Cycloaddition reactivity studies of first-row transition metal-azide complexes and alkynes: An inorganic click reaction for metalloenzyme inhibitor synthesis

Article abstract of DOI:10.1039/c2dt30145h

The studies described herein focus on the 1,3-dipolar cycloaddition reaction between first-row transition metal-azide complexes and alkyne reagents, i.e. an inorganic variant of the extensively used "click reaction". The reaction between the azide complexes of biologically-relevant metals (e.g., Fe, Co and Ni) found in metalloenzyme active sites and alkyne reagents has been investigated as a proof-of-principle for a novel method of developing metalloenzyme triazole-based inhibitors. Six Fe, Co and Ni mono-azide complexes employing salen- and cyclam-type ligands have been synthesized and characterized. The scope of the targeted inorganic azide-alkyne click reaction was investigated using the electron-deficient alkyne dimethyl acetylenedicarboxylate. Of the six metal-azide complexes tested, the Co and Ni complexes of the 1,4,8,11-tetrametyl-1,4,8,11-tetraazacyclotetradecane (Me ₄cyclam) ligand showed a successful cycloaddition reaction and formation of the corresponding metal-triazolate products, which were crystallographically characterized. Moreover, use of less electron deficient alkynes resulted in a loss of cycloaddition reactivity. Analysis of the structural parameters of the investigated metal-azide complexes suggests that a more symmetric structure and charge distribution within the azide moiety is needed for the formation of a metal-triazolate product. Overall, these results suggest that a successful cycloaddition reaction between a metal-azide complex and an alkyne substrate is dependent both on the ligand and metal oxidation state, that determine the electronic properties of the bound azide, as well as the electron deficient nature of the alkyne employed.

Full text of DOI:10.1039/c2dt30145h

View Article Online / Journal Homepage / Table of Contents for this issue
This paper is published as part of a Dalton Transactions themed issue entitled:
New Talent: Americas
Guest Editors: John Arnold, Dan Mindiola, Theo Agapie,
Jennifer Love and Mircea Dincă
Published in issue 26, 2012 of Dalton Transactions
Image reproduced with permission of Richard L. Brutchey
Articles published in this issue include:
Synthesis and reactivity of 2-azametallacyclobutanes
Alexander Dauth and Jennifer A. Love
Dalton Trans., 2012, DOI: 10.1039/C2DT30639E
Perceiving molecular themes in the structures and bonding of intermetallic phases:
the role of Hückel theory in an ab initio era
Timothy E. Stacey and Daniel C. Fredrickson
Dalton Trans., 2012, DOI: 10.1039/C2DT30298E
Cycloruthenated sensitizers: improving the dye-sensitized solar cell with classical
inorganic chemistry principles
Kiyoshi C. D. Robson, Paolo G. Bomben and Curtis P. Berlinguette
Dalton Trans., 2012, DOI: 10.1039/C2DT30825H
Visit the Dalton Transactions website for more cutting-edge inorganic chemistry
www.rsc.org/dalton

Dynamic ArtVicielweALritnicklesOnline
Dalton
Transactions
Cite this: Dalton Trans., 2012, 41, 8010
www.rsc.org/dalton
PAPER
Cycloaddition reactivity studies of first-row transition metal–azide complexes
and alkynes: an inorganic click reaction for metalloenzyme inhibitor
synthesis†
Emi Evangelio,a Nigam P. Rathb and Liviu M. Mirica*a
Received 19th January 2012, Accepted 19th March 2012
DOI: 10.1039/c2dt30145h
The studies described herein focus on the 1,3-dipolar cycloaddition reaction between first-row transition
metal–azide complexes and alkyne reagents, i.e. an inorganic variant of the extensively used “click
reaction”. The reaction between the azide complexes of biologically-relevant metals (e.g., Fe, Co and Ni)
found in metalloenzyme active sites and alkyne reagents has been investigated as a proof-of-principle for
a novel method of developing metalloenzyme triazole-based inhibitors. Six Fe, Co and Ni mono-azide
complexes employing salen- and cyclam-type ligands have been synthesized and characterized. The scope
of the targeted inorganic azide–alkyne click reaction was investigated using the electron-deficient alkyne
dimethyl acetylenedicarboxylate. Of the six metal–azide complexes tested, the Co and Ni complexes of
the 1,4,8,11-tetrametyl-1,4,8,11-tetraazacyclotetradecane (Me4cyclam) ligand showed a successful
cycloaddition reaction and formation of the corresponding metal–triazolate products, which were
crystallographically characterized. Moreover, use of less electron deficient alkynes resulted in a loss of
cycloaddition reactivity. Analysis of the structural parameters of the investigated metal–azide complexes
suggests that a more symmetric structure and charge distribution within the azide moiety is needed for the
formation of a metal–triazolate product. Overall, these results suggest that a successful cycloaddition
reaction between a metal–azide complex and an alkyne substrate is dependent both on the ligand and
metal oxidation state, that determine the electronic properties of the bound azide, as well as the electron
deficient nature of the alkyne employed.
Introduction
The use of fast, irreversible reactions appropriately called “click
reactions“ is a simple and popular approach to the synthesis of
functionalized molecules by joining two fragments together.1
These click reactions proceed under mild conditions with quanti-
tative yields and high regioselectivity. Examples of such reac-
tions include nucleophilic ring opening (with epoxides and
aziridines), non-aldol type carbonyl reactions (formation of
hydrazones and heterocycles), or additions to carbon–carbon
multiple bonds (i.e., Michael additions or oxidative formation of
epoxides).1 From this last group, the Huisgen 1,3-dipolar
cycloadditions are the most commonly used click reactions.2
These cycloaddition reactions involve a 1,3-dipole (i.e., an
Scheme 1 (i) Huisgen 1,3-dipolar cycloaddition; (ii) Cu-catalyzed
azide) and an unsaturated dipolarophile (e.g., an alkyne,
azide–alkyne cycloaddition (CuAAC).
Scheme 1, i). Among various types of 1,3-dipoles, organic
azides are particularly important as they provide an entry into
aDepartment of Chemistry, Washington University in St. Louis, One
Brookings Drive, St. Louis, Missouri 63130-4899, USA. E-mail:
the synthesis of triazoles and tetrazoles.2 These heterocycle
derivatives have found use in important applications such as
pharmaceuticals, chemical biology,3 or energetic materials.4
Since the first report of a regioselective 1,3-dipolar cyclo-
addition reaction in 2002,5 tremendous scientific efforts have
been devoted to studying these click reactions. The Cu-catalyzed
mirica@wustl.edu; Fax: (+1)-314-935-4481
bDepartment of Chemistry and Biochemistry and Center for
Nanoscience, University of Missouri-St. Louis, One University
Boulevard, St. Louis, Missouri 63121-4400, USA
†CCDC reference numbers 864402–864408. For crystallographic data
in CIF or other electronic format see DOI: 10.1039/c2dt30145h
8010 | Dalton Trans., 2012, 41, 8010–8021
This journal is © The Royal Society of Chemistry 2012

View Article Online
azide–alkyne cycloaddition reaction (CuAAC) (Scheme 1, ii)
occurs in various solvents at room temperature and the 1,4-sub-
stituted isomer is obtained regioselectively. All these character-
istics make the CuAAC an excellent prototype for the click
reaction.
The catalyst-free AAC reaction has also led to the develop-
ment of “in situ click reactions” in enzyme inhibitor design for
drug discovery,6 in which the targeted enzymes are allowed to
assemble new inhibitors by linking azides and alkynes that bind
to adjacent sites on the protein.7,8 The increase in their local con-
centration likely accelerates the cycloaddition reaction, a high-
affinity inhibitor being formed (Scheme 2a).7 Thus, the enzyme
itself, serving as the reaction vessel, promotes the synthesis of its
highest-affinity inhibitor. In contrast, although metal ions are
used as catalysts for the cycloaddition reaction,5 the reaction
between a metalloenzyme–azide adduct and an alkyne-derived
substrate has not been explored for metalloenzymes
(Scheme 2b).9
Scheme 3 Reported reactions between metal–azide complexes and
electron-deficient alkynes.
In this context, we are interested in studying the cycloaddition
reaction between a metal–azide complex and an alkyne (i.e., an
“inorganic” click reaction) as a proof-of-concept for the enzyme-
templated reaction.10 Performing this reaction at the active site of
a metalloenzyme and using an alkyne substrate analogue should
lead to a novel approach for the synthesis of high-affinity inhibi-
tors. A successful cycloaddition reaction would lead to formation
of the triazole product, which can be detected upon protein de-
naturation. The identified triazoles can then be independently
synthesized through a non-enzymatic methodology and used for
enzyme inhibition studies.
Scheme 4 Targeted synthesis of metal–triazolate complexes through
the cycloaddition of metal–azide complexes and alkynes. The ligands
employed are shown in Scheme 5.
A survey of the literature reveals that reactions between metal–
azide complexes and electron deficient alkynes have been
observed for a few neutral Co11 and Ni12 coordination com-
pounds, as well as several organometallic complexes of Ta,13
Ru,14–16 Fe,15 Mn,17 Au,18 Os,19 and Mo20 (Scheme 3). In
addition, a “fully” inorganic click reaction between a metal–
azide and a metal–acetylide has been reported recently.21 By
contrast, such cycloaddition reactions for biomimetic metal com-
plexes with N- and O-donor ligand systems are much more
rare.11a,12 In this context, several Fe–, Co– and Ni–azide
Scheme 5 Biomimetic ligands employed in this study.
complexes that are structural models of metalloenzyme active
sites22 have been synthesized and their cycloaddition reactivity
toward alkynes has been investigated (Schemes 4 and 5). The
reaction of these azide complexes with different alkynes was
studied to determine the scope of the cycloaddition chemistry
with respect to the metal coordination geometry and oxidation
state, as well as the electronic properties of the alkyne reagent.
Results and discussion
Synthesis of salen-type ligands
The metal–azide complexes employing salen-type ligands have
been synthesized following literature procedures and employing
two
ligands,
N,N′-bis(salicylidene)-1,2-ethylenediimine
(H2salen1) and N,N′-bis(salicylidene)-1,2-propylenediimine
(H2salen2, Scheme 5). H2Salen1 and H2Salen2 were synthesized
by condensation of 2 equiv. salicylaldehyde and 1 equiv. ethy-
lenediamine or 1,2-diaminopropane,23 respectively, in ethanol
over a period of 2 h.
Scheme 2 (a) Use of the azide–alkyne cycloaddition (AAC) reaction
to synthesize a high-affinity inhibitor from two inhibitors with different
binding sites (ref. 7); (b) an inorganic azide–alkyne cycloaddition reac-
tion proposed herein as a novel metalloenzyme inhibitor design strategy.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 8010–8021 | 8011

View Article Online
Scheme 6 Synthetic method for the synthesis of metal–azide com-
plexes with H2Salen1 and H2Salen2.
A solution of H2Salen1 was mixed with equimolar solutions
of cobalt acetate in ethanol and sodium azide in water
(Scheme 6), and slow evaporation of the filtered solution led to
crystallization of [(salen1)CoIII(N3)]2, 1 (Table 1). When the
same reaction is performed with H2Salen2 and iron nitrate
(Scheme 6), crystals of [(salen2)FeIII(N3)]2, 2 were obtained.
Similarly, the [(salen1)FeIII(N3)]2 and [(salen2)CoIII(N3)]2 com-
plexes were synthesized and characterized by ESI-MS, however
no pure crystalline products could be obtained. Interestingly, for-
mation of a mononuclear [(salen2)FeIIIN3] complex was reported
previously under similar reaction conditions,23a although the
room-temperature X-ray structure for that complex reveals inter-
molecular Fe–O interactions analogous to those observed for 2.
Structures of [(salen1)CoIII(N3)]2 (1) and [(salen2)FeIII(N3)]2 (2)
The asymmetric unit of both 1 and 2 consists of one-half of the
dinuclear structure, the other half being generated by projection
through a crystallographic inversion centre. Fig. 1 shows the
molecular structure and atom labelling of both complexes. The
two MIII ions of the two dinuclear complexes are bridged by two
phenolate μ-O atoms. The octahedral coordination of each MIII
ion is completed by the tetradentate Schiff-base ligand and a
−
−
terminal N3 ion. The two N3 ligands are trans relative to the
M–M vector, as imposed by the crystallographic inversion
centre.
The M1–O2i bond is longer than the M1–O2 bond (Table 2)
is likely due to either the greater trans influence of the N3− com-
pared to the imine N atom23b or a better orbital overlap of the O2
atom with the metal centre. The M1–N3 distances (1.932 Å for
cobalt and 2.015 Å for iron complexes, respectively) are in the
range of the other MIII complexes reported in the literature.23a,24
The M1–O2 bond is longer than the M1–O1 bond due to the
interaction of O2 with M1i. The average M–N(imine) bond
length is 1.883 and 2.103 Å for cobalt and iron complexes,
respectively. The three trans angles at the MIII atom are between
158.6 and 177.6°. The distorted octahedral geometry of the
8012 | Dalton Trans., 2012, 41, 8010–8021
This journal is © The Royal Society of Chemistry 2012

View Article Online
varies from square pyramidal for the nickel complex (τ value of
∼0) to a more distorted trigonal bipyramidal geometry for the
cobalt complex (τ value of 0.45).29
For first-row transition metals, the coordination chemistry of
monofunctionalized cyclam derivatives has mainly focused on
complexes of NiII, CuII and CoII.30 In 2000, Wieghardt et al.,
described the synthesis and characterization of a monofunctiona-
lized FeIII complex of 1,4,8,11-tetraazacyclotetradecane-1-acetic
acid, cyclamAcH, as a precursor for the synthesis of high-valent
Fe complexes.22 As described therein, the cyclamAcH ligand
provides upon deprotonation a monoanionic pentacoordinate
framework for the synthesis of a series of complexes with a
single, variable coordination site. Thus, the cyclamAcH ligand is
particularly appealing for our studies. In addition, the presence
of a carboxylate group makes the corresponding metal com-
plexes more similar to the coordination environment found in
non-heme iron enzymes, where a carboxylate group coordinates
to the Fe centre.31 The synthesis of the cyclamAcH complexes
was performed as described by Wieghardt et al. with a slight
variation (Scheme 8). Ferric chloride was treated with cycla-
mAcH in a refluxing aqueous solution for 90 min. After cooling
this solution at room temperature and addition of an excess of
sodium azide, a clear red solution formed after 2 h. Addition of
Fig. 1 View of the molecular structure and atom labelling of (left)
[(salen1)CoIII(N3)]2 (1) and (right) [(salen2)FeIII(N3)]2 (2). Hydrogen
atoms have been omitted for clarity.
metal centre is also suggested by the cis angles that vary from
77.7 to 107.5° in the case of the Fe complex, and from 85.5° to
95.0° for the Co complex. As a result of the centre of symmetry,
the M2O2 core of both complexes is perfectly planar.
Synthesis of cyclam-type ligands
The cyclam-type ligands were chosen based on their ability to
stabilize mononuclear metal–azide complexes that are more
structurally similar to the metalloenzyme active sites. As such,
an excess of KPF6 yields
a microcrystalline solid of
[(cyclamAc)FeIII(N3)]PF6 (5) upon cooling at 4 °C overnight.
This compound was fully characterized by ESI-MS, UV-vis and
IR spectroscopy, and the results are in full agreement with the
compound previously described.22
1,4,8,11-tetrametyl-1,4,8,11-tetraazacyclotetradecane
(Me4cy-
clam) and 1,4,8,11-tetraazacyclotetradecane-1-acetic acid (cycla-
mAcH) were chosen for the synthesis of metal–azide complexes
(Scheme 5). The synthesis of Me4cyclam (Scheme 7) has been
successfully accomplished following a literature procedure25 to
obtain the ligand in high yield. The isolated product is highly
hygroscopic and was dried under vacuum for 2 h at 50 °C before
the metal complexation reaction. The synthesis of the cycla-
mAcH ligand was also accomplished by employing a modified
version of a published procedure (Scheme 7),26 a higher yield of
the product being obtained by extending the reaction time to 4 h,
washing with CHCl3, and treating the final product with HCl
(vide infra).
When the same reaction is performed in the presence of Co-
(ClO4)·6H2O, a dark violet powder is obtained. Crystals of
[(cyclamAc)CoIII(N3)]ClO4 (6) were obtained by slow diffusion
of diethyl ether in an acetonitrile solution. A summary of the
X-ray experimental data is provided in Table 1. A representation
of the [(cyclamAc)CoIII(N3)]+ cation is shown in Fig. 3, with
selected bond distances and angles listed in Table 2.
The CoIII ion sits in a compressed octahedral environment
with three of the four ligand N atoms (i.e., N1, N4 and N3)
occupying the equatorial positions along with the carboxylate O
atom (Fig. 3); the fourth N atom (N2) occupies one of the axial
sites with the azide ligand situated trans to the N2 atom. By con-
trast to the trans-III ligand configuration in 5,22 the cyclamAc
ligand adopts the cis-V28 configuration in 6, which implies a dra-
matic configuration change. The complex exhibits a longer Co–
N2 distance (1.994 Å) compared to the Co–N1 (1.962 Å) and
Co–N3 (1.982 Å) distances, as expected for the slightly stronger
trans influence of the carboxylate group. The Co–N3 bond dis-
tance (1.982 Å) is slightly longer than the Co–N1 distance
(1.962 Å), likely due to the presence of the acetate arm bonded
to N3. The Co–Nazide and Co–Oacetate distances are 1.936 and
1.899 Å, similar to those reported for the iron complex.21 To the
best of our knowledge, complex 6 represents the first example of
a cyclamAc metal complex where the macrocyclic ligand adopts
the less-common cis-V configuration.32
Synthesis and structures of metal–azide complexes derived from
cyclam-type ligands
As described in the literature,27 mononuclear metal–azide com-
plexes derived from the Me4cyclam ligand were obtained for Co
and Ni: [(Me4cyclam)NiII(N3)]+ (3) and [(Me4cyclam)CoII(N3)]+
(4), respectively. Mixing of 1 equiv. ligand with 1 equiv. Ni or
Co perchlorate in ethanol, followed by addition of 5 equiv.
aqueous NaN3 and heating at 70 °C for 1 h led to formation of
green or violet crystals upon standing at low temperature. A
summary of the X-ray experimental data is provided in Table 1.
The crystal structure of these compounds (Fig. 2) and the
bond lengths and angles (Table 2) are similar to those previously
reported.27 The cyclam adopts the trans-I configuration as
described by Bosnich et al.,28 which is the most favoured
configuration for this kind of complexes. The four methyl substi-
tuents of the ring are positioned on the same side of the molecule
as the azide ligand, which fits into the cavity formed by the
methyl groups. The coordination geometry of the metal centre
IR characterization of the metal–azide complexes
For all metal–azide complexes synthesized, their characteristic
azide stretches were measured by IR spectroscopy. Spectra were
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 8010–8021 | 8013

View Article Online
Table 2 Selected bond distances (Å) and angles (°) for [(salen1)CoIII(N3)]2 (1), [(salen2)FeIII(N3)]2 (2), [(Me4cyclam)NiII(N3)]PF6 (3),
[(Me4cyclam)CoII(N3)]ClO4 (4), [(cyclamAc)CoIII(N3)]ClO4 (6), {(Me4cyclam)NiII[N3C2(CO2Me)2]}ClO4 (7) and [{(Me4cyclam)-
CoII[N3C2(CO2Me)2]}ClO4 (8)
1
2
3
4
6
7
8
M1–N1
M1–N2
M1–N3
M1–O1
M1–O2
M1–O2i
O2–M1i
N3–N4
N4–N5
1.8738(13)
1.8916(14)
1.9319(14)
1.8798(11)
1.9333(11)
2.0217(11)
2.0216(11)
1.209(2)
2.087(4) M1–N1
2.118(4) M1–N1i
2.015(4) M1–N2
1.881(3) M1–N2i
1.981(3) M1–N3
2.169(3) N3–N4
2.169(3) N5–N5
1.209(5)
2.1959(8)
2.1295(8)
2.1323(8)
2.1323(8)
1.9887(12)
1.1805(18)
1.153(2)
2.1192(15) Co–N1
2.1136(15) Co–N2
2.2184(15) Co–N3
2.2357(15) Co–N4
1.9742(13) Co–N5
1.962(2)
1.994(2)
1.982(2)
1.972(2)
1.936(2)
1.8995(11) N3–N4
1.283 (3) N3–N4i
1.224(3)
M1–N1
M1–N1i
M1–N2
M1–N2i
M1–N3
2.1131(12)
2.1131(12)
2.1472(13)
2.1472(13)
2.0017(17)
1.3304(15)
1.3304(15)
2.1990(19)
2.1990(19)
2.1292(18)
2.1292(18)
2.030(3)
1.183(2)
1.160(2)
Co–O1
C12–O1
C12–O2
N5–N6
N6–N7
1.326(2)
1.326(2)
1.150(2)
1.132(6)
1.208(3)
1.151(3)
O2–M1–O1 88.01(5)
O2–M1–N2 91.64(5)
O1–M1–N2 177.65(5)
O2–M1–N1 174.22(5)
O1–M1–N1 95.04(5)
N2–M1–N1 85.51(6)
O2–M1–O2i 81.46(5)
O1–M1–O2i 93.08(5)
N2–M1–O2i 89.17(5)
N1–M1–O2i 93.46(5)
O2–M1–N3 92.32(5)
O1–M1–N3 89.49(6)
N2–M1–N3 88.20(6)
N1–M1–N3 92.62(6)
O1i–M1–N3 173.15(5)
M1–N3–N4 117.20(11)
N3–N4–N5 175.73(17)
107.51(13) N1–M1–N1i 94.24(5)
84.40(14) N2–M1–N2i 91.68(5)
165.83(14) N1–M1–N2 84.86(3)
158.61(15) N1i–M1–N2i 84.85(3)
88.56(15) N2i–M1–N1 164.09(3)
78.08(16) N2–M1–N1i 164.09(3)
77.75(12) N2–M1–N3 94.32(4)
91.48(13) N2i–M1–N3 94.32(4)
83.43(13) N1–M1–N3 101.42(3)
88.05(13) N1i–M1–N3 101.42(3)
94.36(14) M1–N3–N4 133.24(12)
95.53(15) N3–N4–N5 177.4(2)
91.09(15)
93.34(6)
92.11(6)
83.27(6)
83.78(6)
141.50(6)
168.63(5)
94.18(6)
N4–Co–N1 85.79(9)
N4–Co–N3 94.33(9)
N3–Co–O1 87.29(8)
N1–Co–O1 92.61(9)
N1–Co–N3 179.42(9)
N4–Co–O1 177.21(8)
N2–Co–N5 175.68(10)
Co–N5–N6 119.03(18)
N5–N6–N7 175.4(3)
N1–M1–N1i 147.39(8)
N2–M1–N2i 171.73(8)
N1–M1–N2 85.13(5)
N1i–M1–N2i 85.13(5)
N2i–M1–N1 92.54(5)
N2–M1–N1i 92.54(5)
N2–M1–N3 94.14(4)
N2i–M1–N3 94.14(4)
N1–M1–N3 106.31(4)
N1i–M1–N3 106.31(4)
167.65(12)
140.94(12)
83.91(8)
83.91(8)
91.96(8)
91.96(8)
109.53(6)
109.53(6)
96.17(6)
96.17(6)
97.19(6)
109.03(7)
109.42(7)
134.22(12)
178.13(19)
98.12(15)
170.76(14)
124.6(3)
176.4(5)
measured as thin films obtained from dichloromethane solutions
on a NaCl plate. The azide stretching frequencies, ν(N3)asym, for
the dinuclear complexes into mononuclear species in solution
needs to be considered; however, the weak axial metal–O inter-
actions (vide supra) are not expected to have an appreciable
effect on the rate of the azide–alkyne cycloaddition reaction for
the dinuclear vs. mononuclear species.
all these complexes were found between 2015 and 2070 cm−1
typical for such azide adducts (Table 3).
,
Interestingly, treatment of 1 equiv. of [(Me4cyclam)NiII(N3)]-
PF6 (3) with 5 equiv. of dimethyl acetylenedicarboxylate under
stirring in CHCl3 at RT shows the complete disappearance of the
azide peak (2070 cm−1) after 6 h and the appearance of stretches
at 1726 cm−1 (ν(CvO)), 1480 cm−1 (ν(NvN)),19 1270 cm−1
(ν(C–O)), and triazolate ring vibrations at 826, 806 and
778 cm−1, corresponding to the formation of a triazolate species
(Fig. 4).11 Crystals of the product {(Me4cyclam)NiII[N3C2-
(CO2Me)2]} ClO4 (7) were obtained by slow diffusion of hexane
into a dichloromethane solution. Similar results were obtained
for [(Me4cyclam)CoII(N3)]ClO4 (4), which was treated with 5
equiv. dimethyl acetylenenedicarboxylate in chloroform to lead
to the disappearance of the azide stretch at 2070 cm−1 and
appearance of stretches at 1728, 1479, 1261 cm−1, and triazolate
ring vibrations at 822, 805 and 779 cm−1, respectively.11a,19
Crystals of {(Me4cyclam)CoII[N3C2(CO2Me)2]}ClO4 (8) were
obtained by slow diffusion of hexane into a dichloromethane sol-
ution of the isolated product.
Reaction of metal–azide complexes with dimethyl
acetylenedicarboxylate
A summary of the cycloaddition reactions between the syn-
thesized metal–azide complexes and dimethyl acetylenedicar-
boxylate is shown in Scheme 9. Typically, 1 equiv. of metal–
azide complex was treated with an excess of dimethyl acetylene-
dicarboxylate (4–5 equiv.) in chloroform at room temperature or
acetonitrile at 50 °C, and the reaction was followed by IR spec-
troscopy. The acetonitrile reaction at 50 °C was performed for
the complexes where no reaction was observed in chloroform at
RT. In the cases when the 1,3-dipolar cycloaddition occurs, the
formation of the N(2)-bound 4,5-bis(methoxycarbonyl)-1,2,3-
triazolate complex was confirmed by the disappearance of the
ν(N3)asym stretch in the IR spectrum and the appearance of sharp
peaks assigned to the stretching frequencies of the CvO, NvN
and C–O bonds, and ring vibrations of the triazolate product, as
reported previously.11a,19
By contrast, when the (cyclamAc)Fe– and (cyclamAc)Co–
azide complexes 5 and 6 were treated with an excess of dimethyl
acetylenedicarboxylate, no disappearance of the azide stretch
at 2052 cm−1 was observed even after 5 days, either in aceto-
nitrile or chloroform at RT or 50 °C. Overall, the lack of a
cycloaddition reaction suggests that subtle changes in the
electronic properties and ligand environment of the metal centre
When the cycloaddition reaction was attempted for the (salen)
Co– and (salen)Fe–azide complexes 1 and 2, no decrease in the
intensity of the azide stretching frequency was observed even
after 6 days, either at RT or 50 °C in chloroform or acetonitrile,
indicating that the cycloaddition reaction does not occur for
these complexes. For both 1 and 2, the potential dissociation of
8014 | Dalton Trans., 2012, 41, 8010–8021
This journal is © The Royal Society of Chemistry 2012

View Article Online
Scheme 7 Synthesis of cyclam-derived ligands.
Fig. 3 View of the molecular structure and atom labelling of the mono-
cation of [(cyclamAc)CoIIIN3]ClO4 (6). The counter-anion and hydrogen
atoms have been omitted for clarity.
Table 3 Azide stretching frequencies, ν(N3)asym, for the metal–azide
complexes described herein. IR spectra were measured as thin films
Complex
ν(N3)asym/cm−1
[(salen1)CoIII(N3)]2
[(salen2)FeIII(N3)]2
2015
2058
2070
2070
2046
2052
a
[(Me4cyclam)NiIIN3]PF6
[(Me4cyclam)CoIIN3]ClO4
Fig. 2 View of the molecular structure and atom labelling of the mono-
cations of (left) [(Me4cyclam)NiII(N3)]PF6 (3) and (right) [(Me4cyclam)-
CoII(N3)]ClO4 (4). The counteranions and hydrogen atoms have been
omitted for clarity.
[(cyclamAc)FeIIIN3]PF6
b
[(cyclamAc)CoIIIN3]ClO4
a Lit. value 2066 cm−1 (ref. 27b). b Lit. value 2051 cm−1 (ref. 22).
are critical in the successful formation of the triazolate product
(vide infra).
Structures of metal–triazolate complexes 7 and 8
Complexes {(Me4cyclam)CoII[N3C2(CO2Me)2]}ClO4 (7) and
{(Me4cyclam)NiII[N3C2(CO2Me)2]}ClO4 (8) were characterized
by X-ray single crystal analysis (Fig. 5). The experimental crys-
tallographic data are summarized in Table 1 and selected bond
lengths and angles are given in Table 2.
Both complexes 7 and 8 crystallize in the same space group
(monoclinic C2/c). In both structures the metal ion exhibits a
square-pyramidal coordination geometry where the N atoms of
the Me4cyclam ligand occupy the equatorial positions and the
triazolate anion is bound axially (Fig. 5). The 4,5-bis(methoxy-
carbonyl)-1,2,3-triazolate ligand binds symmetrically to the
metal centre, through its central N(3) atom, and the triazole
plane adopts a staggered orientation relative to the methyl groups
of Me4cyclam in order to minimize steric repulsions. Given this
C2 symmetry of the cations of 7 and 8, the two sets of equivalent
M–Nligand distances are 2.113 and 2.147 Å for the Ni complex
and 2.129 and 2.199 Å for the Co complex, while the M–Ntriazole
distances are 2.002 and 2.030 Å for 7 and 8, respectively
(Table 2). To the best of our knowledge, complexes 7 and 8
Scheme 8 Synthesis of metal–azide complexes using cyclam-derived
ligands.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 8010–8021 | 8015

View Article Online
Scheme 9 Survey of the cycloaddition reaction for various metal–azide complexes and the electron-deficient alkyne dimethyl
acetylenedicarboxylate.
represent the first structurally characterized NiII– and CoII–triazo-
late complexes formed upon a cycloaddition reaction; only one
other structurally characterized CoIII–triazolate complex has been
reported to date.33 Moreover, 7 and 8 are the first metal–triazo-
late complexes with a five-coordinated geometry around the
metal ion.32
Factors determining the cycloaddition reaction for metal–azide
complexes
In order to discern the factors that lead to a successful cyclo-
addition reaction between an alkyne and various metal–azide
complexes, we compared relevant metrical parameters of the
azide complexes obtained herein and two other structurally
characterized CoIII–azide complexes33,34 that have been pre-
viously shown to undergo a cycloaddition reaction with dimethyl
acetylenedicarboxylate.11 Analysis of metrical parameters
reveals that metal–azide complexes that lead to triazole for-
mation exhibit longer metal–azide bonds (>1.955 Å, Table 4),
which could be due to either the presence of axial donors with
stronger trans influence (e.g., PPh3 or py),11 or the presence of
metal centers in lower oxidation states (e.g., NiII and CoII).
However, while the Fe–azide bond length in [(salen2)FeIII(N3)]2
is 2.015 Å, no cycloaddition reaction was observed for this
complex. By contrast, shorter proximal N–N bonds (N–Nprox) of
the bound azide ligand are observed for the complexes under-
going the cycloaddition reaction (Table 4). In addition, the
difference between the long and short N–N bonds of the azide
ligand (i.e., N–Nprox and N–Ndist, respectively) is smaller
(<0.029 Å) for these complexes vs. those that do not form
Fig. 4 The cycloaddition reaction between [(Me4cyclam)NiII(N3)]ClO4
and dimethyl acetylenedicarboxylate in CHCl3 monitored by IR: (a)
before addition of alkyne and (b) after 6 h.
triazolate adducts (>0.053 Å). This bond length difference
(N–Nprox − N–N dist) provides a measure of the contribution of
8016 | Dalton Trans., 2012, 41, 8010–8021
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 4 Comparison of relevant structural parameters for metal–azide
complexes
d/Å
(N–N)prox
M–N3 N–Nprox N–Ndist (N–N)dist
−
Complex
[(salen1)CoIII(N3)]2
1.932 1.209
2.015 1.209
1.989 1.180
1.150
1.132
1.153
1.160
1.156
1.151
1.161
1.126
0.059
0.077
0.027
0.023
0.053
0.057
0.019
0.029
a
a
[(salen2)FeIII(N3)]2
a
[(Me4cyclam)NiIIN3]PF6
[(Me4cyclam)CoIIN3]ClO4 1.974 1.183
a
a
[(cyclamAc)FeIIIN3]PF6
1.931 1.209
1.936 1.208
2.014 1.180
1.955 1.155
[(cyclamAc)CoIIIN3]ClO4
a
b
(PPh3)CoIII(DH)2N3
(py)CoIII(DH)2N3
c
The azide adducts highlighted in bold yield triazole products.a This
work. b Ref. 33. c Ref. 34, DH = dimethylglyoximato.
stretch and the appearance of the triazolate product stretches. For
all substrates investigated, no triazolate product was obtained
when either 7 or 8 was used as the starting material, suggesting
that a highly electron deficient alkyne is required for the cyclo-
addition reaction to occur. Overall, these studies imply that the
targeted inorganic azide–alkyne cycloaddition reaction can only
occur for a limited set of metal–azide complexes and only when
very electron deficient alkynes are employed. Alternatively, use
of metal complexes in other oxidation states or stabilized by
ligands with various electronic properties may allow for the
cycloaddition reaction to proceed for less electron-deficient
alkynes. Such additional studies are currently underway.
Fig. 5 View of the molecular structure and atom labelling of the mono-
cation of (top) {(Me4cyclamNiII[N3C2(CO2Me)2]}ClO4 (7) and (bottom)
{(Me4cyclamCoII[N3C2(CO2Me)2]}ClO4 (8). The counteranion and
hydrogen atoms have been omitted for clarity.
Conclusion
the two possible resonance forms (a) or (b) to the electronic
structure of the metal–azide complexes, and strongly suggests
that the azide groups with a greater contribution from the reson-
ance form (b) (i.e., with a more symmetric charge density distri-
bution) are expected to undergo the cycloaddition reaction more
easily.
In summary, we have synthesized and structurally characterized
a series of Fe, Co and Ni mono-azide complexes that employ
biomimetic salen- and cyclam-derived ligands. The obtained
metal–azide complexes were investigated for their ability to
undergo an azide–alkyne cycloaddition reaction – an “inorganic
click reaction”, with several alkyne substates. Our studies reveal
that one Co–azide and one Ni–azide complex of the neutral
ligand Me4cyclam react with the electron deficient alkyne
dimethyl acetylenedicarboxylate to generate the metal–triazolate
products, which were structurally characterized. Use of less elec-
tron deficient alkynes did not generate the corresponding cyclo-
addition products. In addition, use of the anionic ligand
cyclamAc seems to abolish the ability of the corresponding
metal–azide complexes to react with any alkyne, suggesting that
even subtle changes in the electronic properties of the metal
centre leads to an azide ligand whose polarity renders it unreac-
tive toward a cycloaddition reaction. Analysis of the structural
parameters of the investigated metal–azide complexes suggests
that a more symmetric charge density distribution within the
azide moiety is needed for the formation of the metal–triazolate
product. In this context, it will be interesting to study whether
the azide adducts of non-heme iron enzymes, as opposed to
small inorganic model complexes, are capable of undergoing a
cycloaddition reaction with alkyne substrate analogues. It is
expected that the active site of these enzymes will promote the
“click reaction” to a greater extent than a model complex by
Reaction of metal–azide complexes with other alkynes
Our results suggest that the Ni– and Co–azide complexes 7 and
8 with the Me4cyclam ligand can undergo a cycloaddition reac-
tion with dimethyl acetylenedicarboxylate to yield the corre-
sponding 1,4-triazolate products. Since the alkyne employed
above is very electron deficient, we set out to investigate the
click reaction of 7 and 8 with less electron deficient alkynes
(Scheme 10). The selected alkynes were chosen as they resemble
the methylated amine substrates of the histone demethylase
enzymes, a new class of non-heme iron enzymes.35 The cyclo-
addition studies were performed similarly to those with dimethyl
acetylenedicarboxylate, by adding 5 equiv. alkyne to 1 equiv.
metal–azide complex in either chloroform or acetonitrile at RT or
50 °C and followed by IR for the disappearance of the azide
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 8010–8021 | 8017

View Article Online
Kappa Apex-II Charge Coupled Device (CCD) Detector system
single-crystal X-Ray diffractometer equipped with an Oxford
Cryostream LT device. Data were collected using graphite-mono-
chromated Mo-Kα radiation (λ = 0.71073 Å) from a fine-focus
sealed-tube X-ray source. Preliminary unit cell constants were
determined with a set of 36 narrow frame scans. Typical data
sets consist of a combination of ϖ and ϕ scan frames with typical
scan width of 0.5° and counting time of 15–30 s per frame at a
crystal to detector distance of ∼4.0 cm. The collected frames
were integrated using an orientation matrix determined from the
narrow frame scans. Apex II and SAINT software packages37
were used for data collection and data integration. Analysis of
the integrated data did not show any decay. Final cell constants
were determined by global refinement of reflections from the
complete data set. Data were corrected for systematic errors
using SADABS37 based on the Laue symmetry using equivalent
reflections. Structure solutions and refinement were carried out
using the SHELXTL-PLUS software package.38 The structures
were refined with full-matrix least-squares refinement by mini-
mizing ∑w(Fo − Fc ) . All non-hydrogen atoms were refined
anisotropically to convergence. All H atoms were added in the
calculated position and were refined using appropriate riding
models (AFIX m3). Selected crystals data and structure refine-
ment parameters are listed in Tables 1 and 2. All data were col-
lected at 100 K, except complex 7 for which data were collected
at both 100 and 300 K. For the structural characterization of 7,
multiple data sets were collected on different crystals at room
temperature and at 100 K. The perchlorate anion in the structure
of 7 is disordered and sits on a 2-fold rotation axis. To be able to
resolve the disorder, the symmetry equivalent atoms were gener-
ated and refined with half the expected occupancies and with
PART-1/-2 instructions. The Ni(Me4cyclam) fragment of the
complex shows whole molecule disorder and the disorder was
modeled as two overlapping motifs. The relative occupancies
were refined using free variables. The disordered atoms were
refined with geometrical restraints and displacement parameter
restraints as listed in the .CIF file.
Scheme 10 Alkyne substrates employed in the cycloaddition reaction
with the [(Me4cyclam)MII(N3)]ClO4 complexes (M = Ni, Co).
increasing the local concentration of the two substrates and thus
the rate of the cycloaddition reaction. Our current research
efforts are aimed at employing such enzyme-templated cyclo-
addition reactions for the development of specific inhibitors for
non-heme iron enzymes. Of particular interest are histone
demethylases, a new class of non-heme iron enzymes that have
recently been shown to be overexpressed in cancer cells.36 Thus,
high-affinity histone demethylase inhibitors that exhibit
increased specificity can be used in the development of alterna-
tive cancer therapeutics.35
2
2 2
Experimental section
Commercially available reagents were used as received. All
reagents, organic and inorganic, were of high purity grade and
obtained from E. Merck, Fluka, Chemie and Aldrich Co. The
solvents were dried by passing over an alumina column.
Physical measurements
1H (300.121 MHz) NMR spectra were recorded on a Varian
Mercury-300 spectrometer. Chemical shifts are reported in ppm
and referenced to residual solvent resonance peaks. Infrared
spectra were measured as thin films on a KBr or NaCl plate
using a Perkin Elmer Spectrum BX FT-IR spectrometer in the
4000–400 cm−1 range. UV-Vis spectra were recorded on a
Varian Cary 50 Bio spectrophotometer. ESI-MS experiments
were performed on a Bruker Maxis QTOF mass spectrometer
with an electron spray ionization source at the Washington Uni-
versity Mass Spectrometry Resource, a NIH Research Resource
supported by Grant No. P41RR0954.
Synthesis of ligands
The N,N′-bis(salicylidene)-1,2-ethylenediimine (H2salen1) and
N,N′-bis(salicylidene)-1,2-propylenediimine (H2salen2) ligands
were synthesized by stirring 2 equiv. of 1,2-ethyldiamine or 1,2-
diaminopropane, respectively, and 1 equiv. of salicylaldehyde in
ethanol for 2 h.23 The final solution was used for the synthesis
of the corresponding metal complexes.
Synthesis of 1,4,8,11-tetrametyl-1,4,8,11-tetraazacyclotetra-
decane (Me4cyclam). The Me4cyclam ligand was synthesized
following a literature procedure,25 with some slight modifi-
cations. A solution containing 1.5 g of cyclam (7.5 mmol),
8.5 mL of 98% aqueous formic acid, 6.6 mL of formaldehyde
(38% aqueous solution) and 5 mL of water was refluxed for
24 h. The reaction mixture was diluted with 15 mL of water,
transferred to a 100 mL beaker, and cooled in an ice-bath. A
concentrated solution of sodium hydroxide (15 g of NaOH dis-
solved in 50 mL of water) was slowly added with stirring until
pH > 12. The temperature of the solution was kept below 25 °C.
The solution was then extracted with 5 × 50 mL portions of
X-Ray crystallography
X-Ray diffraction quality crystals of 1–4 were obtained by slow
evaporation of the mother liqueur. The Co complex 6 was crys-
tallized by slow diffusion of anhydrous diethyl ether into a sol-
ution of acetonitrile. The metal–triazolate complexes 7 and 8
were obtained by slow diffusion of hexane in a dichloromethane
solution. Suitable crystals of appropriate dimensions were
mounted on Mitgen loops in random orientations. Preliminary
examination and data collection were performed using a Bruker
8018 | Dalton Trans., 2012, 41, 8010–8021
This journal is © The Royal Society of Chemistry 2012

View Article Online
(b) Derived from cyclam ligands
CHCl3, the extracts were combined and dried over sodium sul-
phate, and then were concentrated to yield an oily residue. The
residue crystallized at 4 °C overnight and the product was dried
under vacuum for 2 h at 50 °C. Yield 80%. 1H NMR (δ, CDCl3,
Synthesis of [(Me4cyclam)NiII(N3)] (X = ClO4 or PF6 ):
−
−
[(Me4cyclam)NiII(N3)]ClO4
To a stirred solution of Me4cy-
27a,b
clam (100 mg, 0.4 mmol) in 5 mL of ethanol was added an
aqueous solution (5 mL) of NaN3 (100 mg, 1.5 mmol). After
1 min, solid Ni(ClO4)2·6H2O was added to the stirred solution
(142.85 mg, 0.4 mmol). The solution was heated to 70 °C and
stirred for 1 h. The resulting light green solution was cooled to
room temperature, filtered to remove the excess NaN3, and then
kept at 4 °C. After 48 h, green needle crystals formed. Yield:
60%. FTIR (NaCl, cm–1): 3392, 2996, 2939, 2864 (ν(C–H)),
2070 (ν(N3)), 1471, 1433, 1324, 1307 (ν(C–N)), 1288, 1240,
1088 (ν(ClO4−)). UV-Vis recorded in MeCN, λmax/nm (ε/M−1
cm−1): 683 (85), 375 (2450). Anal. Calc. for C14H32NiN7ClO4
(MW 456.59): C, 36.83; H, 7.06; N, 21.47. Found: C, 36.79; H,
7.62; N, 21.46%. ESI-MS: m/z 359.16, 359.20 expected for
[(Me4cyclam)NiII(N3)]+.
3
300 MHz): 1.643 (q, J(H–H) = 7 Hz, 4H, Ha), 2.198 (s, 12 H,
4Me), 2.427 (m, 16 H, Hb,c). ESI-MS: m/z 257.3, 257.3 expected
for [Me4cyclamH]+.
Synthesis of 1,4,8,11-tetraazacyclotetradecane-1-acetic acid
terahydrochloride (cyclamAc·4HCl). The cyclamAcH ligand
was synthesized following a literature procedure,26 with some
slight modifications. To 1 g (5 mmol) of cyclam dissolved in
15 mL of EtOH and 3 mL of H2O, 48 mg (2 mmol) of LiOH
and 186 mg (1 mmol) of ICH2COOH dissolved in 4 mL of H2O
were added at 5 °C. The mixture was refluxed for 4 h. The EtOH
was then evaporated, and the alkaline aqueous solution was
treated with CHCl3 (4 × 10 mL) to remove non-reacted cyclam.
The aqueous solution was dried at vacuum. Addition of 2 mL of
HCl and 2 mL of EtOH lead to precipitation of the protonated
tetrachloride product, which can be converted to the free-base
Using the same procedure but starting with Ni(NO3)2·6H2O and
upon addition of an excess of KPF6, the complex [(Me4cyclam)
NiII(N3)]PF6 was formed as green needle crystals suitable for
X-ray crystallography. ESI-MS (m/z): 359.16, 359.20 expected
1
form upon neutralization with NaOH. Yield: 41%. H NMR (δ,
for [(Me4cyclam)NiII(N3)]+.
CDCl3, 300 MHz): 1.94 (m, 4H, Ha), 2.8–3.5 (m, 16 H, Hb,c),
3.6 (s, 2H, CH2COOH). ESI-MS: m/z: 259.19, 259.21 expected
for [cyclamAcH]+.
27c
Synthesis of [(Me4cyclam)CoII(N3)]]ClO4
To a stirred sol-
ution of Me4cyclam (100 mg, 0.4 mmol) in 5 mL of ethanol was
added an aqueous solution (5 mL) of NaN3 (100 mg, 1.5 mmol).
After 1 min, solid Co(ClO4)2·6H2O was added to the stirred sol-
ution (143 mg, 0.4 mmol). The solution was heated at 70 °C and
stirred for 1 h. The resulting dark-purple solution was cooled to
room temperature, filtered to remove the excess NaN3, and then
kept at 4 °C. After 48 h, dark-purple crystals formed. Yield:
67%. FTIR (NaCl, cm–1): 3014, 2925, 2867 (ν(C–H)), 2070
(ν(N3)), 1476, 1089, 1275. UV-Vis recorded in MeCN, λmax/nm
(ε/M−1 cm−1): 597 (118), 337 (2973). Anal. Calc. for
C14H32CoN7ClO4 (MW 456.83): C, 36.81; H, 7.06; N, 21.46.
Found: C, 36.35; H, 7.06; N, 21.45%. ESI-MS: m/z = 358.16,
358.21 expected for [(Me4cyclam)CoII(N3)]+.
Synthesis of metal–azide complexes
(a) Derived from salen ligands
Synthesis of [(salen1)CoIII(N3)]2. To a solution of H2Salen1
(27 mg, 0.1 mmol) in EtOH heated at 60 °C, a solution of Co-
(OAc)2·4H2O (20 mg, 0.1 mmol) in 5 mL of EtOH and an
aqueous solution (2 mL) of NaN3 (6.5 mg, 0.1 mmol) were
added. The solution was stirred at 60 °C for 30 min to form a
brown precipitate. The reaction was cooled to room temperature
and the solid was collected by filtration. X-Ray quality crystals
were obtained by slow evaporation of the resulting solution in
air. Yield: 91%. FTIR (NaCl, cm–1): 2015(ν(N3)), 1645, 1557
(ν(CvN)), 1471, 1447, 1338 (ν(C–N)), 1287. UV-Vis recorded
in MeCN, λmax/nm (ε/M−1 cm−1): 390 (8300), 314 (12 400),
257 (43 000). Anal. Calc. for C32H28Co2N10O4·7H2O (MW
860.60): C, 44.66; H, 4.92; N, 16.28. Found: C, 44.57; H, 4.70;
N, 16.00%. ESI-MS: m/z 325.00, 325.04 expected for [(salen1)-
22
Synthesis of [(cyclamAc)FeIII(N3)]PF6 To 150 mg of cycla-
mAcH (0.58 mmol) dissolved in 5 mL degassed water was
added 94.73 mg (0.58 mmol) of FeCl3 in 5 mL degassed water.
The resulting solution was brought to a gentle reflux and heated
for 90 min as an orange color developed. The solution was
cooled to room temperature and excess NaN3 was added
(124 mg, 1.9 mmol) to yield a dark-red solution. The resulting
solution was stirred for 2 h at room temperature and sub-
sequently filtered to yield a clear red filtrate. Addition of an
aqueous solution (5 mL) of 350 mg KPF6 (1.9 mmol) yielded
red–orange microcrystals upon cooling at 4 °C overnight that
were collected by filtration, washed with 2 × 10 mL diethyl
ether, and dried under vacuum. Yield: 45%. FTIR (NaCl, cm−1):
3260 (ν (N–H)), 2046 (ν(N3)), 1651 (ν(CvO)), 838, 558
(ν(PF6−)). UV-Vis recorded in MeCN, λmax/nm (ε/M−1 cm−1):
460 (2500), 305 (4535), 276 (sh), 245 (11175). ESI-MS: m/z =
355.18, 355.14 expected for [(cyclamAc)FeIII(N3)]+.
CoIII]+;
CoIIIN3CoIII(salen1)]+.
692.09,
692.10
expected
for
[(salen1)-
Synthesis of [(salen2)FeIII(N3)]2. To a solution of H2Salen2
(57 mg, 0.2 mmol), an ethanol solution (10 mL) of Fe
(NO3)3·9H2O (80.8 mg, 0.2 mmol) and an aqueous solution
(4 mL) of NaN3 (13 mg, 0.2 mmol) were added under stirring.
After allowing the resulting purple solution to stand in air for 10
days, black crystals were formed upon slow evaporation of the
solvent. Yield: 77%. FTIR (NaCl, cm–1): 2926, 2355 (ν(C–H)),
2058(ν(N3)), 1618, 1598, 1541 (ν(CvN)), 1468, 1444, 1391
(ν(C–N)), 1299. UV-Vis recorded in MeCN, λmax/nm (ε/M−1
cm−1): 454 (2575), 319 (6167), 294 (sh) 256 (13 500). Anal.
Calc. for C34H32Fe2N10O4·4CH3CH2OH·2H2O (MW 976.68):
C, 51.65; H, 6.19; N, 14.34. Found: C, 52.11; H, 5.85; N,
13.94%. ESI-MS: m/z 336.05, 336.08 expected for [(salen2)-
FeIII]+.
Synthesis of [(cyclamAc)CoIII(N3)]ClO4. To 150 mg of cycla-
mAcH (0.58 mmol), dissolved in 5 mL degassed water was
added 214.41 mg (0.58 mmol) of Co(ClO4)2·6H2O in 5 mL
degassed water. The resulting solution was brought to a gentle
reflux and heated for 90 min as an orange color developed. The
solution was cooled to room temperature and an excess of NaN3
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 8010–8021 | 8019

View Article Online
was added (124 mg, 1.9 mmol) to yield a dark-purple solution.
The resulting solution was concentrated to 2 mL and excess
NaN3 was filtered. The solvent was then removed and the purple
product was washed with 2 × 10 mL diethyl ether, collected by
filtration, and dried under vacuum. X-ray quality crystals were
obtained by slow diffusion of ether into a MeCN solution. Yield:
50%. FTIR (NaCl, cm−1): 2923 (ν(N–H)), 2052 (ν(N3)), 1686
(ν(CvO)), 837, 560. UV-Vis recorded in MeCN, λmax/nm (ε/
M−1 cm−1): 545 (200), 317 (4727). Anal. Calc. for C12H25ClCo-
N7O6·7H2O (MW 581.85): C, 24.77; H, 6.41; N, 16.85. Found:
C, 24.44; H, 6.18; N, 16.76%. ESI-MS: m/z = 358.18, 358.14
expected for [(cyclamAc)CoIII(N3)]+.
We also thank Barrie Cascella for the synthesis of propargyl
amine substrates. L. M. M. is a Sloan Fellow.
Notes and references
1 H. C. Kolb, M. G. Finn and K. B. Sharpless, Angew. Chem., Int. Ed.,
2001, 40, 2004.
2 (a) R. Huisgen, Proc. Chem. Soc., 1961, 357; E. Lieber, R. L. Minnis
and C. N. R. Lao, Chem. Rev., 1965, 65, 377; (b) R. Huisgen,
G. Szeimies and L. Möbius, Chem. Ber., 1976, 100, 2949; (c) The Chem-
istry of Azido Group, ed. S. Patai, Interscience, New York, 1971.
3 (a) K. Nomiya, R. Noguchi and M. Oda, Inorg. Chim. Acta, 2000, 298,
24; (b) S. S. van Berkel, A. J. Dirks, M. J. Debets, F. L. van Delft, J. J. L.
M. Cornelissen and F. P. J. T. Tutjes, ChemBioChem, 2007, 8, 1504;
(c) A. Tam, U. Arnold, M. B. Soellner and R. T. Raines, J. Am. Chem.
Soc., 2007, 129, 12670; (d) E. Bokor, T. Docsa, P. Gergely and
L. Somsák, Bioorg. Med. Chem., 2010, 18, 1171; (e) E. M. Sletten and
C. R. Bertozzi, Acc. Chem. Res., 2011, 44, 666.
4 (a) R. P. Singh, R. D. Verma, D. T. Meshri and J. M. Shreeve, Angew.
Chem., Int. Ed., 2006, 45, 3584; (b) B. C. Tappan, M. H. Huynh, M.
A. Hiskey, D. E. Chavez, E. P. Luther, J. T. Mang and S. F. Son, J. Am.
Chem. Soc., 2006, 128, 6589.
5 (a) C. W. Tornoe, C. Christensen and M. Meldal, J. Org. Chem., 2002,
67, 3057; (b) V. V. Rostovtsev, L. G. Green, V. V. Fokin and K.
B. Sharpless, Angew. Chem., Int. Ed., 2002, 41, 2596.
Synthesis of metal–triazolate complexes
Synthesis of {(Me4cyclam)NiII[N3C2(CO2Me)2]}ClO4. To
10 mg [(Me4cyclam)NiII(N3)]ClO4 (0.02 mmol) dissolved in
chloroform were added 14 μl of dimethyl acetylenedicarboxylate
(0.11 mmol). The solution was stirred for 6 h and the formation
of the triazolate product was followed by FT-IR. The product
was precipitated in hexane, filtered, and dried under vacuum.
X-Ray quality crystals were obtained by slow diffusion of
hexane into a dichloromethane solution at 4 °C. Yield: 91%.
FTIR (NaCl, cm−1): 2955, 2870, 1726 (ν(CvO)), 1480
(ν(NvN)), 1441, 1328, 1270 (ν(C–O), 1202, 1171, 1090, 964,
826, 806, 778, 750, 623. UV-Vis recorded in MeCN, λmax/nm
(ε/M−1 cm−1): 636 (88), 514 (149), 386 (731). Anal. Calc. for
C20H38ClNiN7O8·0.5C6H14 (MW 641.79): C, 43.04; H, 7.07; N,
15.28. Found: C, 43.01; H, 6.93; N, 15.34%. ESI-MS:
6 R. A. Copeland, M. R. Harpel and P. J. Tummino, Expert Opin. Ther.
Targets, 2007, 11, 967.
7 W. G. Lewis, L. G. Green, F. Grynszpan, Z. Radić, P. R. Carlier,
P. Taylor, M. G. Finn and K. B. Sharpless, Angew. Chem., Int. Ed., 2002,
41, 1053.
8 (a) H. C. Kolb and K. B. Sharpless, Drug Discovery Today, 2003, 8,
1128; (b) S. K. Mamidyala and M. G. Finn, Chem. Soc. Rev., 2010, 39,
1252; (c) X. Hu and R. Manetsch, Chem. Soc. Rev., 2010, 39, 1316.
9 T. Suzuki, Y. Ota, Y. Kasuya, M. Mutsuga, Y. Kawamura, H. Tsumoto,
H. Nakagawa, M. G. Finn and N. Miyata, Angew. Chem., Int. Ed., 2010,
49, 6817.
10 There are several previous reports describing the cycloaddition reaction
between metal–azide complexes and alkynes, and the term “inorganic
click reaction” has been introduced before. Our current study reports
cycloaddition reactions between first-row metal–azide complexes and an
alkyne as a proof-of-concept for an enzyme-templated cycloaddition reac-
tion as a novel strategy to synthesize metalloenzyme inhibitors. In this
context, the proposed metalloenzyme-templated reaction is an inorganic
variant of the in situ click reaction employed by Sharpless et al. in
enzyme inhibitor design (ref. 7).
11 (a) T. Kemmerich, J. H. Nelson, N. E. Takach, H. Boehme, B. Jablonski
and W. Beck, Inorg. Chem., 1982, 21, 1226; (b) B. T. Hsieh, J.
H. Nelson, E. B. Milosavljevic, W. Beck and T. Kemmerich, Inorg.
Chim. Acta, 1987, 133, 267.
12 P. Paul and K. Nag, Inorg. Chem., 1987, 26, 2969.
13 M. Herberhold, A. Goller and W. Z. Milius, Z. Anorg. Allg. Chem., 2003,
629, 1162.
m/z
=
498.30, 498.23 expected for {(Me4cyclam)
NiII[N3C2(CO2Me)2]}+.
Synthesis of {(Me4cyclam)CoII[N3C2(CO2Me)2]}ClO4. To
10 mg [(Me4cyclam)CoII(N3)]ClO4 (0.02 mmol) dissolved in
chloroform were added 14 μl of dimethyl acetylenedicarboxylate
(0.11 mmol). The solution was stirred for 10 h and the formation
of the triazolate product was followed by FT-IR. The product
complex was precipitated in hexane, filtered, and dried under
vacuum. X-ray quality crystals were obtained by slow diffusion
of hexane into a dichloromethane solution at room temperature.
Yield: 84%. FTIR (NaCl, cm−1): 2961, 2923, 2861, 2064, 1728
(ν(CvO)), 1479 (ν(NvN)), 1454, 1302, 1261, 1226 (ν(C–O),
1170, 1090, 1042, 1019, 960, 842, 822, 806, 779, 732,
623 cm−1. UV-Vis recorded in MeCN, λmax/nm (ε/M−1 cm−1):
552 (30), 483 (43), 325 (sh). Anal. Calc. for C20H38ClCo-
N7O8·2H2O (MW 616.97): C, 38.83; H, 6.67; N, 15.44. Found:
C, 38.42; H, 6.39; N, 15.35%. ESI-MS: m/z = 499.16, 499.23
expected for {(Me4cyclam)CoII[N3C2(CO2Me)2]}+.
14 C. W. Chang and G. H. Lee, Organometallics, 2003, 22, 3107.
15 L. Busetto, F. Marchetti, S. Zacchini and V. Zanotti, Inorg. Chim. Acta,
2005, 358, 1204.
16 (a) K. S. Singh, C. Thone and M. R. Kollipara, J. Organomet. Chem.,
2005, 690, 4222; (b) K. S. Singh, V. Svitlyk and Y. Mozharivskyj, Dalton
Trans., 2011, 40, 1020.
17 J. A. K. Bauer, T. M. Becker and M. Orchin, J. Chem. Crystallogr., 2004,
34, 843.
18 (a) D. V. Partyka, J. B. Updegraff, M. Zeller, A. D. Hunter and T.
G. Gray, Organometallics, 2007, 26, 183; (b) D. V. Partyka, L. Gao, T.
S. Teets, J. B. Updegraff, N. Deligonul and T. G. Gray, Organometallics,
2009, 28, 6171.
19 K. Pachhunga, P. J. Carroll and K. M. Rao, Inorg. Chim. Acta, 2008, 361,
2025.
20 F. C. Liu, Y. L. Lin, P. S. Yang, G. H. Lee and S. M. Peng, Organometal-
lics, 2010, 29, 4282.
21 T. J. Del Castillo, S. Sarkar, K. A. Abboud and A. S. Veige, Dalton
Trans., 2011, 40, 8140.
Acknowledgements
We thank the Department of Chemistry at Washington University
for start-up funds, the Department of Defense, Breast Cancer
Research Program for a Concept Award (BC097014) to
L. M. M., and the National Science Foundation (MRI,
CHE-0420497)
for
the
purchase
of
the
ApexII
diffractometer. E. E. thanks the Generalitat de Catalunya and
Program Fulbright for a one-year Fulbright Fellowship Award.
22 C. G. Grapperhaus, B. Mienert, E. Bill, T. Weyhermüller and
K. Wieghardt, Inorg. Chem., 2000, 39, 5306.
8020 | Dalton Trans., 2012, 41, 8010–8021
This journal is © The Royal Society of Chemistry 2012

View Article Online
23 (a) Z.-D. Liu, M.-Y. Tan and H.-L. Zhu, Acta Crystallogr., Sect. E:
Struct. Rep. Online, 2004, 60, m910; (b) Z.-L. You and H.-L. Zhu, Acta
Crystallogr., Sect. E: Struct. Rep. Online, 2004, 60, m1046.
24 M. Weil and A. D. Khalaji, Anal. Sci., 2008, 24, x19.
25 E. K. Barefield and F. Wagner, Inorg. Chem., 1973, 12, 2435.
26 M. Struder and T. A. Kaden, Helv. Chim. Acta, 1986, 69, 2081.
27 For the Ni complex: (a) A. Escuer, R. Vicente, M. S. El Fallah,
X. Solans and M. Font-Bardia, Inorg. Chim. Acta, 1996, 247, 85; (b) M.
J. D’Aniello Junior, M. T. Mocella, F. Wagner, E. K. Barefield and I.
C. Paul, J. Am. Chem. Soc., 1975, 97, 192. For the Co complex:
(c) S. Reiner, M. Wicholas, B. Scott and R. D. Willett, Acta Crystallogr.,
Sect. C: Cryst. Struct. Commun., 1989, 45, 1694.
32 The Cambridge Structural Database, http://www.ccdc.cam.ac.uk/, search
performed December 2011.
33 J. H. Nelson, N. E. Takach, N. Bresciani-Pahor, L. Randaccio and
E. Zangrando, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 1984,
40, 742.
34 A. Clearfield, R. Gopal, R. J. Kline, M. L. Sipski and L. O. Urban, J.
Coord. Chem., 1978, 8, 5.
35 (a) Y. Zhang and R. J. Klose, Nat. Rev. Mol. Cell Biol., 2007, 8, 307;
(b) A. Spannhoff, A.-T. Hauser, R. Heinke, W. Sippl and M. Jung, Chem-
MedChem, 2009, 4, 1568; (c) N. R. Rose, M. A. McDonough,
O. N. F. King, A. Kawamura and C. J. Schofield, Chem. Soc. Rev., 2011,
40, 4364.
28 B. Bosnich, C.-K. Poon and M. L. Tobe, Inorg. Chem., 1965, 4,
1102.
36 (a) L. A. Boyer, K. Plath, J. Zeitlinger, T. Brambrink, L. A. Medeiros, T.
L. Lee, S. S. Levine, M. Wernig, A. Tajonar, M. K. Ray, G. W. Bell, A.
P. Otte, M. Vidal, D. K. Gifford, R. A. Young and R. Jaenisch, Nature,
2006, 441, 349; (b) P. A. C. Cloos, J. Christensen, K. Agger,
A. Maiolica, J. Rappsilber, T. Antal, K. H. Hansen and K. Helin, Nature,
2006, 442, 307.
29 (a) A. W. Addison, T. N. Rao, J. Reedijk, J. Van Rijn and G.
C. Verschoor, J. Chem. Soc., Dalton Trans., 1984, 1294; (b) R.
D. Willett, G. Pon and C. Nagy, Inorg. Chem., 2001, 40, 4342.
30 P. V. Bernhardt and G. A. Lawrance, Coord. Chem. Rev., 1990, 104, 297–
343.
37 Bruker Analytical X-Ray, Madison, WI, 2008.
31 M. Costas, M. P. Mehn, M. P. Jensen and L. Que Jr., Chem. Rev., 2004,
104, 939.
38 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2007,
64, 112.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 8010–8021 | 8021