Inorganic Chemistry
Article
starting at zero order and ending at first order indicates that UV
light is the limiting factor initially, but this limitation is reduced
as the inner filter from [Cu([24.31]adz)H2O]2+ is removed by
dechelation. Instead, the amount of substrate, also [Cu([24.31]-
adz)H2O]2+, becomes the limitation. Finally, the decline at
285 nm followed that at 600 nm. That was not the case in
5.0 M HCl. The difference in reaching 5% decline (measured
at 310 nm where the decline was fastest) is only 70-fold.
Moreover, the rate depends on which wavelength it is measured
at. Apparently, Cl− protects the chelate from light-induced
dechelation as explained above.
[Cu([24.31]adz)HO]+, indicates that one of the products could
involve deprotonation of the coordinated water.
Consequences for Copper Radiochemistry. Competi-
tive binding of Cu2+ ions from salts and buffers is generally not
considered a problem in radiochemistry, although commonly
used buffers are slightly CuII binding.31,32 In regular chemistry,
with concentrations of Cu2+ and the chelator in at least the
millimolar range, this is an advantage because it helps to keep
the Cu2+ ions in solution. However, in radiochemistry, the
concentration of Cu2+ can be in the subnanomolar range
(under carrier-free conditions).31 Under these conditions,
competitive binding from buffers might be very important.
Presently, Cu2+ labeling of peptides and protein-targeting
molecules relies on heat to increase the reaction speed suf-
ficiently, with the half-life of the isotope in mind. Unfortu-
nately, this either restricts the kinetic stability of the chelator or
excludes heat-labile targeting molecules. UV-induced chelation
of Cu2+ offers a possible alternative. However, like heat, UV
irradiation can denature proteins and cause DNA cleavage so
the use of a suitable UV filter is important.33,34 That said, the
only way UV irradiation is presently used for CuII coordination
is by causing an electron transfer within the ligand, thus
enabling it to coordinate to CuII, and microwaving at high
temperatures, as required for some ligands, is also a harsh
treatment.35,36
It is normal to compare the kinetic stability of copper(II)
complexes by acid decomplexation.24,25 This is a stepwise pro-
cess where the N−Cu coordination bonds one by one are broken
and replaced with protonation of the amine groups. However,
because photoexcitation enables a faster ligand exchange, there is a
risk that the kobs values reported in various papers are much too
high.21,26,27 This would be the case, if the dissociation of tetra-
azacopper complexes in an acid solution was studied kinetically by
spectrophotometry, and the reaction rate was enhanced by UV
irradiation. Furthermore, acido ligands like Cl− can stabilize both
the chelate and free Cu2+, making the results difficult to compare.
Effects of Solute O2 on the Light-Induced Chelation
Rate. Experiments performed under a normal atmosphere and
under argon showed that chelation without O2 was about 130%
higher after 600 s than chelation under a normal atmosphere
(Figure 7). O2 is known to form copper(II) superoxo com-
plexes by the oxidation of CuI.28 These can be quite stable, but
in this case, where CuI in D is oxidized by O2, the O2•− radical
is adjacent to an >NH•+ radical. Therefore, rapid redelivery of
CONCLUSION
■
The UV spectra of a mixture of Cu2+ and [24.31]adz in water
indicates the existence of a long-lived two-coordinated copper(II)
complex (not counting water ligands) with the adamanzane at
pH 6−7.5, which does not form at lower pH. Irradiation of the
labile complex in the LMCT band leads to photodeprotonation
and subsequently to the formation of [Cu([24.31]adz)H2O]2+
at a rate 7800-fold higher at 25 °C. At very low pH, this process
is reversed because photoexcitation of the chelate results in the
breakage of Cu−N bonds and the fact that CuI prefers a
different coordination geometry than CuII and because CuI
does not coordinate well to amines. This leads to increased
acid-decomplexation rates compared to the rates in darkness.
The presence of anions in the solution during chelation was
found to have three major effects: competitive inhibition from
CuII binding anions, inhibition of the photoinduced trans-
chelation from UV-absorbing anions, and photoredox inhib-
ition from ligands able to act as electron donors in an LMCT
reaction.
the electron from O2 to >NH•+ can be expected, with B as
•−
the result. Thus, the overall result of O2 presence is a higher
rate of D → B.
Side Reaction. To investigate possible side reactions due
to UV irradiation, pure Cu([24.31]adz)(ClO4)2 in an aqueous
solution was irradiated with and without a NaBr UV filter
with a cutoff at 221 nm in front of the sample (Figure S6 in
the Supporting Information). Apparently, only UV light with
wavelengths shorter than 220 nm is energetic enough to induce
the side reaction. Mauralidharan and Ferraudi have shown that
high-energy photons can open a macrocycle, forming a C−C
double bond.23,29 In the adamanzanes, it is also possible that
the cross bridge is broken, yielding a cyclen with two side
chains, one of them containing a double bond. The reaction has
been studied by MS with no sign of decomposition products as
in accordance with this model (data not shown): monocyclic
chelates with a double bond resulting from a C−C bond break
will have the same molecular weight as the adamanzane chelate,
so this type of reaction is undetectable by MS. In contrast,
imine formation, as described in a macrocycle by Chrisian et al.,
would be detectable by MS.30 The lack of structure that can
stabilize a double bond prevents the reaction at low energies.
The results of irradiation with and without O2 and through a
NaCl filter and not (Figure 9) speak against cyclen formation as
the sole explanation. That is because the overlapping peaks
between 220 and 280 nm found without O2 change shape and
intensity when O2 is present, indicating that more than one
reaction is active and that one of the reactions has to involve
O2. It is possible that part of the effect of a bromide filter seen
in Figure S6 in the Supporting Information is to prevent a
secondary reaction involving the formation of oxygen radicals.
The isosbestic point at 286 nm, within the error of measurement
of the isosbestic point at 285 nm for [Cu([24.31]adz)H2O]2+ and
If UV irradiation in the LMCT band is to be used as a
chelation technique, the effects of the acido anions, dissolved
O2, and high-energy UV photons have to be taken into account.
ASSOCIATED CONTENT
■
S
* Supporting Information
Supporting Figures S1−S6. This material is available free of
AUTHOR INFORMATION
Corresponding Author
■
ACKNOWLEDGMENTS
■
Bente Nielsen is thanked for synthesis of the adamanzanes and
for some of the spectrophotometric measurements. Ulla Olsen
is thanked for helping with the actinometric experiments.
12712
dx.doi.org/10.1021/ic201839w|Inorg. Chem. 2011, 50, 12705−12713