Ultrasounds in melted poly(ethylene glycol) promote copper-catalyzed cyanation of aryl halides with K4[Fe(CN)6]

Article abstract of DOI:10.1002/cssc.201300997

Melted poly(ethylene glycols) (PEGs) were used for the first time as solvent for the sonochemically promoted cyanation of aryl halides employing inexpensive and safe K₄[Fe(CN)₆] and a relatively low amount of Cu-based catalyst. The M_w (weight-average polymer molecular weight) of PEG proved to notably influence the substrate conversion, which is indicative of a strong dependence of the sonication efficacy on solvent properties. Gel permeation chromatography (GPC), X-ray photoelectron spectroscopy (XPS) and transmission electron microscopy (TEM) contributed to the characterization of the polymer and the elucidation of the catalytic system. Off the PEG: Melted poly(ethylene glycol)s (PEG)s were used for the first time as solvent for the sonochemical cyanation of aryl halides using K ₄[Fe(CN)₆] and a relatively low amount of Cu catalyst (5 %). The weight-average molecular weight of PEG was proven to exert a notable influence on conversion, and CuI proved to be the most efficient catalyst, affording good results within short reaction times with the use of aryl iodides and activated aryl bromides as substrates.

Full text of DOI:10.1002/cssc.201300997

CHEMSUSCHEM
FULL PAPERS
DOI: 10.1002/cssc.201300997
Ultrasounds in Melted Poly(ethylene glycol) Promote
Copper-Catalyzed Cyanation of Aryl Halides with
K4[Fe(CN)6]
Guido Giachi,[a, c] Marco Frediani,[a] Werner Oberhauser,[b] Frederic Lamaty,[c] Jean Martinez,[c]
and Evelina Colacino*[c]
This article is dedicated to Prof. Irina P. Beletskaya on the occasion of her 80th Birthday.
Introduction
Cyanation of aryl halides is of great practical importance as ar-
omatic nitriles constitute a class of widespread organic inter-
mediates as well as fine chemicals with applications as phar-
maceuticals, perfumes, or dyes.[1,2] Traditional cyanation proce-
dures mostly require stoichiometric amounts of extremely
toxic metal cyanides.[3] This latter issue was tackled by Beller
and co-workers[4] by introducing nonpoisonous, inexpensive
potassium hexacyanoferrate(II) as a cyanide source coupled
with a phosphine-free Pd catalyst. Since then, a great deal of
research was carried out and further improvements were pro-
posed, including the use of heterogeneous catalysts,[5,6] large-
scale experimental setups[7] or microwave reactors,[8] or to pro-
mote the cyanation reaction starting from (hetero)arylchlor-
ides.[9]
range. Lower Cu-catalyst loading leads to a dramatic drop in
nitrile formation.
Ultrasonic probes and sonication baths are already acknowl-
edged as useful equipment for processes of industrial rele-
vance like extractions,[17,18] generation of emulsions,[19] polymer
synthesis and degradation,[20,21] and many other applications.
Currently, due to the strong demand for alternative energy
sources for chemical transformations, ultrasonic reactors are at-
tracting the attention of a growing number of organic and
nanomaterial synthetic chemists.[22–24] Indeed, the peculiar phe-
nomena and locally extreme conditions due to the propaga-
tion of sound waves through the reaction medium may facili-
tate diffusion and homogenization, increasing the rate or even
switching the selectivity of a synthesis.[25–27] During the last
decade, nontoxic and relatively cheap PEGs have been effec-
tively employed in catalysis as phase-transfer agents,[28–30]
metal nanoparticles stabilizers,[31,32] and as safe solvents with
high boiling points.[33,34]
A major step forward in this field of research was achieved
by replacing costly and toxic palladium with copper-based cat-
alytic systems[10–14] in volatile organic solvents (toluene, dime-
thylacetamide, or N-methyl-2-pyrrolidone) under prolonged
conventional heating (16–24 h). More recently, a few examples
of microwave-accelerated syntheses in environmentally benign
solvents such as water were reported.[15,16] However, phase-
transfer agents (n-tetrabutylammonium bromide[16] or tetra-
ethylene glycol)[15] were always necessary. A troublesome
common feature of all Cu-catalyzed cyanation reactions is the
need for a very high catalyst loading, in the 10 to 30 mol%
Although some remarkable examples of ultrasound-induced
reactivity of polymer-tagged substrates are known,[26,27,35] to
the best of our knowledge organic syntheses performed by
sonicating a mixture of reactants in melted polymers has never
been reported.
Results and Discussion
Previous reports[36] and some preliminary experiments per-
formed in our laboratory clearly showed that the microwave-
enhanced cyanation reaction failed when PEG alone [PEG-300,
PEG-3400, PEG-4000, or PEG-2000-(OMe)2] was used as the re-
action solvent in palladium-,[36] copper-, or silver-catalyzed
transformations. Convinced that the suitable choice of activa-
tion technique could have a deep impact on the final outcome
of the reaction, we coupled a probe sonicator with melted PEG
and tested the newly developed cyanation procedure on
a series of aryl iodides and bromides. We employed ligand-free
copper catalysts and potassium hexacyanoferrate(II) as cyanide
source. In order to perform ultrasound irradiation and accord-
ing to Equations (1) and (2), the polymer had to be homogene-
ously liquid at the operational temperature. This was experi-
mentally achieved by setting the temperature of the ultra-
[a] G. Giachi, M. Frediani
Dipartimento di Chimica “U. Schiff”
Universitꢀ di Firenze
Via della Lastruccia 3-13, 50019 Sesto Fiorentino, Firenze (Italy)
[b] W. Oberhauser
Istituto di Chimica dei Composti Organometallici (ICCOM-CNR)
Area di Ricerca CNR di Firenze
via Madonna del Piano 10, 50019 Sesto Fiorentino, Firenze (Italy)
[c] G. Giachi, F. Lamaty, J. Martinez, E. Colacino
Institut des Biomolꢁcules Max Mousseron (IBMM)
UMR 5247 CNRS-UM I-UM II
Universite’ de Montpellier II
Place E. Bataillon, 34095 Montpellier CEDEX 5 (France)
E-mail: evelina.colacino@univ.montp2.fr
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201300997.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 919 – 924 919

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
sound (US) probe at 808C, thus allowing polymer melting (PEG
mixture melting point was around 608C). Preliminary tests
were performed with polymers with low molecular weight,
that is, PEG-250-(OMe)2, PEG-300 or monomethyl ether PEG-
1100. No conversion of the starting material was observed. Our
first aim was to determine the influence of the weight-average
molecular weight (Mw) of PEGs on reaction conversion and
yield, using 4-iodoanisole as benchmark substrate and 5 mol%
commercial copper nanopowder (Cu-NP) as catalyst. We fine-
tuned the Mw of the polymer by varying the ratio of a PEG-
3400/PEG-10000 mixture. Based on previous findings,[35] the re-
action mixture was sonicated in a pulsed fashion (i.e., on/off=
5 s/5 s) for a total reaction time of 20 min (i.e., 10 min total
sonication). As a result, we observed a narrow high substrate
conversion and a yield dependence according to the Mw of the
PEG mixture. A set of PEGs with different values of Mw, ob-
tained by mixing different proportions of PEG-3400 and PEG-
10000, was screened using 20 g of mixture (per reaction)[38]
Table 1. Cu-NP-catalyzed cyanation of 4-iodoanisole in melted PEG.
Entry[a]
PEG Mw
[Da]
PEG [g]
3400
t [min]
Conversion[b]
[%]
Yield[c]
[%]
10000
1
3400
4200
5000
5300
5600
5900
6000
6200
7000
5000
5000
20.00
17.58
15.15
14.24
13.33
12.42
–
11.52
10.91
–
0.00
2.42
4.85
5.76
6.76
7.58
–
8.48
9.09
–
20
20
20
20
20
20
20
20
20
20
40
18
26
74
86
94
99
70
89
98
73
0
18
24
63
72
81
75
67
72
73
62
0
2[d]
3[d]
4[d]
5[d]
6[d]
7[e]
8[d]
9[d]
10[f]
11[g]
–
–
[a] 4-iodoanisole (4.0 mmol), K4[Fe(CN)6] (2.4 mmol), Cu-NP (0.2 mmol),
melted PEG (20.0 g), N2 atmosphere, US-probe temperature set at 808C,
pulsed sonication (on/off=5 s/5 s), amplitude (A) of 70%. [b] Conversion
was determined by HPLC and 1H NMR. [c] Yield was determined by
1H NMR spectroscopy using CH2Br2 as internal standard. [d] PEG-3400/
PEG-10000 weighted mixture. [e] Commercially available PEG-6000 was
used, 4-iodoanisole (4.0 mmol), K4[Fe(CN)6] (1.2 mmol), Cu-NP (0.6 mmol),
melted PEG (20.0 g). [f] Commercially available monomethyl ether PEG-
5000 was used. [g] 20 wt% MeO-PEG-5000/water solution (20.0 g).
and showed
a maximum catalytic activity for PEG-5600
(Figure 1 and Table 1, entries 1-9).
The poor substrate conversions observed for low-Mw PEGs
may be rationalized on the basis of the mathematical Equa-
tion (1) which describes the attenuation of acoustic waves:[39]
I ¼ I0 exp ðꢀ2 a lÞ
ð1Þ
where f is the ultrasonic frequency, 1 is the density of the
medium, and c is the speed of sound through the medium.
Throughout the experiments f is kept constant (i.e., 20 kHz),
while the variation of 1 within the screened Mw interval was
negligible. Hence, it is reasonable to assume that the increase
in melt viscosity (which is markedly dependent on molecular
weight)[40] plays a major role in enhancing the absorption of
acoustic energy and, consequently, determines a more efficient
energy-transfer to the molecules of reactants.
I is the acoustic intensity, l corresponds to the distance of
the emitting probe, and a is the absorption coefficient, which
depends on the viscosity of the medium (hs) as given by the
Navier–Stokes Equation (2):
8p2f2
ð2Þ
a ¼
hs
31c3
On the other hand, melted PEGs with Mw higher than
6000 Da (Table 1, entry 7) up to 20000 Da were probably too
viscous to allow an efficient diffusion of reactants. On using
PEG-7000, agitation was visibly difficult and our reactor was
forced to work at a power level close to overload. The lower
homogeneity resulted in a decreased selectivity with 4,4’-dime-
thoxy-biphenyl and 4-methoxy-benzylamine as the main by-
products. The results obtained from a reaction in pure MeO-
PEG-5000, which were similar to those obtained with the PEG-
5000 mixture (Table 1, entry 10 vs. 3) suggested that the effect
of mixing high- and low-molecular weight PEGs was not mark-
edly important. A 20 wt% MeO-PEG-5000/water solution was
tested as solvents in the same experimental conditions and no
conversion was obtained even after prolonged sonication
(Table 1, entry 11). This finding suggested that water might
hamper reactivity by lowering the viscosity of the mixture and
limiting the reaction temperature by partial evaporation. This
clearly showed the interesting effect of ultrasound in promot-
ing the cyanation reaction in pure PEG; with the use of micro-
wave activation,[36] this reaction was ineffective and succeeded
only in PEG/water mixtures.[15,36]
Figure 1. Cu-NP-catalyzed cyanation of 4-iodoanisole in melted PEG; conver-
sion and yield versus Mw of the PEG mixture.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 919 – 924 920

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
Given the strong influence of PEG Mw on the efficacy of the
whole reaction system, it was important to evaluate the stabili-
ty of the polymer backbone under the US-assisted cyanation
reaction conditions. To this end, we recovered the PEG-5600
phase after work-up. It was reused in a second run, and a GPC
(gel permeation chromatography) analysis on native, recov-
ered, and recycled polymer phases was performed. The varia-
tion in the polydispersion index of Mw did not exceed experi-
mental error (see Supporting Information).
idation state in accordance with the low performance ob-
served for CuO (Table 2, entry 4). On the other hand, the close
characteristic binding energies and very low concentration of
copper in PEG made an unambiguous discrimination between
Cu0 and CuI species impossible. TEM investigations on PEG
phases after catalysis showed aggregates of quite similar mor-
phology (Figure 2) in which geometrical and spherical shaped
particles with highly variable size (approximately from 10 to
50 nm) appear to coexist in large clusters for reactions cata-
lyzed by both 5 mol% Cu-NP and CuI. A high resolution trans-
mission electron microscopy–X-ray energy dispersive spectros-
copy (HRTEM–EDS) analysis (see Supporting Information) sug-
gested the presence of major quantities of residual K4[Fe(CN)6]
and Fe0 obtained from the redox process involving the reduc-
tion of the latter by the hydroxyl end-groups of polyol
After selecting the polymer blend with optimal Mw and melt
viscosity (i.e., PEG-5600), we screened the activity of some
copper species at different oxidation states (Table 2, entries 1–
4). As a result, Cu-NP and CuI proved to be the most promising
catalysts, giving almost complete conversions and good yields
even at a 5 mol% catalyst loading. On the other hand, CuI and
CuII oxides showed moderate and poor activity, respectively.
A test was repeated employing a triple concentration of re-
actants in PEG and gave nearly identical results (Table 2,
entry 5 vs 2), suggesting that the throughput of the synthesis
might be easily scaled up. As expected, conversion and yield
values could be further increased by increasing the amount of
catalyst to 15 mol% (Table 2, entries 6 and 7). Once again, Cu-
NP and CuI showed a very similar catalytic performance.
To further investigate the nature of the active catalytic spe-
cies, we performed an XPS analysis on both 15 mol% Cu-NP-
and CuI-containing prereaction mixtures and postreaction PEG
phases recovered after work up (by using precipitation and fil-
tration, see Supporting Information).[33,41,42] Though the easily
recognizable CuII satellite peaks (942 and 962 eV) indicated sur-
face oxidation in native Cu-NP, the two PEG phases did not
contain CuII after reaction, ruling out the latter as the active ox-
Table 2. Cu- and Pd-catalyzed cyanation of 4-iodoanisole in melted PEG-
5600.
Entry[a]
Catalyst
[Catalyst]
[mol%]
t
Conversion[b]
[%]
Yield[c]
[%]
Figure 2. TEM images of PEG phases recovered after reactions catalyzed by:
(a) Cu-NP and (b) CuI.
[min]
1
2
3
4
5[d]
6
Cu-NP
CuI
Cu2O
CuO
CuI
Cu-NP
CuI
–
5
5
5
5
5
15
15
–
1
5
20
20
20
20
20
15
15
20
12
120
30
94
92
68
42
97
>99
>99
64
>99
0
81
84
57
33
84
94
92
54
75
0
PEG,[43,44] as well as a minor quantity of copper (in accordance
with the reaction stoichiometry).
Although they clarify the fate of iron in the cyanation reac-
tion, these observations leave some uncertainty about the
identification of the catalytic active copper species, for which
more thorough characterization will be needed.
7
8[e]
9[f]
10[g]
11
Pd(OAc)2
Pd(OAc)2
–
Inductively coupled plasma–optical emission spectroscopy
(ICP–OES) measurements performed on PEG recovered from
the CuI-catalyzed (5 mol%) 4-iodoanisole cyanation reaction
showed copper concentrations of 570 and 63 ppm in the poly-
mer and product phase respectively, confirming a copper
leaching of about 10% into the organic phase. Consequently,
the organic product contamination was far below the thresh-
old levels for human copper intake (i.e., tolerable upper intake
level=10 mgdayꢀ1).[45] Iron concentrations of 2800 ppm and
<0.01 ppm were found in the polymer phase and product, re-
0
0
0
[a] 4-iodoanisole (4.0 mmol), K4[Fe(CN)6] (2.4 mmol), melted PEG-5600 as
a PEG-3400/PEG-10000 weighted mixture (20.0 g), N2 atmosphere; US-
probe temperature was set at 808C, pulsed sonication (on/off=5 sec./5
sec.), amplitude of 70%. [b] Conversion was determined by HPLC and
1H NMR spectroscopy. [c] Yield was determined by 1H NMR using CH2Br2
as internal standard. [d] 4-iodoanisole (12.0 mmol), K4[Fe(CN)6] (7.2 mmol),
melted PEG-5600 (16.0 g). [e] 2nd cycle, 4-iodo anisole (4 mmol) and
K4[Fe(CN)6] (2.4 mmol) were added. [f] Addition of Na2CO3 (4.0 mmol).
[g] The reaction was performed under microwave activation in a sealed
vessel at 1408C using Na2CO3 (4.0 mmol).
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 919 – 924 921

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
spectively. This experimental result indicated that roughly half
of the stoichiometric amount of K4[Fe(CN)6] was effectively
available for the cyanation reaction in the polymer melt. In ad-
dition, this latter finding was in accordance with the visual ob-
servation that mechanical segregation of the salt to the flask
neck occurred during the first minutes of sonication. This ex-
perimental problem might be resolved by developing a more
sophisticated reactor setup, placing for example an ultrasound
probe-dedicated tight septum between the probe and the
neck, recycling of the segregated salt during sonication. The
recycling of the polymer/catalyst system was performed by
melting the recovered PEG phase and adding fresh substrate
and K4[Fe(CN)6] to the polymer melt. As a result, a drop in sub-
strate conversion was registered for the second cycle (Table 2,
entry 8 vs 2) probably due to the notable aggregation of Cu-
NP displayed by TEM analyses (see Supporting information).
Based on these observations, we hypothesized a reaction
mechanism based on copper reduction by the polymer end
groups followed by oxidative addition of the aryl halide, coor-
dination of the cyanide group with release of the halide, and
reductive elimination yielding the benzonitrile product
(Scheme 1). Both the intermediate Cu0 and CuII species could
be easily stabilized by the PEG matrix.[29] Formation of stable
Cu2[Fe(CN)6] complexes as well as the high affinity of cyanide
ions towards copper-based catalysts could be responsible for
a fast deactivation of the catalytic system.[4,11] PEG polymer
acts not only as the reaction solvent and reducing agent, but
also as a metal-stabilizing protective matrix, thus avoiding the
need for high catalyst loading. Indeed, chelation of potassium
cations, as a result of its crown-ether-like effect,[46] could be re-
a thermally mediated process, several cyanation reactions were
carried out using microwave heating. Because of its intrinsic
properties, poly(ethylene glycol) is a suitable solvent for micro-
wave irradiation, where the heating characteristics of the sol-
vent play a crucial role.[47]
The effective heating of the mixture is further enhanced by
the presence of metallic and/or ionic species (e.g., K4[Fe(CN)6]
or copper salts). The microwave-enhanced cyanation of arylio-
dides was investigated at temperatures up to 1608C in various
PEGs (PEG-300, PEG-3400, PEG-5600, or PEG-2000-(OMe)2) that
differed in molecular weight and in the nature of the end-
groups. Catalytic systems based on palladium, silver, or copper
salts were also tested in the presence of an external ligand
(N,N-dimethylethylendiamine) or KI as additive. In all cases, the
substrates were not converted, confirming the literature re-
ports.[15,36] On the basis of these considerations, thermal effects
were not effective in promoting the reaction. In light of this
finding, other activation mechanisms[26,27] need to be evoked
for the ultrasound promoted cyanation reaction, such as cavi-
tation or mechanochemical effects for heterogeneous systems.
The ability of a medium to absorb the acoustic energy de-
pends on the viscosity of the medium. In the case of PEGs, this
is markedly dependent on its molecular weight and deter-
mines the energy transfer to the molecules of the reactants
[Eqs. (1) and (2)].
A blank reaction (i.e., with 0 mol% Cu) was carried out to ex-
clude reagents or PEGs contamination by any catalytic species.
No conversion was observed even after prolonged sonication
(Table 2, entry 11). On the basis of its good catalytic per-
formance, CuI was selected as the optimal catalyst and used to
study the reaction scope with a number of mono and bi-func-
tional aryl iodides and bromides. Throughout this screening,
a significantly lower amount of copper catalyst (i.e., 5 mol%)
was employed compared to literature reports (in which
amounts varied from 10 to 30 mol%). Good-to-excellent yields
were obtained in relatively short reaction times. As a result,
aryl iodides (Table 3, entries 1, 3–8, and 10) were cyanated
smoothly, regardless of the electronic nature of substituents
with the exception of p-methyl-iodobenzene (Table 3, entry 3)
and p-nitro-iodobenzene (Table 3, entry 4), which showed high
to total conversion but a moderate yield. p-Methyl-benzonitrile
proved to be unstable during purification by column chroma-
tography. In the case of p-nitro-iodobenzene, reduction of the
nitro group occurs quantitatively,[48] assuming that a copper
hydride species might be generated during the reaction
course in the presence of hydroxyl functional groups, as previ-
ously described with other metals,[49] and stabilized by the
presence of the polyoxygenated backbone of PEG. Moreover,
extraction of the p-cyano-aniline from PEG phase was probably
hampered by the affinity between oxygen atoms in the poly-
Scheme 1. Proposed reaction mechanism for copper-catalyzed cyanation
with K4[Fe(CN)6] in melted PEG.
sponsible for the improved reactivity of the “naked” [Fe(CN)6]4ꢀ
counterion under ultrasound activation.
It is worth noting that Pd(OAc)2 (1 mol%) could be em-
ployed as a catalyst with good results under sonochemical
conditions (Table 2, entry 9), though recent attempts described
in the literature[36] and in our laboratory (Table 2 entry 10)
showed that Pd-based species failed to give satisfactory yields
when microwaves were used as the energy source in the pres-
ence of bulk PEG. In order to demonstrate that the successful
outcome of the reaction was due to ultrasounds and not to
mer chain and the polar ꢀNH2/ꢀNH3 groups of the product.[50]
+
When 2-iodobenzylalcohol (Table 3, entry 7) was cyanated in-
stead of 3-iodobenzylalcohol (Table 3, entry 6), the conversion
was quantitative, but the yield was hampered by the formation
of byproducts such as isoindoline, through an intramolecular
reductive alkylation of cyano group with the alcohol,[51] and
isobenzofuran-1(3H)-one.[52] These by-products were detected
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 919 – 924 922

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
was modest. The presence of an electron-donating group is re-
sponsible for the low conversion. At the same time, the newly
formed benzylcyanide can also act as source of cyanide ions,[14]
leading to uncontrolled side reactions. Cyanation of 3-chloro-
anisole was tentatively performed under these conditions but
conversion into benzonitrile failed even after 1 h of sonication.
Table 3. Cu-NP-catalyzed cyanation of arylhalides in melted PEG.
Entry[a] Substrate
Product
t [min] Conversion [%][b] Yield[c] [%]
92
1
20
60
20
20
92
50
(84)
2[d]
3
(13)
27
87
(80)
Conclusions
4
>99
57[e]
We demonstrated the feasibility and effectiveness of the sono-
chemically enhanced cyanation of aryl iodides and electron-
poor aryl bromides in melted PEG using K4[Fe(CN)6] as cyanide
source and a relatively low amount (i.e., 5 mol%) of CuI as cat-
alyst. Interestingly, the substrate conversion into the desired ni-
trile turned out to depend on the viscosity of melted PEG,
showing the highest substrate conversion for PEG-5600. The
reaction was fast and selective, although the activity of the cat-
alytic system decreased upon recycling. The quantity of copper
leached into benzonitrile products was far below the threshold
levels for human copper intake, suggesting that the procedure
is potentially suitable for food or pharmaceutical applications.
84
(97)
5
6
20
20
97
94
79
(91)
7
20
20
60
>99
>99
95
(15)
83
(88)
72
8[f]
9[d]
(75)
89
(95)
10[f]
20
>99
11
20
60
90
60
87
>99
90
(87)[g]
Experimental Section
80
(70)
85
(88)
65
12[d,f]
13[d]
14[d]
Typical experimental procedure for the cynation of aryliodides:
Cyanation reactions were carried out with a 400 W Bioblock Scien-
tific Vibracell 72408 equipped with a 12 mm diameter titanium
alloy probe and a rubber septum. Sonication was pulsed (on/off=
5 s/5 s) for 20 min (total sonication time: 10 min), wave amplitude
was 70%, and probe temperature was set at 808C. PEG (20.0 g)
was slowly melted in a 25 mL round-bottom flask by means of
a hot air gun under vacuum/N2 cycles, then aryliodide (4 mmol),
copper catalyst (5 mol%), and K4[Fe(CN)6] (2.4 mmol) were added
under N2 flux and the probe was inserted into the flask. As the
rubber septum was applied on the flask neck, the N2 flux was
stopped and sonication started. After sonication, the raw product
was analyzed by HPLC–UV; then CH2Cl2 was used to recover the
product from the probe and to dissolve the mixture. The solution
was carefully withdrawn and transferred dropwise into Et2O.[33,41,42]
The light-grey slurry was kept at ꢀ168C for 4 h and then filtered
through a frit. The clear, liquid phase was concentrated to dryness
and the product was redissolved in CDCl3 for NMR and HPLC–UV
86
(82)
15[d]
16[d]
60
60
>99
0
36
(21)
[a] Arylhalide (4.0 mmol), K4[Fe(CN)6] (2.4 mmol), CuI (0.2 mmol), melted
PEG-5600 as a PEG-3400/PEG-10000 weighted mixture (20.0 g), N2 atmos-
phere, US-probe temperature was set at 808C, pulsed sonication (on/
off=5 sec./5 sec.), amplitude (A) 70%. [b] Conversion was determined by
HPLC and 1H NMR spectroscopy. [c] Yield given in parenthesis was deter-
mined by 1H NMR spectroscopy using CH2Br2 as internal standard. [d] KI
(2.4 mmol) was also used. [e] No 4-Nitrobenzonitrile was obtained. [f] Ar-
ylhalide (2.0 mmol), K4[Fe(CN)6] (2.4 mmol), CuI (0.2 mmol), melted PEG-
5600 (20.0 g), N2 atmosphere. [g] Product is a mix of X=Br (44%), X=CN
(24%), and X=I (19%).
1
analysis. H NMR yields were measured using CH2Br2 (0.712 mmol,
50.0 mL) as internal standard. CH2Br2 was added to a weighted ali-
quot of the sample (1:10 of the crude) into CDCl3 and analyzed by
means of 1H NMR spectroscopy. CH2Br2 displays a singlet at d=
4.91 ppm integrating for two protons. The integral areas of both
CH2Br2 and one characteristic signal belonging to the desired prod-
uct (normalized for two protons) were measured and expressed in
cm. They are respectively referred to as ‘cmCH2Br2’ and ‘cmprod-
uct’. The quantity of product (in mmol) is calculated as follows:
0.712 mmolꢁ(cmproduct/cmCH2Br2). When necessary, residual
poly(ethylene glycol) was eliminated by aqueous washing or
column chromatography. For materials and characterization meth-
ods, see Supporting Information.
1
in the crude by H NMR and LC/MS analyses. In the presence
of KI (60% molar ratio), electron-poor or neutral aryl bromides
also gave high yields, although longer sonication times were
needed compared to iodides (Table 3, entries 12–14). On the
other hand, electron-rich aryl bromides were less active
(Table 3, entry 2). Reaction with p-bromo-iodobenzene (Table 3,
entry 11) resulted in high conversion but low chemoselectivity,
as cyanation either occurred at the iodide group (prevalently),
at the bromide group or at both, even without the addition of
KI. Methyl 2-bromobenzoate (Table 3, entry 15) was also fully
converted, but many unidentified byproducts were obtained.
With 1-bromo-2-(bromomethyl)-4-methoxybenzene (Table 3,
entry 16) the yield of 2-(cyanomethyl)-4-methoxybenzonitrile
Typical experimental procedure for the cynation of arylbro-
mides: Arylbromides (4.0 mmol) were cyanated under analogous
conditions to aryliodides, in the presence of KI (2.4 mmol) as addi-
tive. The work-up and product recovery was carried out as already
described for aryliodides.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 919 – 924 923

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
[20] M. M. Caruso, D. A. Davis, Q. Shen, S. A. Odom, N. R. Sottos, S. R. White,
J. S. Moore, Chem. Rev. 2009, 109, 5755–5798.
Typical experimental procedure for recycling: The precipitate
PEG-5600/Cu (PEG–metal catalytic system) obtained after the pre-
cipitation/filtration work-up from a previous experiment (Table 2,
entry 2) was dried under vacuum. Then, 4-iodoanisole (4 mmol)
and K4[Fe(CN)6] (2.4 mmol) were added to the melted precipitate
and the reaction mixture was sonicated under analogous condi-
tions to the first cycle. At the end of the reaction, the product and
the PEG–metal catalytic system were separated by means of the
previously described work-up procedure based on a precipitation/
filtration step[31,34,41,42,49,53]
[21] T. Sato, D. E. Nalepa, J. Coat. Technol. 1977, 49, 45.
[22] R. B. Baig, R. S. Varma, Chem. Soc. Rev. 2012, 41, 1559–1584.
[23] T. Ding, J.-J. Zhu, J.-M. Hong, Mater. Lett. 2003, 57, 4445–4449.
[24] A. Tuulmets, G. Cravotto, S. Salmar, J. Jꢂrv, Mini-rev. Org. Chem. 2010, 7,
204–211.
[25] T. Ando, S. Sumi, T. Kawate, J. Ichihara, T. Hanafusa, J. Chem. Soc. Chem.
Commun. 1984, 439–440.
[26] G. Cravotto, P. Cintas, Angew. Chem. 2007, 119, 5573–5575; Angew.
Chem. Int. Ed. 2007, 46, 5476–5478.
[27] G. Cravotto, P. Cintas, Chem. Sci. 2012, 3, 295–307.
[28] B. Sauvagnat, F. Lamaty, R. Lazaro, J. Martinez, Tetrahedron Lett. 1998,
39, 821–824.
[29] G. E. Totten, N. A. Clinton, J. Matlock, J. Macromol. Sci. Rev. Macromol.
Chem. Phys. 1988, 28, 293–337.
Acknowledgements
[30] G. E. Totten, N. A. Clinton, J. Matlock, J. Macromol. Sci. Rev. Macromol.
Chem. Phys. 1998, 38, 77–142.
[31] E. Colacino, J. Martinez, F. Lamaty, L. S. Patrikeeva, L. L. Khemchyan, V. P.
Ananikov, I. P. Beletskaya, Coord. Chem. Rev. 2012, 256, 2893–2920.
[32] G. Giachi, W. Oberhauser, M. Frediani, E. Passaglia, L. Capozzoli, L. Rosi,
J. Polym. Sci. Part A 2013, 51, 2518–2526.
[33] E. Colacino, L. Daich, J. Martinez, F. Lamaty, Synlett 2007, 1279–1283.
[34] V. Declerck, P. Ribiꢃre, Y. Nꢄdellec, H. Allouchi, J. Martinez, F. Lamaty, Eur.
J. Org. Chem. 2007, 201–208.
[35] C. R. Hickenboth, J. S. Moore, S. R. White, N. R. Sottos, J. Baudry, S. R.
Wilson, Nature 2007, 446, 423–427.
The authors thank for financial support the French CNRS and
MESR. The authors are grateful to Franck Godiard (for TEM analy-
ses, ‘Service de Microscopie Electronique et Analytique’, Universitꢁ
Montpellier II-France), Valꢁrie Flaud (for XPS data, Institut Charles
Gerhardt, Montpellier-France for fruitful discussions) and to
Marco Carlo Mascherpa (for ICP-OES analyzes, ICCOM, UOS-Pisa,
Italy).
[36] G. Chen, J. Weng, Z. Zheng, X. Zhu, Y. Cai, J. Cai, Y. Wan, Eur. J. Org.
Chem. 2008, 3524–3528.
[37] A. Azua, J. A. Mata, P. Heymes, E. Peris, F. Lamaty, J. Martinez, E. Colaci-
no, Adv. Synth. Catal. 2013, 355, 1107–1116.
Keywords: catalysis
glycol) · sonochemistry
· copper · cyanation · poly(ethylene
[38] The resulting Mw of the mixture is given by: (weight fraction of PEG-
3400)ꢁ3400+(weight fraction of PEG-10000)ꢁ10000. For example,
5600=0.33ꢁ10000+0.66ꢁ3400; hence, in order to get 20 g of PEG-
5600 mixture 6.66 g of PEG-10000 and 13.33 g of PEG-3400 were
weighted.
[39] T. J. Mason in Sonochemistry: The Uses of Ultrasound in Chemistry, The
Royal Society of Chemistry, 1990.
[40] L. W. Kessler, W. D. O’Brien, F. Dunn, J. Phys. Chem. 1970, 74, 4096–
4102.
[41] E. Colacino, L. Villebrun, J. Martinez, F. Lamaty, Tetrahedron 2010, 66,
3730–3735.
[42] V. Declerck, J. Martinez, F. Lamaty, Synlett 2006, 3029–3032.
[43] F. Bonet, C. Guꢄry, D. Guyomard, R. Herrera Urbina, K. Tekaia-Elhsissen,
J. M. Tarascon, Int. J. Inorg. Mater. 1999, 1, 47–51.
[44] F. Bonet, C. Guꢄry, D. Guyomard, R. Herrera Urbina, K. Tekaia-Elhsissen,
J. M. Tarascon, Solid State Ionics 1999, 126, 337–348.
[45] Institute of Medicine, Food and Nutrition Board, Dietary Reference In-
takes, National Accademy Press, Washington 2001.
[46] J. Chen, S. K. Spear, J. G. Huddleston, R. D. Rogers, Green Chem. 2005, 7,
64–82.
[1] R. C. Larock, Comprehensive Organic Transformations, Wiley, Chichester,
1999.
[2] Z. Rappoport in Chemistry of the Cyano Group (Ed.: Z. Rappaport), Wiley,
New York, 1970.
[3] G. P. Ellis, T. M. Romney-Alexander, Chem. Rev. 1987, 87, 779–794.
[4] T. Schareina, A. Zapf, M. Beller, J. Organomet. Chem. 2004, 689, 4576–
4583.
[5] K. Chattopadhyay, R. Dey, B. C. Ranu, Tetrahedron Lett. 2009, 50, 3164–
3167.
[6] V. Polshettiwar, P. Hesemann, J. J. E. Moreau, Tetrahedron 2007, 63,
6784–6790.
[7] Y. Ren, Z. Liu, S. He, S. Zhao, J. Wang, R. Niu, W. Yin, Org. Process Res.
Dev. 2009, 13, 764–770.
[8] A. R. Hajipour, K. Karami, G. Tavakoli, A. Pirisedigh, J. Organomet. Chem.
2011, 696, 819–824.
[9] T. D. Senecal, W. Shu, S. L. Buchwald, Angew. Chem. 2013, 125, 10219–
10223; Angew. Chem. Int. Ed. 2013, 52, 10035–10039.
[10] Y. Ren, C. Dong, S. Zhao, Y. Sun, J. Wang, J. Ma, C. Hou, Tetrahedron Lett.
2012, 53, 2825–2827.
[11] Y. Ren, Z. Liu, S. Zhao, X. Tian, J. Wang, W. Yin, S. He, Catal. Commun.
2009, 10, 768–771.
[12] T. Schareina, A. Zapf, M. Beller, Tetrahedron Lett. 2005, 46, 2585–2588.
[13] T. Schareina, A. Zapf, W. Magerlein, N. Muller, M. Beller, Chem. Eur. J.
2007, 13, 6249–6254.
[47] C. Oliver Kappe, Chem. Soc. Rev. 2008, 37, 1127–1139.
[48] J. Santhanalakshmi, L. Parimala, J. Nanopart. Res. 2012, 14, 1090.
[49] X. Bantreil, M. Sidi-Ykhlef, L. Aringhieri, E. Colacino, J. Martinez, F.
Lamaty, J. Catal. 2012, 294, 113–118.
[50] S. Luo, S. Zhang, Y. Wang, A. Xia, G. Zhang, X. Du, D. Xu, J. Org. Chem.
2010, 75, 1888–1891.
[14] Q. Wen, J. Jin, Y. Mei, P. Lu, Y. Wang, Eur. J. Org. Chem. 2013, 4032–
4036.
[51] X. Cui, Y. Zhang, F. Shi, Y. Deng, Chem. Eur. J. 2011, 17, 2587–2591.
[52] F.-T. Luo, A. Jeevanandam, Tetrahedron Lett. 1998, 39, 9455–9456.
[53] R. Spina, E. Colacino, B. Gabriele, G. Salerno, J. Martinez, F. Lamaty, J.
Org. Chem. 2013, 78, 2698–2702.
[15] C. DeBlase, N. E. Leadbeater, Tetrahedron 2010, 66, 1098–1101.
[16] Y. Ren, W. Wang, S. Zhao, X. Tian, J. Wang, W. Yin, L. Cheng, Tetrahedron
Lett. 2009, 50, 4595–4597.
[17] M. Vinatoru, M. Toma, O. Radu, P. I. Filip, D. Lazurca, T. J. Mason, Ultra-
sonics 1997, 4, 135–138.
[18] L. Petigny, S. Perino-Issartier, J. Wajsman, F. Chemat, Int. J. Mol. Sci.
2013, 14, 5750–5764.
Received: September 18, 2013
Revised: December 2, 2013
[19] S. M. Jafari, Y. He, B. Bhandari, J. Food Eng. 2007, 82, 478–488.
Published online on February 12, 2014
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 919 – 924 924