An efficient and reliable procedure for the preparation of highly reactive Rieke zinc

Article abstract of DOI:10.1002/adsc.201201077

Rieke zinc has a wide potential for applications in organic chemistry, notably for the synthesis of highly chemoselective organozinc reagents. However, due to the rather unreliable preparation method, leading to large batch to batch variations, its use has been rather limited. Rieke zinc is commonly prepared by the reduction of zinc chloride with lithium using a stoichiometric amount of naphthalene. In our hands, it was observed that the reaction outcome was highly dependent on the naphthalene source and purity grade. The presence of benzothiophene seems crucial to avoid coagulation of the zinc particles and the amount of benzothiophene has a large effect on the physical properties and the reactivity of the resulting zinc powder. Accordingly, highly reactive Rieke zinc was easily prepared from zinc chloride by adding an optimum amount (3 mol% with regard to ZnCl₂) of benzothiophene into the lithium naphthalenide solution (prepared in situ). The Rieke zinc obtained in this way was successfully employed in the so-called "Rieke method" to prepare regioregular poly(3-hexylthiophene) (P3HT). The novel approach enables preparation of Rieke zinc in a reproducible manner, which considerably facilitates its use in both small molecule and polymer synthesis.

Full text of DOI:10.1002/adsc.201201077

UPDATES
DOI: 10.1002/adsc.201201077
An Efficient and Reliable Procedure for the Preparation of
Highly Reactive Rieke Zinc
Suleyman Kudret,a Jan Dꢀ Haen,b Laurence Lutsen,c Dirk Vanderzande,a,c
and Wouter Maesa,*
a
Hasselt University, Institute for Materials Research (IMO), Design & Synthesis of Organic Semiconductors (DSOS),
Agoralaan 1 – Building D, B-3590 Diepenbeek, Belgium
Fax : (+32)-11-268-299; e-mail: wouter.maes@uhasselt.be
Hasselt University, Institute for Materials Research (IMO), Electrical and Physical Characterization (ELPHYC),
b
Universitaire Campus – Wetenschapspark 1, B-3590 Diepenbeek, Belgium
IMEC, IMOMEC Ass. Lab., Universitaire Campus - Wetenschapspark 1, B-3590 Diepenbeek, Belgium
c
Received: December 7, 2012; Published online: January 25, 2013
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201201077.
Abstract: Rieke zinc has a wide potential for appli-
cations in organic chemistry, notably for the synthe-
sis of highly chemoselective organozinc reagents.
However, due to the rather unreliable preparation
method, leading to large batch to batch variations,
its use has been rather limited. Rieke zinc is com-
monly prepared by the reduction of zinc chloride
with lithium using a stoichiometric amount of naph-
thalene. In our hands, it was observed that the reac-
tion outcome was highly dependent on the naphtha-
lene source and purity grade. The presence of ben-
zothiophene seems crucial to avoid coagulation of
the zinc particles and the amount of benzothio-
phene has a large effect on the physical properties
and the reactivity of the resulting zinc powder. Ac-
cordingly, highly reactive Rieke zinc was easily pre-
pared from zinc chloride by adding an optimum
amount (3 mol% with regard to ZnCl2) of benzo-
thiophene into the lithium naphthalenide solution
(prepared in situ). The Rieke zinc obtained in this
way was successfully employed in the so-called
“Rieke method” to prepare regioregular poly(3-
hexylthiophene) (P3HT). The novel approach ena-
bles preparation of Rieke zinc in a reproducible
manner, which considerably facilitates its use in
both small molecule and polymer synthesis.
and good functional group tolerance and stability.[1]
However, due to their lower reactivity towards
a broad range of electrophiles and moderate yields
compared to Grignard reagents, they have received
rather limited attention, mostly for Reformatsky[2]
and Simmons–Smith reactions.[3] The first organozinc
compound to be reported by Frankland in 1849 was
prepared from the reaction of metallic Zn with ethyl
iodide.[4] However, due to the lack of reactivity of the
Zn metal, this reaction was quite limited in its scope.
Several methods have been used to activate Zn, such
as washing with a hydrochloric acid solution[5] or
adding 1,2-dibromoethane to prevent passivation of
the zinc surface.[6] In spite of these methods, direct ox-
idative addition of the Zn metal was only successful
for relatively reactive alkyl iodides. Thus, metallic Zn
did not generally allow preparation of organozinc re-
agents directly from organic halides. In the early
1970s, Rieke et al. introduced a new method which
utilized potassium metal to reduce zinc salts, affording
highly reactive Zn metal powder.[7] This active form
of Zn was called “Rieke zinc (Zn*)” and displayed
excellent reactivity towards organic halides compared
with bulk Zn metal, Zn dust or metallic Zn activated
by entrainment methods. Later on, in 1991, the same
group reported on an alternative method to prepare
the active Zn species under milder and safer condi-
tions by reduction of the zinc salt with lithium in the
presence of naphthalene as an electron carrier, giving
rise to even more reactive Zn*.[8] Reduction of a zinc
cyanide salt by the same method resulted in Rieke
zinc which reacts at a synthetically useful rate with
alkyl or aryl chlorides at room temperature to pro-
duce the corresponding organozinc halide reagents.[9]
Keywords: benzothiophene; naphthalene; poly(3-
hexylthiophene) (P3HT); reproducibility; Rieke
zinc
Organozinc compounds are important synthetic inter- Liquid ammonia has also been used as an electron
mediates as they exhibit excellent chemoselectivity carrier to produce active Zn powder.[10] Since then,
Adv. Synth. Catal. 2013, 355, 569 – 575
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
569

Suleyman Kudret et al.
UPDATES
Rieke zinc has found application in numerous organic addition reactions. On the other hand, Rieke zinc pre-
syntheses, mostly for the preparation of organozinc pared by this procedure was successfully implemented
reagents. Compared to Grignard and organolithium in Refortmatsky reactions at room temperature or
reagents, organozinc derivatives tolerate the presence below, affording quantitative reaction yields.[21] Later
of a wide range of functional groups, such as ketones, on, a safer and milder procedure which uses Li and
nitriles, esters and other halides, making it a very a catalytic amount of naphthalene (10 mol%) was in-
useful synthetic tool.[11] Rieke zinc has also been used troduced.[8,22] Initially it was believed that the use of
as a reducing agent for common functional groups an additional electron carrier would complicate purifi-
(conjugated aldehydes, alkynes, esters and nitro com- cation of the final organic products, but actually the
pounds).[12]
electron carrier can easily be removed by washing the
In polymer synthesis, the “Rieke method” – refer- reactive Zn powder with fresh THF or DME. More-
ring to the synthesis of regioregular (fully head-to-tail over, the preparation in THF or DME using either
coupled) poly(3-hexylthiophene) (P3HT) (co)poly- a stochiometric amount of lithium naphthalenide[8] or
mers by reaction of 2,5-dibromothiophene monomers Li and a catalytic amount of naphthalene even led to
with Rieke zinc – is of high relevance, as the resulting active Zn of higher reactivity. Reduction of the zinc
semiconducting conjugated polymer materials are of salt with Li and naphthalene requires vigorous stirring
great importance in organic electronics, notably in or- for about 10 h to obtain the active Zn slurry, but also
ganic photovoltaics.[13,14] Side chain functionalized to prevent the reduced Zn to coat the Li surface and
poly(3-alkylthiophenes) were also prepared by the thereby inhibit the reduction. The most convenient
same protocol,[15] and the method was extended to method actually involves reduction of the zinc salt
a limited number of other polyarylenes such as with a stoichiometric amount of lithium naphthalen-
poly(p-phenylene)[16] and poly(dialkylcyclopentadi-
ACHTUNGTRENiNGNU de, shortening the reduction time and eliminating the
thiophene)[17]. Besides these notable exceptions, the coating problem. The procedure is carried out in two
Rieke polymerization protocol has been almost steps. In the first step, the lithium naphthalenide/THF
absent from literature. Thiophene-based homopoly- solution is prepared, which takes about 2 h. In the
mers are generally prepared by the Grignard meta- next step, the solution of the zinc salt in THF is trans-
thesis (GRIM) method,[14] although this procedure is ferred using a cannula into the lithium naphthalenide/
not compatible with all desired functional side chain THF solution over 10–15 min. When stirring is
patterns (e.g., esters).[15]
stopped, the very finely dispersed zinc particles settle
Upon closer inspection of the available literature, out, leaving a clear colorless supernatant above the
one notices that the applicability of Rieke Zn seems Zn powder. In some cases, the supernatant may
mainly limited by obstacles in its preparation, which appear more or less green due to the presence of
later on have a profound effect on the reactivity. some lithium naphthalenide. This method, with some
Kharisov et al. have prepared the active Zn metal in improvements as mentioned below, was utilized for
a
complicated and completely closed system.[18] the preparation of the highly reactive Rieke zinc
Vyvyan et al. reported some difficulties in preparing metal powder during this study.
large quantities of alkylzinc reagents due to batch to
In the standard preparation of highly reactive
batch variations in the reactivity of the Rieke zinc Rieke zinc, naphthalene is used as an electron carri-
metal.[19] Zhou attempted to prepare alkylzinc bro- er.[8] However, when we started to conduct Rieke
mides from commercially available Rieke zinc (Al- polymerization reactions (towards P3HT) with new
drich), but this did not lead to high conversion of the sources of naphthalene, the in situ formed Zn metal
primary alkylzinc bromides, even when a large excess particles were found to strongly coagulate, yielding
of the active Zn metal was used.[20] He also tried to chunky metal clusters. On the other hand, the “old”
prepare the active Zn metal with Li and a stochiomet- naphthalene batch (Merck) still gave the desired
ric amount of naphthalene, following the procedure Rieke zinc as a finely dispersed black slurry. Hence, it
from Rieke et al.,[8] which also gave no improvement. became clear that the old naphthalene had to be
In this paper, we report on a simple preparation somehow different compared with the new one, and
method for highly reactive Rieke zinc under common a number of analytical methods was applied to ach-
laboratory conditions. As a proof of principle, the im- ieve insight on the properties (notably the impurities)
proved synthetic procedure was applied for the prepa- of different naphthalene sources.
ration of both 3-hexylthiophene and P3HT.
Commercially available naphthalene, produced
The highly reactive form of metallic Zn, generally from coal tar (of which it is the most important poly-
referred to as Rieke zinc (Zn*), was first prepared in cyclic aromatic hydrocarbon fraction),[23] contains
1973.[7] The synthetic protocol involved reduction of small amounts of sulfur-containing aromatic com-
a zinc salt in refluxing THF or DME utilizing either pounds. The most abundant among these impurities is
Na or K. These highly reactive Zn powders showed benzothiophene (BT), as previously also observed by
high reactivity towards organic halides in oxidative Hiyoshi et al.[24] In spite of the fact that the thia-het-
570
asc.wiley-vch.de
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 569 – 575

An Efficient and Reliable Procedure for the Preparation of Highly Reactive Rieke Zinc
erocycles are normally eliminated by a desulfurization
process, it is quite difficult to remove them to a full
extent. Consequently, commercial naphthalene gener-
ally contains small amounts of thia-aromatic impuri-
ties.
The metalIic trace elements in the naphthalene
samples and in the Rieke zinc powder were initially
analyzed by inductively coupled plasma-mass spec-
trometry (ICP-MS) for almost all the elements of the
periodic table. Only a trace of sulfur was detected. In
addition, high-performance liquid chromatography
Scheme 1. Optimized procedure for the preparation of
Rieke zinc.
(coupled to a fluorescence detector; HPLC-FLD) procedure for the preparation of Rieke zinc in highly
analysis of the old (Merck) and new (Acros) naphtha- active form by reducing the zinc chloride salt in THF
lene was performed. The contaminant gave a signal with lithium naphthalenide (prepared in situ) and de-
around 230 nm. The HPLC results obtained for the liberate addition of a catalytic amount of BT, as
two different naphthalene samples are summarized in shown in Scheme 1.
Table 1. Analysis showed that both of samples con-
It appears from our analysis that the presence of
tained the same contamination. The amount was, a catalytic amount of BT in the reducing lithium
however, nine times higher in the old batch compared naphthalenide solution is crucial, affording a fine dis-
to the new naphthalene.
persion of the reactive metal powder. Without BT, co-
To identify the exact nature of the main contami- agulation of the Zn metal particles occurred, giving
nant, gas chromatography-mass spectrometry (GC- rise to unreactive chunky metal clusters. According to
MS) analysis was performed on four different naph- Rieke et al., the appearance of the formed active Zn
thalene batches (Table 2). According to this analysis, powder depends on the amount of zinc chloride salt
the main contaminant has
a
molar mass of added into the lithium naphthalenide solution and on
134 gmolÀ1, which (as expected) corresponds to BT. the addition speed.[8] With slow addition, i.e., 3 sec-
Additionally, the mass spectrum of the contaminant onds per drop with no excess of Zn, a finely divided
was in accordance with the spectrum of BT under the black slurry of the active Zn is formed, whereas in
same conditions. The values obtained from GC-MS case of fast addition, i.e., ~1 second per drop, sponge-
were somewhat lower than the HPLC-FLD ratios due shaped active Zn is obtained. This effect was also ob-
to the difference in response factor.
served by Bronk and discussed in his dissertation on
À
From the above, it is clear that the old (Merck) C C bond formation via intramolecular conjugate ad-
naphthalene was contaminated with BT to a larger dition of organozinc compounds.[25] In our hands,
extent and apparently this has an effect on the stabili- however, it appeared that the reason of the coagula-
zation of the Zn particles in THF solution and hence tion of the highly reactive Zn particles was neither
on the success of the Rieke zinc preparation and reac- the addition of an excess of zinc chloride salt nor the
tivity. For this reason, we have pursued a convenient addition speed, but the phenomenon could be attrib-
uted to the BT impurity, which is present in naphtha-
lene in tiny but varying quantities.
At this stage, a number of additional experiments
was conducted to optimize the Rieke Zn formation.
Table 1. Contaminant content in different naphthalene sour-
ces, as determined by HPLC (relative response).
Naphthalene Purity
[%]
Impurity content
[%]
Exact purity
[%]
First, naphthalene with different purity grades from
different companies was purchased and the reduction
process was carried out in parallel with reactions em-
ploying the old (Merck) naphthalene to allow full
comparison. All experiments that were performed
with new naphthalene batches initially failed. Since
the major difference between the naphthalene sam-
ples was the BT content, novel reactions were con-
ducted with addition of some pure BT to the lithium
naphthalenide mixture. Upon addition of the zinc
chloride solution to lithium naphthalenide containing
BT, a fine slurry of the highly reactive Rieke zinc
metal powder was obtained. Further studies revealed
that the amount of BT had a large effect on the physi-
cal properties of the resulting reactive metal powder,
which then later on affects the oxidative addition pro-
Old (Merck) 99+%
5.6
0.6
94.4
99.4
New
99+%
(Acros)
Table 2. Benzothiophene content in different naphthalene
sources, as determined by GC-MS (relative response).
Naphthalene
Purity [%]
BT content [%]
Old (Merck)
Merck
Aldrich
99+%
ꢀ99%
ꢀ99%
99+%
1.69
0.06
0.14
0.20
New (Acros)
Adv. Synth. Catal. 2013, 355, 569 – 575
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
571

Suleyman Kudret et al.
UPDATES
Figure 1. SEM-EDX pictures of the Rieke zinc powder prepared with old (Merck) naphthalene (left), 1.5 mol% BT
(middle), and 3 mol% BT (right) (see EDX figure in the Supporting Information).
cess of the Rieke zinc metal to organic halides. Thus, pared through our optimized procedure by utilizing it
several experiments were carried out to optimize the in the well-known synthesis of P3HT.[14,29,30] The prep-
ratio BT/zinc chloride. It was found that the optimum aration of P3HT starts from the readily available re-
amount of BT to be added is 3 mol% (with respect to agents 3-bromothiophene (3) and hexyl bromide (1).
ZnCl2). The limiting amount to prevent coagulation Their conversion to 2,5-dibromohexylthiophene (5) is
seems to be 1 mol%.
depicted in Scheme 2.
Scanning electron microscopy-energy disperse X-
First, hexyl bromide (1) was treated with an excess
ray spectroscopy (SEM-EDX) studies were conducted of Rieke Zn*. The reactive Rieke Zn metal readily
on the Rieke metal powder prepared with the differ- undergoes oxidative addition in a 1.0–1.2 Zn*/RX
ent methods to study the impact of the addition of molar ratio at room temperataure to afford hexylzinc
BT on the formed Zn metal powder (Figure 1). In bromide (2) within 2 h.[8] The hexylzinc bromide (2)
case of 1.5 mol% of BT (Figure 1, middle), the ob- solution was dark brown in color and stable under an
tained Zn powder was not homogeneous and con- argon (Ar) atmosphere, and the conversion was esti-
tained Zn particles of very different size compared mated to be >99%, as determined by GC-MS (after
with the ones prepared with either the old (Merck) quenching with an iodine solution). The excess of the
naphthalene (Figure 1, left) and the naphthalene con- reactive metal powder could easily be separated from
taining 3 mol% of BT (Figure 1, right).
the solution by allowing the Zn to settle down for
The Zn metal powder obtained from the reaction about 3 h. The hexylzinc bromide (2) solution was
with 1.5 mol% BT was implemented in a reaction then transferred via cannula to another flask for the
with 2,5-dibromo-3-hexylthiophene (5) (towards the next reaction. In the latter, the coupling was conduct-
formation of P3HT). As expected, oxidative addition ed by reaction of n-hexylzinc bromide (2) with 3-bro-
of the Rieke zinc did not proceed to a full extent (as mothiophene (3) in the presence of 5 mol% of
1
analyzed by H NMR). This can be related to the de- Ni
N
creased active surface of the reactive metal by ag- el (II) chloride}.[31] 3-Hexylthiophene is usually pre-
glomeration of the formed small particles, which gives pared via Kumada coupling of 3-bromothiophene
rise to a loss of their reactivity towards organic hal- using a Grignard instead of an organozinc reagent. To
ides. However, when we prepared the Rieke zinc by the best of our knowledge, Rieke chemistry was em-
adding at least 3 mol% of BT into the lithium naph- ployed here for the first time to prepare 3-hexylthio-
thalenide solution, we always obtained a fine black phene. The coupling reaction was performed at room
slurry of Rieke Zn metal, showing excellent reactivity temperature overnight. Additionally, when the cou-
in the oxidative addition towards organic halides. Ad-
ditionally, the active Zn powder kept its reactivity
over several days and no further agglomeration of the
particles (that could lead to loss of reactivity) was ob-
served.[26] In spite of the fact that the mechanism of
the stabilization of the Rieke zinc particles as a fine
dispersion in THF supernatant is not fully under-
stood, one could suggest a possible complex forma-
tion of BT with either Li or zerovalent Rieke Zn.[27,28]
Following Rieke et al., we then demonstrated the
excellent reactivity of the Rieke zinc powder pre- employing Rieke zinc.
Scheme 2. Preparation of 2,5-dibromo-3-hexylthiophene (5)
572
asc.wiley-vch.de
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 569 – 575

An Efficient and Reliable Procedure for the Preparation of Highly Reactive Rieke Zinc
Scheme 3. Regio-controlled synthesis of P3HT.
pling reaction was performed in the presence of LiBr methanol and hexanes (24 h each), removing residual
the reaction time could be shortened from 2 days to monomer and oligomer fractions. The yield was about
18 h and the yield was dramatically enhanced from 42 70% after Soxhlet purification, and the molecular
to 82% (close to the 87% reported for the Kumada weight increased to Mw =36.1ꢂ103, Mn =22.9ꢂ103
coupling of the corresponding Grignard re- (corresponding to 138 repeating units per chain),
agent).[31,32,33,34] Until now, the role of the LiBr is not while the polydispersity narrowed down to 1.6. There-
fully understood. Nonetheless, the beneficial effect of fore, one can conclude that the monomers were quan-
the addition of a Li salt in such reactions is well docu- titatively cross-coupled in the polymerization reaction
mented. Koszinowski and co-workers suggested that with at least 70% of the monomers incorporated in
the Li salt forms complexes with the organozinc spe- the polymer fraction.
cies, which increases their solubility and provides
The optimized Rieke procedure was later on also
a clean metal surface. On the other hand, due to the applied towards the preparation of a number of ester-
electronegativity of the halogen atom, the organozinc functionalized P3AT copolymers, of particular interest
gains nucleophilic character.[35] Subsequently, bromi- for organic solar cell applications, and also there
nation of 3-hexylthiophene with N-bromosuccinimide decent molecular weights were routinely obtained
(NBS) under mild conditions gave 2,5-dibromo-3-hex- with high reproducibility.[15e–i]
ylthiophene (5).[34] Simple distillation of the crude
product yielded the monomer in high purity.
In summary, we have developed a straightforward
procedure to prepare Rieke zinc in highly reactive
The synthesis of regioregular P3HT (7) is shown in form under standard chemical laboratory conditions.
Scheme 3. The highly reactive Rieke zinc (Zn*) metal The key to afford a highly reproducible protocol is
powder undergoes oxidative addition to 2,5-dibromo- the addition of a particular amount (3 mol%) of BT –
3-hexylthiophene in a regiospecific manner, yielding which was identified as the main additive responsible
2-bromo-5-(bromozincio)-3-hexylthiophene (6a) as for previously observed batch to batch variations – to
the major (95%) and 2-(bromozincio)-5-bromo-3-hex- the lithium naphthalenide solution employed for zinc
ylthiophene (6b) as the minor regioisomer (5%). The chloride reduction. The reactivity of the newly pre-
regioselectivity of the oxidative addition of Zn* metal pared active Zn powder was demonstrated in the oxi-
is higher at lower temperature.[29] In our case, the dative addition to hexyl bromide and 2,5-dibromo-3-
total yield for both regioisomers was >99%, as con- hexylthiophene to give the corresponding organozinc
firmed by GC-MS analysis after quenching the crude reagents in excellent yields. The optimized method
reaction mixture with a saturated ammonium chloride also readily enables the preparation of active Rieke
solution. After completion of the reaction, stirring Zn in large quantities with high reproducibility. Ac-
was stopped and unreacted Zn* particles were al- cordingly, we believe that this optimized procedure
lowed to settle down. Removal of the excess Zn* is can increase the synthetic scope of Rieke zinc in both
crucial since it can further react with a bromine general organic and polymer chemistry, as was al-
source and give rise to bis(organozinc) products, ready shown recently by the preparation of polythio-
which lead to low molecular weight species. Polymeri- phene-based conjugated (co)polymers for applications
zation of the resulting organozinc solution was carried in organic photovoltaics.[15e–i]
out by adding a catalytic amount of NiACTHNUTRGEN(UNG dppe)Cl2 cata-
lyst to afford highly regioregular P3HT.[13]
After completion of the reaction the polymer was
precipitated in a solution of 2M HCl/MeOH (2:1).
The yield of the polymerization was over 90% at this
point. The molecular weight and corresponding poly-
dispersity were measured by gel permeation chroma-
tography (GPC) in THF. The polymer before extrac-
tion had Mw =24.3ꢂ103 and Mn =6.8ꢂ103 (corre-
sponding to 41 repeating units per chain) with a poly-
dispersity D=3.6. The crude polymer was then puri-
Experimental Section
Materials and Methods
All manipulations were carried out with the aid of a dual
manifold vacuum/Ar system. Lithium (granular, 99+%)
from Acros was stored in a Schlenk tube under Ar. Lithium,
naphthalene and benzothiophene (Sigma–Aldrich, 98%)
were weighed in air as needed and transferred to the
Schlenk tube under a stream of Ar. Naphthalene with differ-
fied by means of sequential Soxhlet extractions with ent purity grades was purchased from different suppliers
Adv. Synth. Catal. 2013, 355, 569 – 575
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
573

Suleyman Kudret et al.
UPDATES
and each stored in a separate dessicator together with BT
over P2O5. Zinc chloride (Acros, analysis grade 98,5%) was
transferred into small vials in the glove box and stored in
a dessiccator over P2O5. Zinc chloride was dried by treating
it with thionyl chloride and heating with a Bunsen burner,
and subsequently removed under a stream of Ar. THF was
freshly distilled from Na/benzophenone under an N2 atmos-
phere at atmospheric pressure prior to use. Brass cannulas
were stored at 1108C and cleaned immediately after use
with acetic acid (in the case of Zn* remnants), acetone (to
clean non-polymeric residues), or hot chloroform and/or
chlorobenzene (for polymer-based contaminations).
HPLC analysis was carried out using an Agilent 1100
series system incorporating a chrompack 150 mmꢂ4.6 mm
column and a spectra SYSTEM FL2000 type fluorescence
detector. The mobile phase was MeCN/water (53/47 to 100/0
v/v) with a flow rate of 1.6 mLminÀ1 and an injection
volume of 20 mL. SEM was performed with an FEI Quanta
200 FEG-SEM in combination with EDX. ELAN DRC-e.
ICP-MS was employed to analyze the metallic trace ele-
ments in naphthalene. GC-MS analysis was carried out with
Synthesis of 3-Hexylthiophene (4)
Hexyl bromide (4.7 mL, 28.1 mmol) was dissolved in dry
THF (40 mL) and added via a cannula to the active zinc
powder (33.82 mmol) then the reaction mixture was stirred
for 3 h at room temperature. Stirring was stopped and the
solution was allowed to stand for a couple of hours to allow
the excess of zinc to settle down from the dark brown orga-
nozinc bromide solution. In a 300-mL flame-dried Schlenk
vessel, LiBr (3.41 g, 39.25 mmol), NiACTHNUTRGNEN(UG dppe)Cl2 (0.94 g,
1.784 mmol), and 3-bromothiophene (2.5 mL, 26.76 mmol)
were dissolved in dry THF (150 mL). The organozinc bro-
mide solution was then transferred via cannula over
a period of 1 h to this mixture under continuous stirring at
room temperature and the reaction mixture was allowed to
stir overnight. The resulting mixture was then quenched
with saturated NH4Cl, followed by extraction with diethyl
ether, and simple distillation afforded 3-hexylthiophene as
a colorless liquid; yield: 3.84 g (82%).
Synthesis of 2,5-Dibromo-3-hexylthiophene (5)
a
thermoquest TSQ7000, applying a DB-5MS 30 mꢂ
2,5-Dibromo-3-hexylthiophene was obtained by bromination
of 3-hexylthiophene following the literature procedure as re-
ported by Bꢃuerle et al.[34]
0.25 mm column, in the temperature range 35–3208C. Mo-
lecular weights and molecular weight distributions were de-
termined relative to polystyrene standards by GPC. Chro-
matograms were recorded on a Spectra Series P100 (Spectra
Physics) equipped with two mixed-B columns (10 mm,
0.75 cmꢂ30 cm, Polymer labs) and a refractive index detec-
tor (Shodex) at 408C. THF was used as the eluent at a flow
rate of 1.0 mLminÀ1.
Synthesis of Poly(3-hexylthiophene) (P3HT)
A
solution of 2,5-dibromo-3-hexylthiophene (4.08 g,
12.5 mmol) in THF (50 mL) was added via cannula to fresh-
ly prepared (according to the new procedure) Rieke Zn*
(13.79 mmol) at À788C. The mixture was stirred for 1 h at
this temperature and then allowed to warm to 08C gradual-
ly. The unreacted Zn* was left to settle down overnight and
the formed organozinc supernatant was then transferred by
cannula via a 0.45 mm acrodisc filter into a flame-dried
Enhanced Preparation Method for the Highly
Reactive Rieke Zinc (Zn*) Metal Powder
Two 120-mL Schlenk vessels, A and B, were dried by heat-
ing with a Bunsen burner under reduced pressure and
cooled down to room temperature under a stream of Ar.
Schlenk vessel A, filled with Ar, was weighed and then reas-
sembled to the Schlenk line. Under a stream of Ar, ZnCl2
was charged to the vessel. After three Ar/vacuum cycles,
ZnCl2 was wetted with a small amount of SOCl2. The
Schlenk vessel was heated by a Bunsen burner until the
ZnCl2 salt melted and a white fume was released, and the
Schlenk flask was then cooled down under an Ar flow.
Schlenk flask A was weighed again to determine the exact
amount of ZnCl2 and a stirring bar was added. Dried ZnCl2
(1.1 equiv.) was dissolved in freshly distilled THF
(25 mLgÀ1). Li pellets (2.2 equiv.), naphthalene (2.25 equiv.)
and benzothiophene (0.04 equiv.) were weighted in air and
charged into Schlenk vessel B under an Ar stream. Dry
THF (the same amount as added to dissolve ZnCl2) was
added and the solution turned from colorless to dark green
within less than 2 min. It was stirred further for 2 h to dis-
solve the Li pellets. The ZnCl2 solution was transferred
dropwise via cannula to the lithium naphthalenide solution
over 10–15 min. The resulting black suspension might be
stirred for 1 more h to consume the remaining Li or stirring
can be stopped right after the addition. The highly reactive
zinc powder was allowed to settle down for a couple of
hours. The supernatant was siphoned off via cannula leaving
the Zn* powder. The thus prepared Rieke zinc was ready to
use.
Schlenk tube. Via a cannula, 0.2 mol% of NiACTHNURGTNEUNG(dppe)Cl2
(0.013 g, 0.025 mmol) suspended in THF (5 mL) was added
to the ice-cooled organozinc solution. The Schlenk vessel
was immersed into a preheated oil bath at 608C and the
mixture was stirred at this temperature overnight. It was
then poured into a solution of MeOH:HCl (2M) and the re-
sulting dark precipitate was filtered off and washed several
times with MeOH. The crude polymer (Mw =24.3ꢂ103, Mn =
6.8ꢂ103, D=3.6) was transferred into an extraction thimble
and purification was performed by sequential Soxhlet ex-
tractions with MeOH, acetone and hexanes. The polymer
was collected with CHCl3 and the solvent was removed
under reduced pressure. The polymer was redissolved in
CHCl3 and precipitation was again performed upon addition
of MeOH. Filtration and drying under high vacuum afforded
the pure P3HT material; yield: 1.44 g (70%); Mw =36.1ꢂ
103, Mn =22.9ꢂ103, D=1.6.
Acknowledgements
We thank IMEC for the PhD grant of S.K. We also acknowl-
edge the OPV-Life project from the IWT (O&O 080368), the
POLYSTAR project from PV ERA-NET and the IWT-SBO
project POLYSPEC.
574
asc.wiley-vch.de
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 569 – 575

An Efficient and Reliable Procedure for the Preparation of Highly Reactive Rieke Zinc
References
Vanderzande, U. Wurfel, R. Andriessen, R. Rosch, H.
Hoppe, G. Teran-Escobar, M. Lira-Cantu, A. Rivaton,
G. Y. Uzunoglu, D. Germack, B. Andreasen, M. V.
Madsen, K. Norrman, F. C. Krebs, RSC Adv. 2012, 2,
882.
[1] a) Organozinc Reagents, A Practical Approach, (Eds.:
P. Knochel, P. Jones), Oxford University Press, New
York, 1999; b) The Chemistry of Organozinc Com-
pounds, (Eds.: Z. Rappoport, I. Marek), John Wiley &
Sons ltd, West Sussex, England, 2006; c) P. Knochel,
M. A. Schade, S. Bernhardt, G. Manolikakes, A. Metzg-
er, F. M. Piller, C. J. Rohbogner, M. Mosrin, Beilstein J.
Org. Chem. 2011, 7, 1261; d) X.-F. Wu, Chem. Asian J.
2012, 7, 2505; e) X.-F. Wu, H. Neumann, Adv. Synth.
Catal. 2012, 354, 3141.
[2] S. Reformatsky, Ber. Dtsch. Chem. Ges. 1887, 20, 1210.
[3] H. E. Simmons, R. D. Smith, J. Am. Chem. Soc. 1958,
80, 5323.
[4] E. Frankland, Justus Liebigs Ann. Chem. 1849, 71, 171.
[5] M. S. Newman, F. J. Evans, J. Am. Chem. Soc. 1955, 77,
946.
[6] M. Bellassoued, M. Gaudemar, A. El Borgi, B. Baccar,
J. Organomet. Chem. 1985, 280, 165.
[7] R. D. Rieke, S. J. Uhm, P. M. Hudnall, J. Chem. Soc.
Chem. Commun. 1973, 269.
[8] L. Zhu, R. M. Wehmeyer, R. D. Rieke, J. Org. Chem.
1991, 56, 1445.
[9] M. Hanson, R. D. Rieke, Synth. Commun. 1995, 25,
101.
[10] M. Ma˛kosza, K. Grela, Tetrahedron Lett. 1995, 36,
9225.
[16] T. A. Chen, R. A. OꢄBrien, R. D. Rieke, Macromole-
cules 1993, 26, 3462.
[17] a) P. Coppo, H. Adams, D. C. Cupertino, S. G. Yeates,
M. L. Turner, Chem. Commun. 2003, 2548; b) P. Coppo,
D. C. Cupertino, S. G. Yeates, M. L. Turner, Macromo-
lecules 2003, 36, 2705.
[18] B. I. Kharisov, L. A. Garza-Rodrꢅguez, H. M. Leija Gu-
tiꢆrrez, U. Ortiz Mꢆndez, R. Garcꢅa Caballero, A. Y.
Tsivadze, Synth. React. Inorg. Met.-Org. Nano-Met.
Chem. 2005, 35, 755.
[19] J. R. Vyvyan, C. Loitz, R. E. Looper, C. S. Mattingly,
E. A. Peterson, S. T. Staben, J. Org. Chem. 2004, 69,
2461.
[20] J. S. Zhou, Dissertation, Massachusetts Institute of
Technology, 2005.
[21] a) R. D. Rieke, S. J. Uhm, Synthesis 1975, 452; b) R. T.
Arnold, S. T. Kulenovic, Synth. Commun. 1977, 7, 223.
[22] R. D. Rieke, P. T.-J. Li, T. P. Burns, S. T. Uhm, J. Org.
Chem. 1981, 46, 4323.
[23] G. Azpꢅroz, C. G. Blanco, C. Banciella, Fuel Process.
Technol. 2008, 89, 111.
[24] N. Hiyoshi, R. Miura, C. V. Rode, O. Sato, M. Shirai,
Chem. Lett. 2005, 34, 424.
[11] R. D. Rieke, Science 1989, 246, 1260.
[25] B. S. Bronk, Dissertation, Massachusetts Institute of
[12] a) W.-N. Chou, D. L. Clark, J. B. White, Tetrahedron
Lett. 1991, 32, 299; b) J. Kroemer, C. Kirkpatrick, B.
Maricle, R. Gawrych, M. D. Mosher, D. Kaufman, Tet-
rahedron Lett. 2006, 47, 6339.
[13] T. A. Chen, R. D. Rieke, J. Am. Chem. Soc. 1992, 114,
10087.
[14] a) M. T. Dang, L. Hirsch, G. Wantz, Adv. Mater. 2011,
23, 3597; b) A. Smeets, K. Van den Bergh, J. De Wint-
er, P. Gerbaux, T. Verbiest, G. Koeckelberghs, Macro-
molecules 2009, 42, 7638.
Technology, 1995.
[26] R. D. Rieke, M. S. Sell, W. R. Klein, T.-A. Chen, J. D.
Brown, M. V. Hanson, in: Active Metals, Wiley-VCH,
Weinheim, 2007, pp 1–59.
[27] S. Velu, X. Ma, C. Song, Prepr. Pap. - Am. Chem. Soc.
Div. Fuel Chem. 2003, 48, 694.
[28] L. A. Garza-Rodrꢅguez, B. I. Kharisov, O. V. Kharisso-
va, Synth. React. Inorg. Met.-Org. Nano-Met. Chem.
2009, 39, 270.
[29] T.-A. Chen, X. Wu, R. D. Rieke, J. Am. Chem. Soc.
[15] a) M. Pomerantz, Y. Cheng, R. K. Kasim, R. L. Else-
nbaumer, J. Mater. Chem. 1999, 9, 2155; b) F. Saito, Y.
Takeoka, M. Rikukawa, K. Sanui, Synth. Met. 2005,
153, 125; c) M. Katagiri, Adv. Mater. Res. 2011, 409,
502; d) R. D. Rieke, S.-H. Kim, Bull. Korean Chem.
Soc. 2012, 33, 1; e) R. Rosch, D. M. Tanenbaum, M.
Jorgensen, M. Seeland, M. Barenklau, M. Hermenau,
E. Voroshazi, M. T. Lloyd, Y. Galagan, B. Zimmer-
mann, U. Wurfel, M. Hosel, H. F. Dam, S. A. Gevorgy-
an, S. Kudret, W. Maes, L. Lutsen, D. Vanderzande, R.
Andriessen, G. Teran-Escobar, M. Lira-Cantu, A. Riva-
ton, G. Y. Uzunoglu, D. Germack, B. Andreasen, M. V.
Madsen, K. Norrman, H. Hoppe, F. C. Krebs, Energy
Environ. Sci. 2012, 5, 6521; f) D. M. Tanenbaum, M.
Hermenau, E. Voroshazi, M. T. Lloyd, Y. Galagan, B.
Zimmermann, M. Hosel, H. F. Dam, M. Jorgensen,
S. A. Gevorgyan, S. Kudret, W. Maes, L. Lutsen, D.
1995, 117, 233.
[30] Y. Firdaus, A. Khetubol, S. Kudret, H. Diliꢇn, W. Maes,
L. Lutsen, D. Vanderzande, M. Van der Auweraer,
Proc. Soc. Photo-Opt. Instrum. Eng. 2012, 8438,
84381G.
[31] J. Dai, J. L. Sellers, R. E. Noftle, Synth. Met. 2003, 139,
81.
[32] S.-H. Kim, J.-G. Kim, Bull. Korean Chem. Soc. 2009,
30, 2283.
[33] G. T. Achonduh, N. Hadei, C. Valente, S. Avola, C. J.
OꢄBrien, M. G. Organ, Chem. Commun. 2010, 46, 4109.
[34] P. Bꢃuerle, F. Pfau, H. Schlupp, F. Wurthner, K.-U.
Gaudl, M. B. Caro, P. Fischer, J. Chem. Soc. Perkin
Trans. 2 1993, 489.
[35] J. E. Fleckenstein, K. Koszinowski, Organometallics
2011, 30, 5018.
Adv. Synth. Catal. 2013, 355, 569 – 575
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
575