Ultrasound promoted Wurtz coupling of alkyl bromides and dibromides

Article abstract of DOI:10.1016/j.ultsonch.2011.06.002

Sonochemically enhanced Wurtz coupling using lithium metal has been investigated for a number of isomeric alkyl bromides under a variety conditions. The products result from direct coupling of short lived radicals formed at the metal surface rather than the secondary radicals which can be formed during coupling of aromatic halides and thus give rise to a single major product. Coupling has been extended to dibrominated aryl and alkyl compounds as well as showing that aryl-alkyl coupling is possible. Dibrominated alkyls were found to give low molecular weight oligomers although no reaction occurred for 1,2-isomers. The growth of oligomers in THF may be solubility limited. A simple model is proposed to explain these findings.

Full text of DOI:10.1016/j.ultsonch.2011.06.002

Ultrasonics Sonochemistry 19 (2012) 5–8
Contents lists available at ScienceDirect
Ultrasonics Sonochemistry
journal homepage: www.elsevier.com/locate/ultsonch
Short Communication
Ultrasound promoted Wurtz coupling of alkyl bromides and dibromides
⇑
Daniel VandenBurg, Gareth J. Price
Department of Chemistry, University of Bath, BATH BA2 7AY, United Kingdom
a r t i c l e i n f o
a b s t r a c t
Article history:
Sonochemically enhanced Wurtz coupling using lithium metal has been investigated for a number of iso-
meric alkyl bromides under a variety conditions. The products result from direct coupling of short lived
radicals formed at the metal surface rather than the secondary radicals which can be formed during cou-
pling of aromatic halides and thus give rise to a single major product. Coupling has been extended to
dibrominated aryl and alkyl compounds as well as showing that aryl–alkyl coupling is possible. Dibromi-
nated alkyls were found to give low molecular weight oligomers although no reaction occurred for 1,2-
isomers. The growth of oligomers in THF may be solubility limited. A simple model is proposed to explain
these findings.
Received 28 April 2011
Received in revised form 26 May 2011
Accepted 2 June 2011
Available online 16 June 2011
Keywords:
Ultrasound
Sonochemistry
Wurtz coupling
Lithium
Ó 2011 Elsevier B.V. All rights reserved.
Bromoalkanes
1. Introduction
radical scavenger, indicating that the coupling reaction involved
a pathway involving free radical intermediates.
High intensity ultrasound is a useful tool for the acceleration of
many synthetically useful organic and organometallic reactions
[1–3]. A wide range of processes can be promoted and commonly
reported effects include significant increases in rates and yields
as well as, in some cases, an apparent switch in products due to
the promotion of one possible mechanism in preference to others
[4,5].
Among the heterogeneous reactions promoted by ultrasound is
Wurtz-type coupling over reactive metals such as lithium (Scheme
1). Han and Boudjouk were the first to apply ultrasound to this
reaction and reported the coupling between chloro-, bromo- and
iodobenzene to form biphenyl in good yields [6] while the coupling
of tolyl systems gave lower yields.
In related work, Osborne et al. [7] reported that coupling
bromopyridines gave mixtures of isomeric bipyridyls and that
dehalogenation to form pyridine was a major side reaction, indicat-
ing that the mechanism for aryl Wurtz coupling was more complex
than first assumed. This was confirmed by Price and Clifton [8]
who found that reaction of bromotoluenes also gave a mixture of
products. Coupling of 2-bromotoluene yielded a mixture of 2,2-,
2,3- and 2,4-dimethyl biphenyl. Similarly, coupling of the 3- and
4-substituted isomers also gave mixtures in which one ring re-
tained the methyl group in the original position of substitution
while the other ring contained each of the three possible positions.
They also found that these reactions were inhibited by a free
In addition to organic synthesis, ultrasound has also been
applied to the preparation of polymers with controlled structures
[9]. Among these, Wurtz type coupling reactions using sodium of
dialkyl- and diaryl silanes has been used to prepare long chain
poly(organosilanes) [10–12], polymers with all silicon backbones,
and the reaction has also been applied to polymers made with
tin backbones [13]. However, it has not previously been applied
to carbon compounds, presumably due to their lower reactivity.
Coupling dibromoalkanes containing branched alkyl groups offers
the potential for forming polyolefins with controlled short-chain
branches although further detail on the products and reaction
mechanism is needed in order to evaluate the potential of this
reaction. As far as we are aware, to date the Wurtz type coupling
of halogenoalkanes has only been studied for 1-chloropropane
[6] and 1-bromobutane [14]. We therefore present here an investi-
gation into the coupling of bromopentane isomers. In addition we
have extended the ultrasonic Wurtz coupling reaction to dibromo
aryl and alkyl systems to investigate the possibility of the forma-
tion of polymers with controlled branched structures.
2. Experimental
2.1. Materials
All reagents were used as received from Sigma–Aldrich (UK) or
Alfa Aesar (UK). Solvents used were of HPLC grade and used with-
out further purification. Gases were obtained from BOC and were
used without further purification.
⇑
Corresponding author. Tel.: +44 1225 386504.
E-mail address: g.j.price@bath.ac.uk (G.J. Price).
1350-4177/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.ultsonch.2011.06.002

6
D. VandenBurg, G.J. Price / Ultrasonics Sonochemistry 19 (2012) 5–8
Scheme 1. Sonochemical Wurtz coupling.
2.2. Instrumentation
Gas chromatographic analysis was carried out using an Agilent
6890 N Network gas chromatograph fitted with an HP-5 column.
For bromobenzene analysis the initial temperature for the oven
was 70 °C. This was held for 1 min and then increased at
16 °C minꢀ1 to 150 °C. For analysis of bromopentanes, the initial
oven temperature was 40 °C held for 5 min then increased at
16 °C minꢀ1 to 150 °C. 13C NMR spectra were recorded in CDCl3
solution using a 300 MHz Bruker AV NMR spectrometer. 1H NMR
were recorded using a 250 MHz Bruker AV NMR spectrometer.
Chemical shifts are reported with reference to TMS. Gel perme-
ation chromatography was carried out using a Polymer Laborato-
Fig. 1. Consumption of bromobenzene and bromopentanes during sonochemical
Wurtz coupling using lithium.
The product was confirmed to be n-decane by comparison via GC
and 1H NMR with an authentic sample. The yield was 83% by 1H
NMR. No significant amounts of other products were detected via
GC or NMR. The consumption of 2-bromopentane was also faster
than bromobenzene although somewhat slower than the 1-isomer.
The Wurtz coupling of bromoaryls is known to follow a mecha-
nism involving radicals [6]. To investigate this for bromoalkyls, the
reaction was repeated in the presence of excess DPPH, a free radi-
cal scavenger. After 2 h of sonication, little of the 1-bromopentane
had reacted (from GC analysis) and no coupling products were ob-
served, confirming that radical intermediates were also involved in
the reaction of alkyl bromides.
ries PL-GPC-50 Plus equipped with a PLGel 5 lm ‘‘mixed C’’
column. THF was used as the eluting solvent at a flow rate of
1 cm3 minꢀ1 at 30 °C.
2.3. Sonication
Reactions were carried out in a Kerry Ultrasound cleaning bath
operating at 25 kHz delivering 0.4 W of power into the reaction,
measured calorimetrically in the usual manner [15].
2.4. Coupling of bromobenzene and bromopentanes
The coupling of 1-bromopentane was then studied under vary-
ing atmospheres. Fig. 2 shows that the rate of reaction was similar
under Argon and N2. The rate of consumption was somewhat
slower under air but was essentially complete after 90 min. The
rate was slower under O2 and went to only 73% completion. Under
CO2 the reaction stopped after 15 min at a conversion of only 21%.
The major product in each reaction was n-decane and for the reac-
tion with CO2 no products arising from CO2 insertion were seen.
One explanation for these differences in rate could be the differ-
ence in the polytropic ratio of the gases [15]; a bubble filled with a
gas of high polytropic ratio will generate higher temperatures and
pressures on implosion. Carbon dioxide has the lowest polytropic
ratio of the other gases used and produced the lowest yield and
rate of reaction. However, given the nature of the Wurtz reaction
(see below), the effect of the gas is more likely to be explained in
terms of chemical reactivity. Oxygen will act as a free radical scav-
enger and so any radical based reaction will be partially quenched
In a 100 cm3 conical flask with a ground glass neck joint,
10 mmol of brominated compound was dissolved in 25 cm3 THF.
To this was added 200 lL of dihexylether as an internal standard
for GC analysis. The solution was purged with nitrogen for
10 min after which time 12 mmol (1.1 equiv., 83 mg) of lithium
pellets were added and the flask placed in the sonic bath so that
the disturbance of the solution’s surface by ultrasonic action was
maximised. The samples were then sonicated while a slow flow
of nitrogen was maintained. The solutions were sampled regularly
taking 0.5 cm3 aliquots which were diluted to 1 cm3 with THF
before injection into the GC.
2.5. Coupling of dibrominated compounds
The same method was used except that 2.2 equiv. of lithium
were used and no dihexylether was added. After sonication the res-
idues were poured into 200 cm3 of distilled water and the product
extracted with two 100 cm3 portions of diethyl ether. The organic
layers were combined and dried using MgSO4 before removing the
solvent under vacuum.
3. Results and discussion
The reaction time for the sonochemical Wurtz coupling of bro-
mobenzene has previously been reported as ranging between 2
and 10 h [6–8,14]. Under the conditions used in this work, the
reaction was essentially complete with respect to bromobenzene
consumption after 2 h, as shown in Fig. 1. As previously reported,
the only product recovered was biphenyl with a yield of approxi-
mately 70%. The remaining product is assumed to be benzene,
resulting from a competitive dehalogenation reaction.
1-Bromopentane was chosen for an initial study the reaction of
alkyl bromides. The consumption of 1-bromopentane is also shown
in Fig. 1 and indicates a significantly faster reaction than with
bromobenzene with complete consumption after around 40 min.
Fig. 2. Effect of gaseous atmosphere on sonochemical Wurtz coupling of 1-
bromopentane.

D. VandenBurg, G.J. Price / Ultrasonics Sonochemistry 19 (2012) 5–8
7
when run under oxygen or air. Carbon dioxide is known to react
with lithium and its salts so much of the lithium present in our
system could have reacted with carbon dioxide deactivating it to
coupling [16,17].
Further experiments were carried out while varying the stoichi-
ometric equivalents of lithium. As shown in Fig. 3, using two
equivalents of lithium instead of one had little effect on the rate of
bromopentane consumption. When 0.75 equiv. was used, the
reaction was 63% complete and no further consumption of bromo-
pentane took place after 35 min; the corresponding values were
28% and 30 min, respectively when 0.5 equiv. was used. This indi-
cates that the availabilityof lithium is not rate limiting as long as suf-
ficient is available.
In order to determine whether the formation of a single product
was a feature of alkyl systems (in contrast to aromatics), coupling
reactions were conducted using 1- and 2-bromopentane, 1-bromo-
3-methylbutane, 2-bromo-2-methylbutane, 1-bromo-2,2-dimeth-
ylpropane. In each case, only a single product was formed resulting
from the coupling of both fragments at the position of the bromine
atom (Scheme 2).
The formation of a single product is in complete contrast to the
bromoaryls studied by Osborne et al. [7] and by Price and Clifton
[8] (Scheme 3) where mixtures of isomers were produced. How-
ever, in each case, the reactions proceeded via a radical intermedi-
ate. Aryl radicals are generally more stable and hence less reactive
than alkyl radicals and we postulate that the intermediates pro-
duced during coupling of aryl bromides are sufficiently long-lived
for rearrangement to take place. In contrast, in the more reactive
alkyl radicals, no rearrangement can take place prior to coupling
and so only a single product is seen.
Another significant difference between these systems was that
coupling of bromobenzene was found to follow second order kinet-
ics (rate constant, k = 0.01 dm3 mmolꢀ1 minꢀ1) while that of the
bromoalkanes was found to follow first order kinetics (k = 0.07
minꢀ1 and k = 0.04 minꢀ1 for the 1- and 2-isomers, respectively).
This adds further evidence to our suggestion of more reactive alkyl
radical intermediates, the argument being as follows. The generally
accepted mechanism [18] of the reaction initially involves an elec-
tron transferfrom the metalto producelithium bromide and an alkyl
radical followed by a second electron transfer to form an alkyl anion.
Scheme 2. Coupling of alkyl bromides. Only a single isomer is observed.
RꢀLiþ þ R-Br ! R-R þ LiþBrꢀ
k3
The overall kinetics of the process will therefore depend on the
relative rate constants for the steps. Where k2 is relatively faster
(i.e. the radical is short lived), k3 will be the rate limiting step
and the overall reaction will be second order; conversely, if k2 is
the slow step then the observed reaction order will be one (pseu-
do-first).
In a related experiment, a ‘‘cross coupling’’ between bromoben-
zene and bromopentane was attempted. Analysis of the NMR spec-
tra suggested cross reaction had occurred (aromatic proton to alkyl
proton coupling was seen in a COSY experiment to differentiate
from a mixture of pentane and benzene) but the recovered yield
of pentyl benzene was only approximately 10%, the majority prod-
ucts being those from the homocoupling.
Since mono-brominated aryl and alkyl compounds undergo effi-
cient Wurtz coupling, the possibility of forming polymeric materi-
als from dibrominated compounds (Scheme 4) was investigated
using 1,4-dibromobenzene, 1,4-dibromobutane and 1,6-
dibromohexane.
The coupling of 1,4-dibromobenzene led to a sticky brown ‘‘tof-
fee-like’’ material which was found by GPC to have an average rel-
ative molar mass of 716 corresponding to an oligomer with around
nine repeat units. NMR suggested that a short chain polyphenyl-
ene-like material had been formed although the coupling was
not exclusively at the 1,4-positions, as might be expected from
the results on monosubstituted aryls. Similarly 1,6-dibromohexane
and 1,4-dibromobutane coupled under the same conditions
yielded white solids which had molar masses of 516 and 464 mass
units, respectively. The 1H and 13C NMR of both samples were iden-
tical and consistent with the spectra of long chain alkanes or poly-
ethylene with no significant branching or alkyl groups due to
rearrangement.
Coupling of a series of 1,3- and 1,2-dibrominated alkanes shown
in Scheme 5 was attempted. Significantly, the 1,2-isomers did not
react even after prolonged sonication times whereas the 1,3-iso-
mers reacted but formed a complex mixture of products containing
linear and substituted oligomers. Only the isomers with terminal
substituents gave clean products.
R-Br þ Li ! RÅ þ LiþBrꢀ k1
RÅ þ Li ! RꢀLiþ
k2
The coupled product then arises from displacement of a bro-
mide ion from a second molecule of alkyl bromide, usually in a
SN2 reaction.
The short chain lengths of the oligomers produced by these
reactions was disappointing but may well be a consequence of
the rapidly decreasing solubility as the alkane chain length in-
creases. In order to counter the potential insolubility of long chain
polyalkanes in THF at 25 °C coupling of 1,6-dibromohexane in dec-
alin at 70 °C (polyethylene being soluble under these conditions)
was attempted. However no reaction occurred and only the start-
ing materials were recovered. It may be the solvent vapour pres-
sure was too high under these conditions to allow sufficient
cavitation to influence reaction at the metal surface. From the par-
allel with silicon-based materials [19,20], it may also be that a
stronger reducing agent such as sodium is needed, although
poly(organosilanes) are often more soluble than polyethylene
and so may form more easily.
The results can be rationalised by considering the mechanism
outlined above. The initial electron transfer step takes place at or
near to the metal surface. The metal surface would rapidly become
covered with a layer of lithium bromide, preventing access of
Fig. 3. Effect of lithium quantity on sonochemical Wurtz coupling of 1-
bromopentane.

8
D. VandenBurg, G.J. Price / Ultrasonics Sonochemistry 19 (2012) 5–8
Scheme 3. Sonochemical coupling of bromotoluenes [8]. Each of the possible isomers is formed in one ring.
No migration of the radical position occurs, in contrast to the find-
ings in aromatic systems. The process can also be used with dibromi-
nated compounds although the bromines must be separated by at
least two carbon atoms.
References
[1] G. Cravotto, P. Cintas, Forcing and controlling chemical reactions with
ultrasound, Angew. Chem. Int. Ed. 46 (2007) 5476.
[2] J.-L. Luche, Synthetic Organic Sonochemistry, Plenum Press, New York, 1998.
[3] M. Ashokkumar, T. Mason, ‘‘Sonochemistry’’ in Kirk-Othmer Encylcopedia of
Chemical Technology, John Wiley and Sons, 2007.
[4] T. Ando, P. Bauchat, A. Foucaud, M. Fujita, T. Kimura, H. Sohmiya, Sonochemical
switching from ionic to radical pathways in the reactions of styrene and trans-
beta-methylstyrene with lead-tetraacetate, Tetrahedron Lett. 32 (44) (1991)
6379.
[5] M.J. Dickens, J.-L. Luche, Further evidence for the effect of ultrasonic-waves on
Scheme 4. Potential coupling of alkyl dibromides to give poly(alkene)s.
electron-transfer processes
– the case of the Kornblum–Russell reaction,
Tetrahedron Lett. 32 (1991) 4709.
[6] B.H. Han, P. Boudjouk, Organic sonochemistry – ultrasound-promoted coupling
of organic halides in the presence of lithium wire, Tetrahedron Lett. 22 (1981)
2757.
[7] A.G. Osborne, K.J. Glass, M.L. Staley, Ultrasound-promoted coupling of
heteroaryl halides in the presence of lithium wire. Novel formation of
isomeric bipyridines in a Wurtz-type reaction, Tetrahedron Lett. 30 (1989)
3567.
[8] G.J. Price, A.A. Clifton, A re-examination of the sonochemical coupling of
bromoaryls, Tetrahedron Lett. 32 (1991) 7133.
[9] G.J. Price, Polymer Sonochemistry: Controlling the Structure of Macromolecules,
in: L.A. Crum, T.J. Mason, J.L. Reisse, K.S. Suslick (Eds.), Sonochemistry and
Sonoluminescence, NATO ASI Series 524, Kluwer, Dordecht, 1998, pp. 321–344.
[10] G.J. Price, A.M. Patel, The application of ultrasound to the synthesis of
poly(organosilanes), Eur. Polym. J. 32 (1996) 1289.
Scheme 5. Alkyl dibromides for potential sonochemical coupling.
[11] H.K. Kim, K. Matyjaszewski, Degradation of poly(methylphenylsilylene) and
poly(di-n-hexylsilylene), J. Polym. Sci. Polym. Chem. 31 (1993) 299.
[12] G.J. Price, The preparation of poly(organosilanes) using high intensity
ultrasound, Chem. Commun. (1992) 1209.
[13] N. Devylder, M.S. Hill, K.C. Molloy, G.J. Price, Wurtz synthesis of high molecular
weight poly(di-n-butyl stannane), Chem. Commun. 711 (1996).
[14] T.D. Lash, D. Berry, Promotion of organic reactions by ultrasound: coupling of
alkyl and aryl halides in the presence of lithium metal and ultrasound, J. Chem.
Ed. 62 (1985) 85.
[15] T. Leighton, The Acoustic Bubble, Academic Press, London, 1994.
[16] S.O.N. Lill, U. Kohn, E. Anders, Carbon dioxide fixation by lithium amides: DFT
studies on the reaction mechanism of the formation of lithium carbamates,
Eur. J. Org. Chem. (2004) 2868.
further alkyl halide. Using ultrasound ‘cleans’ the surface by
removing this passivating layer and hence allows the reaction to
proceed, explaining why ultrasound is useful in this reaction. This
process becomes more difficult as the alkyl halide becomes bulkier.
It is known [18] that the side reaction to form an alkene becomes
more prevalent with bulkier substituents all at the halogen con-
taining carbon due to radical stabilisation and the higher activation
energy for the second substitution reaction favouring an elimina-
tion pathway. In the 1,2-dibromo alkanes, there may not be suffi-
cient conformational flexibility for both bromine atoms to
approach the surface. Our results suggest that for coupling to occur
the bromines must be separated by at least one methylene group if
they are to interact effectively with the surface.
[17] K. Vyakaranam, J.B. Barbour, J. Michl, Li+-catalyzed radical polymerization of
simple terminal alkenes, J. Am. Chem. Soc. 128 (2006) 5610.
[18] M.B. Smith, J. March, March’s Advanced Organic Chemistry. Reactions,
Mechanisms, and Structure, fifth ed., John Wiley and Sons, New Jersey, 2007.
[19] R.G. Jones, W.K.C. Wong, S.J. Holder, Correlation of structure and molecular
weight distributions during the formation of poly(methylphenylsilylene) by
the Wurtz reductive-coupling reaction, Organometallics 38 (1988) 1633.
[20] G.J. Price, Synthesis and modification of silicon-containing polymers using
ultrasound, Polymer Int. 58 (2009) 290.
4. Conclusion
It has been shown that the Wurtz type coupling of bromoalkyls is
possible and follows a mechanism involving radical intermediates.