered-ring intermediate since migration of the 4-chloro-
substituent cannot be discerned because intermediates 4 and 5
(Scheme 1, X ) Y ) Z ) Cl) each react with chloride ion to
give the same product 1,2,4-trichloro-1,1,2-trifluorobutane (8).
The parent hydrocarbon tetramethylene chloronium ion has been
(3) Banks, R. E.; Smart, B. E., and Tatlow, J. C., Eds. Organofluorine
Chemistry. Principles and Commercial Applications. In Topics in Applied
Chemistry; Plenum Press: New York, 1994; pp 71 and 190. Chambers, R. D.
Fluorine in Organic Chemistry; Wiley: New York, 1973; p 98.
(4) (a) Heine, H. W.; Siegfried, W. J. Am. Chem. Soc. 1954, 76, 489. (b)
Swain, C. G.; Kuhn, D. A.; Schowen, R. L. J. Am. Chem. Soc. 1965, 87, 1553.
(5) Ohta, B. K.; Hough, R. E.; Schubert, J. W. Org. Lett. 2007, 9 (12), 2317.
(6) Van de Sande, C. C.; McLafferty, F. W. J. Am. Chem. Soc. 1975, 97,
2298.
(7) Olah, G. A. J. Am. Chem. Soc. 1968, 90, 4675.
(8) Peterson, P. E.; Tao, E. V. P. J. Am. Chem. Soc. 1964, 86, 4503.
J. Org. Chem. Vol. 73, No. 12, 2008 4533

Shellhamer et al.
TABLE 2. MP2/6-311++G(D,P) Relative Enthalpiesa (∆Ho, kcal/mol)
a Zero-point vibrational energy corrections scaled by 0.9748 (see:. Scott, A. P.; Radom, L.J. Phys. Chem. 1996, 100, 16502-16513). b Two stable
open-ion structures were found, resembling structures A and E. c Did not find an open-ion local minimum for E.
TABLE 3. Products from Reaction of 4-Halo-1,1,2-trifluorobut-1-enes (1, 2, and 3) with Halogen Electrophiles in Methylene Chloride
a Chlorine and bromine in equilibrium with bromine monochloride in methylene chloride gave dichloro (8) and dibromo (9) in a ratio of 10:5.2,
respectively. b Chlorine and bromine in equilibrium with bromine monochloride gave dichloro (14) and dibromo (15) in a ratio of 1.0:2.9, respectively.
c Product ratio at 10-15 min reaction time as the ratios change at longer times. d Rearreanged products were 4-chloro-1,2-dibromo-1,1,2-trifluorobutane
(9) in 38% and 1-bromo-2,4-dichloro-1,1,2-trifluorobutane (10) in 9%. e Product ratio obtained by extrapolating back to t ) 0 since 18 rearranges to 19.
f Product 15 from equilibrium of bromine with iodine monobromine. g Product ratio obtained by extrapolating back to t ) 0 since 20 rearranges to 21.
h Product 24 rearranges to 20 on standing. I Products from reaction of bromine in the 1.0 M IBr solution. They are 24 and 20 in a ratio of 1.0:1.4,
respectively. j Product ratio extrapolated back to t ) 0 since 25 rearranges to 26.
chloronium ion from alkene 2. Structure A was found to be the
local minimum when calculations were started from structure
E. Perhaps the small amount of rearranged product (2%; Table
3, run 5) represents a small contribution from E in methylene
chloride as solvent. Bromine addition to 2 does not provide
evidence for a five-membered-ring intermediate since migration
of the 4-bromosubstituent cannot be discerned. Rearranged
products are not observed for the reactions of ICl or IBr with
alkene 2 (Table 3, runs 8 and 9).
In methylene chloride as solvent, the chlorination of 2 gives
only a small amount of rearranged product (Table 3, run 5).
However, Cl2 and Br2 from dissociation of BrCl gives 9 and
38% rearranged products, respectively (Table 3, run 7). The
reaction of BrCl with alkene 2 is slow and requires about 15
min while the reactions of bromine or molecular chlorine are
very fast. We suggest that Cl2 and Br2 in the presence of BrCl
form a complex (Scheme 2).9 A complex would account for
the slow rate. Also, the resulting halonium cation and anion
4534 J. Org. Chem. Vol. 73, No. 12, 2008

Rearrangement of Trifluorobromonium and Iodonium Ions
SCHEME 2
similar (Table 1, runs 2 and 3) This effect may be due to the
ring-opening difference of the bromine compared to the iodine
in the three-membered halonium ions. Also there is more M
product formed for reaction of ICl than with IBr for reactions
with alkene 2 where identical iodonium ions are formed (Table
3, runs 8 and 9). Perhaps the smaller chloride anion in aprotic
solvent can open the iodonium ion at the internal sterically
hindered number-2 carbon better than the bromide anion (Table
3, runs 8 and 9).
[XBrCl]- pair may be longer lived and allow more time for
rearrangement to the more stable five-membered-ring intermedi-
ate (Scheme 1, step k3).
(5) 4-Bromo-3,3,4,4-tetrafluorobut-1-ene (6). Chlorination
of alkene 6 is very slow and did not give rearranged product.
Ionic addition of Cl2 to 6 is similar in reactivity to ionic reaction
of Cl2 with 1H,1H,2H-perfluorooctene-1.11 Experimental data
in methanol indicated that the chloronium ion from 1H,1H,2H-
perfluorooctene-1 was rather symmetrical (D);11 and calculations
suggest a symmetrical intermediate (C) for the chloronium ion
from 3,3,3-trifluoropropene.2 Thus we expect the chloronium
ion from 6 to be rather symmetrical and that it would not give
rearranged product.
(3) Halonium Ions from 3. Reaction of Cl2 with alkene 3
gives the most rearranged product (25%, Table 3, run 10). The
open-chloronium ion E from alkene 3 is only 1.6 kcal/mol higher
in energy than open-ion A (Table 2, run 7). We suspect that
the rearranged product from chlorination of 3 is from intermedi-
ate E in equilibrium with A. The superb iodine neighboring
group10 accounts for the large amount of rearranged product.
The three-membered-ring bromoniun and iodonium ions from
alkene 3 are quite symmetrical (C D, Table 1, runs 8 and 9),
and no rearranged product is formed with Br2 (Table 3, run
11). The 2% rearranged product from reaction of IBr with 3
probably results from the fact that iodine is a better leaving
group than bromine in the three-membered-ring halonium ion.
(4) Comparison of Halonium Ions from Alkenes 1, 2,
and 3. Three-membered-ring halonium ions are formed better
by iodine than bromine, and bromine bridges better than
chlorine.1 The open-ion structures A for the three chloronium
ions from alkenes 1, 2, and 3 (Table 1, runs 1, 4, and 7)
compared to the more symmetrical structures (C D) for the
bromonium and iodonium ions from these alkenes support our
earlier observations1 that iodine and bromine bridge better than
chlorine. The similar calculated structures for the bromonium
and iodonium ions are surprising.
Our calculations show that the structures of intermediates A
and C D are not dependent on the number-4 substituent. For
example, structures of the open chloronium ions from alkenes
1, 2, and 3 are quite similar (Table 1, runs 1, 4, and 7), as are
the bridged bromonium ions (Table 1, runs 2, 5, and 8) and the
iodonium ions (Table 1, runs 3, 6, and 9). The similar structures
calculated for the bromonium and iodonium ions are consistent
with their nearly comparable product distributions from bro-
mination (Table 3, runs 6 and 11) and iodination (Table 3, note
the similar aM/M product ratios, runs 9 and 12). The small
product differences from these three-membered bromonium vs
iodonium ions cannot be predicted from calculations since their
structures and charge density distributions are similar (Table 1,
compare runs 2, 5, and 8 with runs 3, 6, and 9). The product
differences can perhaps best be explained by the better leaving
group ability of iodine vs bromine in the three-membered
halonium ions (Table 3, compare runs 12 with 11). Also, better
participation by the number-4 iodine compared to bromine may
account for the small amount of rearranged product from the
reaction of IBr with alkene 3 compared to 2 (Table 3, runs 12
and 9). Other factors that may influence the product distribution
from similar calculated structures include the nature of the anion
nucleophile in aprotic solvents (compare the M/aM ratio for
reaction of ICl and IBr with alkene 2, Table 3, runs 8 and 9).
The M/aM product ratio is greater for reaction of IBr than
BrCl with alkene 1 (Table 3, runs 3 and 4) even though their
calculated bromonium and iodonium ions structures are quite
B. Comparison of Intermediates from Addition of a
Proton and Chlorine to Fluorosubstituted Terminal Alkenes.
Acid-catalyzed addition of a proton to the hydrocarbon propene
gives an open unbridged secondary carbocation intermediate.
On the other hand, the chloronium ion from the chlorination of
propene in the gas phase is bridged with a rather symmetrical
structure C containing some small asymmetry approaching B
(C B).2 Chlorination of alkenes like 1, 2, or 3 with three
fluorine atoms results in an open-ion structure A as the lowest
energy intermediate because the alkyl group and back-bond
resonance from the single fluorine on carbon-2 gives a more
stable intermediate than a bridged ion or a cation on the
difluoroterminal carbon. This open-ion structure A is similar to
that expected for addition of open-ion electrophiles like a proton
(H+) to 1,1,2-trifluoroterminal alkenes. Our calculations show
that a proton, like chlorine, also prefers to add to the terminal
carbon of 1,1,2-trifluoropropene placing the positive charge on
the number-2 carbon (Table 4, run 1). The energy difference
between having the charge on the number-2 compared to the
terminal carbon is small (1.3 kcal/mol). Thus one might expect
both regioisomers to undergo addition of a proton to 1,1,2-
trifluoroalkyl-1-enes with the preferred isomer having the added
proton on the terminal carbon. Products from reaction of 70%
perchloric acid with alkene 2 were not stable at the temperatures
required.
Calculations show that the positive charge is greatly preferred
on the number-2 carbon for addition of a proton to 1,2-
difluoropropene and 2-fluoropropene (Table 4, runs 3 and 4).
Hydration of 2-fluorooct-1-ene with perchloric acid/formic acid
catalyst gave only 2-octanone. Similarly, reaction of 1-chlo-
romethyl-4-fluoro-1,4-diazoniabicyclo[2.2.2]octane bi (tetrafluo-
roborate) [F-TEDA-BF4] with 2-fluorooct-1-ene in acetonitrile/
water gave only 1-fluoro-2-octanone (see the Supporting
Information). Earlier calculations predict an open-ion intermedi-
ate for chlorination of 2-fluoropropene,2 and that was confirmed
for chlorination of 2-fluorooct-1-ene in methanol where only
(9) Chiappe, C.; Del Moro, F.; Raugi, M. Eur. J. Org. Chem. 2001, 18, 3501.
(10) Peterson, P. E.; Bopp, R. J.; Chevli, D. E.; Curran, E. L.; Dillard, D. E.;
Kamat, R. J. J. Am. Chem. Soc. 1967, 89, 5902.
(11) Shellhamer, D. F.; Allen, J. L.; Allen, R. D.; Bostic, M. J.; Miller, E. A.;
O’Neil, C. M.; Powers, B. J.; Price, E. A.; Probst, J. W.; Heasley, V.L. J. Fluorine
Chem. 2000, 106, 103.
J. Org. Chem. Vol. 73, No. 12, 2008 4535

Shellhamer et al.
TABLE 4. Comparison of the Intermediates from Proton (H+) and Chlorine (Cl2) Addition to Fluorosubstituted Propenesa
a Calculations were performed at the MP2 level using a 6-311+G** basis set on the Spartan 02 program. b A negative ∆H means the positive charge
is more stable on carbon-1 than carbon-2. c Data from: Shellhamer, D. F.; Gleason, D. C.; Rodriguez, S. L.; Heasley, V. L.; Boatz, J. A; Lehman,
J. J.Tetrahedron, 2006, 62, 11609.
1,2-dichloro-2-fluorooctane and the M product 1-chloro-2-
fluoro-2-methoxyoctane was found.1 The open-ion intermediate
A for chlorination of 1,2-difluoropropene2 is similar to that
expected for addition of a proton (Table 4, run 3).
halogen electrophiles to alkenes 2 or 3 where the number-4
substituent is a bromine or iodine. The three strong electron-
withdrawing vinyl fluorine atoms on alkenes 1, 2, and 3
attenuate the neighboring group effect since we did not find
evidence for a neighboring group effect when the number-4
substituent is chlorine (alkene 1). This is in contrast to the
hydrocarbon where five-membered-ring tetramethylene chlo-
ronium ions are readily formed.6–8 For halogen substituents
on the number-4 carbon of 1,1,2-trifluoroterminal alkenes,
iodine participates in neighboring group rearrangement better
than bromine and the involvement of chlorine is not indicated.
Quantum chemical calculations show that the chloronium ions
from alkenes 1, 2, or 3 are unbridged and open-ions that form
ꢀ-chlorocarbenium ions are similar to those reported by Ohta
in super acid.5 The increase in rearranged products is due to
the increase in nucleophilicity of the number-4 halogen (Z)
where I > Br > Cl. Bromonium and iodonium ions from
these alkenes are rather symmetrical and similar in structure.
The bromonium ions bridge well enough such that no
rearranged products are found, except for the BrCl reaction
where a complex changes the reaction. Experimental product
distributions for bromination and iodination reactions with
alkenes 2 or 3 cannot be predicted from their calculated
structures because their intermediates are similar. Product
distributions do correlate with structural differences of
halonium ions when their calculated structures are different.1,2
Calculations show that an open-ion with positive charge on
the terminal carbon is also greatly preferred for addition of a
proton or a chlorine2 electrophile to 1,1-difluoropropene (Table
4, run 2). However, the positive charge is only slightly favored
(2.5 kcal/mol) on the terminal carbon for protination of
1-fluoropropene (Table 4, run 5). Products were not stable to
the reaction conditions for acid-catalyzed hydration of 1-fluo-
rooct-1-ene. Our earlier calculations show that the intermediate
for chlorination of 1-fluoropropene is bridged and rather
symmetrical in the gas phase represented as C D, but it is
less symmetrical (D C) when corrected for the solvent
methanol.2 Experimentally in methanol as solvent, the chloro-
nium ion is highly unsymmetrical and may even be an open-
ion since chlorination of (E)- or (Z)-1-fluorooct-1-ene gave only
1,2-dichloro-1-fluorooctane and the aM product 2-chloro-1-
fluoro-1-methoxyoctane.1 A bridged chloronium ion forms with
1-fluoro-1-alkenes because neither the terminal cation stabilized
by back-bond resonance from the number-1 fluorine nor the
number-2 secondary carbon cation are as stable as the bridged
chloronium ion.
Conclusion
The chloronium ions from terminal alkenes with vinyl
fluorines tend to be open-ions2 similar to the addition of a proton
We have shown that five-membered-ring trifluorotetram-
ethylene halonium ions (5) are indicated for reaction of some
4536 J. Org. Chem. Vol. 73, No. 12, 2008

Rearrangement of Trifluorobromonium and Iodonium Ions
2-Bromo-4-chloro-1-iodo- (12) and 1-Bromo-4-chloro-2-iodo-
1,1,2-trifluorobutanem (13). IBr, 1, 12 h, 25 °C, 70% by NMR
with benzene as internal standard.
except for the bridged chloronium ion from 1-fluoro-1-alkenes.
A bridged chloronium ion from 1-fluoroterminal alkenes is rather
symmetrical because the terminal fluorine and the alkyl group
on the internal carbon each provide similar charge stabilization.2
4-Bromo-1,2-dichloro- (14) and 1,2,4-Tribromo-1,1,2-trifluo-
robutane (15). See preparative scale synthesis in the Supporting
Information.
1,4-Dibromo-2-chloro- (16)14,15 and 2,4-Dibromo-1-chloro-
1,1,2-trifluorobutane (17). BrCl, 2, 15 min., longer reaction times
gave different product ratios due to thermodynamic rearrangement,
25 °C, 54% isolated yield by preparative GC.
Experimental Section
A. General Methods. Alkenes 1 and 3 were prepared from
commercially available 2 in 37% and 83% yield, respectively.
Alkene 1 was from reaction of dry lithium chloride with 2 in dry
DMSO at 100 °C for 1 h in a pressure bottle. Alkene 3 (bp 117 °C
at 760 Torr) was from reaction of 2 and potassium iodide with
acetone as solvent at 90 °C for 4 h. Alkene 6 was commercially
available. 2-Fluorooct-1-ene12 and 1-fluorooct-1-ene13 were syn-
thesized from literature preparations.
4-Bromo-2-chloro-1-iodo- (18) and 4-Bromo-1-chloro-2-io-
dobutane (19). ICl, 2, 15 min, the product ratio in Table 1 was
obtained by extrapolating back to t ) 0 due to equilibration, 25
°C, 74% by GC with 15 as internal standard. Compounds 18 and
19 were reported earlier.1
Halogenation reactions were carried out as follows: Chlorine gas
was slowly bubbled into a 1.0 M methylene chloride solution of
alkene 1, 2, 3, or 6 at room temperature and the progress followed
by gas chromatography. For Br2, ICl, or IBr, alkene 1, 2, or 3 (1.0
mmol) was added to 1.0 mmol of the halogen or interhalogen in
1.0 mL of methylene chloride at room temperature. Bromine
monochloride (0.62 M in CH2Cl2) was prepared by adding an
equivalent amount of Br2 to a 0.62 M methylene chloride solution
of Cl2.
2,4-Dibromo-1-iodo- (20) and 1,4-Dibromo-2-iodo-1,1,2-tri-
fluorobutane (21). IBr, 2, 12 h, product ratios in Table 1 were
obtained by extrapolating back to t ) 0, 25°, 80% by GC with 15
as internal standard.
1,2-Dichloro-4-iodo- (22) and 2,4-Dichloro-1-iodo-1,1,2,-tri-
fluorobutane (23). Cl2, 3, 15 min, 25 °C, 85% by GC with 14 as
internal standard.
1,2-Dibromo-4-iodo-1,1,2-trifluorobutane (24). Br2, 3, 15 min,
25 °C, 95% by GC with 14 as internal standard.
Product structural descriptions and ratios for reactions of halogen
electrophiles with alkenes 1, 2, 3, and 6 are given in Table 3.
Reaction times and percent yields are in the Analytical section
below. Most of the products were purified by preparative GC with
2-Bromo-1,4-diiodo- (25), 1-Bromo-2,4-diiodo- (26), and
4-Bromo-1,2-diiodo-1,1,2-trifluorobutane (27). IBr, 3, 12 h,
product ratios in Table 1 were obtained by extrapolating back to t
) 0, 50 °C, 72% by GC with 14 as internal standard.
C. Reactions with 2-Fluorooct-1-ene. See the Supporting
Information.
D. Theoretical Methods. Geometry optimizations and vibra-
tional frequency calculations were performed at the second-order
perturbation theory level (MP2, also known as MBPT(2)),16 using
the GAMESS17 quantum chemistry code. The 6-311++G(d,p) basis
set18 was used for all calculations. Harmonic vibrational frequencies
were calculated for each structure to verify that the optimized
structure is a local minimum on the ground-state potential energy
surface. Lo¨wdin atomic charges were obtained by using a Mulliken
population analysis19 based upon symmetrically orthogonalized
orbitals.20
3
a stainless steel 6 ft × /8 in. column of 5% OV-17 on Chromosorb
W 80/100. The remaining products were isolated by distillation
from preparative scale reactions, or they were independently
synthesized and/or converted by SN2 reactions to known compounds.
Product 8 was compared to a commercial sample, and 16 is a
known compound.14,15 We characterized compounds 14, 15, 18,
and 19 earlier.1 Products 7, 9, 12, 13, 17, 21, 22, 23, and 24 are
characterized in the Supporting Information. Several reaction
products were converted by SN2 reaction of halide ion to replace
the number-4 halogen converting it to a known compound. Thus
compounds 10, 11, 22, and 24 were converted by SN2 reactions to
known or characterized compounds 16, 17, 14, and 15 (Supporting
Information). Products 16, 22, and 24 were also independently
synthesized, while 20, 25, 26, and 27 decomposed during attempted
purification and except for 26 they are minor products (Supporting
Information).
Acknowledgment. Support for this work was provided by
the National Science Foundation (NSF-RUI Grant Nos. CHE-
0345551 and CHE-0640547), and Research Associates of PLNU
(alumni support group). We would also like to acknowledge
our use of the 400 MHz NMR at the University of San Diego
obtained by support from the National Science Foundation (NSF
MRI Grant No. CHE-0417731). We thank Dr. Richard Kondratt
at the Mass Spectrometry Center, the University of California,
Riverside for the Exact Mass data.
B. Analytical Reactions. The following reactions of Cl2 and
Br2 with alkene 1 are representative.
1,2,4-Trichloro-1,1,2-trifluorobutane (8). To a stirred solution
of 1 (1.00 mmol) in 1.0 mL of methylene chloride at room
temperature was slowly bubbled Cl2 until all of the alkene was
consumed. Product 8 was formed in 52% yield by GC analysis
with pure 15 as internal standard. Spectral data for 8 were identical
with those of a commercial sample.
1,2-Dibromo-4-chloro-1,1,2-trifluorobutane (9). To a stirred
solution of 1.00 mmol of Br2 in 1.0 mL of methylene chloride at
room temperature was added 145 mg (1.00 mmol) of 1. Product 9
was formed in 78% yield as determined by NMR analysis with
benzene as internal standard. Product 9 is characterized in the
Supporting Information.
Similarly (electrophile, alkene, time, percent yield) gave the
following:
1,4-Dibromo-2-chloro- (10) and 2,4-Dibromo-1-chloro-1,1,2-
trifluorobutane (11). BrCl, 1, 20 min, 25 °C, 80% by GC with 15
as internal standard.
(16) (a) Moller, C.; Plesset, M. S. Phys. ReV. 1934, 46, 618. (b) Pople, J. A.;
Binkley, J. S.; Seeger, R. Int. J. Quantum Chem. S10 1976, 1. (c) Frisch, M. J.;
Head-Gordon, M.; Pople, J. A. Chem. Phys. Lett. 1990, 166, 275. (d) Bartlett,
R. J.; Silver, D. M. Int. J. Quantum Chem. Symp. 1975, 9, 1927.
(17) (a) Gordon, M. S.; Schmidt, M. W., as cited in: Dykstra, C. E.; Frenking,
G.; Kim, K. S.; Scuseria, G. E. Theory and Applications of Computational
Chemistry: The First Forty Years; Elsevier: Amsterdam, The Netherlands, 2005.
(b) Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.; Gordon, M. S.;
Jensen, J. H.; Koseki, S.; Matsunaga, N.; Nguyen, K. A.; Su, S. J.; Windus,
T. L.; Dupuis, M.; Montgomery, J. A. J. Comput. Chem. 1993, 14, 1347–1363.
(18) (a) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem. Phys.
1980, 72, 650. (b) Blandeau, J.-P.; Davis, N. E.; Binning, R. C., Jr.; Radom, L.
J. Chem. Phys. 1995, 103, 6104. (c) Clark, T.; Chandrasekhar, J.; Schleyer, P.
v. R. J. Comp. Chem. 1983, 4, 294.
(12) Eckes, L.; Hanack, M. Synthesis 1978, 207.
(13) Cox, D. G.; Gurusamy, N.; Burton, D. J. J. Am. Chem. Soc. 1985, 107,
2811.
(19) (a) Mulliken, R. S. J. Chem. Phys. 1955, 23, 1833–1840. (b) Mulliken,
R. S. J. Chem. Phys. 1955, 23, 1841–1846. (c) Mulliken, R. S. J. Chem. Phys.
1955, 23, 2338–2342. (d) Mulliken, R. S. J. Chem. Phys. 1955, 23, 2343–2346.
(20) Lo¨wdin, P.-O. AdV. Chem. Phys. 1970, 5, 185–199.
(14) Tarrant, P.; Gillam, E. G. J. Am. Chem. Soc. 1954, 76, 5423.
(15) Hinton, J. F.; Jaques, L.W. J. Magn. Reson. 1975, 17, 95.
J. Org. Chem. Vol. 73, No. 12, 2008 4537

Shellhamer et al.
Supporting Information Available: Procedures for inde-
pendent synthesis of 16, 22, and 24 and preparative scale
reactions to make 7, 14, and 15, along with procedures to
convert 24, 22, 10, and 11 to 15, 14, 16, and 17, respectively,
along with NMR (1H, 19F, 13C) and GC/MS ([CH2Z]•+,
[CF2X]•+, M+, descriptive fragmentation and isotope cluster)
data. This material is available free of charge via the Internet
at http://pubs.acs.org.
JO800472E
4538 J. Org. Chem. Vol. 73, No. 12, 2008