Mechanistic insights into copper(I)-catalyzed azide-alkyne cycloadditions using continuous flow conditions

Article abstract of DOI:10.1002/adsc.200900726

The copper-catalyzed azide-alkyne cycloaddition (CuAAC, "click chemistry") was studied employing copper-in-charcoal (Cu/C) and a variety of copper metal sources as "heterogeneous" catalysts. The type and pretreatment conditions of the different copper sources on the CuAAC were investigated. In addition, the effect of copper leaching from the catalyst over time and in dependence on the reaction mixture composition was studied by ICP-MS analysis in the continuous flow mode. These investigations confirm a "homogeneous" mechanism and suggest surface layer copper(I) oxide as the catalytically active species in CuAAC chemistry involving zerovalent copper metal.

Full text of DOI:10.1002/adsc.200900726

COMMUNICATIONS
DOI: 10.1002/adsc.200900726
Mechanistic Insights into Copper(I)-Catalyzed Azide-Alkyne
Cycloadditions using Continuous Flow Conditions
a
b
c
a,
Michael Fuchs, Walter Goessler, Christian Pilger, and C. Oliver Kappe *
a
Christian Doppler Laboratory for Microwave Chemistry (CDLMC) and Institute of Chemistry, Karl-Franzens-University
Graz, Heinrichstrasse 28, 8010 Graz, Austria
Fax : (+43)-(0)316-380-9840; e-mail: oliver.kappe@uni-graz.at
Institute of Chemistry, Analytical Chemistry, Karl-Franzens-University Graz, Universitꢀtsplatz 1, 8010 Graz, Austria
BASF SE, 67056 Ludwigshafen, Germany
b
c
Received: October 20, 2009; Revised: December 3, 2009; Published online: January 27, 2010
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adsc.200900726.
Abstract: The copper-catalyzed azide-alkyne cyclo-
addition (CuAAC, “click chemistry”) was studied
employing copper-in-charcoal (Cu/C) and a variety
of copper metal sources as “heterogeneous” cata-
lysts. The type and pretreatment conditions of the
different copper sources on the CuAAC were inves-
tigated. In addition, the effect of copper leaching
from the catalyst over time and in dependence on
the reaction mixture composition was studied by
ICP-MS analysis in the continuous flow mode.
These investigations confirm a “homogeneous”
mechanism and suggest surface layer copper(I)
oxide as the catalytically active species in CuAAC
chemistry involving zerovalent copper metal.
Recent preparative synthetic advances in the field
of the CuAAC include the generation of the required
[
3,4]
azides in situ, or alternatively the use of continuous
[
5]
flow processing.
Performing chemical reactions
under flow conditions in micro- or mesofluidic devices
is of growing interest since these conditions offer a
variety of advantages over macroscopic batch reac-
tors, such as reduced consumption of reagents, high
surface-to-volume ratios, and improved control over
[
5]
mass and heat transfer. For the CuAAC, the safety
aspects in combination with the inherent scalability of
continuous flow processing are particularly appealing
since both types of starting materials can – in princi-
[6–9]
ple – be generated in-line using flow chemistry.
In
this communication we report new mechanistic in-
sights into CuAAC chemistry performed both in
batch and under continuous flow conditions using im-
mobilized Cu catalysts in combination with leaching
studies using ICP-MS Cu analysis.
Keywords: 1,3-dipolar cycloaddition; click chemis-
try; copper catalysis; flow chemistry; microwave
chemistry; 1,2,3-triazoles
Our initial experiments were concerned with inves-
tigating the suitability of copper-in-charcoal (Cu/C) as
a catalyst for the CuAAC under continuous flow con-
Since the development of the Cu(I)-catalyzed azide- ditions. The Cu/C system was introduced by Lipshutz
alkyne 1,3-dipolar cycloaddition (CuAAC) independ- and Taft in 2006 as a heterogeneous, inexpensive and
ently by the groups of Medal and Sharpless in 2002, shelf-stable catalyst for CuAAC chemistry, with Cu O
2
this transformation has attracted a significant amount dispersed within the charcoal matrix implicated as the
[
1,2]
[10]
of attention within the scientific community.
The most likely catalytically active species.
Applying
reasons for the popularity of CuAAC are related to microwave heating, near quantitative yields of 1,2,3-
the generated 1,4-substituted-1,2,3-triazole scaffold triazole products were obtained after 3 min at 1508C
and its chemical and biological properties, as well as using dioxane as solvent in the absence of any addi-
[
10]
to the robustness of the cycloaddition process, the tives such as base or a reducing agents. Based on
[
2]
high stereoselectivity and wide substrate scope.
our recent experience with Mizoroki–Heck coupling
Modern applications of the CuAAC are today not chemistry performed with Pd/C in a dedicated high-
[
11]
only widely used in classical organic/combinatorial temperature/pressure flow reactor,
we speculated
synthesis and medicinal chemistry/drug discovery pro- that Cu/C would be a very suitable catalyst for
grams, but have also made a major impact in poly- CuAAC reactions under continuous flow conditions.
mer/material science, and the field of chemical biol-
ogy.
Using the cycloaddition of benzyl azide (1) and
phenylacetylene (2) as a model reaction we initially
[2]
Adv. Synth. Catal. 2010, 352, 323 – 328
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
323

Michael Fuchs et al.
COMMUNICATIONS
repeated the CuAAC using the microwave conditions Table 1. Cu-catalyzed cycloaddition of benzyl azide and phe-
[a]
[
10]
nylacetylene using different Cu(0) sources (Scheme 1).
reported by Lipshutz and Taft. We were pleased to
find that full conversion to the 1,2,3-triazole 3
[b]
Entry Cu source
Conversion
[c]
(
Scheme 1) was obtained under a variety of experi-
[
%]
mental conditions, using different solvents and con-
centrations of starting material. Having continuous
flow processing in mind, we noted that using acetone
as a solvent the reaction not only provided complete
conversion but also remained completely homogene-
ous – apart from the Cu/C catalyst – throughout the
course of the cycloaddition process. Ultimately, suita-
ble processing conditions involved an equimolar mix-
ture of cycloaddition partners in 0.25M concentration,
1
2
3
4
5
turnings (Strem #29-0090)
wire (Strem #93-2972)
6
4
47
powder (À100 mesh, Strem #93-2999)
powder (À140 mesh, Acros #196570250) 98 (92)
powder (À200 mesh, Aldrich #20778-0) 97 (92)
[a]
General conditions: 0.5 mmol benzyl azide (1), 0.55 mmol
phenylacetylene (2), 0.5 mequiv. Cu in 2 mL of acetone,
MW, 1608C, 10 min.
Fresh bottles, used as received.
Calibrated HPLC conversions. Isolated yield in parenthe-
sis.
[
[
b]
c]
10 mol% Cu/C catalyst, acetone as solvent and 1508C
controlled microwave heating for 3 min. These experi-
mental conditions provided a 94% isolated yield of
pure 1,2,3-triazole product 3 by simple filtration from
the catalyst and evaporation of solvent. In addition,
we have studied the Cu/C catalyst system in a one-
pot, three-component modification of the CuAAC
process, whereby azide 1 was generated in situ from
Table 2. Cu-catalyzed cycloaddition of benzyl azide and phe-
nylacetylene using differently treated sources of Cu turnings
[a]
(
Scheme 1).
[b]
[c]
benzyl chloride and sodium azide in a 1:1 t-BuOH/ Entry
Cu Treatment
Conversion [%]
H O mixture. This protocol, originally introduced by
2
1
–
35
11
48
9
Van der Eycken using Cu(I) generated by compropor-
2
conc. HCl
35% H O
conc. HCl
35% H O
tionation from Cu(0) and Cu(II) (Cu turnings/
3
4
5
2
2
[4]
[
[
d]
e]
CuSO ), worked equally well and provided high
4
yields of the desired triazole 3 (MW, 1508C, 10 min,
36
2
2
9
3% yield).
[
a]
General conditions: 0.5 mmol benzyl azide (1), 0.55 mmol
phenylacetylene (2), 2.0 mequiv Cu turnings (Strem #29-
While our work was in progress, an elegant study
by Bogdan and Sach was published in the literature,
where the authors have performed a variety of
CuAAC transformations in the continuous-flow mode
using a 0.75 mm i.d. Cu coil as a reactor. Notably,
without any additives [i.e., a Cu(II) source] successful
cycloadditions were performed at temperatures
0
090) in 2 mL of acetone, MW, 1608C, 10 min.
[b]
Treatment involves 15 min stirring in conc. HCl or 35%
aq. H O solution, followed by acetone washes (2ꢂ
2
2
[
9]
5 mL).
[
[
c]
Calibrated HPLC conversions.
Cu turnings used for the experiment described in entry 3
were dried and reused.
Cu turnings used for the experiment described in entry 2
were dried and reused.
d]
~
1508C using DMF/H O mixtures as solvent system.
2
[e]
With this in mind, we have evaluated the catalytic ac-
tivity of various sources of Cu metal in the CuAAC
reaction. Applying our standard set of conditions
(
MW, 1608C, 10 min) the data shown in Table 1 clear-
ly reveal a strong dependence on surface area were first stirred in concentrated HCl prior to use,
[
13]
(
powder being more effective than turnings or wire), followed by a careful wash with acetone.
Indeed
and on the commercial supplier. In agreement with the conversion in the CuAAC was significantly re-
the generally accepted mechanistic hypotheses con- duced (Table 2, entry 2) compared to untreated Cu
[
2]
cerning the CuAAC reaction, we speculated that turnings (Table 2, entry 1). The opposite effect was
the genuine catalytically active species in the cycload- achieved by pretreatment with 35% H O , a known
2
2
ditions using Cu(0) metal is in fact surface Cu(I). In effective oxidizing agent for Cu metal (Table 2,
[
14]
fact, it is well known that Cu metal itself is oxidized entry 3). To further corroborate our hypothesis we
continuously in the air even at room temperature additionally executed the reversed cycle of washings,
forming non-self-protecting layers of Cu oxides of dif- i.e., previously oxidized (“activated”) Cu turnings
[
12]
ferent oxidation states.
were deactivated with HCl, and HCl-cleaned (“deac-
In order to obtain more detailed information about tivated”) Cu turnings were activated with H O . The
2
2
the catalytically active species in CuAAC reactions results were as expected (Table 2, entries 4 and 5) and
involving zerovalent Cu metal catalysts, samples of confirmed that the catalytically active species in
Cu turnings were pretreated before being subjected CuAAC chemistry involving zerovalent Cu metal is
[12]
to the CuAAC reaction (Table 2). For the purpose of most likely connected to a surface layer of Cu O.
2
eliminating the Cu O/CuO layers, the Cu turnings Thus, employing suitable reaction conditions, it may
2
324
asc.wiley-vch.de
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 323 – 328

Mechanistic Insights into Copper(I)-Catalyzed Azide-Alkyne Cycloadditions
not be necessary to use a Cu(II) source for compro- the conversion in the CuAAC reaction, but instead
portionation purposes in conjunction with the use of led to an enhanced desorption/washing-off of the tri-
[
1,2]
Cu turnings, powders or coils.
ACHTUNGTRENNUaGN zole product from the charcoal support. This kind of
Because of the significantly higher activity and “chromatography” or “adsorption” effect has previ-
better reproducibility of Cu/C compared to the Cu ously been observed in the context of Pd/C-catalyzed
metal catalysts we decided to perform our continu- Mizoroki–Heck coupling chemistry utilizing the same
[
11]
ous-flow experiments with this catalyst. Before start- flow reactor system.
ing flow experimentation a control experiment involv- In the next phase of our experiments the leaching
ing a silicon carbide (SiC) reaction vessel and micro- of Cu from the charcoal support was investigated uti-
wave heating confirmed that the rapid reaction rates lizing ICP-MS. Performing reactions with “heteroge-
achieved under microwave batch conditions were neous” transition metal catalysts under continuous
only related to a bulk temperature effect, and not to flow conditions provides a unique opportunity to gain
[11]
a specific interaction of the electromagnetic field with mechanistic insights otherwise not easily available.
the Cu/C catalyst (see the Supporting Information for In contrast to standard batch processing, leached
[15]
details).
These results ensured that the reaction metal will not simply be readsorbed onto the support
times could be translated to residence times in a con- after the catalytic cycle is completed, but is removed
ventionally heated flow instrument. Flow experiments from the support by the flow process. In the experi-
were subsequently performed in a dedicated high- ments described in Table 3 we have determined the
[
11]
pressure flow reactor (X-Cube, ThalesNano Inc.)
leaching of Cu from a standard X-Cube catalyst car-
where the heterogeneous Cu/C catalyst is immobi- tridge [250 mg, 8.7% (w/w) Cu/C, 22 mg Cu] during
lized in a pre-packed, replaceable stainless steel car- the CuAAC reaction of benzyl azide (1) and phenyl-
tridge (60ꢂ4 mm i.d., ~250 mg Cu/C). The catalyst
ACHTUNGTRENNGNUa cetylene (2) under optimized flow conditions
cartridge can be heated to 2008C and the reaction (Scheme 1). Keeping temperature, pressure, and flow
mixture is pumped through the cartridge using HPLC rate constant, the only changes made were related to
À1
pumps at flow rates of 0.1 to 3.0 mLmin and pres- the composition of the reaction mixture processed
[
11]
sures of up to 150 bar. Since complete reaction ho- through the catalyst cartridge. The results summarized
mogeneity is generally required for performing flow in Table 3 clearly indicate a homogeneous catalysis
[5]
synthesis, the optimized protocol starting from iso- for the CuAAC reaction with the charcoal acting as a
lated benzyl azide (1) and phenylacetylene (2) shown
in Scheme 1 was used for the experiments described
herein.
Optimization of the CuAAC reaction shown in
Scheme 1 demonstrated that full conversion and a
Table 3. Leaching of Cu from Cu/C catalyst for the cycload-
dition of benzyl azide (1) and phenylacetylene (2) under X-
[a]
Cube flow conditions (Scheme 1).
[b]
Entry Reaction mixture composition Leaching [mg Cu]
>
99% isolated triazole yield were readily obtained at
À1
1
2
3
4
5
6
7
acetone
2.1
3.6
59
1708C and a 1.5 mLmin flow rate (20 bar set pres-
acetone/2
acetone/1
acetone/2/1
acetone/Et N
acetone/Et N/2
sure) using a 0.25M concentration of azide 1 in ace-
tone (1.1 equivalents of acetylene 2). Lower reaction
temperatures led to incomplete conversions. Based on
the measured dead volume of ~300 mL of the filled
catalyst cartridge (60ꢂ4 mm i.d., 250 mg of Cu/C), a
residence time over the catalyst of ~12 s can be calcu-
30
38.6
18.7
71
3
3
acetone/3
[a]
À1
Complete reaction mixture composition: 0.25 mmol
lated applying a 1.5 mLmin flow rate (the residence
benzyl azide (1), 0.27 mmol phenylacetylene (2), acetone
time in the complete X-Cube system is ~3 min at this
flow rate). An initial evaluation of the influence of
the flow rate appeared to indicate that higher flow
rates would lead to increased isolated product yields.
More detailed investigations of this effect, however,
revealed that the higher flow rates did not influence
(
2
1 mL), Cu/C catalyst in cartridge (8.7% Cu/C, 250 mg,
2 mg Cu). Conditions: 1708C, 20 bar, 1.5 mLmin flow
À1
rate. A fresh cartridge was used for every run. See the
Supporting Information for details.
b]
[
Determined by ICPMS analysis of the product contained
in a 7.5 mL fraction.
Scheme 1. Cu-catalyzed cycloaddition of benzyl azide and phenylacetylene.
Adv. Synth. Catal. 2010, 352, 323 – 328
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
325

Michael Fuchs et al.
COMMUNICATIONS
À1
reservoir for Cu involving a release-recapture mecha- nation from 600 mgkg
to acceptable levels
[16]
À1
À1
TM
nism. Significant leaching of Cu metal was observed (6 mgkg for charcoal, <1 mgkg for Quadrapure
when the full reaction mixture was passed through TU).
the Cu/C cartridge (Table 3, entry 4). Interestingly, it
In order to obtain information on the lifetime of
was established that the leaching of Cu was minimal the Cu/C catalyst cartridge for larger scale prepara-
when a solution of the terminal alkyne 2 was passed tions, and on the capacity of the charcoal scavenging
through the Cu/C catalyst cartridge (entry 2), compa- cartridge we have performed a scale-up experiment
rable to the background leaching observed with pure using the X-Cube flow reactor. In the event, ~1 L of
solvent (entry 1). In contrast, leaching for the azide reaction mixture (0.25M azide 1+0.27M alkyne 2 in
cycloaddition partner 1 was 20 times higher, indicat- acetone) were processed through the flow reactor
ing a strong complexation of the organic azide with fitted with both a catalyst and a scavenging cartridge
the Cu(I) or Cu(II) species present in the Cu/C over a period of 11 h. Each hour a sample was taken,
matrix (entry 3). These results seem to contradict the which was analyzed for conversion (HPLC-UV), Cu
generally accepted mechanistic hypothesis for the contamination (ICP-MS) and isolated yield. The re-
CuAAC reaction involving a distinct Cu-alkyne com- sults are summarized in Table 4. During the first 6 h
[
2,17]
plex as the key intermediate,
and may here be re- nearly full conversion to 1,2,3-triazole 3 was observed
lated to the base- and ligand-free conditions, the sol- and we were able to isolate 32 g of pure triazole prod-
vent system, or the comparatively high temperatures uct (>99% yield, 98% HPLC purity). After this
used in our studies. Clearly, quite stable complexes period conversion dropped significantly and the iso-
between organic azides and Cu ACHUTNGRNENGU( I/II) ions have been lated crude products were contaminated with substan-
reported in the literature providing support for the tial amounts of both azide and alkyne starting materi-
observed leaching of Cu by azide 1 in our flow experi- als. Disappointingly, all product fractions contained
[18]
ments. In addition, we have performed leaching ex- substantial amounts of Cu metal despite the presence
periments using Et N which is frequently used as base of the Norite scavenging cartridge. Apparently, the
3
in the CuAAC to gain information about a potential scavenging capacity of the Norite charcoal for Cu is
[
1,2]
base-catalyzed pathway.
Indeed, Et N enhances the too low to cope with the amounts of Cu released
3
leaching of Cu from the solid support under our reac- from the Cu/C catalyst cartridge in a large-scale ex-
[
20]
tion conditions (entry 5). The leaching of a mixture of periment. Cu analysis by ICP-MS of the spent Cu/C
alkyne 2 and Et N is in the same magnitude, although catalyst cartridge after 11 h of processing revealed
3
a slightly lower value was obtained (entry 6).
that >60% of the originally impregnated Cu had
For the reaction mixture itself (entry 4) processed been leached (the Cu loading decreased from 8.7 to
on a 0.25-mmol scale (1 mL of reaction mixture in
acetone) 0.14% of the Cu were leached from the Cu/
Table 4. Conversion/yield and leaching of Cu from Cu/C cat-
C catalyst, resulting in a Cu contamination of the tri-
alyst for the cycloaddition of benzyl azide (1) and phenyl-
À1
[a]
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
azole product 3 of ~600 mgg , far above the maxi-
ACHTUNGTRENNUaGN cetylene (2) under X-Cube flow conditions (Scheme 1).
mum allowed Cu contamination in pharmaceuticals of
[b]
À1 [c]
À1 [19]
Entry t [h] Yield [%] Conversion [%]
Cu [mgkg ]
1
5 mgkg .
In order to scavenge the leached Cu
from the reaction mixture, in-line scavenging proto-
cols using either Quadrapure TU (thiourea) resin
or activated charcoal (Norite type A) were evaluated.
For this purpose both scavengers were packed into
empty stainless steel cartridges (60ꢂ4 mm i.d.,
1
2
3
4
5
6
7
8
9
1
1
1
2
3
4
5
6
7
8
93
>99
>99
>99
>99
>99
92
65
59
40
39
297Æ6
260Æ50
199Æ40
179Æ14
156Æ11
162Æ12
87Æ2
TM
[20]
100
100
100
100
100
~
350 mg) and the cartridges placed into the X-Cube
[
d]
99[
reactor in sequence to the Cu/C catalyst cartridge.
d]
d]
TM
99[
96
74Æ1
Quadrapure TU is a functionalized polystyrol resin,
9
10
11
62Æ2
frequently used for metal scavenging via thiol com-
[d]
[d]
0
1
92
95
76Æ1
plexation and has been successfully applied previously
29
47Æ1
[
7,9]
in CuAAC reactions to remove Cu impurities.
Ac-
[a]
Conditions: 0.248 mol benzyl azide (1), 0.267 mol phenyl-
acetylene (2), acetone (990 mL), Cu/C catalyst cartridge
(8.7% Cu/C, 250 mg, 22 mg Cu), Norite A scavenging
tivated charcoal represents a less costly alternative
and possesses substantial metal scavenging proper-
[20]
ties, although in the case of charcoal the scavenging
originates from adsorption of the metal on the sur-
face, thus the interaction is a lot weaker compared to
cartridge
.5 mLmin flow rate. See the Supporting Information
for details.
[b]
(350 mg).
Conditions:
1708C,
20 bar,
À1
1
TM
[20]
the thiol complexation of Quadrapure
TU.
We
Calibrated HPLC conversion.
Determined by ICP-MS analysis of the product. Mean
value of three measurements.
[c]
were pleased to see that on applying the CuAAC
flow reaction conditions given in Table 3 (entry 4)
both scavenging methods decreased the Cu contami-
[d]
Crude product contaminated with starting materials.
326
asc.wiley-vch.de
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 323 – 328

Mechanistic Insights into Copper(I)-Catalyzed Azide-Alkyne Cycloadditions
[
11]
3
.3%). At this point we performed an additional con- tridge holder of the X-Cube flow reactor. The cartridge
was primed with acetone for 10 min at a flow rate of
.5 mLmin , then the instrument was turned on: T=1708C,
.5 mLmin flow rate and p=20 bar. When the set parame-
ters were reached (3–4 min) the reaction mixture, containing
benzyl azide (1) (32 mL, 34 mg, 0.25 mmol) and phenylacety-
lene (2) (30 mL, 28 mg, 0.27 mmol) in acetone (1 mL), was
introduced into the flow system. Subsequent to the introduc-
tion of the reaction mixture, the inlet tube was placed back
trol experiment investigating the complexation prop-
erties of the 1,2,3-triazole product 3 for Cu. As dem-
onstrated in Table 3 (entry 7), the triazole product
itself is capable of coordinating to Cu and leaches Cu
metal from the Cu/C catalyst in substantial amounts.
Therefore, the ineffectiveness of the charcoal-based
scavenging cartridge is not surprising, since the
À1
À1
1
1
formed triazole product will constantly remove Cu into the solvent reservoir. After 20 min of collection the ex-
metal from the charcoal support. It is therefore also periment was stopped and the obtained solution was con-
not practical to use the scavenger cartridge containing centrated under reduced pressure to give pure 1,2,3-triazole
3
as a pale yellow solid; yield: 59 mg (99%); mp 1328C (lit.
adsorbed Cu as a “catalyst cartridge” by reversing the
flow direction. Notably, 1,2,3-triazoles are known to
[23] 1
mp 132–1338C); H NMR (300 MHz, DMSO): d=8.67 (s,
1H), 7.86 (m, 2H), 7.52 (m, 8H), 5.65 (s, 2H); C NMR
(75 MHz, DMSO): d=147.1, 136.5, 131.1, 129.4, 129.3,
1
1
3
[
21]
form strong complexes with Cu(I) salts. Although
TM
the scavenging capacity of Quadrapure TU is three
28.6, 128.4, 125.6, 122.0, 53.5.
For the large-scale run (Table 4), the X-Cube flow reactor
À1
times higher (Norite A: 0.06 mequiv. g , Quadra-
TM
À1 [20]
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
pure TU: 0.17 mequiv. g ) for large-scale CuAAC
was prepared as for the small-scale experiments (vide
supra), but with a scavenging cartridge in the second car-
reactions both methods are clearly unsuitable.
Ultimately, we decided to remove Cu impurities tridge holder containing activated charcoal Norite type A
off-line by a simple washing step with ethylenedi- (350 mg). When the settings were reached the tube was
ACHTUNGTRENNUNG
aminetetraacetic acid (EDTA), a strong complexation changed to the reaction mixture, containing benzyl azide (1)
(
32 mL, 34 g, 250 mmol) and phenylacetylene (2) (30 mL,
agent for Cu species (stability constant=
À1 [22]
28 g, 270 mmol) in acetone (990 mL). For a period of 11 h
each hour the collection flask was changed and 100 mL of
the obtained solution were taken to determine the conver-
sion. The solvent was evaporated to determine the isolated
yield and the product was subjected to ICP-MS analysis.
The product fractions isolated from entries 1–6 (Table 4)
were combined to provide 32 g of triazole 3. The analytical
1
8.80 Lmol ).
For this purpose, the product with
the highest Cu contamination from the scale-up ex-
periment described above (Table 4, entry 1,
À1
2
97 mgkg ) was dissolved in CH Cl and extracted 3
2 2
times with an aqueous EDTA solution. This simple
aqueous washing step reduced the Cu levels to
À1
2
mgkg , thus being almost as efficient as the Quad- and spectral data of this material were as described for the
rapure scavenger.
small-scale experiments.
In conclusion, we have investigated continuous-
flow methods for the Cu(I)-catalyzed cycloaddition of Supporting Information
organic azides and terminal alkynes (CuAAC reac-
General experimental details and further description of ex-
perimental procedures are available as Supporting Informa-
tion.
tion) that rely on the use of copper-in-charcoal (Cu/
C) as an immobilized source of Cu catalyst. By study-
ing different Cu sources and pretreatment methods it
has become clear that the catalytically active species
in CuAAC chemistry involving zerovalent Cu metal is
most likely connected to a surface layer of Cu O. By
performing leaching experiments it was demonstrated
2
Acknowledgements
that a homogeneous mechanism is in operation using This work was supported by BASF AG and a grant from the
Cu/C as a catalyst. Under flow conditions, this leads Christian Doppler Research Society (CDG). We thank
to leaching of Cu from the heterogeneous catalyst. Thales Nanotechnology Inc. for the provision of the X-Cube
reactor and Dr. Toma N. Glasnov and Mag. Bernhard Gut-
mann for experimental assistance.
On a small scale efficient in-line scavenging can be
performed using activated charcoal or Quadrapure
TM
TU resin, while for large-scale experiments an extrac-
tive work-up with EDTA is the method of choice.
References
[
1] a) C. W. Tornøe, C. Christensen, M. Meldal, J. Org.
Chem. 2002, 67, 3057; b) V. V. Rostovtsev, L. G. Green,
V. V. Fokin, B. K. Sharpless, Angew. Chem. 2002, 114,
Experimental Section
2
708; Angew. Chem. Int. Ed. 2002, 41, 2596.
Cu-Catalyzed Cycloaddition of Benzyl Azide (1) and
Phenylacetylene (2) in Continuous Flow Mode
[
2] For reviews, see: a) H. C. Kolb, M. G. Finn, B. K.
Sharpless, Angew. Chem. 2001, 113, 2056; Angew.
Chem. Int. Ed. 2001, 40, 2004; b) V. D. Bock, H. Hiem-
stra, J. H. van Maarseveen, Eur. J. Org. Chem. 2006, 51;
c) P. Wu, V. V. Fokin, Aldrichimica Acta 2007, 40, 7;
(
Scheme 1)
A CatCartridge packed with Cu/C [60ꢂ4 mm i.d., ~250 mg
Cu/C, 8.7% (w/w) loading, 0.34 mmol] was placed in the car-
Adv. Synth. Catal. 2010, 352, 323 – 328
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
327

Michael Fuchs et al.
COMMUNICATIONS
d) J. E. Moses, A. D. Moorhouse, Chem. Soc. Rev. 2007,
b) J. E. Boggio, J. Chem. Phys. 1979, 70, 5054; S.
Suzuki, Y. Ishikawa, M. Isshiki, Y. Waseda, Mater.
Trans. JIM 1997, 38, 1004.
3
6, 1249; e) M. Meldal, C. W. Tornøe, Chem. Rev. 2008,
08, 2952.
1
[
3] a) K. Kacprzak, Synlett 2005, 943; b) J. Anderson, U.
Madsen, F. Bjçrkling, X. Liang, Synlett 2005, 2209; c) J.
Anderson, S. Bolvig, X. Liang, Synlett 2005, 2941;
d) H. S. G. Beckmann, V. Wittmann, Org. Lett. 2007, 9,
[13] K. N. Chen, C. S. Tan, A. Fan, R. Reif, Appl. Phys.
Lett. 2005, 86, 11903.
[14] S. Aksu, L. Wang, F. M. Doyle, J. Electrochem. Soc.
2003, 150, G718-G723 .
1
; e) P. Ko cˇ alka, N. K. Andersen, F. Jensen, P. Nielsen,
[
15] a) D. Obermayer, B. Gutmann, C. O. Kappe, Angew.
ChemBioChem 2007, 8, 2106; f) A. D. Moorhouse, J. E.
Moses, Synlett 2008, 2089.
Chem. 2009, 121, 8471; Angew. Chem. Int. Ed. 2009, 48,
8
321; b) see also: M. Irfan, M. Fuchs, T. N. Glasnov,
[4] P. Appukkuttan, W. Dehaen, V. V. Fokin, E. Van der
C. O. Kappe, Chem. Eur. J. 2009, 15, 11608.
Eycken, Org. Lett. 2004, 6, 4223.
[
16] B. H. Lipshutz, J. B. Unger, B. R. Taft, Org. Lett. 2007,
[
5] For selected reviews on continuous flow processing in
organic synthesis, see: a) G. Jas, A. Kirschning, Chem.
Eur. J. 2003, 9, 5708; b) P. Watts, S. J. Haswell, Drug
Discovery Today 2003, 8, 586; c) I. R. Baxendale, M. R.
Pitts Chim. Oggi 2006, 24, 41; d) A. Kirschning, W. So-
lodenko, K. Mennecke, Chem. Eur. J. 2006, 12, 597;
e) A. Chighine, G. Sechi, M. Bradley, Drug Discovery
Today 2007, 12, 459; f) B. Ahmed-Omer, J. C. Brandt,
T. Wirth, Org. Biomol. Chem. 2007, 5, 733; g) C. Wiles,
P. Watts Eur. J. Org. Chem. 2008, 1655.
6] C. O. Kappe, E. Van der Eycken, Chem. Soc. Rev. 2009,
DOI: 10.1039/b901973c.
7] a) C. D. Smith, I. R. Baxendale, S. Lanners, J. J. Hay-
ward, S. C. Smith, S. V. Ley, Org. Biomol. Chem. 2007,
9
, 1089.
[
17] F. Himo, T. Lovell, R. Hilgraf, V. V. Rostovtesev, L.
Noodleman, K. B. Sharpless, V. V. Fokin, J. Am. Chem.
Soc. 2005, 127, 210.
[
[
18] a) M. Barz, E. Herdtweck, W. R. Thiel, Angew. Chem.
1
998, 110, 2380; Angew. Chem. Int. Ed. 1998, 37, 2262;
b) H. V. R. Dias, S. A. Polach, S.-K. Goh, E. F. Archi-
bong, D. S. Marynick, Inorg. Chem. 2000, 39, 3894.
19] EU Agency, Note for Guidance on Specification Limits
For Residues of Metal Catalysts, Evaluation of Medi-
cines for Human Use. The European Agency for Evalu-
ation of Medicinal Products, London, UK, 2002.
[
[
[
20] A. Hinchcliffe, C. Hughes, D. A. Pears, M. R. Pitts,
Org. Process Res. Dev. 2007, 11, 477.
5
, 1559; b) C. D. Smith, I. R. Baxendale, G. K. Tranmer,
[
21] a) T. R. Chan, R. Hilfgraf, K. B. Sharpless, V. V. Fokin,
Org. Lett. 2004, 6, 2853; b) J. Xiang, Y.-G. Yin, P. Mei,
Inorg. Chem. Commun. 2007, 10, 1168; c) U. Monko-
wius, S. Ritter, B. Kçnig, M. Zabel, H. Yersin, Eur. J.
Inorg. Chem. 2007, 4597; d) Z.-B. Zheng, R.-T. Wu, J.-
K. Li, Y.-F. Sun, J. Mol. Struct. 2009, 928, 78; e) S.
Zubuku, E. Ozkal, C. Jimeno, M. A. Perics, Org. Lett.
M. Baumann, S. C. Smith, R. A. Lewthwaite, S. V. Ley,
Org. Biomol. Chem. 2007, 5, 1562; c) I. R. Baxendale,
S. V. Ley, A. C. Mansfield, C. D. Smith, Angew. Chem.
2
009, 121, 4077; Angew. Chem. Int. Ed. 2009, 48, 4017.
[8] a) J. Wang, G. Sui, V. P. Mocharla, R. J. Lin, M. E.
Phelps, H. C. Kolb, H.-R. Tseng, Angew. Chem. 2006,
1
18, 5402; Angew. Chem. Int. Ed. 2006, 45, 5276; b) Y.
2
009, 11, 4680.
22] a) A. Gissibl, M. G. Finn, O. Reiser, Org. Lett. 2005, 7,
325; b) E. J. Baran, C. C. Wagner, M. H. Torre, J.
Wang, W.-Y. Lin, K. Liu, R. J. Lin, M. Selke, H. C.
Kolb, N. Zhang, X.-Z. Zhao, M. E. Phelps, C. K. F.
Shen, K. F. Faull, H.-R. Tseng, Lab Chip 2009, 9, 2281.
9] A. R. Bogdan, N. W. Sach, Adv. Synth. Catal. 2009, 351,
[
2
Braz. Chem. Soc. 2002, 13, 576; c) C. M. Klemm, A.
Berthelmann, S. Neubacher, C. Arenz, Eur. J. Org.
Chem. 2009, 2788; d) B. Macꢃas, I. Garcꢃa, M. V. Villa,
J. Borrꢄs, M. Gonzꢄlez-ꢅlvarez, A. Castineiras, J.
Inorg. Biochem. 2003, 96, 367.
[
8
49.
10] B. H. Lipshutz, B. R. Taft, Angew. Chem. 2006, 118,
415; Angew. Chem. Int. Ed. 2006, 45, 8235.
11] T. N. Glasnov, S. Findenig, C. O. Kappe, Chem. Eur. J.
009, 15, 1001.
[
[
[
8
2
[23] N. Candelon, D. Lastꢆcouꢇres, A. K. Diallo, J. R. Ara-
nzaes, D. Astruc, J.-M. Vincent, Chem. Commun. 2008,
6, 741.
12] a) J. Iijima, J.-W. Lim, S.-H. Hong, S. Suzuki, K.
Mimura, M. Isshiki, Appl. Surf. Sci. 2006, 253, 2825;
328
asc.wiley-vch.de
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 323 – 328