Tetraalkylphosphonium trihalides. Room temperature ionic liquids as halogenation reagents

Article abstract of DOI:10.1021/jo901735h

(Chemical Equation Presented) Six room temperature ionic liquids (RTILs) comprised of a tetraalkylphosphonium cation (tridecylmethyphosphonium or trihexyltetradecylphosphonium) and a trihalide anion (Br₃ ^-, BrCl₂^-, or ClBr₂^-) have been prepared and characterized. Their ability to effect halogenation reactions with a variety of substrates has been explored. In general, halogenation reactions of alkenes proceed with high yields and stereo- and regioselectivities, whether performed in the absence or presence of a solvent (chloroform). Reactions of an alkyne and electrophilic substitution on an aromatic ring have been investigated as well. The facile preparation of the salts, their ease of handling, and the simplicity of product isolation should make these RTILs useful additions to the repertoire of halogenation reagents and the reagents of choice for specific transformations.

Full text of DOI:10.1021/jo901735h

pubs.acs.org/joc
Tetraalkylphosphonium Trihalides. Room Temperature Ionic Liquids As
Halogenation Reagents§
Rodrigo Cristiano,† Kefeng Ma, George Pottanat, and Richard G. Weiss*
Department of Chemistry, Georgetown University, Washington, D.C. 20057-1227. †Present address:
´
ꢀ
Departamento de Quımica, Universidade Federal de Santa Catarina, Florianopolis, SC, Brasil.
weissr@georgetown.edu
Received August 10, 2009
Six room temperature ionic liquids (RTILs) comprised of a tetraalkylphosphonium cation (tridecyl-
methyphosphonium or trihexyltetradecylphosphonium) and a trihalide anion (Br3-, BrCl2-, or ClBr2
-
)
have been prepared and characterized. Their ability to effect halogenation reactions with a variety of
substrates has been explored. In general, halogenation reactions of alkenes proceed with high yields and
stereo- and regioselectivities, whether performed in the absence or presence of a solvent (chloroform).
Reactions of an alkyne and electrophilic substitution on an aromatic ring have been investigated as well.
The facile preparation of the salts, their ease of handling, and the simplicity of product isolation should
make these RTILs useful additions to the repertoire of halogenation reagents and the reagents of choice
for specific transformations.
Introduction
Because the dihalogens are corrosive, poisonous, and can be
dangerous to handle, methods that require their transport
and manipulation are difficult. A comparison of many halo-
genation techniques and a determination of their environ-
mental impacts with an approach related to the Sheldon
factor3 have appeared recently.4 Although reactions employ-
ing dihalogen molecules are among the most studied in orga-
nic chemistry,4,5 questions still remain about how they occur
in specific cases, especially those involving addition of a mixed
halogen molecule,6 reaction media with varied structures,7
Organic molecules containing halogen atoms are found in
many products, such as pharmaceuticals, agrochemicals,
and materials for advanced technologies. In organic synth-
esis, they are important intermediates in reactions involving
carbon-carbon bond formation, organometallic prepara-
tions (e.g., Grignard and organo-lithium reagents) and metal
cross-coupling reactions.1,2 A common method to insert
halogen atoms into organic substrates, whether they be
free-radical processes or polar additions to olefinic groups
or electrophilic substitutions on aromatic ones, involves the
use of dihalogen molecules which have high vapor pressures
or are gases at room temperature and 1 atm of pressure.
(3) (a) Andraos, J. Org. Process Res. Dev. 2005, 9, 149–163. (b) Ibid.
404-431. (c) Sheldon, R. A. CHEMTECH 1994, 3, 38–47.
(4) Eissen, M.; Lenoir, D. Chem.;Eur. J. 2008, 14, 9830-9841 and
references cited therein.
(5) (a) Brown, R. S. Acc. Chem. Res. 1997, 30, 131–137. (b) Ruasse, M. F.
Adv. Phys. Org. Chem. 1993, 28, 207–291. (c) Lenoir, D.; Chiappe, C.
Chem.;Eur. J. 2003, 9, 1037–1044.
(6) In chloroform solution: (a) Chiappe, C.; Del Moro, F.; Raugi, M.
Eur. J. Org. Chem. 2001, 3501–3510. In aqueous solution: (b) Wang, T. X.;
Kelley, M. D.; Cooper, J. N.; Beckwith, R. C.; Margerum, D. W. Inorg.
Chem. 1994, 33, 5872–5878.
*To whom correspondence should be addressed. Phone: þ1-202-687-
6013. Fax: þ1-202-687-6209.
§ This article is dedicated to the memory of Dr. Paul M. Kuznesof
(1941-2009), an excellent scientist and close friend of R.G.W.
(1) De Meijere, A.; Diederich, F., Eds. Metal-catalyzed Cross-coupling
Reactions, 2nd ed.; Wiley-VCH: Weinheim, Germany, 2004.
(2) (a) Yao, Q.; Kinney, E. P.; Yang, Z. J. Org. Chem. 2003, 68, 7528–
7531. (b) Seyferth, D. Organometallics 2009, 28, 1598–1605.
(7) Carey, F. A.; Sundberg, R. J. Advanced Organic Chemistry, Part A,
5th ed.; Springer: New York, 2007; pp 485-497.
DOI: 10.1021/jo901735h
r
Published on Web 11/06/2009
J. Org. Chem. 2009, 74, 9027–9033 9027
2009 American Chemical Society

Cristiano et al.
JOCArticle
and reagents with the halogen atoms as part of a larger spe-
cies.4,8-11
SCHEME 1. Preparation of Tetraalkylphosphonium Trihalide
Salts, mPnXY2
In that regard, the high polarity of ionic liquids makes
them interesting media in which to conduct and investigate
the selectivity and reactivity of ionic halogenation reactions,
such as the aforementioned addition and substitution reac-
tions.12 Also, ionic liquids avoid many of the problems
associated with the volatility of common organic media,
and many ionic liquids can be separated easily from products
and/or catalysts and reused.13 Another approach to media-
ting the selectivity and reactivity of halogenations involves
to examine the role of the cation and anion structures on the
kinetics of the reactions, the stability of the reaction inter-
mediates, and, thus, the reactivity and selectivity of the
substrates.
-
using trihalide anions, XY2 (X, Y = Cl, Br, I), as the
halogen reagent,11 and studies that combine trihalide ions
and ionic liquids as the medium have been conducted.11e
Finally, the trihalide anion has been made a part of ionic
liquids so that the solvent alone is an efficient halogenating
material.14 For example, halogenations in salts of tetraalk-
ylammonium15 and imidazolium16 trihalides have been used
As part of our continuing studies of phase properties and
uses of phosphonium salts,17 we developed recently an ionic
liquid, tridecylmethylphosphonium tribromide (1P10Br3),
that functions as a brominating reagent in a stacked, “phase-
vanishing”18 reactor that exploits the diffusion of molecular
bromine and the densities and immiscibilities of bromine,
a fluorous liquid, the ionic liquid, and alkane/substrate
layers.14a Our use of phosphonium rather than ammonium
or imidazolium salts is motivated by the fact that Pþ has a
larger size, a greater polarizability, and a lower binding
energy with anions than Nþ 17 and the phosphonium salts
are structurally simpler molecules than the imidazolium
salts. These attributes make phosphonium salts attractive
and interesting alternative halogenation reagents.
Here, we describe the preparation of six room temperature
ionic liquids, mPnXY2 (Scheme 1), made from combinations
of two tetraalkylphosphonium cations (tridecylmethyphos-
phonium and trihexyltetradecylphosphonium) and three tri-
halide anions (Br3-, BrCl2-, and ClBr2-) and theiruse as both
solvents and halogenation reagents for bromination and
bromo-chlorination reactions with unsaturated substrates.
Both the regio- and stereoselectivities of these reactions with
a wide variety of substrates and the influence on these para-
meters of different aliphatic chains of the mPnXY2 are dis-
cussed. It is shown that the reactions proceed in high yields
and selectivities under mild conditions and that the nature of
the mixed halogenations can be controlled somewhat.
(8) For bromination reactions involving Br2: (a) Rajasingh, P.; Cohen,
R.; Shirman, E.; Shimon, L. J. W.; Rybtchinski, B. J. Org. Chem. 2007, 72,
5973–5979. (b) Suresh, P.; Annalakshmi, S.; Pitchumani, K. Tetrahedron
2007, 63, 4959–4967. (c) Okada, Y.; Yokozawa, M.; Akiba, M.; Oishi, K.;
O-kawa, K.; Akeboshi, T.; Kawamura, Y.; Inokuma, S.; Nakamura, Y.;
Nishimura, J. Org. Biomol. Chem. 2003, 1, 2506–2511. (d) Das, J. P.; Say, S.
J. Org. Chem. 2002, 67, 7861–7864. (e) Adimurthy, S.; Ramachandraiah, G.;
Bedekar, A. V.; Ghosh, S.; Ranu, B. C.; Ghosh, P. K. Green Chem. 2006, 8,
916–922. (f) Nair, V.; Panicker, S. B.; Augustine, T.; George, T. G.; Thomas,
S.; Vairamani, M. Tetrahedron 2001, 57, 7417–7422.
(9) For asymmetric halogenations involving dihalogens: (a) Li, K. Y.;
Alexakis, A. Tetrahedron Lett. 2005, 46, 5823–5826. (b) Goswarni, P.;
Baruah, A.; Das, B. Adv. Synth. Catal. 2009, 351, 1483–1487. (c) Ueda,
M.; Kano, T.; Maruoka, K. Org. Biomol. Chem. 2009, 7, 2005–2012. For
chlorination reactions: (d) Murphy, C. D. Nat. Prod. Rep. 2006, 23, 147–152.
(e) Masilamani, D.; Rogic, M. M. J. Org. Chem. 1981, 46, 4486–4489.
(f) Moreno-Dorado, F. J.; Guerra, F. M.; Manzano, F. L.; Aladro, F. J.;
Jorge, Z. D.; Massanet, G. M. Tetrahedron Lett. 2003, 44, 6691–6693.
(g) Hajra, S.; Bhowmick, M.; Maji, B.; Sinha, D. J. Org. Chem. 2007, 72,
4872–4876. (h) Frings, M.; Bolm, C. Eur. J. Org. Chem. 2009, 4085–4090.
(i) Chen, J. S.; Takenaka, N. Chem.;Eur. J. 2009, 15, 7268–7276.
(10) For iodination and fluorination reactions involving dihalogens:
(a) Brazdil, L. C.; Cutler, C. J. J. Org. Chem. 1996, 61, 9621–9622. (b) Beeson,
T. D.; MacMillan, D. W. C. J. Am. Chem. Soc. 2005, 127, 8826–8828.
(11) For halogenation reactions involving trihalide anions: (a) Zolfigol,
M. A.; Chehardoli, G.; Salehzadeh, S.; Adams, H.; Ward, M. D. Tetrahedron
Lett. 2007, 48, 7969–7973. (b) Bose, G.; Mondal, E.; Khan, A. T.; Bordoloi,
M. J. Tetrahedron Lett. 2001, 42, 8907–8909. (c) Kavala, V.; Naik, S.; Patel,
B. K. J. Org. Chem. 2005, 70, 4267–4271. (d) Hajipour, A. R.; Arbabian, M.;
Ruoho, A. E. J. Org. Chem. 2002, 67, 8622–8624. (e) Chiappe, C.; Pieraccini,
D. J. Org. Chem. 2004, 69, 6059–6064. (f) Bianchini, R.; Chiappe, C. J. Org.
Chem. 1992, 57, 6474–6478. (g) Giordano, C.; Coppi, L. J. Org. Chem. 1992,
57, 2765–2766. (h) Bianchini, R.; Chiappe, C.; Lo Moro, G.; Lenoir, D.;
Lemmen, P.; Goldberg, N. Chem.;Eur. J. 1999, 5, 1570–1580. (i) Smith, K.;
James, D. M.; Matthews, I.; Bye, M. R. J. Chem. Soc., Perkin Trans. 1 1992,
1877–1878.
(12) (a) Chiappe, C.; Pieraccini, D.; Pomelli, C. S. Ionic Liquids in Organic
Synthesis; Malhotra, S. V., Ed.; ACS Symp. Ser. No. 950; American Chemical
Society: Washington, DC, 2007; Chapter 1. (b) Pavlinac, J.; Zupan, M.; Laali,
K. K.; Stavber, S. Tetrahedron 2009, 65, 5625–5662.
(13) (a) Abedin, S. Z. E.; Endres, F. Acc. Chem. Res. 2007, 40, 1106–1113.
(b) Rogers, R. D.; Voth, G. A. Acc. Chem. Res. 2007, 40, 1077–1078. (c) Ionic
Liquids III: Fundamentals, Progress, Challenges, and Opportunities; Rogers,
R. D., Seddon, K. R., Eds.; ACS Symp. Ser. Nos. 901-902; American Chemical
Society: Washington, DC, 2005. (d) Dupont, J.; de Souza, R. F.; Suarez, P. A. Z.
Chem. Rev. 2002, 102, 3667–3692.
(14) Halogenation reactions involving trihalide ionic liquids: (a) Ma, K.;
Li, S.; Weiss, R. G. Org. Lett. 2008, 10, 4155–4158. (b) Chiappe, C.; Leandri,
E.; Pieraccini, D. Chem. Commun. 2004, 2536–2537. (c) Borikar, S. P.;
Daniel, T.; Paul, V. Tetrahedron Lett. 2009, 50, 1007–1009. (d) Salazar, J.;
Dorta, R. Synlett 2004, 7, 1318–1320. (e) Bortolini, O.; Bottai, M.; Chiappe,
C.; Conte, V.; Pieraccini, D. Green Chem. 2002, 4, 621–627. (f) Kessat, A.;
Babadjamian, A. Eur. Polym. J. 1996, 32, 193–199. (g) Cristiano, R.; Walls,
A. D.; Weiss, R. G. To be submitted for publication.
Results and Discussion
Synthesis and Characterization of Phosphonium Trihalides.
The mPnXY2 were prepared as shown in Scheme 1. Trihalide
(17) (a) Abdallah, D. J.; Bachman, R. E.; Perlstein, J.; Weiss, R. G.
J. Phys. Chem. B 1999, 103, 9269–9278. (b) Abdallah, D. J.; Robertson, A.;
Hsu, H.-F; Weiss, R. G. J. Am. Chem. Soc. 2000, 122, 3053–3062.
(c) Abdallah, D. J.; Weiss, R. G. Chem. Mater. 2000, 12, 406–413.
(d) Abdallah, D. J.; Lu, L.; Cocker, M. T.; Bachman, R. E.; Weiss, R. G.
€
Liq. Cryst. 2000, 27, 831–837. (e) Hui, C.; Kwait, D. C.; Gonen, Z. S.;
Weslowski, B. T.; Abdallah, D. J.; Weiss, R. G Chem. Mater. 2002, 14, 4063–
4072. (f) Abdallah, D. J.; Wauters, H. C.; Kwait, D. C.; Khetrapal, C. L.; Nagana
Gowda, G. A.; Robertson, A.; Weiss, R. G. Ionic Liquids IIIB: Fundamental,
Progress, Challenges and Opportunities; Rogers, R. D., Seddon, K. R., Eds.;
ACS Symp. Ser. No. 902; American Chemical Society: Washington, DC, 2005;
Chapter 21. (g) Ma, K.; Somashekhar, B. S.; Nagana Gowda, G. A.; Khetrapal,
C. L.; Weiss, R. G. Langmuir 2008, 24, 2746–2758. (h) Ma, K.; Lee, K.-M.;
Minkova, L.; Weiss, R. G. J. Org. Chem. 2009, 74, 2088–2098. (i) Shahkhatuni,
A. A.; Ma, K.; Weiss, R. G. J. Phys. Chem. B 2009, 113, 4209–4217. (j) Ma, K.;
Shahkhatuni, A. A.; Somashekhar, B. S.; Nagana Gowda, G. A.; Tong, Y.;
Khetrapal, C. L.; Weiss, R. G. Langmuir 2008, 24, 9843–9854.
(15) Bora, U.; Chaudhuri, M. K.; Dey, D.; Dhar, S. S. Pure Appl. Chem.
2001, 73, 93–102.
(18) (a) Ryu, I.; Matsubara, H.; Yasuda, S.; Nakamura, H.; Curran, D. P.
J. Am. Chem. Soc. 2002, 124, 12946–12947. (b) Rahman, M. T.; Kamata, N.;
Matsubara, H.; Ryu, I. Synlett 2005, 2664–2666. (c) Jana, N. K.; Verkade,
J. G. Org. Lett. 2003, 5, 3787–3790.
(16) (a) Chiappe, C.; Conte, V.; Pieraccini, D. Eur. J. Chem. 2002, 16,
2831–2837. (b) Chiappe, C.; Pieraccini, D. J. Org. Chem. 2004, 69, 6059–
6064.
9028 J. Org. Chem. Vol. 74, No. 23, 2009

Cristiano et al.
JOCArticle
SCHEME 2. Halogenation Reactions of Substrates with mPnXY2 Reagents
anions were generated in the salts by reaction of a mPnX with
1 mol equiv of a dihalogen molecule, Y2, in a closed flask
protected from light. Although this reaction is slightly
exothermal, it can be carried out in the absence or presence
of an aprotic solvent such as dichloromethane, even when the
mPnX is a solid. Without solvent, the reactants should be
stirred well to ensure that the reaction is complete; crystals of
the phosphonium halide slowly “melt” as they are converted
to a liquid trihalide salt.
All of the mPnXY2 with m=1 and n=10 (1P10XY2) and
m = 14 and n = 6 (14P6XY2) are room temperature ionic
liquids (RTILs). Preliminary investigations with some other
trihalide salts;4P4Br3, 4P4BrI2, 1P14I3, and 1P14BrI2;
were abandoned when it was found that they were solids at
room temperature or when it was found that the trichloride
preparations (from mPnCl and Cl2) did not lead to salts that
could be characterized well; they may be useful in other
halogenation processes, however. The liquid range of the
RTIL trihalides is very broad although some may become
glasses at very low temperatures: differential scanning calo-
rimetry (DSC) analyses did not show any cooling exotherms
to temperatures as low as -60 °C and thermogravimetric
analyses (TGA) indicated <1% weight loss up to 100 °C.
Thermal decomposition of 14P6Br3 and the previously
reported 1P10Br314a involve loss of Br2 and reformation of
the original monobromide salts, mPnBr: the weight loss
curve for 1P10Br3 (see the Supporting Information) shows
a small loss starting at ca. 100 °C and then abruptly increas-
ing at 175 °C to ca. 24%; the weight percent of Br2 in the salt
is 22.35%. The changes that occur upon heating are more
complex in the cases of the mixed halides (N.B., TGA curves
for 1P10ClBr2 and 1P10BrCl2 in the Supporting Infor-
mation); more than one step in the weight loss is clearly
evident. Although dihalogen molecules appear to be re-
leased, their nature is not clear and may be a combination
of BrCl, Br2, and/or Cl2. An additional test of stability was
conducted in which 1P10BrCl2 was placed in a N2 stream
for 30 min at 25 °C; the weight loss after this time, <0.3%
(see the Supporting Information), indicated that the trihalide
remained intact.
The structures of the RTILs were characterized by 1H and
31P NMR spectrometry and elemental analyses (see the
Experimental Section and the Supporting Information).
Visually, qualitative evidence for trihalide formation is
found in comparison of the appearances of the mPnX
precursors and their mPnXY2 products. For example,
1P10ClBr2 is a red-orange liquid and 1P10Cl is a white solid.
The transformations of the solids into liquids was expected
J. Org. Chem. Vol. 74, No. 23, 2009 9029

Cristiano et al.
JOCArticle
as a consequence of the larger sizes and greater electron
delocalizations (resulting in looser association with the
cationic centers) of the trihalide anions.19 For example, the
chemical shifts of the protons of the methylene and methyl
groups linked to the phosphorus atom of 1P10Cl appear
at 2.48-2.42 (m, 6H, CH2-P) and 2.14 (d, J=13.6 Hz, 3H,
P-CH3) ppm, respectively, while those for 1P10ClBr2 are
shifted upfield to 2.34-2.27 (m, 6H, CH2-P) and 2.00 (d,
J = 12.8 Hz, 3H, P-CH3) ppm.20 Similarly, the 31P reso-
nance shifts from 32.64 ppm in 1P10Cl to 31.95 ppm in
1P10ClBr2. Although the elemental analyses are within
acceptable limits, the observed hydrogen/carbon ratios are
always larger than the calculated ones. This may be an
indication that traces of water are present in the salts.17h
Bromination and Bromo-Chlorination Reactions. Halo-
genation reactions at room temperature in solvent-free con-
ditions or in solutions were performed in the dark (to avoid
radical reactions21) and in closed vessels on the structurally
diverse substrates in Scheme 2 with the salts indicated in
Table S1 (Supporting Information). The mixtures were
stirred only in those cases where the substrate and salt did
not form a solution.
The substrates 1-6 were chosen because all of the expected
(and found) products and their NMR spectra are known;
product ratios are based on integration ratios from NMR
spectra and, in some cases, they were verified by GC ana-
lyses; see the Supporting Information. These reactions were
performed neat and in CDCl3 solutions in order to compare
the results in media of different polarity and different
trihalide ion concentrations. From the observation that the
selectivities were similar in both neat and chloroform solu-
tion reactions, we conclude that they are not very sensitive to
polarity, at least within the range investigated here. It is
known that very little dissociation of trihalides (N.B., a
contribution to reaction from dissociation of BrCl2- cannot
be discounted; vide infra) occurs in chloroform6,11e,f and
the use of CDCl3 allowed the progress and course of the
reactions to be followed directly by NMR spectroscopy:
because the 1H signals for the mPnX and mPnXY2 salts are
between 1.8 and 2.4 ppm and the diagnostic proton reso-
nances of the unsaturated substrates and their halogenated
products are in a very different (downfield) region, from
3.5 to 8.0 ppm., the percent of conversion to products could
be determined easily from integration ratios of peak areas.
Unlike many other types of ionic liquids,13 the mPnXY2
are miscible with low polarity organic solvents such as
hexanes and ethyl ether. Although this enhanced solubility
opens some interesting possibilities for applications and
mechanistic studies, it closes another, the ability to separate
the RTILs from the reaction products by extracting with a
low polarity solvent. Also, the water/methanol/hexane mix-
ture used previously to separate products from the RTIL
14P6Br22 did not function here. However, products could be
separated easily from the ionic liquids by chromatography
through silica gel, using hexane or 6:1 hexane:ethyl acetate as
the eluent.
The data, reported in Table S1 (Supporting Information)
show that reactions involving addition to a double bond were
highly stereoselective and anti: With 14P6Br3, ratios of anti:
syn addition products were as high as 99:1 and only trans-1,2-
dibromocyclohexane (8) was found as the product from
cyclohexene (1). Halogenations of cis-stilbene (2) were more
rapid than those of its trans-isomer (3), and both gave anti
addition products, the d,l (10) and meso (12) stereoisomers
for dibromo products and threo (9) and erythro (11) stereo-
isomers for bromo-chloro products, respectively. These re-
sults are indicative of a similar reaction pathway for most of
the alkene substrates studied, probably involving a bridged
intermediate.23 By contrast, bromination of 2 by the RTIL,
1P10Br3, using two and four layer reactors,14a yielded the
syn-addition product preferentially (but the expected anti-
addition product of 3). When reaction of 2 was conducted
with 14P6Br3 in a two-layer fashion, the syn-addition pro-
duct was formed preferentially again, although only a 26%
product yield was isolated.
Reactions involving ClBr2- and BrCl2- again led to anti-
addition products, but with mixtures of bromo-chloro and
dibromo products whose ratio depended upon the RTIL and
the substrate. Only very low yields of dichloro products were
formed and only with 14P6BrCl2 as the reagent and cis-
stilbene (2) or trans-stilbene (3) as the substrate (entries 31
and 37, Table S1, Supporting Information). Reactions with
-
BrCl2 yielded predominantly bromo-chloro products and
very small amounts of dibromo products. The greater than
statistical incorporation of bromine may be a consequence of
its higher polarizability and nucleophilicity as well as its
enhanced electrophilicity when linked to the more electro-
negative chlorine atoms.
By contrast, reaction of cyclohexene (1) with 1P10ClBr2
gave a 6:4 mixture of bromo-chloro (7) and dibromo (8)
products, both of which were trans. The addition was very
fast;the orange color of the trihalide (when not in excess)
occurred in a few minutes and an NMR spectrum recorded
50 min after the initiation of reactiom indicated a complete
loss of substrate. In an attempt to slow the reaction in order
to observe whether the ratios of bromo-chloro and dibromo
products depend on the extent of substrate conversion (i.e.,
the ratio between 1P10ClBr2 and its monohalide analogues),
1.2 mol equiv of 1P10ClBr2 was reacted as one phase and
substrate 1 dissolved in hexadecane as another.14a This
method limits the reaction zone effectively to the interface
between the two phases. After 10 days, ca. 95% of 1 had
disappeared (Figure 1). Analyses of aliquots of the hexade-
cane layer removed periodically during the intervening per-
iod indicated that the 7:8 ratio remained ∼6:4, the ratio
observed in the one-phase reaction system, throughout
(Figure 1 and entries 3-8, Table S1, Supporting Informa-
tion). Because there must be some equilibration among the
-
possible trihalide species in this experiment (N.B., ClBr2
,
Br3-, and ClBr2-),6 we expected a much larger variation in
the 7:8 ratio as a function of substrate 1 conversion; the
results reported in entries 8 and 9 in Table S1 (Supporting
(19) Grodkowski, J.; Neta, P. J. Phys. Chem. A 2002, 106, 11130–11134.
(20) The chemical shifts of the protons on the methyl and methylene
groups attached to phosphorus in these ionic liquids are very slightly
concentration dependent (Supporting Information). The concentration de-
pendence is much less than the shift differences observed between the mono-
and trihalide salts.
€
(21) Schonberg, A. Preparative Organic Photochemistry; Springer-Verlag:
New York, 1968; Chapter 37.
(22) McNulty, J.; Nair, J. J.; Robertson, A. Org. Lett. 2007, 9, 4575–4578.
(23) (a) Chretien, J. R.; Coudert, J.-D.; Ruasse, M.-F. J. Org. Chem. 1993,
58, 1917–1921. (b) Bellucci, G.; Chiappe, C.; Lo Moro, G. J. Org. Chem.
1997, 62, 3176–3182.
9030 J. Org. Chem. Vol. 74, No. 23, 2009

Cristiano et al.
JOCArticle
trihalide anions,6 but they will require additional experimen-
tation to be understood fully.
Furthermore, addition of one chlorine and one bromine
-
-
atom from ClBr2 and BrCl2 to terminal double bonds
(N.B., styrene (4a) and 1-octene (4b)) resulted in different
ratios of regioisomers (as well as different bromo-chloro:
dibromo product ratios). Assuming that the initially formed
intermediate is a bromonium ion or β-bromocarbocation,25
-
4a undergoes only Markovnikov addition with ClBr2
yielding 13a and the dibromo product. When the trihalide
,
-
employed is BrCl2 from 1P10BrCl2, none of the dibromo
product is detected. However, reaction of 4b with 14P6ClBr2
gave the dibromo product in highest yield and almost equal
yields of the Markovnikov (13b) and anti-Markovnikov
(15b) bromo-chloro products. As expected, addition of more
chloride ions, in the form of 14P6Cl, increased the relative
yields of both 13b and 15b and increased slightly the ratio of
anti-Markovnikov (15b) to Markovnikov (13b) addition
products (entry 46 in Table S1, Supporting Information).
Somewhat surprisingly, reaction of 4b with 14P6BrCl2 gave
only bromo-chloro products, a 45:55 13b:15b mixture (entry
47 in Table S1, Supporting Information).
In general, reactions of diphenylacetylene (5) with the
mPnXY2 were much slower and gave lower yields than those
with the alkenes. Alkynes are known to be less reactive than
alkenes toward electrophilic attack, mainly because the
intermediate vinyl cations are less stable than analogous
saturated cations. Both the dibromo E-isomer 17 and the
tetrabrominated 19 were formed upon reaction of 5 with
14P6Br3 (entry 49 in Table S1, Supporting Information).
Reactions of 5 (entry 50 and 51 in Table S1, Supporting
Information) gave mixtures of bromo-chloro (16) and dibro-
mo (17) products when reacted with 14P6ClBr2 (entry 50 in
Table S1, Supporting Information) and only 16 when reac-
ted with 14P6BrCl2 (entry 51 in Table S1, Supporting
Information); neither tetrabrominated (19) nor mixed tetra-
halo products, such as 18, were observed.
FIGURE 1. Percent conversion of 1 in hexadecane by 1P10ClBr2
(9) and relative product yields for 7 (ꢀ) and 8 (O) as a function of
reaction time. Note that the data points at shorter times have larger
errors due to the smaller amounts of products present.
Information) do suggest that trihalide equilibration may
play some role in the relative product yields (vide infra). In
previous work, we demonstrated that no free bromine could
be detected in the hexadecane layer when 1P10Br3 was the
brominating reagent,14a and a similar experiment here with
1P10ClBr2 as well as with polystyrene beads containing
appended methyltriphenylphosphonium tribromide groups
in dichloromethane14g again gave no evidence for the pre-
sence of free halogen molecules in the alkane layer. On
these bases, we believe that dihalogen molecules are not
responsible for the actual electrophilic addition reactions,
although we cannot eliminate the possibility that a very
small steady-state concentration is the active species at this
time.
As in this case, the relative yields of bromo-chloro pro-
ducts are always larger than those of the corresponding
dibromo products (Table S1, Supporting Information). This
result may reflect the greater strength of chloride than
bromide as a nucleophile in aprotic solvents and in the gas
phase.24 To examine this point further, experiments were
performed in the presence of added chloride ions in the form
of 14P6Cl. Consistent with the hypothesis above (as well as
another in which the larger concentration of chloride allows
it to compete more effectively to add to the product-deter-
mining intermediate, whether it be an alkene-dihalogen
π-complex6a or a halonium or halocarbenium ion), the
relative yields of the bromo-chloro products increased and
those of the dibromo products decreased (see entries 12, 13,
26, 27, 28, 35, 36, 41, and 46 in Table S1, Supporting
Information). Related to the latter explanation, we find that
somewhat different product ratios are obtained when the
molar equivalents of the trihalide salt are reduced. As an
example, cyclohexene (1) yields a ∼6:4 bromo-chloro:dibro-
mo product ratio in the presence of 1.2 equiv of 14P6ClBr2
and a 1:1 ratio when only 0.6 equiv of 14P6ClBr2 is present
(entries 8 and 9 in Table S1, Supporting Information). These
results are suggestive of some equilibration by different
Electrophilic substitutions on the aromatic ring of anisole (6)
by the mPnXY2 were also much slower and gave lower yields
than those with the alkenes. They were performed in the
presence of a base, NaHCO3, to capture the HX released in
the reaction. In the absence of base, no product was observed
-
with 14P6Br3; the phosphonium salts with ClBr2- and BrCl2
anions were somewhat more reactive. However, only bromine-
substituted products were observed, regardless of the composi-
tion of the trihalide employed. For example, 14P6BrCl2 con-
verted 97% of 6 to a 92:8 mixture of 4-bromoanisole:2,
4-dibromoanisole, but only after 48 h. We attribute the pro-
pensity for bromination to the higher polarizability of bromine
than chlorine and the greater electronegativity of chlorine than
bromine: both factors tend to enhance the partial positive
charge on bromine in the mixed trihalides, and the result here
is consistent with our hypothesis that bromonium ions are the
predominantly (if not exclusively) formed intermediates during
the additions to alkenes (vide ante).
Conclusions
The use of two RTILs, tetralkylphosphonium cations
with either tribromide or mixed bromo-chloro trihalides, as
(24) In the gas phase as well as in aprotic solvents, nucleophilicity follows
the trend F- > Cl- > Br- > I-. Olmstead, W. N.; Brauman, J. I. J. Am.
Chem. Soc. 1977, 99, 4219–4228.
(25) Braddock, D. C.; Hermitage, S. A.; Kwok, L.; Pouwer, R.; Redmond,
J. M.; White, A. J. P. Chem. Commun. 2009, 1082–1084.
J. Org. Chem. Vol. 74, No. 23, 2009 9031

Cristiano et al.
JOCArticle
halogenating reagents for alkenes, an alkyne, and an aro-
matic molecule has been explored. On the basis of the results
obtained, these reagents offer an easily synthesized and easily
handled alternative to many of the other halogenation
reagents developed previously. The yields with alkenes,
especially, are very high and the products are easily separated
from the reagents and their byproducts. The reactions
proceed with high stereo- and regioselectivities. Because
these reagents are liquids at room temperature, they func-
tion well in solvent-free conditions. Additional studies are
being conducted to characterize further the properties of
these RTILs and to obtain more details about the mecha-
nism of their halogenation reactions. Also, structural modi-
fications to the phosphonium and trihalide parts will be
explored.
Trihexyltetradecylphosphonium Bromodichloride (14P6BrCl2).
The synthesis followed the same procedure described for
1P10BrCl2 with 14P6Br (2.16 g, 3.83 mmol) and chlorine gas:
1
2.45 g (100%) of a viscous yellow liquid. H NMR (CDCl3) δ
2.30-2.23 (m, 8 H), 1.59-1.53 (m, 16 H), 1.35-1.26 (m, 32 H),
0.93-0.86 (m, 12 H). 31P NMR (CDCl3) δ 33.29 ppm. Elemental
Anal. Calcd for C32H68BrCl2P: C, 60.56; H, 10.80. Found: C,
60.82; H, 11.12.
Representative Procedures for Halogenation Reactions. With
Cyclohexene (1). Two-Phase Bromo-Chlorination (entry 2 in
Table S1, Supporting Information). A solution of cyclohexene
(1) (0.047 g, 0.58 mmol) in 2 mL of hexadecane was added
directly to 1P10ClBr2 (0.35 g, 0.70 mmol) in a 14 mm
diameter screw cap vial. The two-phase system was kept in
the dark and without stirring. Aliquots were removed per-
iodically and analyzed by GC to follow the course of the
reaction.
Reaction with Neat Ionic Liquid (entry 1 in Table S1, Support-
ing Information). Cyclohexene (1) (0.022 g, 0.27 mmol) was
added to a vial containing neat 14P6Br3 (0.24 g, 0.33 mmol)
and the mixture was kept at room temperature in the dark
for 3 h. The product, trans-1,2-dibromocyclohexane (0.059 g
(91%)), was isolated by column chromatography (silica gel;
hexanes as eluant) and identified spectroscopically, using lite-
rature assignments for comparison; see the Supporting Infor-
mation.
With cis-stilbene (2). Reaction with Ionic Liquid in Solution
(entry 18 in Table S1, Supporting Information). cis-Stilbene (2)
(0.0418 g, 0.232 mmol) and 14P6ClBr2 (0.190 g, 0.279 mmol)
were dissolved in CDCl3 (0.70 g, 0.46 mL). The reaction
mixture was kept in the dark at room temperature in a closed
vessel. TLC (silica gel; 6:1 hexanes:ethyl acetate as eluant) and
1H NMR analyses indicated complete reaction after 1 h (see
the Supporting Information). The products (0.064 g, 88%)
were isolated via column chromatography (silica gel; 6:1
hexanes:ethyl acetate as eluant) and identified as a ca. 6:4
mixture of threo-1-bromo2-chloro-1,2-diphenylethane and d,
l-1,2-dibromo-1,2-diphenylethane by their spectra, using
literature assignments for comparison; see the Supporting
Information.
With trans-Stilbene (3). Reaction with Ionic Liquid Leading to
Precipitation of the Product (entry 35 in Table S1, Supporting
Information). trans-Stilbene (3) (0.043 g, 0.24 mmol), 14P6Cl
(0.148 g, 0.286 mmol), and 14P6ClBr2 (0.194 g, 0.286 mmol)
were dissolved in CDCl3 (0.71 g, 0.48 mL). The solution
was kept in the dark for 12 h at room temperature in a closed
vessel. Precipitation of a white solid was noted after 2 h of
reaction. The solid was filtered and washed with cool CHCl3 to
give 0.057 g (77% yield), identified spectroscopically using
literature assignments for comparison (see the Supporting
Information) to be a ca. 3:1 mixture of erythro-1-bromo-2-
chloro-1,2-diphenylethane and meso-1,2-dibromo-1,2-dipheny-
lethane.
With Styrene (4a). Reaction in Solution with Additional
Chloride Ions (entry 41 in Table S1, Supporting Information).
Styrene (4a) (0.027 g, 0.26 mmol), 14P6Cl (0.136 g, 0.262 mmol),
and 14P6ClBr2 (0.178 g, 0.262 mmol) were dissolved in CDCl3
(0.79 g, 0.52 mL). The reaction mixture was kept in the dark at
room temperature in a closed flask for 1 h, at which time,
TLC (silica gel, eluant hexanes:ethyl acetate 6:1) and 1H
NMR analyses indicated the complete loss of styrene. The
products (0.032 g, 52%) isolated via column chromatography
(silica gel; 6:1 hexanes:ethyl acetate as eluant) were identified
spectroscopically, using literature assignments for comparison
(see the Supporting Information), as a ca. 2:1 mixture of
2-bromo-1-chloro-1-phenylethane and 1,2-dibromo-1-phenyl-
ethane.
Experimental Section
Synthesis of Phosphonium Trihalides (mPnXY2). Tridecyl-
methylphosphonium Dibromochloride (1P10ClBr2). In a 50 mL
round-bottomed flask immersed in a room temperature water
bath, bromine (0.396 g, 2.48 mmol) was added to solid 1P10Cl
(1.25 g, 2.48 mmol). The reaction mixture is stirred for 30 min in
the dark in the closed vessel to afford 1.65 g (100%) of a red-
orange viscous liquid. 1H NMR (CDCl3) δ 2.34-2.27 (m, 6 H),
2.00 (d, J=12.8 Hz, 3 H), 1.60-1.49 (m, 12 H), 1.35-1.27 (m,
36 H), 0.88 (t, J=6.4 Hz, 9 H) ppm. 13C NMR (CDCl3) δ 31.84,
30.76, 30.62, 29.47, 29.30, 29.25, 28.96, 22.65, 21.83, 21.78,
20.88, 20.40, 14.09 ppm. 31P NMR (CDCl3) δ 31.95 ppm.
Elemental Anal. Calcd for C31H66Br2ClP: C, 55.98; H, 10.00.
Found: C, 56.20; H, 10.39.
Tridecylmethylphosphonium Bromodichloride (1P10BrCl2).
1P10Br (0.51 g, 0.94 mmol) was transferred to a 50 mL
three-necked round-bottomed flask equipped with an inlet
for chlorine gas and an outlet containing a drying tube. The
flask was cooled in a dry ice bath and chlorine gas26 was
slowly released into it. The liquid chlorine slowly dissolved
the solid 1P10Br and the mixture was kept in the dark for 1 h.
After this period, it was warmed to room temperature,
releasing most of the excess chlorine, and a strong stream of
nitrogen was passed over the liquid to remove the remaining
excess chlorine: yield 0.58 g (99%) of a green-yellowish
viscous liquid. 1H NMR (CDCl3) δ 2.33-2.24 (m, 6 H), 1.97
(d, J = 13.2 Hz, 3 H), 1.61-1.50 (m, 12 H), 1.27 (s broad, 36
H), 0.88 (t, J = 6.4 Hz, 9 H) ppm. 31P NMR (CDCl3) δ 31.89
ppm. Elemental Anal. Calcd for C31H66BrCl2P: C, 59.99; H,
10.72. Found: C, 60.24; H, 10.98.
Trihexyltetradecylphosphonium Tribromide (14P6Br3). Bro-
mine (0.614 g, 3.84 mmol) was added to neat 14P6Br (2.16 g,
3.84 mmol) in a 25 mL round-bottomed flask. The flask was
sealed with a rubber stopper and the red liquid mixture was
stirred at room temperature for 6 h to give 2.72 g (98%) of a red
viscous liquid. 1H NMR (CDCl3) δ 2.28 (m, 8 H), 1.57 (m, 16 H),
1.39-1.26 (m, 32 H), 1.02 (m, 12 H). 31P NMR (CDCl3) δ 33.38.
Elemental Anal. Calcd for C32H68Br3P: C, 53.12; H, 9.47.
Found: C, 52.90; H, 10.04.
Trihexyltetradecylphosphonium Dibromochloride (14P6ClBr2).
The synthesis followed the same procedure described for 1P-
10ClBr2 with 14P6Cl (2.64 g, 5.09 mmol) and bromine (0.814 g,
5.09 mmol): 3.45 g (100%) of an orange viscous liquid. 1H NMR
(CDCl3) δ 2.23-2.16 (m, 8 H), 1.53-1.45 (m, 16 H), 1.27-1.17
(m, 32 H), 0.84-0.77 (m, 12 H). 31P NMR (CDCl3) δ 33.25 ppm.
Elemental Anal. Calcd for C32H68Br2ClP: C, 56.59; H, 10.09.
Found: C, 56.90; H, 10.84.
(26) Vogel, A. I. A textbook of practical organic chemistry, 3th ed.;
Longman: Thetford, UK, 1974; p 183.
9032 J. Org. Chem. Vol. 74, No. 23, 2009

Cristiano et al.
JOCArticle
Acknowledgment. We thank the U.S. National Science
~
were made as well as for his sage advice, and to Dr. Shibu
Abraham for performing some of the analyses.
Foundation for its support of this research and Coordenac-ao
de Aperfeic-oamento de Pessoal de Nıvel Superior (CAPES)
´
Supporting Information Available: Thermal characteriza-
tions of ionic liquids, their NMR spectra as well as those of
reaction products, and selected gas chromatograms of reaction
mixtures. This material is available free of charge via the
Internet at http://pubs.acs.org.
of Brazil for a fellowship to R.C. We are very grateful to
Dr. Al Robertson of Cytec Industries for supplying the
trihexyltetradecylphosphonium bromide and the tridecyl-
phosphine from which the tridecylmethylphosphonium salts
J. Org. Chem. Vol. 74, No. 23, 2009 9033